From bryan at eevolved.com  Sat Mar  3 01:31:14 2001
From: bryan at eevolved.com (Bryan Mongeau)
Date: Sat, 03 Mar 2001 06:31:14 GMT
Subject: Adding a method to an instance
References: <G2%n6.360714$JT5.11606234@news20.bellglobal.com> <6621ato2vknc667m3iuqvmm8vcj0h99qqk@4ax.com>
Message-ID: <Sc0o6.361217$JT5.11613656@news20.bellglobal.com>

Kirby Urner wrote:

Thanks Kirby,

> In saying f.__class__, you point back to the superclass foo.
> If you want to poke newMethod into object f alone, then you
> should just go:
> 
> f = newMethod

This I do not want. Maybe I should have been more clear. I would like to 
add "newMethod" to the instance of class foo so that later I can call:

f.newMethod()

> 
> Here's a transcript that addresses your question:
> 
>  >>> class foo:
> def duh(self):
>             print "Duh!"

Is my question really that bad? :)
 
>             
>  >>> o = foo()
>  >>> o
>  <__main__.foo instance at 00B9719C>
>  >>> k = lambda x: x+2
>  >>> k(2)
>  4
>  >>> o.othermethod = k

k is a lambda function. I would need the new method to be able to access 
internal class attributes, thus require the "self" parameter of class 
methods.

>  >>> o.othermethod(3)
>  5
>  >>> dir(o)
>  ['othermethod']
>  >>> dir(foo)
>  ['__doc__', '__module__', 'duh']
>  >>> o.__class__
>  <class __main__.foo at 00BB3ECC>
>  >>> foo
>  <class __main__.foo at 00BB3ECC>
>  >>> o
>  <__main__.foo instance at 00B9719C>
> 
> Note re the last three commands:  o.__class__ is a pointer to
> the class foo in memory, so you would be adding to the superclass
> to put a method in there.  You want to insert your method into
> the instance/object, so don't involve its __class__.

Okay, based on these statements and referring to my previous example, I 
should be able to do this:

def newMethod(self):
  print self.bar

f.newMethod = newMethod

Problem is:

f.newMethod()
TypeError: not enough arguments; expected 1, got 0 

Understandably though, this works in this case:
f.newMethod(f)


> I bet you get it now.

Sorry, I must be really dense or something... :) I don't see how your 
method can assist me.

-- 
<=====================================>
Bryan Mongeau
Lead Developer, Director
eEvolved Real-Time Technologies Inc.
Website:    http://www.eevolved.com
Public key: http://eevolved.com/bcm.pk
<=====================================>

"The eternal mystery of the world is its comprehensibility."-- Einstein



From ngps at post1.com  Wed Mar 14 10:58:42 2001
From: ngps at post1.com (Ng Pheng Siong)
Date: Wed, 14 Mar 2001 23:58:42 +0800
Subject: Packages, modules and exceptions
In-Reply-To: <20010312163907.A23991@freedom.puma-ag.com>; from stephen_purcell@yahoo.com on Mon, Mar 12, 2001 at 04:39:07PM +0100
References: <98ip5c$397$1@clematis.singnet.com.sg> <20010312163907.A23991@freedom.puma-ag.com>
Message-ID: <20010314235842.B2335@madcap.dyndns.org>

On Mon, Mar 12, 2001 at 04:39:07PM +0100, Steve Purcell wrote:
> I assume you also have a file 'M2Crypto/__init__.py'...

Um, yes; missed that out.


> > For M2Crypto/SSL/Connection.py to get access to SSLError which is defined
> > in M2Crypto/SSL/__init__.py, it (Connection.py) contains the following:
> > 
> >     SSLError = 
> >      __import__('M2Crypto.SSL', globals(), locals(), 'SSLError').SSLError
> 
> from M2Crypto.SSL import SSLError
> raise SSLError, "there's a problem"

Ahh, I wasn't clear enough.

Within M2Crypto/SSL/Connection.py, I have "class Connection:..." Then, 
within M2Crypto/SSL/__init__.py, I have "from Connection import
Connection". This is so that clients can say "s = M2Crypto.SSL.Connection()"
instead of "s = M2Crypto.SSL.Connection.Connection()".

Because of this, I can't just do "from M2Crypto.SSL import SSLError"
within Connection.py. __init__.py's "from Connection import Connection"
creates a circular reference.

What's a nice way to fix this fine mess? ;-)

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




From mal at lemburg.com  Fri Mar 30 18:27:25 2001
From: mal at lemburg.com (M.-A. Lemburg)
Date: Sat, 31 Mar 2001 01:27:25 +0200
Subject: ANN: eGenix.com mxODBC Database Interface Version 2.0.1
References: <3AC5158B.D34E8F27@lemburg.com>
Message-ID: <3AC5165D.B129822A@lemburg.com>

________________________________________________________________________

ANNOUNCING:

               eGenix.com mxODBC Database Package for Python
                            Version 2.0.1

      Full Source Python extensions providing important and useful
                  services for Python programmers.


________________________________________________________________________

WHAT IS IT ?:

The eGenix.com mxODBC Database Package for Python is part of the
eGenix.com mx Extension Series for Python, a collection of
professional quality software tools which enhance Python's usability
in many important areas such as ODBC database connectivity, fast text
processing, date/time processing and web site programming.

________________________________________________________________________

WHAT'S NEW ?

This patch release fixes a few problems which were found by users on
Windows platforms:

* Execution of SQL commands lead to a hanging Python processes for
  some users.

* Handling of date/time values resulted in exceptions on some
  Windows configurations.

Due to popular demand, I am now also providing precompiled Windows
binaries for the older Python 1.5.2 version. Binaries for Python 2.1
will be announced after its final release.

________________________________________________________________________

SPECIAL OFFER

theKompany.com has licensed the eGenix.com mx Commercial Package
(which includes mxODBC) for inclusion in their brand new Qt-based
Python IDE BlackAdder. It allows developing portable GUI-based
database applications which run on Windows and Linux platforms without
any change to the source code.

BlackAdder includes a 1 CPU license for this package at no extra cost,
so you may want to check out their great new product. See

http://www.lemburg.com/files/python/eGenix-mx-Extensions.html#BlackAdder

for details.

________________________________________________________________________

EGENIX.COM MX COMMERCIAL PACKAGE OVERVIEW:

mxODBC - Generic ODBC 2.0-3.5 interface for Python

     mxODBC is an extension package that provides a Python Database
     API compliant interface to ODBC capable database drivers and
     managers.

     In addition to the capabilities provided through the standard DB
     API it also gives access to a rich set of catalog methods which
     allow you to scan the database for tables, procedures,
     etc. Furthermore, it uses the mxDateTime package for date/time
     value interfacing eliminating most of the problems these types
     normally introduce (other in/output formats are available too).

     mxODBC allows you to interface to more than one database from one
     process, making inter-database interfacing very flexible and
     reliable.  

     The source version includes a varity of preconfigured setups for
     many commonly used databases such as MySQL, Oracle, Informix,
     Solid, SAP DB, Sybase ASA and ASE, DBMaker and many more. The
     precompiled versions for Windows and Linux include the interfaces
     to the standard ODBC manager on these platforms to allow for a
     more easily configurable setup.

     More details are available at:

          http://www.lemburg.com/files/python/mxODBC.html

________________________________________________________________________

WHERE CAN I GET IT ?

The download archives and instructions for installing the packages can
be found at:

     http://www.lemburg.com/files/python/

Note that in order to use mxODBC you will also need to install the
eGenix.com mx BASE package which can be downloaded from the same
location.

________________________________________________________________________

WHAT DOES IT COST ?

mxODBC comes with a licenses which allows non-commercial use at no
charge, but costs a moderate fee for commercial use. Please see

http://www.lemburg.com/files/python/eGenix-mx-Extensions.html#mxCOMMERCIAL

for details. The package comes with full source code.

________________________________________________________________________

WHERE CAN I GET SUPPORT ?

Commercial quality support for these packages is available from
eGenix.com Software GmbH. Please see

http://www.lemburg.com/files/python/eGenix-mx-Extensions.html#Support 

for details about the eGenix support offerings.

________________________________________________________________________

REFERENCE:

<P><A HREF="http://www.lemburg.com/files/python/">eGenix.com mxODBC
Package</A> - eGenix.com mxODBC Database Interface 2.0.1 with
distutils support and precompiled binaries for Windows and
Linux. (31-Mar-2001)

________________________________________________________________________

Enjoy,
-- 
Marc-Andre Lemburg
______________________________________________________________________
Company & Consulting:                           http://www.egenix.com/
Python Pages:                           http://www.lemburg.com/python/




From alexander.stirzel at daimlerchrysler.com  Tue Mar 13 06:47:58 2001
From: alexander.stirzel at daimlerchrysler.com (Alexander Stirzel)
Date: Tue, 13 Mar 2001 12:47:58 +0100
Subject: What's wrong with me/Python ?
References: <3AAE0652.D68F253E@daimlerchrysler.com> <98l16c$29cvf$1@fu-berlin.de>
Message-ID: <3AAE08EE.7B6925C6@daimlerchrysler.com>

Matthias Huening wrote:
> 
> >What did I do wrong? And why does Python want a "variable
> >named 'string'" here?
> 
> Because you didn't import the string-module...
> 
> Matthias

*ganzrotesGesicht* (peinlich)

Vielen Dank. (*sichinGrundundBodensch?m*)

-- 
Alexander Stirzel  
E-Mail: alexander.stirzel at informatik.uni-ulm.de


From roy at panix.com  Sat Mar 17 12:02:06 2001
From: roy at panix.com (Roy Smith)
Date: Sat, 17 Mar 2001 12:02:06 -0500
Subject: Looking for web indexer in python
Message-ID: <roy-AD5737.12020617032001@news.panix.com>

The software development group I work in has a small internal web site (a 
few hundred documents) which we don't currently have any way of searching.  
I'd like to put an automated indexer/search system in place.  If possible, 
I'd like to use python, since I think it would be a nifty way to introduce 
the group to the language.  There's no funding for this, so it has to be 
free software.

A cursory look around didn't turn up anything that would fit the bill, but 
I have to assume somebody's already done a free python web indexer.  Any 
suggestions?


From aleaxit at yahoo.com  Fri Mar 30 07:57:46 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 30 Mar 2001 14:57:46 +0200
Subject: how query function args
References: <3AC374BC.F90A2894@abaqus-sn.com> <99vvn1$bmh$1@news.mathworks.com>
Message-ID: <9a1vsj0cg@news1.newsguy.com>

"Joshua Marshall" <jmarshal at mathworks.com> wrote in message
news:99vvn1$bmh$1 at news.mathworks.com...
> david winkler <winkler at abaqus-sn.com> wrote:
> > I need to query the names and values of a function's keyword arguments.
> > I
> > suspect that there is an easy way to do this but cannot find it.  Would
> > someone care to make a suggestion?
>
> func.func_defaults will return a tuple of the default values, but not
> the names of the formals.

func.func_code, the code-object for the function, has most of the
info you want: its co_argcount attribute is the number of arguments
(the ones with defaults are the last len(func.func_defaults) of
them), its co_varnames attribute has the names of local variables
starting with the argument names.

So, the names of arguments which have default values are:
    func.func_code.co_varnames[
        func.func_code.co_argcount - len(func.func_defaults)
            :
        func.func_code.co_argcount
    ]
in the same order as the func.func_defaults.

E.g., if you want name/value tuples (pairs):

def dw(func):
    return zip(
        func.func_code.co_varnames[
            func.func_code.co_argcount - len(func.func_defaults)
                :
            func.func_code.co_argcount
        ], func.func_defaults)

It's of course easy to make this into a dictionary or whatever.


Alex






From tuttledon at hotmail.com  Sat Mar  3 22:02:35 2001
From: tuttledon at hotmail.com (Don Tuttle)
Date: Sun, 04 Mar 2001 03:02:35 GMT
Subject: Beginer
References: <3aa18bf8$0$65279$38c90ffe@news1.twtelecom.net>
Message-ID: <ffio6.29000$Z8.5160738@typhoon.southeast.rr.com>

"mark"
> can anyone tell me where i can find a real beginer tutorial for python?
The
> one that comes with the program is still to complicated.thanks>

http://www.crosswinds.net/~agauld/




From loriendNOSPAM at bigpond.com  Sat Mar 17 01:38:22 2001
From: loriendNOSPAM at bigpond.com (Lorien Dunn)
Date: Sat, 17 Mar 2001 17:38:22 +1100
Subject: Meta-class inheritance problem (Python2.0 bug?)
References: <6Kzs6.35062$v5.73895@newsfeeds.bigpond.com>
Message-ID: <5nDs6.35198$v5.74638@newsfeeds.bigpond.com>

I've located the point in the python 2.0 code where this occurs, so my 
question is now: 

        What needs to be changed to make this wotk with meta-classes as well?:

ceval.c, line 1816 (reformatted for clarity):

/*              BEGIN CODE SAMPLE                       */ 

else 
{
        /* Unbound methods must be called with an instance of the class (or a                                   
        derived class) as first argument */
        if (na > 0 && (self = stack_pointer[-n]) != NULL 
                && PyInstance_Check(self)
                && PyClass_IsSubclass((PyObject *) (((PyInstanceObject*)                                                                                                                                        
                self)->in_class), class))
                         /* Handy-dandy */ ;
        else 
        {
                PyErr_SetString(PyExc_TypeError,
                        "unbound method must be called with class
                         instance 1st argument");
                x = NULL;
                break;
        }
}

/*              END CODE SAMPLE         */


Lorien

                   



From flognat at flognat.myip.org  Thu Mar 15 18:35:47 2001
From: flognat at flognat.myip.org (Andrew Markebo)
Date: 16 Mar 2001 00:35:47 +0100
Subject: Should I close after popen??
References: <XFMail.20010315142704.shaleh@valinux.com> <mailman.984697040.2458.python-list@python.org>
Message-ID: <m3lmq6ppv0.fsf@master.athome>

/ "Ken Seehof" <kens at sightreader.com> wrote:
| [...]
| In other words, you are better off closing files explicitly.
| 
| In the case of popen, it seems to me like it wouldn't matter if you
| leave a few open streams lying around, but it just seems sloppy,
| like not washing your dishes after dinner.

I agree, but I tried with close, but I got the following problems:

Traceback (innermost last):
  File "<stdin>", line 1, in ?
  File "/usr/tmp/python-18803PG", line 314, in ?
    mymain()
  File "/usr/tmp/python-18803PG", line 304, in mymain
    keepALive()
  File "/usr/tmp/python-18803PG", line 259, in keepALive
    if PingAway():
  File "/usr/tmp/python-18803PG", line 230, in PingAway
    fd1.close()
IOError: [Errno 10] No child processes

The code is:


    fd1=os.popen("ping -q -c 3 %s 2> /dev/null" % name1)
    fd2=os.popen("ping -q -c 3 %s 2> /dev/null" % name2)
    pid1, res1=os.wait(fd1)
    out1=fd1.read()
    pid2, res2=os.wait(fd2)
    out2=fd2.read()
    fd1.close()
    fd2.close()

        /Andy


From kens at sightreader.com  Wed Mar 14 17:24:19 2001
From: kens at sightreader.com (Ken Seehof)
Date: Wed, 14 Mar 2001 16:24:19 -0600
Subject: compile / exec
Message-ID: <003d01c0acd5$83b2ab70$04090a0a@upcast.com>

The first line in the string is indented, which is a syntax error.  The
first line of the string has to be non-indented.

"eric wilson" wrote:
> hi -
> i am new to python and i'm trying to use the compile / exec feature for
> speeding up a loop.  i'm using python v2.0 with pythonwin.  i don't
> understand why this doesn't work....
> thanks in advance for the help.
> file://eric
> 
> the code snippet looks like this:
>     exec_code = compile("""
>     for eachLine in map(strip, allLines):
>         mySequence = split('\s\s+|\t', eachLine)
>         foundFields = str(gm2local(DateTimeFromString(mySequence[0])))[:-3]
> + '\t' + mySequence[1]+ '\t' + mySequence[4]+ '\t' + mySequence[5]+ '\t' +
> mySequence[9]+ '\t' + mySequence[16] + '\n'
>         fw.write(foundFields)
>         count = count + 1
>     """, '', 'exec')
>     exec exec_code
> 
> the error message that i get when this piece of code runs is:
> Traceback (most recent call last):
>   File "C:\Python20\Pythonwin\pywin\framework\scriptutils.py", line 301, in
> RunScript
>     exec codeObject in __main__.__dict__
>   File "E:\W2\pytest\logman.py", line 114, in ?
>     checkArgs()
>   File "E:\W2\pytest\logman.py", line 110, in checkArgs
>     parmFind(sys.argv[1], sys.argv[2])
>   File "E:\W2\pytest\logman.py", line 55, in parmFindRAM
>     exec_code = compile("""
>   File "<string>", line 2
>      for eachLine in map(strip, allLines):
>      ^
>  SyntaxError: invalid syntax

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20010314/0635f39e/attachment.html>

From erwin at andreasen.com  Wed Mar 28 16:01:48 2001
From: erwin at andreasen.com (Erwin S. Andreasen)
Date: Wed, 28 Mar 2001 21:01:48 +0000 (UTC)
Subject: Changing every other instance of <B> in a file
References: <mailman.985726476.22406.python-list@python.org> <slrn9c22em.6tv.qrczak@qrnik.zagroda>
Message-ID: <99tjfr$16et$1@news.cybercity.dk>

On 27 Mar 2001 21:44:56 GMT, Marcin 'Qrczak' Kowalczyk wrote:

>Here is an obfuscated version:
>
>import List; main = interact $ concat . snd . mapAccumL f 1 . tails
>f x ('<':'B':'>':_) = (3-x, take x "</"); f x s = (x, take 1 s)

You call that obfuscation? :) You even have superfluous whitespace! I'll give
you obfuscation:

/usr/bin/p*rl -pe's`<b\>`$^X++%2?q{</b>}:$&`gei;' datafile

I managed to use only ^`'{}$%&<>?/;+:*\ -- still lot of punctuation left :)

INTERCAL, anyone?


-- 
=======================================================================
<erwin at andreasen.com>           Herlev, Denmark       Software Designer
<URL:http://www.andreasen.org/>       <*>         LASAT^WEicon Networks
=======================================================================


From aahz at panix.com  Sat Mar 17 11:41:58 2001
From: aahz at panix.com (Aahz Maruch)
Date: 17 Mar 2001 08:41:58 -0800
Subject: This math scares me
References: <20010312170356.B3513@better.net> <98qnq0$qc2$1@panix3.panix.com> <98umb1$f2i$1@panix3.panix.com> <slrn9b7as9.10s.grante@tuxtop.visi.com>
Message-ID: <99044m$cfi$1@panix2.panix.com>

In article <slrn9b7as9.10s.grante at tuxtop.visi.com>,
Grant Edwards <grante at visi.com> wrote:
>
>It looks like I had them mixed up with the Babylonians.  I
>swear, the denizens of c.l.p have the widest base of knowege of
>any newsgroup I've read.

You obviously haven't read rec.arts.sf.fandom    ;-)
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

Three sins: BJ, B&J, B&J


From ryan at NOSPAMda-men.comNOSPAM  Tue Mar  6 09:42:06 2001
From: ryan at NOSPAMda-men.comNOSPAM (Tonnage)
Date: Tue, 06 Mar 2001 14:42:06 GMT
Subject: iis and python
References: <F5622749FED6DA5D.AF4594816A4865C9.16CD98F1C1FAC3F6@lp.airnews.net> <Rs0p6.183447$df5.5139760@news1.crdva1.bc.home.com>
Message-ID: <2H6p6.372$oz.53556@newsread1.prod.itd.earthlink.net>

LOL...this is not a flame, I am just chuckling at the same thing.  I have
played that game, and then walked down the hall and beat the newbie with
config scripts and libs.
USE 'EM
LIVE 'EM
LOVE 'EM

Its far better to have 30 scripts, with only a few operations or commands
per script, than to have one huge one with all of them.

The only people that dont go with with, in the PHP/Perl case, are those that
have never had to either optimize, or debug someone elses huge script.

Ryan
"Wayne Izatt" <wayne.izatt at _nospam_myself.com> wrote in message
news:Rs0p6.183447$df5.5139760 at news1.crdva1.bc.home.com...
> > In order to use python it would have to entail cgi or ASP. Is this
> correct.
> > Or is there an alternative. I dont know if Zope is the right thing. I
hate
> > DTML.
>
> Zope is very powerful, and if you want to you can write almost all your
code
> in Python. DTML is there for presentation, not for logic, and should only
be
> used for simple processing.
>
> > Is using python in ASP a pain in the but? Am I going to have to learn a
> > whole technology just to implelment python well into it?
>
> No, its fairly straighforward. However, before you go the ASP route, take
> the time to examine Webware. Webware gives you ASP with Python, as well as
> Python Server Pages and servlets - very powerful, easy to use, and pure
> Python.
>
> > I hear troubleshooting ASP pages is not fun.
>
> Correct.
>
> > The only time I have used iis is with php and it was a total snap.
> Painless.
>
> PHP is fine, but I've had the pleasure of debugging someone else's massive
> 'one-script-wonder', and I remembered what my misgivings about Perl-based
> languages were. Don't flame me, PHPer's, I still like the language!
>
> cheers
>
>
>
>




From larskl at klassekampen.no  Tue Mar 27 11:03:45 2001
From: larskl at klassekampen.no (Lars Klęboe)
Date: Tue, 27 Mar 2001 18:03:45 +0200
Subject: Changing every other instance of <B> in a file
References: <8e41ctkk4l38jmj34070j15iotgv4is2cg@4ax.com> <99qb39$2m0$1@animus.fel.iae.nl>
Message-ID: <scd1ctccls7bnjk1o7gngc04sfp5k4gunm@4ax.com>

On 27 Mar 2001 17:20:09 +0200, Carel Fellinger <cfelling at iae.nl> wrote:
>Lars Kl?boe <larskl at klassekampen.no> wrote:
>...
>> 	Blablabla <B> talktalk <B> blabla blabla balbalblabla
>
>> The resulting file.html (html)
>
>> 	Blablabla <B> talktalk </B> blabla blabla balbalblabla
>
>> As you can tell, every other instance of <B> is to be changed into </B>.
>
>Let's hope those B-tags aren't nested, then the following might work:

No they aren't nested.

Why haven't I thought of dictionaries before ? Ever ?
In any programming context ? At all ? </ashamed> ;)

Thanks !

-- 
Lars Kl?boe




From bedge at troikanetworks.com  Wed Mar 21 10:32:18 2001
From: bedge at troikanetworks.com (Bruce Edge)
Date: Wed, 21 Mar 2001 07:32:18 -0800
Subject: Saving history in interactive Python
References: <mailman.985113411.25785.python-list@python.org>
Message-ID: <3AB8C982.50901FE5@troikanetworks.com>

set /home/<logname>/.python, then here's what I stuck in my ~/.python
file


histfile="/home/<logname>/.python
import rlcompleter
import readline
readline.parse_and_bind("M-?: complete")
readline.parse_and_bind("tab: menu-complete")
import os
import sys
readline.read_history_file( histfile )
def savehist():
	global histfile
	readline.write_history_file( histfile )
import atexit
atexit.register( savehist )	


This gets you command history saving, and readline command line editing.

-Bruce.


Timothy Grant wrote:
> 
> Is there a way to save the history list in the Python
> interpreter.
> 
> --
> Stand Fast,
>     tjg.
> 
> Timothy Grant                         tjg at exceptionalminds.com
> Red Hat Certified Engineer            www.exceptionalminds.com
> Avalon Technology Group, Inc.         <><       (503) 246-3630
> >>>>>>>>>>>>>Linux, because rebooting is *NOT* normal<<<<<<<<<
> >>>>This machine was last rebooted:  62 days 22:52 hours ago<<


From bencr at bigpond.com  Tue Mar 27 06:13:28 2001
From: bencr at bigpond.com (Ben)
Date: Tue, 27 Mar 2001 21:13:28 +1000
Subject: Executing external programs
Message-ID: <vhZv6.4205$45.21324@newsfeeds.bigpond.com>

Hi,

this appears to be a simple problem ... but i can not find any documentation
on this ... could anyone point me to a rescource that would have an answer
to this solution ...

All I want to do is

 run program1.exe -args

 ... when program1.exe is finished

run program2.exe -args ...

run program3.exe -args

etc etc

 ie the next process waits for the preceding processes exit code before
starting ... if I can avoid the overhead of having the command shell open
everytime I run an exe that would be good.

Thanks






From tim.one at home.com  Sat Mar  3 14:02:24 2001
From: tim.one at home.com (Tim Peters)
Date: Sat, 3 Mar 2001 14:02:24 -0500
Subject: the demise of 'from foo import * and its implications?
In-Reply-To: <z8dAGJAyqLo6EwlX@jessikat.fsnet.co.uk>
Message-ID: <LNBBLJKPBEHFEDALKOLCGELAJDAA.tim.one@home.com>

[Robin Becker]
> my preference would be that statements legal in one context should be
> legal in another where they make sense. I cannot see how using import *
> or eval etc in nested scopes is somehow obviously bad or different from
> the same usage in non-nested scopes. The fact that the prohibition comes
> from making things easier for the compiler adds to my confusion.

The prohibition (against import* except at module scope) is there for "make
sense" reasons.  Here's a simple function, where nested scopes don't even
come into play:

def f():
    from mystery import *
    g()

def g():
    print "hmm"

f()

What does that do?  Forget the compiler!  What makes *sense*?

There simply isn't an answer that isn't surprising (although there *could*
have been!).  The Ref Man has always said:

    Whether a name is local or global in a code block is determined
    by static inspection of the source text for the code block: in
    the absence of global statements, a name that is bound anywhere
    in the code block is local in the entire code block; all other
    names are considered global.

By that lovely rule, "g" is a global name within "f" (because static
inspection shows that g is not bound within the body of f).  However, because
of the way Python was first implemented, it just so happens that *if* the
mystery module contains its own global g, *that's* what "g" inside "f" refers
to instead.  The import* magically creates a local "g" inside "f", shadowing
the global g anyone reading the quoted part of the Ref Man expects (and
assuming they ignored the part of the Ref Man saying import* has undefined
behavior here).  And that's very surprising if you think-- as Guido intended,
and as the Ref Man says --that Python is statically scoped.

OTOH, since the first implementation did just happen to have this other
behavior, Guido clung to it over the years for backward compatibility,
despite that it's documented as being undefined, and despite that the
compiler internals have gotten considerably warped *trying* to preserve this
accidental behavior.  So taking it away now is surprising if you think-- as
some protest ever so loudly --that accidents should be preserved in loving
detail forever more.

It's not "a feature" of Python that it exhibits a bizarre form of dynamic
scoping in some cases.  It wasn't intended.  Whenever that pops up, it's a
flaw in the language design, or (in the case of import* at other than module
scope) a (IMO, with hindsight) misguided attempt to preserve accidents of the
initial implementation.

> I have no particular objection to being able to access variables in
> intermediate scopes.
>
> If we really want to improve visibility we could try to make
>
> >>> def bingo(a):
> ...     if a<=1: return 0
> ...     return bingo(a-1)*a
> ...
>
> work naively, but I'm fairly sure this will be rejected as unpythonic or
> somesuch.

Sorry, I can't make out what you're trying to say there.  It's a vanilla
recursive function, and it works exactly the way I expect it to work (which
hasn't changed since the first release, and almost certainly never will).
What about it doesn't work the way *you* expect?

Note that if there's a "from mystery import *" at *module* scope, and mystery
happens to define a global of its own named "bingo", then that latter is the
bingo called inside the body.  That's an example of "flaw in the language
design" I mentionoed above, and I don't give a rat's ass about preserving
*that* accident of the implementation either.

language-behaviors-are-clarified-by-relating-them-to-animal-body-
    parts-ly y'rs  - tim




From f8dy at yahoo.com  Wed Mar 28 13:18:00 2001
From: f8dy at yahoo.com (Mark Pilgrim)
Date: Wed, 28 Mar 2001 13:18:00 -0500
Subject: list vs tuple
References: <hGlw6.782$p5.3210@news1.rivrw1.nsw.optushome.com.au> <99stcb$483$1@panix6.panix.com>
Message-ID: <99t9om$2o2q3$1@ID-77331.news.dfncis.de>

"Aahz Maruch" <aahz at panix.com> wrote in message
news:99stcb$483$1 at panix6.panix.com...
> In article <hGlw6.782$p5.3210 at news1.rivrw1.nsw.optushome.com.au>,
> deadmeat <root@[127.0.0.1]> wrote:
> >
> >Whats the practical difference between a list[] and a tuple() ?
>
> As someone said, tuples are immutable.  This leads to useful
> functionality, such as being able to easily use tuples as keys in
> dictionaries (ditto strings).

It was my understanding that only tuples containing immutable objects could
be used as dictionary keys.  For instance, (1,2,3) can be, and ("a","b","c")
can be, but ([1,2,3], ["a","b","c"]) can not, because the underlying objects
could mutate and cause problems.  (Wow, that sounds like a bad sci-fi
movie.)  Is this true?

-M
You're smart; why haven't you learned Python yet?
http://diveintopython.org/





From tjg at exceptionalminds.com  Tue Mar  6 17:30:10 2001
From: tjg at exceptionalminds.com (Timothy Grant)
Date: Tue, 6 Mar 2001 14:30:10 -0800
Subject: Processing XML
In-Reply-To: <hf166dh2.fsf@propylon.com>; from martin.skott@propylon.com on Tue, Mar 06, 2001 at 04:58:49PM +0000
References: <mailman.983853727.31962.python-list@python.org> <hf166dh2.fsf@propylon.com>
Message-ID: <20010306143010.L3579@trufflehunter.avalongroup.net>

Thanks for the helpful post. String2xTree was just what I
needed. The good word for Pyxie was also a big help!

On Tue, Mar 06, 2001 at 04:58:49PM +0000, Martin Skott wrote:
> Timothy Grant <tjg at exceptionalminds.com> writes:
> 
> > If Pyxie is a good choice, how do I add nodes to a document
> > stored as an xTree?
> 
> Use the String2xTree function to create a mini tree of your sting and
> then use the paste functions to insert it into your main tree.

-- 
Stand Fast,
    tjg.

Timothy Grant                         tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.         <><       (503) 246-3630
>>>>>>>>>>>>>Linux, because rebooting is *NOT* normal<<<<<<<<<
>>>>This machine was last rebooted:  49 days  2:43 hours ago<<



From scarblac at pino.selwerd.nl  Tue Mar 27 03:02:17 2001
From: scarblac at pino.selwerd.nl (Remco Gerlich)
Date: 27 Mar 2001 08:02:17 GMT
Subject: fixing map(None, [1,2,3]) ?
References: <873dc040ny.fsf@mathdogs.com>
Message-ID: <slrn9c0i2o.nup.scarblac@pino.selwerd.nl>

Mike Coleman <mkc at mathdogs.com> wrote in comp.lang.python:
> Shouldn't "map(None, [1,2,3])" return "[(1,), (2,), (3,)]"?
> 
> Is there any hope that this could be fixed?

No, since zip() already does that in 2.0 and higher, changing map would
break lots of code, and it's not better anyway...

-- 
Remco Gerlich


From gerhard.nospam at bigfoot.de  Wed Mar  7 10:51:19 2001
From: gerhard.nospam at bigfoot.de (Gerhard =?iso-8859-1?Q?H=E4ring?=)
Date: Wed, 07 Mar 2001 16:51:19 +0100
Subject: Python script for deMSWordification?
References: <985h02$1pbm$1@swen.emba.uvm.edu>
Message-ID: <3AA658F7.9127B12E@bigfoot.de>

kelvin.chu at uvm.edu wrote:
> 
> Dear fellow Pythoneers;
> 
> Numerous colleagues and administration types send memos, colloqium
> announcements and other things around in Micro$oft word format.  Has anyone
> written a filter in Python that can convert these things into normal text?
> 
> If not, is there a basic set of rules for decoding .doc format files
> anywhere that I can look at?

Hmm, not Python, but wv (http://www.wvWare.com/) does that job. It
converts M$ Word (6, 95, 97, 2000) to various formats.

It's a (C, C++?) library, so you should be able to write a Python
binding, if you need to access the internals of a Word document. For
conversion only, calling the command line program will suffice, of
course.

Gerhard
-- 
Sorry for the fake email, please use the real one below to reply.
contact: g e r h a r d @ b i g f o o t . d e
web:     http://highqualdev.com


From loewis at informatik.hu-berlin.de  Sat Mar 24 07:19:20 2001
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 24 Mar 2001 13:19:20 +0100
Subject: [Python-Dev] Anyone using weakrefs?
References: <15035.46175.599654.851399@localhost.localdomain> <mailman.985433253.2726.python-list@python.org>
Message-ID: <j4ofurmkaf.fsf@informatik.hu-berlin.de>

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

> am I the only one who gets a bit nervous when the guy who
> designed and implemented a feature has a hard time finding a
> non-contrived use case? ;-)

Seriously (and not knowing the exact meaning of contrived), a good
application is the parent link in a DOM tree. The problem is that this
is 
a) probably a too complex example for the documentation, as just
   making parent weak is not enough - ownerDoc also needs to be weak,
   and,
b) nobody has changed any DOM implementation to make use weak
   references; Fred was asking for a real application.

Regards,
Martin

P.S. I've considered using weakrefs in minidom, but put that plan
aside since that would break portability with 1.5.


From jkraska1 at san.rr.com  Sat Mar 24 23:57:41 2001
From: jkraska1 at san.rr.com (Courageous)
Date: Sun, 25 Mar 2001 04:57:41 GMT
Subject: time to invest ???
References: <3abd3a17.7103358@news.vif.com> <Yrcv6.2751$45.12829@newsfeeds.bigpond.com> <vdqqbtk946qh6p7a0tfpf2kng2o0jkk7mm@4ax.com> <slrn9br4j8.tf.grante@tuxtop.visi.com>
Message-ID: <n0vqbtgbe814d9kmuegvdpvqbrrgofu3g1@4ax.com>

>I would think most people familiar with a couple other
>languages could master the commonly used language features in a
>day or two.

Err, yeah. I tought someone the other day, and they had all they needed
to get going on their whiteboard. It all fit. :)

C//



From ewilson at nospam.jamdat.com  Wed Mar 14 20:28:42 2001
From: ewilson at nospam.jamdat.com (eric wilson)
Date: Thu, 15 Mar 2001 01:28:42 GMT
Subject: compile / exec
References: <mailman.984615559.29257.python-list@python.org>
Message-ID: <eVUr6.14842$R_6.1538043@newsread2.prod.itd.earthlink.net>

thanks ken - that's got it!  :-)


"Ken Seehof" <kens at sightreader.com> wrote in message
news:mailman.984615559.29257.python-list at python.org...
The first line in the string is indented, which is a syntax error.  The
first line of the string has to be non-indented.

"eric wilson" wrote:
> hi -
> i am new to python and i'm trying to use the compile / exec feature for
> speeding up a loop.  i'm using python v2.0 with pythonwin.  i don't
> understand why this doesn't work....
> thanks in advance for the help.
> //eric
>
> the code snippet looks like this:
>     exec_code = compile("""
>     for eachLine in map(strip, allLines):
>         mySequence = split('\s\s+|\t', eachLine)
>         foundFields =
str(gm2local(DateTimeFromString(mySequence[0])))[:-3]
> + '\t' + mySequence[1]+ '\t' + mySequence[4]+ '\t' + mySequence[5]+ '\t' +
> mySequence[9]+ '\t' + mySequence[16] + '\n'
>         fw.write(foundFields)
>         count = count + 1
>     """, '', 'exec')
>     exec exec_code
>
> the error message that i get when this piece of code runs is:
> Traceback (most recent call last):
>   File "C:\Python20\Pythonwin\pywin\framework\scriptutils.py", line 301,
in
> RunScript
>     exec codeObject in __main__.__dict__
>   File "E:\W2\pytest\logman.py", line 114, in ?
>     checkArgs()
>   File "E:\W2\pytest\logman.py", line 110, in checkArgs
>     parmFind(sys.argv[1], sys.argv[2])
>   File "E:\W2\pytest\logman.py", line 55, in parmFindRAM
>     exec_code = compile("""
>   File "<string>", line 2
>      for eachLine in map(strip, allLines):
>      ^
>  SyntaxError: invalid syntax




From sholden at holdenweb.com  Thu Mar 15 00:27:58 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 15 Mar 2001 00:27:58 -0500
Subject: PythonWorks: opinions sought
Message-ID: <YsYr6.83498$ML1.3870607@e420r-atl2.usenetserver.com>

Before I part with almost $400 of hard-earned cash, I would be interested to
hear user experiences of PythonWorks. If you don't want /F to see what you
have to say an email would be fine (;-), but otherwise the newsgroup as a
whole might be interested: I haven't seen much discussion here of what seems
to be quite a worthwhile product.  Presumably due to Fredrik's keen sense of
public service, he never mentions it in his fine postings.

regards
 Steve

PS: "XXX is better" responses will *not* be welcome: I have other IDE
products already, and am currently solely interested in PythonWorks rather
than in IDE comparisons.




From stevewilliams at wwc.com  Sun Mar 25 23:36:47 2001
From: stevewilliams at wwc.com (Steve Williams)
Date: Mon, 26 Mar 2001 04:36:47 GMT
Subject: Python classes on east coast?
References: <bobhicks-2503011622160001@va-frontroyal1a-202.wch.adelphia.net>
Message-ID: <3ABEC78A.99C63F07@wwc.com>

Robert Hicks wrote:

> Is there or are there any plans for Python classes on the east coast

[snip]

As far as I know Python classes have no geographical attributes.

Perhaps we need a PEP for 'import from __longitude__' to distinguish the
GMT Python from the PST Python.






From dale at out-think.NOSPAMco.uk  Thu Mar 22 14:02:22 2001
From: dale at out-think.NOSPAMco.uk (Dale Strickland-Clark)
Date: Thu, 22 Mar 2001 19:02:22 +0000
Subject: Automatic install of ActivePython
Message-ID: <elikbtker40ll94076q82tugve0jaqh264@4ax.com>

I've had a pretty good hunt around for an answer to this and I'm somewhat surprised to have drawn a
blank.

How do I do an automated or quiet install of ActivePython 2.0?

It would be nice to be able to set the target directory but I can live without that.

Thanks.
--
Dale Strickland-Clark
Out-Think Ltd
Business Technology Consultants


From tchur at optushome.com.au  Sat Mar 10 16:24:11 2001
From: tchur at optushome.com.au (Tim Churches)
Date: Sun, 11 Mar 2001 08:24:11 +1100
Subject: Urgent Question about Python licensing
References: <98cpfe$5oj$0@216.39.170.247> <20010310110228.L404@xs4all.nl>
Message-ID: <3AAA9B7B.34AE0336@optushome.com.au>

Thomas Wouters wrote:
> 
> On Sat, Mar 10, 2001 at 08:43:26AM +0000, Dave LeBlanc wrote:
> > Hi;
> 
> > While reading up on the recent Python conference on O'Reilly's site
> > (http://python.oreilly.com/news/pythonday1_0301.html - actually on the
> > 'day 2' page), I cam across the following:
> 
> [ The python licence hassles.]
> 
> > My questions are these:
> > 1. What prevents Python and GPL code from being distributed together?
> > If I write some Python code that drives a GPL'd app (but  which does
> > not depend on that app), what prevents me from offering both my python
> > code and the GPL'd app on the same medium (disk/CD/web)?
> 
> 'derives' is a bit of a general and weird term. If your code does not depend
> on a GPL library or binary, you can just distribute it. You can also
> distribute both a GPL application and a Python-licenced application on the
> same medium -- there is no restriction to that, as long as you make clear
> what licence applies to what part. What you cannot do is link a
> Python-licenced piece of software to a GPL-licenced part, *and* distribute
> that resulting binary. Not even using dynamic linking, for some reason.
> (Don't ask me, I'm not a lawyer.) I am not sure if you can attempt to import
> a module, which might be a C module that is covered, possibly indirectly, by
> the GPL. My gut says that that should be just fine, but my gut usually tries
> to empty itself when it comes to lawyerese.

You had me worried there about the implications of all of the above for
the (really exciting, from my point of view) work being done by Duncan
Temple Lang under the aegis of the OmegaHat project (see
http://www.omegahat.org) to make the R statistical environment (see
http://www.r-project.org), which can be built as a shared library,
callable from within Python, and v-v (see
http://www.omegahat.org/RSPython/index.html)? R is GPLed and only GPLed. 

However, the eleventh section (i.e. section number 10 - clearly the GPL
and Python are fellow-travellers) of the GPL (see
http://www.gnu.org/copyleft/gpl.html) states that if you want to
incorporate GPLed software in non-GPLed software (and presumably v-v),
one merely has to ask permission in writing from the authors (or should
that be copyright holders?) of the GPLed software. So there would not
appear to be any absolute impediment to linking Python (as currently
licensed) with GPLed software - you just have to ask permission first. 

Tim Churches



From nospam at collinet.freeserve.co.uk  Sat Mar 24 20:04:06 2001
From: nospam at collinet.freeserve.co.uk (Tony Collins)
Date: Sun, 25 Mar 2001 02:04:06 +0100
Subject: Python script to create HTML gallery, thumbnails
References: <3ABCAC94.B095CB9A@ozemail.com.au>
Message-ID: <99jfsd$l5u$1@news6.svr.pol.co.uk>

You could try albumatic - creates full web pages complete with thumbnails -
link on my site.

--
Tony Collins
Please change nospam in address to 'tony'
nospam at collinet.freeserve.co.uk

Digital Photo website :
http://www.digifotoinfo.co.uk

"ccang" <calcium at ozemail.com.au> wrote in message
news:3ABCAC94.B095CB9A at ozemail.com.au...
> Hi there all,
> I recently got a digital camera (Kodak DC4800) and wanted to
> put all the crappy photos I shoot on the web.
>
> I looked for software to create thumbnails and corresponding
> HTML files but couldnt find any so I wrote my own using Python
> and the PIL.
>
> It creates a consolidated thumbnail and the corresponding HTML files.
>
> No GUI yet so u'll have to edit the script to customize
> your settings eg page headers, comments.
>
> # Does only JPGs for now.
> # Usage : python mkgallery.py [ filenames ]
> #   Will produce 1 HTML file for the thumbnail index.
> #                1 HTML file per JPG image.
> #                1 consolidated thumbnail JPG.
> #
> # There will be GUI for this pretty soon.
> #
> # The HTML file for the thumbnail is generated in the same
> # directory as where the script is run from.
> # However, it assumes that it will reside
> # in a subdirectory above where the other files (HTML, JPGS) reside.
> # ie u'll have to move it.
> #
> # eg
> #   myGallery/
> #   myGallery/topic.html
> #   myGallery/topic/topic_index.jpg
> #   myGallery/topic/zpic1.jpg
> #   myGallery/topic/zpic1.html
> #       ...
> #       ...
> #   myGallery/topic/zpicN.jpg
> #   myGallery/topic/zpicN.html
> #
>
> http://www.ozemail.com.au/~calcium/scripts.html
>
> Sample output
> http://www.ozemail.com.au/~calcium/rogaine.html
>
> Feedback welcome,
> Chai
> --
> | mailto:calcium at ozemail.com.au   | +61 401 688 408
> |  http://www.ozemail.com.au/~calcium
> |  I dont use the voice of Bart when I'm making love
> |  to my husband, but Marge's voice turns him on a
> |  little - Nancy Cartwright.
>




From gmcm at hypernet.com  Sat Mar 10 14:19:00 2001
From: gmcm at hypernet.com (Gordon McMillan)
Date: Sat, 10 Mar 2001 14:19:00 -0500
Subject: ANN: Linux version of Installer for Python 2
Message-ID: <mailman.984404340.32285.clpa-moderators@python.org>

The Linux version of Installer for Python 2 is here:
 http://www.mcmillan-inc.com/installer_dnld.html

Background:
 The Installer is a sort of compilerless Freeze - a way 
of distributing Python apps with all the required support built 
in. The Linux version collects all dependencies and packages 
the pure Python into an archive attached to the executable.
Just tar, gzip and ship.


Old-Python-style license.

Contact: gmcm at hypernet.com

<P><A 
HREF="http://www.mcmillan-inc.com/installer_dnld.html">
Linux Installer release 3j</A> A compiler-less way 
of distributing Python 2.0 apps on Windows. (10-Mar-01)



- Gordon




From johann at physics.berkeley.edu  Tue Mar 27 16:43:30 2001
From: johann at physics.berkeley.edu (Johann Hibschman)
Date: 27 Mar 2001 13:43:30 -0800
Subject: python, OS X, and fcntl
References: <mailman.985641394.29795.python-list@python.org> <bobhicks-2703010954550001@va-frontroyal1a-202.wch.adelphia.net>
Message-ID: <mtk85aga65.fsf@astron.berkeley.edu>

Robert Hicks writes:

> http://www.zope.org/Members/jshell/ZopeOnOSX
> This goes into exactly what you are asking about...

Aha.  Thanks.  I saw that page, but I think the last time I looked at
it, it hadn't been updated to OS X Final yet.

--Johann

-- 
Johann Hibschman                           johann at physics.berkeley.edu


From chrisw at nipltd.com  Wed Mar 14 06:12:24 2001
From: chrisw at nipltd.com (Chris Withers)
Date: Wed, 14 Mar 2001 11:12:24 +0000
Subject: zipfile: how to find out if a zipinfo object is a directory or file
References: <3AAF3BCF.580F3ECC@t-online.de>
Message-ID: <3AAF5218.9EC5DA79@nipltd.com>

Stephane wrote:
> 
> Hi,
> 
> does somebody know how to find out in a zipinfo object represents a file
> or directory.

I think zipinfo objects _always_ represent files. The name in the zipinfo object
includes that path where that file should be expanded to.

cheers,

Chris



From porter at et.byu.edu  Wed Mar 21 13:34:19 2001
From: porter at et.byu.edu (C. Porter Bassett)
Date: Wed, 21 Mar 2001 11:34:19 -0700 (MST)
Subject: IDE with vi key bindings
In-Reply-To: <M_5q6.45279$1D5.2048753@e420r-atl1.usenetserver.com>
References: <mailman.983615170.23313.python-list@python.org> <tN5o6.2746$eH3.14913@NewsReader> <Koco6.7518$1D5.244602@e420r-atl1.usenetserver.com> <mailman.983997252.5628.python-list@python.org> <meAp6.29961$ML1.1514835@e420r-atl2.usenetserver.com> <mailman.984014353.5058.python-list@python.org> <BcCp6.37817$1D5.1582482@e420r-atl1.usenetserver.com> <98amlj01sp6@news2.newsguy.com> <M_5q6.45279$1D5.2048753@e420r-atl1.usenetserver.com>
Message-ID: <Pine.GHP.4.21.0103211130430.1134-100000@jj.et.byu.edu>

Is there any python IDE (windows) that either can integrate with vim or
has vi key bindings?  I know that there are some (Dave) who can't
understand why I would want this, but I just can't seem to break my vi
addiction.  I currently do all my python coding in vim, but I would like
to try an IDE without leaving behind my precious vi.

--------------------------------------------------------------------------
"Pretend like this is a really witty saying." - Anonymous
--------------------------------------------------------------------------





From loewis at informatik.hu-berlin.de  Wed Mar 14 06:12:27 2001
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 14 Mar 2001 12:12:27 +0100
Subject: Process details
References: <mailman.984439220.26863.python-list@python.org>
Message-ID: <j4y9u8li3o.fsf@informatik.hu-berlin.de>

Timothy Grant <tjg at exceptionalminds.com> writes:

> I need to create a list of processes and process information. I
> have created a class that contains much of the information in
> /proc/'pid'/status.
> 
> It works quite well, however, it takes about four seconds to
> populate a list of 90 procs. This seems a bit long. Is there a
> faster way to get ahold of the information than reading /proc?

It depends on you exactly you read them, and what information exactly
you need. os.listdir("/proc") never took more than 10ms on my system.

Regards,
Martin


From tim.one at home.com  Sat Mar  3 15:21:19 2001
From: tim.one at home.com (Tim Peters)
Date: Sat, 3 Mar 2001 15:21:19 -0500
Subject: the demise of 'from foo import * and its implications?
In-Reply-To: <AQyhzIAfSUo6EwQ9@jessikat.fsnet.co.uk>
Message-ID: <LNBBLJKPBEHFEDALKOLCEELLJDAA.tim.one@home.com>

[Robin Becker]
>>> If we really want to improve visibility we could try to make
>>>
>>> >>> def bingo(a):
>>> ...     if a<=1: return 0
>>> ...     return bingo(a-1)*a
>>> ...
>>>
>>> work naively, but I'm fairly sure this will be rejected as unpythonic or
>>> somesuch.

[Tim Hochberg]
>>?????
>>
>> In what sense doesn't this work now?
>> ...
>>
>>What doesn't work is the following:
>>
>>def spam(n):
>>   def fact(a):
>>      if a <= 1: return 1
>>      else: return fact(a-1)*a
>>   return fact(n) + 42

[Robin Becker]
> you're right of course, this is what I meant. I haven't played with the
> new statically scoped stuff, but I probably wrongly assumed that bingo
> would remain invisible inside bingo.

Ah!  Yes, that assumption is wrong.  Making nested recursive functions "work"
as expected was one of the (few) strong *practical* arguments in favor of
adding nested_scopes.

> ...
> I hadn't really appreciated before now that python wishes to enforce a
> fixed scope for references (for optimisation purposes)

See other reply:  "optimization" is incidental here.  Python wants static
scoping *primarily* for sanity.  Dynamic scoping is more error-prone because
more surprise-prone.




From glhzxo at aol.com  Thu Mar  8 10:40:25 2001
From: glhzxo at aol.com (glhzxo at aol.com)
Date: Thu, 08 Mar 2001 15:40:25 GMT
Subject: Napster Zapstered - Downloads Mastered! Get them HERE!  FREE!
Message-ID: <JJNp6.19556$m8.181952@news.direcpc.com>

If Napster won't help - WE WILL!  Everything you want 
is available RIGHT NOW at #MP3_DEPOT on IRC; the server is 

irc.action-irc.net.  

Need speed?  Well, most of our databases are cable 
or T-1.  The servers are all megaspeed - all are on a 
fiber channel. You won't be disappointed.  And we 
have servers in the U.S., Canada, Europe, even 
Australia.  Wherever you are, there is a server close 
enough to give you top speed! 

Everything you need and want, in one convenient place 
and, as a bonus, REAL NICE PPL!  Friendly chat, and 
OPS who are actually users, not pricks.  You have a real 
opportunity to pitch in and help your fellow internet 
junkies, and maybe earn OP or even IRCOP status.

Action_IRC is a small network, owned and operated 
By the users.  We are a friendly bunch - but we 
unquestionably have the largest music library on the 
Internet.  We have 10 servers worldwide and 300 
channels.  Our selection is absolutely incredible.  
Of course we have all the modern pop, rock, and rap.  
But then we also have jazz, blues, bluegrass, country, 
classical, 50's, and anything else you are looking for.

If you are new to IRC, you are in for a real treat.  
There are none of the hassles you have with 
Napster.  You don't get half a file then never see 
the user again.  And, unlike Napster, we don't have 
a million users, each with the same 3 files!  Just 
download a database list, select your file and 
download.  You can choose from over 50 MP3 
databases.  Our MP3 databases range from 100 
gigabytes to a terabyte - that's right, a Terrabyte 
of MP3's in ONE PLACE!  

Looking for Celtic Mouth Music?  No? Well, how 
about  Hilary Hahn, BEETHOVEN Violin Concerto 
BERNSTEIN Serenade - Aristophanes?  Or a rare
Jimmi Hendrix boot?  Frank Zappa Compositions 
by the Kronos Quartet? All of the works of Robert 
Johnson or Howlin' Wolf? Carter Family, anyone? 
Or the Library of Congress collection of great 
American folk?  Or Hank Williams demos? That 
oldie from your Senior Prom - or the newbie, 
as the case may be... we have it all. Did you 
know that the Chordettes did a female barbarshop 
album? Yeah, we got that.

But music is not all we have.  Action_IRC also has the 
largest binary library on the net.  Want movies?  Then 
visit #Movietime or #Movieland for them.  Is videos 
your thing?  Then you will want to check out 
#Musicvideos.  And don't forget to stop by #Warez4u, 
#Hotwarez, #Crackhouse, or #Gametime.

So stop by and see us.  We will be waiting for you!

IF YOU HAVE NEVER USED IRC BEFORE
---------------------------------

If you have never used IRC before, IRC is really 
easy to use.  Any of the following links will directly 
download MIRC, the program you need:

http://www.kissland.com/mirc/mirc582t.exe
http://www.carbosoft.com/mirc/mirc582t.exe
ftp://mirc.entic.net/pub/mIRC/mirc582t.exe
http://mirc.mcclimans.com/downloads/mirc582t.exe

NOTE TO TECHIES: ONLY use MIRC, 5.71 or above.  
That's because you will need to load SpR, a MIRC 
script to access the databases.  It works only on MIRC 
5.71 or higher.
 
Install MIRC, then look for this server:

Action-IRC
 
If you can't find it, or are using an earlier version of 
MIRC (or another client) then either add this server:

irc.action-irc.net

or just type 

/join irc.action-irc.net

When you connect to irc.action-irc.net, a window will 
pop up with lots of channels.  Select them all 
and delete them.  
 
Then add this channel
 
    #MP3_Depot
 
You need the #.  Join that channel.  If you 
can't figure out how to add a channel, once 
you have logged into a server and get the 
welcome message, close the menu box and at 
the bottom of the page just type this:
 
/join #MP3_Depot
 
No space or other character before the /
 
That will bring you to our channel  From there 
just ask any question you have and someone will 
answer it.

jkxoqyguphuuvhfvhnxlebitveqdgukvfwgglbofiptlhqrpltedcjztwzcyfhqxlp



From root at rainerdeyke.com  Thu Mar  8 22:58:37 2001
From: root at rainerdeyke.com (Rainer Deyke)
Date: Fri, 09 Mar 2001 03:58:37 GMT
Subject: from Tkinter import *
References: <mailman.983920449.5230.python-list@python.org> <slrn9ab0g7.s6.matt@happy-hour.mondoinfo.com> <989dot$t6t$1@panix3.panix.com>
Message-ID: <NxYp6.375996$ge4.132135288@news2.rdc2.tx.home.com>

"Aahz Maruch" <aahz at panix.com> wrote in message
news:989dot$t6t$1 at panix3.panix.com...
> Actually, I disagree with the effbot.  I used to think it was okay to
> special-case "from Tkinter import *", but I've come to the conclusion
> that "import *" is _always_ a Bad Idea in any sort of production code.
> Sure it takes more work to write
>
> from Tkinter import Tk, Frame, Button, Entry, Canvas, Text
>
> but I think the payoff in clarity is worth it.

I have used 'from ... import *' to group several imports:

includes.py:
import a, b, c, d, e, f, g # This line is the entire content of includes.py

somethingelse.py:
from includes import * # Imports a, b, c, d, e, f, g
...


--
Rainer Deyke (root at rainerdeyke.com)
Shareware computer games           -           http://rainerdeyke.com
"In ihren Reihen zu stehen heisst unter Feinden zu kaempfen" - Abigor




From evertonreid at netscapeonline.co.uk  Sun Mar 11 23:38:55 2001
From: evertonreid at netscapeonline.co.uk (Everton reid)
Date: Mon, 12 Mar 2001 04:38:55 +0000
Subject: programming h----ng
Message-ID: <3AAC52DF.420FB510@netscapeonline.co.uk>

You might not want to answer this but I have read and  I am just blocked
I know the basic of c/C++ but I am hungryfor more reverse engineer
canyou help please
please please please please please
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20010312/575668ef/attachment.html>

From akuchlin at mems-exchange.org  Thu Mar 22 17:41:59 2001
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 22 Mar 2001 17:41:59 -0500
Subject: Curses under python 2.0
References: <99darm$o3o$00$1@news.t-online.com>
Message-ID: <3dsnk5zars.fsf@ute.cnri.reston.va.us>

"Christian Theune" <ct at gocept.com> writes:
> >>> import curses
> Traceback (most recent call last):
>   File "<stdin>", line 1, in ?
>   File "/usr/local/lib/python2.0/curses/__init__.py", line 15, in ?
>     from _curses import *
> ImportError: No module named _curses

You *did* enable the _curses module in Modules/Setup, right?

--amk


From kalle at gnupung.net  Thu Mar  8 14:11:12 2001
From: kalle at gnupung.net (Kalle Svensson)
Date: Thu, 8 Mar 2001 20:11:12 +0100
Subject: "Wrap" a python-cgi
In-Reply-To: <3aa7c5f9$0$21482@SSP1NO25.highway.telekom.at>; from dusty@strike.wu-wien.ac.at on Thu, Mar 08, 2001 at 06:48:31PM +0100
References: <3aa7c5f9$0$21482@SSP1NO25.highway.telekom.at>
Message-ID: <20010308201112.A6106@father>

Sez Hermann Himmelbauer:
> Hi,
> I am currently using python for several cgi programs. One problem is that 
> if my cgi-script has a bug it exits with an exception and the user gets an 
> ugly error message done by the webserver.
> 
> I would like to put some really simple "bugfree" (if something like this 
> can exist :-) "wrapper" around my cgi-scripts, let it trap Exceptions, log 
> the error and let it output a nice error page to the user.
> 
> My first approach would perhaps be to wrap the whole script into a "try: 
> except" clause but at first I don't know if nested try/except clauses are 
> allowed and moreover the whole script looks somehow ugly as every line is 
> indented.

This is basically a good idea, but you should use functions too.  Something
like this might work (untested):

def main():
    # whatever goes here...

if __name__ == "__main__":
    print "Content-Type: text/html\n"
    try:
        main()
    except:
        import traceback, sys
	print "<h1>Error!</h1>\n<pre>"
	traceback.print_exc(file=sys.stdout)
	print "</pre>"

And yes, try: ... except: blocks can be nested.

Peace,
  Kalle
-- 
Email: kalle at gnupung.net     | You can tune a filesystem, but you
Web: http://www.gnupung.net/ | can't tune a fish. -- man tunefs(8)
PGP fingerprint: 0C56 B171 8159 327F 1824 F5DE 74D7 80D7 BF3B B1DD
[Not signed due to braindamage.  Blame Microsoft Outlook Express.]



From chris.gonnerman at usa.net  Mon Mar  5 00:24:10 2001
From: chris.gonnerman at usa.net (Chris Gonnerman)
Date: Sun, 4 Mar 2001 23:24:10 -0600
Subject: Newbie question about nested lists and constructor functions...
References: <W1Do6.4683$ke1.121750@jekyl.ab.tac.net> <mailman.983764626.21888.python-list@python.org> <csEo6.4686$ke1.122169@jekyl.ab.tac.net>
Message-ID: <01a901c0a534$8701d2e0$a100000a@local>

----- Original Message -----
From: "Norm Lyons" <norm at cybersurf.net>
Sent: Sunday, March 04, 2001 10:18 PM
Subject: Re: Newbie question about nested lists and constructor functions...


> Sorry about that Chris.

S'awright.

> I am trying to create instantiated objects with the following properties:
>
> list_var = [(key, value), left, right]
>
> where the "key, value" pair are a tuple, and the variables "left" and
> "right" are empty lists [ ] which will be manipulated to hold values later
> in the program after they have already been initialized to empty lists.
The
> purpose of this object is to hold the necessary data for nodes in a binary
> search tree.  I know how to implement this type of data structure in C++
but
> I seem to be getting hung up on the syntax differences in the Python class
> and member variable declarations.
>
> An example of one of my attempts would be:
>
> class Node:
>      def __init__(self, key = ''):
>           left = []
>           right = []
>           data = (key, 'value')
>           newnode = [data, left, right]
>      def info(self):
>           return self.data

OKAY.  Now we are getting somewhere.  This seems to be a common mistake for
users of other languages.  Your code likely should be like this:

class Node:
     def __init__(self, key = ''):
          left = []
          right = []
          data = (key, 'value')
          self.data = [data, left, right]
     def info(self):
          return self.data[0]

SPECIFICALLY, you need to assign your list to a data attribute of self.  I
would write
your code more like this:

class Node:
     def __init__(self, key = ''):
          self.left = []
          self.right = []
          self.data = (key, 'value')
     def info(self):  # not really needed with this version IMHO
          return self.data

or perhaps:

class Node:
     def __init__(self, key = '', value = ''):
          self.left = None   # these get object references (later)
          self.right = None  # so I don't think they need to be lists.
          self.key = key
          self.value = value  # is this a default?
     def info(self):  # not really needed with this version IMHO
          return (self.key, self.value)

but that is a personal thing.

> The "info" function is only intended to return the values of the variables
> "key and value".  The other values are used as pointers to child nodes in
> the tree.  As I mentioned earlier, I have been through the online tutorial
a
> few times and have not found any information that clears this issue up for
> me.  Obviously, the code above is incomplete as it should also contain the
> appropriate "if" and "elif" statements to properly branch according to the
> information passed to the constructor function.  Unfortunately, I do not
> know what I am doing wrong.  Any suggestions?

I think I have this all right, but I haven't test run it yet.  Good luck :-)





From sholden at holdenweb.com  Thu Mar 22 00:41:50 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 22 Mar 2001 00:41:50 -0500
Subject: comp.lang.python.travel, anyone?
Message-ID: <fkgu6.4536$WC2.158010@e420r-atl1.usenetserver.com>

Just wondered how many on this news group travel regularly, and if so where?
I was in Albuquerque two weeks ago, New York last week and Boston this (tip
o' the hat to Tim Peters, who formerly lived here).  Sweden in April. Any
Pythoneers out there?

regards
 Steve





From jmg at ecs.soton.ac.uk  Tue Mar 13 14:36:35 2001
From: jmg at ecs.soton.ac.uk (Jacek Generowicz)
Date: 13 Mar 2001 19:36:35 +0000
Subject: Extending Python: rewriting a single method in C
References: <g0wv9vxms0.fsf@scumbag.ecs.soton.ac.uk> <slrn9apn8f.382.jepler@potty.housenet> <g0ofv6xbj4.fsf@scumbag.ecs.soton.ac.uk> <98l1i901k4u@news1.newsguy.com>
Message-ID: <g0lmq9xxz0.fsf@scumbag.ecs.soton.ac.uk>

"Alex Martelli" <aleaxit at yahoo.com> writes:

> > How do I get at the members of a class in C if that class was defined
> > in python ?
> 
> By passing the reference to the instance to the function, and
> receiving it on the C side as a PyObject* -- you can then call
> methods and access attributes on it from C, just as you could
> from Python.  See the "O" format code in the documentation of
> PyArg_ParseTuple (1.7 in the "Extending and Embedding" manual).

This is exactly what I have been trying to do, using exactly that
reference as a guide, but I haven't managed to get a working piece of
code yet.

> > JE> In fast_a and fast_b, you can get arguments by using the ParseTuple
> string
> > JE> "(Oii)" and then use PyObject_SetAttrString and PyObject_GetAttrString
> > JE> to store or retrieve attributes (including, if necessary, instance
> > JE> methods) from the instance by their name.
> >
> > Is there an organized reference where I can find the specifications of
> > all these functions ?  All I have come across shows example code, but
> > no full descriptions of what can/must be done.
> 
> It's somewhat disorganized.  I think Beazley's "Python Essential
> Reference" (New Riders) is your best bet among current books (but
> it only covers Python 1.5.2).

Even now that the local copy of the above has retured from walkabout,
I am not all that much wiser.

I've not seriously used gettattr before, so maybe I'm missing something.

As I understand it, I have to provide a __getattr__ method in my
python class which will get called when I use PyObject_GetAttr. (If
this is so then I don't see why the function isn't automatically
provided; any creative definition---ie differing from the blatantly
obvious (and hence tedious)---surely only opens the door to
horrors. Still, mine is not to reason why.)

In Beazley I see that the signature of the PyObject_GetAttr is 

PyObject * PyObject_GetAttr( PyObject *o, PyObject *attr_name )

Nowhere can I find an explanation of the variables, so I'm guessing
that o should be a pointer to the object whose attribute I want to get
(which I will have obtained by something like 

  PyObject *o;
  if (! PyArg_ParseTuple( args, "O", o ) )
    return NULL;

). But what on earth is attr_name ?  Is it a PyObject* pointing to a
python string containing the name of the attribute I want ?

How about PyObject_GetAttrString ?  Is it the same as PyObject_GetAttr
except that I can use a C string to give the name of the attribute ?

Assuming that is so, and that these funcitons return a pointer to the
attribute in question, how do I do anything with it ?  For example,
assume that I know that PyObject * O points to a python integer, how
do I convert it to a C integer? or is this The Wrong Thing To Do ? 

(Would it be fair to say that the documentaion is, on the whole, less
that satisfactory ?)

> > Maybe some kind soul could show me what the correct code for a minimal
> > extension type is ?
> >
> > For example, I tried to write an extension type equivalent to
> >
> >
> > class foo:
> >     def __init__ ( self, a ):
> > self.a = a
> >     def show ( self ):
> > print a
> >
> > My various attempts, almost work, but there is always something not
> > quite right.
> 
> Getting equivalence to this is not going to be easy -- the
> type/class separation hits hard,

Could you elabore a bit on this, please ? (the bit after --).

Jacek


From tim at vegeta.ath.cx  Thu Mar  8 03:39:35 2001
From: tim at vegeta.ath.cx (Tim Hammerquist)
Date: Thu, 08 Mar 2001 08:39:35 GMT
Subject: "!=" is bad form. Re: sorry....never mind
References: <DZkp6.18811$pF4.3015652@news1.telusplanet.net> <fdlp6.16982$go.2157953@news0.telusplanet.net> <3AA6E308.5B63ECC6@cosc.canterbury.ac.nz> <3AA72253.FB021815@Lugoj.Com>
Message-ID: <slrn9aehgr.33d.tim@vegeta.ath.cx>

James Logajan <JamesL at Lugoj.Com> wrote:
> Greg Ewing wrote:
> > Victor Louie wrote:
> > > <> is obsolescent.
> > 
> > Really? I hope it's not taken away, because that will
> > break ALL of my code! I hate != (looks too much like
> > C), so I always use <>. Nobody ever told me I was doing
> > anything wrong!

I hate '<>' because it's too much like VB (and so many other BASIC
variants).

> Actually, for Python, "!=" IS totally bogus and should be considered the
> unusual variant.

The only confusing thing I've noticed about this is that it comes close
to violating Python's guideline: There should be one, and only one,
reasonable way to do something.  But I, unlike Mr. Logajan, decline to
challenge Guido's better judgement.  (Hell, _he_ wrote it, not you!)

[ snip ]

> Shucks, if one is going to borrow "!=" from "C", why stop there? Why not
> borrow the keywords and concepts such as "register", "auto", "extern",
> pointers, malloc, free, etc.

Heck, why not throw out print(), the '==' operator, the printf()-style
"%f" notation, and the whole dot (.) access operator just because they
were based on something C (or C++ for access op) had?  Or, just maybe,
we can get past this?

-- 
-Tim Hammerquist <timmy at cpan.org>
Programmers are achievement oriented; give them an impossible task,
and they'll do their best to give you what they think you would have
asked for if you had a clue as to what was possible.
	-- Peter Coffee, PC Week


From dav-nospam-id at westco-nospam-ntrol.com  Thu Mar 22 09:23:48 2001
From: dav-nospam-id at westco-nospam-ntrol.com (David Brown)
Date: Thu, 22 Mar 2001 15:23:48 +0100
Subject: Python 2.1 and MySQL 3.23.33 on Windows
Message-ID: <99d1te$ogo$1@news.netpower.no>

I have MySQL 3.23.33 and Python 2.1 installed on my Windows machine.  Both
work fine (I use MySQL along with PHP and Apache), and I have used Python
for a few short scripts.  But I cannot get Python to work with MySQL.  The
mySQLdb windows installation is built for Python 2.0, not 2.1, and I do not
have MSVC to recompile it myself.  Does anyone have any hints or pointers to
get me started?  Thanks.

David Brown
Norway





From ping at lfw.org  Mon Mar 12 00:18:06 2001
From: ping at lfw.org (Ka-Ping Yee)
Date: Sun, 11 Mar 2001 21:18:06 -0800 (PST)
Subject: [Python-Dev] Revive the types sig?
In-Reply-To: <Pine.LNX.4.32.0103111745440.887-100000@localhost.localdomain>
Message-ID: <Pine.LNX.4.10.10103112102030.13108-100000@skuld.kingmanhall.org>

On Sun, 11 Mar 2001, Michel Pelletier wrote:
> As I see it, interfaces satify your first point, remove the need for your
> second and third point, satify your fourth point, and meet the goals of
> your fifth.

For the record, here is a little idea i came up with on the
last day of the conference:

Suppose there is a built-in class called "Interface" with the
special property that whenever any immediate descendant of
Interface is sub-classed, we check to make sure all of its
methods are overridden.  If any methods are not overridden,
something like InterfaceException is raised.

This would be sufficient to provide very simple interfaces,
at least in terms of what methods are part of an interface
(it wouldn't do any type checking, but it could go a step
further and check the number of arguments on each method).

Example:

    >>> class Spam(Interface):
    ...     def islovely(self): pass
    ...
    >>> Spam()
    TypeError: interfaces cannot be instantiated
    >>> class Eggs(Spam):
    ...     def scramble(self): pass
    ...
    InterfaceError: class Eggs does not implement interface Spam
    >>> class LovelySpam(Spam):
    ...     def islovely(self): return 1
    ...
    >>> LovelySpam()
    <LovelySpam instance at ...>

Essentially this would replace the convention of writing a
whole bunch of methods that raise NotImplementedError as a
way of describing an abstract interface, making it a bit easier
to write and causing interfaces to be checked earlier (upon
subclassing, rather than upon method call).

It should be possible to implement this in Python using metaclasses.


-- ?!ng

"Computers are useless.  They can only give you answers."
    -- Pablo Picasso




From bh at abbenay.cs.berkeley.edu  Sun Mar 11 19:43:00 2001
From: bh at abbenay.cs.berkeley.edu (Brian Harvey)
Date: 11 Mar 2001 16:43:00 -0800
Subject: Einstein's Riddle
References: <oNPq6.396644$ge4.138479566@news2.rdc2.tx.home.com> <98glrd$1si1a$1@ID-75892.news.dfncis.de>
Message-ID: <98h62k$knn$1@abbenay.CS.Berkeley.EDU>

"Boris Popov" <no-spam-popov at bigserver.com> writes:
>house         1 2 3 4 5
>color            yellow blue red green white
>nationality   norweigian dane brit german swede
>drink            water tea milk coffee beer
>smoke        dunhill blend pall mall prince bluemaster
>pet              cats horse birds fish dogs
>
>P.S. Did I have to write software to solve it ? :)

I can't resist being (apparently) first to post a computer program.
But it doesn't really belong in any of these newsgroups, because it's
written in Logo.

First get Berkeley Logo from ftp://ftp.cs.berkeley.edu/ucblogo/
(pick the one that matches your platform)

Then get the logic-puzzle-solver from
http://www.cs.berkeley.edu/~bh/logic-code/hybrid.lg

Then here's the part specific to this puzzle:

? po [einstein neighbor neighbor1]
to einstein
cleanup
category "position [1 2 3 4 5]
category "color [red green white yellow blue]
category "nationality [British Swedish Danish Norwegian German]
category "beverage [tea coffee milk beer water]
category "cigar [pallmall dunhill blend bluemaster prince]
category "pet [dog bird cat horse fish]
verify "British "red
verify "Swedish "dog
verify "Danish "tea
justbefore "green "white :position
verify "green "coffee
verify "pallmall "bird
verify "yellow "dunhill
verify 3 "milk
verify 1 "Norwegian
neighbor "blend "cat :position
neighbor "horse "dunhill :position
verify "bluemaster "beer
verify "German "prince
neighbor "Norwegian "blue :position
neighbor "blend "water :position
print []
solution
end

to neighbor :a :b :lineup
falsify :a :b
neighbor1 :lineup count :lineup
end

to neighbor1 :lineup :count
if :count=0 [stop]
foreach bl bf bf :lineup [implies :a first :lineup "true :b ? "false]
neighbor1 (lput first :lineup bf :lineup) :count-1
end

? solve [einstein]
category position [1 2 3 4 5]
category color [red green white yellow blue]
category nationality [British Swedish Danish Norwegian German]
category beverage [tea coffee milk beer water]
category cigar [pallmall dunhill blend bluemaster prince]
category pet [dog bird cat horse fish]
British red -> true
Swedish red -> false
Danish red -> false
Norwegian red -> false
German red -> false
green British -> false
white British -> false
yellow British -> false
blue British -> false
Swedish dog -> true
British dog -> false
red dog -> false
Danish dog -> false
Norwegian dog -> false
German dog -> false
bird Swedish -> false
cat Swedish -> false
horse Swedish -> false
fish Swedish -> false
Danish tea -> true
British tea -> false
red tea -> false
Swedish tea -> false
dog tea -> false
Norwegian tea -> false
German tea -> false
coffee Danish -> false
milk Danish -> false
beer Danish -> false
water Danish -> false
green 5 -> false
white 1 -> false
green coffee -> true
red coffee -> false
British coffee -> false
white coffee -> false
yellow coffee -> false
blue coffee -> false
5 coffee -> false
tea green -> false
Danish green -> false
milk green -> false
beer green -> false
water green -> false
pallmall bird -> true
dunhill bird -> false
blend bird -> false
bluemaster bird -> false
prince bird -> false
dog pallmall -> false
Swedish pallmall -> false
cat pallmall -> false
horse pallmall -> false
fish pallmall -> false
yellow dunhill -> true
red dunhill -> false
British dunhill -> false
green dunhill -> false
coffee dunhill -> false
white dunhill -> false
blue dunhill -> false
pallmall yellow -> false
bird yellow -> false
blend yellow -> false
bluemaster yellow -> false
prince yellow -> false
3 milk -> true
1 milk -> false
2 milk -> false
4 milk -> false
5 milk -> false
tea 3 -> false
Danish 3 -> false
coffee 3 -> false
green 3 -> false
white 4 -> false
beer 3 -> false
water 3 -> false
1 Norwegian -> true
2 Norwegian -> false
3 Norwegian -> false
milk Norwegian -> false
4 Norwegian -> false
5 Norwegian -> false
white Norwegian -> false
British 1 -> false
red 1 -> false
Swedish 1 -> false
dog 1 -> false
Danish 1 -> false
tea 1 -> false
German 1 -> false
blend cat -> false
horse dunhill -> false
yellow horse -> false
bluemaster beer -> true
pallmall beer -> false
bird beer -> false
dunhill beer -> false
yellow beer -> false
blend beer -> false
prince beer -> false
tea bluemaster -> false
Danish bluemaster -> false
coffee bluemaster -> false
green bluemaster -> false
milk bluemaster -> false
3 bluemaster -> false
water bluemaster -> false
German prince -> true
British prince -> false
red prince -> false
Swedish prince -> false
dog prince -> false
Danish prince -> false
tea prince -> false
Norwegian prince -> false
1 prince -> false
pallmall German -> false
bird German -> false
dunhill German -> false
yellow German -> false
blend German -> false
bluemaster German -> false
beer German -> false
Norwegian blue -> false
1 blue -> false
blue 3 -> false
milk blue -> false
blue 4 -> false
blend water -> false


  Assuming 1 yellow
  1 yellow -> true
  2 yellow -> false
  dunhill 2 -> false
  3 yellow -> false
  milk yellow -> false
  dunhill milk -> false
  3 dunhill -> false
  4 yellow -> false
  dunhill 4 -> false
  5 yellow -> false
  dunhill 5 -> false
  1 dunhill -> true
  Norwegian dunhill -> true
  Swedish dunhill -> false
  dog dunhill -> false
  yellow dog -> false
  Swedish yellow -> false
  Danish dunhill -> false
  tea dunhill -> false
  water dunhill -> true
  pallmall water -> false
  bird water -> false
  prince water -> false
  German water -> false
  yellow water -> true
  red water -> false
  British water -> false
  white water -> false
  blue water -> false
  Swedish water -> false
  Norwegian water -> true
  1 water -> true
  2 water -> false
  4 water -> false
  5 water -> false
  dog water -> false
  coffee 1 -> false
  green 1 -> false
  Norwegian green -> false
  yellow Norwegian -> true
  beer Norwegian -> false
  bluemaster Norwegian -> false
  1 bluemaster -> false
  beer 1 -> false
  horse Norwegian -> false
  1 horse -> false
  water horse -> false
  blend Norwegian -> false
  1 blend -> false
  pallmall Norwegian -> false
  bird Norwegian -> false
  1 bird -> false
  pallmall 1 -> false
  coffee Norwegian -> false
  Danish yellow -> false
  tea yellow -> false
  white 2 -> false
  blend 4 -> false
  blend 3 -> false
  milk blend -> false
  horse 4 -> false
  horse 3 -> false
  milk horse -> false

    Assuming 5 beer
    5 beer -> true
    2 beer -> false
    bluemaster 2 -> false
    4 beer -> false
    bluemaster 4 -> false
    5 bluemaster -> true
    pallmall 5 -> false
    bird 5 -> false
    blend 5 -> false
    2 blend -> true
    white blend -> false
    pallmall 2 -> false
    bird 2 -> false
    prince 2 -> false
    German 2 -> false
    cat 2 -> false
    cat 5 -> false
    bluemaster cat -> false
    beer cat -> false
    cat 4 -> false
    prince 5 -> false
    German 5 -> false
    tea 5 -> false
    Danish 5 -> false

      Assuming 5 Swedish
      5 Swedish -> true
      2 Swedish -> false
      blend Swedish -> false
      bluemaster Swedish -> true
      beer Swedish -> true
      coffee Swedish -> false
      green Swedish -> false
      German green -> true
      prince green -> true
      pallmall green -> false
      bird green -> false
      coffee bird -> false
      pallmall coffee -> false
      blend green -> false
      2 green -> false
      4 green -> true
      cat green -> false
      coffee cat -> false
      horse green -> false
      coffee horse -> false
      red 4 -> false
      British 4 -> false
      coffee 4 -> true
      tea 4 -> false
      Danish 4 -> false
      2 Danish -> true
      blend Danish -> true
      pallmall Danish -> false
      bird Danish -> false
      British bird -> true
      red bird -> true
      white bird -> false
      pallmall white -> false
      blue bird -> false
      pallmall blue -> false
      red pallmall -> true
      British pallmall -> true
      4 pallmall -> false
      3 pallmall -> true
      milk pallmall -> true
      tea pallmall -> false
      blend tea -> true
      2 tea -> true
      cat tea -> false
      Danish cat -> false
      bird tea -> false
      milk bird -> true
      3 bird -> true
      4 bird -> false
      dog 3 -> false
      Swedish 3 -> false
      milk Swedish -> false
      dog milk -> false
      cat 3 -> false
      1 cat -> true
      yellow cat -> true
      red cat -> false
      British cat -> false
      white cat -> false
      blue cat -> false
      water cat -> true
      milk cat -> false
      Norwegian cat -> true
      German cat -> false
      prince cat -> false
      dunhill cat -> true
      fish dunhill -> false
      1 fish -> false
      yellow fish -> false
      water fish -> false
      Norwegian fish -> false
      fish 3 -> false
      milk fish -> false
      white 3 -> false
      red 3 -> true
      British 3 -> true
      German 3 -> false
      prince 3 -> false
      4 prince -> true
      horse prince -> false
      fish prince -> true
      milk prince -> false
      coffee prince -> true
      blue prince -> false
      German blue -> false
      white prince -> false
      bluemaster white -> true
      5 white -> true
      beer white -> true
      tea white -> false
      Danish white -> false
      blue Danish -> true
      Swedish blue -> false
      dog blue -> false
      white Swedish -> true
      4 Swedish -> false
      German 4 -> true
      dog 4 -> false
      fish 4 -> true
      2 fish -> false
      blend fish -> false
      5 fish -> false
      beer fish -> false
      bluemaster fish -> false
      green fish -> true
      red fish -> false
      British fish -> false
      white fish -> false
      blue fish -> false
      Danish fish -> false
      German fish -> true
      tea fish -> false
      coffee fish -> true
      dog coffee -> false
      beer dog -> true
      5 dog -> true
      2 dog -> false
      blend dog -> false
      bluemaster dog -> true
      green dog -> false
      white dog -> true
      horse white -> false
      milk white -> false
      red milk -> true
      British milk -> true
      German milk -> false
      coffee German -> true
      horse German -> false
      white German -> false
      2 coffee -> false
      blend coffee -> false
      beer British -> false
      5 British -> false
      bluemaster British -> false
      red bluemaster -> false
      5 red -> false
      beer red -> false
      horse British -> false
      Danish horse -> true
      tea horse -> true
      beer horse -> false
      5 horse -> false
      2 horse -> true
      blend horse -> true
      bluemaster horse -> false
      British blend -> false
      red blend -> false
      blue blend -> true
      bluemaster blue -> false
      5 blue -> false
      2 blue -> true
      beer blue -> false
      tea blue -> true
      red 2 -> false
      British 2 -> false
      horse blue -> true
      red horse -> false

1 yellow Norwegian water dunhill cat
2 blue Danish tea blend horse
3 red British milk pallmall bird
4 green German coffee prince fish
5 white Swedish beer bluemaster dog


From fredrik at pythonware.com  Thu Mar  1 16:35:07 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Thu, 01 Mar 2001 21:35:07 GMT
Subject: [Q] Inspect variables during execution?
References: <ESyn6.12612$Z8.1615647@typhoon.southeast.rr.com>
Message-ID: <fgzn6.16229$Qb7.2646735@newsb.telia.net>

Daehyok Shin wrote:
> Is there any function helping me to inspect variables for debugging during
> execution?

something like this, perhaps?

#
# code-example-3.py

def keyboard(banner=None):
    import code, sys

    # use exception trick to pick up the current frame
    try:
        raise None
    except:
        frame = sys.exc_info()[2].tb_frame.f_back

    # evaluate commands in current namespace
    namespace = frame.f_globals.copy()
    namespace.update(frame.f_locals)

    code.interact(banner=banner, local=namespace)

#
# usage example

def func():
    print "START"
    a = 10
    keyboard()
    print "END"

func()

# START
# Python 1.5.2
# Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
# (InteractiveConsole)
# >>> print a
# 10
# >>> print keyboard
# <function keyboard at 9032c8>
# ^Z
# END

Cheers /F

<!-- (the eff-bot guide to) the standard python library:
http://www.pythonware.com/people/fredrik/librarybook.htm
-->




From stephen_purcell at yahoo.com  Tue Mar 27 11:14:23 2001
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Tue, 27 Mar 2001 18:14:23 +0200
Subject: Eat this one
In-Reply-To: <3AC0B349.EE0AA9B2@hack.org>; from bos@hack.org on Tue, Mar 27, 2001 at 05:35:37PM +0200
References: <3AC0B349.EE0AA9B2@hack.org>
Message-ID: <20010327181423.A1260@freedom.puma-ag.com>

Rikard Bosnjakovic wrote:
> I feel ashamed to show the code, but I didn't find any other solution :)
> 
> What way would be better and _faster_ to wipe out the five chars from
> the list?

Here is a nice readable way:

    >>> import string
    >>> s = 'here is < some > text with& nasty characters'
    >>> for badchar in "<>&\n":
    ...   s = string.replace(s,badchar,'')
    ... 
    >>> s
    'here is  some  text with nasty characters'

And using regular expressions:

    >>> import re
    >>> s2 = re.sub('[<>&\n]', '', s)
    >>> s2 
    'here is  some  text with nasty characters'


-Steve


-- 
Steve Purcell, Pythangelist
Get testing at http://pyunit.sourceforge.net/
Any opinions expressed herein are my own and not necessarily those of Yahoo



From root at [127.0.0.1]  Sat Mar 31 02:22:43 2001
From: root at [127.0.0.1] (deadmeat)
Date: Sat, 31 Mar 2001 07:22:43 GMT
Subject: list vs tuple
References: <hGlw6.782$p5.3210@news1.rivrw1.nsw.optushome.com.au> <3AC20CD3.265A2D94@alcyone.com> <ru69cto61jtmrqsf809e38hb3pbm2lhi7g@4ax.com> <bC1x6.4319$4N4.905946@newsc.telia.net> <71q9ct4le63anefmiajj5cbh3mg04qu1ga@4ax.com> <mailman.985991202.5578.python-list@python.org>
Message-ID: <7Bfx6.989$p5.3791@news1.rivrw1.nsw.optushome.com.au>

> Have you used other languages before coming to Python?  I would
> think it would be confusing to people who were used to languages
> that have a reference/value inconsistency.  Python is totally
> consistent (everything is a object, all names refer to objects).

No it's not.  I come from a Pascal background, the behaviour noted in the
parent is what would happen if two pointers were made equal.  To copy the
contents from one variable to another, a ^ is used to refer to the structure
at the pointer, not the pointer itself.

Python does not do that

>>> a = 1
>>> b = a
>>> a = 5
>>> b
1

>>> a = [1,2,3]
>>> b = a
>>> b[1] = 999
>>> b
[1, 999, 3]

It's dependant on datatype, which most people would think as inconsistant.

Example 2 is the same for function arguments, pass all lists to functions
with a [:] on the end to ensure your function doesn't modify your list
without the parent code knowing about it.





From cjw at sympatico.ca  Sun Mar 11 12:28:04 2001
From: cjw at sympatico.ca (Colin J. Williams)
Date: Sun, 11 Mar 2001 17:28:04 GMT
Subject: Python In A Nutshell
References: <nG5q6.54910$__6.11376601@typhoon.southeast.rr.com> <Ufcq6.140786$47.2078615@news.bc.tac.net>
Message-ID: <3AABB165.959ABD5@sympatico.ca>

Laura,

Two suggestions:

   * Produce the book with a binding so that it will lie flat on a desk when open
   * Document the classes in wxPython

Colin W.

LLewin at oreilly.com wrote:

> Hi,
> Yep--Python Nutshell is "forthcoming"...we're just in the planning stages at this point--very early on in the process.  Any suggestions are welcome!
> Laura
> LLewin at oreilly.com
>
> > Hey Alex,  congrats on landing "Python In A Nutshell" book deal.  One
> > problem.  I went down to the book store and it's not there.  What gives?
> > <wink>
> >
> > Don
> >
> >
>
> ==================================
> Posted via http://nodevice.com
> Linux Programmer's Site
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20010311/2c6600b7/attachment.html>

From piet at cs.uu.nl  Thu Mar 29 08:11:20 2001
From: piet at cs.uu.nl (piet at cs.uu.nl)
Date: 29 Mar 2001 15:11:20 +0200
Subject: Getting all the *files* from a directory -- A better way??
References: <mailman.985732718.3725.python-list@python.org> <99r6v7$4hef2$1@fido.engr.sgi.com> <slrn9c3jf1.qb3.scarblac@pino.selwerd.nl>
Message-ID: <u66gsbtzb.fsf@cs.uu.nl>

>>>>> scarblac at pino.selwerd.nl (Remco Gerlich) (RG) writes:

RG> Or even

RG> dir="a/b/c"
RG> files = [os.path.join(dir, f) for f in os.listdir(dir)
RG>                               if os.path.isfile(f)]

RG> If you do the map() with a list comprehension, why not the filter() as
RG> well.

Perhaps because it would have to be:
                 if os.path.isfile(os.path.join(dir, f))
-- 
Piet van Oostrum <piet at cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: P.van.Oostrum at hccnet.nl


From aahz at panix.com  Mon Mar 19 09:57:37 2001
From: aahz at panix.com (Aahz Maruch)
Date: 19 Mar 2001 06:57:37 -0800
Subject: Problems with 2.0 cgi scripts
References: <fs9bbtg68qne71t853lp89l44uarbb5jk5@4ax.com>
Message-ID: <9956p1$8g0$1@panix3.panix.com>

In article <fs9bbtg68qne71t853lp89l44uarbb5jk5 at 4ax.com>,
Sheila King  <sheila at spamcop.net> wrote:
>
>So, basically, it cannot find the modules when I run it as a cgi-script. I
>have all the paths set correctly, and the permissions set correctly. Grr, I'm
>frustrated with this one by now...

You can't set the path because you don't own the web server.  Simplest
solution is to set sys.path yourself as the first line of the script.
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

There's a difference between a person who gets shit zie doesn't deserve
and a person who gets more shit than zie deserves.  --Aahz


From NoSpam at NoSpam.com  Fri Mar  2 15:27:09 2001
From: NoSpam at NoSpam.com (Tom)
Date: Fri, 02 Mar 2001 20:27:09 GMT
Subject: Unresolved symbols in MSVCRT.DLL when running a py2exe packaged program.
References: <mailman.983556668.16192.python-list@python.org>
Message-ID: <xmTn6.15713$UZ4.4329504@news4.rdc1.on.home.com>

"Tim Peters" <tim.one at home.com> wrote in message
news:mailman.983556668.16192.python-list at python.org...
> [Tom]
> > I disagree.
> >
> > You can distribute system dll's.  You are expected to, and sometimes you
> > have to.  You must insure that the version you load is at least
> > equal to the version that you linked against.
> >
> > I have done this (distribute system DLL's) on all windows except WinME
and
> > the 16bit ones.  Unfortunately I can't say how I did it precisely,
because
> > I just used InstallShield.
>
> And Python just uses the Wise installer.  They aren't that fancy under the
> covers, though -- they just "follow the rules", checking version numbers,
> bumping refcounts on refcounted DLLs, and arranging for a reboot if they
need
> to replace something.  But that stuff is ineffective on the newer systems
> with SFP.
>
> > But I believe that, where SFP exists, you must use specific and new
> > API's to replace DLL's, otherwise the system will back out your changes.
>
> My understanding is that SFP isn't so easily fooled; e.g.,
>
>     http://support.microsoft.com/support/kb/articles/Q253/5/71.ASP
>
>     http://www.microsoft.com/hwdev/sfp/wfp.htm
>
> SFP won't allow replacement of protected files via just using API tricks;
it
> also sucks in a digital signature scheme, and only MS can supply
signatures
> that work.

I didn't mean to suggest 'fooling' it - I agree that safeguards like SFP are
needed - I meant that one had to do it according to the rules.

As for not being able to update the DLL's, this would be a problem.  If a
service pack for VC6 updates the MFC or CRT DLL's (which SP's usually do),
and you ship an app build with it, then you must update these DLL's on your
target system.  Perhaps this is where the new ability to run 'personal'
instances of DLL's comes in: ie. you can't update the system32 version the
DLL, but you can install your own personal copy.

Tom.

> > One solution is to specify the changes in an .INF file and let the
system
> > worry about these details (but this has its limitations).
> >
> > I assume that Python itself updates the MSVCRT.dll.
>
> The PythonLabs Windows installer tries to, yes.
>
> > Might look at how it does it on Win2K.
>
> My understanding is that it can't (but Win2K ships with a "good enough"
set
> of DLLs, so the version check shouldn't trigger there -- yet).
>
> BTW, I'm *glad* MS is doing SFP!  Whatever Windows Life may be in theory,
> DLLs in practice cannot be mixed and matched -- you need a self-consistent
> set else Very Strange Things Happen.  And with about 1,000 DLLs under a
> typical Windows system directory, that's just impossible for app
developers
> to keep straight.  May well be impossible for MS to keep straight too, but
at
> least it's no longer impossible for them to *try* to keep 'em straight
now.
>
> something-to-ponder-each-time-your-win9x-crashes-ly y'rs  - tim
>
>




From stephen_purcell at yahoo.com  Wed Mar 14 12:20:27 2001
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Wed, 14 Mar 2001 18:20:27 +0100
Subject: Packages, modules and exceptions
In-Reply-To: <20010314235842.B2335@madcap.dyndns.org>; from ngps@post1.com on Wed, Mar 14, 2001 at 11:58:42PM +0800
References: <98ip5c$397$1@clematis.singnet.com.sg> <20010312163907.A23991@freedom.puma-ag.com> <20010314235842.B2335@madcap.dyndns.org>
Message-ID: <20010314182027.A30729@freedom.puma-ag.com>

Ng Pheng Siong wrote:
> Within M2Crypto/SSL/Connection.py, I have "class Connection:..." Then, 
> within M2Crypto/SSL/__init__.py, I have "from Connection import
> Connection". This is so that clients can say "s = M2Crypto.SSL.Connection()"
> instead of "s = M2Crypto.SSL.Connection.Connection()".
> 
> Because of this, I can't just do "from M2Crypto.SSL import SSLError"
> within Connection.py. __init__.py's "from Connection import Connection"
> creates a circular reference.

You can. There is no problem with such a circular reference. Just try it out.

Consider the following files and their contents:

./M2Crypto/SSL/Connection.py
	#---------
	from M2Crypto.SSL import SSLError

	class Connection:
	    pass
	#---------
./M2Crypto/SSL/__init__.py
	#---------
	class SSLError(Exception):
	   pass

	from Connection import Connection
	#---------
./M2Crypto/__init__.py
	#---------
	#---------


I can start Python, and everything works as intended:

>>> from M2Crypto.SSL import Connection
>>> Connection
<class M2Crypto.SSL.Connection.Connection at 807f408>


Magic!


-Steve


-- 
Steve Purcell, Pythangelist
Get testing at http://pyunit.sourceforge.net/
Any opinions expressed herein are my own and not necessarily those of Yahoo



From bk at xk7.com  Thu Mar  1 08:34:51 2001
From: bk at xk7.com (Burkhard Kloss)
Date: Thu, 1 Mar 2001 13:34:51 -0000
Subject: python performance
References: <J8fn6.756$me1.378895@news2.cableinet.net>
Message-ID: <983453691.827499@master.nyc.kbcfp.com>

Speed is always a tricky issue - in my experience it's fast enough, but
YMMV.  I would recommend that you write as much as possible in Python to get
it working, then profile it and either improve the algorithms or rewrite low
level routines in C.  Lots of people seem to use this approach for numerical
stuff, and if that combination is fast enough for numerics it ought to be
fast enough for anything else as well.

HTH - Burkhard
> Can  anyone comment on the speed of Python programs under Windows?
>
> I am interested in using Python as a embedded game scripting language.
> Amongst other things, I want to doing AI, but not the low level routines
> like path finding.





From tim at vegeta.ath.cx  Wed Mar  7 02:16:29 2001
From: tim at vegeta.ath.cx (Tim Hammerquist)
Date: Wed, 07 Mar 2001 07:16:29 GMT
Subject: How to use <>?
References: <DZkp6.18811$pF4.3015652@news1.telusplanet.net>
Message-ID: <slrn9abo97.on.tim@vegeta.ath.cx>

Victor Louie <vlouie at telusplanet.net> wrote:
> I was just wondering if any knew what <> does....I've got some sample code
> and can't figure out what's happening
> 
> eg.
> x =10
> if x <> 10:
>  ....

It's an alternate operator which is identical in meaning to !=
(inequality).  The following are equivalent:

	if x != 10: # do stuff
	if x <> 10: # do stuff

HTH

-- 
-Tim Hammerquist <timmy at cpan.org>

Computers are useless. They can only give you answers.
	-- Pablo Picasso


From benjamin.altman at noaa.gov  Wed Mar 28 11:41:55 2001
From: benjamin.altman at noaa.gov (Benjamin.Altman)
Date: Wed, 28 Mar 2001 11:41:55 -0500
Subject: learning python...
References: <3AB8F4E1.184DD9C@noaa.gov> <a2uh99.o71.ln@pingu>
Message-ID: <3AC21453.3A6F8524@noaa.gov>

What about the aspect of it covering a version less than 2.0 when  Programming
Python would be more up to date?

Thanks,
Ben

chris wrote:

> "Core Python Programming "from PH is easily the better book for easing into
> Python. I am an O'Reilly fan and generally buy their books blind if it is
> on a subject I am interested in (since they very rarely disappoint).
> However on this occassion the PH book is a genuinely good book and worth
> buying, considerably more than the Learning Python book and the 2e version
> of Programming Python.
>
> My opinion!
>
> Chris



From alf at leo.logilab.fr  Mon Mar  5 05:46:28 2001
From: alf at leo.logilab.fr (Alexandre Fayolle)
Date: Mon, 5 Mar 2001 10:46:28 +0000 (UTC)
Subject: Editing Python source as XML?
References: <mailman.983546704.10789.python-list@python.org> <j44rxcjgnw.fsf@informatik.hu-berlin.de>
Message-ID: <97vqq4$2ap7$1@norfair.nerim.net>

Martin von Loewis <loewis at informatik.hu-berlin.de> wrote:
> Steven Haryanto <steven at haryan.to> writes:

>> A recent experience of having to modify a long C source file has
>> led me to wonder whether there are any tool/effort towards editing
>> Python source code as XML or other form of structured document.

> Not sure *why* you'd want to do that, but it is certainly possible to
> convert any script automatically to an XML document with the help of
> the parser module.

I'm releasing today a utility we use at Logilab for documenting the code 
of Narval. We call it pypasax. It uses the parser module to extract 
information about classes and methods and generates an XML tree from this.
We are working on XSLT to generate XMI files so that we can import the data 
in some UML tool such as ArgoUML.

Alexandre Fayolle
-- 
http://www.logilab.com 
Narval is the first software agent available as free software (GPL).
LOGILAB, Paris (France).


From skidrow at nownuri.net  Sun Mar 11 05:16:48 2001
From: skidrow at nownuri.net (Seo ji won)
Date: Sun, 11 Mar 2001 19:16:48 +0900
Subject: mod_python or MoinMoin problem.
Message-ID: <98fioq$qud$1@news2-2.kornet.net>

I'm now using FreeBSD 4.2 release, apache 1.3.14, mod_python 2.6.3

and trying to use MoinMoin(one of wiki clones, written in python, and is
kind of CGI program I guess)

but it doesn't work out.

the apache log says,
----------------------------------------------------------------------------
-------------------------
Mod_python error: "PythonHandler mod_python.cgihandler"

Traceback (most recent call last):

  File "/usr/local/lib/python2.0/site-packages/mod_python/apache.py", line
185, in Dispatch
    result = object(self.req)

  File "/usr/local/lib/python2.0/site-packages/mod_python/cgihandler.py",
line 76, in handler
    imp.load_module(module_name, fd, path, desc)

  File "/usr/local/www/cgi-bin/MoinMoin/wiki-moinmoin/moin.py", line 15, in
?
    cgimain.run()

  File "/usr/local/www/cgi-bin/MoinMoin/cgimain.py", line 161, in run
    cgi.print_exception()

  File "/usr/local/lib/python2.0/cgi.py", line 870, in print_exception
    print

  File "/usr/local/lib/python2.0/site-packages/mod_python/apache.py", line
484, in write
    h, v = string.split(line, ":", 1)

ValueError: unpack list of wrong size
----------------------------------------------------------------------------
-----------------------------





What seems to be the problem?..
Thanks in advance. :)
p.s) If this news group is not proper to post this message, plz let me
know. -- with suitable news group. :)





From kens at sightreader.com  Thu Mar 15 16:02:05 2001
From: kens at sightreader.com (Ken Seehof)
Date: Thu, 15 Mar 2001 15:02:05 -0600
Subject: Should I close after popen??
References: <XFMail.20010315142704.shaleh@valinux.com>
Message-ID: <003201c0ad93$30f982f0$04090a0a@upcast.com>

----- Original Message -----
From: "Sean 'Shaleh' Perry" <shaleh at valinux.com>
To: "Andrew Markebo" <flognat at flognat.myip.org>
Cc: <python-list at python.org>
Sent: Thursday, March 15, 2001 4:27 PM
Subject: RE: Should I close after popen??


>
> On 15-Mar-2001 Andrew Markebo wrote:
> > I am hacking away a small background program, using popen to call
> > commands, should I close the file in some way when done or it is taken
> > care of automagically??
> >
> > I do something like this once every 10 mins..
> >
> >         fd1=os.popen("command")
> >         fd2=os.popen("command")
> >         os.wait(fd1)
> >         os.wait(fd2)
> >         values=fd1.read()+fd2.read()
> >
> > And.. anything to worry about, nesting popen/wait like this??
> > (Linux/Solaris)
> >
>
> if that is in a loop, it should be fine:
>
>   while 1:
>      f = open(file) # every time the loop repeats, f gets closed and
opened
>                     # again
>      # do things to f
>
I don't think that's reliable in general.  I'm pretty sure the file gets
closed when
the file object gets finalized, but you aren't supposed to count on garbage
collection to do its job at any particular time.  The while loop above could
in
theory be equivalent to the folowing:

f0 = open(file)
f0.close()
f1 = open(file)
f2 = open(file)
f3 = open(file)
f1.close
f4 = open(file)
f3.close()
f2.close()
f4.close()

...because the objects can be deleted whenever the interpreter feels
like it.  In practice, you'll probably get by with it, but you are setting
yourself up for nasty bugs later.

In other words, you are better off closing files explicitly.

In the case of popen, it seems to me like it wouldn't matter if you
leave a few open streams lying around, but it just seems sloppy,
like not washing your dishes after dinner.

Your usage of wait is fine, (assuming that the fd1 process is never
dependent on the fd2 process).

> --
> http://mail.python.org/mailman/listinfo/python-list
>




From gregj at pobox.com  Sun Mar 11 16:36:52 2001
From: gregj at pobox.com (Gregory Jorgensen)
Date: Sun, 11 Mar 2001 21:36:52 GMT
Subject: Einstein's Riddle
References: <oNPq6.396644$ge4.138479566@news2.rdc2.tx.home.com>
Message-ID: <UdSq6.2184$54.2166@www.newsranger.com>

The 2% that can solve it must be psychic or deranged. The riddle as posed is not
solvable.

We are told of five houses anf five people. The hints mention five colors (red,
green, white, yellow, blue), five drinks (tea, coffee, milk, beer, water), five
smokes (Pall Mall, Dunhill, Blends, Bluemaster, Prince), and five nationalities
(Brit, Swede, Dane, Norwegian, German). But only four types of pet are
mentioned: dogs, birds, cats, horse. We have no evidence that any of the five
owns fish.

More telling: how many people can't quickly determine that this puzzle is
unsolvable?


In article <oNPq6.396644$ge4.138479566 at news2.rdc2.tx.home.com>, Steve Graham
says...
>
>Who would be interested in using his/her brain (and his computer) to solve
>the following?
>
>
>Steve Graham
>
>===
>
>Einstein's Riddle
>
>Albert Einstein wrote this riddle this century [ed. 20th century].  He said
>98% of the world could not solve it.
>
>There are 5 houses in 5 different colors.  In each house lives a person with
>a different nationality.  The 5 owners drink a certain type of beverage,
>smoke a certain brand of cigar, and keep a certain pet.  No owners
>have the same pet, smoke the same brand of cigar or drink the same beverage.
>
>The question is: "Who owns the fish?"
>
>Hints:
>
>The Brit lives in the red house.
>
>The Sweed keeps dogs as pets.
>
>The Dane drinks tea.
>
>The green house is on the left of the white house.
>
>The green house's owner drinks coffee.
>
>The person who smokes Pall Mall rears birds.
>
>The owner of the yellow house smokes Dunhill.
>
>The man living in the center house drinks milk.
>
>The Norwegian lives in the first house.
>
>The man who smokes Blends lives next to the one who keeps cats.
>
>The man who keeps the horse lives next to the man who smokes Dunhill.
>
>The owner who smokes Bluemasters drinks beer.
>
>The German smokes Prince.
>
>The Norwegian lives next to the blue house.
>
>The man who smokes Blends has a neighbor who drinks water.

Greg Jorgensen
Deschooling Society
Portland, Oregon, USA
gregj at pobox.com


From kuncej at mail.conservation.state.mo.us  Tue Mar 20 16:12:33 2001
From: kuncej at mail.conservation.state.mo.us (Jeff Kunce)
Date: Tue, 20 Mar 2001 15:12:33 -0600
Subject: (no subject)
References: <mailman.985042764.18051.python-list@python.org>
Message-ID: <LLPt6.905$%A.1353@news.more.net>

> I want to execute an executable binary as thought through the
Command-Line.

in addition to os.system(),
take a look at win32api.ShellExecute()

  --Jeff





From hgg9140 at seanet.com  Sat Mar 17 22:47:14 2001
From: hgg9140 at seanet.com (Harry George)
Date: 17 Mar 2001 19:47:14 -0800
Subject: installing 2.x over the top of 1.5.2
References: <mailman.984861502.27092.python-list@python.org>
Message-ID: <m33dcbivr1.fsf@fred.local>

Don't know if this qualifies as clean.  I have the rpm'd 1.5.2 under
/usr.  Then I installed py 2.x under /usr/local.  During our
(extended) transition period, we use shell scripts which set
PYTHONHOME and PYTHONPATH before calling python1.5 or
python2.0.  Once all is well in 2.x, we'll remove the rpm'd 1.5.2, and
(conceivably) install 2.x into /usr.

Timothy Grant <tjg at exceptionalminds.com> writes:

> This problem has been bothering me for a while.
> 
> Python v1.5.2 is installed by default on RH6.2 boxes. 
> 
> My problem is that upgrading an RH6.2 box by using the 2.x rpms
> now means that there are two versions of Python installed on my
> box, and any extra modules I have installed have to be copied
> to the new location for v2.x. I then need to uninstall the
> v1.5.2 rpms.
> 
> I'm sure someone somewhere has come up with a clean way of
> upgrading Python. As with all thing Pythonic, there must be an
> easier way.
> 
> -- 
> Stand Fast,
>     tjg.
> 
> Timothy Grant                         tjg at exceptionalminds.com
> Red Hat Certified Engineer            www.exceptionalminds.com
> Avalon Technology Group, Inc.         <><       (503) 246-3630
> >>>>>>>>>>>>>Linux, because rebooting is *NOT* normal<<<<<<<<<
> >>>>This machine was last rebooted:  60 days     4 hours ago<<


From bwalker at earth1.net  Mon Mar 26 15:14:05 2001
From: bwalker at earth1.net (Bill Walker)
Date: Mon, 26 Mar 2001 15:14:05 -0500
Subject: Some Basic Questions about Python
Message-ID: <tbv8g06eg66tec@corp.supernews.com>

I am fairly new to OOP, but I have done quite a bit of programming in
Dos Basic a few years back. I've tried Java -- forget that. Perl too.

I bought the book "Core Python Programming" by Wesley Chun. It had
Python 1.5 on the CD-ROM that came with the book as well as TK graphics.

I have successfully written a Python program to display the graph of a sine
wave in a Tk window. I like the syntax.

Now for the questions:

1) If I write an application and want to pass it on to others who may not
have Python installed, how do I do this? I understand there is no compiler
for
Python and Tk yet that generates .exe files.

2) Can I distribute the Python install file along with my app without
violating a copyright or license?

3) Better yet, can I include only the base files needed to run my
application with the source code? Say Pythonw.exe and only the appropriate
DLLs, libs, etc.

I tried to do #3 above and ran into trouble. It is hard to determine just
what
dlls, libs and the like are actually needed, even for a simple program that
uses
only the Tk and math modules.

4) Is there docs somewhere that tells one what files are needed for what
modules?






From mcherm_python at yahoo.com  Tue Mar 20 11:12:52 2001
From: mcherm_python at yahoo.com (Michael Chermside)
Date: Tue, 20 Mar 2001 10:12:52 -0600
Subject: New PEP: The directive statement
Message-ID: <3AB78184.5060409@yahoo.com>

> Please have a look at the PEP below.

Martin:

After all the heated discussion going on here, I'm glad SOMEONE
finally responded to Tim's request to write it up as a PEP. Now,
as Tim points out, you'll probably hear lots of complaints about
why "directive" is a poor idea. But thanks for writing it up...
we can't have "from __future__ import" being the only target,
now can we? <wink>

Seriously, it looks well-written, TRIES (I think) to incorporate
the vast majority of what people have said on the subject so 
far, and as Tim pointed out, it was high time it got written.

-- Michael Chermside



_________________________________________________________
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com




From tim.hochberg at ieee.org  Tue Mar 20 16:12:58 2001
From: tim.hochberg at ieee.org (Tim Hochberg)
Date: Tue, 20 Mar 2001 21:12:58 GMT
Subject: Insertin **keywords into a class
References: <mailman.984971990.29857.python-list@python.org> <9960i1$tp0$1@nntp6.u.washington.edu> <99661m$scc$1@panix6.panix.com> <99877b$uss$1@nntp6.u.washington.edu> <mailman.985114465.27377.python-list@python.org>
Message-ID: <uJPt6.45663$Xt3.6834656@news1.rdc1.az.home.com>

"C. Porter Bassett" <porter at et.byu.edu> wrote :
> cCan you please tell me why the following code snippet doesn't update the
> value of radius?

Because you are updating self.kw instead. Try:

   setattr(self, kw, keywords[kw])

instead of "self.kw = keywords"

I'll make a bit of attempt to expain this, but no gaurantees that it'll be
helpful. "self.kw = keywords[kw]" in this case is translated into
"setattr(self,  'kw', keywords[kw])". Note the quotes around the first kw.
You don't want that. The way you do it now, sets self.kw is set sucessively
to each value in keywords. You might want to throw a "print self.kw" into
your code after the "self.kw =..." line just to see what I mean.

Incidentally, I think you can accomplish the same things with:

 class myClass:
    def __init__(self, *arguments, **keywords):
       self.radius = 0.0
       self.__dict__.update(keywords)
       print "self.radius =", self.radius

But I haven't tried it.

-tim

> class myClass:
>    def __init__(self, *arguments, **keywords):
>       self.radius = 0.0
>       for kw in keywords.keys():
>          print kw, ":", keywords[kw]
>          self.kw = keywords[kw]
>       print "self.radius =", self.radius
>
>
> b = myClass(radius = 1.0)
>
>
> --------------------------------------------------------------------------
> "Pretend like this is a really witty saying." - Anonymous
> --------------------------------------------------------------------------
>
>
>




From j_wesselink at chello.nl  Wed Mar 21 19:24:47 2001
From: j_wesselink at chello.nl (Wieger Wesselink)
Date: Thu, 22 Mar 2001 00:24:47 GMT
Subject: starting Java VM using JNI?
Message-ID: <3AB9487C.F528E426@chello.nl>

I'm trying to start a Java VM in Python using JNI, with the following
C++ code:

  void start()
  {
    JNIEnv *env;
    JavaVM *jvm;
    JavaVMInitArgs vm_args;
    JavaVMOption options[2];
    options[0].optionString = "-Djava.class.path=D:\\java";
    vm_args.version = 0x00010002;
    vm_args.options = options;
    vm_args.nOptions = 1;
    vm_args.ignoreUnrecognized = JNI_TRUE;
    jint res = JNI_CreateJavaVM(&jvm, (void**)&env, &vm_args);
    if (res < 0) {
      exit(1);
    }
  }

That appears to work, but there is a problem. After invoking the
function
start() in Python, the interepreter generates syntax errors after each
subsequent input (see below). Can someone tell me how to fix this, or is
there a better way to start the JVM?

Python 2.0 (#8, Oct 16 2000, 17:27:58) [MSC 32 bit (Intel)] on win32
Type "copyright", "credits" or "license" for more information.
>>> import mymodule
>>> mymodule.start()
>>>
  File "<stdin>", line 1

    ^
SyntaxError: invalid syntax
>>>

Best regards,

Wieger


From nospam at mega-nerd.com  Thu Mar 15 14:38:12 2001
From: nospam at mega-nerd.com (Erik de Castro Lopo)
Date: Thu, 15 Mar 2001 19:38:12 GMT
Subject: Tkinter : resizing Text widget with main window
Message-ID: <3AB11827.469F6C0C@mega-nerd.com>

Hi all,

How does one force a Tkinter.Text widget to be resized whenever 
the main window is resized?

TIA,
Erik
-- 
-----------------------------------------------------------------
Erik de Castro Lopo     nospam at mega-nerd.com      (Yes its valid)
-----------------------------------------------------------------
IDIOT, n. 
  A member of a large and powerful tribe whose influence in human 
  affairs has always been dominant and controlling. The Idiot's 
  activity is not confined to any special field of thought or action, 
  but "pervades and regulates the whole." He has the last word in 
  everything; his decision is unappealable. He sets the fashions and 
  opinion of taste, dictates the limitations of speech and 
  circumscribes conduct with a dead-line. 
            Ambrose Bierce


From tim.one at home.com  Sat Mar 10 18:03:45 2001
From: tim.one at home.com (Tim Peters)
Date: Sat, 10 Mar 2001 18:03:45 -0500
Subject: Memory leaks in VC6 2 line Python 2.1b1 apps?
In-Reply-To: <98b0a0$2g8i$1@shadow.skypoint.net>
Message-ID: <LNBBLJKPBEHFEDALKOLCGEMFJEAA.tim.one@home.com>

[Craig Holman]
> I'm afraid that both versions 2.0 and 2.1b1 leak memory for a single
> Py_Initialize, Py_Finalize sequence in a Visual C++ MFC application.

The bad news is that there are legit bugs in 2.0 here.  The good/bad news is
that it's got nothing to do w/ MFC (it's much the same on all platforms).
The bad/good news is that 2.1b1 appears to be "leaking" exactly as much as
it's designed to "leak".

> ...
> Python 2.1b1 had 21 memory leaks that resulted in a leakage of a
> total of 6,899 bytes for a single pair of Py_Initialize, Py_Finalize
> calls.  Additional Py_Intitialize, Py_Finalize pairs did not result
> in any additional leakage (10,000 pairs were tested).

Then it's functioning as designed.  Python has no protocol to let modules
know that Python is shutting down, so any number of file statics "leak" at
shutdown.  However, modules are also written so as to reuse those guys at the
next Py_Initialize, so they're "leaks" in a mostly academic sense; i.e., the
cost of hunting all of them down is thought higher than the benefits of
killing them.  Besides, we could well end up with more than 7Kb of runtime
code to track down and nuke that 7Kb of data you're seeing!  So long as
additional (beyond the first) Init/Fini pairs don't systematically leak
additional memory, we're happy enough.

> ,,,
> I'm hoping that some of these leaks can be addressed before the 2.1
> release is issued.

It's unlikely unless somebody submits "obviously safe" patches.  Most leaks
now appear due to interned strings, and I'll push a bit to get those cleaned
up (they already are in 2.1b1, but only when Barry is running *his* leak
tests under Insure++).

the-question-is-how-python-knows-when-barry-is-running-it-ly y'rs  - tim




From aahz at panix.com  Mon Mar 19 18:48:55 2001
From: aahz at panix.com (Aahz Maruch)
Date: 19 Mar 2001 15:48:55 -0800
Subject: Why "from __future__" stinks; a counter-offer
References: <mailman.985029808.7088.python-list@python.org> <memo.20010319213204.8771C@digitig.compulink.co.uk>
Message-ID: <9965t7$s57$1@panix6.panix.com>

In article <memo.20010319213204.8771C at digitig.compulink.co.uk>,
Tim Rowe <tim at digitig.cix.co.uk> wrote:
>
>But there's part of the problem; egoless PEP writing is as hard as
>egoless programming, and having come up with a solution the author
>of that solution is bound to have some emotional involvement in that
>solution.  There have been (rightly IMHO) howls of protest about
>__future__, but Tim has said it's ok for him, routinely dismisses all
>protests, presumably because he genuinely doesn't understand what our
>problem is, and he tells us it's going in anyway. Can you see why we're
>not all convinced that the PEP process is any help to us?

I think that of the various people to accuse of having an ego tied up in
something, the Timbot is one of the last that I would choose.  No, it's
more that Uncle Timmy is *lazy*.  Thus, to get his interest, you have to
appeal to his laziness.

So far, no other person has actually come up with a scheme that addreses
the problem at hand, as opposed to simply whining, suggesting minor and
worthless variations, or putting forth even *more* complicated solutions.
One of the nice things about this solution is that if it is decided to be
an ugly wart, it can be easily dropped for Python 2.2.
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

There's a difference between a person who gets shit zie doesn't deserve
and a person who gets more shit than zie deserves.  --Aahz


From rburnham at cri-inc.com  Fri Mar  2 19:38:12 2001
From: rburnham at cri-inc.com (Roger Burnham)
Date: Sat, 03 Mar 2001 00:38:12 GMT
Subject: Communicating with serial port
References: <ndNn6.5161$t21.152593@news3.oke.nextra.no>
Message-ID: <3aa039a6.210669196@news-server.socal.rr.com>

On Fri, 2 Mar 2001 14:16:58 +0100, "Peter Blom" <pet-blo at online.no> wrote:

>I shall write a python program for downloading data from a memory card in a
>measuring instrument, via a serial port.
>For this purpose, I have downloaded SioModule.zip - a serial comm extension
>for windows 95, upgraded to Python 1.5.1.
>This does not seem to work - when importing the module 'serial.py', my
>system hangs (Windows 98, Activestate Python 2.0).
>Could anybody give a hint?
>
>Peter (a newbie)
>
>

Hi,

I'm the SioModule author.  While I deal with getting my starship account back
active (where this installer will soon be), I've sent you via email the
installer that will work with 2.0.

Cheers,
http://starship.python.net/crew/roger/


Roger Burnham
rburnham at cri-inc.com


From claird at starbase.neosoft.com  Wed Mar 14 11:28:26 2001
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 14 Mar 2001 10:28:26 -0600
Subject: Who's minister of propaganda this week?
References: <B98041329020E007.7DA84CAD9D0562AB.817BBC572F480869@lp.airnews.net> <98nu1m$12e@dispatch.concentric.net>
Message-ID: <660ACCA1E4816F7E.C359CDF4168EF3FB.3B6B05538D619770@lp.airnews.net>

In article <98nu1m$12e at dispatch.concentric.net>,
Phlip  <phlip_cpp at my-deja.com> wrote:
			.
			.
			.
>Robert C. Martin meant two things. A> If you write wall-to-wall unit tests 
>when you code, if you use a typeless language you will add bullet-proof 
>features faster than if you use a statically typed language, even though 
>the former has less built-in error checking at type conversion time. The 
>latter supports refactoring easier.
			.
			.
			.
Along with this, there's an important *empirical* aspect
that I regard as inadequately appreciated.  Uche recently
expressed it this way:
  Type errors are not even close to the majority of
  those I make while programming in Python, and I'm
  quite certain that the code I've written in Python
  is much less buggy than code I've written in
  strongly-typed languages.  Expressiveness, IMO, is
  a far better aid to correctness than artificial
  restrictions (see Java for the example of school-
  marm programming gone amok).
It's OK to think, "gosh, with strict typing we save errors,
and with Hungarian notation we save errors, and ..."  We've
done enough experiments by now, though, to convince at least
me that it is exactly true that
  Expressiveness ... is a far better aid to
  correctness ...

I leave to another time fine points about how there are more
advanced systems of strict typing (Dylan, for example), and
about other dimensions in which managerially-accepted languages
differ from Smalltalk, Python, and Ruby; I just want to empha-
size for now that, among the few reasonably certain results
Software Engineering has given us, the dominance of expressive-
ness is one of the most important.

Myself, I prefer to "write wall-to-wall unit tests" *before* I
code.
-- 

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


From mwh21 at cam.ac.uk  Mon Mar  5 15:13:20 2001
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 05 Mar 2001 20:13:20 +0000
Subject: Static class methods
References: <rdzo6.154$gc4.80445@news.pacbell.net>
Message-ID: <m3wva4as9r.fsf@atrus.jesus.cam.ac.uk>

"Alex Shindich" <alex at shindich.com> writes:

> Just out of my curiosity, have any of you ladies and gentlemen
> experienced a need for an equivalent of C++/Java static method?

Here's a particularly nasty way to do what you want.

class C:
    class _Temp:
        def __call__(self, arg):
            print arg
    static = _Temp()
    del _Temp

    def call_static(self):
        self.static("from meth")

Then:

>>> C.static("from cli")
from cli
>>> C().static("from cli 2")
from cli 2
>>> C().call_static()
from meth

Cheers,
M.

-- 
  There are two kinds of large software systems: those that evolved
  from small systems and those that don't work.
                           -- Seen on slashdot.org, then quoted by amk


From sekter at matavnet.hu  Fri Mar  9 05:31:51 2001
From: sekter at matavnet.hu (Arpad Kiss)
Date: Fri, 09 Mar 2001 11:31:51 +0100
Subject: OODB question
References: <WM1q6.7817$Z%5.6737@NewsReader>
Message-ID: <3AA8B117.1050107@matavnet.hu>

No, you did't dream. Zope has an OODB named ZODB. Look for it on 
www.zope.org or on www.amk.ca/zodb/ there are some good docs and a 
separate distribution (without Zope).
Arpad


chris lamb wrote:

> I am sure that I read somewhere of an OODB implementation developed
> 
> specifically with Python or having a Python interface.
> 
> 
> 
> I am beginning to wonder if I dreamt it up since I cannot remember where I
> 
> read about it!
> 
> 
> 
> Can anyone help me ? (I doubt it!)
> 
> 
> 
> Best wishes
> 
> 
> 
> Chris


-- 
"Don't try to teach a pig to sing;
it wastes your time and it irritates the pig."
Awareness by Anthony de Mello



From rickylee at americasm01.nt.com  Wed Mar  7 12:55:17 2001
From: rickylee at americasm01.nt.com (Lee, Rick)
Date: Wed, 07 Mar 2001 12:55:17 -0500
Subject: Setting interactive flag inside a script?
Message-ID: <3AA67605.B04B1135@americasm01.nt.com>

Is there a way to set the python interactive flag, ie. the '-i' flag in
the command line, inside a script itself?  (Motivation: Sometimes you
give someone else a .py script, s/he doubleclicks on it, something may
have gone wrong, and the command line window just disappears. If the
script can turn the interactive flag on, then the window won't just
disappear by default.  When the script finishes properly, it can turn
the flag off just at the very end.)

- Rick Lee



From pearu at cens.ioc.ee  Thu Mar 22 16:44:31 2001
From: pearu at cens.ioc.ee (Pearu Peterson)
Date: Thu, 22 Mar 2001 23:44:31 +0200
Subject: PEP 240 scares me
Message-ID: <Pine.LNX.4.21.0103222338530.29360-100000@kev.ioc.ee>


Hi!

I very much agree what is in PEP 240 Rationale, but its Proposal scares me
a lot, its like a nightmare (and I actually don't see any logical
connection between the Rationale and the Proposal).
Let me explain why.

It is true that 1/2 == 0 surprise new (Python) programmers (but I quickly
learned to write 1./2 or 1/2. or 1/float(n) etc). 
But making 1.2 to be a rational
                                1
                               ---
                                2
will most definitely surprise new (Python) programmers even more *because*
in ever programming language I know of (C, Fortran, Maple, Mathematica, Matlab
to name few), use 1.2 to denote a float that is equal to
                                6
                               ---
                                5
(up to a computers ability to represent such a number in float).

I apologize if my message is too emotional but after reading the PEP 240
Proposal, I felt very uncomfortable.
 
I think that any of the following representation of rational numbers are
much more appropriate, though none of them is perfect:
         1//2
         1r/2         (like 1j+2 is a complex number)
         1/2r         (like 1+2j)
         1_2
         1:2          (that's probably tricky in Python)

Note that they all raise SyntaxError in current Python, so, just pick one
and there will be no backwards compatibility problems. Using 1.2 for a 
`half' will probably break any Python code that does
(scientific) numerical calculations.

In conclusion, I will look forward for Python to support rational
numbers. Though floats are always imperfect in computers representation,
somebody in c.l.py even said hating floats. I don't love them either but
I *do* need them. Let's not make things worse than they already are.

Regards,
	Pearu




From ccraig at ccraig.org  Wed Mar 28 11:08:25 2001
From: ccraig at ccraig.org (Christopher A. Craig)
Date: 28 Mar 2001 11:08:25 -0500
Subject: Speed problems with Python vs. Perl
References: <99sv33$8hk$1@news.cineca.it> <1znw6.1969$sk3.665386@newsb.telia.net>
Message-ID: <87y9tplvuu.fsf@elbereth.ccraig.org>

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

> assuming you're using Python 2.1, the following version is
> about 16 times faster on my box:
> 
> import sys
> 
> def main():
> 
>     icount = 0
>     for line in sys.stdin.xreadlines():
>         icount += 1
>         f = line.split()
>     print "Total lines read", icount
> 
> if __name__ == '__main__':
>     main()
> 

You can't necessarily blame this on the line I/O performance because
you used line.split() where he used the split method on a regular
expression object.  Testing just the differences in those two I show a
10 fold improvement on line.split().  

In this case that is valid since you are splitting on whitespace in
both Perl and Python, but it would be interesting to see if Python
still holds up under more complex regular expressions.  

I went ahead and used

--- Python Script ---

import re
import sys

r = re.compile(":+")

def main():
    icount = 0
    while 1:
        lines = sys.stdin.readlines(50000)
        if not lines:
                break
        for line in lines:
                icount += 1
                f = r.split(line)

if __name__=="__main__": main()

--- End Python Script ---

vs.

--- Perl Script ---
$icount = 0;

while(<>) {
  $icount++;
  @f = split(/:+/);
}

--- End Perl Script ---


with Python 2.0 and Perl 5.6.0 on a Celeron 300A using a 242000 line
input file and I got the following results:

python: 61.10s user 0.42s system 87% cpu 1:10.37 total
perl:   11.31s user 0.10s system 76% cpu 14.960 total

admittedly Python is being hurt by the slower 2.0 line I/O, but I
would guess that regular expressions are still hurting performance.



-- 
Christopher A. Craig <ccraig at ccraig.org>
"Parity is for farmers." Seymour Cray on his machines lack of parity
"I guess farmers buy a lot of computers." Seymour Cray on including parity


From phd at phd.pp.ru  Mon Mar  5 09:31:26 2001
From: phd at phd.pp.ru (Oleg Broytmann)
Date: Mon, 5 Mar 2001 17:31:26 +0300 (MSK)
Subject: Open Source
In-Reply-To: <9806i4$2e97$1@news3.infoave.net>
Message-ID: <Pine.LNX.4.30.0103051730160.13477-100000@fep132.fep.ru>

On Mon, 5 Mar 2001, Alton Luder III wrote:
> Is there any place on the web that offers open source python programs. It
> might be helpful for me to learn if I see other people's program's source
> code.

   Welcome! There are a lot. Run through Parnasus (vex.net/parnasus) and
downlaod what you may find interesting. Parnasus is not downloading site,
just a catalog of links, but pretty good catalog.

Oleg.
----
     Oleg Broytmann            http://phd.pp.ru/            phd at phd.pp.ru
           Programmers don't die, they just GOSUB without RETURN.




From paulp at ActiveState.com  Wed Mar 14 09:49:34 2001
From: paulp at ActiveState.com (Paul Prescod)
Date: Wed, 14 Mar 2001 06:49:34 -0800
Subject: Who's minister of propaganda this week?
References: <B98041329020E007.7DA84CAD9D0562AB.817BBC572F480869@lp.airnews.net> <98nu1m$12e@dispatch.concentric.net>
Message-ID: <3AAF84FE.F28A2C30@ActiveState.com>

Phlip wrote:
> 
> ...
> 
> B> The full depth of behaviors we should expect from a typeless language
> have yet to be fathomed. Ruby trumps Python in the simple matter of
> closures (imagine my shock when I discovered Python didn't have them). But
> a language "like" those might win the decade.

I don't know if you are serious (I mean closures are NOT that
important!) but anyhow, Python 2.1 has nested scopes and I think that
functions defined in those scopes are closures just like in any other
language.

-- 
Python:
    Programming the way
    Guido
    indented it.
       - (originated with Skip Montanaro?)



From wunschname at kapet.de  Fri Mar 23 12:25:23 2001
From: wunschname at kapet.de (Karsten Petersen)
Date: Fri, 23 Mar 2001 18:25:23 +0100
Subject: Python MPI
References: <mailman.985354471.4871.python-list@python.org>
Message-ID: <3u0g99.vn3.ln@dollerup.csn.tu-chemnitz.de>

Eugene Leitl wrote:
> I've just came back from a 3 day MPI course, and promptly noticed that
> I can use Python with MPI (yay!):
> 	http://www.fysik.dtu.dk/~schiotz/comp/PythonAndSwig/pythonMPI.html
Well, that only describes how to use Python with C-modules using MPI.
I found it even more interesting to call MPI functions directly from
Python - thats possible with e.g. the MPI-module contained in
ScientificPython 2.1.
	http://starship.python.net/crew/hinsen/scientific.html
The above link also describes, that you have to patch your Python
executable.  But my tests with lam-mpi and mpich had shown that it is
not really needed - they finish themself cleanly even if MPI_Finalize is
not called.

> Any MPI Python users here? I'll be interested in hearing your experiences.
I used Python with the MPI-Module from Scientific-Python 2.1 during the
last months to do my MPI-coding homework.  Currently I am trying to
write a parallizing fractal-generator in Python using MPI, NumPy and
PyGTK...

Greets, Karsten
-- 
,-,  Student of Computer Science at Chemnitz University of Technology  ,-,
| |    EMail:  Karsten at kapet.de          WWW:  http://www.kapet.de/    | |
'-'  Home: kapet at dollerup.csn   V72 / 230    Phone: +49-177-82 35 136  '-'


From cce at clarkevans.com  Tue Mar 20 00:39:34 2001
From: cce at clarkevans.com (Clark C. Evans)
Date: Tue, 20 Mar 2001 00:39:34 -0500 (EST)
Subject: Functionalism, aesthetics Was:(RE: I come to praise .join,  not
In-Reply-To: <5.0.2.1.0.20010319224910.021134d0@mail.inet.com.br>
Message-ID: <Pine.LNX.4.21.0103200026570.21343-100000@clarkevans.com>

> I'm lost for words... This summarizes a lot of concerns that sometimes I 
> find so hard to explain. Anyone who has actually *read* Christopher 
> Alexander's books on design should know that the best solutions are the 
> ones that take into account the inner sense of satisfaction of their end 
> users. As Alexander puts it itself, it is a rewarding experience born out 
> of the "completeness" of the solution. The best solutions also share 
> another common trait: they look *and feel* good to the untrained eye. 
> Beware of works that can only be appreciated by experts - they are way too 
> trained to see the obvious.

CA has a nice talk at OOPSLA a few years back, '97 I believe, where
he really got into these kind of items.  The speech challenged 
programmers to take an ethical/humane approach to software 
construction.  As a programmer I really didn't clue in to what
he was talking with.  Recent work at a local museum has made me
see better, the software they use is deeply disturbing -- the 
antithesis of aestetic concern.  Such a sharp contrast with the
beautyful objects on the table less than a meter away.

Mike C. Fletcher wrote:
> The reason I felt it necessary to jump in here again is the implicit
> creation of a "inner circle" of the "initiated" who are considered
> "worthy" to contribute to a discussion about the evolution of the
> language.

In my experience, Inner circles tend to be self serving and I
avoid them at all costs.  Meritocracy works much better.

> The users of a system, whether it is a building, a language, a piece of
> software, or a chair, need to be heard.  A user complaining that something
> is "ugly", "bad" or "wrong" needs the nature of their discomfort
> ascertained, and that knowledge integrated into further design decisions.

Now, a user community is a horse of a different color.  And, indeed,
I would hope that a strong user community emerges with a voice 
equal to the implementers.  However, this maturity has yet to come
to the software world -- I think it will take many many years.

> Designers who ignore their users, assuming that their own 
> understanding of the system is correct, and that anyone who 
> is unable to speaking the dialect of designers cannot possibly
> contribute to a design, will create monstrosities which are next to
> impossible to inhabit.

Hear Hear!  If only you would broadcast your post on the xml-dev
mailing list... XML Schema is an abomination.

> Its unfortunate that in many cases the non-expert user keeps quiet, 
> assuming that whatever opinion the "master" haves, its going to be better 
> than theirs - after all they are the so-called experts.

Well, this is a problem with open source in general.  Most 
coders have the attitude: "If you can't code, get out of
the kitchen".  Sad, but true.

Clark






From jmarshal at mathworks.com  Thu Mar  1 13:01:09 2001
From: jmarshal at mathworks.com (Joshua Marshall)
Date: 1 Mar 2001 18:01:09 GMT
Subject: A counter-proposal to __future__ in PEP 236
References: <j47l29lpo8.fsf@informatik.hu-berlin.de>
Message-ID: <97m2p5$3au$2@news.mathworks.com>

Martin von Loewis <loewis at informatik.hu-berlin.de> wrote:
...
> In this implementation, directive is only considered as a keyword if
> it appears at the beginning of the module. Therefore, code that
> happens to use directive as an identifier will not break (unless it
> assigns to directive as the first thing in the module).

Note that it doesn't need to be implemented this way.  Currently,

  directive foo

is a syntax error.  Meaning can be assigned to this syntax without
breaking any existing code--it is unambiguous.


From tjreedy at udel.edu  Mon Mar 12 17:00:09 2001
From: tjreedy at udel.edu (Terry Reedy)
Date: Mon, 12 Mar 2001 17:00:09 -0500
Subject: This math scares me
References: <3aad4062.1290021@News.CIS.DFN.DE>
Message-ID: <98jgij$si1$1@news.udel.edu>

<costas at springmail.com> wrote in message
news:3aad4062.1290021 at News.CIS.DFN.DE...
> Ok, I can see maybe division having problems. But why does addition of
the two numbers below:
>
> 5.01+5.54
>
> give me this?
>
> 10.550000000000001

Neither 5.01, 5.54, nor 10.55 have exact binary representations.  Floating
point arithmetic is always slightly approximate unless you only use
'binary' fractions like .5, .25, .125, .0625, etc and multiples.
Understanding and controlling such inexactness is a large part of the
discipline of numerical analysis.

> I hope its some setting that can take care of this. I am using Python
under Windows 98.

You can hide the inexactness by displaying results with % formats.

Terry J. Reedy





From erno-news at erno.iki.fi  Tue Mar 27 03:24:03 2001
From: erno-news at erno.iki.fi (Erno Kuusela)
Date: 27 Mar 2001 11:24:03 +0300
Subject: HTTP redirect via Python?
References: <SNVv6.128157$tP3.2040832@news1.rdc1.bc.home.com>
Message-ID: <kug0fzwrfg.fsf@lasipalatsi.fi>

In article <SNVv6.128157$tP3.2040832 at news1.rdc1.bc.home.com>, "John
E. Barham" <bezilla at hotmail.com> writes:

| Apologies if this is too basic a question, but I'm wondering how I can send
| a redirect/URL rewrite command to a browser via Python.

| The context:  we're developing a fairly substantial Web-based application
| and using Python for the CGI coding (which is working beautifully, BTW).  In
| order to partially protect ourselves (and the users ;) against
| back/forward/refresh with forms, we'd like to send them to a "safe" URL
| after a form has been submitted.  How can we do this?  There doesn't appear
| to be anything to do this in module cgi...

| If it's a server dependent thing, we're using Apache.

def print_redirect(url):
        print 'Status: 302 Moved Temporarily'
        print 'Location:', url
        print 'Pragma: no-cache'
        print 'Content-Type: text/html'
        print
        
        print '<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML//EN">'
        print '<title>Redirect (302)</title>'
        print '<h1>302 Moved Temporarily</h1>'
        print '<p>The answer to your request is located at'
        hurl = cgi.escape(url, 1)
        print '<a href="%s">%s</a>.' % (hurl, hurl)

note that the url must be absolute.

  -- erno


From syver at NOSPAMcyberwatcher.com  Wed Mar 14 17:05:39 2001
From: syver at NOSPAMcyberwatcher.com (Syver Enstad)
Date: Wed, 14 Mar 2001 23:05:39 +0100
Subject: windows script host and python
References: <%tKr6.2570$pX.26174@news1.onlynews.com>
Message-ID: <98oq3r$1e0$1@troll.powertech.no>

"Mike Carifio" <carifio.nospam at nospam.usys.com> wrote

> I'm looking for some resources on using python with wsh generally.

Just curious but what do plan to acheive by using python with wsh that you
couldn't do using normal .py files?




From ransen_spam_me_not at nemo.it  Wed Mar  7 13:02:54 2001
From: ransen_spam_me_not at nemo.it (Owen F. Ransen)
Date: Wed, 07 Mar 2001 18:02:54 GMT
Subject: tokenizer.c crashes in 2.1b1
References: <3aa34d5c.1284101@news.newsguy.com> <ulmqiobyp.fsf@roundpoint.com> <3aa5d493.5922346@news.newsguy.com> <984q1m0skr@news1.newsguy.com>
Message-ID: <3aa76191.1071940@news.newsguy.com>

On Wed, 7 Mar 2001 09:03:58 +0100, "Alex Martelli" <aleaxit at yahoo.com>
wrote:

>"Owen F. Ransen" <ransen_spam_me_not at nemo.it> wrote in message
>news:3aa5d493.5922346 at news.newsguy.com...
>> On 06 Mar 2001 12:54:06 -0800, Ben Hutchings
>> <ben.hutchings at roundpoint.com> wrote:
>>
>> >> The debugger points to the guilty party:
>> >> tokenizer.c, the fgets call at line 247 fails.
>> >> My VC project uses Multithreaded Debug compiler settings,
>> >> as does the Pythoncore project.
>    [snip]
>> Debug Multithreaded in my MFC app make settings
>> Debug Multithreaded DLL in the Pythoncore DLL make settings
>>
>> Is this wrong?
>
>Yes:

Oops! Thanks for the pointer...




--
Owen F. Ransen
http://www.ransen.com/
Home of Gliftic & Repligator Image Generators


From costas at meezon.com  Sat Mar 10 23:53:01 2001
From: costas at meezon.com (Costas Menico)
Date: Sun, 11 Mar 2001 04:53:01 GMT
Subject: import headaches/gripes
References: <3aaa3f94.15940553@News.CIS.DFN.DE> <mailman.984250586.21661.python-list@python.org>
Message-ID: <3aaafcdd.344477@News.CIS.DFN.DE>

Paul Prescod <paulp at ActiveState.com> wrote:

>> Why can't the import statement be smart enough to check if a file was
>> changed to import it anyway? All it has to do is compare .py to the
>> last import date/time.
>
>I think there are two reasons:
>
> 1. Even checking file dates has a performance cost. It involves going
>to the hard disk after all. Code like this isn't rare in Python
>
>def foo():
>   import bar
>   ....
>
> 2. What would happen if you built a big, long-running application and
>then changed a file while it was running. Would you really want one
>import statement in the file to load the new version? Then you would
>have objects floating around from the old version alongside objects from
>the new version. When it crashed due to the inconsistency, it would be
>very confusing.
>
>Pythonwin is one big long-running Python app. In that particular
>situation you might prefer it to act differently but Python doesn't
>really have a feature that allows what you want...and even if it did,
>you'd have to be careful because you could get into the same situation
>where you had old objects hanging around with the old behavior. That
>happens even with a reload today.
>
>> if module_exists(utils):
>>         reload(utils)
>> else:
>>         import utils
>
>You can make it simpler than that:
>
>import utils
>reload(utils)
>
>The first operation will continue silently and quickly if the module is
>loaded. Then it will be reloaded. The only "cost" is that the very first
>time you would load the module twice in a row...

Well, I didnt expect anyone to change the way the import works. But
your idea of the two in a row is good. Now if I could write a function
that I can call to do this, then I can enable and disable the reload()
at will.

Maybe a solution:

import aModule1 alias aName1; assert reload(aName1)
import aModule2 alias aName2; assert reload(aName2)

Then turn off debugging when done.

Thanks

Costas
	


From guido at digicool.com  Mon Mar 26 04:46:48 2001
From: guido at digicool.com (Guido van Rossum)
Date: Mon, 26 Mar 2001 04:46:48 -0500
Subject: WANTED: chairs for next Python conference
Message-ID: <200103260946.EAA02170@cj20424-a.reston1.va.home.com>

I'm looking for chairs for the next Python conference.  At least the
following positions are still open: BOF chair (new!), Application
track chair, Tools track chair.  (The Apps and Tools tracks are
roughly what the Zope and Apps tracks were this year.)  David Ascher
is program chair, I am conference chair (again).

We're in the early stages of conference organization; Foretec is
looking at having it in a Southern city in the US, towards the end of
February 2002.

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



From bernie at pacific.net.hk  Mon Mar 12 13:45:18 2001
From: bernie at pacific.net.hk (bernie)
Date: Tue, 13 Mar 2001 02:45:18 +0800
Subject: (no subject)
References: <mailman.984369195.11262.python-list@python.org>
Message-ID: <3AAD193D.EAC30597@pacific.net.hk>

Try www.modpython.org

Bernie

?i???q wrote:

> Hi All, Can I find Apache module for Python ??
> --
> Best Regards, Winson Chang 3F, 2, 150 Lane, Sec. 5, Hsin I Rd.,
> Taipei, Taiwan, R.O.C. Tel: +886-2-27225333 ext 376
> Fax: +886-2-27222330 Email: winson at mdsserv.mds.com.tw
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20010313/c711b8dd/attachment.html>

From igorr at ifi.uio.no  Thu Mar  1 10:55:07 2001
From: igorr at ifi.uio.no (Igor V. Rafienko)
Date: 01 Mar 2001 16:55:07 +0100
Subject: str behaviour
Message-ID: <xjvn1b55vs4.fsf@skepnuskapur.ifi.uio.no>

Hi,


I was playing around with printing tuples, when I noticed how str
behaves with "compound" types (such as tuples):


>>> class A:
...     def __str__( self ):
...         return "I'm A"
... 
>>> a = A()
>>> t = ( A, a )
>>> str( t )
'(<class __main__.A at 80fd498>, <__main__.A instance at 810f090>)'
>>> str( a )
"I'm A"
>>> 


I expected str to be called recursively for the constituent types.
But, alas, that did not happen. Am I misunderstanding something or is
it the intended behaviour? If the latter is the case, could someone
elaborate on why this behaviour was chosen?

TIA,





ivr
-- 
Besides, meat tends to run away when possible, or fights. Either
response presents behavioral challenges too complex for any existing
robot.
		-- Stuart Wilkinson, inventor of the "gastrobot"


From grante at visi.com  Wed Mar  7 12:22:11 2001
From: grante at visi.com (Grant Edwards)
Date: Wed, 07 Mar 2001 17:22:11 GMT
Subject: Can anyone recomend a good intoduction to C...
References: <mailman.983894288.19965.python-list@python.org> <3AA5EF59.507F1D91@olsen.ch> <983958878.976344@newsmaster-04.atnet.at>
Message-ID: <77up6.1289$y6.235541@ruti.visi.com>

In article <983958878.976344 at newsmaster-04.atnet.at>, Werner Schiendl wrote:
>
>> I don't think C deserves all the superlatives you apply to it,
>> but I agree that it's instructive to study C++ as an example of
>> a language that gets it wrong. Take a good language, C; add
>> some good ideas; and end up with a baroque mess. Fascinating.
>
>To be completely honest, I can not see the obvius reason why
>C++ should be such a bad programming language.

It's complicated and unsafe.  Either one is tolerable without
the other.  When you combine the two, you're dead.

>It is widely available (on almost any platform, I would guess).

If 1 billion people do a stupid thing, it's still a stupid thing.

>It supports for a real large number of different design approaches.

That isn't always a good thing. ;)

-- 
Grant Edwards                   grante             Yow!  It was a JOKE!! Get
                                  at               it?? I was receiving
                               visi.com            messages from DAVID
                                                   LETTERMAN!! YOW!!


From whisper at oz.net  Tue Mar 27 14:48:19 2001
From: whisper at oz.net (Dave LeBlanc)
Date: 27 Mar 2001 19:48:19 GMT
Subject: Python scripts as Drop targets in Windows Explorer
References: <99qogo$o6g$1@troll.powertech.no>
Message-ID: <99qqq3$qt7$0@216.39.170.247>

I can't speak for others, but i'd be reluctant to even attempt to
import the following into my registry without knowing exactly what it
does.

How about using the export option in regedit for a human readable
version of this please?

Interesting idea though :-)

Dave LeBlanc

On Tue, 27 Mar 2001 20:53:21 +0200, "Syver Enstad"
<syver at NOSPAMcyberwatcher.com> wrote:

>Run the attached .reg file (or enter the contained information manually) to
>make python scripts behave like .exe in the way that you can drop files on
>them and the files will appear in the sys.argv list as command line
>arguments.
>
What contained information?

>Something to consider for the installer of the win32 port of python maybe?
>
<snip>


From nospam at wanted.com  Tue Mar  6 10:25:51 2001
From: nospam at wanted.com (huber)
Date: Tue, 6 Mar 2001 16:25:51 +0100
Subject: Tkinter eventmask ?
Message-ID: <lm4389.tlg.ln@adsl.bapnt>

how can i get the modifiers ( Control, Shift ..... ) from a KeyPress 
event ? 


From costas at meezon.com  Thu Mar  1 21:17:02 2001
From: costas at meezon.com (Costas Menico)
Date: Fri, 02 Mar 2001 02:17:02 GMT
Subject: compile() source code from a string
References: <3a9ee600.391990@News.CIS.DFN.DE> <etdd7c1c8n5.fsf@pickled-herring.mit.edu>
Message-ID: <3a9f01f5.7549826@News.CIS.DFN.DE>

If you look at my post this is exactly what I am doing. I cannot get
this to work with multiline code (as in my example). Can you give me
an example how to make it work?

costas

Alex <new_name at mit.edu> wrote:

>
>>>> print compile.__doc__
>compile(source, filename, mode) -> code object
>
>Compile the source string (a Python module, statement or expression)
>into a code object that can be executed by the exec statement or eval().
>The filename will be used for run-time error messages.
>The mode must be 'exec' to compile a module, 'single' to compile a
>single (interactive) statement, or 'eval' to compile an expression.
>>>> 
>
>So, you want the 'exec' option.
>
>Alex.
>



From aleaxit at yahoo.com  Fri Mar 16 15:46:33 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 16 Mar 2001 21:46:33 +0100
Subject: Yet another C++ binding (using Extension Classes)
References: <mailman.984760603.29541.python-list@python.org>
Message-ID: <98tu4i06kq@news2.newsguy.com>

"Lee, Jaeho" <Jaeho.Lee at brooks.com> wrote in message
news:mailman.984760603.29541.python-list at python.org...
> > About a year ago, I had to bind a C++ template class into objects that
> > could be used from within python.  I made a wrapper for Extension
> > Classes that extended the basic structure into a C++ class.  Any objects
> > that subclassed this class would automagically become valid python
> > objects and could be extended in much the same way as Extension Classes.
>
> I am researching on how to expose "complex" C++ class in python. I found
> that SWIG does not support well my specific purpose. If you can open your
> source code (even with document - how great !!), it will be very helpfu.

Have a look at www.boost.org, specifically the Boost Python Library --
it's likely to include most or all of the tools you need.


Alex





From emile at fenx.com  Sun Mar 18 21:24:24 2001
From: emile at fenx.com (Emile van Sebille)
Date: Sun, 18 Mar 2001 18:24:24 -0800
Subject: Subsetting a dictionary
References: <993dkm$ttr$1@news6.svr.pol.co.uk>
Message-ID: <993qq3$4bajg$1@ID-11957.news.dfncis.de>

Rather than copying and then deleting from the copy, try building a new
dictionary while adding only those that qualify.

Something like:  (untested)

def subset(self, attr, value):
                 # Copy semantics
                 td = {}
                 td.data = self.data.copy()
                 for ky, val in TrainingData.items():
                          if ky[attr] == value:
                                    td[ky] = val
                 return td


"Will Newton" <will at nospam.misconception.org.uk> wrote in message
news:993dkm$ttr$1 at news6.svr.pol.co.uk...
>
> I need to get a subset of a dictionary:
>
> D = Dictionary of objects (in this case, training examples for a decision
> tree)
>
> D2 = {D | d in D, d under some constraint}
>
> I need to be able to do this, but as it is temporary I do not want to
> destructively remove examples, so:
>
> D.subset(c)
> will return a new dictionary object that is equivalent to D under
> constraint c.
>
> I have code that does this, but it's very slow, consuming typically 75% of
> program runtime - it is called many times, which seems to be unavoidable
> without modifying the algorithm (ID3) too far.
>
> I currently have:
>
> class TrainingData:
>          data = {}
>         def subset(self, attr, value):
>                 # Copy semantics
>
>                 td = TrainingData()
>
>                 td.data = self.data.copy()
>
>                 for item in td.data.keys():
>                          if item[attr] != value:
>                                    del td.data[item]
>
>                 return td
>
> Can anyone see how I can speed this up?
>
> I could conceivably do this with a list and use filter, I'll see how that
> works.




From sheila at spamcop.net  Sun Mar 18 15:13:17 2001
From: sheila at spamcop.net (Sheila King)
Date: Sun, 18 Mar 2001 20:13:17 GMT
Subject: Help with 500 error on cgi script
References: <quv8bto7r8s6f1tckc7qpmuir0kvufp1nb@4ax.com> <mailman.984909623.8814.python-list@python.org> <5739bt04e50ku01rn3dptgqch3l07p2cnf@4ax.com> <mailman.984915263.21558.python-list@python.org>
Message-ID: <fl5abt8vtmldoakfa5t1v83q9t558d58n5@4ax.com>

On Sun, 18 Mar 2001 12:35:11 +0100, Steve Purcell <stephen_purcell at yahoo.com>
wrote in comp.lang.python in article
<mailman.984915263.21558.python-list at python.org>:

:Closing quote is missing.
:
:Try running the scripts from the command line first to weed out such syntax
:errors.
:
:-Steve

Thanks. I finally got my script working this morning. I used your advice, and
ran it from the command line. That was much more efficient than testing it
through the web server, for syntax errors anyhow.

--
Sheila King
http://www.thinkspot.net/sheila/
http://www.k12groups.org/



From kens at sightreader.com  Thu Mar 22 13:53:36 2001
From: kens at sightreader.com (Ken Seehof)
Date: Thu, 22 Mar 2001 10:53:36 -0800
Subject: Good Editor?
References: <3ab9a32b$1_1@news2.one.net> <3AB9FAC9.97EA420E@student.liu.se> <slrn9bjv7m.4qv.grey@teleute.dmiyu.org> <99d780$l2g$1@laurel.tc.umn.edu> <3ABA20F7.AB71D3B4@student.liu.se> <98qu6.166420$Z8.35271037@typhoon.southeast.rr.com>
Message-ID: <002501c0b301$672d11c0$04090a0a@upcast.com>

Yeah, but I want to know: does it actually work?
> "Rikard Bosnjakovic"
> > Anyone sending unwanted advertising e-mail to my address will be
> > charged $250 for network traffic and computing time. By extracting my
> > address from this message or its header, you agree to these terms.
> 
> love it...Love It...LOVE IT!
> 
> Don
----------------------------------------------------
Copyright (c) 2001 by Ken Seehof
This document may not be distributed, copied,
duplicated, or replicated in any form without
express written permission by Ken Seehof.
Express permission is hereby granted.
kens at neuralintegrator.com
----------------------------------------------------





From moshez at zadka.site.co.il  Fri Mar 23 12:13:24 2001
From: moshez at zadka.site.co.il (Moshe Zadka)
Date: Fri, 23 Mar 2001 19:13:24 +0200
Subject: newbie question
In-Reply-To: <575F0D65CC45F12D.7D165177C452F6D5.01903532892CC93D@lp.airnews.net>
References: <575F0D65CC45F12D.7D165177C452F6D5.01903532892CC93D@lp.airnews.net>, <99fll8$996$1@rex.ip-plus.net>
Message-ID: <E14gV7s-0000su-00@darjeeling>

On 23 Mar 2001, claird at starbase.neosoft.com (Cameron Laird) wrote:

> There are few languages that canNOT be used with
> Apache.

I might be talking out of my ass here, in which case Cameron will surely
correct me. Haskell, Ruby, Icon, Scheme, ML and Smalltalk are all languages 
which are highly useful, and which no interfaces to Apache exists as far as I
know. I'd simply go the opposite direction and say

"there are few things Python cannot talk to. Apache is not one of those"
(I like to phrase it: Python plays nice with the other kids. And there
are *always* other kids)

waiting-for-someone-to-prove-me-wrong-(hurd,plan9)-ly y'rs, Z.
-- 
"I'll be ex-DPL soon anyway so I'm        |LUKE: Is Perl better than Python?
looking for someplace else to grab power."|YODA: No...no... no. Quicker,
   -- Wichert Akkerman (on debian-private)|      easier, more seductive.
For public key, finger moshez at debian.org  |http://www.{python,debian,gnu}.org



From donn at oz.net  Fri Mar 16 20:37:12 2001
From: donn at oz.net (Donn Cave)
Date: 17 Mar 2001 01:37:12 GMT
Subject: Should I close after popen??
References: <XFMail.20010315142704.shaleh@valinux.com> <mailman.984697040.2458.python-list@python.org> <m3lmq6ppv0.fsf@master.athome> <98satl$1t8$0@216.39.151.169> <m3lmq6nk9a.fsf@master.athome>
Message-ID: <98uf48$552$1@216.39.151.169>

Quoth Andrew Markebo <flognat at flognat.myip.org>:
| Ahh hmm yep.. the scripts are running in 1.5.2, so what I do is to
| remove the waits, (hmm are ther any possibility of doing something
| similar??? am I supposed to use select/poll?)

Depends on what you want to do.  For sure, select has the potential
to solve some programming problems with pipes.  One thing that you
should bear in mind, though, is that the file object returned by
popen() buffers input, and select() can't see the buffered data,
so you may have "no data" yet have more to read from the file object.
You can ignore the file object, use the fileno() method to obtain 
the pipe file descriptor and use os.read() on it, for example.

Anyway, it would be easier to say what you ought to use, if it were
clear what problem you're trying to solve.  It's probably simpler
than you think.  If you just read the pipes until they return an
empty string, like any file, there's no reason to expect problems.

| Uhm how do I get the exit-code of stuff from popen? Looking at popen2
| it forks and does waitpid.. 

Did I mention that you could find some relevant documentation in
the pclose man page?

  fp = popen('echo OK; exit 9')
  data = fp.read()
  status = fp.close()

You might have to shift the result down 8, on UNIX, I forget.

One unrelated advantage of popen2 is that if you supply a list of
arguments, it will use them directly instead of a shell command
line.  That is much safer.  A command line constructed at run
time can have much more surprising effects if it's wrong, because
the shell is a much more powerful and general purpose program.

	Donn Cave, donn at oz.net


From alan.gauld at bt.com  Mon Mar 12 09:50:21 2001
From: alan.gauld at bt.com (Alan Gauld)
Date: Mon, 12 Mar 2001 14:50:21 +0000
Subject: learning C
References: <mailman.983972829.23481.python-list@python.org> <983975751.18833@newsmaster-04.atnet.at> <slrn9adklr.bv9.scarblac@pino.selwerd.nl> <984047775.592522@newsmaster-04.atnet.at> <9896hk$ig8$1@verence.demon.co.uk> <984134612.781779@newsmaster-04.atnet.at>
Message-ID: <3AACE22D.3A9EEB12@bt.com>

Werner Schiendl wrote:
> Unsafe is somewhat the same topic I think: It's a question 
> of disciplined use. 

I think one problem with C++ is the number of temporary objects 
it creates. Calling a seemingly innocuous function with some 
object references as paramrters may or may not involve the 
creation of many temporary objects. Since these objects may be
subclasses the programmer may not be able to determine how 
much memory will be alloocated by the function call mechanism 
and in a minimalist memory system it can cause a crash which
appears inermittent and non deterministic.

In C, by virtue of its fixed primitive types, is much easier 
to gurantee what will happen by way of memory allocation etc.

> > There's the feeling that you never quite know what it's 
> > going to do.  

Temporary objects being one such case. Others include the use of
operator overloading, function overloading vv default parameters 
etc etc. C++ is a maintenance nightmare compared to C - I speak 
from 18 months painful experience maintaining a large C++ system!

> You can have the same in C++ (using one or more static objects, 
> or construct the objects just once when booting)

Even here the compiler will cause tempoiray object copies to 
occur when passing object references around. Static objects are 
not as 'static' as we might like to think!


> This is nothing more difficult than in C.

Sorry but it is. There's just so much less lurking under 
the hood in C.

> > One of the particular unsafe things in C is error handling, 
> C++ exceptions at least make sure the error _is_ handled 

This bit I tend to agree with. I'm not sure what the original 
poster had in mind here...

Alan g


From dnew at san.rr.com  Fri Mar 16 12:14:34 2001
From: dnew at san.rr.com (Darren New)
Date: Fri, 16 Mar 2001 17:14:34 GMT
Subject: list.sort(cmpfunc) question
References: <m1ae6lg5qt.fsf@pclabs188.labs.trema.com> <98t9dq$qvu$1@panix6.panix.com> <98tf9d$g3v$1@news.mathworks.com>
Message-ID: <3AB249FB.131080CD@san.rr.com>

Joshua Marshall wrote:
> Seems the word "should" should never appear in documentation.

Not quite true. RFC's use "should" all the time, but there's a very strict
definition for what it means. I think "usually" is the other problematic
word. :-)

-- 
Darren New / Senior MTS & Free Radical / Invisible Worlds Inc.
San Diego, CA, USA (PST).  Cryptokeys on demand.


From dan at eevolved.com  Fri Mar  2 11:35:48 2001
From: dan at eevolved.com (Dan Parisien)
Date: Fri, 02 Mar 2001 16:35:48 GMT
Subject: feature wish
Message-ID: <EZPn6.325779$f36.11597531@news20.bellglobal.com>

it seems all the special purpose overloadable methods are to provide type 
emulation through classes
e.g.
__nonzero__
__contains__
__setitem__
etc.

it would be cool if there was another method provided that allowed a 
programmer to overload the value returned by an instance as well.

e.g.

class example:
        def __init__(self, name, data):
                self.name = name
                self.data = data

        def __value__(self):
                return self.data

eg = example("my name", (1,2,3))

>>>print eg
(1,2,3)

You could get to the instance behind the value by using a function like
instance(eg)

What do you guys think?

Dan



From sverker.is at home.se  Thu Mar  1 13:55:32 2001
From: sverker.is at home.se (Sverker Nilsson)
Date: Thu, 01 Mar 2001 19:55:32 +0100
Subject: A counter-proposal to __future__ in PEP 236
Message-ID: <3A9E9B23.23FFCEDF@home.se>

I think the directive directive is better too. When I read about 
the future import I was worried over what the new semantics
of import would be, for example what would happen to the
identifier that would be imported: would it be visible, 
probably not but that's a new semantics to understand & learn.

Seems to me it is better starting afresh with a new directive
directive. So I'd give my vote of agreement with you.

Also it's reasurring as Joshua Marshall noted that directive
wouldn't need to break code even if it was used in an assignment,
because directive <directive> would have a unique meaning anyway.

Sverker



From uche.ogbuji at fourthought.com  Mon Mar 12 02:11:27 2001
From: uche.ogbuji at fourthought.com (Uche Ogbuji)
Date: Mon, 12 Mar 2001 00:11:27 -0700
Subject: [Python-Dev] Revive the types sig? 
In-Reply-To: Message from Jeremy Hylton <jeremy@alum.mit.edu> 
   of "Sun, 11 Mar 2001 20:22:04 EST." <15020.9404.557943.164934@w221.z064000254.bwi-md.dsl.cnc.net> 
Message-ID: <200103120711.AAA09711@localhost.localdomain>

Jeremy Hylton:

> If documentation is really the problem you want to solve, I imagine
> we'd make much more progress if we could agree on a javadoc-style
> format for documentation.  The ability to add return-type declarations
> to functions and methods doesn't seem like much of a win.

I know this isn't the types SIG and all, but since it has come up here, I'd 
like to (once again) express my violent disagreement with the efforts to add 
static typing to Python.  After this, I won't pursue the thread further here.

I used to agree with John Max Skaller that if any such beast were needed, it 
should be a more general system for asserting correctness, but I now realize 
that even that avenue might lead to madness.

Python provides more than enough power for any programmer to impose their own 
correctness tests, including those for type-safety.  Paul has pointed out to 
me that the goal of the types SIG is some mechanism that would not affect 
those of us who want nothing to do with static typing; but my fear is that 
once the decision is made to come up with something, such considerations might 
be the first out the window.  Indeed, the last round of talks produced some 
very outre proposals.

Type errors are not even close to the majority of those I make while 
programming in Python, and I'm quite certain that the code I've written in 
Python is much less buggy than code I've written in strongly-typed languages.  
Expressiveness, IMO, is a far better aid to correctness than artificial 
restrictions (see Java for the example of school-marm programming gone amok).

If I understand Jeremy correctly, I am in strong agreement that it is at least 
worth trying the structured documentation approach to signalling pre- and 
post-conditions before turning Python into a rather different language.


-- 
Uche Ogbuji                               Principal Consultant
uche.ogbuji at fourthought.com               +1 303 583 9900 x 101
Fourthought, Inc.                         http://Fourthought.com 
4735 East Walnut St, Ste. C, Boulder, CO 80301-2537, USA
Software-engineering, knowledge-management, XML, CORBA, Linux, Python





From aleaxit at yahoo.com  Fri Mar 16 08:44:29 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 16 Mar 2001 14:44:29 +0100
Subject: gah! I hate the new string syntax
References: <mailman.983485923.1136.python-list@python.org> <97sudi01i26@news1.newsguy.com> <98elrk$li8$1@nntp.Stanford.EDU> <98fc070thh@news1.newsguy.com> <98iqo2$56o$1@nntp.Stanford.EDU> <j4kr6.68739$ML1.3282814@e420r-atl2.usenetserver.com> <Pine.LNX.4.21.0103161339210.2809-100000@bast.jhb.equinox.co.za>
Message-ID: <98t5bv0l8b@news1.newsguy.com>

"Pieter Nagel" <pieter at nagel.co.za> wrote in message
news:Pine.LNX.4.21.0103161339210.2809-100000 at bast.jhb.equinox.co.za...
    [snip]
> Of course you can. The technique is to modify the String class itself

It's not a class.  It's a non-class type, and does not support
modification.


Alex





From grante at visi.com  Fri Mar 30 10:51:16 2001
From: grante at visi.com (Grant Edwards)
Date: Fri, 30 Mar 2001 15:51:16 GMT
Subject: Creation d'un NG francophone sur Python
References: <1eqzyk6.gxd4jr1tkjykzN%francois.granger@free.fr> <9a1e6f011es@enews3.newsguy.com> <D9305864F80B9ADF.9B74F9BE0F743AE2.DC045B55BF4E1CD8@lp.airnews.net> <mailman.985964201.6565.python-list@python.org>
Message-ID: <UX1x6.9226$SB2.838492@ruti.visi.com>

In article <mailman.985964201.6565.python-list at python.org>, Fran?ois Pinard wrote:

>> >Pardon, mais pourquoi un NG pour Python en fran?ais?
>
>> C'est un choix, une opportunit?.
>
>I directly know some people that do not read English easily,
>and that would be Python users if there were more French
>resources for Python. I'm sure the same applies for people
>speaking many other languages. English should not be necessary
>to use computers, and if Python is meant to be a CP4E, it
>should have wide views about non-English people.

Post away in whatever languge you want.  I see no need to
separate people by linguistic lines.  I suppose if you don't
read English it would be nice not to have to sift through the
newsgroup looking for postings you can read.

Though as a mono-lingual American, I also feel the urge to

P O S T   S L O W L Y   A N D   L O U D L Y   I F   I T   W I L L
H E L P   P E O P L E   W H O   D O N ' T   R E A D   E N G L I S H.

;)

-- 
Grant Edwards                   grante             Yow!  Is this "BIKINI
                                  at               BEACH"?
                               visi.com            


From nde at comp.leeds.ac.uk  Thu Mar 22 07:36:42 2001
From: nde at comp.leeds.ac.uk (Nick Efford)
Date: Thu, 22 Mar 2001 12:36:42 GMT
Subject: Good Editor?
References: <3ab9a32b$1_1@news2.one.net> <Xns906C5F868E6B7PaCmAnRDLM@161.48.76.20>
Message-ID: <GALn19.C1z@leeds.ac.uk>

On 22 Mar 2001 08:22:26 GMT, Fred Pacquier <fredp at mygale.org.nospam> wrote:

> >Can someone recomend a good editor to write python in?  Im just getting
> >started learning python and dont know which ones support python.
> 
> You don't mention the platform. Under Unix/Linux there are a number of 
> options (including Vim, Emacs, etc.). Under Windows, I found WScite 
> (http://www.scintilla.org) the easiest to get started with...

vim and emacs run on Windows, too!  So you can use one editor across
platforms as well as one programming language :)


N.



From stephen_purcell at yahoo.com  Mon Mar  5 11:18:13 2001
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Mon, 5 Mar 2001 17:18:13 +0100
Subject: Name of the Script
In-Reply-To: <cbKjOqPRwNza00jOm7A1HBDkRYj7@4ax.com>; from paul.moore@uk.origin-it.com on Mon, Mar 05, 2001 at 04:40:08PM +0100
References: <20010304.185354.2114937732.8277@localhost.localdomain> <NJCo6.6720$qv3.2834377@nnrp5-w.sbc.net> <FpmjOjvwe5bm0Bg09LDTMPhh5vji@4ax.com> <Pine.LNX.4.21.0103051715150.18263-100000@kev.ioc.ee> <cbKjOqPRwNza00jOm7A1HBDkRYj7@4ax.com>
Message-ID: <20010305171813.A9043@freedom.puma-ag.com>

Paul Moore wrote:
> On Mon, 5 Mar 2001 16:17:38 +0100 , Pearu Peterson <pearu at cens.ioc.ee>
> wrote:
> >--- a.py ---
> >import os,sys
> >print os.path.abspath(sys.argv[0])
> >---
> >might be what you want.
> 
> Close, but not quite...


>>> def current_module_path():
...     import os, sys
...     try:
...         path = sys.modules[__name__].__file__
...     except AttributeError:
...         path = sys.argv[0]
...     return path and os.path.join(os.getcwd(), path) or None
... 
>>> 

Then, in an interactive session:

>>> `current_module_path()`
'None'

And in 'a.py':

'/home/steve/a.py'


-Steve

-- 
Steve Purcell, Pythangelist
Get testing at http://pyunit.sourceforge.net/
Get servlets at http://pyserv.sourceforge.net/
"Even snakes are afraid of snakes." -- Steven Wright



From sholden at holdenweb.com  Wed Mar  7 14:24:38 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Wed, 7 Mar 2001 14:24:38 -0500
Subject: I come to praise .join, not to bury it...
References: <mailman.983485923.1136.python-list@python.org> <97p7iq$kk2$1@nntp6.u.washington.edu> <97rka002k9@news1.newsguy.com> <3AA2F4B3.A7496E76@cosc.canterbury.ac.nz> <97vmoh061p@news2.newsguy.com> <3AA46134.D385208B@cosc.canterbury.ac.nz> <98292h0s0g@news1.newsguy.com> <3AA58BD0.3324C709@cosc.canterbury.ac.nz> <9850ge013ns@news1.newsguy.com> <gRup6.364652$ge4.128679630@news2.rdc2.tx.home.com>
Message-ID: <qZvp6.29197$ML1.1439945@e420r-atl2.usenetserver.com>

"Rainer Deyke" <root at rainerdeyke.com> wrote in message
news:gRup6.364652$ge4.128679630 at news2.rdc2.tx.home.com...
> "Alex Martelli" <aleaxit at yahoo.com> wrote in message
> news:9850ge013ns at news1.newsguy.com...
> > in fact, just about any test for 'IS object X of exact type
> > Y' you can find in the Python sources, unless they are
> > shortcut-like 'accelerators' for a special case _before_
> > the more-general request 'please o mr X try to behave like
> > an Y, are you able to?', can be seen as implementation
> > defects, in my personal opinion.
>
> I'm curious: how do you feel about the automagic conversions of functions
> (and only true function, not function-like objects) into unbound methods,
> and from unbound methods into bound methods?  I'm thinking that the former
> could be eliminated entirely, and the latter replaced by a __bind__ magic
> method that allows the mechanism to apply to function-like objects.
>
>
Surely no type conversion is involved: the interpreter just converts

    instance.method(*args)

into

    class.method(instance, *args)

regards
 Steve





From vecernik at aon.at  Tue Mar 27 03:19:23 2001
From: vecernik at aon.at (Oliver Vecernik)
Date: Tue, 27 Mar 2001 10:19:23 +0200
Subject: sitemap and ftp
Message-ID: <3AC04D0B.5A44A4D5@aon.at>

Hi,

I'm relatively new to Python, but I felt immediately in love because of
it's clear structure and ease to read the code. Unfortenately I still
lack a lot of knowledge. Maybe some can help me or point me to the right
spot in the docs.

I'd like to parse through my website tree and generate an automated
sitemap from the <title> tags. Is there a module that helps me achieving
this task?

Furthermore I'd like to copy all *.html files and all *.jpg and *.gif
files to my ISP preserving the directory structure. Before copying the
*.html files have to be filtered (adding a header and a footer). Has
anybody written such routines already?

Regards,
Oliver


From stephen_purcell at yahoo.com  Fri Mar 30 02:09:36 2001
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Fri, 30 Mar 2001 09:09:36 +0200
Subject: max filename size
In-Reply-To: <3ac416a6$0$4017$6e49188b@news.goldengate.net>; from volucris@hotmail.com on Thu, Mar 29, 2001 at 11:16:41PM -0600
References: <3ac416a6$0$4017$6e49188b@news.goldengate.net>
Message-ID: <20010330090936.C4187@freedom.puma-ag.com>

Voluciris wrote:
> Is there away to find the maximum filename length of the current os? On
> Windows it's 215 characters, but I don't want to hard code that in. It would
> be much more pythonic if it was portable.

I don't think there's a standard function for this.

On some platforms, the maximum filename length varies according to the
location on the disk. If you run UNIX with different filesystems mounted
in various locations on your directory tree, you might see the following
values for the maximum filename length in those locations:

    >>> import statvfs, os
    >>> os.statvfs('/')[statvfs.F_NAMEMAX]
    4032
    >>> os.statvfs('/boot')[statvfs.F_NAMEMAX]
    255

-Steve

-- 
Steve Purcell, Pythangelist
Get testing at http://pyunit.sourceforge.net/
Any opinions expressed herein are my own and not necessarily those of Yahoo



From shredwheat at mediaone.net  Mon Mar 19 10:22:45 2001
From: shredwheat at mediaone.net (Pete Shinners)
Date: Mon, 19 Mar 2001 15:22:45 GMT
Subject: Python In A Nutshell - Numpy essential
References: <3AAE7E0D@MailAndNews.com> <slrn9aolph.ccf.jimhill@meldrick.swcp.com> <3AACE20C.AC8D2EA2@northwestern.edu> <Ikkr6.68807$ML1.3285992@e420r-atl2.usenetserver.com> <3AAE5DA3.B736C2D0@northwestern.edu> <Pine.HPX.4.33.0103182208511.10898-100000@sesame.ee.byu.edu>
Message-ID: <9vpt6.71944$Ch.9136379@typhoon.we.rr.com>

> I used to be one who really wanted Numeric in the Python core.  At this
> point it is not critical to me.  Maybe someday, the multidimensional array
> aspect of Numeric will be distributed with Python.  For now, though what
> is needed instead is an easy to install package (which includes Numeric)
> and provides a competitive numerical enviroment (something like SciPy
> which is coming this summer...)
>
> I'm happy to keep Numeric out of the core as it gives the math-heads more
> control over it's optimization.

well, i would still like to see at least the multidimensional array
part of Numeric included with the python release. the most notable
changes to the package since python2.0 came out has been support for
python2.0 and python2.1 features. the package seems to be settling
down nicely, and is a great value to me (and i'm not even using it
for any scientific purposes, owell)

i just created a fun tutorial on Numeric Python and using it to
do imaging effects here.
http://pygame.seul.org/docs/tut/SurfarrayIntro.html






From gregj at pobox.com  Fri Mar 16 23:39:34 2001
From: gregj at pobox.com (Greg Jorgensen)
Date: Sat, 17 Mar 2001 04:39:34 GMT
Subject: Cookies
References: <mailman.984707839.22839.python-list@python.org> <lmq6m07j.fsf@propylon.com> <mailman.984773062.25819.python-list@python.org>
Message-ID: <aUBs6.581328$U46.17323742@news1.sttls1.wa.home.com>

"Ken Seehof" <kens at sightreader.com> wrote in message
news:mailman.984773062.25819.python-list at python.org...
> Here's a restatement of the problem.  It's really about crossing the
sandbox
> barrier.
>
> Does anyone know of another mailing list more specific to http questions?
>
> 1. Client downloads an execuatable (frozen python application).
> 2. Client downloads a web page.
> 3. Web page javascript generates a statistically unique random identifier.
> 4. Web page sends identifier to the server, which uses it as database key.
> 5. User runs python application
> 6. *** Python application somehow gets the identifier from somewhere ***
> 7. Now the python application and the server both have the same
identifier.

Can you do it like this?

1. Client downloads and runs Python executable
2. Python app sends HTTP request to remote server
3. Server creates unique ID, sends it back to the Python app via HTTP
4. Python app saves the unique ID somewhere (text file, registry) where it
or other Python app can find it later
5. Python app launches web browser, passing server URL + unique ID
6. Web browser opens up remote web page, sending unique ID in the URL
7. Remote server gets unique ID, sends it back to client embedded in the web
page (hidden INPUT field)
8. User fills out form, submits it (along with the unique ID embedded in the
web page)
9. Server saves user's inputs in database record keyed on unique ID

At this point the server could send back a page containing a cookie that has
the unique ID (or whatever) in it. Or you could use JavaScript to create a
bookmark in the user's browser that has your URL + unique ID embedded in it.

Maybe you need two Python apps: one to "prime the pump," i.e. get a unique
ID from the server and save it, and another that does whatever your current
application is supposed to do.

Variations on this could involve writing an HTML page from your Python app
on the user's system, and embedding the unique ID in that page. When the
page is loaded into the browser it sends a URL + unique ID to your server.
For clients using Windows or MacOS you can save a URL as a shortcut on the
desktop; again that URL can be your web page plus the unique ID generated
earlier (e.g. http://www.somesite.com/page.html?id=12345).

Again I advise you to generate the unique IDs on your server, not on the
client side. I've often used a simple scheme that combines a record sequence
number (called an IDENTITY column in SQL Server) and a random number
generated on the server and stored in the database row. The database layout
looks like this:

id: identity column or default value set to a sequence generator (a la
Oracle)
check: default value is a function that generates a random number
user-stuff: name, address, etc.
...

You can send this back as id.check, and pass it in the URL like
id=1234.67132. The server uses the first part to fetch the row, and then
checks to make sure the second bit matches (this can all be done in the SQL
SELECT).

Good luck.

--
Greg Jorgensen / programmer, pedant, raconteur / Portland, Oregon, USA





From jcc.ugm at ix.netcom.com  Wed Mar 28 15:18:20 2001
From: jcc.ugm at ix.netcom.com (Jonathan Claggett)
Date: Wed, 28 Mar 2001 15:18:20 -0500
Subject: Is inheritance broken?
In-Reply-To: <15042.17325.125458.471763@beluga.mojam.com>
Message-ID: <NDBBLAKACLGJJBDNGGCMIEJDCCAA.jcc.ugm@ix.netcom.com>

>How about creating mixins, then composing a set of classes from them:

Interesting. Are there conventions for mixin classes which distinguish them
from normal classes? For example, are mixins not suppposed to use
inheritance (or at least not use mulitple inheritance)? Thanks for the help,

Jonathan





From jdgorman at att.net  Sat Mar 24 22:59:39 2001
From: jdgorman at att.net (John D. Gorman)
Date: Sun, 25 Mar 2001 03:59:39 GMT
Subject: New Tkinter and PIL documentation (free)
References: <99en42$2vl6$1@newshost.nmt.edu> <Y_Vu6.179074$Z8.37733056@typhoon.southeast.rr.com> <99ikps$1qmo$1@newshost.nmt.edu>
Message-ID: <3ABD6D24.260DE481@att.net>

Was this document generated using Adobe Distiller?  I've had similar problems with
fonts and Distiller.  I've had better luck with pdftex.  This produces pdfs that
don't have the font blockiness.  There's a Win32 version at miktex.org.

--John Gorman

John Shipman wrote:

> First I wrote:
>
> +--
> | At this site you will find my newly written reference document
> | for Tkinter:
> |
> |     http://www.nmt.edu/tcc/help/pubs/lang.html
> +--
>
> Then Don Tuttle <tuttledon at hotmail.com> replied:
>
> +--
> | I downloaded the Tkinter PDF, but apparently I don't have the same
> | font you used.  (The text is very fuzzy)  Assuming this is the
> | problem, would you consided remaking the PDFs with a more standard
> | font? Thanks.
> +--
>
> The document was designed for print, not for screen use.  It
> should print cleanly.  If you're reading it with Acrobat,
> try increasing the magnification.
>
> --
> John Shipman (john at nmt.edu), Applications Specialist, NM Tech Computer Center,
> Speare 128, Socorro, NM 87801, (505) 835-5950, http://www.nmt.edu/~john
>   ``Let's go outside and commiserate with nature.''  --Dave Farber
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20010325/24b73203/attachment.html>

From emile at fenx.com  Fri Mar 23 10:10:06 2001
From: emile at fenx.com (Emile van Sebille)
Date: Fri, 23 Mar 2001 07:10:06 -0800
Subject: New PEP: The directive statement
References: <mailman.985085006.24236.python-list@python.org> <slrn9bk7fk.fnm.scarblac@pino.selwerd.nl> <j4k85gkiye.fsf@informatik.hu-berlin.de> <99fnif$gsq$1@news.mathworks.com>
Message-ID: <99fp6h$15i5v$1@ID-11957.news.dfncis.de>

I don't think the interpreter switch works.  Modules will either depend on
the current or future behavior, so the option needs to be there.

--

Emile van Sebille
emile at fenx.com

---------
"Joshua Marshall" <jmarshal at mathworks.com> wrote in message
news:99fnif$gsq$1 at news.mathworks.com...
> Martin von Loewis <loewis at informatik.hu-berlin.de> wrote:
> > scarblac at pino.selwerd.nl (Remco Gerlich) writes:
>
> >> First the inconsistency: the syntax definition in (C) claims that some
> >> restrictions on the placement of directive may be in place, depending
on
> >> the directive. However, (D) notes that directives must appear as the
first
> >> non-string token in a file, for backwards compatibility. So in
practice,
> >> directive can't be used anywhere in a file.
>
> > That's not inconsistent. There is only one directive defined so far,
> > to activate nested scopes, and it must appear at the top of the file.
> > In some later release of Python, directive will become a proper
> > keyword, at which time introducing directives with less strict
> > placement rules is possible.
>
> Please, no more keywords.  It's unnecessary in this case.
> How about an interpreter switch:
>
>   python -directive nested_scopes fu.py




From aleaxit at yahoo.com  Fri Mar  9 10:09:23 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 9 Mar 2001 16:09:23 +0100
Subject: win32com:MakePy -- Early Binding without the size jitters?
References: <mailman.984043211.10915.python-list@python.org> <98ae5r01hn0@news2.newsguy.com> <guKoOvyz2tkgZ9zg6THcL9zanB2o@4ax.com>
Message-ID: <98armv0233m@news2.newsguy.com>

"Paul Moore" <paul.moore at uk.origin-it.com> wrote in message
news:guKoOvyz2tkgZ9zg6THcL9zanB2o at 4ax.com...
    [snip]
> >Yes: study the sources and look at the bForDemand
> >flag (introduced reasonably recently).  That flag
> >ensures that only the wrappers you're actually using
> >are generated (it uses packages rather than modules
> >to house the wrappers).
>
> More specifically, run makepy -d to generate "on demand" support. Support
for
> the basic object is generated, but then sub-object support is built as
> necessary.
>
> This is quite nice, and I didn't know it was available. Thanks for the
> pointer, Alex!

Thank *YOU* for mentioning the -d switch, which I had
completely omitted...!!!


Alex





From greg at cosc.canterbury.ac.nz  Mon Mar 19 20:57:51 2001
From: greg at cosc.canterbury.ac.nz (Greg Ewing)
Date: Tue, 20 Mar 2001 13:57:51 +1200
Subject: This math scares me
References: <3aad4062.1290021@News.CIS.DFN.DE> <20010312170356.B3513@better.net> <mailman.984480856.18097.python-list@python.org> <984486395.407379@newsmaster-04.atnet.at> <slrn9asf1d.ja3.qrczak@qrnik.zagroda> <984518199.127140@newsmaster-04.atnet.at> <jwbnews-09D75F.13313313032001@corp.supernews.com> <98ohuc$du9$1@news.udel.edu> <7bQr6.644$lz1.17686@ruti.visi.com> <98oku1$f6a$1@news.udel.edu> <erRr6.659$lz1.20662@ruti.visi.com> <98qv55$f5d$1@news.udel.edu>
Message-ID: <3AB6B91F.F201AA7C@cosc.canterbury.ac.nz>

Terry Reedy wrote:
> 
> BCD does not eliminate the problems of finite precision arithmetic.

No, but it lets you see exactly what's going on
in terms that most people are familiar with. It
eliminates any *extra* surprises due to conversion
between bases.

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


From Patrick.Bussi at space.alcatel.fr  Thu Mar 29 05:06:52 2001
From: Patrick.Bussi at space.alcatel.fr (Patrick.Bussi at space.alcatel.fr)
Date: Thu, 29 Mar 2001 12:06:52 +0200
Subject: =?iso-8859-1?Q?R=E9f._:_Re:_Why_not_Tcl/Tk=3F?=
Message-ID: <C1256A1E.00384E80.00@vzmta01.netfr.alcatel.fr>


Hello,

I've had the same issues as Alexander's and finally choose Python after reading
some studies about languages comparisons. A list of links is available at
http://www.python.org/doc/Comparisons.html

Hope this helps.

---
Patrick Bussi
patrick.bussi at space.alcatel.fr


Any opinions expressed are my own and not necessarily those of my Company.




grante at visi.com (Grant Edwards) on 27/03/2001 17:54:44

Pour :    python-list at python.org
cc :   (ccc : Patrick Bussi/ALCATEL-SPACE)
Objet :   Re: Why not Tcl/Tk?


-------------- next part --------------


In article <4211ctgvusgcma0c6ok1cpmjou98lv9714 at 4ax.com>, Alexander Kluge wrote:
>Here are some very general newbie questions and a plea for your opinion.
>
>Having some outdated hobby hacking experience (Turbopascal, Modula2,
>DBase) on AtariST and PC, and some recent experience with a 70k perl cgi
>script consecutivly changed by different persons (@!?$@*!!!), I think
>that I actually don't want to use neither C, C++ or Java (overkill) nor
>perl (see above). I confess having used the try and error method of
>learning, therefore my autodidactic abilities stopped every time dynamic
>arrays were needed :).
>I don't want neither spend much time in learning the language nor
>writing professional strength applications but I rather look for a
>replacement of very easy languages as e.g. pascal/basic.
>
>I'm interested in your opinions concerning the following points:
>
>-How would you describe the major strengths and flaws of Tcl vs. Python?

Tcl
 Flaws:
     Only one data type: string
     Incomprehensible quoting semantics
     Roll-your-own control flow
     No OO support
 Strengths:
     Tk integration
     Lots of code out there
     Roll-your-own control flow

Python:
 Flaws:
     No roll-your-own control flow
     Lots of GUI options
 Strengths:
     No roll-your-own control flow
     Good OO support
     Lots of GUI options
     Simple syntax and semantics
     Library modules

>-Why do you prefer the one you use over the other?

I gave up on Tcl after one program and switch to Scheme.  Now I
use Python a lot more than Scheme.

[I don't know much about eithr on Win32, so I'll let others
answer those questions.]

--
Grant Edwards                   grante             Yow!  As President I
                                  at               have to go vacuum my coin
                               visi.com            collection!
--
http://mail.python.org/mailman/listinfo/python-list



From ctavares at develop.com  Sun Mar  4 21:51:40 2001
From: ctavares at develop.com (Chris Tavares)
Date: Mon, 05 Mar 2001 02:51:40 GMT
Subject: Static class methods
References: <rdzo6.154$gc4.80445@news.pacbell.net>
Message-ID: <0bDo6.16804$7Y1.1656672@newsread2.prod.itd.earthlink.net>

Try this:

>>> class Foo:
...     class FooMethods:
...             def foo1(self):
...                     print "FooMethods.foo1"
...     class_methods = FooMethods()
...
>>> foo = Foo()
>>> foo.class_methods.foo1()
FooMethods.foo1
>>> Foo.class_methods.foo1()
FooMethods.foo1
>>>

-Chris

"Alex Shindich" <alex at shindich.com> wrote in message
news:rdzo6.154$gc4.80445 at news.pacbell.net...
> Just out of my curiosity, have any of you ladies and gentlemen experienced
a
> need for an equivalent of C++/Java static method?
> Because I have, but it is impossible add them to Python classes -- all
class
> methods are treated as instance methods.
> This is especially unfortunate given that local functions now work
properly
> thanks to PEP 227.
> I was thinking that a new keyword "unbound" could be used to identify
> functions that belong to a class but do not expect "self" as their first
> attribute.
> From outside the class "unbound" methods would be invoked by specifying
the
> class name -- Foo.my_static_method (). From within the class they would
work
> similar to local functions that are defined at the class level. The __
trick
> should work for the "unbound" methods as well to allow for creation of
> private static methods.
>
> Examples:
> >>>class Foo:
>     """This is a sample class..."""
>
>     __myPrivateStaticVariable  = None
>
>     unbound getMyPrivateStaticVariable ():
>         """getMyPrivateStaticVariable (obj) -> obj
>
>         This method returns the value stored in private static variable...
>         """
>         return Foo.__myPrivateStaticVariable
>
>
>     unbound modifyMyPrivateStaticVariable (obj):
>         """modifyMyPrivateStaticVariable (obj) -> void
>
>         This method assigns obj to the private static variable...
>         """
>         Foo.__myPrivateStaticVariable = obj
>
> >>> type (Foo.modifyMyPrivateStaticVariable)
> <type 'unbound method'>
> >>> type (Foo.getMyPrivateStaticVariable ())
> <type 'None'>
> >>> Foo.modifyMyPrivateStaticVariable ("abc")
> >>> Foo.getMyPrivateStaticVariable ()
> "abc"
> >>>
>
>
>
>




From volucris at hotmail.com  Fri Mar 30 00:16:41 2001
From: volucris at hotmail.com (Voluciris)
Date: Thu, 29 Mar 2001 23:16:41 -0600
Subject: max filename size
Message-ID: <3ac416a6$0$4017$6e49188b@news.goldengate.net>

Is there away to find the maximum filename length of the current os? On
Windows it's 215 characters, but I don't want to hard code that in. It would
be much more pythonic if it was portable.

thanks
greg




From tim.one at home.com  Thu Mar  1 14:30:26 2001
From: tim.one at home.com (Tim Peters)
Date: Thu, 1 Mar 2001 14:30:26 -0500
Subject: A counter-counter-proposal for PEP 236: #pragma( nested_scopes)
In-Reply-To: <0Swn6.3324$TW.17061@tor-nn1.netcom.ca>
Message-ID: <LNBBLJKPBEHFEDALKOLCOEBKJDAA.tim.one@home.com>

[Warren Postma]
> Has anyone considered borrowing a C notation:
>
> #pragma( nested_scopes )
>
> Benefits:
>     - It's a comment therefore it doesn't break anything else.

But that's dead-wrong behavior for future_statements.  If your module needs
nested_scopes semantics, it will do unpredictable wrong things at runtime if
it doesn't get them.  That's why PEP 236 is emphatically *not* a "pragma"
proposal.  That doesn't mean a pragma facility isn't *also* desirable, but
does mean it's unsuitable for what PEP 236 is doing.  future_statements can
have profound effects on both syntax and semantics (indeed, that's what
they're *addressing*); they're not about pragmatics (from whence "pragma").

>     - It compiles fine and is ignored on pre-2.x systems.

Again dead-wrong behavior for future_statements.

>     - More pragmas can be added later, for warnings, code generation,
> etcetera.

Yes, pragmas can be useful for lots of things.  But not for cross-release
incompatible changes in semantics or syntax.




From Randy.L.Kemp at motorola.com  Thu Mar 22 09:03:23 2001
From: Randy.L.Kemp at motorola.com (Kemp Randy-W18971)
Date: Thu, 22 Mar 2001 08:03:23 -0600
Subject: Good Editor?
Message-ID: <E566B020833BD311B6610008C791A397050ECEAD@il93exm04.css.mot.com>

There are many good free, cheap editors.  You can look at emacs on Unix, Ultradev in Windows (www.ultradev.com (load down the python extension), or jedit.  I'm sure there are others. 

-----Original Message-----
From: Courageous [mailto:jkraska1 at san.rr.com]
Sent: Thursday, March 22, 2001 1:38 AM
To: python-list at python.org
Subject: Re: Good Editor?


On Wed, 21 Mar 2001 23:00:58 -0800, "fjs" <fjs6 at usa.net> wrote:

>Can someone recomend a good editor to write python in?  Im just getting
>started learning python and dont know which ones support python.

Vim has a language highliting for python and has friendly tab control.
I have mine set, for example, so that "every tab fills in enough spaces
to take you to the next tab stop." On all my python projects, actual tab
characters are completely forbidden.

C//




From jbauer at rubic.com  Fri Mar 16 20:21:24 2001
From: jbauer at rubic.com (Jeff Bauer)
Date: Fri, 16 Mar 2001 19:21:24 -0600
Subject: Guido on SlashDot
Message-ID: <3AB2BC14.73FA5F26@rubic.com>

Guido discussed on SlashDot:

  http://slashdot.org/articles/01/03/16/0349216.shtml

ministry-of-propaganda-dept-ly y'rs,

-Jeff



From chris.gonnerman at usa.net  Wed Mar  7 23:11:42 2001
From: chris.gonnerman at usa.net (Chris Gonnerman)
Date: Wed, 7 Mar 2001 22:11:42 -0600
Subject: News from Python9!!!
References: <NUCp6.65206$Z8.11697516@typhoon.southeast.rr.com>
Message-ID: <00ac01c0a785$efa17a00$a100000a@local>

I've just got one problem...  When I go to the links given, I get this:

<html><head><title>Site Not Found</title></head>
<body>No web site is configured at this address.</body></html>

Huh.  What happened?


----- Original Message -----
From: "Don Tuttle" <tuttledon at hotmail.com>
Newsgroups: comp.lang.python
To: <python-list at python.org>
Sent: Wednesday, March 07, 2001 9:21 PM
Subject: News from Python9!!!


> For those of you, like myself, who are NOT at the Python bash in Long
Beach,
> some very good news from there  has been posted at ActiveState.
>
> 1) The "Python Software Foundation" has been created
>
http://www.activestate.com/Corporate/Communications/Releases/Press983863928.
> html
>
> 2)The Programmers' Choice for Python award winner is Christian Tismer
>
http://www.activestate.com/Corporate/Communications/Releases/Press983917761.
> html
>
> 3) A "Python Cookbook" project has started.
>
http://www.activestate.com/Corporate/Communications/Releases/Press983583358.
> html
>
> 4)And a bugfix release of Python 2.0 has been released. It includes PPM
> (Programmer's Package Manager ) to make finding and installing Python
> modules simple.
>
http://www.activestate.com/Corporate/Communications/Releases/Press983916629.
> html
>
> Don





From JohnH at PHM.GOV.AU  Sun Mar 18 20:22:54 2001
From: JohnH at PHM.GOV.AU (Hirsch, John)
Date: Mon, 19 Mar 2001 12:22:54 +1100
Subject: Using ActiveX Control from Python
Message-ID: <A90DD8D7B82FD311B4D00008C7DBD6EAB5EBE2@phm-exch.phm.gov.au>

Hi All
I am trying to use an ActiveX Control from Python 2.0 with little success.
The Interactive session goes a little like this.

First I ran AwusbIO ActiveX Control module through makepy.py -i.

>>> AwusbIO ActiveX Control module
 {95438942-F98A-11D2-AC25-00104B306BEE}, lcid=0, major=1, minor=0

   >>> # Use these commands in Python code to auto generate .py support

   >>> from win32com.client import gencache

   >>> gencache.EnsureModule('{95438942-F98A-11D2-AC25-00104B306BEE}', 0, 1,
0)

>>> from win32com.client import gencache

>>> gencache.EnsureModule('{95438942-F98A-11D2-AC25-00104B306BEE}', 0, 1, 0)
<module 'win32com.gen_py.95438942-F98A-11D2-AC25-00104B306BEEx0x1x0' from
'c:\python20\win32com\gen_py
\95438942-F98A-11D2-AC25-00104B306BEEx0x1x0.pyc'>

>>> aw = win32com.client.Dispatch("AWUSBIO.AwusbIOCtrl.1")

>>> aw
<win32com.gen_py.AwusbIO ActiveX Control module._DAwusbIO>

>>> aw.AboutBox
<method _DAwusbIO.AboutBox of _DAwusbIO instance at 01F5B38C>

>>> aw.AboutBox()
Traceback (innermost last):
  File "<interactive input>", line 1, in ?
  File
"c:\python20\win32com\gen_py\95438942-F98A-11D2-AC25-00104B306BEEx0x1x0.py",
line 29, in AboutBox
    return self._oleobj_.InvokeTypes(0xfffffdd8, LCID, 1, (24, 0), (),)
com_error: (-2147418113, 'Unexpected failure', None, None)
>>> 

I am running ActiveState ActivePython2.0 on Win98.
The Control works fine with the supplied demo software from ActiveWire (not
to be confused with ActiveState).
They provide a demo that runs the control in a web browser using Java
script.

Any suggestions?




From mmealman at loki.thebisgroup.com  Fri Mar 23 12:29:21 2001
From: mmealman at loki.thebisgroup.com (Mark)
Date: 23 Mar 2001 11:29:21 -0600
Subject: Commercial Products in Python?
References: <3abb86ee.142862695@localhost>
Message-ID: <slrn9bn8tr.avv.mmealman@loki.thebisgroup.com>

In article <3abb86ee.142862695 at localhost>, Victor Muslin wrote:
>A friend of mine is considering developing a shrink-wrapped commercial
>product. Python seems like a possible choice, but it is not clear
>whether there are any precedents (though probably similar concerns
>apply to a Perl-based product).
>
>I would very much appreciate an advice in this area and possibly
>answers to some of the following questions:
>
>1. Are there any existing shirnk-wrap commercial products written in
>Python?

Hard to say. Most shrink wrap developers don't advertise what language they
code in.

But there's zero reason against using Python for commercial apps. I'd
personally feel more comfortable using Python than say, Java, in a
commercial app. IMHO Python would have fewer deployment issues.

>3. What is the best way to go about it? Is it better to have customers
>install Python and give them bytecode files or use one of the programs
>that creates an executable?

I'd try to compile Python down to an executable. It'd probably make it
easier to deploy.

That's assuming you're targetting Windows.

>4. There is an easy shrink-wrap install for Python (at least on
>Windows NT). Are there such on Unix/Linux platforms? (Typically
>customers don't like having to compile software from source).

Compile Python down to an exe for 'doze and install using Install Shield or
something similar.

That way it looks nice and professional.

For Linux I'd probably pop my program into a .rpm and just list the
dependencies. I'd also create a .deb because I'm a Debian user and shipping
in multiple package dists shows some taste.

>6. Are there any licensing issues to be aware of?

Just pop in a seperate copyright notice somewhere for Python and you should
be covered.

>Anything else that comes to mind in this direction would be very
>helpful.

Yeah, who's your target audience?

If you're expecting John Q Public running Windows 95 to use this, then it
needs to be easy enough for a monkey to install and the install has to look
pretty.

But if you're targeting developers you can get a little more flexible, we
appreciate multiple methods of installing software.

-Mark


-----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
-----==  Over 80,000 Newsgroups - 16 Different Servers! =-----


From winkler at abaqus-sn.com  Thu Mar 29 13:19:07 2001
From: winkler at abaqus-sn.com (david winkler)
Date: Thu, 29 Mar 2001 18:19:07 GMT
Subject: how query function args
Message-ID: <3AC37CAB.50937BD6@abaqus-sn.com>

I need to query the names and values of a function's keyword arguments.
I
suspect that there is an easy way to do this but cannot find it.  Would
someone care to make a suggestion?



From phrxy at csv.warwick.ac.uk  Thu Mar 29 13:31:41 2001
From: phrxy at csv.warwick.ac.uk (John J. Lee)
Date: Thu, 29 Mar 2001 19:31:41 +0100
Subject: Linux GPS program
In-Reply-To: <99sud0$m7a$1@newsserver.rrzn.uni-hannover.de>
References: <3AC01637.4EF3A6F2@crynwr.com>    <99q3ob$ehr$1@newsserver.rrzn.uni-hannover.de>    <3AC0D79E.ADED2832@crynwr.com> <99sud0$m7a$1@newsserver.rrzn.uni-hannover.de>
Message-ID: <Pine.SOL.4.30.0103291926580.13267-100000@primrose.csv.warwick.ac.uk>

On 28 Mar 2001, Bernhard Reiter wrote:

> In article <3AC0D79E.ADED2832 at crynwr.com>,
> 	Russell Nelson <nelson at crynwr.com> writes:
> > Bernhard Reiter wrote:
[...]
> If you want it to be Free Software,
> attach a license (maybe MIT style or LGPL).
[...]
> > It's free software now -- I just don't want rogue versions that would be
> > hard for me to integrate back into the mainline branch.  Feel free
> > to suggest changes.
[...]

Erm, are you two sure you're not talking at cross-purposes?  Does Russell
know the usual meaning of 'Free Software' is free as in freedom rather
than beer?

try

http://www.gnu.org/philosophy/free-sw.html

and

http://www.opensource.org/


John



From lbates at syscon-computers.com  Thu Mar  1 12:11:49 2001
From: lbates at syscon-computers.com (Larry Bates)
Date: Thu, 1 Mar 2001 11:11:49 -0600
Subject: Newbie createing .PYC for import
Message-ID: <3a9e83fa$1@news1.dbtech.net>

I have spent the last couple of hours (unsuccessfully) attempting to figure
out how to generate a .PYC file for a module that I've defined.  Everything
appears to imply that it's "automatic", but I'm not getting anywhere.  I
have a module and I wish to generate a .PYC file so that I can import it
into future projects.  Can anyone shed any light on how to accomplish this?

-Larry




From qrczak at knm.org.pl  Thu Mar  8 15:19:59 2001
From: qrczak at knm.org.pl (Marcin 'Qrczak' Kowalczyk)
Date: 8 Mar 2001 20:19:59 GMT
Subject: I come to praise .join, not to bury it...
References: <mailman.983485923.1136.python-list@python.org> <97p7iq$kk2$1@nntp6.u.washington.edu> <97rka002k9@news1.newsguy.com> <3AA2F4B3.A7496E76@cosc.canterbury.ac.nz> <97vmoh061p@news2.newsguy.com> <3AA46134.D385208B@cosc.canterbury.ac.nz> <98292h0s0g@news1.newsguy.com> <3AA58BD0.3324C709@cosc.canterbury.ac.nz> <9850ge013ns@news1.newsguy.com> <gRup6.364652$ge4.128679630@news2.rdc2.tx.home.com> <qZvp6.29197$ML1.1439945@e420r-atl2.usenetserver.com>
Message-ID: <slrn9afqbf.tth.qrczak@qrnik.zagroda>

Wed, 7 Mar 2001 14:24:38 -0500, Steve Holden <sholden at holdenweb.com> pisze:

> Surely no type conversion is involved: the interpreter just converts
> 
>     instance.method(*args)
> 
> into
> 
>     class.method(instance, *args)

No. This is what I would like it to do, and proposed a few weeks ago.
A conversion happens during extraction from a class, and only if it's
a function (not function-like object).

If C is a class, then after
    foo = whatever
    C.foo = foo
    foo = C.foo
foo is the same as whatever *unless* it was a function, in which case
it got converted to an unbound method (and can only accept C instances
as the first argument).

If you try to implement a method using a function-like object instead
of a function, it won't work.

-- 
 __("<  Marcin Kowalczyk * qrczak at knm.org.pl http://qrczak.ids.net.pl/
 \__/
  ^^                      SYGNATURA ZAST?PCZA
QRCZAK


From jmarshal at mathworks.com  Fri Mar 16 12:19:06 2001
From: jmarshal at mathworks.com (Joshua Marshall)
Date: 16 Mar 2001 17:19:06 GMT
Subject: list.sort(cmpfunc) question
References: <m1ae6lg5qt.fsf@pclabs188.labs.trema.com> <98t9dq$qvu$1@panix6.panix.com> <98tf9d$g3v$1@news.mathworks.com> <3AB249FB.131080CD@san.rr.com>
Message-ID: <98thua$i40$1@news.mathworks.com>

Darren New <dnew at san.rr.com> wrote:
> Joshua Marshall wrote:
>> Seems the word "should" should never appear in documentation.

> Not quite true. RFC's use "should" all the time, but there's a very strict
> definition for what it means. I think "usually" is the other problematic
> word. :-)

What does "should" mean, in the context of RFCs?

Although an RFC is different than documentation.


From aahz at panix.com  Sat Mar 24 10:51:53 2001
From: aahz at panix.com (Aahz Maruch)
Date: 24 Mar 2001 07:51:53 -0800
Subject: New PEP: The directive statement
References: <mailman.985414232.22176.python-list@python.org>
Message-ID: <99ifqp$jni$1@panix3.panix.com>

In article <mailman.985414232.22176.python-list at python.org>,
Tim Peters <tim.one at home.com> wrote:
>[James_Althoff]
>> ok, how about:
>>
>> from  __readyOrNot__  import nested_scopes
>>
>> or perhaps,
>>
>> from  __likeItOrNot__  import nested_scopes
>
>[Aahz Maruch]
>> This is in some respects excessively flippant, but that very flippancy
>> calls attention to it.  I'm not going to push this idea, but I do think
>> it's worth considering.
>
>Aahz, "__future__" was intended to be flippant enough to get peoples'
>attention, and for some people that works (e.g., /F originally said "it
>makes me smile", which was half the point).  But the people offended by
>an ounce of irreverence aren't going to be happier with a full pound.

Yes, I understand.  But "__future__" isn't excessively flippant, so I
thought it might be worth considering pushing the flippancy factor even
higher.  Or not.  But this *is* Python, after all.
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

"I won't accept a model of the universe in which free will, omniscient
gods, and atheism are simultaneously true."  -- M


From sinisa_pavlovic at yahoo.com  Thu Mar 29 06:55:08 2001
From: sinisa_pavlovic at yahoo.com (Simpa)
Date: Thu, 29 Mar 2001 11:55:08 GMT
Subject: Newbie Tkinter problem
Message-ID: <3ac32211.18968775@news.iskon.hr>

Hi.

When I start the following script twice, without leaving Python I got
TclError message (first time everything is OK). What am I doing wrong?

I am using Python 1.5.2 for Win32 with Mark Hammond's PyWin, but same
things happend on IDLE too.

Thanks in advance.
Simpa.




from Tkinter import *

class Wind:
    def __init__(self):
        self.root = Tk()
        self.root.title('Show image')
        
        self.c = Canvas(self.root,background = 'green')
        self.photo = PhotoImage(name = 'wedda',file="wedda.gif")
        self.c.create_image(0,0,image=self.photo,anchor=NW)
        self.c.pack(fill=BOTH,expand=1)

        self.root.mainloop()

MyWind = Wind()



>>> Traceback (innermost last):
  File "D:\PythonW32\Pythonwin\pywin\framework\scriptutils.py", line
237, in RunScript
    exec codeObject in __main__.__dict__
  File "D:\Private\PythonProba\Tkinter\Image.py", line 17, in ?
    MyWind = Wind()
  File "D:\Private\PythonProba\Tkinter\Image.py", line 11, in __init__
    self.photo = PhotoImage(name = 'wedda',file="wedda.gif")
  File "D:\Python\Lib\lib-tk\Tkinter.py", line 1874, in __init__
    apply(Image.__init__, (self, 'photo', name, cnf, master), kw)
  File "D:\Python\Lib\lib-tk\Tkinter.py", line 1839, in __init__
    self.tk.call(('image', 'create', imgtype, name,) + options)
TclError: can't invoke "image" command:  application has been
destroyed



From jkraska1 at san.rr.com  Wed Mar 21 19:05:12 2001
From: jkraska1 at san.rr.com (Courageous)
Date: Thu, 22 Mar 2001 00:05:12 GMT
Subject: New PEP: The directive statement
References: <mailman.985085006.24236.python-list@python.org> <slrn9bi7q3.2sa.hzhu@rocket.knowledgetrack.com>
Message-ID: <pogibt8r69iqs7v2fv6nqfjjfm2lj5gafs@4ax.com>

>from __experimental__ import stackless

While I understand what you're trying to say, stackless requires alterations
to the base runtime environment in which python lives. This is not really
amenable to an import.


C//



From n8gray at caltech.edu.is.my.e-mail.address  Tue Mar 20 22:31:04 2001
From: n8gray at caltech.edu.is.my.e-mail.address (Nathan Gray)
Date: Tue, 20 Mar 2001 19:31:04 -0800
Subject: New PEP: The directive statement
References: <mailman.985085006.24236.python-list@python.org>
Message-ID: <3AB82078.8060005@caltech.edu.is.my.e-mail.address>

Martin von Loewis wrote:

> Please have a look at the PEP below. It is intended to replace the
> __future__ import suggested in PEP 236.

Excellent work, Martin.  As it stands, I would prefer this to __future__
on grounds of clarity alone.  Last time I checked, though, my name wasn't 
Guido, so you can take that for what it's worth.  Thanks for putting your
money where your mouth was.  Or putting your additional keystrokes where 
your original keystrokes were.  Usenet stretches the boundaries of our 
cliches...

I've got a couple of comments about the design, though.

> 
> Syntax
> 
>     A directive_statement is a statement of the form
> 
>         directive_statement: 'directive' NAME [atom] [';'] NEWLINE
> 
>     The name in the directive indicates the kind of the directive; it
>     defines whether the optional atom can be present, and whether
>     there are further syntactical or semantical restrictions to the
>     atom. In addition, depending on the name of the directive, certain
>     additional syntactical or semantical restrictions may be placed on
>     the directive (e.g. placement of the directive in the module may be
>     restricted to the top of the module).

This syntax works but leaves little room for the future (no pun intended).  
Only one directive can be placed on a line, and using whitespace to link NAME
with its possible atom seems inconsistent with the way similar constructs
in Python work. (e.g. import general.lessgeneral.specific)

Would you consider using this sort of syntax instead?

   directive_statement: "directive" directive_ID ("," directive_ID)* 
                            [";"] NEWLINE
   directive_ID:        identifier ("." identifier)*

I understand that there are those who want to keep things as dead-simple as
possible, and there are good reasons for *not* wanting people to put multiple
directives on one line, but at the very least consider changing the syntax to 
NAME ["." atom] to allow for more elegant extension in the future if we 
decide that this functionality is desirable.

My second comment is that you may want to consider adopting a form of the 
__future__ proposal's runtime semantics for directive statements.  Perhaps 

    directive transitional.nested_scopes

could create a "_Feature object" called nested_scopes identical to the one
proposed in "back to the __future__".

    print nested_scopes.getOptionalRelease()   # prints "(2, 1, 0, 'beta', 1)"

Note that not all directive statements would need to have runtime 
semantics.  In the case of "transitional", though, it seems like a 
useful idea.

Well, that's all for now.  Lest it sound like I'm blasting Martin's PEP, 
I should reiterate that I think it's the way to go.  I've just got a 
quibble and a suggestion here and there.

Thanks again, Martin,

-n8

-- 
_.~'^`~._.~'^`~._.~'^`~._.~'^`~._.~'^`~._
               Nathaniel Gray
     California Institute of Technology
      Computation and Neural Systems
       n8gray <at> caltech <dot> edu
_.~'^`~._.~'^`~._.~'^`~._.~'^`~._.~'^`~._



From fredrik at pythonware.com  Wed Mar 21 12:18:56 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 21 Mar 2001 17:18:56 GMT
Subject: Stupid String Question
References: <3AB8327C.493C70A5@direct.ca>
Message-ID: <4o5u6.907$sk3.295107@newsb.telia.net>

"Brian & Colleen" wrote:
> What is the pretty way to generate a string of 30 asterisks.

how about:

a = "*" * 30

Cheers /F




From mcalla at home.com  Wed Mar  7 21:54:38 2001
From: mcalla at home.com (Mike Callahan)
Date: Thu, 08 Mar 2001 02:54:38 GMT
Subject: Simple Tkinter question
References: <DKXo6.12857$YF4.3052373@news1.rdc1.tn.home.com> <3AA482C0.D92E3CC4@theworld.com>
Message-ID: <OvCp6.23402$YF4.4646076@news1.rdc1.tn.home.com>

Oh, so if I use root.quit() and then follow it up with root.mainloop() it
would start the GUI again? Where if I use root.destroy(), it destroys the
widgets and also quits the loop with no chance of starting it again unless I
create another root. Is that correct?

"David Lees" <deblNoNospammy at theworld.com> wrote in message
news:3AA482C0.D92E3CC4 at theworld.com...
> As I understand it, .quit simply exits the main loop, but leaves all
> widgets and everything else you have created, while .destroy removes the
> widgets you have created from the namespace.
>
> david lees
>
> Mike Callahan wrote:
> >
> > I know this is probably a stupid question but after reading the Tkinter
> > documentation I am still confused between root.quit() and
root.destroy().
> > What are the differences between these two methods? Thanks.
> >
> > Mike Callahan




From dsavitsk at e-coli.net  Fri Mar  9 01:29:30 2001
From: dsavitsk at e-coli.net (dsavitsk)
Date: Fri, 9 Mar 2001 00:29:30 -0600
Subject: Another Tkinter question
References: <PUPp6.14231$tV.1448780@newsrump.sjc.telocity.net> <slrn9afn9a.ps.matt@happy-hour.mondoinfo.com>
Message-ID: <iL_p6.14466$tV.1946120@newsrump.sjc.telocity.net>

Wow ... Amazing ... Thanks.  I think I just needed some confidence to try
what VB would never do for me.

ds


"Matthew Dixon Cowles" <matt at mondoinfo.com> wrote in message
news:slrn9afn9a.ps.matt at happy-hour.mondoinfo.com...
> On Thu, 8 Mar 2001 12:08:47 -0600, dsavitsk <dsavitsk at e-coli.net>
> wrote:
>
> >I have a small tkinter app that contains a (PMW) ScrolledFrame which
> >in turn has a grid of entry widgets (it sort of emulates a small
> >spread sheet).  What I would like to do is add a new row of entry
> >widgets while the thing is running, and I'm not sure how to go about
> >this as I've already .packed them in.
>
> Tkinter is very dynamic: that your entry widgets have already been
> packed shouldn't make a difference. Try just adding more. If that
> doesn't work, I'd need to see your code to figure out what's up.
>
> >(Also, if anyone knows a better way to do this than tons of entry
> >widgets that would be great.)
>
> If the easy way works, I'd do it the easy way. Naturally, it's also
> possible to move the data around in a bunch of static widgets. That a
> bit annoying since you have to do your own scrolling. If you'd like to
> see some code of mine that does that, I'd be glad to point you at it.
> I've seen combinations of the two approaches done as well, but that
> was ugly.
>
> Regards,
> Matt




From emile at fenx.com  Wed Mar 14 12:04:06 2001
From: emile at fenx.com (Emile van Sebille)
Date: Wed, 14 Mar 2001 09:04:06 -0800
Subject: Who am I: can a class instance determine its own name?
References: <mailman.984034150.9635.python-list@python.org> <DYV4H3fqAHA.187@typhoon.i3s.net> <mailman.984324974.25480.python-list@python.org> <vZJg10qqAHA.210@typhoon.i3s.net> <xqxzoeol3ju.fsf@cola.ca.boeing.com>
Message-ID: <98o8fu$2s4dj$1@ID-11957.news.dfncis.de>

Has locals changed so that it may be written to reliably?  My documentation
still warns:

locals ()
Return a dictionary representing the current local symbol table. Warning:
The contents of this dictionary should not be modified; changes may not
affect the values of local variables used by the interpreter.

Emile van Sebille
emile at fenx.com



"Harry George" <hgg9140 at cola.ca.boeing.com> wrote in message
news:xqxzoeol3ju.fsf at cola.ca.boeing.com...
> class MyName:
>     #--------------------------
>     def init(self,name):
>         self.name=name
>         try:
>             raise None
>         except:
>             c=sys.exc_info()[2].tb_frame.f_back.f_back
>         c.f_locals[name]=self
> ...
>
> MyName("me")
> print me.name
>
> "Kevin Cazabon" <kevin_cazabon at hotmail.nospamplease!.com> writes:
>
> > The reason I say it's not elegant is that there is the possibility (read
> > "probability") that the assigned name and the actual name can differ.
> >
> > We're assuming that the programmer assigns exactly the same name to
both,
> > which doesn't seem like a hard task, but it introduces the possibility
of
> > problems, and probably hard ones to debug too.
> >
> > If there were an automatic method, it would be MUCH more reliable.
> >
> > Just MHO.
> >
> > Kevin.
> >
> >
> >
> > "Paul Prescod" <paulp at ActiveState.com> wrote in message
> > news:mailman.984324974.25480.python-list at python.org...
> > | Kevin Cazabon wrote:
> > | >
> > | > ...
> > | > You could also
> > | > make it an initialization parameter.
> > | >
> > | > FooBar = Foo(name="FooBar")
> > | >
> > | > Just trying to help... not elegantly though.
> > |
> > | I disagree Kevin. This is not only the canonical way to give an
instance
> > | a name, it is also the elegant way. It gives you a lot more
flexibility
> > | because the name it has at runtime is unrelated to the variable name
you
> > | happened to have given it in your code. If you need to change one or
the
> > | other for any reason, you can do so safely.
> > |
> > |  Paul Prescod
> > |
> >
> >
>
> --
> Harry George                E-mail:  harry.g.george at boeing.com
> The Boeing Company          Renton:  (425) 237-6915
> P. O. Box 3707  02-CA       Everett: (425) 266-3868
> Seattle, WA 98124-2207      Page:    (425) 631-8803




From quickdry at users.sourceforge.net  Sun Mar 11 01:36:47 2001
From: quickdry at users.sourceforge.net (Steven Adams)
Date: Sun, 11 Mar 2001 06:36:47 GMT
Subject: Python & Wireless Apps
References: <98b4ra$107$1@bashir.peak.org>
Message-ID: <32Fq6.10772$pG5.38040@news1.rdc1.nsw.optushome.com.au>

"prince.kattare.com" <louis at decipherinc.com> wrote in message
news:98b4ra$107$1 at bashir.peak.org...
> Hello!   I just got back from the Python Conference and glad to see
that
> there is a lot of motivation for the language.  I am just starting
out with
> python and was wondering how it's being used in the wireless space,
meaning
> mostly palm devices.  Has anyone ported an app to a palm pilot?
What kind
> of restrictions should I be considering when thinking about this?
>
> Thanks for any insight you might have!
>
> Louis
> louis at decipherinc.com

Python has been around on the WindowsCE platform for quite some time,
and is really handy -

http://starship.python.net/crew/skippy/ce/

while I haven't done much serious work on it, my advice would be to
really pay attention to what data structures you're going to use.
Efficiency in terms of execution speed and size is very important,
where convenience is fine on the PC, on a 20MHz Palm you'll feel the
hit that the convenience brings.

IIRC pippy (python's palm port) doesn't support floating point ops,
and there are a few other limitations.

Happy hacking,

Steven





From ^NOSPAM^ifi at ifi.it  Wed Mar  7 09:54:50 2001
From: ^NOSPAM^ifi at ifi.it (accattatevillo)
Date: Wed, 07 Mar 2001 15:54:50 +0100
Subject: Using a jar with jython
References: <3AA6115E.7B2ACDEE@ifi.it> <3aa62b59_1@nnrp1.news.uk.psi.net>
Message-ID: <3AA64BBA.DCECD14F@ifi.it>

> What are the problems that you are having? Error messages?
> 
> BTW, I'm cross-posting this the Jython on AS/400 mailing list. If
> you are using Jython to access an AS/400, you might want to join -
> see <http://groups.yahoo.com/group/JythonAS400>.

I found the problem: I was sleeping when I wrote JT400.JAR instead of
jt400.jar!!!

Thanks a lot.

Regards
Stefano

P.S.: I'll join the group you told me about


From jdries at mail.com  Thu Mar  8 20:23:02 2001
From: jdries at mail.com (Jan Dries)
Date: Fri, 09 Mar 2001 02:23:02 +0100
Subject: Can anyone recomend a good intoduction to C...
References: <mailman.983894288.19965.python-list@python.org> <3AA5EF59.507F1D91@olsen.ch> <983958878.976344@newsmaster-04.atnet.at> <77up6.1289$y6.235541@ruti.visi.com> <985vqs01ipr@drn.newsguy.com> <ewwp6.1466$y6.241005@ruti.visi.com> <slrn9aeblb.cg1.scarblac@pino.selwerd.nl> <djagat4kk3cs9hg0c3k39cudk6goblhkg6@4ax.com>
Message-ID: <3AA83076.B968CCED@mail.com>

Courageous wrote:
> 
> >All this bloat added to C++ was not necessary for making an OO language.
> 
> Echo that. What's "object oriented" about templates?
> 
> A: NOTHING.
> 

If Stroustrup's goal had been to come up with an object oriented
language, and nothing more, he might just as well have produced
SmallTalk.
Templates may not be OO per se, but that does not mean they're of no
use. OO is just one of around 40 commonly used architectural styles to
develop software. OO is a means of achieving some goal, not a goal in
itself, and templates are simply another technique, complementary to OO,
to assist programmers in achieving the goal of developing robust,
reusable, entensible software. Quoting from "The ANSI/ISO C++
Professional Programmer's Handbook": "Generic programming proceeds one
step beyond object oriented programming in pursuing reusability. Two
important features of C++, templates and operator overloading, are the
basis of generic programming."

Regards,
Jan



From dog at ERC.MsState.Edu  Tue Mar 13 22:11:05 2001
From: dog at ERC.MsState.Edu (David OGwynn)
Date: Tue, 13 Mar 2001 21:11:05 -0600 (CST)
Subject: Problems with Irix (stack overload)
Message-ID: <200103140311.f2E3B5i05941@ERC.MsState.Edu>

I'm building a visualization tool for debugging polygonal mapping trees using 
python and the python/c API. In the extension module, I have both togl and glut 
interfaces. On SUN, everything works swimmingly. On SGI/irix, I'm having 
problems. To preface, the module takes a binary file, parses it and builds an  
"Octree" object. The parser then inserts facets and octants (both singly linked 
lists of c structs). Then the preparedraw Octree object function builds a set of 
OpenGL Display lists by marching down the facet and octant lists. I say all this 
to give exposition to my delema. 

Like I said, solaris works like a dream for both glut and togl. SGI, though, 
throws up on the togl preparedraw function. When I run preparedraw function, it 
segfaults and tells me that the program has exceeded its stack allotment. I 
tried increasing the stack allotment to 500M, but that still didn't work. Here's 
another twist: I tried rloging into the same machine and everything worked fine. 
So, I can run the code if I am rloging but not when I'm on the machine.

Can anyone give me a clue?
_____________
David O'Gwynn
dog at erc.msstate.edu

    "Life begins where math breaks down."
	--B.D.O.




From no-spam-popov at bigserver.com  Sun Mar 11 15:05:39 2001
From: no-spam-popov at bigserver.com (Boris Popov)
Date: Sun, 11 Mar 2001 12:05:39 -0800
Subject: Einstein's Riddle
References: <oNPq6.396644$ge4.138479566@news2.rdc2.tx.home.com>
Message-ID: <98glrd$1si1a$1@ID-75892.news.dfncis.de>

"Steve Graham" <js.graham at home.com> wrote in message
news:oNPq6.396644$ge4.138479566 at news2.rdc2.tx.home.com...
> Who would be interested in using his/her brain (and his computer) to solve
> the following?
>
>
> Steve Graham
>
> ===
>
> Einstein's Riddle
>
> Albert Einstein wrote this riddle this century [ed. 20th century].  He
said
> 98% of the world could not solve it.
>
> There are 5 houses in 5 different colors.  In each house lives a person
with
> a different nationality.  The 5 owners drink a certain type of beverage,
> smoke a certain brand of cigar, and keep a certain pet.  No owners
> have the same pet, smoke the same brand of cigar or drink the same
beverage.
>
> The question is: "Who owns the fish?"
>
> Hints:
>
> The Brit lives in the red house.
>
> The Sweed keeps dogs as pets.
>
> The Dane drinks tea.
>
> The green house is on the left of the white house.
>
> The green house's owner drinks coffee.
>
> The person who smokes Pall Mall rears birds.
>
> The owner of the yellow house smokes Dunhill.
>
> The man living in the center house drinks milk.
>
> The Norwegian lives in the first house.
>
> The man who smokes Blends lives next to the one who keeps cats.
>
> The man who keeps the horse lives next to the man who smokes Dunhill.
>
> The owner who smokes Bluemasters drinks beer.
>
> The German smokes Prince.
>
> The Norwegian lives next to the blue house.
>
> The man who smokes Blends has a neighbor who drinks water.
>

I don't beleive he said that only 2% of the world could solve it. It's not
hard, you just need to pay attention to the details. The German owns the
fish. It was supposed to be a table that I used to layout all the details,
but I prefer to post plain-text:

house         1 2 3 4 5
color            yellow blue red green white
nationality   norweigian dane brit german swede
drink            water tea milk coffee beer
smoke        dunhill blend pall mall prince bluemaster
pet              cats horse birds fish dogs

-Boris
P.S. Did I have to write software to solve it ? :)




From chris at home.com  Fri Mar 30 19:52:03 2001
From: chris at home.com (Chris)
Date: Sat, 31 Mar 2001 00:52:03 GMT
Subject: Know of Substantial Apps Written in Python?
References: <3ac4e4a8_1@news4.newsfeeds.com> <mailman.985984122.20341.python-list@python.org>
Message-ID: <3AC529CD.70406@home.com>

Have a look at the Gadfly database engine.  Not all that "large", but 
complex, as demonstrates the power of python.


Moshe Zadka wrote:

> On Fri, 30 Mar 2001 12:02:14 -0800, "Dry Ice" <nomail at nomail.com> wrote:
> 
> 
>> Know of substantial apps written in Python?
> 
> 
> Two big ones come to mind: Mailman (www.list.org) and Zope (www.zope.org)
> Oh, and Sketch (sketch.sourceforge.net)




From greg at cosc.canterbury.ac.nz  Sun Mar  4 21:06:43 2001
From: greg at cosc.canterbury.ac.nz (Greg Ewing)
Date: Mon, 05 Mar 2001 15:06:43 +1300
Subject: I come to praise .join, not to bury it...
References: <mailman.983485923.1136.python-list@python.org> <97p7iq$kk2$1@nntp6.u.washington.edu> <97rka002k9@news1.newsguy.com>
Message-ID: <3AA2F4B3.A7496E76@cosc.canterbury.ac.nz>

Alex Martelli wrote:
> 
> I have not heard ANY _technical_ arguments opposed to
> this last time the discussion went around -- nothing but
> vague aesthetic, "it should be the other way", "I find it
> ugly" kind of complaints.

While your arguments are _technically_ sound, we are yet
to be convinced that the benefits outweigh the drawbacks
of being forced to use what many find to be an obscure and 
hard-to-follow syntax for all string joining.

The vast majority of string-joining operations in Python
code that I've seen and written haven't had any need for
polymorphism at all. On the rare occasions when I want a
polymorphic joiner[1], I'm quite willing to build the
necessary machinery and use whatever syntax is necessary.
But I don't want to pay the price of having to use that
syntax *every* time I join strings!

Footnotes:
[1] So rare, in fact, that so far I've *never* wanted one.
    (Although I *have* written functions to do the join-the-
    last-pair-with-something-different thing, I've never had
    to choose between the two polymorphically.)

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


From petasis at iit.demokritos.gr  Mon Mar 12 01:54:23 2001
From: petasis at iit.demokritos.gr (Petasis George)
Date: Mon, 12 Mar 2001 08:54:23 +0200
Subject: Einstein's Riddle
References: <oNPq6.396644$ge4.138479566@news2.rdc2.tx.home.com>
Message-ID: <3AAC729F.77068663@iit.demokritos.gr>

Well, I saw the same in Greek a month ago and it was somewhat
unclear in its formation. Now that I see the English version, the ambiguity
remains. If we assume that the fifth owner owns the fish, then:

There is only one solution if you *asume* that
"The green house is on the left of the white house." means that
the green house in next to the white house, on the left side.
On the other hand, if you assume that the green house is on the left
of the white house, but other houses can be in between, then the are more
than one solutions. In a simple C++ program that a friend wrote to find
all solutions, it find at least 8.

Supposing the first case (green house next to white, on the left), everyone
that tried to solve it succeded. Ok, some people simply need more time.
Even my 16-year little brother manage to solve it. So, where is this 98%
not able to solve it?

But unfortunately, I have to agree with Greg Jorgensen: the way it is
stated is not solvable:-) You have to make assumptions for solving it...

George

Steve Graham wrote:
> 
> Who would be interested in using his/her brain (and his computer) to solve
> the following?
> 
> Steve Graham
> 
> ===
> 
> Einstein's Riddle
> 
> Albert Einstein wrote this riddle this century [ed. 20th century].  He said
> 98% of the world could not solve it.
> 
> There are 5 houses in 5 different colors.  In each house lives a person with
> a different nationality.  The 5 owners drink a certain type of beverage,
> smoke a certain brand of cigar, and keep a certain pet.  No owners
> have the same pet, smoke the same brand of cigar or drink the same beverage.
> 
> The question is: "Who owns the fish?"
> 
> Hints:
> 
> The Brit lives in the red house.
> 
> The Sweed keeps dogs as pets.
> 
> The Dane drinks tea.
> 
> The green house is on the left of the white house.
> 
> The green house's owner drinks coffee.
> 
> The person who smokes Pall Mall rears birds.
> 
> The owner of the yellow house smokes Dunhill.
> 
> The man living in the center house drinks milk.
> 
> The Norwegian lives in the first house.
> 
> The man who smokes Blends lives next to the one who keeps cats.
> 
> The man who keeps the horse lives next to the man who smokes Dunhill.
> 
> The owner who smokes Bluemasters drinks beer.
> 
> The German smokes Prince.
> 
> The Norwegian lives next to the blue house.
> 
> The man who smokes Blends has a neighbor who drinks water.


From xeger at usa.net  Wed Mar  7 00:15:10 2001
From: xeger at usa.net (Regular Expression)
Date: Wed, 07 Mar 2001 05:15:10 GMT
Subject: C/C++ extension: subclassing an extension type with Python code?
Message-ID: <ytjp6.25967$jg1.598879@news1.alsv1.occa.home.com>

I'd like to create an extension type which can be further extended by end
users of the type. Ideally, I'd like to define certain properties that are
read-only, and others that cannot be occluded in "subclasses" of the type.
I'd also like to define certain methods that cannot be overridden.

Does anybody have an idea how to do this?




From aleaxit at yahoo.com  Fri Mar 23 18:47:01 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sat, 24 Mar 2001 00:47:01 +0100
Subject: PEP 245: Python interfaces
References: <mailman.985377931.22542.python-list@python.org>
Message-ID: <99gnb306gu@news2.newsguy.com>

"Michel Pelletier" <michel at digicool.com> wrote in message
news:mailman.985377931.22542.python-list at python.org...
    [snip]
> >    d) What is also missing from Python is
> >       an "auto-delegation".  An analysis
> >       of how this fits in or doesn't fit
> >       in would be helpful.
>
> What is that?  I've neve heard of auto-delegation.

More properly "automatic delegation" -- in Python, you
code it up in __getattr__; in my dream-version of Java
(not existing afaik) you could just declaratively state
"and all methods of interfaces X and Y which I have
not implemented just delegate to the following expression
based on my attributes", with at least interface
granularity (if not method by method).


Alex





From careye at spamcop.net  Thu Mar 22 05:06:38 2001
From: careye at spamcop.net (Carey Evans)
Date: 22 Mar 2001 22:06:38 +1200
Subject: Evaluating the contents of a string
References: <snitl5r6oy.fsf@motorola.com>
	<mwst6.4525$Dp6.681183@news-rep.ab.videon.ca>
Message-ID: <87lmpy6rtd.fsf@psyche.dnsalias.org>

"Matt Barnett" <mbarnett at ualberta.ca> writes:

> In article <snitl5r6oy.fsf at motorola.com>, Aaron Ginn 
> <aaron.ginn at motorola.com> wrote:

> > Is there a simple way to evaluate the contents of a string assuming
> > that string contains only nunbers and mathematical operators?

[...]

> eval("1.1 * 1.15")

You can do that more safely, by passing in a globals() argument:

>>> eval('1.1 * 1.15', {'__builtins__': {}})
1.2649999999999999

Without this (or checking the string's syntax), something like this is
possible (don't try this at home):

>>> eval(r'__import__("os").system(r"echo I\'m so worried | mail root")')
0
>>> eval(r'__import__("os").system(r"echo I\'m so worried | mail root")',
...   {'__builtins__': {}})
Traceback (innermost last):
  File "<stdin>", line 2, in ?
  File "<string>", line 0, in ?
NameError: __import__

You could allow just mathematical operations, too:

>>> import math
>>> gl = {'__builtins__': {}}
>>> for k in dir(math):
...   if k[0] != '_':
...     gl[k] = getattr(math, k)
...
>>> eval('cos(2 * pi)', gl)
1.0

More advanced forms of this are handled by the rexec and Bastion
modules.

If this goes beyond what Aaron wants, I'm sure there's someone out
there that found it interesting.

-- 
	 Carey Evans  http://home.clear.net.nz/pages/c.evans/

	    "Quiet, you'll miss the humorous conclusion."


From cribeiro at mail.inet.com.br  Wed Mar 28 13:31:55 2001
From: cribeiro at mail.inet.com.br (Carlos Alberto Reis Ribeiro)
Date: Wed, 28 Mar 2001 15:31:55 -0300
Subject: What *instance* called me & proxy methods
In-Reply-To: <003701c0b7b2$039b1250$04090a0a@upcast.com>
References: <3AC1BE92.ACE82741@ua.ac.be>
 <5.0.2.1.0.20010328104602.021cfc40@mail.inet.com.br>
Message-ID: <5.0.2.1.0.20010328152624.021a45e0@mail.inet.com.br>

At 10:07 28/03/01 -0800, Ken Seehof wrote:
>You want 'im_self'.

Ken, I checked this *before* posting my first message. In fact I cited this 
in my original message... First, in this paragraph, where I'm stating that 
I'm not looking for self:

> > Can I know what *instance* called a method? I'm not talking about "self"
> > here, because we *can* call a method using other techniques, such as
> > copying the instance method pointer directly:

And then here, when I said that I know about the im_self attribute (and 
about its companions as well).

> > In both cases, hello() received the same value as the self parameter. This
> > is exactly what I need; in fact, the im_self attribute of the instance
> > method stores the pointer to self, which explains this behavior.

Please check my examples. My question is not about how to know about "self" 
or "im_self". When I attach a instance method to my proxy, the value of 
im_self is kept the same(see the first example). I would like to have some 
way to know if the call was redirected through my proxy. My second example 
shows how I did manage to achieve it, but I'm not satisfied with the solution.


Carlos Ribeiro





From jeremy at alum.mit.edu  Wed Mar 14 17:37:18 2001
From: jeremy at alum.mit.edu (Jeremy Hylton)
Date: Wed, 14 Mar 2001 17:37:18 -0500 (EST)
Subject: Thoughts about extensions to the Python language
In-Reply-To: <5.0.2.1.0.20010314181331.0213eac0@mail.inet.com.br>
References: <5.0.2.1.0.20010314181331.0213eac0@mail.inet.com.br>
Message-ID: <15023.62110.155494.764345@w221.z064000254.bwi-md.dsl.cnc.net>

Guido did a poll at developers day last week.  The question was: "Is
Python development moving too fast or at just the right speed?"  (He
was being dictatorial and didn't even consider votes from the
more-faster crowd.)

I believe the results were 42 too fast and 50 just right.  I noted
that two of the more vocal proponents of the too fast camp proceeded
to suggest new features.  So at least some of the "too fast" people
probably meant "too fast unless it's a feature I want" :-).

Several people also suggested that Python 2.0 was a fine release, but
there were too many new things in Python 2.1.  Other than nested
scopes, which aren't slated for standard arrival until 2.2, there
aren't very many new features in 2.1 -- rich comparisons, a warnings
framework, support for weak references.  Rich comparisons, to pick an
example, have been on the table for ages.  The numeric people seem
happy, because NumPy has already added support for them.

You seem to have a very abstract concern about the speed of
develpoment and the fact that we write design documents that describe
changes to the language.  I voted for "just right" in the poll, so we
probably don't agree on the former.  As for the latter, what
alternative is there?  Make changes to the language without
describing the changes in advance? <wink>  I don't expect Guido's good
judgement is impaired by having written proposals to consider.  He's
been known to reject PEPs.

Jeremy



From enrico.spinielli at marconi.com  Thu Mar  8 03:46:37 2001
From: enrico.spinielli at marconi.com (Enrico Spinielli)
Date: 08 Mar 2001 09:46:37 +0100
Subject: LONG_BIT error when compiling python2.0
Message-ID: <m14rx4d4wi.fsf@gepc4223.it.marconicomms.com>

Hi all,
I am trying to install python2.0 on RedHat 7.0 Linux.
I got the sources and successfully run ./configure
but after a little while down in the compilation I got
the following error message:
[...]
gcc -g -O2 -Wall -Wstrict-prototypes -I./../Include -I.. -DHAVE_CONFIG_H   -c -o node.o node.c
In file included from ../Include/Python.h:54,
from node.c:3:
../Include/pyport.h:390:2: #error "LONG_BIT definition appears wrong for platform (bad gcc config?)."
make[1]: *** [node.o] Error 1
make[1]: Leaving directory `/mysrc/Python-2.0/Parser'
make: *** [Parser] Error 2

I am using gcc2.96-54

Any idea about a possible solution?
In the meanwhile I'll install gcc2.96-69 and retry...

Thanks a million
bye


From sheila at spamcop.net  Wed Mar 28 20:17:51 2001
From: sheila at spamcop.net (Sheila King)
Date: Thu, 29 Mar 2001 01:17:51 GMT
Subject: list vs tuple
References: <hGlw6.782$p5.3210@news1.rivrw1.nsw.optushome.com.au> <3AC1EB93.7BC87292@hack.org> <wEmw6.785$p5.3426@news1.rivrw1.nsw.optushome.com.au> <3AC1F93A.BAC9E34B@hack.org> <99teuu02p5f@news1.newsguy.com> <O4tw6.6798$m04.497167@e420r-atl1.usenetserver.com>
Message-ID: <k835ctknej8rhhprvg7cr91cd63iu291qr@4ax.com>

On Wed, 28 Mar 2001 16:50:58 -0500, "Steve Holden" <sholden at holdenweb.com>
wrote in comp.lang.python in article
<O4tw6.6798$m04.497167 at e420r-atl1.usenetserver.com>:

:"Alex Martelli" <aleaxit at yahoo.com> wrote in message
:news:99teuu02p5f at news1.newsguy.com...

:> Actually, tuples have NO methods -- including those that would
:> require no mutability at all:

...<snipped>...

:> Note the lack of 'count' and 'index', which do not mutate a list
:> when applied to it, but are still missing from tuples.
:>
:So, tuples are not only immutable, they are also uncooperative.

Tuples can be used as keys in dictionaries. Lists cannot.

--
Sheila King
http://www.thinkspot.net/sheila/
http://www.k12groups.org/



From sholden at holdenweb.com  Mon Mar  5 20:55:59 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Mon, 5 Mar 2001 20:55:59 -0500
Subject: Little problem with the "+" operator
References: <983802310.155838463@news.t-online.de> <mailman.983804468.16428.python-list@python.org> <980i6q$crs$1@news.mathworks.com>
Message-ID: <iwXo6.19376$ML1.967381@e420r-atl2.usenetserver.com>

"Joshua Marshall" <jmarshal at mathworks.com> wrote in message
news:980i6q$crs$1 at news.mathworks.com...

[ ... ]
> Comparing Python to Perl here doesn't seem very fair.  They're very
> different languages.  In general, I'm surprised by the urgency with
> which people often jump to compare them.

A very fair point, although of course the original enquiry was about an
exception that Perl (nor VBScript, for that matter), would have raised.
Methinks we Pythonistas should not protest too much.

regards
 Steve





From fredrik at pythonware.com  Mon Mar 12 17:25:09 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Mon, 12 Mar 2001 22:25:09 GMT
Subject: binary comparison of file object
References: <mailman.984429074.25277.python-list@python.org>
Message-ID: <91cr6.17919$Qb7.3117741@newsb.telia.net>

Jaeho Lee wrote:
> I am trying to compare the contents of binary file. Checking if the files
> are same or not is enough.

use the filecmp module:

    result = filecmp.cmp(file1, file2, shallow=0)

for more info, see:

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

Cheers /F

<!-- pythonworks 1.2 coming soon:
http://www.pythonware.com/products/works/index.htm
-->




From phlip_cpp at my-deja.com  Wed Mar 28 15:26:04 2001
From: phlip_cpp at my-deja.com (Phlip)
Date: Wed, 28 Mar 2001 12:26:04 -0800
Subject: Python and SWIG - passing object pointer param to an embedded script.
References: <3ac1a999@pron.i-way.net.uk>
Message-ID: <BQrw6.322$HL3.46475@news.pacbell.net>

Andrew Bonello wrote:

> PyDict_SetItemString( globals, "myobjptr", myobj);
> 
> PyObject* obj = PyRun_File( fp, (char*)fn, 0, globals, locals );
> 
> This just causes an access violation. (I am using Microsoft Visual C++
> V6.0 with Python V2.0, Debug build).

We ran with SWIG for a spell, and then I auditioned the Boost Python 
Library (www.boost.org) and never looked back.

For a way-cool example of reflection in native C++ without any extra 
parsers or mucking Boost rocks.

-- 
  Phlip                          phlip_cpp at my-deja.com
============== http://phlip.webjump.com ==============
  --  It's a small Web, after all...  --


From scarblac at pino.selwerd.nl  Thu Mar 22 11:01:16 2001
From: scarblac at pino.selwerd.nl (Remco Gerlich)
Date: 22 Mar 2001 16:01:16 GMT
Subject: learning python...
References: <3AB8F4E1.184DD9C@noaa.gov> <3ABA1148.ABBC50C0@sewardweb.com>
Message-ID: <slrn9bk895.fnm.scarblac@pino.selwerd.nl>

Lloyd Sommerer <lsommerer at sewardweb.com> wrote in comp.lang.python:
> One final bit of advice:  When you're wondering how to use a specific
> module, the best documentation is very often within the module itself.  I
> didn't actually come across that advice anywhere, and it would have helped
> me to have known it sooner.

And all of that info is now available on http://www.pydoc.org under "Module
Documentation by PyDoc". All the docstrings extracted from the code. Pretty
cool.

-- 
Remco Gerlich


From dg at pearl.tao.co.uk  Tue Mar 13 06:59:08 2001
From: dg at pearl.tao.co.uk (David Given)
Date: Tue, 13 Mar 2001 11:59:08 +0000
Subject: Atomic Event.{wait,clear}
Message-ID: <c22l89.2v1.ln@127.0.0.1>

I'm using the threading module, with the Event class to handle
synchronisation. It's all working nicely and is very easy to use.

However, I can't seem to find a way of atomically waiting for an Event,
and then clearing the flag, which is very strange. Currently I do
Event.wait() and then Event.clear(). If another thread does a Event.set()
in between the two, the event will be lost. For my current application,
this isn't a problem, but what shold I do in the future?

-- 
+- David Given ---------------McQ-+ 
|  Work: dg at tao-group.com         | Ugl?k u bagronk sha pushdug Internet-glob
|  Play: dgiven at iname.com         | b?bhosh skai.                              
+- http://wired.st-and.ac.uk/~dg -+ 


From cfelling at iae.nl  Fri Mar  2 19:12:50 2001
From: cfelling at iae.nl (Carel Fellinger)
Date: 3 Mar 2001 01:12:50 +0100
Subject: Thinking Pythonically (was Re: gah! I hate the new string syntax)
References: <mailman.983485923.1136.python-list@python.org> <B7In6.16281$Qb7.2658902@newsb.telia.net> <97ohh2$polt0$1@ID-77331.news.dfncis.de> <97opsb$nhu$1@nntp6.u.washington.edu> <97oupq$qijrh$1@ID-77331.news.dfncis.de>
Message-ID: <97pcu2$n6n$1@animus.fel.iae.nl>

Mark Pilgrim <f8dy at yahoo.com> wrote:
...
>>>> output = os.popen("cleartool lsvob -l").readlines()
...
>>>> [line for line in output if line[:4]=='Tag:']
...
>>>> vobs = [line.replace('Tag:', '').strip() for line in output if
> line[:4]=='Tag:']
...
>>>> "cleartool lock %s" % " ".join(vobs)
> 'cleartool lock /vobs/intranet /vobs/project'

> And that's what I call thinking Pythonically.  YMMV.

nai, that's unix thinking, like with

$ cleartool lock `cleartool lsvob -l | grep ^Tag: data | cut -c 6-`



-- 
groetjes, carel


From gerhard.nospam at bigfoot.de  Fri Mar 23 21:10:47 2001
From: gerhard.nospam at bigfoot.de (Gerhard =?iso-8859-1?Q?H=E4ring?=)
Date: Sat, 24 Mar 2001 03:10:47 +0100
Subject: Install Python on Apache
References: <99fph5$a2n$1@rex.ip-plus.net> <EDC008F28CADC736.9881C22ECC137D33.5745C372F32CC002@lp.airnews.net>
Message-ID: <3ABC0227.BC35664@bigfoot.de>

No need for weird and ugly hacks like the ones mentioned before 8-)

Just add this *one* line at the end of your httpd.conf:

ScriptInterpreterSource Registry

Assuming .py, .pyc, .pyw and .pyo are registered to be handled by
python{w}.exe, like any installer from ActiveState or python.org does,
you are ready to go. Just give your cgi-scripts an extension registered
to Python. My scripts look like this:

#!/usr/bin/env python
print 'Content-type: text/plain\n\nPython cgi works!'

They then run unchanged on win32 and Unix.

Gerhard

PS: The ScriptInterpreterSource is mentioned in the Apache docs. It just
makes Apache use the Windows assosiations of the various extensions to
invoke the right interpreter for cgi, such as .pl for P*rl, .py for
Python, .bat and .cmd for batch files, etc.
-- 
Sorry for the fake email, please use the real one below to reply.
contact: g e r h a r d @ b i g f o o t . d e
web:     http://highqualdev.com


From kknbk2 at bellsouth.net  Sat Mar  3 22:17:24 2001
From: kknbk2 at bellsouth.net (Kyle Wayne Kelly)
Date: Sat, 3 Mar 2001 21:17:24 -0600
Subject: Java guy interested in Python
Message-ID: <003801c0a459$a29c97f0$63da1542@wisdom>

Can anyone tell me some of the things that Python offers which are not available in Java?  I like Python's syntax, but what can Python do that Java can not do?  Any comments are helpful.

Kyle Wayne Kelly
Computer Science Student
University of New Orleans
http://www.cs.uno.edu/~kkelly
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20010303/426c929b/attachment.html>

From shaleh at valinux.com  Wed Mar  7 14:37:32 2001
From: shaleh at valinux.com (Sean 'Shaleh' Perry)
Date: Wed, 07 Mar 2001 11:37:32 -0800 (PST)
Subject: Can Python 1.5.2 code run on Python 2.x?
In-Reply-To: <985c9m$3gg$1@hobbes2.crc.com>
Message-ID: <XFMail.20010307113732.shaleh@valinux.com>

On 07-Mar-2001 Glen Mettler wrote:
> Can code written in Python 1.5.2 run without modification on Python 2.x?
> If not, what modification in code is required?
> 

The what's new docs show places where old code could fail.



From grante at visi.com  Tue Mar  6 11:47:57 2001
From: grante at visi.com (Grant Edwards)
Date: Tue, 06 Mar 2001 16:47:57 GMT
Subject: Can anyone recomend a good intoduction to C...
References: <3aa366c6_2@nnrp1.news.uk.psi.net> <8gNo6.440$y6.128357@ruti.visi.com> <3aa79a8a.466828531@nntp.interaccess.com>
Message-ID: <1x8p6.1046$y6.188683@ruti.visi.com>

In article <3aa79a8a.466828531 at nntp.interaccess.com>, Thaddeus L. Olczyk wrote:

>>>I'm going to have to bite the bullet and learn a bit of C. I'd at
>>>least like to learn the *right* way to do it.
>>
>>The "right" way to do it is with a PDP-11, a Lear-Siegler ADM3a
>>and the v7 source code.  ;)  Some would argue in favor of an
>>ASR-33, but they tend to annoy the downstairs neighbors.

>Recently the original  UNIX was released( isn't the v7 source a bit
>old? I would think you get something a bit earlier).

Yes, v7 is farily "modern".  It happens to be the first one I
used, so that's what came to mind.

-- 
Grant Edwards                   grante             Yow!  Where does it go when
                                  at               you flush?
                               visi.com            


From aleaxit at yahoo.com  Fri Mar  9 08:52:53 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 9 Mar 2001 14:52:53 +0100
Subject: gah! I hate the new string syntax
References: <mailman.983615170.23313.python-list@python.org> <tN5o6.2746$eH3.14913@NewsReader> <Koco6.7518$1D5.244602@e420r-atl1.usenetserver.com> <mailman.983997252.5628.python-list@python.org> <m2vgpkyfjo.fsf@mycroft.actrix.gen.nz> <mailman.984063199.18779.python-list@python.org>
Message-ID: <98an7i01tgp@news2.newsguy.com>

"D-Man" <dsh8290 at rit.edu> wrote in message
news:mailman.984063199.18779.python-list at python.org...
> On Thu, Mar 08, 2001 at 06:51:07PM +1300, Paul Foley wrote:
> | On Wed, 7 Mar 2001 15:33:30 -0500, D-Man  wrote:
> |
> | > What about one who is both English and Scots, yet never set foot on
> | > that island ;-)?
> |
> | AFAIK, nobody outside the US would admit the existence of such a
> | thing.  The American habit of calling themselves "Irish" or "Italian"
> | or whatever when they're not is *very* strange.  I've never heard
> | anything like it anywhere else.
>
> Heh.  Maybe Americans are very odd.  Wouldn't surprise me at all.  ;-)

Maybe.  OTOH, Italians born abroad (from parents who consider
themselves temporary immigrants, and fully intend to return "home"
when they've put enough savings aside) have traditionally called
themselves "Italians" whether that "abroad" was Northern America,
Argentina, Germany, Libia, or any of the several other countries
where Italian emigrants used to go.  (Of course, intentions to
return are not always carried out -- few are aware, for example,
that such an icon of French popular culture as Yves Montand was
Italian -- real name Ivo Livi, born of just-as-Italian parents
in a village near Florence; his father, a socialist, had to flee
[and ended up in France] when the Fascists took power, and Ivo,
as well as the rest of the family, followed him a while later;
I'm not sure when he changed his citizenship over to France).


> that I am not really American since none of my ancestors came from
> America.  "Real" Americans would be the so-called "Indians", or more

Why?  *THEIR* ancestors came from Asia over the Bering Straits,
after all (and following the line a while longer, one gets back
to Africa).



Alex





From tim.one at home.com  Sat Mar 24 17:26:10 2001
From: tim.one at home.com (Tim Peters)
Date: Sat, 24 Mar 2001 17:26:10 -0500
Subject: New PEP: The directive statement
In-Reply-To: <j466gzo3tb.fsf@informatik.hu-berlin.de>
Message-ID: <LNBBLJKPBEHFEDALKOLCGECAJIAA.tim.one@home.com>

[Martin von Loewis]
> That was rather meant as a comment to PEP 236: Instead of giving a
> mandatory release, it appears that all you can specify is a
> is_mandatory (isMandatory, or mandatory_p) flag; this is also all you
> said an application might need.

I don't think I followed that.  Since the future isn't certain until it
happens (and half the time not even then <0.9 wink>), I don't see that it's
possible "even in theory" to do better than give a best guess, and one that
gets updated if the once-predicted future arrives in a form other than hoped
for.

> ...
> it is also a fact that I do like to see the possibility to specify
> source encodings.

Me too.

> It's just that I cannot hope to get that now - but I had the hope that
> the future statement would be removed before the 2.1 release. Now that
> the final beta still has it, that hope is vanishing...

Note that Guido didn't reject your patch, he postponed it.  He even said he
*liked* the idea of directives!  The "transitional" form of the patch didn't
show up until Tuesday (three tiny days before the 2.1b2 scheduled release:
the worst possible time to try to get mindshare), and it didn't come with
(for example) doc changes, or patches to IDLE or python-mode.el to colorize
stuff the way people will expect.  We couldn't possibly make time to take up
that slack, and 2.1b2 is supposed to be the last chance for *anything* new to
get in.  The PEP was also too late, and now that you've written it it's clear
enough that it hasn't been met with universal delight either.

So I agree w/ Guido's patch comment in all respects:

    I'm postponing this now.

    I like the idea of directives, but I don't think we should
    disrupt the 2.1 release for this.

    So let's look at this again after 2.1.

You have to work pretty hard to get depressed by that <wink>.




From mwh21 at cam.ac.uk  Fri Mar  9 06:52:04 2001
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 09 Mar 2001 11:52:04 +0000
Subject: Static typing (was Re: Java guy interested in Python)
References: <mailman.984131592.17401.python-list@python.org>
Message-ID: <m3u2539n2z.fsf@atrus.jesus.cam.ac.uk>

=?iso-8859-1?q?Hamish=20Lawson?= <hamish_lawson at yahoo.co.uk> writes:

> Michael Hudson wrote:
> 
> > If you type-annotate this, you might write (using entirely made up
> > syntax):
> > 
> > def write16(file : <file>, int : <int>):
> >     file.write(struct.pack("h",int))
> > 
> > but then what about passing in a StringIO object, or some user 
> > wrapper of a file object?  What makes a "<file>"?  At the moment, 
> > it's just a loose collection of methods, some of which are optional 
> > in some circumstances.
> 
> Wouldn't there be a hierarchy of interface types for file-like
> objects, with some interfaces specifying more methods than others?
> You'd then choose the interface type that specified the collection
> of methods that you require of your passed object.

Well, that might be possible, but 
(a) there isn't one, presently
(b) you'd certainly have fun trying to make the heirachy a DAG.

I suspect that you'd basically end up listing the methods required,
eg:

def func(file : <has 'write', 'read', 'seek'>, ...):
    pass

I'm not sure that makes a "type system".  Hmm.  More thinking
required.

Cheers,
M.

-- 
  The use of COBOL cripples the mind; its teaching should, therefore,
  be regarded as a criminal offence.
           -- Edsger W. Dijkstra, SIGPLAN Notices, Volume 17, Number 5


From fredrik at pythonware.com  Tue Mar 13 14:57:00 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Tue, 13 Mar 2001 19:57:00 GMT
Subject: Extending Python: rewriting a single method in C
References: <g0wv9vxms0.fsf@scumbag.ecs.soton.ac.uk> <slrn9apn8f.382.jepler@potty.housenet> <g0ofv6xbj4.fsf@scumbag.ecs.soton.ac.uk> <98l1i901k4u@news1.newsguy.com> <g0lmq9xxz0.fsf@scumbag.ecs.soton.ac.uk>
Message-ID: <gYur6.119$4d3.23079@newsb.telia.net>

Jacek Generowicz wrote:
> (Would it be fair to say that the documentaion is, on the whole, less
> that satisfactory ?)

are you sure you read the C API docs before posting?

http://www.python.org/doc/current/api/objects.html
explains how PyObject's work.

http://www.python.org/doc/current/api/concrete.html
explains how to deal with objects of a known type.

http://www.python.org/doc/current/api/object.html
explains the abstract object interface, including the functions
you mention:

    PyObject* PyObject_GetAttrString (PyObject *o, char *attr_name)

    Return value: New reference.

    Retrieve an attribute named attr_name from object o. Returns
    the attribute value on success, or NULL on failure. This is the
    equivalent of the Python expression "o.attr_name".

etc.

Cheers /F




From grante at visi.com  Mon Mar 12 16:50:31 2001
From: grante at visi.com (Grant Edwards)
Date: Mon, 12 Mar 2001 21:50:31 GMT
Subject: This math scares me
References: <3aad4062.1290021@News.CIS.DFN.DE>
Message-ID: <Hwbr6.1139$Tg.127571@ruti.visi.com>

In article <3aad4062.1290021 at News.CIS.DFN.DE>, costas at springmail.com wrote:

>Ok, I can see maybe division having problems. But why does addition of
>the two numbers below:
>
>5.01+5.54
>
>give me this?
>
>10.550000000000001

Sorry. That's how floating point numbers on computers work.

Many numbers (including 10.01) can't be represented exactly
in the binary representation used by your computer.

Therefore floating-point results are never going to be exact.  

>I hope its some setting that can take care of this. I am using Python
>under Windows 98.

Use integer or fixed-point BCD math if you need exact results.

-- 
Grant Edwards                   grante             Yow!  MMM-MM!! So THIS is
                                  at               BIO-NEBULATION!
                               visi.com            


From fredrik at pythonware.com  Wed Mar 14 17:15:57 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 14 Mar 2001 22:15:57 GMT
Subject: gah! I hate the new string syntax
References: <mailman.983485923.1136.python-list@python.org> <97sudi01i26@news1.newsguy.com> <98elrk$li8$1@nntp.Stanford.EDU> <98fc070thh@news1.newsguy.com> <98iqo2$56o$1@nntp.Stanford.EDU> <98iv3s01sc2@news2.newsguy.com> <slrn9avqlm.8ee.quinn@retch.ugcs.caltech.edu>
Message-ID: <x4Sr6.331$4d3.75760@newsb.telia.net>

Quinn Dunkan wrote:
> Note that in ruby, any time you write
>
> class Foo
>     ...
> end
>
> you are adding to, not overriding, 'Foo'.  Thus,
>
> class String
>     def cap
>         s = (if self[0].betwwen?(97, 122) then
>                 self[0] - 32
>             else
>                 self[0]
>             end).chr
>         s + self[1 .. self.length]
>     end
> end
>
> is perfectly valid ruby that does what he wanted

does all classes live in the same namespace?

Cheers /F




From jziniti at speakeasy.org  Tue Mar  6 08:21:05 2001
From: jziniti at speakeasy.org (mapMonkey)
Date: Tue, 06 Mar 2001 18:21:05 +0500
Subject: Open Source Organic Chemistry software project
Message-ID: <taaseuid59k047@corp.supernews.com>

Pythonistas,

I am encouraging my current employer to GPL at least some of
our code and I would like to see what kind of interest I can
generate from the Python community ( everyone's favorite
language ;).  The amount of interest is very likely to swing the
vote ...

The current code-base is geared to Organic Chemistry molecules 
and the IUPAC nomenclature system and it works quite well.  It
was written entirely in Python, and entirely by me. My
goals are to re-engineer this code for added flexibility, and to 
increase the scope of the model to include all (or most) of organic
chemistry (reactions, submolecular thingies, etc.).

Anyone? Anyone?

Thanks for your time,

John Ziniti


From patrick at swdev.com  Fri Mar 23 02:58:45 2001
From: patrick at swdev.com (Patrick Curtain)
Date: Fri, 23 Mar 2001 07:58:45 GMT
Subject: learning python...
References: <3AB8F4E1.184DD9C@noaa.gov>
Message-ID: <patrick-2203012358450001@patrickspb>

Count me in.  I'd LOVE a better integrated environment for development, but
retain the vi keybindings.  :)   Yes, we're strange. 
--p

In article <3AB8F4E1.184DD9C at noaa.gov>, "Benjamin.Altman" <ben at .> wrote:

> Hello.  From the perspective of someone who already has programming
> experience, would anyone know if O'reillys "Learning Python" book would
> be better than say Prentice Hall's "Core Python Programming"?  Or would
> it be better to go straight into something like "Programming Python"?
> 
> Thanks,
> Ben

-- 
Patrick Curtain,  Husband & Father  ( i also write software )
http://www.swdev.com/


From aleaxit at yahoo.com  Tue Mar 20 01:39:03 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Tue, 20 Mar 2001 07:39:03 +0100
Subject: [Very Long (11K)] Numeric PEPs, first public posts
References: <mailman.984748941.31248.python-list@python.org> <slrn9bconk.ns2.qrczak@qrnik.zagroda> <1DBt6.120680$ML1.5626083@e420r-atl2.usenetserver.com>
Message-ID: <996u310128n@news2.newsguy.com>

"Steve Holden" <sholden at holdenweb.com> wrote in message
news:1DBt6.120680$ML1.5626083 at e420r-atl2.usenetserver.com...
    [snip]
> > >     Should the // operator be renamed to "div"?
> >
> > If % is % and not "mod", I would probably spell it //. But I would
> > call % "mod", except that it's too late.
> >
> No it isn't. It's never too late to _suggest_ a change :-) I seem to

I think div and mod might be built-in functions; sure, infix
operator syntax is nice, but, two new keywords...

[BTW, this discussion is supposed to move to the Python
Numerics list...].

> > I guess that sin(1.5) converts the argument to float?
> >
> Well, it shouldn't need to, unless the default numeric literal is a
> rational: 1.5 already _is_ floating point, isn't it?

It is now, but not in Moshe's PEP (it would be rational; 1.5e0
would be the float literal).


Alex





From db3l at fitlinxx.com  Fri Mar 30 16:57:35 2001
From: db3l at fitlinxx.com (David Bolen)
Date: 30 Mar 2001 16:57:35 -0500
Subject: Should I close after popen??
References: <m38zm6r8ih.fsf@master.athome> <fhOs6.510$Fb.777424@typhoon2.ba-dsg.net> <mailman.985919149.7252.python-list@python.org> <uofujnng0.fsf@ctwd0143.fitlinxx.com> <mailman.985983779.19709.python-list@python.org>
Message-ID: <u66gqc434.fsf@ctwd0143.fitlinxx.com>

pinard at iro.umontreal.ca (=?iso-8859-1?q?Fran=E7ois?= Pinard) writes:

> As it was pointed on this list quite a while ago, Python has no language
> definition per itself.  It is rather defined by the documentation of its
> main implementation.  It is true that the documentation of (C)Python does
> not commit it on the particular issue above, and this is why I checked
> with this list before starting to firmly rely on this feature, in real code.

Except that Python (the language) does have a Language Reference,
which does a reasonably good job of defining the language, sans
implementation, along with some implementation notes.  In particular,
section 3.1 talks about finalization and garbage collection as
implementation specific, and even "strongly" recommends using methods
such as close() on objects that involved external resources.

Clearly CPython has some deterministic behavior in this respect, which
has a low probability of changing at this point, and taking advantage
of that is just fine, as long as there's clarity on what the
dependence is, which is important in forum discussions such as this.

> In the meantime, I think we might continue to safely define Python as per
> (C)Python, rather than by the intersection of Jython and Python.  I do know
> that if I ever move from Python to Jython, I'll have to revise many things
> in my code, but until then, I prefer to be as clean and legible as possible.

I think "Python" should be defined as Python, and in discussions about
implementation issues, it's perfectly valid to base decisions on an
implementation such as CPython or JPython/Jython if one is willing to
have implementation-dependent code.  I guess perhaps my point was
somewhat pedantic, but in a discussion about implementation dependent
items, I think it's best to be clear on naming the implementations.

> The close *is* properly done.  This is guaranteed by the fact I do not
> keep any reference to the file.  On the first example, the close is done
> before proceeding with the next Python instruction.  On the second example,
> the close is done even before the first "process(line)" is executed.

In current CPython yes, but not guaranteed by Python the language (per
the reference which is very clear on that fact).

> About errors, shouldn't we just rely on the fact that they raise exceptions,
> automatically?  One of the beauties of Python, compared to C or Perl, is that
> errors are always trapped and handled by the system, when not by the user.

Except that if the finalization was postponed, that exception might
crop up in the strangest place (assuming it was propagated up at all).

But aren't exceptions that occur as part of finalization suppressed?
Assuming that the internal file type works similarly to Python code
objects, where exceptions are definitely suppressed in __del__
processing (see language reference section 3.3.1, under the Warning:
note for __del__).

> So, we have all guarantees that the data got written to the logfile, given
> the script does not fail with an I/O error.  Errors on implicit close will
> also be properly detected.  So, there is no need to feel uncomfortable! :-)

Just as an FYI, but that's not true in CPython.  The close that is
done during object deallocation ignores any error codes, unlike that
done in response to a close() method call.  This is true in at least
1.5.2 and 2.x which I had source around for.  So errors on implicit
closes will be silently ignored.

Note that I'm definitely not trying to force a certain comfort level
on yourself - you have the right to feel as comfortable or
uncomfortable depending on implementation-specific behavior as you
like :-)

But at the same time, I think it's worth pointing out the risks in
such dependence (note your assumption about exceptions), since it's up
to each developer to make that risk/ease tradeoff.

For me, while I may make the trade-off with read-only operations, I
don't think I'd do it if I were writing.  And even for reading, for
best use of resources in the most portable way, explicit close() is
better.

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


From dsh8290 at rit.edu  Thu Mar  8 18:34:31 2001
From: dsh8290 at rit.edu (D-Man)
Date: Thu, 8 Mar 2001 18:34:31 -0500
Subject: Can Python replace Visual Basic? Should it?
In-Reply-To: <iSBp6.14005$hn5.2035325@news1.rdc1.mb.home.com>; from bbollenbach@homenospam.com on Thu, Mar 08, 2001 at 02:10:22AM +0000
References: <5DZo6.11126$hn5.1578410@news1.rdc1.mb.home.com> <mailman.984010931.26257.python-list@python.org> <iSBp6.14005$hn5.2035325@news1.rdc1.mb.home.com>
Message-ID: <20010308183431.D3723@harmony.cs.rit.edu>

On Thu, Mar 08, 2001 at 02:10:22AM +0000, Brad Bollenbach wrote:
[snip]
| 
| Maybe the fact that only one person who saw my original post has
| actually migrated from VB to Python is an answer in itself.

Maybe there are other reasons entirely.  For example, I am not a VB
programmer, nor have I ever used VB.  Maybe there just aren't many
people on this list who have used VB, or more specifically used VB
before using Python.  And again, maybe not.  I come from an Eiffel,
C++, Java, C background (pretty much in that order) mostly on *nix
systems. 

-D




From gerhard.nospam at bigfoot.de  Tue Mar 13 14:14:20 2001
From: gerhard.nospam at bigfoot.de (Gerhard =?iso-8859-1?Q?H=E4ring?=)
Date: Tue, 13 Mar 2001 20:14:20 +0100
Subject: Subject with SMTPLIB?
References: <3AADE48A.5D8AA43@daimlerchrysler.com> <ORlr6.2821$54.2794@www.newsranger.com>
Message-ID: <3AAE718C.DA8B99B9@bigfoot.de>

Gregory Jorgensen wrote:
> 
> You need to add a subject header to your message. Email message format is
> simple: headers, blank line, message body.
> 
> import smtplib
> 
> msg = "subject: hello, world!\n\nmessage body"
> 
> s = smtplib.SMTP(host)
> s.sendmail(fromaddr, toaddrs, msg)
> 
> You'll find a good SMTP quick reference on Dan Bernstein's site (he's the author
> of qmail). See http://cr.yp.to/immhf.html.
> 
> In article <3AADE48A.5D8AA43 at daimlerchrysler.com>, root says...
> [...]

And, btw. you don't need to be root on a mission critical mailserver to
use it ;-)

Gerhard
-- 
Sorry for the fake email, please use the real one below to reply.
contact: g e r h a r d @ b i g f o o t . d e
web:     http://highqualdev.com


From cce at clarkevans.com  Tue Mar 20 22:57:12 2001
From: cce at clarkevans.com (Clark C. Evans)
Date: Tue, 20 Mar 2001 22:57:12 -0500 (EST)
Subject: Yet Another PEP:  Query Protocol Interface or __query__ 
In-Reply-To: <B43D149A9AB2D411971300B0D03D7E8B0FBFBA@natasha.auslabs.avaya.com>
Message-ID: <Pine.LNX.4.21.0103202251040.16274-100000@clarkevans.com>

On Wed, 21 Mar 2001, Delaney, Timothy wrote:
> So far as I can see, this PEP does not solve one basic 
> problem - interfaces which the class writer doesn't know about.

I think that this is a problem it does not set out to solve,
Alex Martelli's last post gave some insightful ways this
can be done.   Something like this can go into another PEP.
As this PEP doesn't prevent such a mechanism.

> I think it would be better if an interface was actually a class instance
> which specified exactly what it required. This could not only be methods (or
> other attributes), but also the (minimum) parameters required (and possibly
> their types) and even any pre- or post- conditions required by the
> interface. It could also mark "optional" features which if available
> will be used.

Right.  This is a complicated issue, and this PEP doesn't 
attempt to solve this.  It only attempts to introduce a 
simple mechanism to allow "declarations" in such a way
so that "discovery" can be added later.

> This instance then gets passed to the __query__() method. The __query__()
> method then inspects the requirements, and returns either a class instance
> or None. The class instance could either be self or a proxy to self or
> something completely separate. A None of course means that it cannot meet
> the requirements for the interface.

I like returning None!  You are I think the third person to 
suggest this, so I'll update the PEP with this suggestion.

> If no __query__() method is defined on the class, the query() function would
> then inspect the methods and other attributes of the class to determine if
> it meets the requirements of the interface. Naturally, interfaces with
> post-conditions would have to fail immediately in this case, and any
> pre-conditions which could not be readily determined would also have to fail
> the interface test immediately. A "readily-determined" pre-condition would
> be something like "attribute x == 3" (not a good example, but a simple one).
> 
> This seems to me a much more general and useful way of specifying
> interfaces, whilst covering all the things an interface may need. It allows
> the __query__() method to return different things on different invocations
> for the same interface, depending on the state of the class. For example, a
> file-like object may be unable to meet the "file-like" interface until it
> has all its data in memory. Until this is the case, it would return None
> from its __query__() method for an interface which required the read()
> method.

Right.  All this is good, and I'm specifically not trying to 
prevent this kind of stuff.  I just wanted a simple way to 
get at interfaces that are declared _without_ diving into
this type of stuff... 

> I may well just try implementing this some time ... most of it shouldn't be
> too hard. Unfortunately, I've got a nasty deadline coming up ... :(

Kewl.

Thank you so much for the feedback!

Clark




From stephen_purcell at yahoo.com  Mon Mar  5 03:03:35 2001
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Mon, 5 Mar 2001 09:03:35 +0100
Subject: Missing the functional mind set
In-Reply-To: <2j35at4v2u75l507v4djdqjv6if8rsq10u@4ax.com>; from urner@alumni.princeton.edu on Sun, Mar 04, 2001 at 10:53:13AM -0800
References: <mlb2at8ptrr7sh5dgq8qlk2201kquh6tfl@4ax.com> <mailman.983716566.6511.python-list@python.org> <2j35at4v2u75l507v4djdqjv6if8rsq10u@4ax.com>
Message-ID: <20010305090335.A4486@freedom.puma-ag.com>

Kirby Urner wrote:
> >The problem is that sorting a list can't be done in a functional way.
> 
> Maybe you didn't see the first responder's solution:
> 
>   >>> d = {3: 'd', 2: 'b', 1: 'a', 0: 'z'}
>   >>> templist = [d[n] for n in range(len(d))]
>   >>> templist
>   ['z', 'a', 'b', 'd']
> 
> That's a good 2.0 and up solution.

Except that he wanted the keys sorted, and didn't claim that they would
always be consecutive integers in range(len(d)).

(in which case I'd go for 'map(d.get, range(len(d)))' anyway)

-Steve

-- 
Steve Purcell, Pythangelist
Get testing at http://pyunit.sourceforge.net/
Get servlets at http://pyserv.sourceforge.net/
"Even snakes are afraid of snakes." -- Steven Wright



From sarnold at earthling.net  Mon Mar 12 01:08:56 2001
From: sarnold at earthling.net (Stephen L Arnold)
Date: Sun, 11 Mar 2001 22:08:56 -0800
Subject: font problem
Message-ID: <20010312060857.494931F661@shiva.arnolds.bogus>

Howdy:

I'm trying to get started with ActivePython, but the windoze html 
help thingie uses the wrong font for identifiers (it's trying to 
use the line drawing font).  I have the latest IE5 and the html 
help update (running on win95 OSR2 with all the updates).

I tried the old windoze un-install/re-install ting, with no luck.  
Anybody know the fix, or which font to install (if one is missing). 
 I've got a ton new ones after installing IE and Python...

Thanks in advance, Steve




From kevin_cazabon at hotmail.nospamplease!.com  Sun Mar 11 02:33:41 2001
From: kevin_cazabon at hotmail.nospamplease!.com (Kevin Cazabon)
Date: Sun, 11 Mar 2001 00:33:41 -0700
Subject: Who am I: can a class instance determine its own name?
References: <mailman.984034150.9635.python-list@python.org>
Message-ID: <DYV4H3fqAHA.187@typhoon.i3s.net>

You could "assign" it a name when you create the instance:

FooBar = Foo()
FooBar.name = "FooBar"

This would let you trace it later... not elegant though.  You could also
make it an initialization parameter.

FooBar = Foo(name="FooBar")

Just trying to help... not elegantly though.


Kevin




"Tim CHURCHES" <TCHUR at doh.health.nsw.gov.au> wrote in message
news:mailman.984034150.9635.python-list at python.org...
This is probably an elementary question and the answer is probably writ
large in multiple places in the Python documentation, but...

...can an instance of a class determine the name of the variable to which it
is assigned? For example:

###########################
class Foo:
    def whoami(self):
         return "You are a Foo() but I do not know your name"

FooBar = Foo()

print FooBar.whoami()
###########################

How does one define the method whoami() so that it returns "FooBar"? This
sort of navel gazing is formally called introspection, I think (therefore I
am)?

Tim Churches
Sydney, Australia
(where, due to the Coriolis effect, the Python prompt does indeed look like
this: <<< - or maybe its because we are upside-down)









From charlie at gabrielinteractive.com  Thu Mar 15 14:37:06 2001
From: charlie at gabrielinteractive.com (Charlie Barrows)
Date: Thu, 15 Mar 2001 14:37:06 -0500
Subject: PyImport_ImportModuleEx() ?
Message-ID: <98r5lv$1mbi$1@allnight.news.cais.net>

I'm just starting to tinker with embedding Python in a C app and I'm having
trouble getting PyImport_ImportModuleEx() to import [*] of a module into
__main__.

More specifically: I can get it to import the module fine as a separate
namespace, but, just for the sake of thinking I know what I'm doing, I
wanted to import all the module's functions into __main__'s namespace.

I guess you could also say I'm looking for the PyImport_ version of this
line:

PyRun_SimpleString("from TestScript import *\n");

I was thinking some combination of PyImport_ImportModuleEx would do the
trick but haven't been able to figure it out.

Am I missing one or more points here?

Thanks.

Charlie Barrows
Gabriel Interactive




From claird at starbase.neosoft.com  Wed Mar 14 17:06:41 2001
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 14 Mar 2001 16:06:41 -0600
Subject: BCD applications (was: This math scares me)
References: <3aad4062.1290021@News.CIS.DFN.DE> <7bQr6.644$lz1.17686@ruti.visi.com> <98oku1$f6a$1@news.udel.edu> <erRr6.659$lz1.20662@ruti.visi.com>
Message-ID: <96182B7D15FEECF4.22799D8BFE69044A.17D2E54CAD32E9D6@lp.airnews.net>

In article <erRr6.659$lz1.20662 at ruti.visi.com>,
Grant Edwards <grante at visi.com> wrote:
			.
			.
			.
>I believe that there are plans (on some exchanges) to switch
>pricing from base-2 fractions to decimal fractions.
			.
			.
			.
Various exchanges have made that switch at various
times during the last few centuries.  Most famous
was 29 January of this year, when the NYSE made the
change for most common equities.  Most of NASDAQ,
though, remains in eighths, I believe.
-- 

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


From bezilla at hotmail.com  Tue Mar 27 00:45:22 2001
From: bezilla at hotmail.com (John E. Barham)
Date: Tue, 27 Mar 2001 05:45:22 GMT
Subject: HTTP redirect via Python?
Message-ID: <SNVv6.128157$tP3.2040832@news1.rdc1.bc.home.com>

Apologies if this is too basic a question, but I'm wondering how I can send
a redirect/URL rewrite command to a browser via Python.

The context:  we're developing a fairly substantial Web-based application
and using Python for the CGI coding (which is working beautifully, BTW).  In
order to partially protect ourselves (and the users ;) against
back/forward/refresh with forms, we'd like to send them to a "safe" URL
after a form has been submitted.  How can we do this?  There doesn't appear
to be anything to do this in module cgi...

If it's a server dependent thing, we're using Apache.




From warren-postma at home.com  Tue Mar 27 17:31:57 2001
From: warren-postma at home.com (Warren Postma)
Date: Tue, 27 Mar 2001 22:31:57 GMT
Subject: WANTED: chairs for next Python conference
References: <mailman.985642835.3364.python-list@python.org>
Message-ID: <xx8w6.92515$A6.21119288@news4.rdc1.on.home.com>

"Guido van Rossum" wrote in message:
> I'm looking for chairs for the next Python conference.

OfficeMax has some nice ones, with good lumbar support.

[sorry!]

Warren





From aleaxit at yahoo.com  Fri Mar  9 10:27:54 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 9 Mar 2001 16:27:54 +0100
Subject: I come to praise .join, not to bury it...
References: <mailman.984074812.27842.python-list@python.org>
Message-ID: <98aspo024jg@news2.newsguy.com>

"Andrew MacIntyre" <andymac at bullseye.apana.org.au> wrote in message
news:mailman.984074812.27842.python-list at python.org...
> On Wed, 7 Mar 2001, Alex Martelli wrote:
>
> > Suppose I'm building a line of a CSV file, a popular
> > textual format where fields are separated (aka joined)
> > with commas -- the acronym stands for 'comma-separated
> > values'.  So, what I want to do is
> >     'comma-separate these values'
> > and, since 'separate' equates to join, this equates to:
> >     'comma-join these values'
> > so, I code:
> >     comma.join(these_values)
> > *WHAT'S BACKWARDS ABOUT IT*?!
>
> Well thats fine from your point of view, but to me you've just argued
> against your point of basing programming around natural language
> (although I may have misconstrued your argument).

I'm definitely not arguing that design should be based on
natural language; rather, I'm taking exception to people
who claim some programming-language expression is 'backwards'
*when it even happens to be basically homomorphic to some
expressions in THEIR OWN native natural language*!


> From my POV, a method is something that logically "sends a message" to its
> object.

...requesting the object to perform some action (execute
some method).  Fine with me.

> In this context, having the joiner be the object is backwards,

No it isn't.  We send to the joiner the request to execute
its method, thus performing the joining-action.

> because it makes more sense to me to think in terms of:
> "send message <join members with separator parameter and return string> to
> a sequence"

But that's exactly what I'm arguing against!  That's *NOT*
optimal.  We do NOT want a sequence to have code for the
purpose of 'joining its items' -- it's not a special task
that every sequence should somehow perform or delegate;
rather, the sequence's task is just to *ENUMERATE* its
items, for whatever purpose that enumeration happens to
be needed this time.  Further, we want the enumeration to
happen *identically*, whether it's for the purpose of
joining items one after the other inside some joiner
object's join-method, or for the purpose of joining items
one after the other inside a file-object's writelines
method, or for any of a zillion other purposes yet.

Consider .writelines again, for example.  I have never
heard objections to it being a method of the file object,
taking the sequence as its argument and accessing it via
the sequence's standard interface (i.e. just asking the
sequence to enumerate its items one after the other).

And, indeed, it's an excellent architecture -- some file
like objects may implement .writelines as just a loop
over their own .write method, others may take advantage
of performance shortcuts (e.g., if they hold their data
internally as a list of pieces, they'll be able to directly
implement .writelines by delegating to .extend).

But the .join/.writelines parallel is VERY close, so why
would the SAME architectural choice (to have the sequence
object as an argument, accessing it through its ability
to enumerate its items, only) cause problems in one case
and not the other?!

I think it's a misperception due to thinking of strings
as somehow 'passive', 'data', while a file-object is
thought of as 'active', 'a truly object-y object', or
something.  I further opine that this misperception is
susceptible of being corrected _in individuals who are
more interested in furthering the effectiveness of their
perceptive mechanisms than in trying to win a point in
a Usenet discussion_ -- which, I hope, describes a vast
majority of Pythonistas:-).


Alex





From tim.hochberg at ieee.org  Wed Mar 28 10:51:53 2001
From: tim.hochberg at ieee.org (Tim Hochberg)
Date: Wed, 28 Mar 2001 15:51:53 GMT
Subject: Speed problems with Python vs. Perl
References: <99sv33$8hk$1@news.cineca.it>
Message-ID: <tMnw6.77278$Xt3.10616558@news1.rdc1.az.home.com>


> Last week I wrote a simple Python program and found out that it was
> terribly slow. Therefore I retried with Perl and found a much better
> performance. The programs simple read a file and split the lines on white
> space, something I have to do very often for data elaboration

[SNIP]

> As you can see, without the splitting (only reading and counting the
lines)
> Perl is twice as fast, but once I split on whitespace, Python gets more
> than 5 times slower than Perl.
>
> These results have been achieved on a AMD with 650 MHerz. On my home
> machine, a 266 Celeron, the performance gap is worse. Without line
> splitting the programs
> are about the same speed, but with splitting enabled, the Python version
> becomes 10 times slower than the Perl version (5 secs against 0.5 secs)

Why the Perl's REs are faster, I'll leave for someone else to explain.
Perhaps they special case common REs more than Python does. However, you are
suffering from Overuse of Regular Expressions (ORE [tm]). Specifically, you
can use string.split or in Python 2.0 you can use the split method on
strings:

def main152(): # Should work on python 1.5.2
    import string
    icount = 0
    for line in sys.stdin.readlines():
        icount = icount + 1
        f = string.split(line)
    print "Total lines read: " + `icount`

or

def main20(): # Should work on python >= 2.0
    icount = 0
    for line in sys.stdin.readlines():
        icount += 1
        f = line.split()
    print "Total lines read: " + `icount`

or

def main21(): # Should? work on Python >= 2.1
    icount = 0
    for line in sys.stdin.xreadlines(): # Might speed things up a bit....
        icount += 1
        f = line.split()
    print "Total lines read: " + `icount`

I haven't actually benchmarked (or even tried) any of these, but I'd be
curious to see how the last one in particular stacks up against it's Perl
equivalent. I don't have Perl installed at the moment, though so I leave
that as an excercise for the reader....

-lazy tim









From gregj at pobox.com  Tue Mar 20 19:07:09 2001
From: gregj at pobox.com (Greg Jorgensen)
Date: Wed, 21 Mar 2001 00:07:09 GMT
Subject: deleting files from a script
References: <RSRt6.2154$nd.2670928@typhoon.ne.mediaone.net>
Message-ID: <NgSt6.596790$U46.18042286@news1.sttls1.wa.home.com>

You can get the name of an open file from the file object's name attribute:

>>> f = open('foo')
>>> f.name
'foo'

If you always open your files with full absolute paths (rather than relative
to the current working directory) you can then retrieve that path from the
f.name attribute. With a full path, deleting the file is simple:

>>> f.close()
>>> os.remove(f.name)

The os.path module has functions for converting relative paths to absolute
paths.

"Gary" <glevenson at mediaone.net> wrote in message
news:RSRt6.2154$nd.2670928 at typhoon.ne.mediaone.net...
> Is the best way to do this to just issue a shell command using getstatus
or
> something similar? I have a file descriptor created with open. I'd rather
> not deal with constructing the entire path string if I can avoid it.

--
Greg Jorgensen / programmer, pedant, raconteur / Portland, Oregon, USA




From thomas at xs4all.net  Thu Mar  1 17:21:07 2001
From: thomas at xs4all.net (Thomas Wouters)
Date: Thu, 1 Mar 2001 23:21:07 +0100
Subject: [Python-Dev] Re: d = {}; d[0:1] = 1; d[0:1] = 2; print d[:]
In-Reply-To: <200103012101.QAA18516@cj20424-a.reston1.va.home.com>; from guido@digicool.com on Thu, Mar 01, 2001 at 04:01:52PM -0500
References: <15006.46879.607192.367739@w221.z064000254.bwi-md.dsl.cnc.net> <200103012101.QAA18516@cj20424-a.reston1.va.home.com>
Message-ID: <20010301232107.O9678@xs4all.nl>

On Thu, Mar 01, 2001 at 04:01:52PM -0500, Guido van Rossum wrote:
> > Quoth Robin Thomas <robin.thomas at starmedia.net>:

[ Dicts accept slice objects as keys in assignment, but not in retrieval ]

> > | 1) Is this behavior considered a bug by the BDFL or the community at large?

> I can't speak for the community, but it smells like a bug to me.

Speaking for the person who implemented the slice-fallback to sliceobjects:
yes, it's a bug, because it's an unintended consequence of the change :) The
intention was to eradicate the silly discrepancy between indexing, normal
slices and extended slices: normal indexing works through __getitem__,
sq_item and mp_subscript. Normal (two argument) slices work through
__getslice__ and sq_slice. Extended slices work through __getitem__, sq_item
and mp_subscript again.

Note, however, that though *this* particular bug is new in Python 2.0, it
wasn't actually absent in 1.5.2 either!

Python 1.5.2 (#0, Feb 20 2001, 23:57:58)  [GCC 2.95.3 20010125 (prerelease)]
on linux2
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> d = {}
>>> d[0:1] = "spam"
Traceback (innermost last):
  File "<stdin>", line 1, in ?
TypeError: object doesn't support slice assignment
>>> d[0:1:1] = "spam"
>>> d[0:1:] = "spam"
>>> d
{slice(0, 1, None): 'spam', slice(0, 1, 1): 'spam'}

The bug is just extended to cover normal slices as well, because the absense
of sq_slice now causes Python to fall back to normal item setting/retrieval.

I think making slices hashable objects makes the most sense. They can just
be treated as a three-tuple of the values in the slice, or some such.
Falling back to just sq_item/__getitem__ and not mp_subscript might make
some sense, but it seems a bit of an artificial split, since classes that
pretend to be mappings would be treated differently than types that pretend
to be mappings.

-- 
Thomas Wouters <thomas at xs4all.net>

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



From clee at gnwy100.wuh.wustl.edu  Wed Mar  7 08:00:18 2001
From: clee at gnwy100.wuh.wustl.edu (Christopher Lee)
Date: 07 Mar 2001 07:00:18 -0600
Subject: How to installing NumPy?
References: <3AA57C0D.BBC5E956@itu.edu.tr> <7u3dcqqq2z.fsf@gnwy100.wuh.wustl.edu> <3AA60D2D.4E31AA50@imkt.uni-hannover.de>
Message-ID: <7usnkpentp.fsf@gnwy100.wuh.wustl.edu>

>>>>> "Harald" == Harald Rosemann <rosemann at imkt.uni-hannover.de> writes:

    Harald> Christopher Lee wrote:
    >>  >>>>> "Nuzhet" == Nuzhet Dalfes <dalfes at itu.edu.tr> writes:

    Harald> Got the error message: ....  gcc -O2 -m486 -fpic -IInclude
    Harald> -I/usr/include/python2.0 -c Src/_numpymodule.c -o
             ^^^^^^^^^^^^^^^^^^^^^^^
    Harald> build/temp.linux-i686-2.0/Src/_numpymodu Src/_numpymodule.c:1:
    Harald> Python.h: No such file or directory error: command 'gcc' failed
    Harald> with exit status 1 ....


    Harald> In Python-2.0 I found: Python-2.0/Include/Python.h but that
    Harald> doesn't seem to work.

It sounds like Suse may not install the headers where python expects, or if
you installed from source, perhaps you didn't do a "make install" ?  I
can't tell based upon your description but either way...


I would:

   - check to see if the directory /usr/include/python2.0 exists
     and create it if it doesn't exist yet

   - copy all the files that you found in Python-2.0/Include to
     /usr/include/python2.0.  This should allow the compiler to find
     Python.h using the -I/usr/include/python2.0 flag.

   - try "python setup.py install" again


good luck


From bowman at montana.com  Mon Mar 26 23:44:30 2001
From: bowman at montana.com (bowman)
Date: Mon, 26 Mar 2001 21:44:30 -0700
Subject: warrior geeks
References: <mailman.985653937.22321.python-list@python.org>
Message-ID: <QfUv6.185$8%6.1115@newsfeed.slurp.net>

Delaney, Timothy <tdelaney at avaya.com> wrote in message
news:mailman.985653937.22321.python-list at python.org...
> That's easy. With Python you can be sure that you have shot yourself in
the
> foot. With Perl, you think you've *probably* shot yourself in the foot,
but
> due to a difficult-to-find typo you have also shot several other people in
> the foot.

Or, as Stroustrup says about C++, it is much more difficult to shoot
yourself in the foot than when using C -- but when you do, you blow your
whole leg off.





From jarober at mail.com  Sat Mar 24 20:24:49 2001
From: jarober at mail.com (James A. Robertson)
Date: Sun, 25 Mar 2001 01:24:49 GMT
Subject: Collection interfaces
References: <3a90fa97.0@news1.mweb.co.za>   <3A9A86B1.9A8C69A2@ix.netcom.com> <MPG.1504d14563dce6499896af@news.earthlink.net> <97fjla128ir@news2.newsguy.com> <MPG.1505a76831c34b749896b3@news.earthlink.net> <97h5s40vhc@news2.newsguy.com> <MPG.1506007591cf818b9896bc@news.earthlink.net> <slrn99opop.gr.amk@mira.erols.com> <MPG.1506265ebf6d934d9896be@news.earthlink.net> <97ibmc0k47@news1.newsguy.com> <MPG.1506cf3293b094b49896c1@news.earthlink.net> <97m0c7080c@news2.newsguy.com> <MPG.15085517debf4b649896ca@news.earthlink.net> <zb%n6.31033$o%6.586026@news1.rdc1.bc.home.com> <MPG.150b383cea226d409896d0@news.earthlink.net> <kNko6.6986$eq3.707121@news.uswest.net> <MPG.150c46db9cec283e9896d4@news.earthlink.net> <3AA2A898.450B659E@mail.com> <gRCt6.12224$Im6.1308560@newsread1.prod.itd.earthlink.net> <3AB755F8.67E84126@mail.com> <%a%t6.17245$227.1601569@newsread2.prod.itd.earthlink.net> <3AB8A730.4BE13D22@mail.com> <61bv6.8693$P4.676800@newsread1.prod.itd.earthlink.net>
Message-ID: <3ABE9A43.90576C8A@mail.com>

topmind wrote:
> 
> "James A. Robertson" <jarober at mail.com> wrote in message
> news:3AB8A730.4BE13D22 at mail.com...
> > topmind wrote:
> > >
> >
> > > > http://www.cincom.com/smalltalk
> > > >
> > >
> > > I only see a front page, not a parsing example.
> > >
> >
> > Follow the tutorial link - it's right there on that page.
> >
> 
> I cannot find any "tutorial" link.

Well, it was there, but the site was redesigned.  Follow this link:

http://www.cincom.com:80/scripts/smalltalk.exe/gettingstarted/index.asp?content=trysmalltalknc


> 
> Perhaps you are afraid that if you give me the full tutorial link, you
> will have to expose to the world the FACT that the site runs on
> Microsoft ASP VSscript instead of Smalltalk?
> 
> Do Chevy salesmen drive Fords?
> 
> -tmind-

-- 
James A. Robertson
Product Manager (Smalltalk), Cincom
jarober at mail.com
<Talk Small and Carry a Big Class Library>


From loewis at informatik.hu-berlin.de  Fri Mar  2 11:01:01 2001
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 02 Mar 2001 17:01:01 +0100
Subject: A counter-counter-proposal for PEP 236: #pragma( nested_scopes)
References: <mailman.983480343.14274.python-list@python.org>
Message-ID: <j4g0gwjh36.fsf@informatik.hu-berlin.de>

"Tim Peters" <tim.one at home.com> writes:

> Hmm.  Maybe that's why I so dislike Martin's attempt to hijack PEP 236:  his
> current formulation of an alternative leaves only the raw names, without the
> __future__ or __possible_world__ qualifiers.  

If *that* is the complaint, then the patch can be easily changed into reading

directive future nested_scopes

(future would be the directive name, and nested_scopes the atom
argument). In that case, I'd prefer /F's interpretation of the state
of things, i.e.

directive experimental nested_scopes

since this does not talk about the future, but about the present (just
as the future import does not talk about the future, but about
directives that are presently needed).

Regards,
Martin


From sholden at holdenweb.com  Thu Mar  1 09:41:14 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 1 Mar 2001 09:41:14 -0500
Subject: To the point of defeat.....IDLE
References: <3A9AE62B.5070402@yahoo.com> <slrn99s6tu.8rl.garrett@falcon.cs.colostate.edu>
Message-ID: <%ftn6.12742$zL.491362@e420r-atl1.usenetserver.com>

"Deon Garrett" <garrett at cs.colostate.edu> wrote in message
news:slrn99s6tu.8rl.garrett at falcon.cs.colostate.edu...
> The best way to go is to completely wipe any traces
> of python rpms from your system before installing
> Python 2.0 from source.
>
> rpm -e python
>
> will fail because a bunch of stuff needs python.  But
> it will list all the packages that need it.  You should
> be able to look through the dependencies and find all
> the stuff you want to remove.
>
> rpm -e --nodeps python python-tools tkinter.......
>
> I assume that the python executable is installed in
> /usr/bin on RedHat 7.  If so, then run configure in
> the python 2.0 root directory with the --prefix=/usr
> option and all your scripts should still work.
>
> cd your/python2.0/root/dir
> ./configure --prefix=/usr (and any other options you want)
> make
> make install
>
> Then build any add-ons you want (Tkinter, NumPy, PIL, etc).
>
Has anyone actually verified that the myriad pieces of RedHat Python aren't
incompatible with 2.x? RedHat makes heavy use of Python for various system
configuration tasks, and I've always been cautious about replacing the
RPM-installed Python (so I installed newer versions
with --prefix=/usr/local).

regards
 Steve





From dwmurphy at home.com  Mon Mar 12 22:43:12 2001
From: dwmurphy at home.com (David Murphy)
Date: Tue, 13 Mar 2001 03:43:12 GMT
Subject: Einstein's Riddle
References: <oNPq6.396644$ge4.138479566@news2.rdc2.tx.home.com> <UdSq6.2184$54.2166@www.newsranger.com> <984380185.768920@news.aei.ca> <uP%q6.2375$54.2207@www.newsranger.com> <4w6r6.193645$df5.5192565@news1.crdva1.bc.home.com> <r62qats4341i3hcn7h6cum5mlfrtso60gi@4ax.com>
Message-ID: <kHgr6.194719$df5.5198492@news1.crdva1.bc.home.com>

"{R} Richard Ashton" <{R}@news.semolina.org> wrote in message
news:r62qats4341i3hcn7h6cum5mlfrtso60gi at 4ax.com...
> On Mon, 12 Mar 2001 16:08:32 GMT, "David Murphy" <dwmurphy at home.com>
> wrote:
>
> }*IF* Einstein was just talking about people's ability to deal with just
the
> }facts, and not to make assumptions, I think 2% +/- 5% is probably a good
> }estimate.   Take a look at the first draft of the next project
specification
> }you get *grin*
>
> He was. There are many variants of these, and they have been around along
> time. A lot of psychological papers have been written about "mindsets" and
> jumping to conclusions, reading more into the information than the data
> warrants. I could go on ...
>
> {R}

Interesting.

    Although I haven't read any of those papers,  i have to wonder what
types of people are inclined to make such assumptions.  i mean, for varies
activities, it might be essential.

    For example,  anything creative may require the ability to lead 'beyond
the facts' to come up with something new. (so long as the individual is able
to work back and fill in the facts)

dave




From tjreedy at udel.edu  Sat Mar 17 01:56:54 2001
From: tjreedy at udel.edu (Terry Reedy)
Date: Sat, 17 Mar 2001 01:56:54 -0500
Subject: Python style questions
References: <98tb91$9t9$1@saltmine.radix.net> <98tdu1$cfk$1@panix2.panix.com> <1eqdo5u.1xxmzfi2kjk0qN%francois.granger@free.fr>
Message-ID: <98v1vi$gre$1@news.udel.edu>

"Fran?ois Granger" <francois.granger at free.fr> wrote in message
news:1eqdo5u.1xxmzfi2kjk0qN%francois.granger at free.fr...
> Aahz Maruch <aahz at panix.com> wrote:
>
> > In article <98tb91$9t9$1 at saltmine.radix.net>,
> > Cary O'Brien <cobrien at Radix.Net> wrote:
>
> > >3. I really miss not having a "switch" or "case" statement.  Sniff.
>
> > Use a dictionary.
>
> Some code snippet as exemple ?
> At this point, I can't imagine a good way, and I don't remember seeing
> something on this.

I posted an example about a month ago ... and about a year ago.
Don't know what archive search facilities we currently have.

Terry J. Reedy





From tuttledon at hotmail.com  Thu Mar 15 18:04:34 2001
From: tuttledon at hotmail.com (Don Tuttle)
Date: Thu, 15 Mar 2001 23:04:34 GMT
Subject: Python and ADO
References: <98r9qn$tbi$1@troll.powertech.no>
Message-ID: <6Ubs6.118581$Z8.23975952@typhoon.southeast.rr.com>

"Syver Enstad" <syver at NOSPAMcyberwatcher.com> wrote in message
news:98r9qn$tbi$1 at troll.powertech.no...
> Has anyone considered making a DB API compliant wrapper around MS ADO? Or
> isn't this considered worthwhile?
>
> As ADO is the database api that I use, I could have a look at making a
> wrapper for it if there was any interest in it and if making such a thing
is
> not too involved.

Sounds like a good idea to me too!  I'm not currently doing an database work
but I will be.

Don




From grante at visi.com  Wed Mar  7 15:05:30 2001
From: grante at visi.com (Grant Edwards)
Date: Wed, 07 Mar 2001 20:05:30 GMT
Subject: Can anyone recomend a good intoduction to C...
References: <mailman.983894288.19965.python-list@python.org> <3AA5EF59.507F1D91@olsen.ch> <983958878.976344@newsmaster-04.atnet.at> <77up6.1289$y6.235541@ruti.visi.com> <985vqs01ipr@drn.newsguy.com>
Message-ID: <ewwp6.1466$y6.241005@ruti.visi.com>

In article <985vqs01ipr at drn.newsguy.com>, Grant Griffin wrote:

>>>To be completely honest, I can not see the obvius reason why
>>>C++ should be such a bad programming language.
>>
>>It's complicated and unsafe.  Either one is tolerable without
>>the other.  When you combine the two, you're dead.
>
>I dunno...I think you've hit on exactly its _strength_: it's complicated and
>unsafe.
>
>As a virtually 100.0% backwards-compatible superset of C, it
>has little choice but to be complicated and unsafe.  By
>"unsafe", maybe you are referring to its
>backwards-compatibility with C's pointers: C++ is unsafe
>because C is unsafe.

Exactly.

>By "complicated", maybe you are refererring to the fact that it
>supports both procedural programming (from C) and OO
>programming (which is its whole reason to exist).

Well, I'm referring to templates, overloading, and all of the
other features that confuse people.

>But those of us who like C++ don't regret its complication or
>its unsafety; we lthink of that more as being "powerful" and
>"flexible".

I think that being too flexible is a bad thing when it comes to
large projects with many developers.

>I would agree that C++ is somewhat _unnecessarily_ compilicated
>(primarily in terms of its silly protection features), but then
>again, nothing's perfect. (Heck, even Python has "warts".
><wink>)
>
>Still, in terms of its design goal (of grafting OO onto C while
>being 100.0% backwards-compatible),I personally think C++ is a
>real Work of Art.

Right!  As in most software project failures, the problem lay
in the requirements.  If you ask for something with both the
unsafe features required for backwards compatibility with C
_plus_ all the bells and whistles they thought they needed for
OO programming, there's no way you can end up with something
that isn't an unholy mess.  I think asking for a language that
did both low-level stuff like C and OO stuff was a mistake.

>(In contrast, I find Perl's approach to grafting on OO to be
>distinctly hokey.)

I never liked Perl, and doubt that duct-taping stuff to the
side of it would make me like it any better.

>I think that C++'s choice of design goal, in combination with
>its success in meeting that goal, adequately explains why large
>numbers of lemmings use it.
>
>>>It is widely available (on almost any platform, I would guess).
>>
>>If 1 billion people do a stupid thing, it's still a stupid thing.
>
>(And you gotta admire the chutzpah of that one guy who thinks
>that the 1 billion people who disagree with him must, ipso
>facto, be wrong. <wink>)

Hey, if I thought I was wrong, I'd change my mind.  :)

I think most of the projects using C++ would be far better off
using Modula-3, Ada, Smalltalk, or Python.  The rest should
probably be using C.  

Despite Ada's bad rap as a DoD project, it's a pretty decent
language (esp for embedded systems), and I think Modula-3 got
OO-in-a-procedureal-language pretty close to right right.

>isn't-it-nice-to-know-that-we-grants-don't-always
>   -agree?-<wink>-ly y'rs

Even when we're both from Iowa.  

-- 
Grant Edwards                   grante             Yow!  Look DEEP into the
                                  at               OPENINGS!! Do you see any
                               visi.com            ELVES or EDSELS... or a
                                                   HIGHBALL??...


From root at rainerdeyke.com  Wed Mar 21 11:42:34 2001
From: root at rainerdeyke.com (Rainer Deyke)
Date: Wed, 21 Mar 2001 16:42:34 GMT
Subject: Yet Another PEP:  Interface Adapter Mechanism __adapt__
References: <Pine.LNX.4.21.0103201837570.21343-100000@clarkevans.com> <Pine.LNX.4.21.0103210418490.21377-100000@clarkevans.com>
Message-ID: <_R4u6.466172$ge4.162590641@news2.rdc2.tx.home.com>

"Clark C. Evans" <cce at clarkevans.com> wrote in message
news:Pine.LNX.4.21.0103210418490.21377-100000 at clarkevans.com...
>     def adapt(obj,ident, options =  None):
>         global check_base
>         def check_base(bas,cmp):
>             if bas == cmp: return 1
>             for base in bas.__bases__:
>                 if check_base(base,cmp): return 1
>             return 0
>         if hasattr(obj, '__adapt__'):
>             retval = obj.__adapt__(ident)
>         if retval == None:
>             if check_base(obj.__class__,ident):
>                 retval = obj
>         # options flag used to enable:
>         #   - reverse lookup via ident (future PEP) goes here?
>         #   - signature based lookup (future PEP) goes here?
>         #   - automatic signature checking (future PEP) goes here?
>         return retval

This 'check_base' stuff is a mess.  "A is an instance of a superclass of B"
does not imply "B implements interface A".  I also don't see why 'ident'
should be a class instance instead of just a class or a new built-in type.
What functionality does 'ident' provide?  How are extension types supported?


--
Rainer Deyke (root at rainerdeyke.com)
Shareware computer games           -           http://rainerdeyke.com
"In ihren Reihen zu stehen heisst unter Feinden zu kaempfen" - Abigor




From jmg at ecs.soton.ac.uk  Mon Mar 12 06:13:51 2001
From: jmg at ecs.soton.ac.uk (Jacek Generowicz)
Date: 12 Mar 2001 11:13:51 +0000
Subject: Extending Python: rewriting a single method in C
Message-ID: <g0wv9vxms0.fsf@scumbag.ecs.soton.ac.uk>

I have an object-oriented numerical code written in Python. It has
come to the stage where the speed of the code is seriously impeding
progress. Profiling reveals (surprise, surprise) that most of the time
is spent in only 2 methods.

Is there any way of re-writing just these methods in C ?

I would imagine that this sort of thing has been done/attempted many
times by the Scientific Python community, but I haven't managed to
find mcuh.

Thanks,

Jacek


From sholden at holdenweb.com  Fri Mar  9 18:55:26 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Fri, 9 Mar 2001 18:55:26 -0500
Subject: gah! I hate the new string syntax
References: <mailman.983615170.23313.python-list@python.org> <tN5o6.2746$eH3.14913@NewsReader> <Koco6.7518$1D5.244602@e420r-atl1.usenetserver.com> <mailman.983997252.5628.python-list@python.org> <m2vgpkyfjo.fsf@mycroft.actrix.gen.nz> <mailman.984063199.18779.python-list@python.org> <98an7i01tgp@news2.newsguy.com>
Message-ID: <e7eq6.47946$ML1.2142738@e420r-atl2.usenetserver.com>

"Alex Martelli" <aleaxit at yahoo.com> wrote in message
news:98an7i01tgp at news2.newsguy.com...
> "D-Man" <dsh8290 at rit.edu> wrote in message
> news:mailman.984063199.18779.python-list at python.org...
> > On Thu, Mar 08, 2001 at 06:51:07PM +1300, Paul Foley wrote:
> > | On Wed, 7 Mar 2001 15:33:30 -0500, D-Man  wrote:
> > |
> > | > What about one who is both English and Scots, yet never set foot on
> > | > that island ;-)?
> > |
> > | AFAIK, nobody outside the US would admit the existence of such a
> > | thing.  The American habit of calling themselves "Irish" or "Italian"
> > | or whatever when they're not is *very* strange.  I've never heard
> > | anything like it anywhere else.
> >
> > Heh.  Maybe Americans are very odd.  Wouldn't surprise me at all.  ;-)
>
> Maybe.  OTOH, Italians born abroad (from parents who consider
> themselves temporary immigrants, and fully intend to return "home"
> when they've put enough savings aside) have traditionally called
> themselves "Italians" whether that "abroad" was Northern America,
> Argentina, Germany, Libia, or any of the several other countries
> where Italian emigrants used to go.  (Of course, intentions to
> return are not always carried out -- few are aware, for example,
> that such an icon of French popular culture as Yves Montand was
> Italian -- real name Ivo Livi, born of just-as-Italian parents
> in a village near Florence; his father, a socialist, had to flee
> [and ended up in France] when the Fascists took power, and Ivo,
> as well as the rest of the family, followed him a while later;
> I'm not sure when he changed his citizenship over to France).
>
>
> > that I am not really American since none of my ancestors came from
> > America.  "Real" Americans would be the so-called "Indians", or more
>
> Why?  *THEIR* ancestors came from Asia over the Bering Straits,
> after all (and following the line a while longer, one gets back
> to Africa).
>
>
>
> Alex
>
The topic drift on this thread could only be explained by the Gondwanaland
theory!

property-is-theft-ly y'rs  - steve





From diverdown at parsmail.com  Thu Mar  1 12:02:53 2001
From: diverdown at parsmail.com (diverdown at parsmail.com)
Date: 1 Mar 2001 17:02:53 GMT
Subject: *CD SALE*
Message-ID: <97lvbt$4j0$15652@iac5.navix.net>

--_NextPart_00009057-000011FA-0093ADA1-52FD
Content-Type: text/plain
Content-Transfer-Encoding: 7bit

SUPER CD SALE!!!

Jam Packed with information.  Reports, How-to's, screensavers, game cheats, recipes, freebies, Internet
business opportunities, and much, much more.  This CD is packed full of valuable information.  No, this CD
is not intended to make you rich.  However, it can save you money and possibly make you a buck or two.
Order now, supplies are limited.  This CD sells on Ebay for $21.95.  As part of our special promotion, you can
order today for only $4.00 and $1.00 S/H.  That's only $5.00 total.  Order today.  We accept cash, checks and
money orders.  Checks will delay shipping until the check clears.  All other forms of payment gets the CD
shipped out same day.


Please, print out this order form and send it along with your payment. If you do not have a printer, just include
the included information on a separate sheet of paper with your order.  Send all orders to the following address.

Software
510 Alma Avenue
Mulberry, AR  72947




Print Clearly


Name:  _______________________________________________________

Address:  _____________________________________________________

City: _________________________________________________________

State/Province:  __________________ Zip/Postal code:  _______________

Country if other than U.S.:  _______________________________________

E-mail address:________________________________________________


Quantity:______ X 5.00 ______

Total:________





--_NextPart_00009057-000011FA-0093ADA1-52FD--


From aleaxit at yahoo.com  Mon Mar 26 02:55:34 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Mon, 26 Mar 2001 09:55:34 +0200
Subject: Yet Another PEP:  Interface Adapter Mechanism __adapt__
References: <mailman.985588473.13759.python-list@python.org>
Message-ID: <99msll02o92@news2.newsguy.com>

"Alex Shindich" <alex at server01.shindich.com> wrote in message
news:mailman.985588473.13759.python-list at python.org...
    [snip]
> Let's see... The abstract section of the PEP says: "This proposal allows
> a standard and extensible method to ask two types of questions:  Is a
> particular object compliant with a given protocol?" The following
> describes the strategy pattern: "The Strategy pattern provides a means of
> parameterizing a component to accept an algorithm.  This keeps the
> component simpler, and lets clients pick and choose an algorithm
> specifically tailored to their needs." See any similarities?

Some vague ones, maybe.  Protocols need not be algorithms, nor
do components need to 'accept' protocols (it can well be the
other way round, for example).


> > > Another way of looking at your PEP is to say that you are
> > > asking for Python Java-like interfaces.
> >
> > No, actually there are other PEPs that do that.
> >
> That is true, there is PEP-245. But in some sense, an ability to determine
> "is-a" relationship is no different from asking "do you support
> protocol Blah?"

Clark's PEP Without A Number most often ends up asking a more
general question than 'are-you'/'do-you-support', namely:
    "Are you now, *or do you know how to wrap yourself to
        provide*, a supporter of protocol X"?
(when this is asked of the object); or:
    "Does object Y support you, *or do you know how to wrap
     it to obtain such a supporter*"?
(when this is asked of the protocol).

The duality is important, because protocols can be developed
after objects are, OR vice-versa, and this PEP lets either
case be supported 'non-invasively' wrt the pre-existing
component[s].

The question[s] don't receive a simple yes/no answer, which
would be of little use by itself; rather, in the 'yes' case,
the return is the required object-supporting-protocol-X, be
that the input object Y or some wrapper around it (this can
be ignored, and used as a simple boolean yes/no, in the rare
cases where the client only DOES care about a yes/no check,
of course; commonly, it will be saved and used for calls to
the relevant protocol methods).


> Other comments:
> 1. You are making an assumption that each class will provide adaptors to
> some set of protocols, thus entangling the class and available set of
> protocol conversions. Why can't protocol conversion live outside of the
> class?

If I have a pre-existing protocol, say "FileObject", and I design
a new class which I intend to support this pre-existing protocol,
then the most natural residence for the code providing that support,
and the knowledge that the support is provided, seems to be in
the new class itself.

If I have a pre-existing type of objects, say builtin file-objects,
and I design a new protocol that is meant to encompass (maybe with
wrapping) said pre-existing objects, then the natural residence
seems to be in the protocol.

The case for residence in a third place yet seems more remote (both
objects and protocols pre-exist, I just want to add an adaptation)
and might be left as a future enhancement IMHO.


> 2. Is adaption symmetric? If class A provides an adaptor for class B,
> mustn't class B provide an adaptor for class A?

***NO***!!!  That would be absurd even if all classes were protocols.

E.g., a ReadableAndWriteableFileObject can surely be adapted (trivially,
by not wrapping) to satisfy a protocol ReadableFileObject, but that
in no way implies adaptation makes any sense the other way 'round --
read-only files may perfectly well exist (even physically, on a RO
media), why deny the fact in quest of nonexistent adaptation symmetry?

Further, *MANDATING* mutual (circular) dependency is about as far
from best software development practices as one can get.  If A
depends on (needs to know about) B, it's BY FAR better if we can
avoid any dependency of B on A.  Please read all that R. Martin
has to say on his www.objectmentor.com site (publications page)
then move on to Lakos' book (it's written for C++, but it's really
about dependency management as THE key issue with large scale SW).


> 3. Versioning is not mentioned anywhere in your PEP. Once a class defines
> an adaptor for a protocol, the protocol better not change. Otherwise every
> adaptor needs to change to match the new definition of the
> protocol.

Looking at protocols as contracts between suppliers and consumers,
they had better never change unless ALL existing suppliers AND
consumers can be refactored appropriately -- whether one provides
a formally recognized idiom to adapt a few more suppliers to the
protocol, or not.


Alex





From gmunsey at adobe.com  Tue Mar 27 20:15:48 2001
From: gmunsey at adobe.com (Grant Munsey)
Date: Tue, 27 Mar 2001 17:15:48 -0800
Subject: wavlets?
References: <tc27anovquqm79@corp.supernews.co.uk>
Message-ID: <aXaw6.22728$qv3.9950577@nnrp5-w.sbc.net>

What platform are you using. For Windows there is a pretty nice tuned
library called IPP
that does wavelet stuff and a lot more. I have a pretty complete set of
cover routines for
the Windows 1.0b2 version. Intel is about to release version 1.0 and when
they do I was
going to put the Python module source up on the Adobe Open Source site.

http://developer.intel.com/software/products/ipp/index.htm


"Taz" <tariq.rashid at iname.com> wrote in message
news:tc27anovquqm79 at corp.supernews.co.uk...
> i'm about to write a wavelets module for python - does one already
exist...
> couldn't find one?
>
> taz
>
>
>




From aleaxit at yahoo.com  Mon Mar 19 18:15:21 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Tue, 20 Mar 2001 00:15:21 +0100
Subject: windows script host and python
References: <%tKr6.2570$pX.26174@news1.onlynews.com><98oq3r$1e0$1@troll.powertech.no> <8Aps6.104$pl.1515@newsfeed.slurp.net><98tc4d0rmk@news1.newsguy.com> <lvf8bt4p6e6naq0vimgnkhs175eoqbajfr@4ax.com> <mailman.985028069.29930.python-list@python.org>
Message-ID: <9963te04s9@news2.newsguy.com>

"Don Dwiggins" <dwig at advancedmp.net> wrote in message
news:mailman.985028069.29930.python-list at python.org...
    [snip]
> >> processes = GetObject("winmgmts:{impersonationLevel=impersonate}").\
> >> InstancesOf("Win32_process")
    [snip]
> > Cool!  Is this supposed to be NT-only?  It doesn't fly on my Win98 box.
>
> Actually, it fails on my NT workstation also:
    [snip]
>   File "e:\python20\win32com\client\__init__.py", line 87, in Moniker
>     moniker, i, bindCtx = pythoncom.MkParseDisplayName(Pathname)
> pywintypes.com_error: (-2147221020, 'Invalid syntax', None, None)

...but do you have WMI installed?  AFAIK, it doesn't come as default
with either 98 or NT (perhaps 2000 and/or ME -- not sure) -- it has
to be downloaded from the MS site and installed.


Alex





From ecastro at cicei.ulpgc.es  Fri Mar 30 09:48:12 2001
From: ecastro at cicei.ulpgc.es (Enrique)
Date: 30 Mar 2001 13:48:12 -0100
Subject: Printing mixtures in integers and strings
References: <3ac38cf6$0$12246$ed9e5944@reading.news.pipex.net>
Message-ID: <3AC47F42.B545A084@cicei.ulpgc.es>

Hi Neil,

Neil Benn wrote:
> 
> Hello,
> 
>             I'm trying to use something like this:-
> 
> def move(row): #Moves the dispenser to the row on the plate
>     print "Moved to Row - " + row

You are trying to add (operator "+") a number and a string

> 
>     Where row is expected to be a number.  However when I import this into
> teh interpreter I get an error message - stating that __add__ and __radd__
> have not been defined:-
> 
> Traceback (innermost last):
>   File "<console>", line 1, in ?
>   File "c:\projects\scripting\TrialScripting\AcmeBRD.py", li
> TypeError: __add__ nor __radd__ defined for these operands
> 

Which is the expected result since built-in strings and ints
do not add together
You can modify __add__ or __radd__ methods in these object
so they know how to add a number a string or the opposite.


It's easier to convert number to string : print "Moved to
Row - " + str(row), 
or let's 'print' to do this conversion : print "Moved to Row
- ", row
(prints string "Moved to Row - " and then prints printable
form of var "row" which happens to be str(row))

Enrique


From hinsen at cnrs-orleans.fr  Thu Mar 22 10:03:54 2001
From: hinsen at cnrs-orleans.fr (Konrad Hinsen)
Date: 22 Mar 2001 16:03:54 +0100
Subject: PEP 242 Released
References: <mailman.985133006.5845.python-list@python.org>
Message-ID: <m3ofutzvz9.fsf@chinon.cnrs-orleans.fr>

"Paul F. Dubois" <paul at pfdubois.com> writes:

> PEP: 242
> Title: Numeric Kinds

Looks good overall. I am just wondering how much of a problem this is
in real life - I have used Python on several platforms, and on all of
them a float was an IEEE double.

> Coercion
> 
>     In an expression, coercion between different kinds is to the
>     greater kind.  For this purpose, all complex kinds are "greater
>     than" all floating-point kinds, and all floating-point kinds are
>     "greater than" all integer kinds.

This might lead to a loss of precision when combining a high-precision
float with a low-precision complex number. It would be better to use a
complex type whose precision is determined by coercion between the
real part of the complex number and the other float.

> Open Issues
> 
>     The assertion that a decimal literal means a binary floating-point
>     value of the largest available kind is in conflict with other
>     proposals about Python's numeric model.  This PEP asserts that
>     these other proposals are wrong and that part of them should not
>     be implemented.

Do you mean Moshe Zadka's proposal for rational numbers and
corresponding literals? One could reconcile both proposals by deciding
that Moshe's rationals are equivalent to infinite-precision floats.
Then literals would always be rationals and the kind objects would
convert to the desired float type.
-- 
-------------------------------------------------------------------------------
Konrad Hinsen                            | E-Mail: hinsen at cnrs-orleans.fr
Centre de Biophysique Moleculaire (CNRS) | Tel.: +33-2.38.25.56.24
Rue Charles Sadron                       | Fax:  +33-2.38.63.15.17
45071 Orleans Cedex 2                    | Deutsch/Esperanto/English/
France                                   | Nederlands/Francais
-------------------------------------------------------------------------------


From fdrake at acm.org  Sat Mar  3 14:54:31 2001
From: fdrake at acm.org (Fred L. Drake, Jr.)
Date: Sat, 3 Mar 2001 14:54:31 -0500 (EST)
Subject: Bug fix releases
In-Reply-To: <200103031944.OAA21835@cj20424-a.reston1.va.home.com>
References: <LNBBLJKPBEHFEDALKOLCMEEPJBAA.tim.one@home.com>
	<mailman.982897324.9109.python-list@python.org>
	<200103031821.NAA24060@panix3.panix.com>
	<200103031910.OAA21663@cj20424-a.reston1.va.home.com>
	<15009.17989.88203.844343@cj42289-a.reston1.va.home.com>
	<200103031944.OAA21835@cj20424-a.reston1.va.home.com>
Message-ID: <15009.19447.154958.449303@cj42289-a.reston1.va.home.com>

Guido van Rossum writes:
 > But of course the library repackaging in itself would cause enormous
 > outcries, because in a very real sense it *does* break code.

  That's why it has to be a feature release.  ;-)


  -Fred

-- 
Fred L. Drake, Jr.  <fdrake at acm.org>
PythonLabs at Digital Creations




From scarblac at pino.selwerd.nl  Tue Mar 13 03:12:31 2001
From: scarblac at pino.selwerd.nl (Remco Gerlich)
Date: 13 Mar 2001 08:12:31 GMT
Subject: Very Cool
References: <g13r6.201$qt5.10920@news3.mia> <OYer6.280$qt5.17754@news3.mia>
Message-ID: <slrn9arlfg.bjg.scarblac@pino.selwerd.nl>

Brendhan Horne <brendhanhorne at bellsouth.net> wrote in comp.lang.python:
> Everybody thank you for your help you may have broken the one barrier that
> was causing me not to understand this. Now let explain what happened and
> maybe you can clarify something even futher for me.
> I went in to my Python 20 command line interpreter the DOS based one ( am I
> saying that right )  And wrote in what you guys sent me.
> Items in parenthsis () are notes about what I did or what happen nothing in
> those parenthesis was ever typed in.
> 
> >>>while 1:    ( I wrote this in )(press enter key
> ....    print "Hello World"  ( the dots would appear as soon as I entered in
> while 1:)
> File "<stdin>", line 2
>    print  "Hello World"
>        (there is a control symbol under the t in print)
> Indentation Error: expected an indented block


You didn't indent the line with print on it.  You have to put some
whitespace (say, a space) before the print otherwise it won't be a block
belonging to the while:.

In the editor, you probably did write the spaces.

When you put the spaces and hit enter after the print line, it will give
another line with dots, where you can continue the block. Just hit enter and
it will run.

-- 
Remco Gerlich


From calcium at ozemail.com.au  Fri Mar 30 07:05:55 2001
From: calcium at ozemail.com.au (ccang)
Date: Fri, 30 Mar 2001 22:05:55 +1000
Subject: General CGI advice...
References: <mailman.985798478.5593.python-list@python.org>
Message-ID: <3AC476A3.7739E221@ozemail.com.au>

http://www.ozemail.com.au/~calcium/publications.html

"Vincent A. Primavera" wrote:

> Hello,
>         Looking for some general guidance...  I am relatively new to Python and
> pretty much totally green to CGI programming.  However, I would like to start
> learning more about CGI using Python as my means.  Do I need to focus on
> generic CGI concepts first or are there any really good resources available
> to integrate the two immediately?
>
>         Thank You,
>         Vincent A. Primavera

--
| mailto:calcium at ozemail.com.au   | +61 401 688 408
|  http://www.ozemail.com.au/~calcium
|  To me, boxing is like a ballet except there's no music, no
|  choreography, and the dancers hit each other - Jack Handy




From ssartor at bellatlantic.net  Thu Mar 29 07:19:53 2001
From: ssartor at bellatlantic.net (Steven Sartorius)
Date: Thu, 29 Mar 2001 12:19:53 GMT
Subject: How to update status bar in wxPython?
References: <5Tpw6.2172$K%.667185@typhoon1.ba-dsg.net> <OEtw6.238$nyH.5111956@news.randori.com>
Message-ID: <JLFw6.2918$j_3.1182024@typhoon2.ba-dsg.net>

In article <OEtw6.238$nyH.5111956 at news.randori.com>, "Robin Dunn"
<robin at stop.spam.alldunn.com> wrote:

> "Steven Sartorius" <ssartor at bellatlantic.net> wrote in message
> news:5Tpw6.2172$K%.667185 at typhoon1.ba-dsg.net...
>> I'm a newbie to GUI programming in general so apologies if this is a
> FAQ....
>>
>> I've got a Python script that does some heavy duty number crunching and
> file
>> I/O.  To track the progress of the script I print a series of messages
>> to
> the
>> console/x-term ("Working on file #....", etc,etc).  As a project, I
> decided to
>> guify the script with wxPython and Boa-constructor.  I've successfully
> built a
>> window with a single button, a couple of text input fields and a status
> bar;
>> clicking the button fires off the script.  What I'm trying to do (so
>> far unsuccessfully) is redirect the progress messages from the console
>> to the status bar on the window.  Because I'm retrofitting an existing
>> script,
> I've
>> just replaced the original 'print' statements with
>> 'statusBar.SetStatusText("Working on file....") and (for good measure)
>> 'statusBar.Refresh()'.  This approach kind of works....the status bar
> shows
>> the final progress message ("All data processed")  but only after I
>> force
> the
>> window to redraw (moving it or covering and then uncovering it).  I'm
>> sure
> I'm
>> missing something basic here so any help would be much appreciated.
>>
> 
> When you click the button and "fire off the script" you are effectivly
> blocking the GUI until the script is complete.  You need to allow the
> GUI's main loop to run in order for new events to be dispatched and
> other housework types of things to happen.  When you call
> statusBar.Refresh() you are simply requesting that a paint event be sent
> ot the statusBar, but as you've seen if the event is never dispatched to
> the window then it is never actually refreshed.
> 
> One way to allow the main loop to run from within the middle of some
> long running task is to call wxYield, which will return when there are
> no more events waiting to be proccessed.  Unfortunatly there are
> potential reentrancy issues you need to deal with when using wxYield,
> for example what if the user clicks on your launcher button while a
> script is already running, and then the code calls wxYield?
> 
> A better way is to refactor your long running task so that it can be
> executed a chunk (or iteration) at a time from an EVT_IDLE event
> handler. Each time the handler is called do the next chunk of the task
> and if there is more to be done call event.RequestMore() before
> returning.  This lets the code return to the main loop frequently so the
> GUI continues receiving events and can be updated, but all extra time is
> spent running your task.
> 
> If the task is not easily broken into chunks then you can run it in a
> background thread, but then there are a few things you need to do
> specially to allow your thread to communicate back to the GUI thread. 
> See the demo and also the wxpython-users archives for details.
> 
> --
> Robin Dunn Software Craftsman robin at AllDunn.com       Java give you
> jitters? http://wxPython.org      Relax with wxPython!


Thanks for the response.  I kind to thought I might have to restructure
my code.  Good thing I'm not doing anything this weekend....


Steve


From cg at schlund.de  Mon Mar  5 10:29:20 2001
From: cg at schlund.de (Carsten Gaebler)
Date: Mon, 05 Mar 2001 16:29:20 +0100
Subject: Access to sybase API
References: <obOo6.1158$f07.1696858@nnrp6.proxad.net>
Message-ID: <3AA3B0D0.90876BA7@schlund.de>

Neotaku wrote:
> I search a code source for python to access Sybase thru API connexion the
> Sybase DB-Library API or Sybase CT-Library (aka the Client Library) API

Have a look at Dave Cole's Sybase module:
http://www.object-craft.com.au/projects/sybase/

cg.


From tim.one at home.com  Sat Mar 17 01:38:17 2001
From: tim.one at home.com (Tim Peters)
Date: Sat, 17 Mar 2001 01:38:17 -0500
Subject: Pep 238 Why new // operator (Non-integer Division)
In-Reply-To: <3AB2ED5E.A9B44B1E@seebelow.org>
Message-ID: <LNBBLJKPBEHFEDALKOLCAEEFJGAA.tim.one@home.com>

[Grant Griffin]
> BTW, I guess I'm not the first one to notice that this new operator
> looks an awful lot like a C/C++ comment.

No, you're the first!  329 people did note that // looks an awful lot like
Fortran's string concatenation operator.  I expect these were the same 329
who complained to the C++ committee about abusing *their* prior art.

> So maybe some other symbol should be used.

Why bother?  Python's # looks an awful lot like C++'s preprocessor directive
flag too.  Guido would probably prefer infix "div" (a la Pascal), but a new
keyword is a very hard sell.

> Hey, isn't "$" still open? <wink>

Barry has first dibs on that one, to make

    print $ fileobject, 2, "+ 5% tax =", 2*1.05

print

    $2.00 + $0.10 = $2.10

Curiously, he will ignore the fileobject in this case, unless it's spelled

    print $ >> fileobject, 2, "+ 5% tax =", 2*1.05

I forget what

    print >> $fileobject, 2, "+ 5% tax =", 2*1.05

is going to mean.  I *think* it reads the next line from fileobject, strips
the newline, then uses that as the name of the file to print to (after
opening for append).

> just-tryin'-to-make-a-buck-ly y'rs,

sell-documentation-for-barry's-proposals!-ly y'rs  - tim




From n8gray at caltech.edu.is.my.e-mail.address  Thu Mar 22 22:32:56 2001
From: n8gray at caltech.edu.is.my.e-mail.address (Nathaniel Gray)
Date: Thu, 22 Mar 2001 19:32:56 -0800
Subject: Good Editor?
References: <3ab9a32b$1_1@news2.one.net>
Message-ID: <99eggq$mvk@gap.cco.caltech.edu>

Sorry if this goes through twice, you never know with news server errors 
whether your post got posted.

fjs wrote:

> Can someone recomend a good editor to write python in?  Im just getting
> started learning python and dont know which ones support python.
> 
> thanks

On *NIX check out NEdit (http://www.nedit.org).  It's got syntax 
highlighting support for a ton of languages, a simple and powerful 
scripting language, rock-solid stability, and it's light on resources.  
Plus, you don't have to time-warp backwards 30 years to understand the 
keyboard accellerators (unless you already think C-x C-f for "open file" is 
intuitive.  I'd bash the vi command too if I'd ever bothered to learn what 
it was...).

-n8

-- 
_.~'^`~._.~'^`~._.~'^`~._.~'^`~._.~'^`~._
             Nathaniel Gray
   California Institute of Technology
     Computation and Neural Systems
     n8gray <at> caltech <dot> edu
_.~'^`~._.~'^`~._.~'^`~._.~'^`~._.~'^`~._



From LLewin at oreilly.com  Fri Mar  9 16:51:48 2001
From: LLewin at oreilly.com (LLewin at oreilly.com)
Date: Fri, 09 Mar 2001 21:51:48 GMT
Subject: Python In A Nutshell
References: <nG5q6.54910$__6.11376601@typhoon.southeast.rr.com>
Message-ID: <Ufcq6.140786$47.2078615@news.bc.tac.net>

Hi,
Yep--Python Nutshell is "forthcoming"...we're just in the planning stages at this point--very early on in the process.  Any suggestions are welcome!
Laura
LLewin at oreilly.com

> Hey Alex,  congrats on landing "Python In A Nutshell" book deal.  One
> problem.  I went down to the book store and it's not there.  What gives?
> <wink>
> 
> Don
> 
> 


==================================
Posted via http://nodevice.com
Linux Programmer's Site


From vsbabu at erols.com  Mon Mar  5 18:56:38 2001
From: vsbabu at erols.com (Satheesh Babu)
Date: Mon, 5 Mar 2001 18:56:38 -0500
Subject: iis and python
References: <F5622749FED6DA5D.AF4594816A4865C9.16CD98F1C1FAC3F6@lp.airnews.net>
Message-ID: <98191m$88j$1@bob.news.rcn.net>

Hi,

ASP with Python works really well. Additionally, Python's vast collection of
modules
make a short work of many tasks ASP/VBScript pundits would need a compiled
COM object for.

We've some ASP/VB code. To extend it to connect to some network services
I started using Python, mixed with VB and it rocks. When we get time,
I'm certainly going to convert all VB stuff to Python.

I  use ActiveState Python..

--
v.s.babu
vsbabu at erols.com
http://vsbabu.csoft.net
"tech" <im_cne at hotmail.com> wrote in message
news:F5622749FED6DA5D.AF4594816A4865C9.16CD98F1C1FAC3F6 at lp.airnews.net...
> I am going to do a project building a web front end  while my friend is
> going to build the back end.
>
> The backend will be MS SQL. The front end will be a web interface, IIS 5.0
> or whatever comes with win2k .
>
> I have a choice of doing php or python.
>
> In order to use python it would have to entail cgi or ASP. Is this
correct.
> Or is there an alternative. I dont know if Zope is the right thing. I hate
> DTML.
>
> Is using python in ASP a pain in the but? Am I going to have to learn a
> whole technology just to implelment python well into it?
>
> I hear troubleshooting ASP pages is not fun.
>
> The only time I have used iis is with php and it was a total snap.
Painless.
>
>




From shaleh at valinux.com  Mon Mar 26 18:12:03 2001
From: shaleh at valinux.com (Sean 'Shaleh' Perry)
Date: Mon, 26 Mar 2001 15:12:03 -0800 (PST)
Subject: fixing map(None, [1,2,3]) ?
In-Reply-To: <873dc040ny.fsf@mathdogs.com>
Message-ID: <XFMail.20010326151203.shaleh@valinux.com>

On 26-Mar-2001 Mike Coleman wrote:
> Shouldn't "map(None, [1,2,3])" return "[(1,), (2,), (3,)]"?
> 
> Is there any hope that this could be fixed?
> 

map runs the function on each element of the list it is passed and returns an
element of a new list.  Why should it magically return a list of one-tuples?



From sourin at alumnux.com  Thu Mar 29 09:36:38 2001
From: sourin at alumnux.com (Sourin Sarkar)
Date: Thu, 29 Mar 2001 20:06:38 +0530
Subject: (no subject)
Message-ID: <NFBBJDKCIKBNLKFLDOBGMEAGCAAA.sourin@alumnux.com>

Can anyone explain the usage of:

   win32api.Apply() (in detail please.) function of the win32 extension add
on to the python library.





From mcfletch at home.com  Mon Mar 19 03:42:39 2001
From: mcfletch at home.com (Mike C. Fletcher)
Date: Mon, 19 Mar 2001 03:42:39 -0500
Subject: Help Rewriting/Refactoring/Rethinking Parsing Algorythm
In-Reply-To: <LNBBLJKPBEHFEDALKOLCOEJAJGAA.tim.one@home.com>
Message-ID: <000701c0b050$8f566510$a9807018@cr706570a>

Thanks Tim.  I'd actually downloaded it when it was first announced and
played with it back with Dragon 3 (I uninstalled it 'cause I was crashing
too often).  Re-downloaded and played with it this evening.  Don't have time
to do it up myself, but looks like it'd be quite simple to make a PythonWin
(or IDLE or Boa)-specific grammar with all sorts of neat macros (for x
indexing y, for instance).

Biggest need I can see in the project is a template using wxPython (or
Tkinter) and a styled text control (aka Scintilla) to do "proper"
update-the-dictation-object-with-minimal-diffs-and-current-context work (as
described in the source files).  From there it should be possible to start
building voice-aware editors with neato features.

Quicky script to load up the "py-commands" seems to crude by comparison :) .
Mike


-----Original Message-----
From: Tim Peters [mailto:tim.one at home.com]
Sent: Sunday, March 18, 2001 11:51 PM
To: python-list at python.org
Subject: RE: Help Rewriting/Refactoring/Rethinking Parsing Algorythm
...
Note that Joel Gould (Dragon NatSpeak's chief architect) developed a system
that lets you script NatSpeak from Python:

    http://www.synapseadaptive.com/joel/
...




From paul at pfdubois.com  Tue Mar 20 18:56:06 2001
From: paul at pfdubois.com (Paul F. Dubois)
Date: Tue, 20 Mar 2001 15:56:06 -0800
Subject: PEP 242 Released
Message-ID: <ADEOIFHFONCLEEPKCACCGEANCHAA.paul@pfdubois.com>

PEP: 242
Title: Numeric Kinds
Version: $Revision: 1.1 $
Author: paul at pfdubois.com (Paul F. Dubois)
Status: Draft
Type: Standards Track
Created: 17-Mar-2001
Python-Version: 2.2
Post-History:


Abstract

    This proposal gives the user optional control over the precision
    and range of numeric computations so that a computation can be
    written once and run anywhere with at least the desired precision
    and range.  It is backward compatible with existing code.  The
    meaning of decimal literals is clarified.


Rationale

    Currently it is impossible in every language except Fortran 90 to
    write a program in a portable way that uses floating point and
    gets roughly the same answer regardless of platform -- or refuses
    to compile if that is not possible.  Python currently has only one
    floating point type, equal to a C double in the C implementation.

    No type exists corresponding to single or quad floats.  It would
    complicate the language to try to introduce such types directly
    and their subsequent use would not be portable.  This proposal is
    similar to the Fortran 90 "kind" solution, adapted to the Python
    environment.  With this facility an entire calculation can be
    switched from one level of precision to another by changing a
    single line.  If the desired precision does not exist on a
    particular machine, the program will fail rather than get the
    wrong answer.  Since coding in this style would involve an early
    call to the routine that will fail, this is the next best thing to
    not compiling.


Supported Kinds

    Each Python compiler may define as many "kinds" of integer and
    floating point numbers as it likes, except that it must support at
    least two kinds of integer corresponding to the existing int and
    long, and must support at least one kind of floating point number,
    equivalent to the present float.  The range and precision of the
    these kinds are processor dependent, as at present, except for the
    "long integer" kind, which can hold an arbitrary integer.  The
    built-in functions int(), float(), long() and complex() convert
    inputs to these default kinds as they do at present.  (Note that a
    Unicode string is actually a different "kind" of string and that a
    sufficiently knowledgeable person might be able to expand this PEP
    to cover that case.)

    Within each type (integer, floating, and complex) the compiler
    supports a linearly-ordered set of kinds, with the ordering
    determined by the ability to hold numbers of an increased range
    and/or precision.


Kind Objects

    Three new standard functions are defined in a module named
    "kinds".  They return callable objects called kind objects.  Each
    int or floating kind object f has the signature result = f(x), and
    each complex kind object has the signature result = f(x, y=0.).

    int_kind(n)
        For n >= 1, return a callable object whose result is an
        integer kind that will hold an integer number in the open
        interval (-10**n,10**n).  This function always succeeds, since
        it can return the 'long' kind if it has to. The kind object
        accepts arguments that are integers including longs.  If n ==
        0, returns the kind object corresponding to long.

    float_kind(nd, n)
        For nd >= 0 and n >= 1, return a callable object whose result
        is a floating point kind that will hold a floating-point
        number with at least nd digits of precision and a base-10
        exponent in the open interval (-n, n).  The kind object
        accepts arguments that are integer or real.

    complex_kind(nd, n)
        Return a callable object whose result is a complex kind that
        will will hold a complex number each of whose components
        (.real, .imag) is of kind float_kind(nd, n).  The kind object
        will accept one argument that is integer, real, or complex, or
        two arguments, each integer or real.

    The compiler will return a kind object corresponding to the least
    of its available set of kinds for that type that has the desired
    properties.  If no kind with the desired qualities exists in a
    given implementation an OverflowError exception is thrown.  A kind
    function converts its argument to the target kind, but if the
    result does not fit in the target kind's range, an OverflowError
    exception is thrown.

    Kind objects also accept a string argument for conversion of
    literal notation to their kind.

    Besides their callable behavior, kind objects have attributes
    giving the traits of the kind in question.  The list of traits
    needs to be completed.


The Meaning of Literal Values

    Literal integer values without a trailing L are of the least
    integer kind required to represent them.  An integer literal with
    a trailing L is a long.  Literal decimal values are of the
    greatest available binary floating-point kind.


Concerning Infinite Floating Precision

    This section makes no proposals and can be omitted from
    consideration.  It is for illuminating an intentionally
    unimplemented 'corner' of the design.

    This PEP does not propose the creation of an infinite precision
    floating point type, just leaves room for it.  Just as int_kind(0)
    returns the long kind object, if in the future an infinitely
    precise decimal kind is available, float_kind(0,0) could return a
    function that converts to that type.  Since such a kind function
    accepts string arguments, programs could then be written that are
    completely precise.  Perhaps in analogy to r'a raw string', 1.3r
    might be available as syntactic sugar for calling the infinite
    floating kind object with argument '1.3'.  r could be thought of
    as meaning 'rational'.


Complex numbers and kinds

    Complex numbers are always pairs of floating-point numbers with
    the same kind.  A Python compiler must support a complex analog of
    each floating point kind it supports, if it supports complex
    numbers at all.


Coercion

    In an expression, coercion between different kinds is to the
    greater kind.  For this purpose, all complex kinds are "greater
    than" all floating-point kinds, and all floating-point kinds are
    "greater than" all integer kinds.


Examples

    In module myprecision.py:

        import kinds
        tinyint = kinds.int_kind(1)
        single = kinds.float_kind(6, 90)
        double = kinds.float_kind(15, 300)
        csingle = kinds.complex_kind(6, 90)

    In the rest of my code:

        from myprecision import tinyint, single, double, csingle
        n = tinyint(3)
        x = double(1.e20)
        z = 1.2
        # builtin float gets you the default float kind, properties unknown
        w = x * float(x)
        w = x * double(z)
        u = csingle(x + z * 1.0j)
        u2 = csingle(x+z, 1.0)

    Note how that entire code can then be changed to a higher
    precision by changing the arguments in myprecision.py.

    Comment: note that you aren't promised that single != double; but
    you are promised that double(1.e20) will hold a number with 15
    decimal digits of precision and a range up to 10**300 or that the
    float_kind call will fail.


Open Issues

    The assertion that a decimal literal means a binary floating-point
    value of the largest available kind is in conflict with other
    proposals about Python's numeric model.  This PEP asserts that
    these other proposals are wrong and that part of them should not
    be implemented.

    Determine the exact list of traits for integer and floating point
    numbers.  There are some standard Fortran routines that do this
    but I have to track them down.  Also there should be information
    sufficient to create a Numeric array of an equal or greater kind.


Copyright

    This document has been placed in the public domain.




Local Variables:
mode: indented-text
indent-tabs-mode: nil
End:




From ws-news at gmx.at  Tue Mar  6 16:27:50 2001
From: ws-news at gmx.at (Werner Schiendl)
Date: Tue, 6 Mar 2001 22:27:50 +0100
Subject: Can anyone recomend a good intoduction to C...
References: <mailman.983894288.19965.python-list@python.org> <983896344.701456@newsmaster-04.atnet.at> <m3lmqibw0a.fsf@atrus.jesus.cam.ac.uk> <983911749.217622@newsmaster-04.atnet.at> <roy-32CDC5.15461906032001@news.panix.com>
Message-ID: <983914348.247943@newsmaster-04.atnet.at>

This is exactly my opinion.

I think right now, given the instruction sets and really hard to predict
design of CPUs (parallel execution, which instruction support that, which
sequence is better...) makes it a really hard task to do hand optimization.

There, for sure, are some experts that still can. But they, I guess, do that
all the time over...

Personally I try to avoid asm whereever possible :-)

regards
werner


Roy Smith <roy at panix.com> wrote in message
news:roy-32CDC5.15461906032001 at news.panix.com...
> "Werner Schiendl" <ws-news at gmx.at> wrote:
> > Years ago I had a lot of discussions why use C instead of Assembler.
> > The arguments were mainly for performance.
>
> I wouldn't be surprised if a good optimizing C compiler can actually
> produce faster code than hand-coded assembler in many cases.




From jhauser at ifm.uni-kiel.de  Fri Mar 23 10:24:48 2001
From: jhauser at ifm.uni-kiel.de (Janko Hauser)
Date: 23 Mar 2001 16:24:48 +0100
Subject: newbe   how to read in a string?
References: <3ABB4E70.54885C9F@noos.fr> <99fo39$gsq$2@news.mathworks.com>
Message-ID: <87ofush5j3.fsf@lisboa.ifm.uni-kiel.de>

As posted numerous times, there is

ftp://ftp.python.org/pub/python/contrib-09-Dec-1999/All/sscanfmodule.c

HTH,
__Janko

-- 
  Institut fuer Meereskunde             phone: 49-431-597 3989
  Dept. Theoretical Oceanography        fax  : 49-431-565876
  Duesternbrooker Weg 20                email: jhauser at ifm.uni-kiel.de
  24105 Kiel, Germany


From jmarshal at mathworks.com  Thu Mar  1 12:44:26 2001
From: jmarshal at mathworks.com (Joshua Marshall)
Date: 1 Mar 2001 17:44:26 GMT
Subject: PEP 227 (was Re: Nested scopes resolution -- you can breathe again!)
References: <LNBBLJKPBEHFEDALKOLCMEEPJBAA.tim.one@home.com> <200102230259.VAA19238@cj20424-a.reston1.va.home.com> <mailman.983384008.21434.python-list@python.org> <97lqsr0311d@news2.newsguy.com>
Message-ID: <97m1pq$3au$1@news.mathworks.com>

Alex Martelli <aleaxit at yahoo.com> wrote:
> "Jeremy Hylton" <jeremy at alum.mit.edu> wrote in message
> news:mailman.983384008.21434.python-list at python.org...
>     [snip]
>>   Rebinding names in enclosing scopes
>     [snip]
>>       opposed to it.  It is difficult to support, because it would
>>       require a new mechanism that would allow the programmer to specify
>>       that an assignment in a block is supposed to rebind the name in an
>>       enclosing block; presumably a keyword or special syntax (x := 3)
>>       would make this possible.

> If this were to be allowed (and I'm not saying it should be!), the
> most natural way to allow it would appear to be to be a new clause
> on the global statement -- most natural, because "allow rebinding
> in an outer scope" is exactly global's job (of course, up to Python
> 2, the only 'outer scope' was the module-one:-).

> Maybe something like...:

>     def spam():
>         def inner():
>             global in spam, x
>             x += 1
>         x = 23
>         inner()

Or how about just using "global" as is?  If you declare a variable
reference global, then search the enclosing scopes until you find the
right name.  It will break a little code, but on the same order as
this incompatible change will already break.


From paulp at ActiveState.com  Mon Mar 12 07:35:46 2001
From: paulp at ActiveState.com (Paul Prescod)
Date: Mon, 12 Mar 2001 04:35:46 -0800
Subject: Extending Python: rewriting a single method in C
References: <g0wv9vxms0.fsf@scumbag.ecs.soton.ac.uk>
Message-ID: <3AACC2A2.B7EC96D4@ActiveState.com>

Jacek Generowicz wrote:
> 
> ..
> 
> Is there any way of re-writing just these methods in C ?

Sure! In fact, it is down-right easy if you are a C programmer. You just
put the two functions in an extension module and call them from C. 

http://www.python.org/doc/current/ext/ext.html

Also, look at this example module: (you'll have to paste together the
URL)

http://cvs.sourceforge.net/cgi-bin/cvsweb.cgi/python/dist/src/Modules/xxmodule.c?rev=2.22&content-type=text/x-cvsweb-markup&cvsroot=python

-- 
Python:
    Programming the way
    Guido
    indented it.
       - (originated with Skip Montanaro?)



From aleaxit at yahoo.com  Sun Mar 25 11:56:19 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sun, 25 Mar 2001 18:56:19 +0200
Subject: Referencing a container object's attributes
References: <mailman.985486832.26248.python-list@python.org>
Message-ID: <99l80s11enf@news1.newsguy.com>

"Tim Churches" <tchur at optushome.com.au> wrote in message
news:mailman.985486832.26248.python-list at python.org...
> Is there any way for a method of a class instance which is embedded in a
> container class instance to look up attributes of that container class
> instance (i.e. the one which contains it)? At the moment I am putting

Zope uses metaclasses for something like that == "inheritance" (sort
of) from a container.  Not sure how applicable that is to your needs...


Alex





From qrczak at knm.org.pl  Mon Mar  5 11:49:57 2001
From: qrczak at knm.org.pl (Marcin 'Qrczak' Kowalczyk)
Date: 5 Mar 2001 16:49:57 GMT
Subject: Java guy interested in Python
References: <003801c0a459$a29c97f0$63da1542@wisdom> <mailman.983688785.707.python-list@python.org>
Message-ID: <slrn9a7gtl.ca6.qrczak@qrnik.zagroda>

Sat, 3 Mar 2001 22:54:16 -0800, Ken Seehof <kens at sightreader.com> pisze:

> Many programmers are confused in that they think the purpose of type
> declarations is to protect the programmer from errors by providing
> enforcement of "programming by contract".

I use static typing to protect myself from errors, and to define
interfaces, and to dispatch polymorphic code (on types rather than
values), and to think about a piece of code before I write it. So it
definitely is a purpose.

> IMHO, type declarations are a sacrifice necessary to give compilers
> enough information to produce efficient code.

This is not true. It's simply not the only purpose.

> Type declarations make sense in C and C++, but they are (IMHO)
> a design flaw in high level languages.

Please don't attribute flaws of C, C++ and Java to static typing
itself. Haskell, ML, Mercury are at least as high level as Python and
have rich type systems, which IMHO makes these languages convenient
and powerful.

Neither static typing nor dynamic typing is better suited than the
other for all purposes. Both approaches have very strong good reasons.
>From this it follows that static vs. dynamic typing flamewars will
never end :-)

-- 
 __("<  Marcin Kowalczyk * qrczak at knm.org.pl http://qrczak.ids.net.pl/
 \__/
  ^^                      SYGNATURA ZAST?PCZA
QRCZAK


From philmi at pliocene.cenozoic  Mon Mar 26 22:27:59 2001
From: philmi at pliocene.cenozoic (Phil Mitchell)
Date: 26 Mar 2001 22:27:59 -0500
Subject: python or smalltalk for persistence-based app?
Message-ID: <m3d7b3x54w.fsf@pliocene.cenozoic>

I need an interactive, OOP language with good support for persistence. I'm
looking at python plus ZODB versus smalltalk and gemstone. I don't know either
language at this point (I know C, Java, Perl). The intended use is to develop
some ideas for a desktop productivity tools framework; there'll be some gui
stuff and a lot of generic info management stuff (eg., managing text data
organized in trees).

Right now I'm leaning towards python, b/c ZODB is open-source and gemstone isn't
(although there's some talk over at squeak of working on such a thing). I don't
have a good sense of the relative depth/quality of libraries (and this is not a
funded project that will be able to buy commercial ones). 

I know that's not a lot to go on, but I'd welcome any thoughts...


-- 
sentinel %at pobox $dot com


From embed at geocities.com  Wed Mar 14 13:40:59 2001
From: embed at geocities.com (Warren Postma)
Date: Wed, 14 Mar 2001 13:40:59 -0500
Subject: Dynamic construction of a print format string
References: <3AAFB663.57FCCE7@troikanetworks.com> <yVOr6.9485$TW.38415@tor-nn1.netcom.ca>
Message-ID: <YYOr6.9486$TW.38270@tor-nn1.netcom.ca>

whoops... how about...
fmtstr = " %"+`namelen`+"s - %s"




From qrczak at knm.org.pl  Mon Mar 19 11:05:44 2001
From: qrczak at knm.org.pl (Marcin 'Qrczak' Kowalczyk)
Date: 19 Mar 2001 16:05:44 GMT
Subject: Who's minister of propaganda this week?
References: <B98041329020E007.7DA84CAD9D0562AB.817BBC572F480869@lp.airnews.net> <98nu1m$12e@dispatch.concentric.net> <660ACCA1E4816F7E.C359CDF4168EF3FB.3B6B05538D619770@lp.airnews.net> <3AAFA13D.76E32C3F@divalsim.it> <98oa4o01i4v@news2.newsguy.com> <mailman.984619039.7856.python-list@python.org> <98q9h60122v@news2.newsguy.com> <mailman.984937946.5318.python-list@python.org> <9937en0bge@news2.newsguy.com> <slrn9bagbm.toq.qrczak@qrnik.zagroda> <994ab801h5o@news2.newsguy.com> <slrn9bbin2.11r.qrczak@qrnik.zagroda> <994tg9026qa@news1.newsguy.com>
Message-ID: <slrn9bcbio.3tr.qrczak@qrnik.zagroda>

Mon, 19 Mar 2001 13:19:22 +0100, Alex Martelli <aleaxit at yahoo.com> pisze:

> > OTOH Haskell and OCaml don't work by implicit coercion of elements
> > to a common supertype.
> 
> Isn't that what the "O" in O'Caml adds to plain good old Caml?

Well, coercion upwards is explicit, but it's not as often as in C++ or
Java because functions generally accept "object of any type providing
these methods of these types" instead of "object coerced to the class
which declares these methods".

There is no coercion downwards (except unsafe Obj.magic) nor typecase.

The class system is rarely used. The module system, higher order
functions or algebraic types are often more appropriate and better
fit the rest of the language. There is "O" in the language, but in
practice it often doesn't matter.

> > Note that in these languages it's meaningless to mix all types of
> > objects together in a container. If they are unrelated, you can't
> > apply any operation to an unspecified member of them.
> 
> Not sure if by "these" you mean, here, "the former" (C++, Java, ...)
> or "the latter" (ML, Haskell, ...).

ML, Haskell.

> Assuming now that by 'these' you mean 'Haskell' (or ML or Caml
> without the 'O'), it's not true that I don't _want_ to built
> heterogeneous, structured containers -- it's that I _can't_ do
> it conveniently,

It's both.

> because of the language limitations imposed by the desire to allow
> static type-checking.

I don't call this a limitation. You could embed a complete Python-like
object model in Haskell or ML. Python's classes won't correspond
to types in Haskell or ML, but to values, with explicit method
dictionaries etc.

Syntax might be a bit awkward, and the library would have to be
adapted, but in can be done. High level functional languages are
especially convenient in building a sublanguage with its own rules
inside them. Although I don't think that anybody would make a totally
isolated object model, because of necessary glue with the rest of
libraries.

Viewed from that point, it's not a limitation wrt. Python's model,
but an extension.

> Objects are rarely mutable _in functional programming_ -- they
> are commonly mutable _in object oriented programming_, and this
> makes a very substantial difference.

What about object oriented functional programming?

I guess they are often immutable, as opposed to functional object
oriented programming :-)

-- 
 __("<  Marcin Kowalczyk * qrczak at knm.org.pl http://qrczak.ids.net.pl/
 \__/
  ^^                      SYGNATURA ZAST?PCZA
QRCZAK


From loewis at informatik.hu-berlin.de  Fri Mar  2 11:06:24 2001
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 02 Mar 2001 17:06:24 +0100
Subject: A counter-proposal to __future__ in PEP 236
References: <j47l29lpo8.fsf@informatik.hu-berlin.de> <97ll2u02qmp@news2.newsguy.com> <3A9FBE5C.95615241@sewardweb.com>
Message-ID: <j4ae74jgu7.fsf@informatik.hu-berlin.de>

Lloyd Sommerer <lsommerer at sewardweb.com> writes:

> This is obviously very distinct from its use with slices.  But it seems
> reasonable to me that new python programmers would wonder if this:
> 
> directive: nested_scopes
> 
> might be equivalent to:
> 
> directive:
>      nested_scope

Indeed, that is also what I would expect, and there is nothing wrong
with that expectation. The confusing part would be that you'd expect
to put statements into the directive block, when you really only put
individual names there.

Regards,
Martin



From dvdeug at x8b4e53cd.dhcp.okstate.edu  Thu Mar  1 12:58:45 2001
From: dvdeug at x8b4e53cd.dhcp.okstate.edu (David Starner)
Date: 1 Mar 2001 17:58:45 GMT
Subject: [ANNOUNCE] pyAda, first public release
References: <3A9DB9E1.8DD0E36D@bigfoot.de> <Xvsn6.7$R4.940@news.clarkson.edu> <3A9E5AD7.FAB962F5@bton.ac.uk>
Message-ID: <97m2kl$7hu1@news.cis.okstate.edu>

On Thu, 01 Mar 2001 14:21:11 +0000, John English <je at bton.ac.uk> wrote:
>Brad Clements wrote:
>> 
>> Plugging in a "non-verifiable" script plug-in kinda defeats this purpose..
>> doesn't it?
>
>You're asking for a pure Ada implementation of Python, then. 

What's the point of being theoritically verifiable if you can't actually
do it? If I'm not mistaken, Python is large enough to make verification
of even an Ada or SPARK version unfeasible. 

In any case, I can't think of where you would have a verified system
that you would want to give the users the ability to execute arbitrary
code on. A Python script plug-in is nice for things like the Gimp, or
a spreadsheet, or a word-processor; stuff where the worst case scenario
is annoyance, not anything that would get verified.

-- 
David Starner - dstarner98 at aasaa.ofe.org
Pointless website: http://dvdeug.dhis.org
"I don't care if Bill personally has my name and reads my email and 
laughs at me. In fact, I'd be rather honored." - Joseph_Greg


From sholden at holdenweb.com  Mon Mar 19 18:05:02 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Mon, 19 Mar 2001 18:05:02 -0500
Subject: list.sort(cmpfunc) question
References: <m1ae6lg5qt.fsf@pclabs188.labs.trema.com> <98t9dq$qvu$1@panix6.panix.com> <98tf9d$g3v$1@news.mathworks.com>
Message-ID: <Tjwt6.115997$ML1.5401185@e420r-atl2.usenetserver.com>

"Joshua Marshall" <jmarshal at mathworks.com> wrote in message
news:98tf9d$g3v$1 at news.mathworks.com...
[ ... ]
> Seems the word "should" should never appear in documentation.

... and '"should" should' should _definitely_ never appear in documentation!

regards
 Steve





From winson at mdsserv.mds.com.tw  Fri Mar 30 03:23:19 2001
From: winson at mdsserv.mds.com.tw (=?big5?B?sWm4zq9x?=)
Date: Fri, 30 Mar 2001 16:23:19 +0800
Subject: Jython and Win2000
Message-ID: <004201c0b8f2$b04ce640$340210ac@winson>

Hi all,

Maybe post to wrong place. But help me...

My win2000 cannot work with Jython. When I type jython under c>, sometimes there nothing happened and response me a c>. Why ?? I installed Sun's Java 1.3 SDK already.

--
Best Regards,

Winson Chang

3F, 2, 150 Lane, Sec. 5, Hsin I Rd., Taipei, Taiwan, R.O.C.

Tel: +886-2-27225333 ext 376
Fax: +886-2-27222330

Email: winson at mdsserv.mds.com.tw
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20010330/95a688f5/attachment.html>

From sholden at holdenweb.com  Fri Mar  9 09:39:57 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Fri, 9 Mar 2001 09:39:57 -0500
Subject: gah! I hate the new string syntax
References: <mailman.983615170.23313.python-list@python.org> <tN5o6.2746$eH3.14913@NewsReader> <Koco6.7518$1D5.244602@e420r-atl1.usenetserver.com> <mailman.983997252.5628.python-list@python.org> <meAp6.29961$ML1.1514835@e420r-atl2.usenetserver.com> <mailman.984014353.5058.python-list@python.org> <BcCp6.37817$1D5.1582482@e420r-atl1.usenetserver.com> <98amlj01sp6@news2.newsguy.com>
Message-ID: <M_5q6.45279$1D5.2048753@e420r-atl1.usenetserver.com>

"Alex Martelli" <aleaxit at yahoo.com> wrote in message
news:98amlj01sp6 at news2.newsguy.com...
> "Steve Holden" <sholden at holdenweb.com> wrote in message
> news:BcCp6.37817$1D5.1582482 at e420r-atl1.usenetserver.com...
>     [snip]
> > > | > What about one who is both English and Scots, yet never set foot
on
> > > | > that island ;-)?
>     [snip]
> > Either you were born in England, or you were born in Scotland, or you
are
> > neither English nor a Scot. Part-either-or-both, maybe, but not fully
>
> Wrong!  You're talking about people whose seafaring skills and
> traditions are (deservedly) legendary.  It's perfectly acceptable
> for either an English or a Scot to be born aboard a ship...!!!
>
OK, you got me!

regards
 Steve





From ws-news at gmx.at  Tue Mar  6 16:04:16 2001
From: ws-news at gmx.at (Werner Schiendl)
Date: Tue, 6 Mar 2001 22:04:16 +0100
Subject: Can anyone recomend a good intoduction to C...
References: <mailman.983902329.13001.python-list@python.org>
Message-ID: <983912934.638039@newsmaster-04.atnet.at>

Hello Laura,

>Hello Werner
>
>you write:
>
> >I further think it is easier to move 'back' from C++ to C, when the need
> >arises, than vice versa. This is simply because you are forced to use a
> >different technique if the one you are used to does not work on that
> >platform.
> >Personally I started with procedural programming and moved to the object
> >oriented technique afterwards. This is why I think it is hard to get used
to
> >objects when you are used to malloc, free, memcpy and friends.
> >
> >Did I miss an important point?
>
>2 points.  First of all, while you may be correct that your first exposure
>to procedural programming has made it hard for you to get used to
>object oriented programming, it is also possible that your problem was that
>C++ is a widely-avialable-but-not-particularily-good programming language.

This is the point to cut in.

1) I think a language is not particularly good or bad at all. Not Python,
not C, not Eiffel or whatever language. Each has advantages and
disadvantages. Using the right language is an art that needs a lot of
experience and knowing a lot of languages is an important issue to get
there.

2) The original post asks for C, probably to add functionality not available
otherwise to Python. Therefore I did not bring up any other language, but
C++ because you have full access to C libraries but it's still more powerful
and from my point of view a better start.
Personally I wish the procedural C stuff had never even existed.

3) I have a heavy background in C because I am working in an industrial
environment and lots of these small devices have nothing else than a C
compiler or they only perform well if C is used. If such a thing comes with
a C++ compiler I consider that a large improvement (and I'm lucky in that
point - this is becoming more common).


>People whose first OO programming language is Smalltalk have, in my
experience
>a far easier time learning OO programming.  I do not know anybody whose
>first OO language was Eiffel, but I suspect they have an easier time as
well.
>I think that the Smalltalk courses which I took at university, which
>were optional, and I only took because I *like* learning languages ended
>up being the most valuable courses I ever took.
>

Never had something to do with Smalltalk or Eiffel. I think those languages
are not very well supported on to many platforms. And this is of course a
point, you need some sort of library or framework to get a GUI application
up and running in a reasonable amount of time.

But because of your emphasis on it and some other sources maybe I will once
take a look at it.

But to come back to your point. Yes, I agree with you that C++ is probably
not the best way to dive into objects. If I had the choice to start with
either Java or C++ I would take the former, even knowing that is not es
widely supported and not always (or often, sadly) not fast enough.

And, just to mention it, I like to learn languages too. This is the exact
reason why I am just working with Python. I cannot use it much for my
business right now (appart from some build tools and such stuff), but who
knows - hope never dies :-)

>So the question becomes, why is it that the original poster decided to
>learn C or C++?  If he has an application which he thinks is suited to
>one or the other, more power to him, though it is hard to make such
>decisions until you have more experience with a host of computer languages.

I agree with that. At least that it is difficult to learn use the right
tool. But C++ is almost always an considerable alternative to plain C. And
if you use it for nothing more than to hold the functions as part of the
data (make them methods).

>Maybe he just has a compiler and wants to play with it.  Maybe he
>has a linux or unix system and wants to find out exactly what it is
>that his system is doing.  But finally, it may be that it is more
>experience with computer languages in general which is what he wants,
>in which case I would say, yes, by all means learn C.  Then go out and
>learn Smalltalk, or Lisp or Haskell.  And make sure you learn an  assembler

I would agree with the completeness, but not with the order. I would take
the most advanced technique first. Simply because it is closest to the human
mind.
I know a lot of newcomers to programming that can cope with objects much
easier than with C's way of doing things, pointers, memory allocation, etc.

>as well on the way, because that too will make your brain go off and work
>in amazing new directions.   This is a keen source of personal pleasure.
>They say that you cannot feel new neural connections being made in your
>brain, but I am not so sure about that.
>
>Laura Creighton
>

Regards

Werner





From vlouie at telusplanet.net  Wed Mar  7 01:57:39 2001
From: vlouie at telusplanet.net (Victor Louie)
Date: Wed, 07 Mar 2001 06:57:39 GMT
Subject: How to use <>?
Message-ID: <DZkp6.18811$pF4.3015652@news1.telusplanet.net>

Hi,

I was just wondering if any knew what <> does....I've got some sample code
and can't figure out what's happening

eg.
x =10
if x <> 10:
 ....

thanks
Victor




From bsass at freenet.edmonton.ab.ca  Fri Mar 23 17:08:18 2001
From: bsass at freenet.edmonton.ab.ca (Bruce Sass)
Date: Fri, 23 Mar 2001 15:08:18 -0700 (MST)
Subject: New PEP: The directive statement
In-Reply-To: <j4hf0kkilb.fsf@informatik.hu-berlin.de>
References: <mailman.985301265.2425.python-list@python.org>
 <j4hf0kkilb.fsf@informatik.hu-berlin.de>
Message-ID: <Pine.LNX.4.33.0103231414360.26538-100000@bms>

On 23 Mar 2001, Martin von Loewis wrote:
> Bruce Sass <bsass at freenet.edmonton.ab.ca> writes:
>
> > For example, I'd rather do...
> >
> > use:
> >     new_math.//
> >     result = new_math_based_function()
> > except NewMathError:
> >     result = old_math_based_function()
> >
> > ...and have the compiler use the appropriate source depending on what
> > the interpreter can handle, rather than do a check of the
> > "sys.version"[1] or try to catch a run-time exception.
>
> And you would supply both new_math_based_function and
> old_math_based_function in the same module, outside of the use: block?
> That is probably difficult to implement, as the compiler *when
> processing new_math_based_function* needs to know what / means.

I'd have figured too costly (timewise), rather than too difficult...
but since it is a compile-time operation (done once, or infrequently,
for most users), maybe the hit would be worth the expense.

In general, have the compiler scan the source to find out what the
program is trying to use, then compile according to that information.
With this specific example, presumably there will be broken, unused,
old_math_based_function-(s) kicking around in the .pyc (removing them
for the .pyo would probably be difficult).

However, it may be necessary to have two pieces in place,  e.g.,

#!/usr/bin/env python
use:
    new_feature
.
.
.
use:
    new_feature
    # new code
except:
    # old code

...but that would really depend on new_feature and the programmer,
with some it may only make sense to do...

#!/usr/bin/env python
use:
    new_feature
except:
    print "Sorry, you need new_feature to run this program!"
    sys.exit(1)

...and if always having declarative use: blocks at the top of the file
would help with compiling, then it should be required.


- Bruce





From MarkH at ActiveState.com  Wed Mar 21 19:22:00 2001
From: MarkH at ActiveState.com (Mark Hammond)
Date: Thu, 22 Mar 2001 00:22:00 GMT
Subject: Little red lines in PythonWin!
References: <mailman.985105228.27509.python-list@python.org> <9OQt6.13652$992.84056@news-server.bigpond.net.au> <u3dc8q9xg.fsf@ctwd0143.fitlinxx.com>
Message-ID: <3AB946F0.3050106@ActiveState.com>

David Bolen wrote:


 > Interestingly enough when I looked at this in response to the first
 > post
 >  (I don't normally use Pythonwin), I noticed that the default  setup
 >  seems to have a tab size of 4 rather than 8.  So Pythonwin was
 > expanding
 >  literal tabs visually to a 4-position stop rather than 8,  which
 > is different from the language definition.

Yes, because I prefer them this way, and it helps keep people honest. 
Anyone who codes assuming that every editor on the planet uses 8 
character tab stops needs a dose of reality.  Every other editor I use 
has 4 character tab stops (if they didn't out of the box they do now).

 > default (with or without the red indicators) would at least present
 > a  visual indication of what the Python parser was going to "see".

I tend to be more concerned with what other editors will see.  At least 
Python will give an immediate error if it is wrong.  If you mix tabs and 
spaces in a way that Python likes, but still looks bad in many editors, 
finding out earlier is better than later IMO.

If you don't like it, simply turn it off.  However, my recomendation is 
to leave it turned on and heed its warnings regardless of the legality 
of the source to Python.

Editors that still insist tabs characters can be 8 characters now have 
about 2 decades worth of catching up to do.  However, I realize this may 
spawn its own "tabs are 8 chars the way God intended" religous thread, 
that I refuse to participate in - mine (and others with a dissenting 
opinion) minds simply can not be changed on this.

Mark.



From phrxy at csv.warwick.ac.uk  Thu Mar  1 07:14:33 2001
From: phrxy at csv.warwick.ac.uk (John J. Lee)
Date: Thu, 1 Mar 2001 12:14:33 +0000 (GMT)
Subject: [Distutils] Re: CPAN functionality for python - requirements
In-Reply-To: <20010228170725.B1781@tummy.com>
Message-ID: <Pine.SOL.4.30.0103011156380.1448-100000@mimosa.csv.warwick.ac.uk>

On Wed, 28 Feb 2001, Sean Reifschneider wrote:

> On Wed, Feb 28, 2001 at 09:59:38AM -0000, Moore, Paul wrote:
[...]
> >The h2xs program
> >----------------
>
> That seems like a distutils tool to me.
>
> >A social structure
> >------------------
>
> Based on my experimenting with Distutils last night, I don't know that this
> will be a problem.  Distutils is totally cool -- I can't imagine going back.
>
> >Interestingly, developers probably don't "expect" to have to include
> >dependency information, and hence many don't - resulting in the problems you
>
> I think we can deal with that in an iterative manner.  First get them to build
> distutils packages, then when it fails to install on some user's machine
> because they don't have foo.py we can educate them on the joys of listing
> third-party module requirements.

I think an iterative approach is a recipe for an archive full of packages
that only very patchily take advantage of all the facilities on offer.
The perl approach of making 'stubs' generation very easy for tests, docs,
etc. is a very good idea -- it encourages you to actually put something
more useful in the stubs (if not immediately, then later on).  It's self
perpetuating: they're always generated and the standard instructions for
uploading stuff say you should include them (presumably), so it becomes
widespread in the archive; people making new packages then see that
everyone makes packages that way, so they'll do the same thing.

I'm not suggesting everyone's about to start writing huge comprehensive
manuals and test suites for everything, but a little encouragement and
standardisation can go a long way, with a relatively small effort on the
part of the developer.  Same for dependencies, even more so probably.

> >still (relatively) new, and so there are LOTS of packages which don't come
> >with a setup.py yet. Often, adding one isn't hard, but it is still to
>
> The biggest win of using distutils is that it makes it easier for the
> developer to run their software on multiple machines.  That selfish
> reason is enough for me.

And, as selfish users of other people's software, you can use that as a
hook to get people to start making documentation and tests at the same
time.  Getting started is often the 'rate limiting step'.

> >Things are looking better with Python 2.1, though. Included with 2.1, it
> >looks like there will be a standard unit testing module, and a new "pydoc"
> >package, which will extract documentation from module docstrings. So the
>
> Should be sweet...
[...]

Especially if this and PyUnit, and whatever else, can all be tied together
in Distutils.  But *only* if there are actually some proper docs and good
howto examples for Distutils itself!!  What happened to the effort a while
back on that?  What remains to be done -- a todo list would be useful to
spread the load, wouldn't it?


John




From djc at object-craft.com.au  Thu Mar  8 20:06:18 2001
From: djc at object-craft.com.au (Dave Cole)
Date: 09 Mar 2001 12:06:18 +1100
Subject: class fields vs __getattr__
References: <9898al$j3j$1@news.mathworks.com>
Message-ID: <m3u253sqd1.fsf@vole.object-craft.com.au>

>>>>> "Joshua" == Joshua Marshall <jmarshal at mathworks.com> writes:

Joshua> If you try to get a field of an object, it seems __getattr__
Joshua> is called only if no such field exists on the class.  For
Joshua> exmaple:

Joshua>  >>> class T:
Joshua>  ...   x = 1
Joshua>  ...   def __getattr__(self, name):
Joshua>  ...     return 2
Joshua>  ... 
Joshua>  >>> ob = T() 
Joshua>  >>> ob.foo
Joshua>  2
Joshua>  >>> ob.x
Joshua>  1

Joshua> I would have expected ob.x to return 2.  Can anyone explain
Joshua> the rationale behind this behavior?

__getattr__ is only used when Python cannot find the named attribute.

- Dave

-- 
http://www.object-craft.com.au


From SEND at YOUR.SPAM  Mon Mar 12 19:12:48 2001
From: SEND at YOUR.SPAM (Dave Simons)
Date: Tue, 13 Mar 2001 01:12:48 +0100
Subject: Einstein's Riddle
References: <oNPq6.396644$ge4.138479566@news2.rdc2.tx.home.com> <984380185.768920@news.aei.ca> <uP%q6.2375$54.2207@www.newsranger.com> <4w6r6.193645$df5.5192565@news1.crdva1.bc.home.com>
Message-ID: <98jofp$2jdr$1@norfair.nerim.net>

David Murphy wrote:

I'm certain that if Albert had wanted to lead people up the garden path, he 
could have devised a much more subtle way than "na na na na na - I never 
said there was any fish!"

Presupposing, of course, that it realy does come from Einstein.

I used to buy a monthly French magazine called Logica that had nothing but 
this exact type of puzzle from cover to cover. I can tell you they start to 
get boring after doing about two hundred of them. I remember learning of 
the magazine's existence from a poetry-writing tramp who wandered into a 
caf? where I was one day. You meet all sorts...

Dave

-- 
I think, therefore I am ... or at least there are some people who think I 
am.


From kbk at shore.net  Wed Mar  7 12:22:31 2001
From: kbk at shore.net (Kurt B. Kaiser)
Date: Wed, 07 Mar 2001 17:22:31 GMT
Subject: Can anyone recomend a good intoduction to C...
References: <3aa366c6_2@nnrp1.news.uk.psi.net>
Message-ID: <3AA66D6A.7021E451@shore.net>

And once you've read K&R, try Kernighan and Pike, "The Practice of
Programming", which has good advice on C style, design, and test.

Harbison and Steele, "C: A Reference Manual", when you need a language
lawyer.

Learning a language is more than absorbing the syntax, you need to
internalize quality examples to gain fluency. Look at at P. J. Plauger,
"The Standard C Library" for many short codes written in the C idiom. 

A most remarkable book: Don Knuth's "Stanford Graphbase", which is an
*extended* example of C programming in the literate programming style.
You can actually sit down and read the code like literature, there's a
gem on almost every page.

In a hurry? K&R followed by browsing Plauger. 

Kurt B. Kaiser


From urner at alumni.princeton.edu  Sat Mar  3 12:16:14 2001
From: urner at alumni.princeton.edu (Kirby Urner)
Date: Sat, 03 Mar 2001 09:16:14 -0800
Subject: Hitting two targets:  OO + group theory (pedagogy)
References: <u2k0atkgkiatdp9ukj3mpd1ej0b648ldsm@4ax.com> <3aa1077a.1110803@nntp.sprynet.com> <0932atgt9v8d8aboq90evoi5ipj5c0sk8s@4ax.com> <3aa11a91.5998558@nntp.sprynet.com>
Message-ID: <ko72atgf5hmg66suefbenp8lvn6vhblajv@4ax.com>

ullrich at math.okstate.edu (David C. Ullrich) wrote:

>Some people would say that the reference to Genesis
>is out of place here... probably you know that some
>people would say that. Might be regarded as just
>generally inappropriate - some other people might
>think that the bit about Abel is misleading, since
>"Abelian" _does_ refer to a man named Abel, but
>not the one in Genesis.

True, but it's a useful mnemonic connected to a 
well-known story.  Making allusions to Biblical 
characters need be no more or less churchy than 
alluding to Greek mythology or to cartoon characters
like Bugs Bunny (or Coyote in Native American lore)
-- just taking advantage of some shared cultural 
heritage.  

Anyway, I learned this mnemonic from a very strait-
laced group theory book (which I've since returned 
to the library), plus saw it mentioned on another 
list frequented by math heads -- sort of assumed 
this clever CAIN/Abel thing was cited routinely 
by those more familiar with the literature.  Agreed 
though: we should make clear who the Abel in Abelian 
was.

>But why not start with a class Group instead
>of special cases? (Would be a class with a
>__mul__ method that does nothing but raise
>an exception "must implement __mul__ in
>a subclass", an Id attribute, etc. I can't think
>of anything offhand that Group would actually
>_do_, but if we're bent on this OO-math thing
>it seems like groups _should_ be descendants
>of a class Group...)
>

In my example, it's the elements of a group which
get instantiated as separate objects, with each 
element inheriting the same meaning for *, whatever
that might be.  The modulus and order of the group
(= totient in this example) are also saved at the 
class level, where all members have access to the
same bytes in memory.  What's unique to each member
is its integer value.

For example:

 >>> a = G(3,20)  # modulus 20, 3 is relatively prime
 >>> b = G(7,20)  # modulus 20, 7 is relatively prime
 >>> a*b          # product
 1
 >>> pow(a,2)     # a^2
 9
 >>> G.totient    # order of group (a & b share access)
 8
 >>> G.n          # modulus of group (ditto)
 20
 >>> a.val        # unique to a
 3
 >>> b.val        # unique to b
 7

This approach parallels my implementation of a 
Vector class and Quaternion class in other curriculum
segments.  Every v "looks within" to find a shared
definition of cross product, dot product or whatever,
such that we write v1.dot(v2).  But for people who
prefer to write dot(v1,v2), we can define a function:

   def dot(v1,v2):
      return v1.dot(v2)

Some might argue that Vector and Quaternion should 
be integrated in some higher level superclass.  Whereas
I do, in fact, subclass Vector into Svector (spherical 
coordinates) and Qvector (quadray coordinates), I don't 
generalize in the other direction.

I will think more about your suggestion as I continue
to develop OO + math curriculum segments.  It could be
that the abstractions re <G,*> inform the thinking of
students such that they imagine special case groups as
subclasses of the generic group class, without needing
to literally code it this way.  In other words, seeing 
a class hierarchy from a given starting point on down, 
helps them think upwards to higher levels of abstraction,
even though the computer language by itself doesn't 
really go there (that's where we stay with a math 
notation -- looking for when it starts to make sense 
to implement more concretely as we descend to the 
applications level).

Kirby



From leo at iems.nwu.edu  Tue Mar 27 15:42:01 2001
From: leo at iems.nwu.edu (Leonardo B Lopes)
Date: Tue, 27 Mar 2001 14:42:01 -0600
Subject: order by instead of two queries
Message-ID: <3AC0FB19.7291F7DF@iems.nwu.edu>

I have some data in a table that I would like to obtain order by a
specific instance of the data. ex:

Name:		Age:

Joe		50
Jill		60
Jack		65

What I would like to do is have a query that returns Jack first, then
the regular name order. Something like:

SELECT Name FROM clients ORDER BY Name='Jack',Name

Can this be done with one query?

Thanks!
Leo.
-- 
=======================================================================
Leonardo B. Lopes                                      leo at iems.nwu.edu 
Ph.D. Student                                             (847)491-8470
IEMS - Northwestern University             http://www.iems.nwu.edu/~leo


From stephen_purcell at yahoo.com  Wed Mar 21 09:40:57 2001
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Wed, 21 Mar 2001 15:40:57 +0100
Subject: [Newbie] line separators in file-output
In-Reply-To: <3AB8AEDA.2D2BD105@bg-sys.de>; from daniel.orlowski@bg-sys.de on Wed, Mar 21, 2001 at 02:38:34PM +0100
References: <3AB8AEDA.2D2BD105@bg-sys.de>
Message-ID: <20010321154057.B26814@freedom.puma-ag.com>

Daniel Orlowski wrote:
> If I in Python uder Windows do smth like this:
> 
> >>> File=open("test","w")
> >>> File.write("a line"+"\n")
> >>> File.close()
> 
> I have data in file test:
> 'a line\015\012'
> 
> How can I prevent output of  '\015'?

Open the file in binary mode:  File = open('test','wb')

Windows automatically changes line endings for files opened in text mode.

-Steve

-- 
Steve Purcell, Pythangelist
Get testing at http://pyunit.sourceforge.net/
Any opinions expressed herein are my own and not necessarily those of Yahoo



From aleaxit at yahoo.com  Mon Mar  5 04:57:03 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Mon, 5 Mar 2001 10:57:03 +0100
Subject: I come to praise .join, not to bury it...
References: <mailman.983756766.27347.python-list@python.org>
Message-ID: <97vnu3076q@news2.newsguy.com>

"Delaney, Timothy" <tdelaney at avaya.com> wrote in message
news:mailman.983756766.27347.python-list at python.org...
    [snip]
> I truly have no complaints with the technical merits of ''.join().

Good.

> However, I have *severe* complaints with the aesthetics and consistency.
> Quite simply, the way we look at things tends to be
>
> "result" = "thing to be acted on to get the result" modified by "thing
which
> does the action"

This is bass-ackwards.  The 'thing which does the action', i.e. the
object which will house the method code, is the one that NEEDS to come
first, because of how dispatching is coded in the Python language!


> This is consistently followed in the standard python libraries in almost
all

We seem to be using two VERY different languages/libraries - funny
that they're both named 'Python', innit.  For example, in the Python
_I_ use, the re module defines objects (compiled regular expressions)
which are the 'things which do the action', and, sure enough, they
have methods which are passed the 'thing to be acted upon' as their
argument -- result = thing_doing_the_action.method(thing_acted_on).

Similarly, the pickle module defines objects (pickler objects) whose
methods are passed 'things to be acted upon' (e.g., a list to be
pickled can be passed to method dumps of the pickler, which returns
the result, a string).

And again, the urllib module defines object (URLopener objects)
whose methods are passed 'things to be acted upon' (e.g., open
is given a fullurl and optionally data) to return a result (here,
a file-like object).

And so on, and so forth.  It's really very consistent -- it HAS
to be, because it's mostly dictated by technical considerations
of 'which object do we need full polymorphism on, which one is
best handled through a standard interface instead'.

> cases ... but ''.join() is a glaring inconsistency. As a result, it seems
> unintuitive and ugly.

What inconsistency?!  It's exactly like the other cases -- the
object which does the action (here, a string-like object, who
works as a joiner) defines a method which is passed as argument
the object to be acted upon.


Alex





From ixokai-nf at myNOSPAM.org  Thu Mar 29 15:45:26 2001
From: ixokai-nf at myNOSPAM.org (Ixokai)
Date: Thu, 29 Mar 2001 12:45:26 -0800
Subject: Looking for Python equivalent to a Perl'ism
References: <3AC375A7.A4BA87F1@spammraqia.com>
Message-ID: <3ac3a14a$1_2@goliath.newsfeeds.com>

While this is not exactly what you want, were you aware of the following
syntax:

dict = {"foo":"yechfoo", "mumble":"yukkomumble"}

print "This is %(foo)s this is junk %(mumble)s and more stuff" % dict

--Stephen
(replace 'NOSPAM' with 'seraph' to respond)

David Lees wrote in message <3AC375A7.A4BA87F1 at spammraqia.com>...
>
>I am trying to figure out how to do the Python equivalent of the
>following perl code, which finds all the groups tagged by 'xx' and
>replaces them with text looked up in a dictionary:
>----------
>my %dict;
>my $string="this is xxfooxx this is junk xxmumblexx and more stuff";
>%dict=(foo=>"yechfoo",mumble=>"yukkomumble");
>
>$string =~ s/xx(\S+)xx/$dict{$1}/g;
>
>print $string;
>---------------------
>
>My Python equivalent without the backreference dictionary lookup is:
>
>dict ={'xxx':'foo',
>       'yyy':'barf',
>       }
>p=re.compile(r'zz(\w+)zz')
>s='this is and so is zzxxxzz abc that an zzyyyzz abc zzxxxzz or else'
>print re.sub(p,'junk',s)
>
>--------------
>
>But I do not understand how to include backreferences into the regex
>substitution as is conveniently done in perl.  Can someone tell me if
>there is an easy Python solution to this?
>
>Thanks in advance.
>
>David Lees




-----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
-----==  Over 80,000 Newsgroups - 16 Different Servers! =-----


From mwh21 at cam.ac.uk  Thu Mar 29 17:22:05 2001
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 29 Mar 2001 23:22:05 +0100
Subject: KeyBoardInterrupt error
References: <mailman.985802077.12779.python-list@python.org>
Message-ID: <m3ae64cj1u.fsf@atrus.jesus.cam.ac.uk>

Headers indicate evolution, so Linux, right?

Jason "Straw." <jstraw at towhee.com> writes:

> What can trigger a Keyboard Interrupt?

Receiving a SIGINT.  The terminal driver will send one of those if you
hit Ctrl-C (by default).

> I need Ctrl-C to be ignored and Ctrl-D to map to a function.

Two options for ignoring Ctrl-C: use termios to tuen off the ISIG
flag, along the lines of

import termios
import TERMIOS
n = termios.tcgetattr(0)
n[3] &=~ TERMIOS.ISIG
termios.tcsetattr(0, TERMIOS.TCSANOW, n)

but with some kind of error checking & stuff or use signal to ignore
SIGINTs:

signal.signal(signal.SIGINT,signal.SIG_IGN)

Note that when you're playing around at the console that readline
tends to reset both of these, which is nice, and that both of these
will leave ^C in the input stream (it seems).

Ctrl-D just causes reads to abort early.  If you're using raw_input()
(a) don't, if you want this level of control (b) just catch EOFErrors
and restart your loop.

> gettting a single char at a time is a bad idea for my project

Eh?  How does this relate to the above?

If you're interested in line input, you *might* want to look at my
python line reader, which was tentatively dubbed "pyrl", but needs a
new name ("pyrl" is taken).  It's not really finished; find it at:

  http://www-jcsu.jesus.cam.ac.uk/~mwh21/hacks/pyrl-0.2.0.tar.gz

To test drive, unpack & go:

$ python python_reader.py

(requires Python 2).

I've almost got incremental history searching done on my hard drive,
but a few bugs and some additional commitments have stalled me...
Anybody interested in seeing more, or have a better name than
pyttyinput?

Cheers,
M.

-- 
  I'll write on my monitor fifty times 'I must not post self-indulgent
  wibble nobody is interested in to ucam.chat just because I'm bored
  and I can't find the bug I'm supposed to fix'.
                                            -- Steve Kitson, ucam.chat


From guido at digicool.com  Mon Mar 12 18:41:01 2001
From: guido at digicool.com (Guido van Rossum)
Date: Mon, 12 Mar 2001 18:41:01 -0500
Subject: [Python-Dev] Revive the types sig?
In-Reply-To: Your message of "Mon, 12 Mar 2001 18:15:15 EST."
             <15021.22659.616556.298360@anthem.wooz.org> 
References: <jeremy@alum.mit.edu> <15020.9404.557943.164934@w221.z064000254.bwi-md.dsl.cnc.net> <200103120711.AAA09711@localhost.localdomain>  
            <15021.22659.616556.298360@anthem.wooz.org> 
Message-ID: <200103122341.SAA23054@cj20424-a.reston1.va.home.com>

> >>>>> "UO" == Uche Ogbuji <uche.ogbuji at fourthought.com> writes:
> 
>     UO> I know this isn't the types SIG and all, but since it has come
>     UO> up here, I'd like to (once again) express my violent
>     UO> disagreement with the efforts to add static typing to Python.
>     UO> After this, I won't pursue the thread further here.
> 
> Thank you Uche!  I couldn't agree more, and will also try to follow
> your example, at least until we see much more concrete proposals from
> the types-sig.  I just want to make a few comments for the record.

Barry, you were supposed to throw a brick at me with this content at
the meeting, on Eric's behalf.  Why didn't you?  I was waiting for
someone to explain why this was a big idea, but everybody kept their
face shut!  :-(

> First, it seemed to me that the greatest push for static type
> annotations at IPC9 was from the folks implementing Python on top of
> frameworks other than C.  I know from my own experiences that there is
> the allure of improved performance, e.g. JPython, given type hints
> available to the compiler.  While perhaps a laudable goal, this
> doesn't seem to be a stated top priority of Paul's.
> 
> Second, if type annotations are to be seriously considered for
> inclusion in Python, I think we as a community need considerable
> experience with a working implementation.  Yes, we need PEPs and specs
> and such, but we need something real and complete that we can play
> with, /without/ having to commit to its acceptance in mainstream
> Python.  Therefore, I think it'll be very important for type
> annotation proponents to figure out a way to allow people to see and
> play with an implementation in an experimental way.

+1

> This might mean an extensive set of patches, a la Stackless.  After
> seeing and talking to Neil and Andrew about PTL and Quixote, I think
> there might be another way.  It seems that their approach might serve
> as a framework for experimental Python syntaxes with minimal overhead.
> If I understand their work correctly, they have their own compiler
> which is built on Jeremy's tools, and which accepts a modified Python
> grammar, generating different but compatible bytecode sequences.
> E.g., their syntax has a "template" keyword approximately equivalent
> to "def" and they do something different with bare strings left on the
> stack.

I'm not sure this is viable.  I believe Jeremy's compiler package
actually doesn't have its own parser -- it uses the parser module
(which invokes Python's standard parse) and then transmogrifies the
parse tree into something more usable, but it doesn't change the
syntax!  Quixote can get away with this because their only change
is giving a different meaning to stand-alone string literals.  But for
type annotations this doesn't give enough freedom, I expect.

> The key trick is that it all hooks together with an import hook so
> normal Python code doesn't need to know anything about the mechanics
> of PTL compilation.  Given a homepage.ptl file, they just do an
> "import homepage" and this gets magically transformed into a .ptlc
> file and normal Python objects.

That would be nice, indeed.

> If I've got this correct, it seems like it would be a powerful tool
> for playing with alternative Python syntaxes.  Ideally, the same
> technique would allow the types-sig folks to create a working
> implementation that would require only the installation of an import
> hook.  This would let them build their systems with type annotation
> and prove to the skeptical among us of their overwhelming benefit.

+1

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



From sholden at holdenweb.com  Mon Mar  5 03:06:28 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Mon, 5 Mar 2001 03:06:28 -0500
Subject: ftp a list to a remote server
References: <tCHo6.8449$v5.32212@newsfeeds.bigpond.com>
Message-ID: <VRHo6.19945$1D5.757100@e420r-atl1.usenetserver.com>

"Austin Wilson" <austinw at bhpvillage.com> wrote in message
news:tCHo6.8449$v5.32212 at newsfeeds.bigpond.com...
> Hi
>
> I would like to save a list as a text file on a remote server using ftp.
>
> Is this possible? If so, could you please provide some sample code.
>
> Or do I have to save the file locally first then ftp that file to the
remote
> server?
>
> Thanks
> Austin
>
>
Take a look at ftpStream in

    http://www.holdenweb.com/Python/index.html

This allows you to read from and write to files on remote FTP servers.

regards
 Steve





From dnew at san.rr.com  Tue Mar 13 18:21:52 2001
From: dnew at san.rr.com (Darren New)
Date: Tue, 13 Mar 2001 23:21:52 GMT
Subject: This math scares me
References: <3aad4062.1290021@News.CIS.DFN.DE> <20010312170356.B3513@better.net> <mailman.984480856.18097.python-list@python.org> <984486395.407379@newsmaster-04.atnet.at> <slrn9asf1d.ja3.qrczak@qrnik.zagroda> <984518199.127140@newsmaster-04.atnet.at> <jwbnews-09D75F.13313313032001@corp.supernews.com> <yTwr6.1235$Tg.167429@ruti.visi.com>
Message-ID: <3AAEAB92.DC8E984A@san.rr.com>

Grant Edwards wrote:
> Of course we'd never imagined we'd ever see hardware FP and BCD
> libraries were only marginally slower than binary ones. Now
> that FP is usually in hardware, BCD FP (or fixed point) will be
> pretty slow by comparison.

One mainframe I worked on (Xerox 560, AKA Sigma 9) had the "scientific
module" which was hardware floating point, and the "business module" which
was the hardware BCD. The latter added instructions like "Edit byte string"
that would copy packed BCD into a string according to COBOL display-edit
formatting. (Or whatever COBOL called the "print using" formatting. Too many
years. Not enough COBOL. Or something like that.)

-- 
Darren New / Senior MTS & Free Radical / Invisible Worlds Inc.
San Diego, CA, USA (PST).  Cryptokeys on demand.


From ben.hutchings at roundpoint.com  Tue Mar  6 19:31:53 2001
From: ben.hutchings at roundpoint.com (Ben Hutchings)
Date: 06 Mar 2001 16:31:53 -0800
Subject: Remote script execution with Python ...
References: <7Ybp6.7729$r4.7336@news.indigo.ie>
Message-ID: <ulmqil8qu.fsf@roundpoint.com>

"Robert Gahan" <gahanr at gofree.indigo.ie> writes:

> Hi there,
> 
> I have installed python on a Windows PC and I am looking to
> remotely execute a script on a UNIX box from a python program.
> I need to be able to pipe a "y" to stdin also. Is there a windows
> equivalent of the unix "rexec" that could help me ?. What is the
> most appropriate way of doing this ?.

You might be able to do this with something like:

    script_in, script_out = os.popen2("rsh otherhost -c myscript")
    print >>script_in 'y'
    result = script_out.readlines()

Windows 2000 includes an rsh client but I'm not sure about other
versions of Windows.  You might want to use ssh instead for greater
security.

-- 
Any opinions expressed are my own and not necessarily those of Roundpoint.


From dnew at san.rr.com  Wed Mar 14 13:37:16 2001
From: dnew at san.rr.com (Darren New)
Date: Wed, 14 Mar 2001 18:37:16 GMT
Subject: Who's minister of propaganda this week?
References: <B98041329020E007.7DA84CAD9D0562AB.817BBC572F480869@lp.airnews.net> <98nu1m$12e@dispatch.concentric.net> <660ACCA1E4816F7E.C359CDF4168EF3FB.3B6B05538D619770@lp.airnews.net> <3AAFA13D.76E32C3F@divalsim.it> <98oa4o01i4v@news2.newsguy.com>
Message-ID: <3AAFBA5D.63AAF897@san.rr.com>

Alex Martelli wrote:
> So, all the compile-time checking is buying is catching (a small
> subset of) the errors that would be caught in testing anyway, a
> little bit earlier (thus, a little bit cheaper) 

Actually, if you do the XP thing, you write the tests before the code, so
you're actually catching the type error before you even have something that
compiles. ;-)

-- 
Darren New / Senior MTS & Free Radical / Invisible Worlds Inc.
San Diego, CA, USA (PST).  Cryptokeys on demand.


From grante at visi.com  Sun Mar 11 23:50:12 2001
From: grante at visi.com (Grant Edwards)
Date: Mon, 12 Mar 2001 04:50:12 GMT
Subject: Poll'able Queue objects.
References: <txkr904gwu9.fsf@vortex.ukshells.co.uk> <cuRq6.240807$o91.31035696@typhoon.nyroc.rr.com>
Message-ID: <slrn9aolc9.sci.grante@isis.visi.com>

In article <cuRq6.240807$o91.31035696 at typhoon.nyroc.rr.com>, Dan Maas wrote:

>> As this didn't seem to exist i have coded a selectable Queue object based
>> on pipes (which should thus work under unix and win32).  I have included
>> the code below.  If you go to the following URL :
>
>I hate to say it, but this code won't work on Win32 since you can't select()
>or WaitForMultipleObjects() on anonymous pipes (thanks, MS!).

The mind wobbles.  One of the things that Unix got right is that you can
select() (or poll()) on pipes, sockets, serial ports (any properly written
character device that's properly written), or, in degenerate-case sort of
way, "regular" files.  The fact that MS could take such a giant step
backwards demonstrates either a monumental lack of knowledge about OS's or a
monumental amount of conceit.  Probably a combination of both.

It's really sad and shocking.

-- 
Grant Edwards                   grante             Yow!  Do you think the
                                  at               "Monkees" should get gas on
                               visi.com            odd or even days?


From hzhu at mars.localdomain  Tue Mar 20 19:16:56 2001
From: hzhu at mars.localdomain (Huaiyu Zhu)
Date: Wed, 21 Mar 2001 00:16:56 GMT
Subject: comp.lang.python.numeric?  (was Re: Numeric PEPs, first public posts)
References: <99669h$so4$1@panix6.panix.com>, <slrn9bcp3h.ns2.qrczak@qrnik.zagroda> <mailman.984748941.31248.python-list@python.org> <Q5Ns6.432845$ge4.152574719@news2.rdc2.tx.home.com> <mailman.985035504.26190.python-list@python.org> <mailman.985080535.7447.python-list@python.org>
Message-ID: <slrn9bfsa6.rv.hzhu@mars.localdomain>

On Tue, 20 Mar 2001 11:24:49 +0200, Moshe Zadka <moshez at zadka.site.co.il> wrote:
>On 19 Mar 2001 15:55:29 -0800, aahz at panix.com (Aahz Maruch) wrote:
> 
>> Because a lot of people don't have time to subscribe to every mailing
>> list that comes along.
>
>I respectfully disagree. The numerics sig has managed to get many people
>that probably don't read c.l.py subscribed to it. 

Different peopel have different preferences.  Some like mail clients, some
like news clients, some even prefer web access.  Any one choice will make
others unhappy.

Since there have been several numerical discussions popping up here and
there, does anybody else think it's a good idea to create a news group

	comp.lang.python.numeric?

It could be gatewayed such that posting to any of the numerical mailing
lists would be submitted to the news group as well.

I do not have the expertise to manage a call for vote and other necessary
things about creating a news group.  Here is a list several advantages that
might be included in a CfV if someone does it.

1. There exists a substantial need for dicussion of numerical computing in
   Python.  This is evident from the existence of several specialized
   discussion forums.

2. A news group offers a unique type of venue that is unmatched by news
   groups in many ways.  Many potential discussants would use news group if
   if they do not use other forums.  Furthermore, it can be used in
   conjunction with others.

3. Numerical related discussions pop up in c.l.p quite often.  But the
   content is often quite specialized that it is not of general interest to
   majority of readers here at any given moment.  The discussions are also
   often very long as well.

On second thought, maybe we could broaden the scope a bit, say
comp.lang.python.math or even comp.lang.python.sci. This would allow
discussions about symbolic algebra as well.

Anybody having the energy and know-how to do this?

Huaiyu


From gervaserules at yahoo.com  Tue Mar 13 21:00:45 2001
From: gervaserules at yahoo.com (Nathan Heagy)
Date: Wed, 14 Mar 2001 02:00:45 GMT
Subject: Console column Width
Message-ID: <hhAr6.3889$L7.27822@news1.sshe1.sk.home.com>

I can't figure out how to learn the (linux) console's column width from
within a script? That is, without using curses.

I thought it would be so simple that an included module would already be
able to do it. I can't find one. I resorted to looking in the ls source and
found that a call to ioctl is how ls does it. However trying to use python's
fnctl.ioctl() on sys.stdout gives me an error. Even if it did work, I'm not
sure how to parse the results!

Any help here is greatly appreciated, especially replies be email : )

Thanks,

nathan




From barry at digicool.com  Wed Mar 21 12:40:21 2001
From: barry at digicool.com (Barry A. Warsaw)
Date: Wed, 21 Mar 2001 12:40:21 -0500
Subject: PEP 1, PEP Purpose and Guidelines
Message-ID: <15032.59269.4520.961715@anthem.wooz.org>

With everyone feeling so PEPpy lately (yay!) I thought it was time to
do an updating pass through PEP 1.  Attached below is the latest copy,
also available (as soon as uploading is complete) via

    http://python.sourceforge.net/peps/pep-0001.html

Note the addition of the Replaces: and Replaced-By: headers for
formalizing the PEP replacement policy (thanks to Andrew Kuchling for
the idea and patch).

Enjoy,
-Barry

-------------------- snip snip --------------------
PEP: 1
Title: PEP Purpose and Guidelines
Version: $Revision: 1.16 $
Author: barry at digicool.com (Barry A. Warsaw),
    jeremy at digicool.com (Jeremy Hylton)
Status: Draft
Type: Informational
Created: 13-Jun-2000
Post-History: 21-Mar-2001


What is a PEP?

    PEP stands for Python Enhancement Proposal.  A PEP is a design
    document providing information to the Python community, or
    describing a new feature for Python.  The PEP should provide a
    concise technical specification of the feature and a rationale for
    the feature.

    We intend PEPs to be the primary mechanisms for proposing new
    features, for collecting community input on an issue, and for
    documenting the design decisions that have gone into Python.  The
    PEP author is responsible for building consensus within the
    community and documenting dissenting opinions.

    Because the PEPs are maintained as plain text files under CVS
    control, their revision history is the historical record of the
    feature proposal[1].
    

Kinds of PEPs

    There are two kinds of PEPs.  A standards track PEP describes a
    new feature or implementation for Python.  An informational PEP
    describes a Python design issue, or provides general guidelines or
    information to the Python community, but does not propose a new
    feature.


PEP Work Flow

    The PEP editor, Barry Warsaw <barry at digicool.com>, assigns numbers
    for each PEP and changes its status.

    The PEP process begins with a new idea for Python.  Each PEP must
    have a champion -- someone who writes the PEP using the style and
    format described below, shepherds the discussions in the
    appropriate forums, and attempts to build community consensus
    around the idea.  The PEP champion (a.k.a. Author) should first
    attempt to ascertain whether the idea is PEP-able.  Small
    enhancements or patches often don't need a PEP and can be injected
    into the Python development work flow with a patch submission to
    the SourceForge patch manager[2] or feature request tracker[3].

    The PEP champion then emails the PEP editor with a proposed title
    and a rough, but fleshed out, draft of the PEP.  This draft must
    be written in PEP style as described below.

    If the PEP editor approves, he will assign the PEP a number, label
    it as standards track or informational, give it status 'draft',
    and create and check-in the initial draft of the PEP.  The PEP
    editor will not unreasonably deny a PEP.  Reasons for denying PEP
    status include duplication of effort, being technically unsound,
    or not in keeping with the Python philosophy.  The BDFL
    (Benevolent Dictator for Life, Guido van Rossum
    <guido at python.org>) can be consulted during the approval phase,
    and is the final arbitrator of the draft's PEP-ability.

    The author of the PEP is then responsible for posting the PEP to
    the community forums, and marshaling community support for it.  As
    updates are necessary, the PEP author can check in new versions if
    they have CVS commit permissions, or can email new PEP versions to
    the PEP editor for committing.

    Standards track PEPs consists of two parts, a design document and
    a reference implementation.  The PEP should be reviewed and
    accepted before a reference implementation is begun, unless a
    reference implementation will aid people in studying the PEP.
    Standards Track PEPs must include an implementation - in the form
    of code, patch, or URL to same - before it can be considered
    Final.

    PEP authors are responsible for collecting community feedback on a
    PEP before submitting it for review.  A PEP that has not been
    discussed on python-list at python.org and/or python-dev at python.org
    will not be accepted.  However, wherever possible, long open-ended
    discussions on public mailing lists should be avoided.  A better
    strategy is to encourage public feedback directly to the PEP
    author, who collects and integrates the comments back into the
    PEP.

    Once the authors have completed a PEP, they must inform the PEP
    editor that it is ready for review.  PEPs are reviewed by the BDFL
    and his chosen consultants, who may accept or reject a PEP or send
    it back to the author(s) for revision.

    Once a PEP has been accepted, the reference implementation must be
    completed.  When the reference implementation is complete and
    accepted by the BDFL, the status will be changed to `Final.'

    A PEP can also be assigned status `Deferred.'  The PEP author or
    editor can assign the PEP this status when no progress is being
    made on the PEP.  Once a PEP is deferred, the PEP editor can
    re-assign it to draft status.

    A PEP can also be `Rejected'.  Perhaps after all is said and done
    it was not a good idea.  It is still important to have a record of
    this fact.

    PEPs can also be replaced by a different PEP, rendering the
    original obsolete.  This is intended for Informational PEPs, where
    version 2 of an API can replace version 1.

    PEP work flow is as follows:

        Draft -> Accepted -> Final -> Replaced
          ^
          +----> Rejected
          v
        Deferred

    Some informational PEPs may also have a status of `Active' if they
    are never meant to be completed.  E.g. PEP 1.


What belongs in a successful PEP?

    Each PEP should have the following parts:

    1. Preamble -- RFC822 style headers containing meta-data about the
       PEP, including the PEP number, a short descriptive title, the
       names contact info for each author, etc.

    2. Abstract -- a short (~200 word) description of the technical
       issue being addressed.

    3. Copyright/public domain -- Each PEP must either be explicitly
       labelled in the public domain or the Open Publication
       License[4].

    4. Specification -- The technical specification should describe
       the syntax and semantics of any new language feature.  The
       specification should be detailed enough to allow competing,
       interoperable implementations for any of the current Python
       platforms (CPython, JPython, Python .NET).

    5. Rationale -- The rationale fleshes out the specification by
       describing what motivated the design and why particular design
       decisions were made.  It should describe alternate designs that
       were considered and related work, e.g. how the feature is
       supported in other languages.

       The rationale should provide evidence of consensus within the
       community and discuss important objections or concerns raised
       during discussion.

    6. Reference Implementation -- The reference implementation must
       be completed before any PEP is given status 'Final,' but it
       need not be completed before the PEP is accepted.  It is better
       to finish the specification and rationale first and reach
       consensus on it before writing code.

       The final implementation must include test code and
       documentation appropriate for either the Python language
       reference or the standard library reference.


PEP Style

    PEPs are written in plain ASCII text, and should adhere to a
    rigid style.  There is a Python script that parses this style and
    converts the plain text PEP to HTML for viewing on the web[5].

    Each PEP must begin with an RFC822 style header preamble.  The
    headers must appear in the following order.  Headers marked with
    `*' are optional and are described below.  All other headers are
    required.

        PEP: <pep number>
        Title: <pep title>
        Version: <cvs version string>
        Author: <list of authors' email and real name>
      * Discussions-To: <email address>
        Status: <Draft | Active | Accepted | Deferred | Final | Replaced>
        Type: <Informational | Standards Track>
        Created: <date created on, in dd-mmm-yyyy format>
      * Python-Version: <version number>
        Post-History: <dates of postings to python-list and python-dev>
      * Replaces: <pep number>
      * Replaced-By: <pep number>

    Standards track PEPs must have a Python-Version: header which
    indicates the version of Python that the feature will be released
    with.  Informational PEPs do not need a Python-Version: header.

    While a PEP is in private discussions (usually during the initial
    Draft phase), a Discussions-To: header will indicate the mailing
    list or URL where the PEP is being discussed.  No Discussions-To:
    header is necessary if the PEP is being discussed privately with
    the author, or on the python-list or python-dev email mailing
    lists.

    PEPs may also have a Replaced-By: header indicating that a PEP has
    been rendered obsolete by a later document; the value is the
    number of the PEP that replaces the current document.  The newer
    PEP must have a Replaces: header containing the number of the PEP
    that it rendered obsolete.

    PEP headings must begin in column zero and the initial letter of
    each word must be capitalized as in book titles.  Acronyms should
    be in all capitals.  The body of each section must be indented 4
    spaces.  Code samples inside body sections should be indented a
    further 4 spaces, and other indentation can be used as required to
    make the text readable.  You must use two blank lines between the
    last line of a section's body and the next section heading.

    Tab characters must never appear in the document at all.  A PEP
    should include the Emacs stanza included by example in this PEP.

    A PEP must contain a Copyright section, and it is strongly
    recommended to put the PEP in the public domain.

    You should footnote any URLs in the body of the PEP, and a PEP
    should include a References section with those URLs expanded.


References and Footnotes

    [1] This historical record is available by the normal CVS commands
    for retrieving older revisions.  For those without direct access
    to the CVS tree, you can browse the current and past PEP revisions
    via the SourceForge web site at

    http://cvs.sourceforge.net/cgi-bin/cvsweb.cgi/python/nondist/peps/?cvsroot=python

    [2] http://sourceforge.net/tracker/?group_id=5470&atid=305470

    [3] http://sourceforge.net/tracker/?atid=355470&group_id=5470&func=browse

    [4] http://www.opencontent.org/openpub/

    [5] The script referred to here is pep2html.py, which lives in
    the same directory in the CVS tree as the PEPs themselves.  Try
    "pep2html.py --help" for details.

    The URL for viewing PEPs on the web is
    http://python.sourceforge.net/peps/


Copyright

    This document has been placed in the public domain.



Local Variables:
mode: indented-text
indent-tabs-mode: nil
End:



From skip at pobox.com  Wed Mar 28 15:03:57 2001
From: skip at pobox.com (Skip Montanaro)
Date: Wed, 28 Mar 2001 14:03:57 -0600 (CST)
Subject: Is inheritance broken?
In-Reply-To: <NDBBLAKACLGJJBDNGGCMOEJBCCAA.jcc.ugm@ix.netcom.com>
References: <15042.11167.848530.564353@beluga.mojam.com>
	<NDBBLAKACLGJJBDNGGCMOEJBCCAA.jcc.ugm@ix.netcom.com>
Message-ID: <15042.17325.125458.471763@beluga.mojam.com>

    Jonathan> Ugly and error prone, no?

Well, perhaps there is a better way...  ;-)

How about creating mixins, then composing a set of classes from them:

    class VaryingAnglesMixin:
	def angle(self): return "variable"

    class SameAnglesMixin:
	def angle(self): return "same"

    class NinetyDegreesMixin:
	def angle(self): return "90 degree"

    class VaryingSidesMixin:
	def side(self): return "variable"

    class SameSidesMixin:
	def side(self): return "same"

    class parallelogram(VaryingAnglesMixin, VaryingSidesMixin):
	pass

    class rhombus(VaryingAnglesMixin, SameSidesMixin):
	pass

    class rectangle(NinetyDegreesMixin, VaryingSidesMixin):
	pass

    class square(NinetyDegreesMixin, SameSidesMixin):
	pass

    s = square()
    print "squares have", s.angle(), "angles"
    print "squares have", s.side(), "sides"

    r = rhombus()
    print "rhombuses have", r.angle(), "angles"
    print "rhombuses have", r.side(), "sides"

Skip



From wware at world.std.com  Tue Mar  6 19:27:04 2001
From: wware at world.std.com (Will Ware)
Date: Wed, 7 Mar 2001 00:27:04 GMT
Subject: Open Source Organic Chemistry software project
References: <taaseuid59k047@corp.supernews.com> <983tj3$fio$1@nntp9.atl.mindspring.net>
Message-ID: <G9sx94.5uE@world.std.com>

"mapMonkey" <jziniti at speakeasy.org> wrote:
> The current code-base is geared to Organic Chemistry molecules
> and the IUPAC nomenclature system and it works quite well.  It
> was written entirely in Python, and entirely by me...

Greg Landrum (gReGlAnDrUm at earthlink.net) wrote:
> I'll express provisional interest...

Likewise. There is in fact a long and venerable history of
computational chemistry in the Python world, most notably the
MMTK molecular modeling package:
http://starship.python.net/crew/hinsen/MMTK/

Also of some interest, but currently inactive to the best of
my knowledge, is the OpenChem project:
http://www.openchem.org/

Here's some stuff by Konrad Hinsen:
ftp://www.ccl.net:/pub/chemistry/software/SOURCES/PYTHON/HINSEN/
Konrad was IIRC the original author of MMTK, and an early
contributor to the development of NumPy.

There's also something interesting on SourceForge called PyMOL:
http://pymol.sourceforge.net/
It's a graphics system for making images and animations.

I don't know if any of this stuff will be of any use to your
project, but at least it should give you a comfortable feeling
about mixing Python and computational chemistry.

-- 
-----------------------------------+---------------------
 22nd century: Esperanto, geodesic | Will Ware
 domes, hovercrafts, metric system | wware at world.std.com


From aleaxit at yahoo.com  Sat Mar 17 03:30:07 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sat, 17 Mar 2001 09:30:07 +0100
Subject: Who's minister of propaganda this week?
References: <mailman.984595938.11127.python-list@python.org> <98otuq0nnf@news1.newsguy.com> <slrn9b0cc2.gsa.neelk@alum.mit.edu> <98q6ro2uoa@news2.newsguy.com> <slrn9b2u0r.j74.neelk@alum.mit.edu> <98t8830nlb@news1.newsguy.com> <slrn9b5sbv.lu3.neelk@alum.mit.edu>
Message-ID: <98v77v11qcg@news2.newsguy.com>

"Neelakantan Krishnaswami" <neelk at alum.mit.edu> wrote in message
news:slrn9b5sbv.lu3.neelk at alum.mit.edu...
    [snip]
> >> I actually have a function that walks the class tree checking it
> >> against the dictionary, which solves this problem.  You don't need to
> >> add a special attribute in Python since __class__ gives you what you
> >> need for free.
> >
> > ...except it doesn't, because it's _not_ inherited -- you have
> > to 'manually' walk the directed acyclic graph (tree? what tree?
> > we have _multiple_ inheritance:-) or iterate through the dictionary's
> > keys (assumed to be types) using isinstance for each on your
> > object.
>
> Okay, I don't understand what you're saying at all.
>
> If I call __class__ on an object, I get its class. Now, I can ask
> each class what its superclasses are using the __bases__ attribute on
> each class. This is okay to leave as a function, because Python
> promises to walk the class heterarchy (happy? :) in a depth-first
> left-to-right order.

So far, so good.


> So I have a dictionary like this
>
> >>> d
> {<class Bar at 80ea570>:   2,
>  <class Glarp at 80e9bd8>: 3,
>  <class Foo at 80dcff8>:   1}
>
> and when I call lookup(d, Baz()) where Baz is a direct subclass of
> Foo, it will return 1. (C types are easy to handle, because they don't
> admit any subtyping.)

That depends on how you have coded your lookup function,
of course; basically, it needs to duplicate the logic of Python's
bases-lookup (left-first, depth-first) unless you want to risk
some big surprise.

Consider, for example:

class Baz(Bar, Foo):
    pass

Now, Baz IS a direct subclass of Foo -- but I would expect the
lookup to yield 2, not 1, because Bar 'dominates' Foo in Baz's
ancestors-DAG -- and directness is less important than other
issues, e.g:

class Baz2(Baz,Foo):
    pass

this inherits from Foo twice, _and_ directly too -- but Bar
STILL dominates Foo, as left-first comes before depth-first.


Nothing wrong in duplicating this logic in your lookup function,
except that I'd much rather have things done ONCE whenever
feasible.  Affixing classes with an attribute, and using that
attribute's value as a key, lets us take advantage of Python's
inheritance mechanism *directly* -- because our getattr for
that attribute uses just that mechanism.  So we KNOW we
have no semantic variation in any subtle aspect, etc.  (It's
also likely to be faster, but that's a secondary issue).


Alex





From pearu at cens.ioc.ee  Sat Mar 10 15:17:14 2001
From: pearu at cens.ioc.ee (Pearu Peterson)
Date: Sat, 10 Mar 2001 22:17:14 +0200
Subject: Can anyone recomend a good intoduction to C...
In-Reply-To: <98dpoe12ckn@news1.newsguy.com>
References: <mailman.983894288.19965.python-list@python.org> <ewwp6.1466$y6.241005@ruti.visi.com> <slrn9aeblb.cg1.scarblac@pino.selwerd.nl> <djagat4kk3cs9hg0c3k39cudk6goblhkg6@4ax.com> <989mbk$24nuv$1@fido.engr.sgi.com> <slrn9ah66m.k09.qrczak@qrnik.zagroda> <98auhn0270j@news2.newsguy.com> <slrn9aiaq7.rmu.qrczak@qrnik.zagroda> <98dpoe12ckn@news1.newsguy.com>
Message-ID: <Pine.LNX.4.21.0103102137300.7964-100000@kev.ioc.ee>

On Sat, 10 Mar 2001, Alex Martelli wrote:

> "Expected" is correct; it's a social issue, NOT a technical one --
> technically, C++ is a better choice for most Python extensions &c
> on a Linux platform.  But, if one wants acceptance, one writes in
> C (e.g., that's what I did for GMPY, because I'd like people to use
> it -- but it would be already out of alpha if had dared use C++, and
> it would be just as fast &c).
> 
> Your opinion is widespread, and self-perpetuating, but it has no
> current technical basis for such tasks as Python extensions on Linux.

I must say that earlier I had also an opinion that no matter what I would
not use C++ for extending Python --- the reasons of my opinion are
now irrelevant after I learned about Boost. Though I had a gap about 2
years of programming in C++, interfacing a C++ library to Python using
Boost was so easy that I caught up with C++ and started to use it on
Linux.

Boost is an excellent tool, so is gcc. Thanks to the authors of these
tools!

But I must say that there is also a dark side in using C++. The
C++ compiler (gcc 2.95.2, in particular) takes more memory (>100MB) than
in my computer when compiling a (still incomplete) Python interface
to the GiNaC C++ library. And this interface has only approx. 300 lines of
code that uses mostly Boost constructs. I don't yet know if I am able to
compile this extension module on my laptop when it will be complete...

In conclusion, extending Python with C++ libraries in connection with 
Boost works very fine on Linux **if** you have plenty of memory available 
in your computer for compilation. At least, that's my experience.

Pearu 



From whisper at oz.net  Thu Mar 29 15:15:40 2001
From: whisper at oz.net (Dave LeBlanc)
Date: 29 Mar 2001 20:15:40 GMT
Subject: Python scripts as Drop targets in Windows Explorer
References: <99qogo$o6g$1@troll.powertech.no> <99qqq3$qt7$0@216.39.170.247> <99r043$qqr$1@troll.powertech.no> <vdqs99.p16.ln@haegar.fqdn.th-h.de> <8kAw6.780$R6.22057@news1.oke.nextra.no>
Message-ID: <9a055c$n2q$1@216.39.170.247>

Yes, i'm using NT 4.0 currently.

The second posting where you said you where putting an ascii version
was also in "gooberish"

Regards,

Dave LeBlanc

On Thu, 29 Mar 2001 08:10:08 +0200, "Syver Enstad"
<syver at NOSPAMcyberwatcher.com> wrote:

>
>"Michael Linnemann" <m.linnemann at addcom.de> wrote
>
>> Yeah, looks cool. What is it, perl? Czech? EBCDIC?
>As it said in the posting:
>>>default on Win2k is to save as unicode.
>> Come on, tune your W2k a little and make it talk plain :-)
>I don't know what you mean, isn't Unicode one of the most plain ways to talk
>nowadays? And I have also posted it as plain text so it shouldn't be too
>difficult for you.
>
>
>



From jmcmonagle at velseis.com.au  Wed Mar 21 23:37:13 2001
From: jmcmonagle at velseis.com.au (John McMonagle)
Date: Thu, 22 Mar 2001 14:37:13 +1000
Subject: Is there python code highlighting in emacs ?
References: <mailman.985223430.22832.python-list@python.org> <3AB978A9.353A0E4B@yahoo.com> <3AB97F1F.70FF97DD@velseis.com.au>
Message-ID: <3AB98179.679DC4A@velseis.com.au>

> 
> > >
> > > The hilit19 package in emacs provides keyword highlighting when editing
> > > code.  For example, if you are writing in C, comments are italicised in
> > > red, strings are in grey, import statements are in lavender, functions
> > > are in bold-blue, etc.
> > >
> > > Does anyone know if there is python support in the hilit19 emacs lisp
> > > package ?
> >
> > I don't know about hilit19, but I use python-mode which highlights
> > using font-lock-mode. Comes with GNU Emacs 20.whatever. Try M-x
> > python-mode and see if anything happens.
> >
> > python-mode does many other useful things, too. I can't imagine
> > working without it at this point.
> >
> 
> I already use python-mode which works fine for indenting, etc.  How do I
> configure this font-lock-mode for python  ? What do I add to my .emacs
> file ?
> 

Worked it out.

Add the following to your .emacs file:

(add-hook 'python-mode-hook '(lambda() (font-lock-mode 1)))




From dsavitsk at e-coli.net  Sat Mar 10 15:39:57 2001
From: dsavitsk at e-coli.net (dsavitsk)
Date: Sat, 10 Mar 2001 14:39:57 -0600
Subject: Tkinter focus_get()
Message-ID: <ziwq6.18374$tV.3430599@newsrump.sjc.telocity.net>

I have a grid (actually a dictionary of dictionaries) of entry widgets that
are dynamically created.  Each has <KeyPress> bound to it.  I am trying to
determine which one has focus, so i used

self.root.focus_get()

which returns something that looks like

.8854796.8898804.18282492.18282532.18283228.18248316.17887444.17854708

is there a way to interpret this? or a better way to get focus?

how the entries are created is shown below (this creates a row of entries in
a ScrolledCanvis)

-----------------------------------------------------
def AddRow(self):
        y = {}
        self.itemnumber = self.itemnumber + 1
        y['lbl'] = str(self.itemnumber) + '.'
        self.data_dict[self.itemnumber] = y
        y['frm'] = Frame(self.ft2.interior(), bg='#AAAAFF')
# frame for entry and label
        Label(y['frm'], width=11, bg='#eeeeee', justify='right',
wraplength=172, text=y['lbl']).pack(side=LEFT, fill=X, expand=YES)
        for i in ['a', 'b', 'c', 'd', 'e']:
            y[i + 'd'] = StringVar()                                     #
entry's data
            y[i + 'e'] = Entry(y['frm'], relief='solid', borderwidth=0,
textvariable=y[i + 'd'])
            y[i + 'e'].bind('<KeyPress-Up>', self.reportUp)
            y[i + 'e'].bind('<KeyPress-Down>', self.reportDown)
            y[i + 'e'].bind('<KeyPress-Left>', self.reportLeft)
            y[i + 'e'].bind('<KeyPress-Right>', self.reportRight)
            y[i + 'e'].pack(side=LEFT, expand=YES, fill=X, padx=1, pady=1)
        y['frm'].pack(side=TOP, expand=YES, fill=X, padx=0, pady=0)
        self.grid[self.itemnumber] = y
------------------------------------------------------

thanks again,

doug




From shaleh at valinux.com  Fri Mar  2 16:06:56 2001
From: shaleh at valinux.com (Sean 'Shaleh' Perry)
Date: Fri, 02 Mar 2001 13:06:56 -0800 (PST)
Subject: the demise of 'from foo import * and its implications?
Message-ID: <XFMail.20010302130656.shaleh@valinux.com>

I understand that the 'from foo import *' syntax will be deprecated some day
soon.  How will this affect things like thr gtk python binding which has many
many functions, defines, etc?  Will I have code where every other word is
'gtk.Foo'?



From Markus_Reitz at yahoo.com  Wed Mar 21 06:44:09 2001
From: Markus_Reitz at yahoo.com (Markus Reitz)
Date: Wed, 21 Mar 2001 12:44:09 +0100
Subject: Bug or feature?
Message-ID: <99a433$b8b$1@sun.rhrk.uni-kl.de>

Hi,

I'm using Python 2.0 and have found a, according to my opinion, strange
behaviour.

a=[]
a.extend('Test')

Printing a results in:

a=['T','e','s','t']

I expected a TypeError message, because a String is not a list and so it
can't be used to extend an existing list.

On the other hand, a string can be seen as a kind of "list of characters"
...

Is this behaviour a bug or a feature?

Markus




From dsh8290 at rit.edu  Tue Mar 27 16:44:29 2001
From: dsh8290 at rit.edu (D-Man)
Date: Tue, 27 Mar 2001 16:44:29 -0500
Subject: Good Editor? (What's the big deal with syntax colouring?)
In-Reply-To: <99erbn$3e@gap.cco.caltech.edu>; from n8gray@caltech.edu.is.my.e-mail.address on Thu, Mar 22, 2001 at 10:38:00PM -0800
References: <008F0A63472BD311AF9800104BCD102561CD3E@minirex.vmlabs.com> <mailman.985295730.14145.python-list@python.org> <99erbn$3e@gap.cco.caltech.edu>
Message-ID: <20010327164429.B3707@harmony.cs.rit.edu>

On Thu, Mar 22, 2001 at 10:38:00PM -0800, Nathaniel Gray wrote:
| Timothy Grant wrote:
| > Now if only else didn't highlight but else: did<wink>

$ vim $VIM/syntax/python.vim   

;-)  you can adjust the regexes to your preferences

| Ah, but that's what auto-indent is for!  When NEdit doesn't indent the 
| first line of a block I know I forgot the colon at the end of the last line.
| 
| One of these days I'll send in my ultra-improved auto-indent Python macros 
| for NEdit.  I'm still working on the auto-dedent macros... <wink>

I can't wait for auto-dedent in vim, it has most of the auto-indent
correct.  Anyone want to work on making it?  Vim is python-scriptable
as well.

-D




From joelg at alum.mit.edu  Mon Mar 12 07:42:37 2001
From: joelg at alum.mit.edu (Joel Gould)
Date: Mon, 12 Mar 2001 12:42:37 GMT
Subject: Tkinter with PIL crashing on large images - help
Message-ID: <3aacc208.1842603@news.ne.mediaone.net>

Is this expected behavior?  If I load a 3.3 megapixel JPG file into a
PIL (Python Image Library) PhotoImage object and then draw that on a
Tkinter.Canvas object, Python crashes with an access violation (under
WIndows 98).

I have attached the code that fails. You can ge a suitable image from
this link:
http://www.gouldhome.com/Travel/StThomas00/images/001229_1106_charlotteAmalie3.jpg

I have tried this using the latest py20 distribution (Python 2.0, PIL
1.1.1, etc.) from Secret Labs.  I also tried it using the PythonWorks
Testdrive 1.1 (which uses Python 1.5.2) also from Secret Labs.

I am using Windows 98 SE on a Pentium III with 256MB of RAM.

Can anyone tell me why this is failing?

P.S.  I know the image is very big and I can make it smaller before
displaying it.  I just do not think it should GPF in this case.

Joel Gould
joelg at alum.mit.edu

----- snip ----

import Tkinter
import Image
import ImageTk

class MainDialog:
    def __init__(self, parent):
        self.canvas = Tkinter.Canvas(parent,width=700,height=500)
        self.canvas.pack()
        self.image = Image.open("001229_1106_charlotteAmalie3.jpg")
        self.tkim = ImageTk.PhotoImage(self.image)
        self.item = self.canvas.create_image(
            0,0,image=self.tkim,anchor='nw')

if __name__ == '__main__':
    root = Tkinter.Tk()
    dialog = MainDialog(root)
    root.mainloop()



From kammie at commie.com  Sun Mar 11 13:38:05 2001
From: kammie at commie.com (kammie)
Date: Sun, 11 Mar 2001 18:38:05 GMT
Subject: Python and ObjectSpace Voyager
Message-ID: <4chnatkpmpvo29v515orlcjcfuoii5gcte@4ax.com>

Hello --

Has anyone heard of writing Python CORBA apps for use with the Voyager
orb?  It would be neat if that orb had its own Python mappings (I
don't think so), otherwise IIOP would be nice.


From ws-news at gmx.at  Wed Mar 14 17:46:02 2001
From: ws-news at gmx.at (Werner Schiendl)
Date: Wed, 14 Mar 2001 23:46:02 +0100
Subject: This math scares me
References: <3aad4062.1290021@News.CIS.DFN.DE> <20010312170356.B3513@better.net> <mailman.984480856.18097.python-list@python.org> <984486395.407379@newsmaster-04.atnet.at> <slrn9asf1d.ja3.qrczak@qrnik.zagroda> <984518199.127140@newsmaster-04.atnet.at> <jwbnews-09D75F.13313313032001@corp.supernews.com> <98ohuc$du9$1@news.udel.edu> <7bQr6.644$lz1.17686@ruti.visi.com>
Message-ID: <984610248.564504@newsmaster-04.atnet.at>

> In fiancial calculations you also need to represent numbers
> that aren't amounts of money.  Interest rates for example.
>

How about 1/100 of a percent?
So you can stay with integer arithmetics, especially if you consider pythons
'unlimited' long

If 0.01 % does not suffice in precision, you can store it as 0.0001 % units
(say, store 1.0625% as 10625, no comma whatsoever)

regards
werner





From gerhard.nospam at bigfoot.de  Thu Mar 22 11:59:36 2001
From: gerhard.nospam at bigfoot.de (Gerhard =?iso-8859-1?Q?H=E4ring?=)
Date: Thu, 22 Mar 2001 17:59:36 +0100
Subject: Python 2.1 and MySQL 3.23.33 on Windows
References: <99d1te$ogo$1@news.netpower.no>
Message-ID: <3ABA2F78.87E037CA@bigfoot.de>

David Brown wrote:
> 
> I have MySQL 3.23.33 and Python 2.1 installed on my Windows machine.  Both
> work fine (I use MySQL along with PHP and Apache), and I have used Python
> for a few short scripts.  But I cannot get Python to work with MySQL.  The
> mySQLdb windows installation is built for Python 2.0, not 2.1, and I do not
> have MSVC to recompile it myself.  Does anyone have any hints or pointers to
> get me started?  Thanks.

A new version of the MySQLdb Windows binary distribution is available at
my homepage. This one contains binaries for Python 1.5.2, 2.0 and 2.1
Beta 1.

Gerhard
-- 
Sorry for the fake email, please use the real one below to reply.
contact: g e r h a r d @ b i g f o o t . d e
web:     http://highqualdev.com


From shaleh at valinux.com  Thu Mar 22 13:56:15 2001
From: shaleh at valinux.com (Sean 'Shaleh' Perry)
Date: Thu, 22 Mar 2001 10:56:15 -0800 (PST)
Subject: Python and emacs
In-Reply-To: <upuf9k5i3.fsf@advancedmp.net>
Message-ID: <XFMail.20010322105615.shaleh@valinux.com>

On 22-Mar-2001 Don Dwiggins wrote:
> A semi-idle question: has anyone looked into embedding Python as an
> extension language in Emacs, alongside Elisp?  I'm toying with the idea of
> doing some Emacs enhancements/modifications, and given the choice, I'd much
> rather do them in Python.  (One such enhancement would be to add
> Pythonwin-style "intellisense" popups to python-mode.)
> 

An editor that spoke python would not be a bad thing.  However, trying to get
emacs to speak python strikes me as an amazingly good way to waste whatever
brain cells you have left.



From hzhu at mars.localdomain  Fri Mar  9 03:36:37 2001
From: hzhu at mars.localdomain (Huaiyu Zhu)
Date: Fri, 09 Mar 2001 08:36:37 GMT
Subject: polymorphjsm &c (was Re: I come to praise .join, not to bury it...)
References: <mailman.983485923.1136.python-list@python.org> <97p7iq$kk2$1@nntp6.u.washington.edu> <97rka002k9@news1.newsguy.com> <9811a8$k5k$1@nntp6.u.washington.edu> <982cja0v5c@news1.newsguy.com> <983u1l$6mk$1@animus.fel.iae.nl> <9851gf014vq@news1.newsguy.com>
Message-ID: <slrn9ah53f.vh.hzhu@mars.localdomain>

On Wed, 7 Mar 2001 11:11:17 +0100, Alex Martelli <aleaxit at yahoo.com> wrote:
>
>but neither do you get what you want, because a.split(b) works
>irregularly...:
>
>    -- if a is a re object, or if it's module-object os.path,
>        then string b is split using splitter-object a
>    -- if a is a string object (including a Unicode string),
>        then string a is split using splitter-object b (which
>        must also be a string)
[...]
>
>It's no doubt too late to remedy the discrepancy between
>the semantics of a.split(b) depending on a's type (string
>versus re) and the resulting irregularity (which we no
>doubt will have to live with anyway) is unpleasant, to say
>the least.

Maybe not too late?

sep.splits(thing)     => list 
sep.joins(list)       => thing

list.joinswith(sep)   => thing
thing.splitswith(sep) => list

This would allow sep to be either string, re or os.path, etc. The words
joinswith and splitswith are comparable to startswith, endswith.

In any case, we have to live with two orders of split whichever way the
original design goes.  How else would one reconcile the following?

re.compile.split         # sep.split(string)
file.read.splitline      # string.split(sep)

But honestly, after seeing all these arguments and switching opinion twice I
come to believe only multimethod would be the true solution.  Wait for Py3k?

Just-realize-I-was-promoting-TMTOWTDI-ly yr's

Huaiyu


From max at alcyone.com  Wed Mar 28 11:09:55 2001
From: max at alcyone.com (Erik Max Francis)
Date: Wed, 28 Mar 2001 08:09:55 -0800
Subject: list vs tuple
References: <hGlw6.782$p5.3210@news1.rivrw1.nsw.optushome.com.au>
Message-ID: <3AC20CD3.265A2D94@alcyone.com>

deadmeat wrote:

> Whats the practical difference between a list[] and a tuple() ?
> 
> A tuple seems to be a very basic list, so unless it's faster, what's
> the
> point of it?

A list is mutable, whereas a tuple is immutable.  That means that
tuples, for instance, cannot be changed behind your back through other
references, and that tuples can be used as dictionary keys where lists
can't.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ 'Tis man's to fight, but Heaven's to give success.
\__/ Homer
    REALpolitik / http://www.realpolitik.com/
 Get your own customized newsfeed online in realtime ... for free!


From richard at loudcloud.com  Tue Mar 27 15:24:12 2001
From: richard at loudcloud.com (Richard Weth)
Date: Tue, 27 Mar 2001 12:24:12 -0800
Subject: How does one set a value via win32com
Message-ID: <3AC0F6EC.314B103D@loudcloud.com>

All

I have been banging away on an assignment that requires me to alter com
plus settings
(something one should never do normally). Here is the pertinent code
fragment:
import win32com.client
import regex
wcc = win32com.client

adm = wcc.Dispatch( "COMAdmin.COMAdminCatalog" )
app = adm.GetCollection("Applications")
app.Populate()

for thingy in app:
  if regex.match("^Hello World.*",thingy.Name)!= -1:
    print "what we have here is ...",thingy.Name
    thingy.Value("Name") = thingy.Name + "t0"          #
<================ THE PROBLEM
    break

  else:
    print thingy.Name
app.SaveChanges()


Now the code in VBS
    thingy.Value("Name") = thingy.Name & "t0"
works just fine but in python I get:
>>> thingy.Value("Name") = "joemama"
SyntaxError: can't assign to function call

Is there a way to set that (and other) properties in Python via
win32com. I really
would like to learn as little VB as humanly possible and not write this
in it's own VBScript.
Any and all suggestions welcome!

--
Sincerely
  Richard A. Weth
  Sr. Software Engineer
  LoudCloud, Inc
  richard at loudcloud.com





From sheila at spamcop.net  Sun Mar 18 15:12:26 2001
From: sheila at spamcop.net (Sheila King)
Date: Sun, 18 Mar 2001 20:12:26 GMT
Subject: Help with 500 error on cgi script
References: <quv8bto7r8s6f1tckc7qpmuir0kvufp1nb@4ax.com> <mailman.984909623.8814.python-list@python.org> <5739bt04e50ku01rn3dptgqch3l07p2cnf@4ax.com> <3AB492FE.4D963C32@stroeder.com>
Message-ID: <7k5abt0vkd55rbbnrb818fprrc1nslu02g@4ax.com>

On Sun, 18 Mar 2001 11:50:38 +0100, Michael Str?der <michael at stroeder.com>
wrote in comp.lang.python in article <3AB492FE.4D963C32 at stroeder.com>:

:What keeps you away from correcting the syntax error mentiond in the
:error traceback?

I guess I was too tired to notice the missing "

I went to bed shortly after I posted this. Thanks for your help.

--
Sheila King
http://www.thinkspot.net/sheila/
http://www.k12groups.org/



From jafo at tummy.com  Mon Mar 26 03:35:58 2001
From: jafo at tummy.com (Sean Reifschneider)
Date: Mon, 26 Mar 2001 01:35:58 -0700
Subject: ANNOUNCE: Distutils upload patch (PEP243)
Message-ID: <20010326013558.A9281@tummy.com>

I've got the server and client side of my PEP243 implementation (for
uploading distributions made by distutils to the repository server) built.
I've also created the code for distutils (2.1b2) to do the upload.  For
example:

   guin:python-netstring$ ./setup.py sdist --submit
   [...]
   gzip -f9 dist/python-netstring-1.14.tar
   removing 'python-netstring-1.14' (and everything under it)
   Submitting to repository server...
       sending dist/python-netstring-1.14.tar.gz...
         Succeeded.
     done.

Currently, I have the repository server set up on our "public service"
server "community.tummy.com".

The files you will need are:

   ftp://ftp.tummy.com/pub/tummy/swalow/swalow-1.13/distutils-swalow.patch
   ftp://ftp.tummy.com/pub/tummy/swalow/swalow-1.13/swalowsupp.py

See the README below for information on how to apply it.

The server for handling the requests is also available at the above site,
you'll need a few support modules as well (like the tar module for
extraction of the PKG-INFO file).

I need to update the repository server and client for PEP241, but that
should be pretty straight-forward.  Also, if anyone else wants to get
copies of the data sent to my repository, I can certainly arrange that...

Any comments?

Sean
=========================
swalow + Distutils README
Sean Reifschneider <jafo-swalow at tummy.com>
==========================================

The file "distutils-swalow.patch" contains a patch against the 2.1b2
distutils which allows "setup.py sdist --submit" to upload the completed
distribution file to the repository server.  This will allow automatic or
semi-automatic submissions of modules.

To gain this functionality, apply the above patch, and copy "swalowsupp.py"
into the main distutils package directory.

If the environment variable "PYTHON_MODULE_SERVER" is set, that value
overrides the default host for uploading the distribution to.
-- 
 That weapon will replace your tongue. You will learn to speak through
 it.  And your poetry will now be written with blood. -- _Dead_Man_
Sean Reifschneider, Inimitably Superfluous <jafo at tummy.com>
tummy.com - Linux Consulting since 1995. Qmail, KRUD, Firewalls, Python



From ben.hutchings at roundpoint.com  Tue Mar 27 18:59:32 2001
From: ben.hutchings at roundpoint.com (Ben Hutchings)
Date: Tue, 27 Mar 2001 16:59:32 -0700
Subject: daemon app in win32
References: <99nk49$a2q$02$1@news.t-online.com> <mailman.985619314.10299.python-list@python.org> <99nsn3$q39$06$1@news.t-online.com> <99o2ge$9on$0@216.39.170.247>
Message-ID: <uu24ebw63.fsf@roundpoint.com>

whisper at oz.net (Dave LeBlanc) writes:

> On Mon, 26 Mar 2001 19:02:27 +0200, Michael Bauer <Mike_B at T-Online.de>
> wrote:
> 
> >Steve Purcell wrote:
> >
> >> Michael Bauer wrote:
> >>> i am writing a daemon application that runs both under windows and unix.
> >>> Since i am not very experienced in win32 - programming, i really would
> >>> appreciate some hints how to register my app in win32 that i can do some
> >>> cleanup-functions when windows is shut down.
> >>> 
> >>> At the moment it is just a standard console application that simply gets
> >>> killed when windows shuts down. The app does not have/need a GUI, it
> >>> should just run in the background.
> >> 
> >> 
> >> If you're using NT, take a look at the win32service API. I think it
> >> provides hooks like SvcStop that Windows will call for you (if you ask it
> >> nicely).
> >
> >Hmm, seems like this is only available under NT. Is there a more general 
> >way to accomplish this task independently of the underlying Windows-version?
> >
> >But maybe i should just get a life and write different wrappers for Win95, 
> >Win98, WinME, Win200, WinNT... :-(

You should only need two.  The Apache web server code for Windows
might be instructive.

> >still hoping... Mike

> AFAIK, Every Windows app gets sent a WM message when it's about to be
> terminated.

If it has a GUI, then its top-level window will be sent WM_CLOSE.

Console applications can register to receive such notifications in the
same way they receive control-C notifications under *some* versions of
Windows.

> Most apps just let this fall through to the default
> message processor (which you have to do anyway) without doiing
> anything to clean up. You can add a message handler for this such that
> the app can gracefully (make out it's will?) prepare to end it all.

The DefWindowProc (default message handler) responds to WM_CLOSE by
calling DestroyWindow() (which runs synchronously).  The message
handler for a top-level window should respond to this by calling
PostQuitMessage(), which posts a WM_QUIT message to the calling
thread's message queue.  (The message handler for a dialog should do
something else again.)  Application frameworks like MFC should do
this stuff for you.

> Just make sure that in the end, it passes the message on to the
> default message handler so that windows can do it's own cleanup of
> your app.

Actually, what's supposed to happen is that your main message loop
calls GetMessage(), which finds a WM_QUIT message and returns 0; the
loop then exits and your cleanup code runs on the way out.

-- 
Any opinions expressed are my own and not necessarily those of Roundpoint.



From glennj at pcard31e.ca.nortel.com  Thu Mar  8 11:24:49 2001
From: glennj at pcard31e.ca.nortel.com (Glenn W Jackman)
Date: 8 Mar 2001 16:24:49 GMT
Subject: grabbing return codes from os.system() call
References: <%wFp6.9481$dL4.126404@vixen.cso.uiuc.edu> <u66hky36l.fsf@ktgroup.co.uk>
Message-ID: <988boh$kl5$1@bcarh8ab.ca.nortel.com>

Pat Knight wrote:
>Look at the process management section of the os module documentation in
>Python. This contains functions to decode exit codes.
>
>Also, check the UNIX man pages for system, and the exec, exit and waitpid
>system calls. You'll find details of the encoding, and probably find that you
>can't have the large exit status values you want.
>
>If you're wanting to return strings or large numbers, start investigating
>os.popen.

Also have a look at the 'commands' module.

-- 
Glenn 


From dnew at san.rr.com  Thu Mar 15 13:15:41 2001
From: dnew at san.rr.com (Darren New)
Date: Thu, 15 Mar 2001 18:15:41 GMT
Subject: I come to praise .join, not to bury it...
References: <mailman.983485923.1136.python-list@python.org> <97p7iq$kk2$1@nntp6.u.washington.edu> <97rka002k9@news1.newsguy.com> <3AAF7E47.294617E1@sage.att.com> <98qeco02hn7@news1.newsguy.com>
Message-ID: <3AB106CD.7D259EBC@san.rr.com>

Alex Martelli wrote:
> And then again, the _appearance_ that too much emphasis
> was being placed on syntax-sugary issues kept me *AWAY*
> from Python for longer than it should have

Actually, what I found quite cool about Python is that the procedural syntax
maps to OO semantics. I.e., in Smalltalk, everything looks like a method
call, including "if". In Eiffel, everything that isn't a method call doesn't
look like a method call, and vica versa. But in Python, the convenient and
familiar "x = y[5:10]" maps to a real live method call you can override.

Well, *I* thought it was neat. It lets you be OO where you want, and keeps
it out of your face otherwise.

-- 
Darren New / Senior MTS & Free Radical / Invisible Worlds Inc.
San Diego, CA, USA (PST).  Cryptokeys on demand.


From steinar.eriksen at fikse.net  Tue Mar 20 11:28:45 2001
From: steinar.eriksen at fikse.net (Steinar Eriksen)
Date: Tue, 20 Mar 2001 17:28:45 +0100
Subject: Jython and IDE/debugger
Message-ID: <9980fp$sm0$1@news.ost.eltele.no>

I would like to prototype my Java application in Jython, byt calling up
Jython scripts from my Java classes. Whilst developing my Jython scripts; I
am looking for some IDE or debugging tool to help me test my code. Anyone
seen that available for Jython ?

I know the Python IDE's are excellent, but since the libraries differ I
suppose they are not applicable,


Regards

Steinar Eriksen




From harisri at bigpond.com  Thu Mar 29 06:47:05 2001
From: harisri at bigpond.com (Srihari Vijayaraghavan)
Date: Thu, 29 Mar 2001 21:47:05 +1000
Subject: A newbie question on ftplib.
Message-ID: <87Ew6.5454$45.29758@newsfeeds.bigpond.com>

Hello,

When I execute the following statement under Linux it uploads the binary 
file properly, but under Windows NT it creates only 512 bytes file at the 
ftp server.

ftp.storbinary('STOR ' + xyz, open(xyz), 1024)

I tried various values (1, 128, 256, 515) for the "number of blocks" 
parameter.

I am using Python 2.0 on both Windows NT (Windows NT 4.0 / SP6a), and on 
Linux. (Linux 2.2.x and 2.4.x)

Can I someone help me correcting if I am doing something wrong?

Advance thanks,
-- 
Hari.


From mal at lemburg.com  Wed Mar 14 13:46:20 2001
From: mal at lemburg.com (M.-A. Lemburg)
Date: Wed, 14 Mar 2001 19:46:20 +0100
Subject: cursor.execute fails under mxODBC
References: <3AABF77F.7B77D51@northwestern.edu> <3AAE59AE.630C148C@lemburg.com> <3AAEF113.E928A434@northwestern.edu>
Message-ID: <3AAFBC7C.2A1CD536@lemburg.com>

Louis Luangkesorn wrote:
> 
> > Thank you for forwarding the messages. I have had numerous questions
> > about this problem and will look into it later this week (just came
> > back from the Python conference and am still fighting the jet lag).
> >
> > One thing I've noted about the problem reports is that all seem
> > to use ActiveState Python 2.0. There may be an issue here, since
> > the mxODBC (alas the egenix-commercial) package was compiled against
> > the BeOpen Python 2.0 version which is available from python.org.
> >
> > Could you perhaps check whether you experience the same problems
> > using BeOpen Python 2.0 ?
> >
> > Thanks,
> > Marc.
> 
> Marc
> 
> No, that did not work. Here it is under the BeOpen-Python-2.0  After the
> cursor.execute(...) the Python shell hangs.

I've uploaded a new version of the Windows installer to:

http://www.lemburg.com/files/python/egenix-mx-commercial-2.0.0.win32-py2.0.exe

There seems to have been some build error during the release
session which was introduced after the final tests. Please check
whether you still get the indicated problems with the new 
build.

One thing this teaches is that distutils should probably clear
the build directory before rebuilding a package... I'll add that
to my batch file.

Thank you for looking into this and sorry for the trouble 
this has caused you. I hope the new build works for you as well
(it does for me on Windows 9x -- don't have a W2k to test on).

--
Marc
 
> Louis
> ---begin output---
> Python 2.0 (#8, Oct 16 2000, 17:27:58) [MSC 32 bit (Intel)] on win32
> Type "copyright", "credits" or "license" for more information.
> IDLE 0.6 -- press F1 for help
> >>> import mx.ODBC.Windows
> >>> channel = mx.ODBC.Windows.Connect('sqltutor')
> 
> >>> cursor = channel.cursor()
> >>> cursor.execute('select FirstName, LastName, Address, City, State FROM
> EmployeeAddressTable')
> 
> ---end output---
> --
> K Louis Luangkesorn
> lluang at northwestern.edu  http://pubweb.nwu.edu/~kll560  PGP:0xF3E2D362
> 
> Whatsoever things are true, ... honest, ... just, ... pure, ... lovely, ... of
> good report; if there be any virtue, and if there be any praise, think on
> these things.
>      - motto - Northwestern University (Phil 4:8)

-- 
Marc-Andre Lemburg
______________________________________________________________________
Company & Consulting:                           http://www.egenix.com/
Python Pages:                           http://www.lemburg.com/python/



From pae at dim.com  Thu Mar 29 00:43:53 2001
From: pae at dim.com (Phil Earnhardt)
Date: Wed, 28 Mar 2001 22:43:53 -0700
Subject: Collection interfaces
References: <3a90fa97.0@news1.mweb.co.za> <MPG.15085517debf4b649896ca@news.earthlink.net> <zb%n6.31033$o%6.586026@news1.rdc1.bc.home.com> <o1jv6.9635$P4.789083@newsread1.prod.itd.earthlink.net> <99l6s8$gpu$1@nyheter.chalmers.se> <m38zlq4w34.fsf@chaos.resnet.gatech.edu> <lv13ct8qbni7f3h7q9cujj6d3iequ8to2s@4ax.com> <m3g0fxg8qt.fsf@chaos.resnet.gatech.edu>
Message-ID: <ffi5ct0v5qlt2ckiqusfe0254fto2ppkit@4ax.com>

On 28 Mar 2001 11:26:12 -0500, Lex Spoon <lex at cc.gatech.edu> wrote:

>> >>     Yea, use a Java applet instead. Guaranted not to fiddle with your
>> >> DLL's :-)
>> >
>> >Oh great, so we get to fiddle with JAR's instead of DLL's!  At least
>> >JAR is easier to pronounce.  :)
>> 
>> Can you explain your comment, Lex? How is downloading and running
>> applets from a jarfile have anything to do with fiddling with DLLs?
>
>Java programs tend to require you to have various extensions installed
>as JAR files, just the same way that Windows EXE's tend to require
>that you have various extensions installed as DLL's.

Not really. A single Jar can be used to deploy your entire
application.

>  I'd hazard that
>JAR's are *worse* than DLL's -- DLL's at least *pretend* to do
>something about versions.

What you're talking about does not apply in the case where a single
application is deployed as a single Jar. In that case, you don't
really need version numbers -- you just overwrite the old jarfile with
the new one.

What you're talking about is the case where an API shared between
multiple applications is implemented in a Jar. In that case, you're
correct. As a practical matter, there is often less need for such APIs
in Java because the core class libraries provide rich functionality.

>Admittedly, I missed the "applet" part.  Applets run with a standard
>set of classes available, so they don't have that problem.  (Instead,
>they have the problem that *no* extensions are available.

What exactly does that mean?

>-Lex

--phil



From digitig at cix.co.uk  Wed Mar 21 17:02:00 2001
From: digitig at cix.co.uk (Tim Rowe)
Date: Wed, 21 Mar 2001 22:02 +0000 (GMT Standard Time)
Subject: I'm sure there's a good reason!
Message-ID: <memo.20010321220225.57471D@digitig.compulink.co.uk>

Can any of the assembled experts explain to me why:
	print r"\hello"
is fine, but
	print r"hello\"
isn't? In other words, why does the backslash escape quotes in a raw 
string? At one level I'm going to guess (without reading the spec) that 
it's because "that's what the spec says", but why? I thought the idea of 
raw strings was that they /didn't/ handle escapes!

TIA,

Tim



From jwbnews at scandaroon.com  Mon Mar 12 22:09:27 2001
From: jwbnews at scandaroon.com (John W. Baxter)
Date: Mon, 12 Mar 2001 19:09:27 -0800
Subject: This math scares me
References: <3aad4062.1290021@News.CIS.DFN.DE> <slrn9aqkth.aki.scarblac@pino.selwerd.nl>
Message-ID: <jwbnews-3E2DF8.19092712032001@corp.supernews.com>

In article <slrn9aqkth.aki.scarblac at pino.selwerd.nl>,
 scarblac at pino.selwerd.nl (Remco Gerlich) wrote:

> If you can avoid them, don't use floats. If you're working with money, count
> cents instead (and use longs if you have numbers over a few hundred million
> cents).

Or perhaps mils, for a little extra room for tax computations.

  --John

-- 
John W. Baxter   Port Ludlow, WA USA  jwbnews at scandaroon.com


From tan at pobox.com  Fri Mar  2 17:12:07 2001
From: tan at pobox.com (tan at pobox.com)
Date: Fri, 2 Mar 2001 15:12:07 -0700
Subject: ImportError with extensions in python2.1 installed in home dir
Message-ID: <20010302151207.A4418@tan.powerandlove.com>

folks,

I've installed python 2.1 in my home directory on the web
hosting service I use (because they have no plans to move
from python 1.5.2 to 2.x anytime soon, and I don't have
access to /usr.) Installation succeeds, and modules that
don't import extension modules work fine, but when I try to
import an extension module, I get

ImportError: No module named [name of extension module]

If I uncomment the module in Modules/Setup, the module gets
compiled into libpython2.1.a correctly, I don't get the
error, and the module seems to work as it should.

sys.path is correct (my python2.1 binary is looking for
python2.1, plat-linux2, lib-tk, lib-dynload and
site-packages in the right place.)

The system is:
Linux 2.0.36 i686 unknown

Before I installed python 2.1, I installed a package
(4Suite) that contains a few extension modules and tried using
them with the python1.5 binary in /usr/bin. Same
problem. 

The python1.5 binary can find and run extension modules in
/usr/lib/python1.5/lib-dynload. But neither it nor the
python2.1 binary I compiled can find extension modules in my
home directory tree.

I suppose whenever I add a package I could compile any
extension modules it contains into libpython2.1.a, but I'd
rather not.

What am I missing?

Please cc any replies to tan at pobox.com.

thanks,
Tom
-- 
Tom Newman	  tan at pobox.com



From dsh8290 at rit.edu  Wed Mar  7 13:41:36 2001
From: dsh8290 at rit.edu (D-Man)
Date: Wed, 7 Mar 2001 13:41:36 -0500
Subject: I come to praise .join, not to bury it...
In-Reply-To: <97rka002k9@news1.newsguy.com>; from aleaxit@yahoo.com on Sat, Mar 03, 2001 at 09:29:55PM +0100
References: <mailman.983485923.1136.python-list@python.org> <97p7iq$kk2$1@nntp6.u.washington.edu> <97rka002k9@news1.newsguy.com>
Message-ID: <20010307134135.A29445@harmony.cs.rit.edu>

On Sat, Mar 03, 2001 at 09:29:55PM +0100, Alex Martelli wrote:
| "Russell E. Owen" <owen at astrono.junkwashington.emu> wrote in message
| news:97p7iq$kk2$1 at nntp6.u.washington.edu...
|     [snip]
| > However, I do agree that join is not intuitive. I believe the problem
| > (at least for the way I look at it) is that join should be a list
| > method, not a string method. I.e.:
| >
| >   joined_string = ['a', 'b'].join(', ')
| >
| > makes a lot of sense to me. The current join string method does not.

This was my opinion as well.

| 
| This seems to be a widespread opinion, and I've already tried
| to explain why my take on it differs, but that was a few months
| ago, and apparently the current crop of discussants hasn't read
| those discussions, so, let's give it one more spin.
| 
| Python does not have multi-methods: it's a single-dispatch

Where can I learn about multi-methods and their
usefulness/applicability?

| 
[snip]
| 
| So, since we want general polymorphism on the joiner
| object, but are quite content with polymorphism through
| the standard sequence interface on the sequence object,
| it is _just right_ that .join be a method on the joiner
| object (e.g., a string) and that it take the sequence of
| string to be joined as its argument.

Ok, so instead of calling "join" a "string method" how about using the
following Java-Python hybrid snippet to describe it :

interface Joiner :
    abstract Object join( Sequence s ) 

class String implements Joiner :
    String join( Sequence s ) :
        """
        oh, cool.  I can mix and match the signatures sensibly without
        casting ;-)
        """
        ...


This does make sense to me.  It looks logical and reasonable.  You
have convinced me Alex.  Your English (and writing) classes have paid
off ;-).  You must have been a writer before computers took over ;-).

-D


PS.  BTW I haven't finished reading this thread yet



From rikbo716 at student.liu.se  Sat Mar 24 01:31:53 2001
From: rikbo716 at student.liu.se (Rikard Bosnjakovic)
Date: Sat, 24 Mar 2001 07:31:53 +0100
Subject: Install Python on Apache
References: <99fph5$a2n$1@rex.ip-plus.net>
Message-ID: <3ABC3F59.336D3296@student.liu.se>

Sharp wrote:

> (cgi.bin is set to the right place because php and perl do work)
...
> [Fri Mar 23 15:54:50 2001] [error] [client xxx.xxx.xxx.xxx] Premature end of
> script headers: c:/web/cgi-bin/python/python.exe

What you need to do with CGI-scripts is to send headers before the
actual CGI is run. The headers tell what kind of data the script will
send. To send a header, you just need a "Content-type: <your type>\n"
before the rest of the output.

In your case, you are sending html-tags. Thus add a print "Content-type:
text/html\n" as first CGI-statement.


-- 
Rikard Bosnjakovic - http://a214.ryd.student.liu.se/cv/ - ICQ: 1158217

Anyone sending unwanted advertising e-mail to my address will be
charged $250 for network traffic and computing time. By extracting my
address from this message or its header, you agree to these terms.


From wayne.izatt at _nospam_myself.com  Tue Mar  6 02:37:21 2001
From: wayne.izatt at _nospam_myself.com (Wayne Izatt)
Date: Tue, 06 Mar 2001 07:37:21 GMT
Subject: iis and python
References: <F5622749FED6DA5D.AF4594816A4865C9.16CD98F1C1FAC3F6@lp.airnews.net>
Message-ID: <Rs0p6.183447$df5.5139760@news1.crdva1.bc.home.com>

> In order to use python it would have to entail cgi or ASP. Is this
correct.
> Or is there an alternative. I dont know if Zope is the right thing. I hate
> DTML.

Zope is very powerful, and if you want to you can write almost all your code
in Python. DTML is there for presentation, not for logic, and should only be
used for simple processing.

> Is using python in ASP a pain in the but? Am I going to have to learn a
> whole technology just to implelment python well into it?

No, its fairly straighforward. However, before you go the ASP route, take
the time to examine Webware. Webware gives you ASP with Python, as well as
Python Server Pages and servlets - very powerful, easy to use, and pure
Python.

> I hear troubleshooting ASP pages is not fun.

Correct.

> The only time I have used iis is with php and it was a total snap.
Painless.

PHP is fine, but I've had the pleasure of debugging someone else's massive
'one-script-wonder', and I remembered what my misgivings about Perl-based
languages were. Don't flame me, PHPer's, I still like the language!

cheers





From wcanopy at covad.net  Mon Mar 12 18:17:49 2001
From: wcanopy at covad.net (Wiwih Gunadi)
Date: Mon, 12 Mar 2001 17:17:49 -0600
Subject: wxPython UI refresh
Message-ID: <98jku0$ifi$1@newstest.laserlink.net>

Hi,

I used the ActiveXWrapper_IE.py to load an Excel spreadsheet.  The problem
is whenever the spreadsheet is loaded, the screen doesn't update and I have
to minimize the window and restore it again (or do other things such as
switching the active notebook page and back again)

Strangely, this doesn't happen when I'm loading ordinary html file, so I
suppose this is IE's quirk, does anybody know how to fix this?

I've tried Refresh(), Validate, wxYield() on the OnSize event handler to no
avail.

Appreciate any help!
BTW, Excel is not an ActiveX control is it? i.e. it's impossible to "wrap"
Excel like IE with the activexwrapper, right?

-Will--




From Michael.Husmann at teleatlas.com  Thu Mar 15 08:28:47 2001
From: Michael.Husmann at teleatlas.com (Michael Husmann)
Date: Thu, 15 Mar 2001 14:28:47 +0100
Subject: Umlaut and Tkinter
Message-ID: <3AB0C38F.34218EEB@teleatlas.com>

typing in idle.py using Python 2.1b1:

>>> print "??"

results in an:

UnicodeError: ASCII encoding error: ordinal not in
range(128)


Is there someone who can help me.

Michael



From syver at NOSPAMcyberwatcher.com  Tue Mar 27 17:12:03 2001
From: syver at NOSPAMcyberwatcher.com (Syver Enstad)
Date: Wed, 28 Mar 2001 00:12:03 +0200
Subject: Newbie question : string processing with RE
References: <mailman.985707278.14550.python-list@python.org>
Message-ID: <99r3br$s06$1@troll.powertech.no>

<Patrick.Bussi at space.alcatel.fr> wrote in message
news:mailman.985707278.14550.python-list at python.org...

>But when I try to read it again, it acts as if the index of
> a file had to be "rewinded" (e.g. as would do a "seek" on a "true" file).
Yes, that is correct, it is a file object (not a disk file in this case) and
it behaves like a file.
>I
> thought that f3.read was to be considered as a string.
Yes, and no, f3.read() returns a string that is the whole file from the
point in the file where you are to the end. But this means that the next
time you call read on the file it will be at the end and return the empty
string.

If you want to search the string twice you'll have to keep the string
returned by read, like this:

content = f3.read()

Then you can use content as many times as you like.




From sholden at holdenweb.com  Thu Mar 22 22:24:55 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 22 Mar 2001 22:24:55 -0500
Subject: Good Editor?
References: <3ab9a32b$1_1@news2.one.net> <3AB9FAC9.97EA420E@student.liu.se> <slrn9bjv7m.4qv.grey@teleute.dmiyu.org> <99d780$l2g$1@laurel.tc.umn.edu> <3ABA20F7.AB71D3B4@student.liu.se> <98qu6.166420$Z8.35271037@typhoon.southeast.rr.com> <mailman.985286969.17591.python-list@python.org>
Message-ID: <Upzu6.8997$G72.209371@e420r-atl1.usenetserver.com>

"Ken Seehof" <kens at sightreader.com> wrote in message
news:mailman.985286969.17591.python-list at python.org...
> Yeah, but I want to know: does it actually work?
> > "Rikard Bosnjakovic"
> > > Anyone sending unwanted advertising e-mail to my address will be
> > > charged $250 for network traffic and computing time. By extracting my
> > > address from this message or its header, you agree to these terms.
> >
> > love it...Love It...LOVE IT!
> >
> > Don
> ----------------------------------------------------
> Copyright (c) 2001 by Ken Seehof
> This document may not be distributed, copied,
> duplicated, or replicated in any form without
> express written permission by Ken Seehof.
> Express permission is hereby granted.
> kens at neuralintegrator.com
> ----------------------------------------------------
>
Now that's what I call an open source legal threat.

regards
 Steve

>
>




From tjreedy at udel.edu  Wed Mar 28 00:51:36 2001
From: tjreedy at udel.edu (Terry Reedy)
Date: Wed, 28 Mar 2001 00:51:36 -0500
Subject: Newbie question : string processing with RE
References: <mailman.985707278.14550.python-list@python.org>
Message-ID: <99rue5$brh$1@news.udel.edu>

I believe the indicated changes will work.

>  f3 = urllib.urlopen ("www.such-a-nice-webpage.com/index.html",
encoded_data)

f3 = urllib.urlopen ("www.such-a-nice-webpage.com/index.html",
encoded_data).read()

>  p=re.compile("window.open")
>  q=re.compile("document.location =" )
>  print p,'\n',q
>  t1=p.search(f3.read())
>  t2=q.search(f3.read())

t1=p.search(f3)
t2=q.search(f3)






From cobrien at Radix.Net  Fri Mar 16 10:25:21 2001
From: cobrien at Radix.Net (Cary O'Brien)
Date: 16 Mar 2001 10:25:21 -0500
Subject: Python style questions
Message-ID: <98tb91$9t9$1@saltmine.radix.net>

I have a couple of style questons regarding larger Python programs.
I'd be interested in comments.

1. I have a class that handles communications with a remote service.
   It is fairly complex, having to deal with timeouts and retransmits
   and the like.  The class definition is getting bigger than I'd like.
   Should I

    a. Live with it
    b. Factor into mix-in classes
    c. Define some methods outside of the class definition.
    d. ???


2. One thing I've done is to define classes used only by another class
   inside that class.  For example, if a class A needs to keep track of a
   bunch of "things", then the class definition for the "things" is inside
   of the class definition of A.  This is

   a. SOP
   b. Brilliant
   c. Idiotic

3. I really miss not having a "switch" or "case" statement.  Sniff.

4. I need to change things like "88aa99bb" (hex string) into integers.  Should I

   a. use expr("0x"+s) 
   b. not use such things
   c. ???


That's it for now.  I'm sure I'll have more questions as time goes on.
I guess what I'm really interested is how to create readable, maintainable,
code for complex classes.

-- cary





From tm at kernelconsult.com  Wed Mar 28 22:01:48 2001
From: tm at kernelconsult.com (TM)
Date: Thu, 29 Mar 2001 03:01:48 GMT
Subject: Which Python version is better on Win32
Message-ID: <20010329.3014838@mis.configured.host>

Hi,

I've seen recently that yhere were several releases of Python for the MS 
Windows platform:
ActivePython is one of them there's also PythonWin which is not branded, 
beOpenPython and maybe others.

Which one would you advise me to use and why?

Thanks in advance

TM


From jmg at ecs.soton.ac.uk  Wed Mar 28 13:12:34 2001
From: jmg at ecs.soton.ac.uk (Jacek Generowicz)
Date: 28 Mar 2001 19:12:34 +0100
Subject: Boost documentation.
References: <g0wv99x256.fsf@scumbag.ecs.soton.ac.uk> <g0u24dx087.fsf@scumbag.ecs.soton.ac.uk>
Message-ID: <g0r8zhwynh.fsf@scumbag.ecs.soton.ac.uk>

Jacek Generowicz <jmg at ecs.soton.ac.uk> writes:

> Well, I've managed to get an attempt past the compiler now, so I'll
> post it here for a critique.
> 
> It provides one function called `hmm', which takes a tuple (of
> integers), and should print out those integers. Unfortunately it
> thinks all the integers are zero (though it does get the correct
> number of them).

OK, i was not initialized to zero here:

>   for ( int i; i<tup.size(); i++ ) {

(But surely this calls the default constructor for int, which should
initialize it to zero ?)

> When I try to do something similar with lists rather than tuples, the
> get() method appears to be unavailable.

I'd still be interested in knowing how to do it with lists.

Jacek


From tim.one at home.com  Sat Mar 24 16:25:10 2001
From: tim.one at home.com (Tim Peters)
Date: Sat, 24 Mar 2001 16:25:10 -0500
Subject: [Python-Dev] Anyone using weakrefs?
In-Reply-To: <018e01c0b456$6710b830$e46940d5@hagrid>
Message-ID: <LNBBLJKPBEHFEDALKOLCGEBNJIAA.tim.one@home.com>

[Fred Drake]
> Is anyone out there playing with the weak references support yet?
> I'd *really* appreciate receiving a short snippet of non-contrived
> code that makes use of weak references to use in the documentation.

Alas, it's kinda like asking for a "short snippet" illustrating coroutines:
the need isn't apparent in tiny examples.

Still, I've been happily playing w/ this little module:

import weakref
id2obj_dict = weakref.mapping()

def remember(obj):
    id2obj_dict[id(obj)] = obj

def id2obj(id):
    return id2obj_dict.get(id, None)

In various introspective tools (like Cyclops.py), there's a need to keep
track of "other peoples'" class instances.  Before weakrefs, it was
impossible to do this without keeping the instances artificially alive.  Now
I can pass arbitrary instances to remember() above, then just use their ids
internally.  This doesn't keep the objects alive, but id2obj() above can
still get the objects back so long as they're still alive.

I suppose I could work with weakrefs directly instead of indirecting through
id()s, but most code of this nature I have is already mucking w/ ids anyway
(ints are small, fast, cheap and make great dict keys themselves).




From ws-news at gmx.at  Wed Mar  7 12:59:55 2001
From: ws-news at gmx.at (Werner Schiendl)
Date: Wed, 7 Mar 2001 18:59:55 +0100
Subject: Can anyone recomend a good intoduction to C...
References: <mailman.983894288.19965.python-list@python.org> <983896344.701456@newsmaster-04.atnet.at> <m3lmqibw0a.fsf@atrus.jesus.cam.ac.uk> <983911749.217622@newsmaster-04.atnet.at> <UScp6.1097$y6.196347@ruti.visi.com> <983916572.286918@newsmaster-04.atnet.at> <9852b8015jm@news1.newsguy.com> <983967284.246700@newsmaster-04.atnet.at> <985oev0216v@news1.newsguy.com>
Message-ID: <983988272.976222@newsmaster-04.atnet.at>

Thank you for your really thorough reply

I would agree that inspecting a C++ compiler to every extent is impossible.
But I think so are most other languages, apart from extremely simply ones
like assembler.

Do you think one can inspect a Python interpreter and make sure there is no,
not a single, error or security problem in it?
I would never expect that possible... nor for C... and of course not for C++

I do not think that a review (this is basically what human expecting source
code means) can prove code to be error free. Not for any language, otherwise
there would not be errors in production software.

The one way I could imagine to test an application for safety is to test all
possible data on the smallest part in software.
But of course this is not practically manageable for normal size
application.

How can you be sure the CPU is doing correct calculation? Think about that
Pentium FDIV bug years ago...
You can use a parallel executing different platform and compare each and
every result.
This does however not support recovery because you do not know which one is
correct.
That is not a way, because it would not be safe either.
So you can use three different architectures (and make sure that one is not
somehow copied/related from an other) and do a majority decision.
The alternative is to inspect the CPU (and other hardware) the same.

What I think is that there is no absolute safety.
You can test as much as you like and review as much as you like.
Never can you be completely sure that there can not be a problem.

Maybe that sounds pessimistic, but this is what I think.
Safety inspection just greatly reduces the number of problems that could
arise by doing much more thorough tests but in the end is limited as well.

>From the educational standpoint it does not matter if the standard library
uses templates or not.
the average user can use it in a cookbook fashion and does not need to know
how the machanics work.

If you need to inspect that as well, yes this is hard.
But if you need the full flexibility you'll end up with code the same
complex than the original.
If the application only needs more trivial containers and I cannot afford
using the template based standard ones, I can still resort to writing an
simpler one myself.

So in the end I think it is an issue, but probably not for the question the
original poster had.
There are other reasons as well, like customers that insist on plain C
because they do not have people experienced in C++.

Again, I really enjoyed your thorough reply

regards
werner




From chris.gonnerman at usa.net  Thu Mar  1 23:26:46 2001
From: chris.gonnerman at usa.net (Chris Gonnerman)
Date: Thu, 1 Mar 2001 22:26:46 -0600
Subject: compile() source code from a string
References: <3a9ee600.391990@News.CIS.DFN.DE>
Message-ID: <008a01c0a2d1$05507f40$a100000a@local>

There IS a syntax error in your code.  To wit:

----- Original Message -----
From: "Costas Menico" <costas at meezon.com>
Subject: compile() source code from a string
[ vicious hack ]
> However I would like to execute the following example from string
> variable:
>
> if x == 1:
>   print "1"
>   print "2"
>   print "3"
> elif x = 2:

ERROR   ^^^
Should be:

elif x == 2:

>   print 'A'
>   print 'B'
>   print 'C'
>
> print 1+2
>
[ brutal hack ]

Whenever you do something fancy like this and it doesn't work, you should
build a stripped example and test that.  Your code doesn't work in a string,
but hey, it doesn't work as a normal module file either.

I don't intend to be mean about it; I've been there too.

Attached is an example using your sample, corrected, which works for me.

-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: testcompile.py
URL: <http://mail.python.org/pipermail/python-list/attachments/20010301/ae939545/attachment.ksh>

From syver at NOSPAMcyberwatcher.com  Wed Mar 21 08:56:49 2001
From: syver at NOSPAMcyberwatcher.com (Syver Enstad)
Date: Wed, 21 Mar 2001 14:56:49 +0100
Subject: Html entities
Message-ID: <Sp2u6.3034$mh4.198592@news3.oke.nextra.no>

Is there an easy way to convert ISO Latin-1 characters that are above 127
ascii to their HTML, XML entity form?




From ransen_spam_me_not at nemo.it  Mon Mar  5 06:47:28 2001
From: ransen_spam_me_not at nemo.it (Owen F. Ransen)
Date: Mon, 05 Mar 2001 11:47:28 GMT
Subject: tokenizer.c crashes in 2.1b1
References: <3aa34d5c.1284101@news.newsguy.com>
Message-ID: <3aa37c7e.74688@news.newsguy.com>

If anyone more expert than I am wants to look at the
program which crashes I have a VC6 workspace, mostly
empty except for calls to the Python API...email
me removing the spam breaker part...
--
Owen F. Ransen
http://www.ransen.com/
Home of Gliftic & Repligator Image Generators


From claird at starbase.neosoft.com  Fri Mar 23 09:59:33 2001
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 23 Mar 2001 08:59:33 -0600
Subject: email address verification
References: <99d5qi$95l$1@solaria.cc.gatech.edu> <NFzu6.9043$G72.216162@e420r-atl1.usenetserver.com> <99fjv8$bhf$1@solaria.cc.gatech.edu>
Message-ID: <E006A65FA23EFF2C.2D74AEB35552E0B6.065EC0B29EF187C6@lp.airnews.net>

In article <99fjv8$bhf$1 at solaria.cc.gatech.edu>,
Holland King  <insanc at cc.gatech.edu> wrote:
			.
			.
			.
>despite the way it sounds i am not trying to set up a mass mailer program
>or trying to spam people. i am trying to set my departments system so
>that when an account expires it will parse the .forward file and use 
>addresses in there to make an alias. however, we don't like the idea of 
>sending mail to a list or to a non-existant address for obvious reason
>(especially consider how large we are).
			.
			.
			.
I don't understand your description.  In general,
comp.mail.misc might be a more productive place to
ask these questions than c.l.p.

Are you asking about verifying e-mail addresses *on
hosts you control*?  That's entirely different!
Well, it's significantly different, at least.
-- 

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


From sheila at spamcop.net  Sun Mar 18 21:09:01 2001
From: sheila at spamcop.net (Sheila King)
Date: Mon, 19 Mar 2001 02:09:01 GMT
Subject: Get Python2 Win32 working with Apache
References: <rKat6.3985$GD.2280188@typhoon.snet.net>
Message-ID: <ihqabto0eurnh30hlk6ah7j5umv9bj3ske@4ax.com>

On Sun, 18 Mar 2001 17:33:15 -0500, "Ryan" <ryan at ryanflynn.com> wrote in
comp.lang.python in article <rKat6.3985$GD.2280188 at typhoon.snet.net>:

:Any hints?

http://www.aota.net/ubb/Forum14/HTML/000218-1.html

--
Sheila King
http://www.thinkspot.net/sheila/
http://www.k12groups.org/



From sholden at holdenweb.com  Sun Mar  4 12:41:55 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Sun, 4 Mar 2001 12:41:55 -0500
Subject: Subclassing builtin types
References: <i6u4atslg4euv732pvaujk6u036kemcidf@4ax.com>
Message-ID: <obvo6.14838$1D5.510512@e420r-atl1.usenetserver.com>

"Daniel Klein" <danielk at aracnet.com> wrote in message
news:i6u4atslg4euv732pvaujk6u036kemcidf at 4ax.com...
> One of the classes I have to write  is merely a wrapper around a 'list'
object.
> A very abreviated example,
>
> class MyParams:
>       def __init__(self):
>             self.parameters = []
>       def append(self, object):
>             self.parameters.append(object)
>       def insert(self, i, object):
>             self.parameters.insert(i, object)
>       def del(self, i):
>             del self.parameters[i]
> etc, etc...
>
> This would allow me to do things like
>
> p = MyParams()
> p.append('foo')
>
> Since I know you are going to ask :-), there are two main reasons I need
to do
> this:
>
> 1) I need to ensure that a certain 'capacity' is not exceeded. I will
probably
> have class variable which is checked in all methods which add objects to
> 'parameters' and then raise an exception if 'capacity' is exceeded.
>
> 2) I can only allow certain types of objects to be added to the list, so
these
> too will be checked where necessary and appropriate exceptions raised.
>
> What I would _REALLY_ like to do is subclass list something like
>
> class MyParams(List):
>
> so that I can inherit all of the list methods and only override the ones I
need
> to. In Smalltalk this would be accomplished simply with
>
> OrderedCollection subclass: #MyParams
>
> I had actually thought of creating a generic List class that could be
reused in
> cases like this.
>
> So two questions:
>
> 1) Are wrapper classes for the builtin types a feasible (and pythonic)
thing to
> do?
>
Yes and yes: so feasible the library includes a UserList class which
provides esxactly what you seem to want: it wraps a list inside a class so
you can inherit from it!

> 2) Are there another ways to accomplish what I'm trying to do here without
> creating my own class hierarchy?
>
> Thanks for any and all help, you people have been a wonderful lot to deal
with
> during this new-language learning period.
>
> Daniel Klein
> Portland OR USA
>
Delegation is also a useful trick. I've just published an ftpStream class
which delegates most of its actions to an ftplib.FTP object by providing its
own __getattr__() method. Follow the public-domain Python link from
http://www.holdenweb.com/

regards
 Steve





From parkw at better.net  Thu Mar  1 15:05:58 2001
From: parkw at better.net (William Park)
Date: Thu, 1 Mar 2001 15:05:58 -0500
Subject: [Q] Question from Python tutorial
In-Reply-To: <hxxn6.6969$dL4.91744@vixen.cso.uiuc.edu>; from ylee12@uiuc.edu on Thu, Mar 01, 2001 at 01:38:02PM -0600
References: <Lgxn6.6964$dL4.91439@vixen.cso.uiuc.edu> <3A9EA472.F15AF847@cybermesa.com> <hxxn6.6969$dL4.91744@vixen.cso.uiuc.edu>
Message-ID: <20010301150558.A1284@better.net>

On Thu, Mar 01, 2001 at 01:38:02PM -0600, Young-Jin Lee wrote:
> > > class MyClass:
> > >     "A simple example class"
> > >     i = 12345
> > >     def f( x ):
> > >         return 'hello world'
> > >
> > > x = MyClass()
> > >
> > > x.counter = 1
> > > while x.counter < 10:
> > >     x.counter = x.counter * 2
> > > print x.counter
> > >       ^
> > > SyntaxError: invalid syntax
> > >
> > > There is no typo, I have no idea of where this SyntaxError came from.
> > >
> > > Thanks in advance. Any comment would be greatly appreciated.

On my computer, it goes something like this:

>>> x.counter = 1
>>> while x.counter < 10:
...     x.counter = x.counter * 2
...
>>> print x.counter
16


---William Park, Open Geometry Consulting, Linux/Python, 8 CPUs.



From scarblac at pino.selwerd.nl  Mon Mar 19 19:43:11 2001
From: scarblac at pino.selwerd.nl (Remco Gerlich)
Date: 20 Mar 2001 00:43:11 GMT
Subject: (no subject)
References: <mailman.985042764.18051.python-list@python.org>
Message-ID: <slrn9bd9nv.ac9.scarblac@pino.selwerd.nl>

Brad Jennings <bjennings at visionael.com> wrote in comp.lang.python:
> I want to execute an executable binary as thought through the Command-Line.
> Python provides some powerful methods in the os package that use very quirky
> syntax.  The two methods I've used are "startfile()" and "execv()". Neither
> have performed the action I  need.  
> 
> I'm just trying to execute "ss.exe" and pass it some args but the method
> syntax is laborious and prone to mistakes.  Do anyone know of  a better way
> to call an executable from a python script without using a function to do
> it?  

Without using a function it's not possible.

But what's the problem with using os.system ?!

import os
os.system("ss.exe some args")

-- 
Remco Gerlich


From aleaxit at yahoo.com  Mon Mar 12 11:56:33 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Mon, 12 Mar 2001 17:56:33 +0100
Subject: gah! I hate the new string syntax
References: <mailman.983485923.1136.python-list@python.org> <97sudi01i26@news1.newsguy.com> <98elrk$li8$1@nntp.Stanford.EDU> <98fc070thh@news1.newsguy.com> <98iqo2$56o$1@nntp.Stanford.EDU>
Message-ID: <98iv3s01sc2@news2.newsguy.com>

"Amit Patel" <amitp at Xenon.Stanford.EDU> wrote in message
news:98iqo2$56o$1 at nntp.Stanford.EDU...
> Alex Martelli <aleaxit at yahoo.com> wrote:
> | "Amit Patel" <amitp at Xenon.Stanford.EDU> wrote in message
> | news:98elrk$li8$1 at nntp.Stanford.EDU...
> |     [snip]
> | > What I mean to say is that I cannot write new methods on strings.  So
> |
> | Perfectly true, just as you cannot on dictionaries, lists, tuples,
> | file objects, and so on.  Only classes and extension-classes
> | are extensible (by inheritance and overriding), not types.
>
> Even if strings were a class, and I could extend it, it is NOT the
> thing to do to add new functionality like capitalize().  Suppose

If you wanted to MODIFY the functionality of the hypothetical string
class, so the new functionality applied to all objects of the class,
then you would indeed have to alter the class-object -- just as you
would have to alter the module-object to alter the functionality of
that module, in an exact parallel.


> The advantage of method functions like string.capitalize is that they
> do not require me to change something I can't change (the string
> type).

A class-object is just as alterable as a module-object -- neither
more, nor less.  (Whether it's _wise_ to do so, of course, is a
different issue -- but the ability is there in either case).

Type-objects are different, but there's no inherent reasons why
it should be any harder to perform
    someobj.capitalize = mycapfun
for someobj being a type-object, than for its being a module --
it can be seen as an accident of current implementation if only
one case out of the two possibilities is allowed.

> I can write capitalize(x) when x is a string.  I can't write
> x.capitalize() when x is a string.

But you would probably be writing string.capitalize in the
former case anyway -- unless you're advocating a "from
string import *"?  So, to add (e.g.) a capitalize function
to the string module, you'd have to set string.capitalize
rather than inject the 'capitalize' name into your global
(or local) namespace in other (dirtier?) ways.


> Defining methods is like running things as root.  Yes, you sometimes
> need to do it, but it's best to limit yourself to only those things
> that need superuser access.  Run as much as you can as non-root.  And
> IMO that goes for methods and functions too -- string type methods are
> "root" (where "root" = "Guido & pals") and have access to the
> internals of a string, whereas functions on strings are user-level and
> go through the abstractions.

Focusing on the polymorphism needs, I see the issue in the mirror
way from you: it's a _weakness_ (of Python and most other languages)
that I can't add methods 'post-facto, from the outside' to existing
objects -- it impedes the most natural, elegant, typeswitch-free way
to add functionality.  If you pursued your "capitalize as a function"
idea, you'd be unable to properly capitalize Unicode (e.g.) without
a typeswitch -- a very unadvisable construct.  Something like Haskell's
typeclasses, on the other hand, would provide the best of both worlds
(multimethods would be an even more powerful alternative) -- let the
compiler/interpreter/runtime do the dispatching, and just code in
the most natural, polymorphic way.

If you have to work with both existing and new objects, some of
which provide a method while others don't, then a wrapper function
(or class, &c) is indeed best -- and inside it you can do the
weakest (safest, least-unadvisable) form of typeswitching, e.g.:

def capitalize(self):
    try: return self.capitalize()
    except AttributeError: pass
    # proceed implementing the default way of
    # capitalization

If you can perform your capitalization-default more effectively
in a mixin class (this is often true, since keeping state can
be quite helpful), then extend this to:

def capitalize(self):
    try: return self.capitalize()
    except AttributeError: pass

    # if it's a class we can add our mixin once & for all:
    try: self.__class__.__bases__.append(MixinCapitalizer)
    except AttributeError: pass
    else: # good, we were able to add it, so init & use it:
        MixinCapitalizer.__init__(self)
        return self.capitalize()

    # proceed implementing the default way of
    # capitalization as a function, sigh


The ability to perform such shehanigans does argue for using
functions _from client-code that needs to be very robust in
dealing with all sorts of weird objects including legacy ones_,
but it does NOT argue against adding suitable methods on the
implementation side -- anything BUT, in fact... where the
method is right on the object we get the fastest path through
the wrapper-function, even in client-code that needs to go
through such wrappers for legacy-support reasons.


Alex





From Sibylle.Koczian at Bibliothek.Uni-Augsburg.de  Fri Mar 23 05:02:59 2001
From: Sibylle.Koczian at Bibliothek.Uni-Augsburg.de (Sibylle Koczian)
Date: Fri, 23 Mar 2001 11:02:59 +0100
Subject: win32net, documentation
References: <3AB8CD29.C69CC1E2@Bibliothek.Uni-Augsburg.de> <3AB949C4.1040501@ActiveState.com>
Message-ID: <3ABB1F53.EEEEB8C1@Bibliothek.Uni-Augsburg.de>

Thank you - I've got that to work after some trial and error.

Mark Hammond wrote:
> 
> > I want to do the Python equivalent of
> > net use k: \\server\share
> >
> I suggest you use win32wnet - that will work on Win9x too.
> 
That's not an issue for me - are there other advantages? It seems a
little simpler, though - and I _did_ get it to work, which should be
enough, but I'm curious (and there isn't anything in your book about it,
as far as I remember?).

> Look in win32\demos\win32wnet for some examples...
> 
Only one - which got me started, anyway.

Thank you again,
Sibylle

-- 
----
Dr. Sibylle Koczian
Universitaetsbibliothek , Abt. Naturwiss.
D-86135 Augsburg

Tel.: (0821) 598-2400, Fax : (0821) 598-2410
e-mail : Sibylle.Koczian at Bibliothek.Uni-Augsburg.DE


From kens at sightreader.com  Tue Mar 13 05:54:45 2001
From: kens at sightreader.com (Ken Seehof)
Date: Tue, 13 Mar 2001 02:54:45 -0800
Subject: This math scares me
References: <3aad4062.1290021@News.CIS.DFN.DE> <20010312170356.B3513@better.net>
Message-ID: <009301c0abac$05352ef0$a325fea9@his>

You're not alone, but there is a good explanation for what you see, and it
is not generally considered a bug in python, though some would like to
change python to give more visually correct results.

This is also true of many other programming languages, such as C and C++.
The reason is that the computer stores your number in binary.  Turns out
that 10.55 (or 0.55) does not have an exact representation in binary,
therefore you get a "rounding error".  To give you an idea of what I am
talking about, consider that in decimal, you can't represent 1/3 exactly
(you get 0.33333333...).

Let's take a simpler example:

>>> 0.2
0.20000000000000001

What happened here?  Well 0.2 = 1/5, which works fine in decimal.  When you
convert that to binary, it looks like this:
0.2 (decimal)  =  0.0011001100110011001100110011001100110011001100110011...
(binary).

The reason your desk calculator gets the right result is that it represents
numbers in decimal internally.  If python did the same, it would run
significantly slower than it does because most computers prefer to do math
in binary.  However, switching to decimal floating point is a proposed
change for a future version of python, along with making 1/2 = 0.5 (or some
other equivalent answer (right now, 1 / 2 = 0 (try it) ) ) and other stuff.

Keep in mind that any floating point system will have rounding errors.
Doing math in decimal merely hides the wierd unexpected ones like the one
you found (at the cost of slowing things down).  Some people propose that
python keeps fractional numbers in rational form (i.e. A/B).  This would
completely solve the rounding error problem (for division and
multiplication, that is), but would make python really really really slow.
There are various other compromises but there is unfortunately no good
solution to this dilemma at all.

These numerical issues cause endless debates among python people.  As you
will see, there will probably be a stream of endless innane babble in
response to this message.  I think there's a SIG or a PEP or something
(couldn't find it) where this debate is supposed to be instead of here.

----- Original Message -----
From: "William Park" <parkw at better.net>
To: <costas at springmail.com>
Cc: <python-list at python.org>
Sent: Monday, March 12, 2001 2:03 PM
Subject: Re: This math scares me


> On Mon, Mar 12, 2001 at 09:34:58PM +0000, costas at springmail.com wrote:
> > Ok, I can see maybe division having problems. But why does addition of
> > the two numbers below:
> >
> > 5.01+5.54
> >
> > give me this?
> >
> > 10.550000000000001
>
> It's called 'Round-off error'.
>
> ---William Park, Open Geometry Consulting, Linux/Python, 8 CPUs.
>
> --
> http://mail.python.org/mailman/listinfo/python-list
>




From francois_forest at yahoo.fr  Fri Mar 16 09:00:03 2001
From: francois_forest at yahoo.fr (Francois Forest)
Date: Fri, 16 Mar 2001 14:00:03 GMT
Subject: Python with the VBA recording capabilities
Message-ID: <3AB21C3E.95DAB3D2@yahoo.fr>

I compared Python to VBA (not for their syntax):

    VBA gives the programming capability to Applications

    Python gives the programming capability at the OS level
    (as shell scripts traditionally did )

VBA has the RECORD capability that transform any sequence of mouse clics
into a program that can be edited and then adapted. This recording
capability is amazingly efficient.

Can we imagine to add this feature to any current GUI (kde and others) ?



From fredrik at pythonware.com  Thu Mar 29 08:29:20 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Thu, 29 Mar 2001 13:29:20 GMT
Subject: Defining class files
References: <3ac33343$0$12248$ed9e5944@reading.news.pipex.net>
Message-ID: <QMGw6.4159$4N4.847815@newsc.telia.net>

Neil Benn wrote:
> However, I then tried to write the class as an external text file,
> using:-
>
> class YourClass:
>     "A simple example class"
>     i = 12345
>     def f(x):
>         return 'hello world'

that's a module containing a class, not a "class file"

as you noticed, "import YourClass" gives you the module.

to get at the class itself, refer to it as "YourClass.YourClass"
(the YourClass object in the YourClass module).

you can also write "from YourClass import YourClass".

> The text file is saved as YourClass.py - is this the problem, should
> class files have different terminaters in their filename??

there are no such thing as a "class file" in Python.  just modules...

for more info, see chapter 6 in the tutorial:

    http://www.python.org/doc/current/tut/node8.html

Cheers /F




From phrxy at csv.warwick.ac.uk  Thu Mar 15 19:05:02 2001
From: phrxy at csv.warwick.ac.uk (John J. Lee)
Date: Fri, 16 Mar 2001 00:05:02 +0000
Subject: tiny extension crashes
In-Reply-To: <zUXr6.571195$U46.16976599@news1.sttls1.wa.home.com>
References: <Pine.SOL.4.30.0103141957240.10526-100000@mimosa.csv.warwick.ac.uk> <zUXr6.571195$U46.16976599@news1.sttls1.wa.home.com>
Message-ID: <Pine.SOL.4.30.0103152349300.26347-100000@mimosa.csv.warwick.ac.uk>

On Thu, 15 Mar 2001, Greg Jorgensen wrote:

> You have an uninitialized pointer, intp, in your iround() function. Do this:
[...]

Oops, thanks.

What I still don't get, though, is why it was crashing even when only the
hcf()  function was being imported and called by my (pure) Python code.
Surely if it doesn't get called, the uninitialised pointer can't scribble
over anything?

> Since you posted your source code, let me suggest an alternative
> implementation of array_max:
>
> static long array_max(PyArrayObject *array, int absolute)
> {
>     long *a = array->data;
>     long stride = array->strides[0];
>     long max = 0;
>
>     if (absolute) {
>         for (i = 0; i < array->dimensions[0]; i++) {
>             if (labs(*a) > max)
>                 max = labs(*a);
>             a += stride;
>         }
>     }
>     else {
>         for (i = 0; i < array->dimensions[0]; i++) {
>             if (*a > max)
>                 max = *a;
>             a += stride;
>         }
>     }
>     return max;
> }
>
> The test on absolute only happens once, not at every loop iteration. (I
> would make this two functions, one that works on the actual values and one
> that works on the absolute values.)

Yeah, so would I, but my C is so rusty that I couldn't be bothered with
the effort (not that my knowledge of C was ever clean and shiny).  :)
That's what these interpreted languages do to you -- I'm spoilt.

> The array->data[] index calculation is
> changed to pointer addition. And I used the labs() function rather than the
> unnecessary floating-point fabs() version. And I changed max to a long, and
> made the function return a long.

All good tips, thanks.  (Although as it happens, in this particular case,
I had intended to replace that function with the built-in Numeric max(),
when I had got round to looking up how to call Python functions from C.
In fact I'll probably scrap all of these functions eventually -- they're
only there because they were the simplest functions I had lying around,
and I knew I'd certainly get tangled up in C if I picked something more
complicated for my first extension.)


John



From thomas.heller at ion-tof.com  Fri Mar 30 14:31:47 2001
From: thomas.heller at ion-tof.com (Thomas Heller)
Date: Fri, 30 Mar 2001 21:31:47 +0200
Subject: Application distribution with python
References: <mailman.985964564.7170.python-list@python.org> <tc9db41hcu8417@corp.supernews.com>
Message-ID: <9a2mv5$3995a$1@ID-59885.news.dfncis.de>

> >
> > Can any one knows what are files to ship to client machine or
> > list of distribution files.
>
> http://starship.python.net/crew/gmcm/distribute.html
>
>
and http://starship.python.net/crew/theller/py2exe/





From loewis at informatik.hu-berlin.de  Sun Mar 11 15:55:33 2001
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 11 Mar 2001 21:55:33 +0100
Subject: Python 2.0 and Tcl/Tk 8.3.2?
References: <3A997A0A.67CA82E1@UTS.Itron.com> <3AAAFFE5.15A7A878@python.net>
Message-ID: <j47l1wm3ei.fsf@informatik.hu-berlin.de>

"Thomas A. Bryan" <tbryan at python.net> writes:

> If Python works as well with Tcl/Tk 8.3 as it does with Tcl/Tk 8.0,
> I wonder why the default Python-2.0 still defaults to the older
> version.  Does this line get updated to the latest version of Tcl/Tk
> that was known to work at the time of release, or is it frozen at
> 8.0 for the time being?
>
> From Modules/Setup.in:
> # *** Uncomment and edit to reflect your Tcl/Tk versions:
> #       -ltk8.0 -ltcl8.0 \

Well, there is a comment in the beginning of the line - if you remove
the comment, you have to adjust the line; this is what you do with all
other lines as well, don't you?

The Linux RPM of Python 2.0 required Tcl 8.0 since that is what was
available on Redhat 6; most likely, this is also the reason that the
Setup refers to 8.0 'by default' (if you look carefully, you'll see
that 'by default', tkinter is not built at all).

In any case, Python 2.1 will use distutils for building _tkinter; it
will search for Tcl 8.4, 8.3, 8.2, 8.1, 8.0 in that order.

Regards,
Martn



From MarkH at ActiveState.com  Wed Mar 14 17:52:01 2001
From: MarkH at ActiveState.com (Mark Hammond)
Date: Wed, 14 Mar 2001 22:52:01 GMT
Subject: [Q]: How can I get the logged in users username on Win32?
References: <3aaf7ab4.3152221@news.is.co.za>
Message-ID: <3AAFF55A.9090107@ActiveState.com>

nomad wrote:

> Howzit,
> 
> After searching through the Python Docs, I found that on Un*x I could
> use getpass.getuser(), but there is no equivalent for the Win32
> platform.
> 
> In other words, how can I get the username in Windows?
> 
> As a side note, how can I enable Windows Scripting Host support for
> python - there is mention of python in the original WSH docs - and if
> I can use python scripts in WSH how would I call the built-in
> Wscript.<methods>
> 
> TIA
> 
> nomad

win32api.GetUserName()

For WSH examples, check out the win32com\test directory.

Mark.



From shaleh at valinux.com  Thu Mar 15 15:19:11 2001
From: shaleh at valinux.com (Sean 'Shaleh' Perry)
Date: Thu, 15 Mar 2001 12:19:11 -0800 (PST)
Subject: Dynamic append in a dictionnary object ...
In-Reply-To: <3ab11de3$0$389$446d705c@news.skynet.be>
Message-ID: <XFMail.20010315121911.shaleh@valinux.com>

On 15-Mar-2001 Furax wrote:
> Hi,
> I'm pretty new to Python.
> I cannot figure out, reading the documentation, how I can dynamicaly add a
> new pair in a dictionnary ?
> 

if not foo.has_key(key):
    foo[key] = value





From mwh21 at cam.ac.uk  Fri Mar  2 08:00:18 2001
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: Fri, 2 Mar 2001 13:00:18 +0000 (GMT)
Subject: python-dev summary 2001-02-15 - 2001-03-01
Message-ID: <Pine.LNX.4.10.10103021258550.18596-100000@localhost.localdomain>

Thanks for all the positive feedback for the last summary!

 This is a summary of traffic on the python-dev mailing list between
 Feb 15 and Feb 28 (inclusive) 2001.  It is intended to inform the
 wider Python community of ongoing developments.  To comment, just
 post to python-list at python.org or comp.lang.python in the usual
 way. Give your posting a meaningful subject line, and if it's about a
 PEP, include the PEP number (e.g. Subject: PEP 201 - Lockstep
 iteration) All python-dev members are interested in seeing ideas
 discussed by the community, so don't hesitate to take a stance on a
 PEP if you have an opinion.

 This is the second python-dev summary written by Michael Hudson.
 Previous summaries were written by Andrew Kuchling and can be found
 at:

   <http://www.amk.ca/python/dev/>

 New summaries will appear at:

  <http://starship.python.net/crew/mwh/summaries/>

 and will continue to be archived at Andrew's site.

   Posting distribution (with apologies to mbm)

   Number of articles in summary: 400

       |                         ]|[                            
       |                         ]|[                            
    60 |                         ]|[                            
       |                         ]|[                            
       |                         ]|[                            
       |                         ]|[                     ]|[    
       |                         ]|[     ]|[             ]|[    
       |                         ]|[     ]|[             ]|[    
    40 |                         ]|[     ]|[             ]|[ ]|[
       |                         ]|[     ]|[             ]|[ ]|[
       |     ]|[                 ]|[ ]|[ ]|[             ]|[ ]|[
       | ]|[ ]|[                 ]|[ ]|[ ]|[             ]|[ ]|[
       | ]|[ ]|[                 ]|[ ]|[ ]|[             ]|[ ]|[
       | ]|[ ]|[                 ]|[ ]|[ ]|[             ]|[ ]|[
    20 | ]|[ ]|[                 ]|[ ]|[ ]|[             ]|[ ]|[
       | ]|[ ]|[             ]|[ ]|[ ]|[ ]|[         ]|[ ]|[ ]|[
       | ]|[ ]|[             ]|[ ]|[ ]|[ ]|[         ]|[ ]|[ ]|[
       | ]|[ ]|[             ]|[ ]|[ ]|[ ]|[ ]|[     ]|[ ]|[ ]|[
       | ]|[ ]|[     ]|[     ]|[ ]|[ ]|[ ]|[ ]|[     ]|[ ]|[ ]|[
       | ]|[ ]|[     ]|[ ]|[ ]|[ ]|[ ]|[ ]|[ ]|[     ]|[ ]|[ ]|[
     0 +-033-037-002-008-006-021-071-037-051-012-002-021-054-045
        Thu 15| Sat 17| Mon 19| Wed 21| Fri 23| Sun 25| Tue 27|
            Fri 16  Sun 18  Tue 20  Thu 22  Sat 24  Mon 26  Wed 28

 A slightly quieter week on python-dev.  As you can see, most Python
 developers are too well-adjusted to post much on weekends.  Or
 Mondays.

 There was a lot of traffic on the bugs, patches and checkins lists in
 preparation for the upcoming 2.1b1 release.


    * backwards incompatibility *

 Most of the posts in the large spike in the middle of the posting
 distribution were on the subject of backward compatibility.  On of
 the unexpected (by those of us that hadn't thought too hard about it)
 consequences of nested scopes was that some code using the dreaded
 "from-module-import-*" code inside functions became ambiguous, and
 the plan was to ban such code in Python 2.1.  This provoked a storm
 of protest from many quarters, including python-dev and
 comp.lang.python.  If you really want to read all of this, start
 here:

  <http://mail.python.org/pipermail/python-dev/2001-February/013003.html>

 However, as you will know if you read comp.lang.python, PythonLabs
 took note, and in:

  <http://mail.python.org/pipermail/python-dev/2001-February/013125.html>
 
 Guido announced that the new nested scopes behaviour would be opt-in
 in 2.1, but code that will break in python 2.2 when nested scopes
 become the default will produce a warning in 2.1.  To get the new
 behaviour in a module, one will need to put

    from __future__ import nested_scopes

 at the top of the module.  It is possible this gimmick will be used
 to introduce further backwards compatible features in the future.


    * obmalloc *

 After some more discussion, including Neil Schemenauer pointing out
 that obmalloc might enable him to make the cycle GC faster, obmalloc
 was finally checked in.

 There's a second patch from Vladimir Marangoz implementing a memory
 profiler.  (sorry for the long line)

  <http://sourceforge.net/tracker/index.php?func=detail&aid=401229&group_id=5470&atid=305470>

 Opinion was muted about this; as Neil summed up in:

  <http://mail.python.org/pipermail/python-dev/2001-February/013205.html>

 noone cares enough to put the time into it and review this patch.
 Sufficiently violently wielded opnions may swing the day...


    * pydoc *

 Ka-Ping Yee checked in his amazing pydoc.  pydoc was described in

  <http://mail.python.org/pipermail/python-dev/2001-January/011538.html>

 It gives command line and web browser access to Python's
 documentation, and will be installed as a separate script in 2.1.


    * other stuff *

 It is believed that the case-sensitive import issues mentioned in the
 last summary have been sorted out, although it will be hard to be
 sure until the beta.

 The unit-test discussion petered out.  Nothing has been checked in
 yet.

 The iteraators discussion seems to have disappeared.  At least, your
 author can't find it!

Cheers,
M.




From qrczak at knm.org.pl  Sat Mar 31 16:29:17 2001
From: qrczak at knm.org.pl (Marcin 'Qrczak' Kowalczyk)
Date: 31 Mar 2001 21:29:17 GMT
Subject: list vs tuple
References: <hGlw6.782$p5.3210@news1.rivrw1.nsw.optushome.com.au><3AC20CD3.265A2D94@alcyone.com><ru69cto61jtmrqsf809e38hb3pbm2lhi7g@4ax.com><bC1x6.4319$4N4.905946@newsc.telia.net><71q9ct4le63anefmiajj5cbh3mg04qu1ga@4ax.com><mailman.985991202.5578.python-list@python.org><7Bfx6.989$p5.3791@news1.rivrw1.nsw.optushome.com.au><Ukhx6.4421$4N4.950603@newsc.telia.net> <mailman.986049281.32365.python-list@python.org> <uGqx6.1052$p5.3997@news1.rivrw1.nsw.optushome.com.au> <slrn9cchb4.3oa.qrczak@qrnik.zagroda> <aPrx6.1056$p5.4182@news1.rivrw1.nsw.optushome.com.au>
Message-ID: <slrn9ccj1d.484.qrczak@qrnik.zagroda>

Sat, 31 Mar 2001 21:16:54 GMT, deadmeat <root@[127.0.0.1]> pisze:

> > thus it is consistent.
> 
> The syntax is not.

It's not different than Pascal. It is self-consistent.

Please show an example which proves that
   a = b
ever does something different than making the name 'a' refer to object
to which 'b' refers.

-- 
 __("<  Marcin Kowalczyk * qrczak at knm.org.pl http://qrczak.ids.net.pl/
 \__/
  ^^                      SYGNATURA ZAST?PCZA
QRCZAK


From tonetheman at hotmail.com  Thu Mar 22 00:22:17 2001
From: tonetheman at hotmail.com (Tonetheman)
Date: Thu, 22 Mar 2001 05:22:17 GMT
Subject: Dynamically reloading extension modules
References: <007f01c0b252$852bb370$04090a0a@upcast.com> <20010321145901.M24341@trufflehunter.avalongroup.net> <00e201c0b25f$b95d4a70$04090a0a@upcast.com> <20010321154600.Q24341@trufflehunter.avalongroup.net> <mailman.985218751.8244.python-list@python.org>
Message-ID: <d_fu6.103265$lk6.8537437@typhoon.midsouth.rr.com>

Ok, I am an idiot.

Sorry about the last post.

I see that the beos module is only using this to cleanup on exit. I will
have to poke around some more to see how you could get at a del module;

Sorry for the bad advice.


"Ken Seehof" <kens at sightreader.com> wrote in message
news:mailman.985218751.8244.python-list at python.org...
>
> > On Wed, Mar 21, 2001 at 03:36:15PM -0800, Ken Seehof wrote:
> > > Not exactly.  I didn't state the problem very clearly.  The
> > > problem is that while an extension module is loaded, the
> > > module file (pyd on windows) is read-only.
> > >
> > > At least this is true on windows.  On unix, I think maybe you
> > > can rename the module file, rebuild it, and do a reload(module).
> > > Is this true?  Windows doesn't allow you to rename or rewrite
> > > a dll while it is in use, so something like FreeLibrary would be
> > > necessary.
> >
> > Sorry, I can't answer Windows questions. I haven't had to run
> > Windows in years.
> >
> > But I would still think that reload() should work as long as
> > you are using python modules, *not* c/c++ modules.
>
> I -am- talking about c/c++ modules (which I have been referring
> to as "extension" modules).  Hence there's a bit more to it than
> reload().
>
> I want to recompile and reload a c/c++ extension module from
> within a continuously running python program.
>
> > Beazley has no disclaimers about reload working only on
> > specific platforms.
> >
> > There is a caveat, objects created with the old module won't be
> > updated.
>
>
>




From nperkins7 at home.com  Fri Mar 23 22:41:25 2001
From: nperkins7 at home.com (Nick Perkins)
Date: Sat, 24 Mar 2001 03:41:25 GMT
Subject: newbie IDLE, Tkinter
Message-ID: <FHUu6.68719$A6.14534907@news4.rdc1.on.home.com>

I have been playing around with Python and Tkinter, using the IDLE prompt
and edit windows.

I am a little confused about what happens when I run my tkinter programs,
then close them (by closing their window).  Sometimes I come back to the
prompt, sometimes the interactive window just seems to hang, and sometimes
the whole IDLE program just quits.

Can someone explain the best way to interactively edit and test a tkinter
program with IDLE?  Do these window-closing behaviours depend on how I
actually run my program?





From korpo at 01019freenet.de  Thu Mar 22 17:24:09 2001
From: korpo at 01019freenet.de (Oliver Korpilla)
Date: Thu, 22 Mar 2001 23:24:09 +0100
Subject: python, perl, lisp, tcl / got a decision to make, can you give me a 
 view point please ? and a joke for python users :)
References: <3aba3076.14395287@news.vif.com>
Message-ID: <3ABA7B89.3A542D1C@01019freenet.de>

You really got the point by publishing the joke...

Ciao, Korpo



From tchur at optushome.com.au  Thu Mar 22 13:10:03 2001
From: tchur at optushome.com.au (Tim Churches)
Date: Fri, 23 Mar 2001 05:10:03 +1100
Subject: python, perl, lisp, tcl / got a decision to make, can you give me a 
 view point please ? and a joke for python users :)
References: <3aba3076.14395287@news.vif.com>
Message-ID: <3ABA3FFB.50CF5714@optushome.com.au>

Yvon Boulianne wrote:
> 
> Hello,  i'm a bit confused with so many language ;-)
> theres the story, i like to learn a programming language for 2 reasons
...
> 2- i want to learn to keep my mind trained, i'm a mystic and i want to
> make this mental grow in ability ;-)
...

Well, assuming your mysticism extends to clairvoyance, Python is
definitely the language for you. After many seances and much work with
the ouija board, GvR and his gang have decided to include support for
the following syntax:

from __future__ import *

Crystal-ball-gazingly yours,

Tim C



From info at pythonware.com  Mon Mar 26 07:52:32 2001
From: info at pythonware.com (PythonWare)
Date: Mon, 26 Mar 2001 14:52:32 +0200
Subject: ANN: PythonWorks Pro 1.2
Message-ID: <021001c0b5f4$367a3440$0900a8c0@SPIFF>

a pythonware newsflash:

    Secret Labs AB has released a new, much improved version of
    their integrated development environment for Python.

    PythonWorks is a rapid application development tool, designed
    for efficient development in Python. PythonWorks includes editors,
    a debugger, a user-interface design tool, project browser,
    deployment tools, integrated on-line help, and a scriptable
    environment.

    PythonWorks is well suited for many tasks, including:

    + Cross-platform system administration (system automation)
    + Prototyping of user interfaces and full applications
    + Building test frameworks
    + Rapid development and deployment of complete applications

    New features in 1.2 include improved support for Python 2.0,
    support for multiple pads and views, directory workspaces, lots
    of editor improvements, basic class browser, and improved EXE
    deployment support.

    Beginning with version 1.2, Secret Labs is also offering OEM
    versions of PythonWorks, for customers who want to provide a
    powerful IDE environment as part of their own Python-powered
    applications, or want a customized version of PythonWorks for
    use in their organization.

    To learn more, visit the PythonWorks product site:

        http://www.pythonworks.com

    You can also mail us at info at pythonware.com

regards,
the pythonworks team




From sholden at holdenweb.com  Wed Mar  7 21:20:35 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Wed, 7 Mar 2001 21:20:35 -0500
Subject: Can Python replace Visual Basic? Should it?
References: <5DZo6.11126$hn5.1578410@news1.rdc1.mb.home.com> <mailman.984010931.26257.python-list@python.org>
Message-ID: <j3Cp6.30579$ML1.1553640@e420r-atl2.usenetserver.com>

"Don Dwiggins" <dwig at advancedmp.net> wrote in message
news:mailman.984010931.26257.python-list at python.org...
[ ... ]
> I'd say go for it -- get the Python on Windows book and convert one of
your
> VB, Powerbuilder, or Delphi projects to Python and see what you think (or
if
> you're adventurous, just use the documentation included in the
> distribution).  I converted one of our ASP pages to Python from VBScript
to
> see if it would be faster (it wasn't), and was pleasantly surprised at how
> easy it was.
>
The main problem here is the overhead of starting the Python interpreter
(and recompiling the top-level script) each time a page is called. I've done
some work with Xitami to develop long-running web processes to server HTML
from realational data, and the speedup was, to say the least, startling.

My results may not translate to IIS, though, since I did this on the road
using Win98 and PWS. YMMV.

Of course, structuring your scripts so they just call a function in an
imported module will win you some performance, since then most everything is
pre-compiled into a .pyc.

> I should say, though, that we haven't used any of the GUI toolkits --
maybe
> someone else can recommend the best one to use when migrating from VB
Forms.
>
I haven't yet seen anything that stirs me, but it's early days yet.
BlackAdder looks promising, though they could certainly use an interface
design consultant on their own product...

regards
 Steve





From jurgen.defurne at philips.com  Fri Mar 16 02:12:50 2001
From: jurgen.defurne at philips.com (jurgen.defurne at philips.com)
Date: Fri, 16 Mar 2001 08:12:50 +0100
Subject: This math scares me
Message-ID: <0056900016651646000002L062*@MHS>

grante at visi.com@SMTP at python.org on 15/03/2001 19:20:46
Sent by:	python-list-admin at python.org
To:	python-list at python.org@SMTP
cc:	 
Subject:	Re: This math scares me
Classification:	

In article <98qv55$f5d$1 at news.udel.edu>, Terry Reedy wrote:

>So, .1% gets represented as the binary equivalent of, perhaps,
>.0010000000000000001.
>So the interest on 2,000,000,000 pennies is calculated as
>2000000.0000000002 instead of
>2000000 pennies.  When rounded off, they are the same.  To me, this is, as
>I claimed. 'well enough' (for many purposes, at least), even though not
>'exact'.

If that's good enough for you, then go ahead and use FP.

But, the people who actually write financial SW for a living
seem to think it's not good enough.  You would have to talk to
one of them to find out the exact situations where FP proved
insufficient.


Try summing all financial operations of one day together, or
better, all operations involving percentages and then adding them
together : cumulative errors.

Jurgen



From kalle at gnupung.net  Sat Mar  3 16:55:45 2001
From: kalle at gnupung.net (Kalle Svensson)
Date: Sat, 3 Mar 2001 22:55:45 +0100
Subject: Opposite of PyLong_FromString
In-Reply-To: <slrn9a2gnv.hs6.qrczak@qrnik.zagroda>; from qrczak@knm.org.pl on Sat, Mar 03, 2001 at 07:16:15PM +0000
References: <slrn9a2gnv.hs6.qrczak@qrnik.zagroda>
Message-ID: <20010303225545.A2165@apone.network.loc>

Sez Marcin 'Qrczak' Kowalczyk:
> What is the best way to convert a Python's Long to a C string
> (matching "-?[0-9]+")?

Perhaps something like
char *s = PyString_AsString(longobj->ob_type->tp_str(longobj));
or maybe use PyString_Format ?

I don't know what's idiomatically correct, though.

Peace,
  Kalle
-- 
Email: kalle at gnupung.net     | You can tune a filesystem, but you
Web: http://www.gnupung.net/ | can't tune a fish. -- man tunefs(8)
PGP fingerprint: 0C56 B171 8159 327F 1824 F5DE 74D7 80D7 BF3B B1DD
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 240 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/20010303/3708fe17/attachment.sig>

From someone at microsoft.com  Tue Mar 20 12:24:43 2001
From: someone at microsoft.com (Miki)
Date: Tue, 20 Mar 2001 12:24:43 -0500
Subject: Python2.0 rpm's don't have IDLE?
References: <OaEt6.2328$Up.121023@sea-read.news.verio.net>
Message-ID: <UpMt6.831$z_.35480631@news.sisna.com>

I had the same problem. What I did was get Python 2.0 as a tarball and
install it that way. Even though IDLE doesn't install, it is still in the
"Tools" subdirectory. I just copied the entire Tools folder to
/user/local/lib/python and changed the shell script "idle" in /usr/bin to
the new path of "idle.py".
This worked for me on Redhat 6.1
Miki

"Martin Brown" <mjbjr at best.com> wrote in message
news:OaEt6.2328$Up.121023 at sea-read.news.verio.net...
> I've downloaded and installed the 3 Python2.0 rpm's retrieved:
>
>    BeOpen-Python-2.0-1.i386.rpm
>    Tkinter-2.0-8.0-3.i386.rpm
>    expat-1.1-1.i386.rpm
>
> I've checked the installation directories, but there doesn't seem to be a
> copy of IDLE.  The web site says IDLE 0.6 is included in Python2.0.  Also,
> as a separate package, apparently, only version 0.5 is available.
>
> Suggestions on how to get version 0.6 of IDLE into my Python2.0
> installation greatly appreciated.
>
> --
>                             - Martin J. Brown, Jr. -
>                             - mjbjr at beaudesign.com -
>
>     PGP Public Key ID: 0xCED9BD8A  Key Server:
http://www.keyserver.net/en/




From root at [127.0.0.1]  Sat Mar 31 16:24:33 2001
From: root at [127.0.0.1] (deadmeat)
Date: Sat, 31 Mar 2001 21:24:33 GMT
Subject: list vs tuple
References: <mailman.986066384.30712.python-list@python.org> <d1rx6.1053$p5.4024@news1.rivrw1.nsw.optushome.com.au> <slrn9cciaj.3ug.qrczak@qrnik.zagroda>
Message-ID: <lWrx6.1057$p5.4152@news1.rivrw1.nsw.optushome.com.au>

> THERE IS NO DIFFERENCE.

What happens to the objects is not the point, what happens to what I put
into a and b is the point.

>>> a = 1
>>> b = a
>>> a = 2
>>> b
1

>>> a = [1,2,3]
>>> b = a
>>> a[0] = 0
>>> b
[0, 2, 3]

Are you saying there is no difference (IN THE *RESULT*) between these?  Are
you able to read?

What goes on in the background is NOT THE POINT. I only care about the
values I put into them, and what happens after b = a.






From sdm7g at virginia.edu  Wed Mar 14 12:45:27 2001
From: sdm7g at virginia.edu (Steven D. Majewski)
Date: Wed, 14 Mar 2001 12:45:27 -0500 (EST)
Subject: Who's minister of propaganda this week?
In-Reply-To: <3AAFA861.C773D977@san.rr.com>
Message-ID: <Pine.NXT.4.21.0103141238460.229-100000@localhost.virginia.edu>


On Wed, 14 Mar 2001, Darren New wrote:

> Phlip wrote:
> > Ruby trumps Python in the simple matter of
> > closures (imagine my shock when I discovered Python didn't have them). 
> 
> Can someone summarize the important difference between closures and object
> instances? Other than convenience, I mean, it would seem that something like
> [ ... ] 

An object is a piece of data with procedures attached to it.
A closure is a procedure with a piece of data attached to it. 

Obviously, they are complete opposites! 

Well -- maybe "mirror images" is a better term.

If you read the HOPL articles on Lisp and Smalltalk, it's
clear that both groups were in communication with each other
and were aware of this -- they called it "orthigonality" --
both a different syntax, and a different way of looking at the 
same thing. 


-- Steve Majewski




From embed at geocities.com  Thu Mar  1 13:49:00 2001
From: embed at geocities.com (Warren Postma)
Date: Thu, 1 Mar 2001 13:49:00 -0500
Subject: A counter-counter-proposal for PEP 236: #pragma( nested_scopes)
References: <j47l29lpo8.fsf@informatik.hu-berlin.de> <mrwn6.16173$Qb7.2637930@newsb.telia.net>
Message-ID: <0Swn6.3324$TW.17061@tor-nn1.netcom.ca>

Has anyone considered borrowing a C notation:

#pragma( nested_scopes )

Benefits:
    - It's a comment therefore it doesn't break anything else.
    - It compiles fine and is ignored on pre-2.x systems.
    - More pragmas can be added later, for warnings, code generation,
etcetera.

Drawbacks:
    - It means a comment can be made to Do Something.  The express intent of
pragma
    is that it was the least evil means to accomplish the given purpose
though.

The comment nature might make it less egregious.  I can imagine other
#pragma stuff
if bytecode compilers for the Microsoft Net CLR and Jython compiler had code
generation options that needed to be taken into account.

I'm willing to attempt a patch if anybody else likes this idea.  The pragma
thing can be useful for other things as well.  As python grows, this
mechanism can be used instead of lots of other little cumulatively uglier
hacks.

Warren Postma





From aahz at panix.com  Sat Mar  3 16:58:40 2001
From: aahz at panix.com (Aahz Maruch)
Date: 3 Mar 2001 13:58:40 -0800
Subject: Bug fix releases
References: <LNBBLJKPBEHFEDALKOLCMEEPJBAA.tim.one@home.com> <200103031910.OAA21663@cj20424-a.reston1.va.home.com> <15009.17989.88203.844343@cj42289-a.reston1.va.home.com> <mailman.983648765.842.python-list@python.org>
Message-ID: <97rpeg$54p$1@panix3.panix.com>

In article <mailman.983648765.842.python-list at python.org>,
Guido van Rossum  <guido at digicool.com> wrote:
>
>That's how all of us here at PythonLabs are feeling...  I feel a
>community task coming.  I'll bless a 2.0.1 release and the general
>idea of bugfix releases, but doing the grunt work won't be a
>PythonLabs task.  Someone else inside or outside Python-dev will have
>to do some work.  Aahz?

I'm certainly committing to be the cheerleader/explainer (meaning that
every time someone complains about the lack of a bugfix release, I tell
them how to get one -- not just flippantly like Uncle Timmy, but in the
sense that there's a real process available for making it happen).

Not sure how much of the actual code I can do, but we'll see what
happens.
-- 
                      --- Aahz (Copyright 2001 by aahz at pobox.com)

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

Nostalgia just ain't what it used to be


From dougfort at downright.com  Thu Mar 22 04:14:40 2001
From: dougfort at downright.com (Doug Fort)
Date: Thu, 22 Mar 2001 04:14:40 -0500
Subject: automatic cookie handling with urllib?
References: <3AB9BFA9.56301B22@ugcs.caltech.edu>
Message-ID: <3AB9C280.391E5025@downright.com>

This is our cookie module.  I've been meaning to rewrite it as a Python
library Cookie object.

--
Doug Fort (dougfort at downright.com)
Senior Meat Manager
Downright Software LLC
http://www.dougfort.net


-------------- next part --------------
#!/usr/bin/env python
"""
CookieContainer

This object stores and retrieves cookies IAW RFC 2109 & RFC 2068

$Id: cookiecontainer.py,v 1.7 2001/02/16 00:27:46 dougfort Exp $
"""
__author__="""
Downright Software LLC
http://www.downright.com
"""
__copyright__="""
Copyright (c) 2000 Downright Software LLC. All Rights Reserved.

Distributed and Licensed under the provisions of the WebNudge
Open Source License (Version 1.0) which is included by reference.

The WebNudge Open Source License can be found in the file WOSLV10.TXT
in the source distribution kit.
"""
__version__="$Revision: 1.7 $"[11:-2]

import re
import time
import urlparse
import string

import webnudge.util.misc
        
def CookieContainerException(Exception):
    def __init__(self, message):
        self._message = message
    def __str__(self):
        return self._message

###########################################################
class CookieContainer:
###########################################################
    """
    This object stores and retrieves cookies IAW RFC 2109 & RFC 2068 
    """

    #----------------------------------------------------------
    def __init__(self, listelement=None):
    #----------------------------------------------------------
        """
        Constructor
        """
        self._cookiedict = {}

        # Match dates  of this form:
        # Monday, 05-Feb-2001 08:00:00 GMT
        self._DatePattern = re.compile(r"""
            (?P<weekday>               # Start of group 'weekday'
            [A-za-z]+                  # Any word of at least one letter
            )                          # End of group 'weekday'
            \s*\,\s*                   # a literal comma after weekday
            (?P<day>                   # Start of group 'day'
            \d\d                       # two digits
            )                          # End of group 'day'
            -                          # literal hyphen 
            (?P<month>                 # Start of group 'month'
            [A-za-z]+                  # three letters
            )                          # End of group 'month'
            -                          # literal hyphen 
            (?P<year>                  # Start of group 'year'
            \d+                        # some digits
            )                          # End of group 'year'
            \s+                        # a space or more
            (?P<hour>                  # Start of group 'hour'
            \d\d                       # two digits
            )                          # End of group 'hour'
            :                          # a colon
            (?P<minute>                # Start of group 'minute'
            \d\d                       # two digits
            )                          # End of group 'minute'
            :                          # a colon
            (?P<second>                # Start of group 'second'
            \d\d                       # two digits
            )                          # End of group 'second'
            \s+                        # some whitespace
            GMT                        # literal 'GMT
            """, re.VERBOSE | re.IGNORECASE)

        self._DateFormat = "%4d-%2s-%2s %2s:%2s:%2s" # yyyy-mm-dd hh:mm:ss
        self._MonthDict = {
            "jan" : "01",
            "feb" : "02",
            "mar" : "03",
            "apr" : "04",
            "may" : "05",
            "jun" : "06",
            "jul" : "07",
            "aug" : "08",
            "sep" : "09",
            "oct" : "10",
            "nov" : "11",
            "dec" : "12"
            }

    #----------------------------------------------------------
    def __str__(self):
    #----------------------------------------------------------
        """
        Report ourself as a string
        """
        return str(self._cookiedict)
        
    #----------------------------------------------------------
    def isempty(self):
    #----------------------------------------------------------
        """
        Report presence of cookies
        """
        return not self._cookiedict

    #----------------------------------------------------------
    def clear(self):
    #----------------------------------------------------------
        """
        Empty out the cookies
        """
        self._cookiedict.clear()

    #----------------------------------------------------------
    def loadFromHeaders(self, defaultdomain, headers):
    #----------------------------------------------------------
        """
        Extract 'set-cookie' from headers from RawHTMLPage
        Return a count of the new cookies added
        """
        count = 0
        
        cookieheaderlist = headers.getallmatchingheaders("set-cookie")
        for cookieheader in cookieheaderlist:
            cookie = {
                "domain" : defaultdomain,
                "path" : "/",
                "secure" : "no"
                }
            
            # split on ';' after dropping 'set-cookie:'
            tokenlist = string.split(cookieheader[12:],";")
            
            # assume name is the  first token
            token =  string.strip(tokenlist[0])
            index = string.find(token, "=")
            if index <= 0:
                continue
            cookie["name"] = token[:index]
            cookie["value"] = token[index+1:]
            
            for token in tokenlist[1:]:
                # split on the first '=', except for secure
                token = string.strip(token)
                if token == "secure":
                    cookie["secure"] = "yes"
                    continue
                index = string.find(token, "=")
                if index <= 0:
                    continue
                key = string.lower(token[:index])
                value = token[index+1:]
                if key == "expires":
                    cookie[key] = self._convertExpirationDate(value)
                else:
                    cookie[key] = value
            self._cookiedict[cookie["name"]] = cookie
            count = count + 1

        return count

    #----------------------------------------------------------
    def returnCookieList(self, url):
    #----------------------------------------------------------
        """
        Return a list of name:value tuples for cookies that
        fit the url
        """
        scheme,netloc,path,parameters,query,fragment = urlparse.urlparse(
            url
            )
        returnlist = []
        for cookie in self._cookiedict.values():
            if len(netloc) < len(cookie["domain"]):
                continue
            # The url must be in the domain the cookie specifies
            if netloc[len(netloc)-len(cookie["domain"]):] != cookie["domain"]:
                continue
            # the path must include the path the domain specifies
            if path and string.find(path, cookie["path"]) != 0:
                continue
            # if we have an expiration date, check for it
            expirationdate = cookie.get("expires", None)
            if expirationdate and expirationdate <= time.time():
                continue
            returnlist.append((cookie["name"],cookie["value"]))

        return returnlist

    #----------------------------------------------------------
    def _convertExpirationDate(self, value):
    #----------------------------------------------------------
        """
        convert a date string of the form
        'Dayofweek, dd-mmm-yyyy hh:mm:ss GMT'
        into a python date.
        """
        # look for a date we can understand
        match = self._DatePattern.search(value)
        if not match:
            return None

        # allow for a two digit year
        # redhat is one of the offenders
        if len(match.group("year")) == 4:
            year = int(match.group("year"))
        else:
            year = 2000 + int(match.group("year"))
            
        # kludge alert!  I can't find a slick way to convert
        # this date, so I'm going to convert it to our database
        # format and use code that I know works
        datestr = self._DateFormat % (
            year,
            self._MonthDict[string.lower(match.group("month"))],
            match.group("day"),
            match.group("hour"),
            match.group("minute"),
            match.group("second"),
            )
        
        return webnudge.util.misc.strtime(datestr)
        
#----------------------------------------------------------
if __name__ == "__main__":
#----------------------------------------------------------
    """
    Code for commandline testing
    """
    import sys
    if len(sys.argv) != 2:
        print "Usage: cookiecontainer.py <url>"
        sys.exit(-1)

    cookiecontainer = CookieContainer()

    import webnudge.util.rawhtmlpage
    page = webnudge.util.rawhtmlpage.RawHTMLPage()
    page.load(sys.argv[1], "GET", [], cookiecontainer, debuglevel=1)
    if not page:
        print "*** Error *** %s" % (page._message)
        sys.exit(-1)

    print "*" * 30
    print page._data
    print "*" * 30

    print "cookie dict"
    for key, value in cookiecontainer._cookiedict.items():
        sys.stdout.write("%s = %s\n" % (key, value))

    print "cookies returned"
    for item in page._cookiesreturned:
        print item

From News at Titanic.co.uk  Sat Mar 10 10:14:46 2001
From: News at Titanic.co.uk (Simon Faulkner)
Date: Sat, 10 Mar 2001 15:14:46 +0000
Subject: Sucking in a text file
Message-ID: <53hkats5d5noic51fskjv0dm72f1rgc29g@4ax.com>

I have a .py cgi script that writes a page to the web and it works!

To allow the users to customise the out put I would like the script to
suck in the first part of the page from a text file and then to spit
this out to the web server.

Is this easy or do I need to open the file and read each byte in turn?


Simon
Simon Faulkner


From aahz at panix.com  Fri Mar 23 10:46:54 2001
From: aahz at panix.com (Aahz Maruch)
Date: 23 Mar 2001 07:46:54 -0800
Subject: Python T-shirts
References: <mailman.985019605.8781.python-list@python.org>
Message-ID: <99fr5e$sp4$1@panix6.panix.com>

[posted to c.l.py with cc to python-dev]

In article <mailman.985019605.8781.python-list at python.org>,
Guido van Rossum  <guido at digicool.com> wrote:
>
>At the conference we handed out T-shirts with the slogan on the back
>"Python: programming the way Guido indented it".  We've been asked if
>there are any left.  Well, we gave them all away, but we're ordering
>more.  You can get them for $10 + S+H.  Write to Melissa Light
><melissa at digicool.com>.  Be nice to her!

If you're in the USA, S&H is $3.50, for a total cost of $13.50.  Also,
at the conference, all t-shirts were size L, but Melissa says that
she'll take size requests (since they haven't actually ordered the
t-shirts yet).
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

"I won't accept a model of the universe in which free will, omniscient
gods, and atheism are simultaneously true."  -- M


From ben.hutchings at roundpoint.com  Tue Mar 20 15:31:52 2001
From: ben.hutchings at roundpoint.com (Ben Hutchings)
Date: 20 Mar 2001 12:31:52 -0800
Subject: Insertin **keywords into a class
References: <mailman.984971990.29857.python-list@python.org> <9960i1$tp0$1@nntp6.u.washington.edu> <99661m$scc$1@panix6.panix.com> <99877b$uss$1@nntp6.u.washington.edu> <Pine.GHP.4.21.0103201151090.4991-100000@don.et.byu.edu>
Message-ID: <uvgp46v2f.fsf@roundpoint.com>

"C. Porter Bassett" <porter at et.byu.edu> writes:

> cCan you please tell me why the following code snippet doesn't update the
> value of radius?

Yes...

> class myClass:
>    def __init__(self, *arguments, **keywords):
>       self.radius = 0.0
>       for kw in keywords.keys():
>          print kw, ":", keywords[kw]
>          self.kw = keywords[kw]

This binds/re-binds self.kw, not the attribute of self named by kw.  I
think you can accomplish what you want by putting self.__dict__[kw] on
the left hand side of the assignment.

>       print "self.radius =", self.radius
> 
> 
> b = myClass(radius = 1.0)

-- 
Any opinions expressed are my own and not necessarily those of Roundpoint.


From tim.one at home.com  Thu Mar  1 17:59:17 2001
From: tim.one at home.com (Tim Peters)
Date: Thu, 1 Mar 2001 17:59:17 -0500
Subject: A counter-proposal to __future__ in PEP 236
In-Reply-To: <97meh4$c0g$1@news.mathworks.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCEECMJDAA.tim.one@home.com>

[Joshua Marshall]
> ...
> Note that it doesn't need to be implemented this way.  Currently,
>
>   directive foo
>
> is a syntax error.  Meaning can be assigned to this syntax without
> breaking any existing code--it is unambiguous.

[Tim]
> You should study Python's parser before getting too optimistic about that
> <wink>.

[back to Joshua]
> So anybody care to post an example of how this can be ambiguous (in
> context) so I don't need to go study Python's parser...?

It's not a question of ambiguity.  You really need to study the parser!  Some
changes are easy, others are hard, due to the way the parser is implemented.
Make "directive" a keyword and it's trivial, because then it can be captured
directly by a grammar production.  It's not possible-- in Python's
parser --to make the identifier "directive" a keyword in some contexts but
not others; if it's not a keyword, it would have to be expressed in terms of
a generic sequence-of-NAME production followed by post-processing.

You also have to consider the parsers in tools, from the Emacs python-mode
thru whatever the heck PythonWorks uses.  Then also the tools building on the
parser module's form of parse trees, which won't even have a case now for
*considering* that "sequence of NAME" is a possible construct.  Unless
"directive" is a keyword, it's a pain in the ass for all of them.

Doesn't mean it can't be hacked, but it's ugly, and you have to get a lot of
different people to sign up to do all that hacking.

"possible"-is-necessary-not-sufficient-ly y'rs  - tim




From robin at jessikat.fsnet.co.uk  Wed Mar 28 03:47:22 2001
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Wed, 28 Mar 2001 09:47:22 +0100
Subject: mailman location spoofing
Message-ID: <BOrBkWAaUaw6Ewxb@jessikat.fsnet.co.uk>

I have setup mailman at x.co.uk for accessibility/control reasons and it
all works fine. Our principal web address is not x.co.uk, but x.com and
it's possible to divert mail to the lists at x.com to lists at x.co.uk. I
would like to know if there's an easy way to spoof the return address
information in mailman or if that makes sense? I know that the web side
cannot in general be moved unless we run mailman on the x.com server.
-- 
Robin Becker


From carribeiro at yahoo.com  Wed Mar 14 17:13:54 2001
From: carribeiro at yahoo.com (Carlos Ribeiro)
Date: Wed, 14 Mar 2001 19:13:54 -0300
Subject: Thoughts about extensions to the Python language
Message-ID: <5.0.2.1.0.20010314181331.0213eac0@mail.inet.com.br>

(flames off please ...)

First of all, some background information... I'm watching this list for the 
past few days. Thats my first contribution to the list. I work on computer 
related jobs for some time (since 1984), and I have saw a lot of things in 
the past decade or so. I'm learning Python since mid-2000, mainly for use 
as a tool to gather network management information.

What attracted me to Python was the simplicity and cleanliness of the 
language. Code written in Python is as clean as code can be. I have written 
systems using several other languages, including C, C++ and Perl, and none 
of these languages can even come close of Python in terms of readability. I 
think that most people on this list will agree with me that making code 
readable is one of the most important (and relegated) steps to improve 
programming skills.

I'm not saying that Python is perfect - as a "work in progress" we can 
surely think about a lot of things that can be improved. However, I'm 
seeing a lot of effort these days toward language extensions. Call it 
"featuritis" or anything like it, but I dont like it, and something inside 
tells me that this is the wrong approach for Python's evolution. I can 
number several ways this approach is failed:

1) Python is not an application, is a programming language. However it 
seems that the PEP process is making Python development look just like any 
other generic application. Someone else on the list has already posted on 
this (he was asking for a maturation time for extensions to the language 
definition). Maybe a lot of this is related to the crescent confusion 
between Python (the language) and CPython - the C implementation that most 
of us know and use.

2) Python has reached a stated where it's useful for a large number of 
people and a wide range of applications. It is absolutely expected that 
people come up with lots of imaginative extensions that fit theirs 
particular needs.

3) Another reason is that some people come from other languages and 
environments. They try to translate their mental models in Python and fail, 
not because something is missing in Python, but because they have a hard 
time learning the "Python Way".

4) There is a lot of pressure from the "my-language-is-better-than-yours" 
department. Some of this people are particularly loud on making their 
requests heard, and this gives an unproportional weight to some proposals 
that are silly at best. The vast majority of the Python users don't listen 
to the list. They are going to have quite a surprise with the next releases...

5) Written proposals are dangerous ;-) Ok, this is an exageration. This is 
another way to state some of the points above. Dumb ideas have somehow 
developed an ability to make themselves look much better when written in a 
formal proposal (Alan Sokal, please?). That's a problem that plagues a lot 
of mailing lists. You have to be alert to avoid it.

6) Most important of all, it's clear that Guido had used a 'common sense' 
approach to design the original language. At that time there were much less 
exposition than today. It seems that some of this "practical" approach is 
being lost in the transition to the open source world. I think that all the 
reasons above apply, mainly because the pressure is much higher.


We could elaborate a lot on this, and I think that some of the older 
"Pythoneers" are doing exactly this. However, for the extensions proposed 
for the language itself, I saw a lot of discussions for particular issues 
*without* the amplitude needed for a generic solution. I fear that this may 
lead to a distorted perception of reality. Is everyone sure that Python 
needs as much extensions to the language as proposed now? Maybe I'm way too 
much conservative, but for me the answer is no. Let's see what everyone 
else thinks.



Carlos Ribeiro
carribeiro at yahoo.com
cribeiro at mail.inet.com.br


_________________________________________________________
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com




From tuttledon at hotmail.com  Tue Mar 13 10:24:19 2001
From: tuttledon at hotmail.com (Don Tuttle)
Date: Tue, 13 Mar 2001 15:24:19 GMT
Subject: win32all buld 137 win32gui fails
References: <98l8qk$oeo$1@troll.powertech.no>
Message-ID: <DYqr6.83020$__6.16207072@typhoon.southeast.rr.com>

"Syver Enstad" <syver at NOSPAMcyberwatcher.com> wrote in message
news:98l8qk$oeo$1 at troll.powertech.no...
> I've just installed the win32all build 137 that contains the support for
> Vtable interfaces in win32com. I've suddenly found that import win32gui
> fails with the message:
>
> ImportError: DLL load failed: The specified module could not be found.
>
> This is mighty strange as the pyd file is lying there among the other .pyd
> files like win32help and so on. Maybe there are some other modules that
> win32gui depends on that it didn't depend on before?
>
> Anybody have a suggestion to fix this?
>
> For the record I am running Win2000 Pro with no service packs, VC++ 6 to
> service pack 3, IE 5.


Don't know.  It imports fine for me.  Perhaps unistall 137, then reinstall.

Don




From claird at starbase.neosoft.com  Fri Mar 23 13:54:16 2001
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 23 Mar 2001 12:54:16 -0600
Subject: newbie question
References: <575F0D65CC45F12D.7D165177C452F6D5.01903532892CC93D@lp.airnews.net> <99fll8$996$1@rex.ip-plus.net> <mailman.985367791.3511.python-list@python.org>
Message-ID: <73125560B5E4EF16.CC3FC0DB3FD693E3.4AF0ACCBD2EFF29B@lp.airnews.net>

In article <mailman.985367791.3511.python-list at python.org>,
Moshe Zadka  <moshez at zadka.site.co.il> wrote:
>On 23 Mar 2001, claird at starbase.neosoft.com (Cameron Laird) wrote:
>
>> There are few languages that canNOT be used with
>> Apache.
>
>I might be talking out of my ass here, in which case Cameron will surely
>correct me. Haskell, Ruby, Icon, Scheme, ML and Smalltalk are all languages 
>which are highly useful, and which no interfaces to Apache exists as far as I
>know. I'd simply go the opposite direction and say
>
>"there are few things Python cannot talk to. Apache is not one of those"
>(I like to phrase it: Python plays nice with the other kids. And there
>are *always* other kids)
>
>waiting-for-someone-to-prove-me-wrong-(hurd,plan9)-ly y'rs, Z.
			.
			.
			.
I was elliptic.  And also intuitive--I speculated that
CGI would fit what the original poster had in mind for
"an interface to Apache".  Subsequent correspondence
turned out to confirm that.  So:  out-of-the-box, Apache 
installs with CGI capabilities, and Icon, Scheme, Ruby,
... are all languages I've seen used in CGI (with Apache).

I make the point not as a mere legalism.  Many of the
questioners who come here are thrilled to learn, "Ah, you
already have a standard Linux distribution Apache and
Python installed; all you need is to write ..., and
you'll see your existing configuration already supports
Python-savvy Web service."  That's a lot more exciting
for them than hearing that, with a mere week or two of
study, they can install Zope on their own.
-- 

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


From padiolea at merlin.irisa.fr  Wed Mar 14 07:26:10 2001
From: padiolea at merlin.irisa.fr (Yoann Padioleau)
Date: 14 Mar 2001 13:26:10 +0100
Subject: Einstein's Riddle
References: <oNPq6.396644$ge4.138479566@news2.rdc2.tx.home.com> <UdSq6.2184$54.2166@www.newsranger.com> <984380185.768920@news.aei.ca> <uP%q6.2375$54.2207@www.newsranger.com> <4w6r6.193645$df5.5192565@news1.crdva1.bc.home.com> <3aae4114.63604234@nntp.interaccess.com> <3AAE4BFD.9A7B0D9C@cs.man.ac.uk> <3AAF1240.C446D19@iit.demokritos.gr>
Message-ID: <wi67l1spme5.fsf@merlin.irisa.fr>

Petasis George <petasis at iit.demokritos.gr> writes:

> "Donal K. Fellows" wrote:
> > 
> > Thaddeus L Olczyk wrote:
> > > I also figure the fish thing was simply a typo, and that this guy (
> > > who hasn't resdponded ) was probably given this as a homework
> > > assignement, and is simply looking for a cheap way to get it done.
> > 
> > That'd be a very lame homework assignment.  Not unless you had to write
> > a program in $FAVOURITE_LANGUAGE to compute the answer automatically
> > (which is much more interesting, especially if you make the program
> > start from the natural language expression of the problem!  :^)
> > 
> Well, try this:-)
> Written by a friend of mine. Its amazing how small can programs be if you
> know STL. To bad I don't know stl :-)
> Of course it will need some time, as it randomly checks all combinations...
But your program dont really do the job cos it is too slow.
You first generate the permutations and then test, the pb is that
there is too many permutations : 
p[nation][english] can be either 0 1 2 3 4 or 5, same for the other variables
you have 5*(5!) possibililty = 24 883 200 000
if you are able to do 1 000 000 ValidConstraints per second (cos you have perhaps
a very very fast computer) you will need 24 883 seconds =~ 6hours to compute (in the worst case
i admit)

the prolog program i propose take 0.1 second to find the solution :)
who say that prolog is slower than c++ :))

> 
> George
> 
> //A simple randomiser to find the solution to the riddle, by J. Y. Goulermas
> 
> #include <vector>
> #include <ctime>
> #include <algorithm>
> #include <fstream>
> #include <iomanip>
> #include <iostream>
> 
> using namespace std;
> 
> enum element_type { nation,  colour,  pet,         drink,  cigs       };
> enum nation_type  { english, swedish, danish,      german, norwegian  };
> enum colour_type  { red,     blue,    yellow,      green,  white      };
> enum pet_type     { cat,     bird,    horse,       dog,    fish       };
> enum drink_type   { tea,     water,   milk,        beer,   coffee     };
> enum cigs_type    { palmal,  dunhill, bluemasters, prince, blends     };
> 
> char* codes[][5] = { { "english", "swedish", "danish",      "german", 
> "norwegian" },
>                      { "red",     "blue",    "yellow",      "green",  
> "white"     },
>                      { "cat",     "bird",    "horse",       "dog",    "fish" 
>       },
>                      { "tea",     "water",   "milk",        "beer",   
> "coffee"    },
>                      { "palmal",  "dunhill", "bluemasters", "prince", 
> "blends"    }
>                    };
> 
> inline bool ValidConstraints(const vector< vector<int> >& p)
> {
>   return   p[nation][norwegian] == 0                   //Rule 9
>          &&
>            p[colour][blue] == 1                        //Rule 14
>          &&
>            p[drink][milk] == 2                         //Rule 8
>          &&
>            p[nation][english] == p[colour][red]        //Rule 1
>          &&
>            p[nation][swedish] == p[pet][dog]           //Rule 2
>          &&
>            p[nation][danish] == p[drink][tea]          //Rule 3
>          &&
>            p[colour][yellow] == p[cigs][dunhill]       //Rule 7
>          &&
>            p[pet][bird] == p[cigs][palmal]             //Rule 6
>          &&
>            p[drink][beer] == p[cigs][bluemasters]      //Rule 12
>          &&
>            p[nation][german] == p[cigs][prince]        //Rule 13
>          &&
>             p[colour][green] == p[drink][coffee]       //Rule 5
>          &&
>            abs(p[cigs][blends] - p[pet][cat]) == 1     //Rule 10
>          &&
>            abs(p[cigs][blends] - p[drink][water]) == 1 //Rule 15
>          &&
>            abs(p[pet][horse] - p[cigs][dunhill]) == 1  //Rule 11
>          &&
>            p[colour][green] < p[colour][white];        //Rule 4
> }
> 
> void Output(const vector< vector<int> >& p,
>                   unsigned               c
>            )
> {
>   ofstream text("solution.txt");
> 
>   for (int i = nation; i <= cigs; ++i)
>   {
>     for (int j = 0; j < p[i].size(); ++j)
>       text <<setw(20)
>            <<codes[i][ find(p[i].begin(), p[i].end(), j) - p[i].begin() ]; 
> //get inverse permutation
>     text <<endl;
>   }
>   text <<"\n\nTotal attempts: " <<c;
> 
>   text.close();
> }
> 
> void main(void)
> {
>   //srand( (unsigned) time(NULL) ); //optional RNG seeding
> 
>   int                   ramp[] = { 0, 1, 2, 3, 4 };
>   vector< vector<int> > permutations;
>   unsigned              counter(0);
>   for (int i = nation; i <= cigs; i++ )
>     permutations.push_back( vector<int>(ramp, ramp + 5) );
> 
>   do
>   {
>     for (int i = nation; i <= cigs; i++)
>       random_shuffle( permutations[i].begin(), permutations[i].end() );
>     if ( ! (++counter % 100000) )
>       cout <<"\rRe-Randomisations: " <<counter;
>   }
>   while ( ! ValidConstraints(permutations) );
> 
>   Output(permutations, counter);
> }

-- 
Yoann  Padioleau,  INSA de Rennes, France,   http://www.irisa.fr/prive/padiolea
Opinions expressed here are only mine. Je n'?cris qu'? titre personnel.
**____   Get Free. Be Smart.  Simply use Linux and Free Software.   ____**


From cg at schlund.de  Sat Mar 10 13:36:48 2001
From: cg at schlund.de (Carsten Gaebler)
Date: Sat, 10 Mar 2001 19:36:48 +0100
Subject: How to efficently build a nested dictionary
References: <3AAA2042.E69A53FC@gmx.de> <98dgjb$iee$1@panix3.panix.com> <3AAA71F1.7FC7597A@schlund.de>
Message-ID: <3AAA7440.E0DBACAF@schlund.de>

Carsten Gaebler wrote:
> 
> Aahz Maruch wrote:
> > >
> > >traffic = {dir: {year: {month: {day: 42}}}}
> >
> > I think Emile's suggestion to use a single compound key is a good one.
> > If that's not sufficient *and* you can use Python 2.0, take a look at
> > the dict.setdefault() method.
> 
> How could the setdefault() method help me out?

By just saying

traffic[dir][year][month][day] =
traffic.setdefault(dir,{}).setdefault(year,{}).setdefault(month,{}).setdefault(day,
0) + bytes

Yes, I know, sometimes I'm a bit slow ... :-)

cg.


From fredrik at pythonware.com  Mon Mar 12 17:21:23 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Mon, 12 Mar 2001 22:21:23 GMT
Subject: This math scares me
References: <3aad4062.1290021@News.CIS.DFN.DE> <Hwbr6.1139$Tg.127571@ruti.visi.com> <3aad4485.2348499@News.CIS.DFN.DE>
Message-ID: <DZbr6.17917$Qb7.3117518@newsb.telia.net>

costas at springmail.com wrote:
> I will also tell my customers that this is the new math.

if you don't understand how floating point numbers work, you
probably shouldn't use them in customer projects.

(unless you like to get sued for incompetence, of course...)

Cheers /F




From wolfson at midway.uchicago.edu  Tue Mar 13 03:14:31 2001
From: wolfson at midway.uchicago.edu (Ben Wolfson)
Date: Tue, 13 Mar 2001 08:14:31 GMT
Subject: Stupid Python trick
Message-ID: <HFkr6.150$L4.10426@uchinews>

I'm sure this can be put to creatively bad use.

Python 2.1b1 (#3, Mar  2 2001, 20:37:19) 
[GCC 2.96 20000731 (Red Hat Linux 7.0)] on linux2
Type "copyright", "credits" or "license" for more information.
IDLE 0.6 -- press F1 for help
>>> class D:
	def __init__(self, otherself, v):
		self.val = v
		try: otherself.val = v
		except: pass

>>> import new
>>> d = D(None, 5)
>>> d.val
5
>>> m = new.instancemethod(D, d, D)
>>> m(6)
<__main__.D instance at 0x82df1dc>
>>> d.val
6

-- 
Barnabas T. Rumjuggler
Fried Something with Noodle:  Very good for ugly, sad, lonely, sick
people with the blood and the grasses all over the floor...
 -- Joe Frank, _A Call in the Night_


From johnw at gnu.org  Sun Mar 11 23:49:10 2001
From: johnw at gnu.org (John Wiegley)
Date: 11 Mar 2001 21:49:10 -0700
Subject: Going ZODB
References: <mailman.984324988.25505.python-list@python.org> <m3wv9w3x6h.fsf@gnu.org> <slrn9aoi0i.rt.amk@mira.erols.com>
Message-ID: <m3bsr7h9rt.fsf@gnu.org>

>>>>> On Sun Mar 11, A writes:

> Current versions of the ZODB *do* support __getattr__ and
> __setattr__, so you may want to take another look at the ZODB.

Now THAT sounds cool.  Checking now...


From vsbabu at erols.com  Sat Mar 31 06:33:08 2001
From: vsbabu at erols.com (Satheesh Babu)
Date: Sat, 31 Mar 2001 06:33:08 -0500
Subject: HTML Tidy and Python wrapper
References: <985945106.28359.0.nnrp-02.c1c3e154@news.demon.co.uk>
Message-ID: <9a4f6t$qn1$1@bob.news.rcn.net>

I was searching for some thing like this. I had to use it with ASP. I've a
small article about this at
http://vsbabu.csoft.net/techbits/tidywithasp.html
Also has a snippet on using Tidy with Python.

This probably might be useless

--
v.s.babu
vsbabu at erols.com
http://vsbabu.csoft.net
"Paul Brian" <pbrian at demon.net> wrote in message
news:985945106.28359.0.nnrp-02.c1c3e154 at news.demon.co.uk...
> Dear all,
>
> I was sure that I recently read that there was a python wrapper around the
> HTML Tidy (http://www.w3.org/People/Raggett/tidy/). Unfortunately I cannot
> find the reference. Does any one know if such a thing exists, as simple
> searches through deja have come up nought.
>
> thank you
>
>




From phrxy at csv.warwick.ac.uk  Sun Mar 25 17:05:59 2001
From: phrxy at csv.warwick.ac.uk (John J. Lee)
Date: Sun, 25 Mar 2001 23:05:59 +0100
Subject: boolean values in COM / VB / Python
Message-ID: <Pine.SOL.4.30.0103252251210.18627-100000@mimosa.csv.warwick.ac.uk>

ie.Document.execCommand("SaveAs", false, pathname)

where ie is an Internet Explorer object, pops up a dialogue box, no matter
what value for false I try, from (1==0, 0, "0", None). According to the MS
docs at

http://msdn.microsoft.com/workshop/author/dhtml/reference/constants/saveas.asp

and

http://msdn.microsoft.com/workshop/author/dhtml/reference/methods/execCommand.asp

this isn't supposed to happen, because when the second argument to
execCommand is 'false', the command 'Does not display a user interface'.
According to Mark Hammond just last month, 0 should do fine as a way of
representing 'false':

http://groups.google.com/groups?hl=en&lr=&safe=off&ic=1&th=1cd0ed3943dead44&thn=Setting+boolean+COM+properties&seekd=926499672

Any clues as to what is happening?

Tips on the differences between C++ / VB and Python in general would also
be appreciated -- presumably there must be some reliable and mechanical
way of doing this?


John



From not.this at seebelow.org  Thu Mar  1 09:17:36 2001
From: not.this at seebelow.org (Grant Griffin)
Date: Thu, 01 Mar 2001 08:17:36 -0600
Subject: is and == (is not other_thread)
References: <mailman.983428868.21795.python-list@python.org>
Message-ID: <3A9E5A00.E276C54A@seebelow.org>

Robin Thomas wrote:
> 
> At 10:33 PM 2/28/01 -0600, Grant Griffin wrote:
> >Hi Gang,
> >
> >I've had a few of my own little questions about "is" and "==", vis-a-vis
> >strings:

<good answers snipped>

> Use the source, Luke.

Yup, I'd like to one day.  But for now, I don't have time to figure out
how 100k lines of C code works: I only have enough time to ask people
other folks who _do_ have enough time to figure out how 100k lines of C
code. <wink>

> All multipart questions are thus answered. This reply
> was your freebie.

You're so kind...may I ask just one more favor?  My car needs an oil
change...

give-in-to-the-dark-side-of-automotive-lubrication,-Luke-<wink>
   -ly y'rs,

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation            http://www.iowegian.com


From dsavitsk at e-coli.net  Mon Mar  5 17:51:21 2001
From: dsavitsk at e-coli.net (dsavitsk)
Date: Mon, 5 Mar 2001 16:51:21 -0600
Subject: Python COM server to VB interface
References: <keyo6.156158$Ch.31627133@newsrump.sjc.telocity.net>
Message-ID: <KLUo6.158450$Ch.32565541@newsrump.sjc.telocity.net>

for anyone interested, this turns out to be a corrupted registry i think.
it works fine on other computers.

ds


"dsavitsk" <dsavitsk at e-coli.net> wrote in message
news:keyo6.156158$Ch.31627133 at newsrump.sjc.telocity.net...
> I am attempting to learn to use python as a com server to a VB interface.
> When i run my little example i get the following error.
>
> Run-time error'-2147467259 (800004005)':
> Unexpected Python Error: AttributeError: CreateInstance
>
> Here is the Python end of things...
>
>     class pyserv_test:
>         _public_methods_ = [ 'AddString' ]
>         _reg_progid_ = 'PyServ.Test'
>         _reg_clsid_ = '{EA08AFC9-E0AA-4C9B-8E42-1170A21D59AC}'
>
>         def AddString(self, val):
>             return 'the string was: ' + str(val)
>
>     def Register():
>         import win32com.server.register
>         return win32com.server.register.UseCommandLine(pyserv_test)
>
>     if __name__=='__main__':
>         print "Registering COM server..."
>         Register()
>
>
> and here is the relevant VB code...
>
>     Private Sub Command1_Click()
>         Dim x
>         Dim y
>         Set x = CreateObject("PyServ.Test")
>         y = x.AddString(Text2.Text)
>         Text1.Text = y
>     End Sub
>
> The error occurs on the "CreateObject" line. Does anyone know what this
> might be?
>
> Thanks,
> doug
>
>




From neilh at scintilla.org  Tue Mar 20 17:26:13 2001
From: neilh at scintilla.org (Neil Hodgson)
Date: Tue, 20 Mar 2001 22:26:13 GMT
Subject: Little red lines in PythonWin!
References: <mailman.985105228.27509.python-list@python.org>
Message-ID: <9OQt6.13652$992.84056@news-server.bigpond.net.au>

Bill Bell:
> When I view the whois client using PythonWin I see that lots of the
> Python indenting is lost and that numerous red lines have been
> introduced which are (presumably) meant to inform me that
> something is wrong.

   Even though Bill has found his problem, I'd like to point out the meaning
of the little red lines. They are there to show that there is potentially
ambiguous indentation in the file and that the indentation should be fixed,
preferrably to all tabs or all spaces.

   This feature is turned on by default in PythonWin and can be controlled
through View | Options... | Tabs and Whitespace | Tab-Timmy. IMO all Python
IDEs should have a similar feature. For those IDEs that use Scintilla, the
feature is already implemented for you so it is almost free to add.

   Neil




From aleaxit at yahoo.com  Mon Mar 12 04:14:52 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Mon, 12 Mar 2001 10:14:52 +0100
Subject: Static typing (was Re: Java guy interested in Python)
References: <mailman.984334635.27345.python-list@python.org> <98glhc02sga@news2.newsguy.com> <m3hf1080or.fsf@atrus.jesus.cam.ac.uk> <98gt3f064l@news2.newsguy.com> <m3elw39a7m.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <98i42a0oas@news2.newsguy.com>

"Michael Hudson" <mwh21 at cam.ac.uk> wrote in message
news:m3elw39a7m.fsf at atrus.jesus.cam.ac.uk...
    [snip]
> > to spot this.  But on re-reading it falls pretty flat -- sorry!
>
> Actually, I thought
>
> #define int foo
>
> was pedantically illegal C++ that was almost certain to work, but on
> checking I see that it's probably legal.  I also hadn't taken this
> possibility into account when replying above; catching people out with
> the preprocessor isn't really fair...

Yes it's legal, and yes it IS 'unfair' -- to the point of making
my original post on this a waste of time (I repeat my apologies).


Alex





From loewis at informatik.hu-berlin.de  Thu Mar 29 02:13:43 2001
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 29 Mar 2001 09:13:43 +0200
Subject: Building "non-standard" standard extensions
References: <719vbt8tlkb05h4n1mcvh65l0f0i62tgsj@4ax.com>
Message-ID: <j4u24dm4ig.fsf@informatik.hu-berlin.de>

Paul Moore <gustav at morpheus.demon.co.uk> writes:

> However, I do have a binary version of the readline library
> available. Can I build the readline module for Windows usingh this?
> Ideally, I'd like to do this *without* having to rebuild the whole
> of Python for Windows (I'm happy with just using the normal build
> for everything else). I've never built Python for myself, so
> apologies if the answer to this is "obvious"...

Building the readline module alone should be possible. I'm not sure
whether the interactive interpreter will use it - it only tries if
stdin isatty.

> (BTW, this doesn't just apply to readline - I have various other libraries
> such as a curses port, which I wouldn't mind trying to integrate as well...)

That should work as well.

Regards,
Martin


From flognat at flognat.myip.org  Thu Mar 15 18:45:20 2001
From: flognat at flognat.myip.org (Andrew Markebo)
Date: 16 Mar 2001 00:45:20 +0100
Subject: inverse of asctime
References: <98rfn6016rf@drn.newsguy.com>
Message-ID: <m3d7bippf3.fsf@master.athome>

/ Grant Griffin <not.this at seebelow.org> wrote:
| Hi Gang,
| 
| I'm in the market for the inverse of asctime().  I looked in the time module but
| couldn't find it.  (Nope, strptime isn't available on my Windows installation.) 
| Seems like there oughtta be sucha thing somewhere.

Ehm.. I actually have a strptime.py, used by RogueWave ;-) *stretches
his back* You can locate it at my mismaintained code-page, will have
to make it come back to life some day.

http://www.fukt.hk-r.se/~flognat/hacks/
http://www.fukt.hk-r.se/~flognat/hacks/strptime.py

*stretching-his-backly-yours*

        /Andy


BTW Have you actually tried searching the python site :-) 

Number-crunching NumPy Extensions

... strptime.py - Produce time-module style time tuples (Andy
Eskilsson) The inverse of strftime, convert a string to a tuple that
the time module can handle. ...
www.python.org/topics/scicomp/numbercrunching.html - 14k - Cached -



From lluang at northwestern.edu  Tue Mar 13 12:49:23 2001
From: lluang at northwestern.edu (Louis Luangkesorn)
Date: Tue, 13 Mar 2001 11:49:23 -0600
Subject: Python In A Nutshell - Numpy essential
References: <3AAE7E0D@MailAndNews.com> <slrn9aolph.ccf.jimhill@meldrick.swcp.com> <3AACE20C.AC8D2EA2@northwestern.edu> <Ikkr6.68807$ML1.3285992@e420r-atl2.usenetserver.com>
Message-ID: <3AAE5DA3.B736C2D0@northwestern.edu>

Steve Holden wrote:

> "Louis Luangkesorn" <lluang at northwestern.edu> wrote in message
> news:3AACE20C.AC8D2EA2 at northwestern.edu...
> > I have to second that Numeric Python belongs in the ESSENTIAL column.  For
> me, the fact that Python can replace Matlab was the reason I started using
> Python.

> <stuff deleted>

> Well, just for balance, I still haven't downloaded NumPy, and don't ever
> expect to.
>
> one-man's-essential-is-another-man's-redundant-ly y'rs  - steve

Oh, oh, I hope this does not mean that we have to have an election.



From joconnor at cybermesa.com  Thu Mar  1 14:35:14 2001
From: joconnor at cybermesa.com (Jay O'Connor)
Date: Thu, 01 Mar 2001 12:35:14 -0700
Subject: [Q] Question from Python tutorial
References: <Lgxn6.6964$dL4.91439@vixen.cso.uiuc.edu>
Message-ID: <3A9EA472.F15AF847@cybermesa.com>


Young-Jin Lee wrote:

> Hi, I have a question from Python tutorial.
> The code segment in the page 60 (chapter 9. classes) is not executed.
> I don't know what I did wrong. Here is the code and the error I got.
>
> class MyClass:
>     "A simple example class"
>     i = 12345
>     def f( x ):
>         return 'hello world'
>
> x = MyClass()
>
> x.counter = 1
> while x.counter < 10:
>     x.counter = x.counter * 2
> print x.counter
>       ^
> SyntaxError: invalid syntax
>
> There is no typo, I have no idea of where this SyntaxError came from.
>
> Thanks in advance. Any comment would be greatly appreciated.

You probably cut and pasted the code from the tutorial.  When I cut and
pasted your code into IDLE, I got the same error.  When I manyually typed
the code, it works fine.  My guess is that the cut and paste didn't paste
properly and the interactive interpreter didn't make the right break
between the 2 and the word print

--
Jay O'Connor
joconnor at cybermesa.com
http://www.cybermesa.com/~joconnor

Python Language Forum -
http://pub1.ezboard.com/fobjectorienteddevelopmentpython




From Pieter_Claerhout at CreoScitex.com  Wed Mar  7 04:37:05 2001
From: Pieter_Claerhout at CreoScitex.com (Pieter Claerhout)
Date: Wed, 7 Mar 2001 10:37:05 +0100 
Subject: GadFly and gfplus.py working with Python 2.0?
Message-ID: <2B1262E83448D211AE4B00A0C9D61B0301DDDB81@msgeuro1.177.115.149.IN-ADDR.ARPA>

I tried it and it seems to work fine...

Pieter

-----Original Message-----
From: Eugene Leitl [mailto:Eugene.Leitl at lrz.uni-muenchen.de]
Sent: Wednesday, March 07, 2001 10:42 AM
To: python-list at python.org
Subject: GadFly and gfplus.py working with Python 2.0?



See subject.
Aye, nay, maybe?

TIA,
-- Eugene
______________________________________________________________
ICBMTO  : N48 10'07'' E011 33'53'' http://www.lrz.de/~ui22204 
57F9CFD3: ED90 0433 EB74 E4A9 537F CFF5 86E7 629B 57F9 CFD3


-- 
http://mail.python.org/mailman/listinfo/python-list



From sholden at holdenweb.com  Fri Mar  9 09:33:00 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Fri, 9 Mar 2001 09:33:00 -0500
Subject: COM support (was Re: Can Python replace Visual Basic? Should it?)
References: <5DZo6.11126$hn5.1578410@news1.rdc1.mb.home.com> <mailman.984010931.26257.python-list@python.org> <iSBp6.14005$hn5.2035325@news1.rdc1.mb.home.com> <mailman.984094514.29514.python-list@python.org> <984133550.452523@newsmaster-04.atnet.at> <98ai2j01o8j@news2.newsguy.com> <984145785.263773@newsmaster-04.atnet.at>
Message-ID: <fU5q6.45262$1D5.2046670@e420r-atl1.usenetserver.com>

"Werner Schiendl" <ws-news at gmx.at> wrote in message
news:984145785.263773 at newsmaster-04.atnet.at...
> There are definitely areas in COM - a lot of such, to be precise - where
> Visual Basic is _not_ suited (at all).

Well, "three thousand six hundred seventeen" would be precise. "A lot" just
sounds like you've got(ten) tired of stubbing your toe on an inadequate
platform.

> Thing is, there are a lot of things that do not need to care on how the
> marshaling is done in detail (because the default is good enough for a lot
> of applications, IF you have well designed interfaces - if not, writing
your
> own marshaling code is the wrong answer to the problem) or how the
security
> is handled in detail (becuase, e. g. it is not intended to use even by a
> local client - only inproc like active x controls or utility classes to
> provide functions Visual Basic is missing by default).
>
COM is quite good at hiding (often-irrelevant) implementation detail from
the inexperienced user. The major problems with it seem to be its underlying
complexity and Microsoft's insistence on inventing things for themselves
which solve problems for whiuch there are already perfectly adequate
solutions, quite often in open source.

> As I mentioned, I do not have much experience with Python and COM and I
> _thought_, based on that very small insight, Visual Basic would be the
more
> convenient environment. I read a lot of rather headache stuff on
generating
> interface classes and things like that and this is one point Visual Basic
> handles nicely. ActiveX control containment is also one thing I would have
> no idea on how to start.
>
 COM is much more  fun in Python than in VB, until you stub your toe against
somethig that really does require an understanding of the underlying
implementation. My usual solution (and that of many others) at this point is
to scream for help on c.l.py, and Mark Hammond amazingly often comes up with
an answer. I somethimes don't LIKE the answer, of course, but that's *my*
problem.

> BTW: I intend to get a copy of O'Reillys book "Python Programming on
Win32"
> to deepen my insight on that topic. If you know
> other links or books worth reading (apart from the docs that come with
> python, of course) I would appreciate to hear about them.
>
Run, don't walk. It's an amazingly broad and useful book.

> regards
> werner
>
> P.S.: Of course one should (or must, to bring it to the point) understand
> the technology in use. This applies heavily to COM, and I know quite a lot
> about that stuff. This is one of the reasons I think that well designed
> interfaces are more important for remoting than the actual programming
> language. And you can use IDL defined Interfaces with VB, otherwise it
would
> be a headache some time...
>
[martellibot wisdom snipped]





From dnew at san.rr.com  Mon Mar  5 11:37:26 2001
From: dnew at san.rr.com (Darren New)
Date: Mon, 05 Mar 2001 16:37:26 GMT
Subject: Missing the functional mind set
References: <mlb2at8ptrr7sh5dgq8qlk2201kquh6tfl@4ax.com> <uDao6.335972$ge4.119457077@news2.rdc2.tx.home.com> <t6f2at870pk82q96bnd5a2od4n8ns5kf83@4ax.com> <ntg2atcemm3hp9rb26hn7c5uin4dfn8lc9@4ax.com> <97rqgb$lt2$1@tyfon.itea.ntnu.no> <3AA298F5.7742810E@wwc.com>
Message-ID: <3AA3C0C8.2F33233C@san.rr.com>

Steve Williams wrote:
> To be consistent with the "index == offset" mindset, counting should
> start at zero.

Most humans don't start counting until after they've already found one.
Since computers only do one thing at a time (either count, or find one),
computers make explicit that you're starting at zero.

-- 
Darren New / Senior MTS & Free Radical / Invisible Worlds Inc.
San Diego, CA, USA (PST).  Cryptokeys on demand.


From ws-news at gmx.at  Tue Mar  6 11:59:26 2001
From: ws-news at gmx.at (Werner Schiendl)
Date: Tue, 6 Mar 2001 17:59:26 +0100
Subject: absolute namespace for callbacks?
References: <3AA50AFB.83C7459F@troikanetworks.com> <983895797.296667@newsmaster-04.atnet.at> <3AA51144.3DD3FC3F@troikanetworks.com>
Message-ID: <983898242.925306@newsmaster-04.atnet.at>

Con you specify how you intend to register your callback?

somewhat like this?

>>> def test_func(msg):
...  print "Hello in Testfunc, message is:", msg
...  return len(msg)
...
>>> def use_callback(func):
...  print "Function provided as callback is %s", func
...  testdata = "Hello, World!"
...  result = func(testdata)
...  print "Invoked function with %s, result is %d" % (testdata, result)
...  return result
...
>>> use_callback(test_func)
Function provided as callback is %s <function test_func at 0146CBBC>
Hello in Testfunc, message is: Hello, World!
Invoked function with Hello, World!, result is 13
13

(make sure for proper identation)

So it in fact does not matter, where test_func or use_callback is
declared...
Did I miss something?

hth
werner


Bruce Edge <bedge at troikanetworks.com> wrote in message
news:3AA51144.3DD3FC3F at troikanetworks.com...
> My callback func is defined in one module, and the callback is executed
> from another module which does not know in advance what module the
> callback func resides in. So, by default, the caller cannot find the
> callback func in it's own namespace.
> The callback registration procedure needs to specify the namespace where
> the callback func exists.
>
> OTOH, maybe I'm missing something here :-)
>
> -Bruce.
>
> Werner Schiendl wrote:
> >
> > What's the matter?
> >
> > You do not pass the name, but a reference to the function
> > (the function is an object for python like anything else)
> > It does not matter where the function is defined...
> >
> > or did I miss anything in your post?
> >
> > hth
> > werner
> >
> > Bruce Edge <bedge at troikanetworks.com> wrote in message
> > news:3AA50AFB.83C7459F at troikanetworks.com...
> > > Tcl has a func to get a fully qualified namespace from a context. This
> > > is essential for registering callback functions.
> > > How does one register a callback in Python, when the callback func
> > > exists in a different namespace?
> > >
> > > Is there any method to get the current absolute namespace?
> > >
> > > Thanks, Bruce.
> > >




From aleaxit at yahoo.com  Thu Mar 29 09:25:23 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 29 Mar 2001 16:25:23 +0200
Subject: setvbuf() ??
References: <slrn9c50mp.me7.wcuddy@lserv.ja10629.home> <3AC29961.9937EC59@ucdavis.edu>
Message-ID: <99vglp02gdd@news2.newsguy.com>

"Bruce Wolk" <bawolk at ucdavis.edu> wrote in message
news:3AC29961.9937EC59 at ucdavis.edu...
> Putting a trailing comma after the print statement eliminates the newline,
but
> adds a trailing space.  Would periods separated by spaces work?
>
> I would be curious if anyone knows how to eliminate the trailing space.

The simplest way is probably

    import sys
    sys.stdout.write('.')


Alex





From lluang at northwestern.edu  Sat Mar 17 00:47:58 2001
From: lluang at northwestern.edu (Louis Luangkesorn)
Date: Fri, 16 Mar 2001 23:47:58 -0600
Subject: MS Access query fails on Date/Time field
Message-ID: <3AB2FA8E.F00ED88E@northwestern.edu>

This has got to be a FAQ, but I can't find it anywhere

I am trying to use the mx.ODBC.Windows module to query an Access
database (for some data laundering it into a database application we are
using at my church).  But when I try a query that returns a Date/Time
field, it returns an error

>>> import mx.ODBC.Windows
>>> connect =
mx.ODBC.Windows.connect('ptcdonors',user='',password='christ')
>>> c1 = connect.cursor()
>>> c1.execute('select [Receipt #], City, State from Person')
>>> a = c1.fetchone()
>>> a
(3268, None, None)
Works fine, but
>>> c1.execute('select Date, [Receipt #], City, State from Person')
>>> a = c1.fetchone()
Traceback (innermost last):
  File "<interactive input>", line 1, in ?
InternalError: ('S1003', 0, '[Microsoft][ODBC Driver Manager] Program
type out of range', 3013)
>>>
I've narrowed it down to the date field.  When I try to use ADO
(following Hammond's chapter 13) I can get the date object, but then I
have to figure out how to get each of the returned data fields out from
their tuples.

So,
1)  Is it possible to get the information out of an MS Access date field
using ODBC?  -or-
2)  How do you unpack all those tuples that ADO queries return?  I
figured out the date one, using int(aDate), then aDate.Format(), but not
the others.

Thank you.

Louis

--
K Louis Luangkesorn
lluang at northwestern.edu  http://pubweb.nwu.edu/~kll560  PGP:0xF3E2D362

Whatsoever things are true, ... honest, ... just, ... pure, ... lovely,
... of good report; if there be any virtue, and if there be any praise,
think on these things.
     - motto - Northwestern University (Phil 4:8)




From fgeiger at datec.at  Mon Mar 12 02:25:27 2001
From: fgeiger at datec.at (Franz GEIGER)
Date: Mon, 12 Mar 2001 08:25:27 +0100
Subject: question , Does pytyon excist on a pda, like psion?
References: <xDJq6.28348$KE1.1127955@zwoll1.home.nl>
Message-ID: <98htoa$a87$1@newsreaderm1.core.theplanet.net>

I tried to install Jython on my Psion but yet did not succeed. There are
some special considerations installing Java programs on EPOC. But I'm not a
Java programmer...

Regards
Franz

"SonoBull" <me at home.nl> wrote in message
news:xDJq6.28348$KE1.1127955 at zwoll1.home.nl...
> Hello all,
>
> Is Python already ported to the psion or other pda, s?
>
> greetings.
>
> Sono.
>
>
>
>




From qrczak at knm.org.pl  Sat Mar 31 16:00:20 2001
From: qrczak at knm.org.pl (Marcin 'Qrczak' Kowalczyk)
Date: 31 Mar 2001 21:00:20 GMT
Subject: list vs tuple
References: <hGlw6.782$p5.3210@news1.rivrw1.nsw.optushome.com.au><3AC20CD3.265A2D94@alcyone.com><ru69cto61jtmrqsf809e38hb3pbm2lhi7g@4ax.com><bC1x6.4319$4N4.905946@newsc.telia.net><71q9ct4le63anefmiajj5cbh3mg04qu1ga@4ax.com><mailman.985991202.5578.python-list@python.org><7Bfx6.989$p5.3791@news1.rivrw1.nsw.optushome.com.au><Ukhx6.4421$4N4.950603@newsc.telia.net> <mailman.986049281.32365.python-list@python.org> <uGqx6.1052$p5.3997@news1.rivrw1.nsw.optushome.com.au>
Message-ID: <slrn9cchb4.3oa.qrczak@qrnik.zagroda>

Sat, 31 Mar 2001 19:59:22 GMT, deadmeat <root@[127.0.0.1]> pisze:

> but thats not the point - which when the statement a = b copies
> the data across for one type it's expected to do it for others.
> It doesn't, so it's inconsistant, contrary to the original claim.

It *never* copies the data across, thus it is consistent.

-- 
 __("<  Marcin Kowalczyk * qrczak at knm.org.pl http://qrczak.ids.net.pl/
 \__/
  ^^                      SYGNATURA ZAST?PCZA
QRCZAK


From aaldv97 at student.removethispart.vxu.se  Mon Mar 26 09:23:24 2001
From: aaldv97 at student.removethispart.vxu.se (David Andreas Alderud)
Date: Mon, 26 Mar 2001 16:23:24 +0200
Subject: Lists and Files
Message-ID: <99njdo$oa3$1@news.lth.se>

I'm new to Python, started yesterday, is there a good way to handle lists
and files?

For example:

    stooges = ['val1', 'val2', 1, [['val2_1', 'val2_2', 0], ['val2_1',
'val2_2', 0]]]
    fds = open("stooges", "w")
    # fds.write(stooges) doesn't work, maybe I can encapsulate and pack it
to some other format, other than str that is, one that works?
    fds.write(str(stooges))
    fds.close()
    fds = open("stooges", "r")
    no_longer_stooges = list(fds.read())
    print no_longer_stooges
        =>
            ['[', "'", 'v', 'a', 'l', '1', "'", ',', ' ', "'", 'v', 'a',
'l', '2', "'", ',', ' ', '1', ',', ' ', '[', '[', "'", 'v', 'a', 'l', '2',
'_', '1', "'", ',', ' ', "'", 'v', 'a', 'l', '2', '_', '2', "'", ',', ' ',
'0', ']', ',', ' ', '[', "'", 'v', 'a', 'l', '2', '_', '1', "'", ',', ' ',
"'", 'v', 'a', 'l', '2', '_', '2', "'", ',', ' ', '0', ']', ']', ']']

and that's not what I want, I could write a parser for the dynamic data but
I rather not if there is a good way to handle it.
Someone _must_ have stumbled across this problem before me, as it seams like
a common thing to do, and solved it, so I rather not duplicate the effort,
any hits and pointers or full solutions?

/Kind regards,
    David A. Alderud




From loewis at informatik.hu-berlin.de  Sat Mar 24 06:26:53 2001
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 24 Mar 2001 12:26:53 +0100
Subject: Is setup.py a "good thing"? (long)
References: <3aba76e2.167083508@24.0.228.33> <3dvgp1zasw.fsf@ute.cnri.reston.va.us> <99fsdu$p2u$1@hop.gmd.de> <3dpuf8zb4t.fsf@ute.cnri.reston.va.us>
Message-ID: <j4zoebmmpu.fsf@informatik.hu-berlin.de>

Andrew Kuchling <akuchlin at mems-exchange.org> writes:

> woetzel at gmd.de (Gerd Woetzel) writes:
> > Modules/Setup has buit zlibmodule.so and socketmodule.so which worked
> > fine during the test phase, but setup.py has (secretly?) created
> > zlib.so (socket.so) too! Finally "make install" has installed all of them.
> 
> That certainly shouldn't have happened.  Can you reproduce the problem?

It could happen if you use *shared* in Modules/Setup and if setup.py
is older than 1.30; before, it would use on builtin_module_names to
exclude modules. So that was a problem once and should be fixed since
2.1b1.

Regards,
Martin


From qrczak at knm.org.pl  Sun Mar 18 18:15:02 2001
From: qrczak at knm.org.pl (Marcin 'Qrczak' Kowalczyk)
Date: 18 Mar 2001 23:15:02 GMT
Subject: Who's minister of propaganda this week?
References: <B98041329020E007.7DA84CAD9D0562AB.817BBC572F480869@lp.airnews.net> <98nu1m$12e@dispatch.concentric.net> <660ACCA1E4816F7E.C359CDF4168EF3FB.3B6B05538D619770@lp.airnews.net> <3AAFA13D.76E32C3F@divalsim.it> <98oa4o01i4v@news2.newsguy.com> <mailman.984619039.7856.python-list@python.org> <98q9h60122v@news2.newsguy.com> <mailman.984937946.5318.python-list@python.org> <9937en0bge@news2.newsguy.com>
Message-ID: <slrn9bagbm.toq.qrczak@qrnik.zagroda>

Sun, 18 Mar 2001 21:55:59 +0100, Alex Martelli <aleaxit at yahoo.com> pisze:

> For example, say that you want to organize certain objects in a
> container/contained hierarchy, with no nesting limits.
> 
> All existing objects will be leaves in the resulting tree.  Why does
> this imply that such existing objects need to inherit from a certain
> interface asserting "I may or may not be a container" (perhaps with
> a default implementation stating "I am _not_ a container after all")?

It doesn't imply that.

> In a dynamically typed language, you will define an interface
> (and perhaps a default implementation thereof) *only for container
> objects*, i.e., the "new" organizing superstructure.  Each container
> will hold a set (e.g. a list, or dict) of (suitably tagged)
> references to other objects; for recursive walks on the tree, one
> will dynamically test whether a given object is-a container (else,
> it's a leaf).

In a statically typed language I would just recognize leafs on the
container side, without relying on any properties of objects in
leafs themselves.

You solved a problem which did not exist.

-- 
 __("<  Marcin Kowalczyk * qrczak at knm.org.pl http://qrczak.ids.net.pl/
 \__/
  ^^                      SYGNATURA ZAST?PCZA
QRCZAK


From robin.thomas at starmedia.net  Fri Mar 30 11:58:15 2001
From: robin.thomas at starmedia.net (Robin Thomas)
Date: Fri, 30 Mar 2001 11:58:15 -0500
Subject: best practice? null/None vs 'None' vs '' (in web/sql)
In-Reply-To: <9a2b9v+lrgp@eGroups.com>
Message-ID: <4.3.1.2.20010330113737.00e39ab0@exchange.starmedia.net>

At 04:12 PM 3/30/01 +0000, fluxent at yahoo.com wrote:
>(Python 2.x, Win2K, IIS5)
>
>I've got some simple pages for viewing and editing SQL records via
>Python in ASP.
>
>I would really like to distinguish between fields with null values
>and fields that might actually include the text string 'None'. Right
>now both cases come out from a SELECT as showing up as 'None', and
>then when I submit and edit they definitely get changed to 'None', so
>I end up with no nulls.
>
>I can see that the ODBC result set distinguishes them. But of course
>when things get rendered out to HTML you don't want to see single
>quotes around every string value, so 'None' and None become the same
>thing.

This about the limitations of "query arguments" to HTTP requests, not about 
any limitations of ODBC, Python, or ODBC/Python interfaces.

HTTP query arguments are of one type: STRING. There is no concept of a 
"null" value; the closest thing you have is an "empty" string like "". In 
most database engines, there is a clear difference between NULL and "" in 
any textual or character data column.

If your HTTP-based application wishes to consider some special string value 
as meaning "null", then you must choose which special string value will be 
considered "null". Then you must have your application transform data in 
two ways: transforming the special string to the null value (when 
processing HTTP request input), and transforming the null value to the 
special string value (on output of HTTP response).



>So, I guess this isn't a platform bug, it's a nasty area calling for
>some sort of interface standard. Is there a best practice I should
>adopt?

The best practice is consistency in your application. Pick a special value 
and go with it.

The "lack of standard" is instead a lack of (or lack of need for) as 
built-in string-to-null-value conversion function. Python has int() and 
float() and others to convert strings to numeric objects. But a null() 
function is overkill. You can write it yourself:

if a == "": a = None

If you instead wish for a magical standard "convert a string into the 
neatest possible object" function, the magical standard may not do what you 
ask. Consider:

magic_function("")
magic_function("None")
magic_function('"is this string quoted?"')
magic_function("[]")
magic_function("{'a': 5}")
magic_function("6L") # form submitter's apartment number, not a long integer
magic_function("11201") # zip code for ZIP_CODE VARCHAR column
magic_function("6K5 B1E") # Canadian zip code for ZIP_CODE VARCHAR column

Would these do the conversions you want in your application? In the case of 
"6L" as apartment number in an address form in HTML, definitely not.

Because the policies of string-to-object conversion are so 
application-dependent, Python does not impose a magical standard that may 
cuase you headaches. Instead, it gives you specific type-conversion 
function like int() and float(), and lets you do the rest.

>Maybe (a) transform the outputs of each SELECT to a different
>representation of None (either '' or 'null'?), then (b) transform all
>UPDATE query strings before executing them to change any ='' (or
>='null') to =NULL?

Yes. You've stated above that you do *not* wish to have "None" be the 
special string in your application. So you can pick something else. Many 
people choose the empty string "", which means that no input argument via 
HTTP can really be an empty string; it will always get converted to 
null/None by your application.

def output(object):
     if object is None: return ""
     else: return str(object)

def input(string):
     if not string: return None
     else:
         # maybe do other conversions to int() or float()
         return string # or use string unchanged


--
Robin Thomas
Engineering
StarMedia Network, Inc.
robin.thomas at starmedia.net




From ct at gocept.com  Fri Mar 23 15:48:51 2001
From: ct at gocept.com (Christian Theune)
Date: Fri, 23 Mar 2001 21:48:51 +0100
Subject: Curses under python 2.0
References: <99darm$o3o$00$1@news.t-online.com> <3dsnk5zars.fsf@ute.cnri.reston.va.us>
Message-ID: <99gcrj$b43$03$1@news.t-online.com>

Ok. I did. Now.

Tnx & best regards

Christian Theune

PS: why didn't i check this out before .... stupid baby myself ...

In article <3dsnk5zars.fsf at ute.cnri.reston.va.us>, "Andrew Kuchling"
<akuchlin at mems-exchange.org> wrote:

> "Christian Theune" <ct at gocept.com> writes:
>> >>> import curses
>> Traceback (most recent call last):
>>   File "<stdin>", line 1, in ? File
>>   "/usr/local/lib/python2.0/curses/__init__.py", line 15, in ?
>>     from _curses import *
>> ImportError: No module named _curses
> 
> You *did* enable the _curses module in Modules/Setup, right?
> 
> --amk


From dbrueck at edgix.com  Wed Mar 21 13:17:36 2001
From: dbrueck at edgix.com (Dave Brueck)
Date: Wed, 21 Mar 2001 11:17:36 -0700
Subject: re.match
Message-ID: <066601c0b233$3db4f700$450514ac@PRODUT2KDAVE>

"Bob Cannard" <bob_cannard at mentor.com> wrote in message news:3AB8E6DF.A2C32AD8 at mentor.com...
>    - Is this string a Thing? (re.match - er, no, that doesn't work,
>      it only asks if this string begins with a Thing.)
> 

Hi Bob,

'$' matches 'end of string', so you _can_ use re.match to test "Is this string a Thing".

-Dave

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20010321/4115aaf1/attachment.html>

From ransen_spam_me_not at nemo.it  Fri Mar  2 11:04:54 2001
From: ransen_spam_me_not at nemo.it (Owen F. Ransen)
Date: Fri, 02 Mar 2001 16:04:54 GMT
Subject: PyRun_SimpleFile - impossible to call it under Windows?
Message-ID: <3aa07704.2672581@news.newsguy.com>

I've seen that there is an odd bug in PyRun_SimpleFile
probably because of version incompatibilities in the
FILE structure and using DEBUG modes in VC6

I'd like to call PyRun_SimpleFile in the debug version
of my VC6 program, is it impossible currently with Python 2?

TIA

Owen

--
Owen F. Ransen
http://www.ransen.com/
Home of Gliftic & Repligator Image Generators


From andreas at andreas-jung.com  Wed Mar 28 07:30:38 2001
From: andreas at andreas-jung.com (Andreas Jung)
Date: Wed, 28 Mar 2001 07:30:38 -0500
Subject: dynamic web pages?
References: <99q701$jlr10@news-1.bank.dresdner.net> <3AC0B262.6F0E30DE@gmx.net>
Message-ID: <99tvno$d0k$1@slb7.atl.mindspring.net>

And we have the small gun: PMZ (Poor Men's Zope): http://pmz.sourceforge.net

Andreas
"Robert Kiendl" <rkiendl at gmx.net> schrieb im Newsbeitrag
news:3AC0B262.6F0E30DE at gmx.net...
> we have the BIG GUN here: Zope (www.zope.org) dynamic app.server in
> native python logic including an OODBMS!
>
> or use cgi programming with apache





From tjg at exceptionalminds.com  Tue Mar 20 13:37:54 2001
From: tjg at exceptionalminds.com (Timothy Grant)
Date: Tue, 20 Mar 2001 10:37:54 -0800
Subject: Saving history in interactive Python
Message-ID: <20010320103754.V3109@trufflehunter.avalongroup.net>

Is there a way to save the history list in the Python
interpreter.

-- 
Stand Fast,
    tjg.

Timothy Grant                         tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.         <><       (503) 246-3630
>>>>>>>>>>>>>Linux, because rebooting is *NOT* normal<<<<<<<<<
>>>>This machine was last rebooted:  62 days 22:52 hours ago<<



From stephen_purcell at yahoo.com  Mon Mar 26 03:56:55 2001
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Mon, 26 Mar 2001 10:56:55 +0200
Subject: file locking
In-Reply-To: <5zCv6.583$p5.2667@news1.rivrw1.nsw.optushome.com.au>; from root@excalibur.iks-jena.de on Mon, Mar 26, 2001 at 07:52:33AM +0000
References: <5zCv6.583$p5.2667@news1.rivrw1.nsw.optushome.com.au>
Message-ID: <20010326105655.A4191@freedom.puma-ag.com>

deadmeat wrote:
> How do I lock a file on opening it?  the bultin open() only allows me to set
> the read/write mode, not any deny modes.

On Unix, look at the 'flock' function in the module 'fcntl'.

You can also try the 'posixfile' module -- posixfile objects support locking.

-Steve

-- 
Steve Purcell, Pythangelist
Get testing at http://pyunit.sourceforge.net/
Any opinions expressed herein are my own and not necessarily those of Yahoo



From sholden at holdenweb.com  Wed Mar  7 19:15:48 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Wed, 7 Mar 2001 19:15:48 -0500
Subject: gah! I hate the new string syntax
References: <mailman.983615170.23313.python-list@python.org> <tN5o6.2746$eH3.14913@NewsReader> <Koco6.7518$1D5.244602@e420r-atl1.usenetserver.com> <mailman.983997252.5628.python-list@python.org>
Message-ID: <meAp6.29961$ML1.1514835@e420r-atl2.usenetserver.com>

"D-Man" <dsh8290 at rit.edu> wrote in message
news:mailman.983997252.5628.python-list at python.org...
> On Sat, Mar 03, 2001 at 03:19:05PM -0500, Steve Holden wrote:
[ ... ]
>
> What about one who is both English and Scots, yet never set foot on
> that island ;-)?
>
No such animal!

regards
 Steve





From matt at mondoinfo.com  Wed Mar 21 14:52:45 2001
From: matt at mondoinfo.com (Matthew Dixon Cowles)
Date: Wed, 21 Mar 2001 19:52:45 GMT
Subject: Substitute for join()
References: <Mo7u6.58$Hp.2896@typhoon.aracnet.com>
Message-ID: <slrn9bi1kd.n9.matt@happy-hour.mondoinfo.com>

On Wed, 21 Mar 2001 11:35:53 -0800, Daniel Klein <DanielK at jBASE.com>
wrote:

>I've got to make a script (written in version 2.0) to work on a machine
>running 1.5.2. The only incompatible line is one that uses the
>"".join(mylist) method.

>>> import string
>>> string.join(mylist,"")

Will do what you want.

Regards,
Matt


From tim.one at home.com  Fri Mar 16 19:12:15 2001
From: tim.one at home.com (Tim Peters)
Date: Fri, 16 Mar 2001 19:12:15 -0500
Subject: list.sort(cmpfunc) question
In-Reply-To: <3ab2a132@si-nic.hrz.uni-siegen.de>
Message-ID: <LNBBLJKPBEHFEDALKOLCIEDGJGAA.tim.one@home.com>

[Simon Budig]
> ...
> Is the sort-function guaranteed to be stable? Does it keep the order,
> when cmpfunc returns 0 for a pair of items?

No, Python's list.sort() is not stable.  Stability can be "faked", though, by
sorting a derived list y, where each element y[i] == (x[i], i).  Then no two
list elements are equal.  Because tuples are compared lexicographically,
cmp((x, i), (x, j)) has the same sense as cmp(i, j), thus preserving the
original order of equal elements in x.

Caution:  if a list is "short enough", list.sort() falls back to a binary
insertion sort, and IIRC that *is* stable.  So trying out small examples may
fool you.

forewarned-is-forearmed-ly y'rs  - tim




From belrick at home.com  Sun Mar 11 15:58:59 2001
From: belrick at home.com (Bruce Elrick)
Date: Sun, 11 Mar 2001 20:58:59 GMT
Subject: Einstein's Riddle
References: <oNPq6.396644$ge4.138479566@news2.rdc2.tx.home.com> <98glrd$1si1a$1@ID-75892.news.dfncis.de>
Message-ID: <3AABE712.499CCF4C@home.com>

I swear there is no evidence that any of them own a fish.  The German
owns an emu.

Bruce

Boris Popov wrote:
> 
> "Steve Graham" <js.graham at home.com> wrote in message
> news:oNPq6.396644$ge4.138479566 at news2.rdc2.tx.home.com...
> > Who would be interested in using his/her brain (and his computer) to solve
> > the following?
> >
> >
> > Steve Graham
> >
> > ===
> >
> > Einstein's Riddle
> >
> > Albert Einstein wrote this riddle this century [ed. 20th century].  He
> said
> > 98% of the world could not solve it.
> >
> > There are 5 houses in 5 different colors.  In each house lives a person
> with
> > a different nationality.  The 5 owners drink a certain type of beverage,
> > smoke a certain brand of cigar, and keep a certain pet.  No owners
> > have the same pet, smoke the same brand of cigar or drink the same
> beverage.
> >
> > The question is: "Who owns the fish?"
> >
> > Hints:
> >
> > The Brit lives in the red house.
> >
> > The Sweed keeps dogs as pets.
> >
> > The Dane drinks tea.
> >
> > The green house is on the left of the white house.
> >
> > The green house's owner drinks coffee.
> >
> > The person who smokes Pall Mall rears birds.
> >
> > The owner of the yellow house smokes Dunhill.
> >
> > The man living in the center house drinks milk.
> >
> > The Norwegian lives in the first house.
> >
> > The man who smokes Blends lives next to the one who keeps cats.
> >
> > The man who keeps the horse lives next to the man who smokes Dunhill.
> >
> > The owner who smokes Bluemasters drinks beer.
> >
> > The German smokes Prince.
> >
> > The Norwegian lives next to the blue house.
> >
> > The man who smokes Blends has a neighbor who drinks water.
> >
> 
> I don't beleive he said that only 2% of the world could solve it. It's not
> hard, you just need to pay attention to the details. The German owns the
> fish. It was supposed to be a table that I used to layout all the details,
> but I prefer to post plain-text:
> 
> house         1 2 3 4 5
> color            yellow blue red green white
> nationality   norweigian dane brit german swede
> drink            water tea milk coffee beer
> smoke        dunhill blend pall mall prince bluemaster
> pet              cats horse birds fish dogs
> 
> -Boris
> P.S. Did I have to write software to solve it ? :)


From aahz at panix.com  Mon Mar 26 18:41:48 2001
From: aahz at panix.com (Aahz Maruch)
Date: 26 Mar 2001 15:41:48 -0800
Subject: Namespaces question
References: <mailman.985627029.26291.python-list@python.org>
Message-ID: <99ok3s$mkg$1@panix3.panix.com>

In article <mailman.985627029.26291.python-list at python.org>,
Chris Withers  <chrisw at nipltd.com> wrote:
>
>def create(self):
>
>    for name in ['x','y','z']:
>        value = session.get(name,None)
>        if value is None:
>            errors.append('No %s supplied.' % name)
>        else:
>            locals()[name]=value
>
>    if errors:
>        return errors
>
>    print x
>
>...now I know it's not right, 'cos I get a NameError on x at the print
>statement.
>
>What should I be using instead of locals()?

Just use a dictionary: "print dict['x']" works fine
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

"Boost the stock market -- fire someone"


From nadavh at envision.co.il  Wed Mar  7 02:27:23 2001
From: nadavh at envision.co.il (Nadav Horesh)
Date: Wed, 07 Mar 2001 09:27:23 +0200
Subject: NumPy's undocumented type code
Message-ID: <3AA5E2DB.45BFD417@envision.co.il>

An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20010307/e8873680/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: nadavh.vcf
Type: text/x-vcard
Size: 360 bytes
Desc: Card for Nadav Horesh
URL: <http://mail.python.org/pipermail/python-list/attachments/20010307/e8873680/attachment.vcf>

From peter at schneider-kamp.de  Fri Mar  9 14:02:20 2001
From: peter at schneider-kamp.de (Peter Schneider-Kamp)
Date: Fri, 09 Mar 2001 20:02:20 +0100
Subject: "!=" is bad form. Re: sorry....never mind
References: <BLEOJGEMBNBNEAKMIFDHEELCPLAA.dbrueck@edgix.com>
Message-ID: <3AA928BC.1DAB4284@schneider-kamp.de>

Dave Brueck wrote:
> 
> I don't really care if Python keeps <>, but the reason I don't like it is
> because of the example listed above: to me seeing "<>" implies something
> about the ordering of a and b, namely (a < b) or (a > b), even when such an

Well, Python *does* define an ordering of a and b, no matter what
a and b may be. E.g.:

>>> 1 > (1,2)
0
>>> 1 < (1,2)
1
>>> {"a":1} < u"hello world"
1
>>> {"a":1} > u"hello world"
0

So, yes, interpreting "a <> b" as "(a < b) or (a > b)" makes
perfect sense in Python.

not-sure-comparing-strings-and-dictionaries-is-a-sane-thing-to-do-ly
y'rs
Peter



From Denys.Duchier at ps.uni-sb.de  Sat Mar 24 18:13:20 2001
From: Denys.Duchier at ps.uni-sb.de (Denys Duchier)
Date: 25 Mar 2001 00:13:20 +0100
Subject: [Python-Dev] Anyone using weakrefs?
References: <mailman.985469191.25050.python-list@python.org>
Message-ID: <is7l1ekbfz.fsf@elk.ps.uni-sb.de>

"Tim Peters" <tim.one at home.com> writes:

> Still, I've been happily playing w/ this little module:
> 
> import weakref
> id2obj_dict = weakref.mapping()
> 
> def remember(obj):
>     id2obj_dict[id(obj)] = obj
> 
> def id2obj(id):
>     return id2obj_dict.get(id, None)

If memory serves, in T3 (Yale Scheme) the printed representation of an
object included an integer id which you could use as a key to look up
the object in a weak table.  Quite convenient for interactive use.

Cheers,

-- 
Dr. Denys Duchier			Denys.Duchier at ps.uni-sb.de
Forschungsbereich Programmiersysteme	(Programming Systems Lab)
Universitaet des Saarlandes, Geb. 45	http://www.ps.uni-sb.de/~duchier
Postfach 15 11 50			Phone: +49 681 302 5618
66041 Saarbruecken, Germany		Fax:   +49 681 302 5615


From phd at phd.pp.ru  Tue Mar 13 12:18:39 2001
From: phd at phd.pp.ru (Oleg Broytmann)
Date: Tue, 13 Mar 2001 20:18:39 +0300 (MSK)
Subject: Sortin dictionaries in python
In-Reply-To: <3aae4757.68617130@News.CIS.DFN.DE>
Message-ID: <Pine.LNX.4.30.0103132017170.7032-100000@fep132.fep.ru>

On Tue, 13 Mar 2001 costas at springmail.com wrote:
> Is there a simple way to sort a dictionary? It seems the only way to
> do this is by indirectly creating  a sorted list.

   There is no.

> And while on the topic of sorting lists. I tried sorting a list and
> wanted to to pass a compare function to the list.sort() method. The
> documentaion gives no example.
>
> import string
>
> x=['a','c','B']
>
> def ignoreCase(left, right):
>      if string.upper(left) < string.upper(right):
>          return -1
>      else:
>          return 1
>
> x.sort(ignoreCase(left, right))

   You are close:

x.sort(ignoreCase)

   Pass the function, not call it :)

   BTW, ignoreCase (like any other comparison function) should return 0 if
left == right!

Oleg.
----
     Oleg Broytmann            http://phd.pp.ru/            phd at phd.pp.ru
           Programmers don't die, they just GOSUB without RETURN.




From stevewilliams at wwc.com  Tue Mar 13 11:11:13 2001
From: stevewilliams at wwc.com (Steve Williams)
Date: Tue, 13 Mar 2001 16:11:13 GMT
Subject: This math scares me
References: <3aad4062.1290021@News.CIS.DFN.DE> <Hwbr6.1139$Tg.127571@ruti.visi.com> <3aad4485.2348499@News.CIS.DFN.DE> <DZbr6.17917$Qb7.3117518@newsb.telia.net> <zrcr6.1144$Tg.129254@ruti.visi.com>
Message-ID: <3AAE46B5.963FC3CE@wwc.com>

Grant Edwards wrote:

> In article <DZbr6.17917$Qb7.3117518 at newsb.telia.net>, Fredrik Lundh wrote:
> >costas at springmail.com wrote:
> >> I will also tell my customers that this is the new math.
> >
> >if you don't understand how floating point numbers work, you
> >probably shouldn't use them in customer projects.
>
> Fredrik is right.  Using floating point numbers requires that
> you know what you're doing.  They are _not_ exact.  If your
> customers are expecting exact results and you're using floating
> point numbers, then you did a bad thing.
>
> People should have to pass a test before they're allowed to use
> floating point numbers.    1/2 ;)
>

[snip]

Languages supporting serious commercial numeric processing have the dreaded
static typing.  Viz:

    Picture 99v99 in COBOL

or

    Decimal(4,2) in DDL

That way you 'do the right thing' in your *intermediate* calculations without
having to sit the exam for floating point or remembering to 'adjust the scale
and precision' for fixed point.

And using print formatting to solve the problem is [wait for it] missing the
point.




From MarkH at ActiveState.com  Fri Mar 23 18:15:44 2001
From: MarkH at ActiveState.com (Mark Hammond)
Date: Fri, 23 Mar 2001 23:15:44 GMT
Subject: Python Script to dial a VPN connection, automate VPN routing table updates.
References: <FCNu6.66711$A6.13912607@news4.rdc1.on.home.com>
Message-ID: <3ABBDA6D.2000305@ActiveState.com>

Warren Postma wrote:

> Sorry for the nearly-off-topic nature of this post, but ....
> 
> I work from home now, and there's something in Windows 2000 that's
> annoyingly manual:
> 
> I have a permanent internet connection via @home cable modem, and I "dial"
> through that to create a VPN connection to the office.  The problems are
> two:
> 
> (1) Windows won't accept or save my userid, password and the domain name, so
> I have to enter it manually.
> 
> (2) After connecting, I have two default gateways in my routing tables
> (ROUTE PRINT) and I need to update it so only the traffic that is destined
> to the office goes through that interface.  The IP addresses I get given are
> different each time, making a batch file insufficient, so voila, a perfect
> use for python.

I have very similar problems.  Here is a script I wrote for the purpose. 
  It doesnt handle authentication sa Windows does remember the password 
etc for me.

When run with a command-line option being the VPN connection name, it 
establishes a connection to the named site.

Once connected, or if an argument is not given it attempts to find the 
IP of the VPN connection and manually add a route for the VPN.

Thus, I generally use it with no options after manually connecting - not 
for any good reason actually, just that is the way I do it :)

import os
import sys
import win32ras

def Connect(rasEntryName, numRetries = 5):
     # See if already connected
     for info in win32ras.EnumConnections():
         if info[1] == rasEntryName:
             print "Already connected to '%s'" % (rasEntryName,)
             return 1
     dial, have_pw = win32ras.GetEntryDialParams(None, rasEntryName)
     if not have_pw:
         print "Error: The password is not saved for this connection"
         print "Please connect manually selecting the 'save password' 
option and try again"
         sys.exit(1)
     retryCount = numRetries
     bValid = 0
     print "Connecting to:", rasEntryName
     while retryCount > 0:
         rasHandle, errCode = win32ras.Dial(None, None, dial, None)
         if win32ras.IsHandleValid(rasHandle):
             bValid = 1
             break
         print "Retrying..."
         win32api.Sleep(5000)
         retryCount = retryCount - 1
     return bValid

def FixRoute():
     data = os.popen("ipconfig", "r").readlines()
     have_ppp = 0
     ip_str = None
     for line in data:
         if line.startswith("PPP"):
             have_ppp = 1
         if have_ppp and line.strip().startswith("IP Address"):
             ip_str = line.split(":")[1].strip()

     if ip_str is None:
         print "ERROR: could not find the PPP ip address"
     else:
         print "Your PPP IP address is", ip_str
         cmd = "route add 192.168.2.0 mask 255.255.255.0 %s" % (ip_str,)
         print cmd
         os.system(cmd)
         print "Route updated."

if len(sys.argv) == 2:
     if not Connect(sys.argv[1]):
         print "Could not connect to '%s'" % (sys.argv[1],)
         sys.exit(1)
else:
     print "No connection name specified - just adding route..."
FixRoute()



From dnew at san.rr.com  Sun Mar 18 17:15:34 2001
From: dnew at san.rr.com (Darren New)
Date: Sun, 18 Mar 2001 22:15:34 GMT
Subject: Who's minister of propaganda this week?
References: <B98041329020E007.7DA84CAD9D0562AB.817BBC572F480869@lp.airnews.net> <98nu1m$12e@dispatch.concentric.net> <3AAFA861.C773D977@san.rr.com> <992qbh$et7$1@nntp.Stanford.EDU>
Message-ID: <3AB53387.9AEF272D@san.rr.com>

> Don't underestimate convenience.  :-) If you had to declare each
> string value in your program as a class like the above, it would make
> strings less useful in a practical sense although they'd still be
> equivalent in a theoretical sense. 

True. On the other hand, there are things like co-routines which, to get the
same effect, you have to basically implement co-routines rather than just
use an inconvenient syntax. I think there's actually a rather big difference
between "inconvenience" and "impossible".  :-)

-- 
Darren New / Senior MTS & Free Radical / Invisible Worlds Inc.
San Diego, CA, USA (PST).  Cryptokeys on demand.
A million monkeys in a room with a million typewriters 
              will only yield half a million pregnant monkeys.


From phlip_cpp at my-deja.com  Wed Mar 14 09:08:54 2001
From: phlip_cpp at my-deja.com (Phlip)
Date: 14 Mar 2001 14:08:54 GMT
Subject: Who's minister of propaganda this week?
References: <B98041329020E007.7DA84CAD9D0562AB.817BBC572F480869@lp.airnews.net>
Message-ID: <98nu1m$12e@dispatch.concentric.net>

Proclaimed Cameron Laird from the mountaintops:

> Sure; Bruce "Mr. Java" Eckel says Python's his favorite language.
> That was last week's news.  What would you think if the former
> editor of the *C++ report* gave his
>     prediction for this decade: Keep an
>     eye on languages like Python, Ruby,
>     and Smalltalk. They are likely to
>     become extremely important.

Robert C. Martin meant two things. A> If you write wall-to-wall unit tests 
when you code, if you use a typeless language you will add bullet-proof 
features faster than if you use a statically typed language, even though 
the former has less built-in error checking at type conversion time. The 
latter supports refactoring easier.

B> The full depth of behaviors we should expect from a typeless language 
have yet to be fathomed. Ruby trumps Python in the simple matter of 
closures (imagine my shock when I discovered Python didn't have them). But 
a language "like" those might win the decade.

> I heard you wiseacres in the back there, with the, "But Python
> and Smalltalk *already* are extremely important" comment.

Many pointy-haired bosses are unaware of the former, and think 
the latter is some academic curiosity. That's why selling Java (a 
statically typed language) to these bosses was so easy.

-- 
  Phlip                          phlip_cpp at my-deja.com
============== http://phlip.webjump.com ==============
  --  Please state the nature of the programming emergency  --


From adaminthedomaindevtty.net  Sat Mar  3 01:15:45 2001
From: adaminthedomaindevtty.net (Adam Logghe)
Date: Fri, 2 Mar 2001 22:15:45 -0800
Subject: Editing Python source as XML?
References: <mailman.983546704.10789.python-list@python.org>
Message-ID: <3a9fe475_1@news.nwlink.com>

It sounds like the feature you are really looking for is frequently called
"code folding".

I also find it to be very useful. You have many options in editors
unfortunately IDLE does not yet support code folding.

www.activestate.com has a project called "Komodo" which is written with a
combination of the Mozilla codebase and Python (my understanding is that
originally this was also to have a substantial portion written in Perl but
the developers could code so much faster in Python that Perl ended up as a
minor part). Despite some primitive features I have begun to use this
full-time, I find it the easiest to use for editing multiple files. I'm
hoping they start supporting XML folding as well.

www.activestate.com also supports "PythonWin" which is basically a Windows
specific enhanced version of  IDLE.

http://archaeopteryx.com/ is selling an Editor for Linux which also supports
code folding and looks very capable. It is available for free if you are
using it to develop free software.

I know I am  missing one more fairly significant editor, I am sure someone
else will point that out (extra points for not mentioning vi or emacs).

By the way, there are several shareware editors available that support
Python and code folding. Search on hotfiles.com

Adam
in the domain devtty.net


"Steven Haryanto" <steven at haryan.to> wrote in message
news:mailman.983546704.10789.python-list at python.org...
> A recent experience of having to modify a long C source file has
> led me to wonder whether there are any tool/effort towards editing
> Python source code as XML or other form of structured document.
>
> What I would like is an editor that presents the source file as a
> tree outline, where I can collapse selected class or function
> definition body, certain long if's/for's, some section of code,
> view/hide the asserts, etc. So then I only have to deal with the
> section I am concentrating with and don't have look at other
> distracting parts (especially in C with all those #ifdef/#endif's,
> argh...).
>
> I visited conglomerate.org some months ago, and it seems the
> project has been abandoned.
>
> Thanks,
> Steve
>
>




From fig at monitor.net  Wed Mar 21 19:17:32 2001
From: fig at monitor.net (Stephen R. Figgins)
Date: Wed, 21 Mar 2001 16:17:32 -0800
Subject: Puzzling input problem
Message-ID: <3AB9449C.A1D14738@monitor.net>

I ran into a puzzling problem combining print with raw_input(). 
Normally I would pass the prompt string to raw_input, but instead I
tried something like this: 

  print "prompt string: ",
  answer = raw_input()
  print answer

When it prints answer, it is indented one space.  Why is raw_input()
putting a space in there?  Is it a bug, a feature, an unintended side
effect?  input() does this too.   I am currently using Active Python 2.0
build 202.

When I tried it with Jython 2.0, I didn't get the extra space. 


Stephen


From dwig at advancedmp.net  Mon Mar 19 13:53:45 2001
From: dwig at advancedmp.net (Don Dwiggins)
Date: 19 Mar 2001 10:53:45 -0800
Subject: windows script host and python
In-Reply-To: Tim Roberts's message of "Sat, 17 Mar 2001 20:57:12 -0800"
References: <%tKr6.2570$pX.26174@news1.onlynews.com>
	<98oq3r$1e0$1@troll.powertech.no> <8Aps6.104$pl.1515@newsfeed.slurp.net>
	<98tc4d0rmk@news1.newsguy.com> <lvf8bt4p6e6naq0vimgnkhs175eoqbajfr@4ax.com>
Message-ID: <uofuxa8ue.fsf@advancedmp.net>

Tim Roberts writes:
> "Alex Martelli" <aleaxit at yahoo.com> wrote:
>> 
>> Here, for example, is the "list all running processes" WMI scripting
>> example translated into Python:
>> 
>> from win32com.client import GetObject
>> 
>> processes = GetObject("winmgmts:{impersonationLevel=impersonate}").\
>> InstancesOf("Win32_process")
>> 
>> for process in processes:
>> print process.Name

> Cool!  Is this supposed to be NT-only?  It doesn't fly on my Win98 box.

Actually, it fails on my NT workstation also:

Traceback (most recent call last):
  File "<stdin>", line 3, in ?
  File "e:\python20\win32com\client\__init__.py", line 72, in GetObject
    return Moniker(Pathname, clsctx)    
  File "e:\python20\win32com\client\__init__.py", line 87, in Moniker
    moniker, i, bindCtx = pythoncom.MkParseDisplayName(Pathname)
pywintypes.com_error: (-2147221020, 'Invalid syntax', None, None)

Just FYI, I haven't tried to debug it.

-- 
Don Dwiggins                    "Solvitur Ambulando"
Advanced MP Technology
dwig at advancedmp.net




From alex at server01.shindich.com  Sun Mar 25 19:01:54 2001
From: alex at server01.shindich.com (Alex Shindich)
Date: Mon, 26 Mar 2001 00:01:54 +0000 (UTC)
Subject: Yet Another PEP:  Interface Adapter Mechanism __adapt__
Message-ID: <Pine.LNX.4.21.0103251549140.8511-100000@server01.shindich.com>

It sounds like your PEP suggests adding Strategy pattern behavior to all
Python objects. My question is WHY? Implementing strategy pattern,
i.e. asking object if it supports a particular interface, is trivial. In
fact, you can do it easily by using introspection, i.e. hasattr (obj,
'blah').
There are other ways of implementing strategies -- flyweights, query
interface, etc.

Another way of looking at your PEP is to say that you are asking for
Python Java-like interfaces. In my humble opinion, interface driven
programming is not as needed in Python because of introspection. Instead
of asking if the object supports a set of methods, you can ask if
object supports individual methods.

-- 
Alex Shindich
mailto:alex at shindich.com
Visit http://www.shindich.com/


-- 
Posted from adsl-63-194-17-248.dsl.lsan03.pacbell.net [63.194.17.248] 
via Mailgate.ORG Server - http://www.Mailgate.ORG


From austinw at bhpvillage.com  Mon Mar  5 03:07:10 2001
From: austinw at bhpvillage.com (Austin Wilson)
Date: Mon, 5 Mar 2001 19:07:10 +1100
Subject: ftp a list to a remote server
Message-ID: <tCHo6.8449$v5.32212@newsfeeds.bigpond.com>

Hi

I would like to save a list as a text file on a remote server using ftp.

Is this possible? If so, could you please provide some sample code.

Or do I have to save the file locally first then ftp that file to the remote
server?

Thanks
Austin




From gerhard.nospam at bigfoot.de  Wed Mar 14 08:57:09 2001
From: gerhard.nospam at bigfoot.de (Gerhard =?iso-8859-1?Q?H=E4ring?=)
Date: Wed, 14 Mar 2001 14:57:09 +0100
Subject: Python interacting with Java Script webpage?
References: <KFur6.8651$Vg3.600799@bgtnsc06-news.ops.worldnet.att.net>
Message-ID: <slrn9auv1g.1b9.gerhard.nospam@lilith.hqd-internal>

On Tue, 13 Mar 2001 19:37:14 GMT, William Dandreta wrote:
>I would like to write a Python program that will be able to connect to a
>webpage that runs a Java Script, post data to the form displayed, and save
>the data retrieved to a file.

Hmm. As I see it, the JavaScript is irrelevant here. What you probably want is
to have a CGI script output a HTML page with a form included. And the form
fields should be already filled with default values or values that were stored
earlier, like user preferences. Am I right?

I have attached a quick and dirty CGI script that should show the basics of how
this can be done.

>I have been looking for a book, tutorial, or any docs that will take me from
>not having a clue as to how to do it to being able to write the program.

For an example, you can look at the source code of the CGI module, there is a
test function in there that uses most of the modules functionality and is also
useful for debugging.

Gerhard

#####################################################################

#!/usr/bin/env python

# FillForm demo script

import string, cgi

def printRetrievedData():
    # could write data to file instead
    print cgi.FieldStorage(), '<br>' * 3

def printForm( fields, submiturl ):
    print '<form action="%s">' % ( submiturl )
    for ( key, value ) in fields.items():
            print '%s: <input type="text" name="%s" value="%s"><br>'\
              % ( key, key, value )
    print '<input type="submit"><br>'
    print '</form>'

# Mapping of name of form fields to values
form_fields = { 'Name' : 'Gerhard', 'favlanguage' : 'Python' }

# URL of the CGI-Script, probably to be adjusted
cgi_url = '/cgi-bin/ff.py'

print 'Content-type: text/html\n'

print '<html><body><p>'
printRetrievedData()
printForm( form_fields, cgi_url )
print '</p></body></html>'

-- 
mail:   gerhard <at> bigfoot <dot> de
web:    http://highqualdev.com


From jmarshal at mathworks.com  Thu Mar 29 13:42:41 2001
From: jmarshal at mathworks.com (Joshua Marshall)
Date: 29 Mar 2001 18:42:41 GMT
Subject: how query function args
References: <3AC374BC.F90A2894@abaqus-sn.com>
Message-ID: <99vvn1$bmh$1@news.mathworks.com>

david winkler <winkler at abaqus-sn.com> wrote:
> I need to query the names and values of a function's keyword arguments.
> I
> suspect that there is an easy way to do this but cannot find it.  Would
> someone care to make a suggestion?

func.func_defaults will return a tuple of the default values, but not
the names of the formals.


From just at letterror.com  Thu Mar 15 03:40:46 2001
From: just at letterror.com (Just van Rossum)
Date: Thu, 15 Mar 2001 09:40:46 +0100
Subject: Cameron Laird Quote on www.unixinsider.com -- Probably old news to 
 y'all, but new to me.
References: <bJOr6.9481$TW.38047@tor-nn1.netcom.ca> <3AB00593.D1C4BBBF@fxtech.com>
Message-ID: <3AB08001.9532A3F0@letterror.com>

"Paul T. Miller" wrote:
> 
> > I apologize for trotting it out again, if I am repeating something earlier
> > discussed, but I couldn't find any direct reference to it (my nntp server
> > sucks, only goes back 7 days).
> >
> > All I'll add is: Well said, Cameron!
> 
> I love Python. I especially love it as an embedded scripting language.
> But it is not going to replace graphics applications where speed is
> paramount. Many of these articles fail to take that into account. But
> then again, many problem domains do not have the interaction constraints
> that 3D or compositing applications have.

Yeah, I was going to say: there's an awfully large application area _between_
scripting and interactive 3d graphics apps! Python already does quite well
with interactive 2d graphics, though: see eg. Sketch.

Just


From pinard at iro.umontreal.ca  Thu Mar 29 21:11:23 2001
From: pinard at iro.umontreal.ca (=?iso-8859-1?q?Fran=E7ois?= Pinard)
Date: 29 Mar 2001 21:11:23 -0500
Subject: I come to praise .join, not to bury it...
In-Reply-To: <98tacr0q5s@news1.newsguy.com>
References: <mailman.983485923.1136.python-list@python.org>
	<97p7iq$kk2$1@nntp6.u.washington.edu> <97rka002k9@news1.newsguy.com>
	<3AAF7E47.294617E1@sage.att.com> <98qeco02hn7@news1.newsguy.com>
	<98rmod$c8p$1@animus.fel.iae.nl> <98tacr0q5s@news1.newsguy.com>
Message-ID: <oqwv98m2es.fsf@lin2.sram.qc.ca>

[Alex Martelli]

> Somebody who's unable to do anything but talk of aesthetics when
> discussing a choice in the architecture of a software component is simply
> not contributing to the discussion.  It looks good to you, it looks
> bad to him, it looks indifferent to her, yawn, great debate, now let's
> get beyond this to the REAL meat of the issue, please -- the technical
> implications of this, that, or the other choice in a specific matter.

I just want to emphasize, as others replied already, that aesthetics (or
better said in this precise context, legibility and understandability) *is*
real meat.  Consider that Python is art turned into technique :-).  By being
technical for the only sake of it, we are escaping something important.

> If you _can't_ get down to concrete specifics, then either you're not
> investing enough time & energy to contribute (in which case, staying
> out of a debate WOULD be a nice matter of courtesy)

Sorry if I'm looking impolite to you, then (:-).  I still think that minor
contributions, and even opinions, should not be merely or wholly dismissed
because their author is not available enough to write and sustain a PEP!

> IF the huge investment of time and energy doesn't faze you [...] striving
> to help others articulate their objections, in great detail, by first
> laying out in just as much details my reason for NOT objecting at all
> to the architecture Guido has chosen for .join; it won't of course have
> any direct repercussion on how Python behaves in the matter, but it MAY
> have positive implications on technical growth of the people involved,
> and thus, very indirectly, on future efforts on their part).

I'm not trying to systematically argue, here, but yet, I would like to
counter-weight a tiny bit the attitude which is depicted above.  Investing
a lot of energy into a project undoubtedly helps someone to develop more
complex and structured opinions, and is often a good step towards acquiring
more competence.  However, the amount of work invested does not necessarily
guarantee competence, and someone may mislead himself into dubious tracks
for quite a long while.  I presume we all know people that did such mistakes.

Huge investment of work is never a complete guarantee of being right, and
simple remarks from casual users are not necessarily wrong, even if they
do not show commitment to spoil all their sweat and blood arguing with you.

> it _still_ boils down to aesthetics being AT BEST a mental and verbal
> shorthand to avoid the considerable investment required to articulate
> the concrete underlying realities in detail.

The scheme is usual, to me at least.  It takes a tremendous amount of
energy for me to argue with people which are already set on an idea, and
even after having exhausted myself, I have no guarantee at all that any
progress will result of all that expense.  After many years, I learned
how to spare myself a bit better.  Yet, I still do not accept, for one,
to hear that my opinion is not worth being listened, just because I'm not
ready to fight hard for it.

> AT WORST (and, I suspect, more often, particularly in Usenet debates)
> a way for somebody to keep spewing words on a subject on which they
> actually have nothing concrete, constructive, and useful to say.

Note that I understand that you, too, need some protection against abuse.
We all lived it.  Best, however, is to not become over-protective! :-)

In the case under discussion, here, I suspect that the community has shown
enough reluctance to consider that there was at least something to ponder.
Details may well have been articulated, but if they are not even listened
at, there is no point in repeating them endlessly: this is a lost cause.

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



From jepler at inetnebr.com  Mon Mar 26 20:19:01 2001
From: jepler at inetnebr.com (Jeff Epler)
Date: Tue, 27 Mar 2001 01:19:01 GMT
Subject: Newbie needs help to find bug in his first Perl program
References: <3ABCC757.6C6F17CA@earthlink.net> <3ABE73B2.CAFCA177@student.liu.se> <3ABD67C6.7B714D7A@earthlink.net>
Message-ID: <slrn9bvqk4.42i.jepler@potty.housenet>

On Mon, 26 Mar 2001 04:24:14 GMT, Ron Stephens
 <rdsteph at earthlink.net> wrote:
>This is a multi-part message in MIME format.
>--------------FD2A49219370BD01A077FCD5
>Content-Type: text/plain; charset=us-ascii
>Content-Transfer-Encoding: 7bit
>
>Thanks, Richard, you solved my problem, despite my fuzzy question ;-))) and
>to the rest of you, yes, I will try to be much more careful and precise from
>now on when asking for help! Point acknowledged.

Ron,
To help introduce you to the power of Python, I have written a program with
a similar intent to yours.  It lets a user enter any number of options
and criteria, but it uses some of Python's versatile data types (mostly
the "dictionary", but a few lists), functions, and loops to reduce the
amount of code which must be repeated by the programmer (imagine when your
client asks you to make your program support 99 options and 100 criteria!
Without loops, that's well over 10,000 lines the programmer has to write).

I've tried to add comments that will explain what is going on.  I think
that almost all of the ideas in this program are present in the Tutorial
document.  The "sort" at the end is probably the hardest to understand.

Good luck with Python!

------------------------------------------------------------------------
# This code is placed in the public domain

def get_list(heading, prompt):
	""" get_list(heading, prompt) -> list

This function prompts for a list of things.  The heading is printed on
a line by itself, and the prompt must have a %d substitution for the
number of the item within the list.
"""
	print heading
	print
	print "(enter a blank line to end the list)"
	ret = []
	i = 1
	while 1:
		line = raw_input(prompt % i)
		if not line:
			break
		ret.append(line)
		i=i+1
	print
	return ret

def get_number(prompt):
	""" get_number(prompt) -> float

This function prompts for a number.  If the user enters bad input, such as
"cat" or "3l", it will prompt again.
"""
	res = None
	while res is None:
		try:
			res = float(raw_input(prompt))
		except ValueError: pass
	return res

# First, ask the user to enter the lists
options = get_list("Enter your options", "Option %d: ")
criteria = get_list("Enter your criteria ...", "Criterion %d: ")

# Next, get the user to rank his criteria.  I use a system where higher
# is better, so that an undesirable characteristic can be given a negative
# weight.
#
# {} is a dictionary, it can be indexed by (nearly) any expression,
# and we will index it with the names of the criteria.
# (For a more traditional program, we could use a list and index by the
# number of the criterion)

rankings = {}
print
print "Enter relative importance of criteria (higher is more important)"
print
for c in criteria:
		rankings[c] = get_number("Criterion %s: " % c)

# Next, get the user to score each option on all the criteria.
# Here, we index the dictionary on the pair (option, criterion).
# This is similar to a two-dimensional array in other languages

score = {}
print
print "Enter score for each option on each criterion"
print
for o in options:
	print
	print "Scores for option %s" % o
	print
	for c in criteria:
		score[o, c] = get_number("Criterion %s: " % c)

# Calculate the resulting score for each option.  This equation
# is different from Rod Stephen's original program, because I
# make more important criteria have higher rankings, and even let
# bad criteria have negative rankings.

# The "result" dictionary is indexed with the names of the options.
result = {}
for o in options:
	value = 0
	for c in criteria:
		print o, c, rankings[c], score[o, c]
		value = value + rankings[c] * score[o, c]
	result[o] = value

# Now, I want to take the dictionary result, and turn it into a ranked list.
results = result.items()	# A list of tuples (key, value)
results.sort(lambda x, y: -cmp(x[1], y[1]))
				# Sort the list using the reverse of the
				# "value" of the entry, so that higher
				# values come first

print
print "Results, in order from highest to lowest score"
print
print "%5s %s" % ("Score", "Option")

# Take the pairs out of results in order, and print them out
for option, result in results:
	print "%5s %s" % (result, option)
------------------------------------------------------------------------


From aleaxit at yahoo.com  Sat Mar 10 17:47:41 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sat, 10 Mar 2001 23:47:41 +0100
Subject: Can anyone recomend a good intoduction to C...
References: <mailman.983894288.19965.python-list@python.org> <ewwp6.1466$y6.241005@ruti.visi.com> <slrn9aeblb.cg1.scarblac@pino.selwerd.nl> <djagat4kk3cs9hg0c3k39cudk6goblhkg6@4ax.com> <989mbk$24nuv$1@fido.engr.sgi.com> <slrn9ah66m.k09.qrczak@qrnik.zagroda> <98auhn0270j@news2.newsguy.com> <slrn9aiaq7.rmu.qrczak@qrnik.zagroda> <98dpoe12ckn@news1.newsguy.com> <3AAA85C1.81FB90AA@shore.net>
Message-ID: <98eb1f02uqq@news1.newsguy.com>

"Kurt B. Kaiser" <kbk at shore.net> wrote in message
news:3AAA85C1.81FB90AA at shore.net...
> Alex Martelli wrote:
>
> > Your opinion is widespread, and self-perpetuating, but it has no
> > current technical basis for such tasks as Python extensions on Linux.
> >
> My understanding is that the C++ compiler for Linux is still
> considered to be unstable and that that is a significant technical
> factor for sticking with C?

Interesting; we developed and deployed C++ applications for years
on gcc on various Unix boxes around the mid-90's, and we had no
problems with "unstable compilers" (then, the market forced our
hand so we switched to Win/NT and VC++ -- now THAT was another
ball of wax at the time, though it HAS gotten better since).

It's no doubt possible that gcc has significantly worsened in the
last 5 or 6 years, but I'll confess that would really surprise me.


Alex





From cerutti at together.net  Thu Mar  1 10:12:58 2001
From: cerutti at together.net (Neil Cerutti)
Date: 1 Mar 2001 15:12:58 GMT
Subject: Allow anything in identifiers?
References: <3A9DC13A.1647EE51@home.se>
Message-ID: <slrn99sq84.a2.cerutti@fiad06.norwich.edu>

Sverker Nilsson posted:
>This would be a new kind of identifiers, I suppose one
>can make them by writing for example id <string> like in:
>
>id 'a>b' = a > b
>id 'x.b' = x.b
>alu.id'+' = lambda x, y...
>
>I think it could make programs easier to read by allowing
>more evident identifiers in many cases. The extended identifier
>syntax should not break anything, and it seems it might be easy to
>implement...

I have to admit I love being able to use function names like
number? and list->word in scheme. is_number and list_to_word are
just harder to type and to read.

-- 
Neil Cerutti <cerutti at together.net>
  "The Analytical Engine weaves algebraic patterns just as the Jacquard
  loom weaves flowers and leaves." -- Lady Lovelace


From stephen_purcell at yahoo.com  Thu Mar 29 09:23:28 2001
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Thu, 29 Mar 2001 16:23:28 +0200
Subject: Whats in your $PYTHONSTARTUP$ ?
In-Reply-To: <20010329083827.B4611551@vislab.epa.gov>; from aa8vb@yahoo.com on Thu, Mar 29, 2001 at 08:38:27AM -0500
References: <FJ2w6.5358$GD.2921338@typhoon.snet.net> <20010329083827.B4611551@vislab.epa.gov>
Message-ID: <20010329162328.A631@freedom.puma-ag.com>

Randall Hopper wrote:
> Just a little fragment to colorize my python prompts.  Nowadays, I'm always
> on a VT100 color capable terminal (recent xterm, aterm, FreeBSD/Linux
> console, etc.).
> 
>      import sys,os
>      if os.environ.get( 'TERM' ) in [ 'xterm', 'vt100' ]:
>        sys.ps1 = '\001\033[0:1;32m\002>>> \001\033[0m\002'
>        sys.ps2 = '\001\033[0:31m\002... \001\033[0m\002'
> 

That's nice. And you can set the xterm title string using:

    if os.environ['TERM'] in ('rxvt', 'xterm'):
      sys.stdout.write('\033]0;python in %s\007' % os.getcwd())

You could add in all kinds of nice stuff like the Python version number and
the hostname (sys.version, socket.gethostname)...

-Steve

-- 
Steve Purcell, Pythangelist
Get testing at http://pyunit.sourceforge.net/
Any opinions expressed herein are my own and not necessarily those of Yahoo



From chris.gonnerman at usa.net  Fri Mar  9 09:03:51 2001
From: chris.gonnerman at usa.net (Chris Gonnerman)
Date: Fri, 9 Mar 2001 08:03:51 -0600
Subject: learning C
References: <mailman.983972829.23481.python-list@python.org> <983975751.18833@newsmaster-04.atnet.at> <3AA82422.53026D3D@cosc.canterbury.ac.nz> <20010308221519.A4367@harmony.cs.rit.edu> <mailman.984113052.19927.python-list@python.org> <984135226.858948@newsmaster-04.atnet.at>
Message-ID: <006301c0a8a1$c856b2e0$a100000a@local>

----- Original Message -----
From: "Werner Schiendl" <ws-news at gmx.at>
Subject: Re: learning C
> The compiler vendors already did it on a lot of platforms.
> C++ will swallow C, because C is nearly a subset of C++.

The COMPILER swallows C, C++, and (in the case of the newer GCC) a bunch of
other
languages.  Fortran is in there, but you won't catch me using it.

> If you have a full cupboard of tools installed, what sense does it make to
> ignore 90% of it??

I forget who first quoted the famous bear, but I too am of "little brain".
This
doesn't mean I'm not a competent programmer, but only that the cost of
learning
the "new tools" of C++ is just too high for the benefits I perceive.  I also
agree with the assertion that C is more predictable than C++; yeah, I know I
could
learn all the rules for performance etc. but that is exactly what I am
complaining
about.

For OOP I choose Python.  For raw power and control I choose C.  If I need
both,
I extend Python with C.  My Python extensions tend to be quite small; in
fact,
the Windows Conio (WConio) package that I wrote got smaller with each
version,
as I removed code from the C extension and put it in the Python wrapper
layer.

> Why not have a look at all of it and use what's suited for the application
> best?

Like I said, been there, done that, couldn't afford the T-shirt.  Rewiring
my
brain for C++ is just too hard.  I'm glad that it works for you, and many
others,
but there are also many like me that can't see why you are pushing so hard.
I
guess you just can't see our point of view, though I think I see yours just
fine.






From kens at sightreader.com  Wed Mar 28 13:07:35 2001
From: kens at sightreader.com (Ken Seehof)
Date: Wed, 28 Mar 2001 10:07:35 -0800
Subject: What *instance* called me & proxy methods
References: <3AC1BE92.ACE82741@ua.ac.be> <5.0.2.1.0.20010328104602.021cfc40@mail.inet.com.br>
Message-ID: <003701c0b7b2$039b1250$04090a0a@upcast.com>

You want 'im_self'.

Also, you might find 'im_func' handy as well.  It returns the
function, so you can unbind a bound method.

>>> class X:
...  def a(self):
...   return 1
...
>>> x = X()
>>> x.a
<method X.a of X instance at 01390DFC>
>>> x.a.im_self
<__main__.X instance at 01390DFC>
>>> x.a.im_func
<function a at 0139105C>

- Ken Seehof

----- Original Message -----
From: "Carlos Alberto Reis Ribeiro" <cribeiro at mail.inet.com.br>
To: <python-list at python.org>
Sent: Wednesday, March 28, 2001 8:57 AM
Subject: What *instance* called me & proxy methods


> I've seen some talk on how to get the name of the caller. I got a related
> problem to solve, only slightly more complicated.
>
> Can I know what *instance* called a method? I'm not talking about "self"
> here, because we *can* call a method using other techniques, such as
> copying the instance method pointer directly:
>
>     >>> class myclass:
>     ...   def hello(self):
>     ...     print "hi! I'm ", self
>     >>> class myproxy:
>     ...   pass
>     >>> a = myclass()
>     >>> a.hello()
>     hi! I'm  <__main__.myclass instance at 015093EC>
>     >>> m = a.hello
>     >>> p = myproxy()
>     >>> p.m = m
>     >>> p.m()
>     hi! I'm  <__main__.myclass instance at 015093EC>
>
> In both cases, hello() received the same value as the self parameter. This
> is exactly what I need; in fact, the im_self attribute of the instance
> method stores the pointer to self, which explains this behavior. However,
I
> would also like to have some information to differentiate between these
two
> calls. My goal is to implement some security checking as part of my
> proxy/adapter module [1]. I have already tried some simple minded
> approaches (too many to list here), so far with no success.
>
> The best solution the I have thought so far can be implemented as a Python
> extension - if I manage to put my Linux box up and running (I'm gcc-less
> right now :-).
>
> 1) Implement a new object called 'ProxyMethod', derived of 'instance
>     method', with the following differences:
>
>     * It does not make the im_* attributes visible to the outside
>       world;
>
>     * It stores a read-only attribute to the proxy that 'owns' the
>       ProxyMethod.
>
> 2) The 'ProxyMethod' constructor would take a instance method and
>     build the proxy method from it.
>
> 3) The information about the proxy would need to be stored somewhere
>     in the call stack. I dont know if it is possible without changing
>     some of the Python internals.
>
> 4) The runtime semantics of the 'ProxyMethod' would be about the
>     same as the ones of the instance method - just a direct call,
>     without any extra intermediation.
>
> I would like to know what the community thinks about turning this into a
> PEP. Maybe we can find a simpler solution, or it's better to keep this as
a
> module outside the distribution.
>
> For now I have a very simple Python implementation. It's far from perfect,
> and it has some undesirable side effects because it is not completely
> transparent, and it needs some intermediate processing. The only way that
I
> found to signal to the proxied object about the existence of the proxy was
> to create a temporary "proxy" attribute (see point (3) above). However, it
> serves it purpose as a proof-of-concept:
>
> ----------------------------------------------------------------------
> TEST RESULT
>
>  >>> n = NuclearArsenal()
>  >>> n.NuclearAttack("NY")
> Only the proxy can order the attack
>  >>> p = Proxy()
>  >>> p.attack = ProxyMethod(n.NuclearAttack, p)
>  >>> p.attack("NY")
> Boooom! NY destroyed
>
> ----------------------------------------------------------------------
> CODE
>
> """ ProxyMethod proof-of-concept
>      (c) 2001 Carlos Ribeiro - cribeiro at mail.inet.com.br
>      Released in public domain for use "as-is"
> """
>
> from types import *
>
> class Proxy:
>    pass
>
> class ProxyMethod:
>
>    def __init__(self, im, p):
>      if type(im) != MethodType:
>        raise "A proxy needs a method"
>      if p.__class__ != Proxy:
>        raise "A proxy method is bounded to a proxy"
>      self.im = im
>      self.p  = p
>
>    def __call__(self, *a):
>      # set a flag inside the target instance
>      setattr(self.im.im_self, 'proxy', self.p)
>      # do the call
>      self.im(*a)
>      # clear the flag
>      delattr(self.im.im_self, 'proxy')
>
> class NuclearArsenal:
>    def NuclearAttack(self, city):
>      if getattr(self, 'proxy', None):
>        # called through the proxy
>        print "Boooom! %s destroyed" % (city)
>      else:
>        print "Only the proxy can order the attack"
>
> ----------------------------------------------------------------------
>
> Carlos Ribeiro
>
>
>
> --
> http://mail.python.org/mailman/listinfo/python-list





From stowasser.h at idowa.de  Wed Mar 14 08:44:42 2001
From: stowasser.h at idowa.de (Stowasser Harald)
Date: Wed, 14 Mar 2001 14:44:42 +0100
Subject: Better string.translate?
Message-ID: <3aaf758f$1_2@news.zet.net>

Hello,
i am very new in Python. I don't know how i can do this better:

_deltable=string.maketrans('.,:;-_?!.=()?[]{}\'\"\r\n\f\t\v','
')
ret = string.translate(s,_deltable)

This works fine, but i need a 'anti-maketrans'.
I want to define what characters NOT to change in a space-character.

How can i do this in Python?




From db3l at fitlinxx.com  Thu Mar  1 13:57:28 2001
From: db3l at fitlinxx.com (David Bolen)
Date: 01 Mar 2001 13:57:28 -0500
Subject: Newbie createing .PYC for import
References: <3a9e83fa$1@news1.dbtech.net>
Message-ID: <u7l29pbaf.fsf@ctwd0143.fitlinxx.com>

"Larry Bates" <lbates at syscon-computers.com> writes:

> I have spent the last couple of hours (unsuccessfully) attempting to figure
> out how to generate a .PYC file for a module that I've defined.  Everything
> appears to imply that it's "automatic", but I'm not getting anywhere.  I
> have a module and I wish to generate a .PYC file so that I can import it
> into future projects.  Can anyone shed any light on how to accomplish this?

It's "automatic" if you are importing the module and Python has the
ability (permissions, free space, etc...) to write the compiled module
back to the directory.

Note that running Python on a top level script is not considered an
import and thus no pyc will be created by default.

You can manually compile any module using the py_compile module, and,
for example, the compile() function in that module, so interactively:

    >>> import py_compile
    >>> py_compile.compile('filename')

and this will write the pyc to the same location as filename (or you
can override that with the optional parameter cfile).

You can also automatically compile all files in a directory or
directories using the compileall module, which can also be run
straight from the command line.

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


From robin at jessikat.fsnet.co.uk  Thu Mar  1 13:41:37 2001
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Thu, 1 Mar 2001 18:41:37 +0000
Subject: PEP 227 (was Re: Nested scopes resolution -- you can breathe again!)
References: <LNBBLJKPBEHFEDALKOLCMEEPJBAA.tim.one@home.com>
 <200102230259.VAA19238@cj20424-a.reston1.va.home.com>
 <mailman.983384008.21434.python-list@python.org>
 <97lqsr0311d@news2.newsguy.com>
 <mailman.983466367.2968.python-list@python.org>
Message-ID: <KNSz0UAhfpn6EwZj@jessikat.fsnet.co.uk>

In article <mailman.983466367.2968.python-list at python.org>, Jeremy
Hylton <jeremy at alum.mit.edu> writes
>Or: global NAME in FUNCTION
>
>def eggs():
>    z = 2
>    def spam():
>        def inner():
>            global x in spam, z in eggs
>            x += 1
>            z += x
>        x = 23
>        inner()
>
>Jeremy
>
I like that a lot better than having it default possibly wrongly, but
how would this be doable/spelled in the lambda case?
-- 
Robin Becker


From kens at sightreader.com  Fri Mar 16 18:08:38 2001
From: kens at sightreader.com (Ken Seehof)
Date: Fri, 16 Mar 2001 17:08:38 -0600
Subject: Pep 238 Why new // operator (Non-integer Division)
References: <LNBBLJKPBEHFEDALKOLCMEDGJGAA.tim.one@home.com>
Message-ID: <003501c0ae6e$27e4de40$04090a0a@upcast.com>

Also, integer division is fairly common and a direct implementation is
significantly faster than floor(a/b).

Hey, I know: maybe "a//b" should be "a log b".  That way // is
the inverse operator of **.  :-)  Dang, if only I had time to
write PEPs. :-)

----- Original Message ----- 
From: "Tim Peters" <tim.one at home.com>
To: <python-list at python.org>
Sent: Friday, March 16, 2001 6:14 PM
Subject: RE: Pep 238 Why new // operator (Non-integer Division)


> [Gordon Williams]
> > Why do we need a new // operator to do the divide operation.  If
> > we want to do an integer divide with two reals why wont people just
> > use int(a/b).  For
> 
> 1. int() truncates.  Current i/j returns the floor.
> 
> 2. There is no bound on the sizes of i and j; it's very easy to
>    get Python long ints that don't fit in a Python float.
> 
> 
> -- 
> http://mail.python.org/mailman/listinfo/python-list
> 





From scarblac at pino.selwerd.nl  Thu Mar 29 15:15:41 2001
From: scarblac at pino.selwerd.nl (Remco Gerlich)
Date: 29 Mar 2001 20:15:41 GMT
Subject: Quoted identifiers in Python?
References: <3ac3387f$0$12248$cc9e4d1f@news.dial.pipex.com>
Message-ID: <slrn9c75sj.98.scarblac@pino.selwerd.nl>

Eddie and Babs <kca17 at dial.pipex.com> wrote in comp.lang.python:
> While reading an article about the Dylan language a few years ago, I was 
> surprised to find that you could have quoted identifiers (ie, variable and
> function names, etc). So you could have a function named "turn around" (with
> quotes) and then call it using something like:
> 
>     "turn around"(xxx)
> 
> The idea is that you can put any characters within the quotes.

So when I do

x = "turn around"

How does Python know whether I mean the string or the identifier?

-- 
Remco Gerlich


From lac at cd.chalmers.se  Wed Mar 28 08:54:09 2001
From: lac at cd.chalmers.se (Laura Creighton)
Date: Wed, 28 Mar 2001 15:54:09 +0200 (MET DST)
Subject: __init__.py files in PMW
Message-ID: <200103281354.PAA13594@boris.cd.chalmers.se>

PMW comes in several directories, and each one has as file named
__init__.py which contains this line:
# File to allow this directory to be treated as a python 1.5 package.

I don't understand.  Who needs this file and for what?

Thanks very much,
Laura Creighton
lac at cd.chalmers.se



From pmj769 at concentric.net  Fri Mar 30 22:28:02 2001
From: pmj769 at concentric.net (pmj769 at concentric.net)
Date: 31 Mar 2001 03:28:02 GMT
Subject: mxODBC chokes on dates! Any Help?
References: <MPG.152e4c579fa0baef9896a5@news.pacbell.net> <ulmpnnmte.fsf@ctwd0143.fitlinxx.com> <MPG.152e9338351a7e1b9896a7@news.pacbell.net>
Message-ID: <3ac54daa.33659169@news.xo.com>

The older version 1.1.1 version of mxODBC works fine on dates.

I just ran into the same problem that you did and sent a report to
Marc. He has been excellent at getting back to me in the past.

For now you can revert back to the older (and free) version.
However, you may run into an issue with the older version in the way
it (doesn't) handles UNICODE data coming from MS-SQL 7.0

Marshall


On Fri, 30 Mar 2001 12:58:02 -0800, Owen Ap Owen <no at thank.you> wrote:

>Thanks for the info. It is SQL Server that I am connecting to. But I got 
>the same error with date fields when I set up a test connection to 
>Access.
>
>It's nice that a fix is in the works. But this seems like something of a 
>show stopper (at least for the win32 side of things). 
>
>Considering that the creator of mxODBC is trying to charge a laughably 
>exorbitant license fee for commercial use he really needs to fix this 
>kind of stuff very quickly. He also needs to add a lot more functionality 
>and even then his licensing concept is pretty extreme.
>
>
>In article <ulmpnnmte.fsf at ctwd0143.fitlinxx.com>, db3l at fitlinxx.com 
>says...
>
>> I ran into the same problem with some date fields in an MS SQL Server
>> database, and Marc-Andre was able to revert back to an earlier way of
>> handling date fields (he preferred a different C data type if defined
>> in 2.0.0 than in earlier versions) to resolve it.  I was traveling and
>> was a little tardy getting back to verify that the fix worked, but I
>> expect it will show up in the next release of mxODBC.



From trentm at ActiveState.com  Sat Mar 31 02:32:19 2001
From: trentm at ActiveState.com (Trent Mick)
Date: Fri, 30 Mar 2001 23:32:19 -0800
Subject: very basic, how do you find square roots?
In-Reply-To: <9a408a$1mpp$1@raewyn.paradise.net.nz>; from gordonj@paradise.net.nz on Sat, Mar 31, 2001 at 07:05:43PM +1200
References: <9a408a$1mpp$1@raewyn.paradise.net.nz>
Message-ID: <20010330233219.A27969@ActiveState.com>

On Sat, Mar 31, 2001 at 07:05:43PM +1200, johnston112 wrote:
> Yes, that is my question :)
> I am trying to make a program that dose the quadratic equation.
> 
> johnston112


import math
print math.sqrt(4)


Trent

-- 
Trent Mick
TrentM at ActiveState.com



From fredrik at pythonware.com  Thu Mar 29 05:41:51 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Thu, 29 Mar 2001 10:41:51 GMT
Subject: Getting all the *files* from a directory -- A better way??
References: <mailman.985816117.6859.python-list@python.org> <99uu6401u9c@news2.newsguy.com>
Message-ID: <PjEw6.4142$4N4.841410@newsc.telia.net>

Alex Martelli wrote:
> So wrap it up...:
>
> def joined_listdir(dir):
>     return [os.path.join(dir,f) for f in os.listdir(dir)]

or: glob.glob(os.path.join(dir, "*"))

> then use your wrapper:
>
> files1 = [f for f in joined_listdir(dir) if os.path.isfile(f)]

or: files1 = filter(os.path.isfile, joined_listdir(dir))

(shorter, faster -- and works for all versions of Python)

Cheers /F




From SCook at pobox.com  Tue Mar 13 06:40:32 2001
From: SCook at pobox.com (Steve)
Date: Tue, 13 Mar 2001 11:40:32 GMT
Subject: Compiling python under Windows
Message-ID: <QGnr6.17986$0N3.109906@news-server.bigpond.net.au>

Hi,

Is it possible to build python on win32 with either mingw (the win32 port of
gcc) or the borland free compiler?  As I don't want to have to buy MS VC++
just to try a changes that I plan to the core.




From costas at meezon.com  Fri Mar  2 08:03:15 2001
From: costas at meezon.com (Costas Menico)
Date: Fri, 02 Mar 2001 13:03:15 GMT
Subject: compile() source code from a string
References: <3a9ee600.391990@News.CIS.DFN.DE> <mailman.983506923.29610.python-list@python.org>
Message-ID: <3a9f98b0.2420376@News.CIS.DFN.DE>

"Chris Gonnerman" <chris.gonnerman at usa.net> wrote:

>This is a multi-part message in MIME format.
>
>------=_NextPart_000_0085_01C0A29E.B3939520
>Content-Type: text/plain;
>	charset="Windows-1252"
>Content-Transfer-Encoding: 7bit
>
>There IS a syntax error in your code.  To wit:
>
Sorry, I should test my examples before posting (duh). The problem I
have been having is with this code below.This code will not work
unless the """ are on their own line.  It seems Python needs an extra
CR at the end of the code. I guess it kind of makes sense. 

Thanks for your help, 
costas

#!/usr/bin/env python
# This code does not work unless you put
# the """ on their own line
codestr = """
if x == 1:
  print "1"
  print "2"
  print "3"
elif x == 2:
  print 'A'
  print 'B'
  print 'C'"""
codeobj = compile(codestr, "codestr", "exec")
x = 2
exec(codeobj)


#!/usr/bin/env python
# This code works
codestr = """
if x == 1:
  print "1"
  print "2"
  print "3"
elif x == 2:
  print 'A'
  print 'B'
  print 'C'
"""
codeobj = compile(codestr, "codestr", "exec")
x = 2
exec(codeobj)



From info at black-karma.de  Fri Mar 23 08:50:44 2001
From: info at black-karma.de (Harvest T. Moon)
Date: Fri, 23 Mar 2001 14:50:44 +0100
Subject: Slightly Off-Topic: Python featured in Severance: Blade of Darkness
References: <NMHu6.144$NR.1628@news3.oke.nextra.no>
Message-ID: <99fk8p$13h82$1@ID-22517.news.dfncis.de>

| Writing mods/extensions in Python would be cool in any game. :->
| Wouldn't Python fit just perfectly for this purpose?

i remember someone said a Quake1 MOD was written in Python, some BOTs too
... dunno if anyone knows something more specific about it.

Harvest T. Moon
-- a life less ordinary!!




From cfelling at iae.nl  Thu Mar 29 18:06:00 2001
From: cfelling at iae.nl (Carel Fellinger)
Date: 30 Mar 2001 01:06:00 +0200
Subject: Accessing variables in __main__ from modules....
References: <mailman.985891179.24578.python-list@python.org> <9a0ett$eqb$1@animus.fel.iae.nl>
Message-ID: <9a0f4o$eu6$1@animus.fel.iae.nl>

Carel Fellinger <cfelling at iae.nl> wrote:

> The most straightforwarth way to achieve this would be to import Debug
> from main, or better to have a seperate options module and import from
> there. like:
...
> other module:

Oh, and importing something in many modules is relative cheap.  Only
the first import compiles and executes that module, subsequent imports
only bind the module object to a name in the current scope.
-- 
groetjes, carel


From tim.one at home.com  Sun Mar 11 01:01:08 2001
From: tim.one at home.com (Tim Peters)
Date: Sun, 11 Mar 2001 01:01:08 -0500
Subject: Urgent Question about Python licensing
In-Reply-To: <98ejhr$38q$0@216.39.170.247>
Message-ID: <LNBBLJKPBEHFEDALKOLCCENBJEAA.tim.one@home.com>

[Dave LeBlanc]
> When UCITA first came out, there was a lot of press about it's evils.

There still is!  There was also a lot of discussion about it on
comp.lang.python, specifically in relation to the 1.6 CNRI license at the
time.  You need to read the license and the actual text of UCITA to get
anywhere with this -- or ask a lawyer.  Whether UCITA is good or bad, the 1.6
license grants you very strong rights, and UCITA can't magically negate
those.

> ...
> However, after reviewing information at http://www.ucitaonline.com/
> and at http://www.2bguide.com, I came to the conclusion that i'd been
> mislead. Imagine! being mislead by the press! <g>.  (I should have
> known better - i'm a firm believer in the "believe none of what you
> hear and only half of what you read" school of thought.)

Both of those are UCITA advocacy sites, and set up by the same lawyer (Carol
Kunze, who "attended the UCITA drafting committee meetings from 1996 -
1999").  So on *those* sites, believe the negative halves (if you can find
any <0.5 wink>).

> ...
> As it turns out, after the fact license changes are NOT binding,
> "shrink wrap" or "click wrap" licenses ARE binding and have been
> upheld by the courts, can't make a bogus venue choice, and that a
> licensor CAN unilaterally disable software but only after a 2 week
> warning in many cases.

Note that what UCITA says is irrelevant unless read *in conjunction* with
what the 1.6 license says.  For example, section 308(2)(B) of UCITA holds
that the license is *perpetual* if "the license expressly grants the right to
incorporate or use the licensed information or informational rights with
information or informational rights from other sources in a combined work for
public distribution or public performance.".  The CNRI 1.6 license does so
expressly grant both rights.  I am not a lawyer, but I can read -- so can
you.  UCITA simply doesn't kick in except where the 1.6 license leaves holes,
and CNRI didn't intend to leave any holes.  IMO (and IANAL), the GPL does
leave holes, and that's the fundamental source of the conflict (when CNRI
plugs a hole that the GPL doesn't, that's "a restriction" not in the GPL,
hence automatically leaves the CNRI license "incompatible" with the latter,
which forbids additional restrictions of *any* kind).

> ...
> Moving along, I find it ironic that the Open Software Foundation calls
> the GPL an open source license since it violates the rules given for
> such. GPL software has source available, but it's use is constrained
> and contaminating (unlike the GLL library license).

I see that they've added a footnote to point 9 at

    http://www.opensource.org/docs/definition.html

It reads:

    Yes, the GPL is conformant with this requirement.  GPLed
    libraries "contaminate" only software to which they will actively
    be linked at runtime, not software with which they are merely
    distributed.

I'm unsure whether they added that before or after your post, though <wink>.

> P.S Guido is welcome to my first born if I ever have one - but he has
> to do his own cooking.

guido-does-not-eat-children-neither-cooked-nor-raw-ly y'rs  - tim




From robin at jessikat.fsnet.co.uk  Fri Mar  2 10:51:41 2001
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Fri, 2 Mar 2001 15:51:41 +0000
Subject: gah! I hate the new string syntax
References: <mailman.983485923.1136.python-list@python.org>
 <0vpn79.331.ln@127.0.0.1>
Message-ID: <UY5kBKANG8n6EwqG@jessikat.demon.co.uk>

In article <0vpn79.331.ln at 127.0.0.1>, Carl Banks <idot at vt.edu> writes
>Sean 'Shaleh' Perry <shaleh at valinux.com> wrote:
>> return ";".join(["%s=%s" % (k, params[k]) for k in params.keys()])
>>
>> every day python seems to be moving closer to the line noise aspect of
>> coding.
>
>I find it ok for methodlike functions such as split.
>
>For unmethodlike functions such as join, I would use the old
>string.join syntax (for regular strings, anyways).
>
>"".join is just creepy.
>
>
it looks plain wrong, but would ['a','b','c'].join(';') look any better?

In fact there may be some benefit in being able to say

pathJoiner = (sys.platform=='win32' and ';' or ':').join
.....
path1 = pathJoiner([d1,d2,.....])
path2 = pathJoiner([e1,e2,.....])

-- 
Robin Becker


From DSHIN at nc.rr.com  Thu Mar 15 01:15:53 2001
From: DSHIN at nc.rr.com (Daehyok Shin)
Date: Thu, 15 Mar 2001 06:15:53 GMT
Subject: [Q] Bug in Python?
Message-ID: <t6Zr6.92712$__6.18256422@typhoon.southeast.rr.com>

Recently, I met a strange behavior of Python 6.1.
When I tried the following module,
I got an error message in "print var":
    UnboundLocalError: var

But, when I changed "var=None" to "pass",
the error message was gone and I got an expected result:--printing None and
None.
What happens here? Is this a bug in Python or I missed something?

############# test.py ###############
var = None

def a():
     print var #UnboundLocalError
     if var != None:
          var = None  # >> pass

def b():
     print var
     a()

b()
#################################

--
Daehyok Shin (Peter)
Terrestrial Hydrological Ecosystem Modelers
Geography Department
UNC-CH




From tim at vegeta.ath.cx  Wed Mar  7 20:21:24 2001
From: tim at vegeta.ath.cx (Tim Hammerquist)
Date: Thu, 08 Mar 2001 01:21:24 GMT
Subject: gah! I hate the new string syntax
References: <mailman.983615170.23313.python-list@python.org> <tN5o6.2746$eH3.14913@NewsReader> <Koco6.7518$1D5.244602@e420r-atl1.usenetserver.com> <mailman.983997252.5628.python-list@python.org> <meAp6.29961$ML1.1514835@e420r-atl2.usenetserver.com>
Message-ID: <slrn9adnr9.7b0.tim@vegeta.ath.cx>

Steve Holden <sholden at holdenweb.com> wrote:
> "D-Man" <dsh8290 at rit.edu> wrote in message
> news:mailman.983997252.5628.python-list at python.org...
> > On Sat, Mar 03, 2001 at 03:19:05PM -0500, Steve Holden wrote:
> [ ... ]
> >
> > What about one who is both English and Scots, yet never set foot on
> > that island ;-)?
> >
> No such animal!

If you refer to being "English" and "Scots" as a native only, then that may
be true.

Otherwise, you've just negated my existence!!  ;)

-- 
-Tim Hammerquist <timmy at cpan.org>
American society is a sort of flat, fresh-water pond which absorbs
silently, without reaction, anything which is thrown into it.
	-- Henry Brooks Adams


From tjg at exceptionalminds.com  Thu Mar 22 13:26:39 2001
From: tjg at exceptionalminds.com (Timothy Grant)
Date: Thu, 22 Mar 2001 10:26:39 -0800
Subject: Good Editor?
In-Reply-To: <99cb42013df@news2.newsguy.com>; from aleaxit@yahoo.com on Thu, Mar 22, 2001 at 08:54:42AM +0100
References: <3ab9a32b$1_1@news2.one.net> <99cb42013df@news2.newsguy.com>
Message-ID: <20010322102639.S24341@trufflehunter.avalongroup.net>

On Thu, Mar 22, 2001 at 08:54:42AM +0100, Alex Martelli wrote:
> "fjs" <fjs6 at usa.net> wrote in message news:3ab9a32b$1_1 at news2.one.net...
> > Can someone recomend a good editor to write python in?  Im just getting
> > started learning python and dont know which ones support python.
> 
> I mostly use VIM (www.vim.org), which among other things
> supports the use of Python as a macro-language.  However,
> VIM is a development of vi -- some of us love its modal
> basic operations (it minimizes fingermovement wrt any
> alternative -- highly ergonomic), some other detest it
> (I think it's because they try to grasp it with their
> brains rather than just let their _fingers_ learn it).

As a vim newbie, I can vouch for this brains/fingers thing.
Once my fingers got things mostly figured out, I find myself
using vim keystrokes everywhere by force of habit. I even ended
up changing from bash to zsh because zsh has such amazingly
good vim emulation.

I'm faster, more productive, and use *far* fewer system
resources with vim than I did with emacs.

I'd like to give a plug to Glimmer as an editor. It's still
quite young, but it's native tongue is Python. It is quite
light and has some nice interface things. It's what I used
between emacs and vim.

-- 
Stand Fast,
    tjg.

Timothy Grant                         tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.         <><       (503) 246-3630
>>>>>>>>>>>>>Linux, because rebooting is *NOT* normal<<<<<<<<<
>>>>This machine was last rebooted:  64 days 22:39 hours ago<<



From thomas at cintra.no  Thu Mar  8 06:18:23 2001
From: thomas at cintra.no (Thomas Weholt)
Date: Thu, 8 Mar 2001 12:18:23 +0100
Subject: Added namespaces; now nothing works!
Message-ID: <3UJp6.10068$t21.271693@news3.oke.nextra.no>

Hi,

I have a xml-document I've decided to add namespaces to. Tried to convert my
old stylesheet to reflect the changes and parse it to HTML with XT, but now
nothing but the stylesheet ends up in the resulting document. What am I
doing wrong?

The xml-document :

<?xml version="1.0" encoding="iso-8859-1"?>
<container xmlns="http://www.w3c.org/TR/REC-html40"
 xmlns:jcd="http://www.weholt.no/jeezez/jcd/1.0"
 xmlns:jeezez="http://www.weholt.no/jeezez/">
  <jeezez:header>
   <jcd:test>TestValue</jcd:test>
  </jeezez:header>
 <jeezez:items>
  <jcd:item>
          <jcd:ID>1</jcd:ID>
          <jcd:filename>alice.txt</jcd:filename>
  </jcd:item>
 </jeezez:items>
</container>

The stylesheet, just a simple thing to get something into the browser or
result document :
<?xml version='1.0'?>
<xsl:stylesheet xmlns:xsl="http://www.w3c.org/1999/XSL/Transform"
  xmlns="http://www.w3c.org/TR/REC-html40"
 xmlns:jcd="http://www.weholt.no/jeezez/jcd/1.0"
 xmlns:jeezez="http://www.weholt.no/jeezez/">

<xsl:output method="html"/>

<xsl:template match="/">
    <HTML>
      <BODY>
         <xsl:apply-templates/>
     </BODY>
   </HTML>
</xsl:template>

<xsl:template match="jeezez:header">
 <b><xsl:value-of select="jcd:test"/></b>
</xsl:template>
</xsl:stylesheet>

I only get the entire stylesheet as result. No formatting done whatsoever.
If somebody could provide a dead-simple stylesheet that just enters the
value of the jcd:test-tag in the result document I'd be happy.

Just started using namespaces so I just need a kick to get going. Thanks!!

Best regards,
Thomas Weholt




From support at internetdiscovery.com  Tue Mar 27 07:36:58 2001
From: support at internetdiscovery.com (Mike Clarkson)
Date: Tue, 27 Mar 2001 12:36:58 GMT
Subject: Python wrapper for TkHTML?
Message-ID: <3ac08765.3126948@24.0.228.33>

Does anyone have a Python wrapper for TkHTML, the c-coded HTML widget
for Tk <http://www.hwaci.com/sw/tkhtml>?

Many thanks in advance,

Mike


From tjg at exceptionalminds.com  Mon Mar 19 14:11:08 2001
From: tjg at exceptionalminds.com (Timothy Grant)
Date: Mon, 19 Mar 2001 11:11:08 -0800
Subject: Evaluating the contents of a string
In-Reply-To: <snitl5r6oy.fsf@motorola.com>; from aaron.ginn@motorola.com on Mon, Mar 19, 2001 at 10:48:13AM -0700
References: <snitl5r6oy.fsf@motorola.com>
Message-ID: <20010319111108.C3109@trufflehunter.avalongroup.net>

On Mon, Mar 19, 2001 at 10:48:13AM -0700, Aaron Ginn wrote:
> 
> Is there a simple way to evaluate the contents of a string assuming
> that string contains only nunbers and mathematical operators?
> 
> For example, I'm parsing a file that contains expressions contained in 
> parantheses such the following:
> 
> 12 net64 a vss vss n w=1.1u l=0.5u ad=(1.1 * 1.15) as=(1.1 * 1.15)
>                                       ^^^^^^^^^^^^    ^^^^^^^^^^^^
> 
> I'm using a stack to determine when an expression can be evaluated
> (i.e., when a closing paranthesis is found), and I'm storing the
> contents of the parantheses in a string.  When the first closing
> parenthesis is found in the example above, the string to be evaluated
> will be '1.1 * 1.15'.  Is there a module that will mathematically
> evaluate this string?

Using your example string above, the following appears to work:

>>> import re
>>> x = re.findall(r'[a-z][a-z]=\(.*?\)', x)
>>> for y in x:
...     exec y
... 
>>> dir()
['__builtins__', '__doc__', '__name__', 'a', 'ad', 'as', 're',
'x', 'y', 'z']
>>> 
>>> print ad
1.265
>>> print as
1.265
>>> 


-- 
Stand Fast,
    tjg.

Timothy Grant                         tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.         <><       (503) 246-3630
>>>>>>>>>>>>>Linux, because rebooting is *NOT* normal<<<<<<<<<
>>>>This machine was last rebooted:  61 days 23:25 hours ago<<



From djc at object-craft.com.au  Sat Mar 24 22:50:51 2001
From: djc at object-craft.com.au (Dave Cole)
Date: 25 Mar 2001 13:50:51 +1000
Subject: Referencing a container object's attributes
References: <200103250239.SAA10240@mail.novus-tele.net> <mailman.985488572.31554.python-list@python.org>
Message-ID: <m3ae6azeuc.fsf@vole.object-craft.com.au>

>>>>> "Tim" == Tim Churches <tchur at optushome.com.au> writes:

>> Is so, there there is no language feature that would allow you to
>> do this without passing an instance of 'b' to 'a'. Standard
>> practice would be to do something like this:
>> 
>> class a:
>>     def __init__(self, parent):
>>         self.parent = parent
>> 
>> def doSomething(self):
>>     self.parent.blah( )
>> 
>> class b:
>>     __init__(self):
>>         self.a = a( self )
>> 

Tim> OK, many thanks.

Tim> One question: does the circularity inherent in the above (in
Tim> which a reference to the container object is placed in an
Tim> attribute of each containee) have any potential for causing
Tim> problems - particularly since I want to pickle the container
Tim> objects?

You should be able to pickle the container, but you will have to
manually remove all of the circular references yourself when you want
to free the object(s).

In Python 2.x I understand that the garbage collector can detect and
handle circular references.

- Dave

-- 
http://www.object-craft.com.au


From emile at fenx.com  Mon Mar 12 07:23:48 2001
From: emile at fenx.com (Emile van Sebille)
Date: Mon, 12 Mar 2001 04:23:48 -0800
Subject: I have no idea
References: <g13r6.201$qt5.10920@news3.mia>
Message-ID: <98ifa7$26rae$1@ID-11957.news.dfncis.de>

while 1:
    print "Hello World"

>"Brendhan Horne" <brendhanhorne at bellsouth.net> wrote in message
news:g13r6.201$qt5.10920 at news3.mia...
> In basic you use to be able to do this:
> 10 print "Hello World"
> 20 goto 10
> Run
>
> If you did that it would run up the screen in an endless loop until you
hit
> the ESC key. If you added   ; it would cover the whole screen. How do you
> write the equivalent in python?
>
> --
> Thanks,
> Brendhan
>
>




From erno-news at erno.iki.fi  Sun Mar  4 19:29:10 2001
From: erno-news at erno.iki.fi (Erno Kuusela)
Date: 05 Mar 2001 02:29:10 +0200
Subject: Concurrency and shelve
References: <3AA26D5E.2FD28B40@stroeder.com> <kuk865l6e1.fsf@lasipalatsi.fi> <3AA2C1F0.980A23F7@stroeder.com> <kuelwdkvq0.fsf@lasipalatsi.fi>
Message-ID: <ku8zmlkqi1.fsf@lasipalatsi.fi>

In article <kuelwdkvq0.fsf at lasipalatsi.fi>, Erno Kuusela
<erno-news at erno.iki.fi> writes:

| In article <3AA2C1F0.980A23F7 at stroeder.com>, Michael Str?der
| <michael at stroeder.com> writes:

|| BTW: I'm trying to use shelve to store web session data. How does
|| others solve this problem with the the standard lib without being to
|| DB-specific? Unpossible?

| don't keep it open all the time.

erm, i seem to have missed the "not db-specific" part. i guess
the answer is to do the locking yourself. there are some functions
in the fcntl module to do it. it's probably best to use a separate
empty file (like dbfilename + '.lock') that you do the locking on,
so that it does not interfere with gdbm's locking.

  -- erno


From tim.one at home.com  Mon Mar 12 00:55:06 2001
From: tim.one at home.com (Tim Peters)
Date: Mon, 12 Mar 2001 00:55:06 -0500
Subject: From SETL to Python translation
In-Reply-To: <t7Yq6.12$N22.336@typhoon.nyu.edu>
Message-ID: <LNBBLJKPBEHFEDALKOLCIEPMJEAA.tim.one@home.com>

[cyberian bear]
> Hi guys I have 2 programs below that I must have in Python but I have
> them in SETL(whatever the hell that is) and I don't know anything about
> it. I tried finding stuff about it on the web but all I found was two
> small, irrelevant pages and no tutorials.

You must have missed:

    http://birch.eecs.lehigh.edu/~bacon/doc.html

You can download Robert Dewar's introductory "The SETL Programming Language"
paper via a link on that page.

> The thing is, if at least I knew what the 'Euler path construction', or
> 'Prime factors' are, I could just program them on my own without looking
> at the SETL code, but i don't.

prime_factors(n) takes an integer as input and returns a list of its prime
factors, via repeated trial division by 2, then by the odd numbers from 3 up
to the ceiling of the square root of n.

Euler_path(G) takes an undirected graph represented as a set of 2-element
lists (representing edges via the nodes they connect), and returns an Euler
path as a list of the nodes of G in an Euler-path order; it returns OM
(SETL's spelling for "there's no meanigful result") if no Euler path exists.
An Euler path is a traversal of the graph such that each edge is taken
exactly once.

These are standard algorithms in their fields, and as such can be found all
over the web.

> Basically I just want to ask some good soul if he could translate these
> two programs for me from SETL to Python. I know chances of finding anyone
> who knows SETL are negligibe but i guess it's worth a try.
> ...

Finding someone who knows SETL is easier than finding someone who's keen to
solve what appears to be a homework problem <0.5 wink>.  Study the algorithms
and reimplement them from scratch; line-by-line translation of SETL is going
to be very frustrating.  For example, the closest simple and readable
translation of the SETL:

>         if #(odds := {x in nodes | odd(#G{x})}) > 2 then

is the Python:

    odds = []
    for x in nodes:
        successors = [y for (z, y) in G if z == x]
        if len(successors) % 2:
           odds.append(x)
    if len(odds) > 2:

but even then lots of representation changes have been made (the SETL set G
is taken to be a list of 2-tuples in Python instead; and the SETL set odds is
constructed instead as a Python list, exploiting that I happen to know no
element appears more than once in the SETL-set / Python-list "nodes").  The
SETL is a more direct translation of an English algorithm, reading "Let
'odds' be the set of nodes with an odd number of successors.  If there are
more than two of those, then ...".

prime_factors is (by far) the easier of the two, so start with that.

setl-is-to-ltes-as-python-is-to-nohtyp-ly y'rs  - tim




From tim.one at home.com  Thu Mar  1 22:41:57 2001
From: tim.one at home.com (Tim Peters)
Date: Thu, 1 Mar 2001 22:41:57 -0500
Subject: define a new func on the fly?
In-Reply-To: <3a9f0ce3.10347756@News.CIS.DFN.DE>
Message-ID: <LNBBLJKPBEHFEDALKOLCIEEEJDAA.tim.one@home.com>

[Costas Menico]
> Here is the exact error I get when running it
> 
> >>>> exec("def f():\n\treturn 'hello'\t")
> Traceback (innermost last):
>   File "<interactive input>", line 1, in ?
>   File "<string>", line 2
>     return 'hello'
>                   ^
> SyntaxError: invalid syntax
>
> I've trying all day to get this to work. No luck...

>>> exec "def f():\n\treturn 'hello'\n"
>>> f()
'hello'
>>>

1. exec is a statement, not a function, so you don't need the parens
   (although they shouldn't hurt).

2. You *intended* to end your string with a newline, not a tab.  The
   tab is what yields the SyntaxError.  A good clue about that is the
   caret pointing after the final ' in 'hello' in the error message.




From fredrik at pythonware.com  Thu Mar 15 12:45:22 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Thu, 15 Mar 2001 17:45:22 GMT
Subject: Random Access Files
References: <9e4s6.42245$p66.13534448@news3.rdc1.on.home.com>
Message-ID: <Sc7s6.42$4N4.8297@newsc.telia.net>

Colin Meeks wrote:
> I would like to be able to do something like
>
>     1.    Seek the potion to write

f.seek(offset)

>     2.    Write

f.write(data)

>     3.    Close file

f.close()

> I don't mean writing code that hides the methods of the first example.  If I
> remember correctly you can do random access files in GWBasic.
>
> Anyone have any pointer??

http://www.python.org/doc/current/lib/bltin-file-objects.html

Cheers /F

<!-- (the eff-bot guide to) the standard python library:
http://www.pythonware.com/people/fredrik/librarybook.htm
-->




From tuttledon at hotmail.com  Sat Mar 31 23:09:45 2001
From: tuttledon at hotmail.com (Don Tuttle)
Date: Sun, 01 Apr 2001 04:09:45 GMT
Subject: PythonLabs Team Moves to ActiveState
References: <mailman.986086501.5576.python-list@python.org>
Message-ID: <dSxx6.237223$Z8.51066000@typhoon.southeast.rr.com>

"Guido van Rossum"
> PythonLabs Team Moves to ActiveState

To paraphrase a popular saying in my neck of the woods...
"Guido said it.  I believe it.  That settles it!"

(or was that just one of the reasons I deconverted?  ;-)

Don




From mbarnett at ualberta.ca  Mon Mar 19 13:49:14 2001
From: mbarnett at ualberta.ca (Matt Barnett)
Date: Mon, 19 Mar 2001 11:49:14 -0700
Subject: Evaluating the contents of a string
References: <snitl5r6oy.fsf@motorola.com>
Message-ID: <mwst6.4525$Dp6.681183@news-rep.ab.videon.ca>


----------
In article <snitl5r6oy.fsf at motorola.com>, Aaron Ginn 
<aaron.ginn at motorola.com> wrote:


>
> Is there a simple way to evaluate the contents of a string assuming
> that string contains only nunbers and mathematical operators?
>
> For example, I'm parsing a file that contains expressions contained in
> parantheses such the following:
>
> 12 net64 a vss vss n w=1.1u l=0.5u ad=(1.1 * 1.15) as=(1.1 * 1.15)
>                                       ^^^^^^^^^^^^    ^^^^^^^^^^^^

eval("1.1 * 1.15")

>When the first closing
>parenthesis is found in the example above, the string to be evaluated
>will be '1.1 * 1.15'.  Is there a module that will mathematically
>evaluate this string?

Yup


From f8dy at yahoo.com  Fri Mar  9 13:26:48 2001
From: f8dy at yahoo.com (Mark Pilgrim)
Date: Fri, 9 Mar 2001 13:26:48 -0500
Subject: I come to praise .join, not to bury it...
References: <mailman.984074812.27842.python-list@python.org> <98aspo024jg@news2.newsguy.com>
Message-ID: <98b75u$15les$1@ID-77331.news.dfncis.de>

"Alex Martelli" <aleaxit at yahoo.com> wrote in message
news:98aspo024jg at news2.newsguy.com...
> But the .join/.writelines parallel is VERY close, so why
> would the SAME architectural choice (to have the sequence
> object as an argument, accessing it through its ability
> to enumerate its items, only) cause problems in one case
> and not the other?!
>
> I think it's a misperception due to thinking of strings
> as somehow 'passive', 'data', while a file-object is
> thought of as 'active', 'a truly object-y object', or
> something.

Exactly!  I have taught Python to half a dozen people in the last few
months, and every one of them had trouble with this concept, because they
all came from a C background where they only thought of "string" as "packed
array of char", or some other language (like Powerbuilder) where "string" is
just a native datatype and isn't an object in any sense of the word.  Nobody
minded file objects, even if they were only used to using integer file
handles in their previous language.  I think that's because they could build
a mental model of a file object as a kind of secretary who has a copy of a
document on their desk and can read you various parts of it, or all of it,
or tell you its title, or whatever.  But everybody had trouble with string
objects, because the mental model is different.

IIRC, the post that started the thread that spawned this thread quoted a
code sample from my book, something like this:
    return ";".join([...])
There are obviously lots of ways to do this (assign ";" to a variable first,
or use the old string.join function, or don't use join at all), but I did it
this way on purpose.  Call it shock value, for lack of a better term -- to
force the point that the literal string ";" is an object, not just string
variables.  I have gotten multiple e-mails from readers telling me that this
was their "a-ha!" moment when they realized that everything in Python really
*is* an object.  After that moment, they don't mind seeing modules that have
attributes, or functions passed around as parameters, or anything other
weird/cool/wicked-ness.  "Sure, you can do that, you can do anything,
because everything is an object; just look at what you can do with string
literals..."

-M
You're smart; why haven't you learned Python yet?
http://diveintopython.org/





From db3l at fitlinxx.com  Tue Mar 20 18:48:11 2001
From: db3l at fitlinxx.com (David Bolen)
Date: 20 Mar 2001 18:48:11 -0500
Subject: Little red lines in PythonWin!
References: <mailman.985105228.27509.python-list@python.org> <9OQt6.13652$992.84056@news-server.bigpond.net.au>
Message-ID: <u3dc8q9xg.fsf@ctwd0143.fitlinxx.com>

"Neil Hodgson" <neilh at scintilla.org> writes:

>    This feature is turned on by default in PythonWin and can be controlled
> through View | Options... | Tabs and Whitespace | Tab-Timmy. IMO all Python
> IDEs should have a similar feature. For those IDEs that use Scintilla, the
> feature is already implemented for you so it is almost free to add.

Interestingly enough when I looked at this in response to the first
post (I don't normally use Pythonwin), I noticed that the default
setup seems to have a tab size of 4 rather than 8.  So Pythonwin was
expanding literal tabs visually to a 4-position stop rather than 8,
which is different from the language definition.  Expanding to 8 by
default (with or without the red indicators) would at least present a
visual indication of what the Python parser was going to "see".

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


From lluang at northwestern.edu  Mon Mar 12 17:13:07 2001
From: lluang at northwestern.edu (Louis Luangkesorn)
Date: Mon, 12 Mar 2001 16:13:07 -0600
Subject: sent a note to Lemburg
References: <mailman.983993842.26448.python-list@python.org> <tUzp6.29925$ML1.1507874@e420r-atl2.usenetserver.com> <3AA7FF0E.50E6D70F@northwestern.edu> <Oesq6.12063$v5.52124@newsfeeds.bigpond.com>
Message-ID: <3AAD49F3.B9FF9480@northwestern.edu>

An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20010312/0ac946f2/attachment.html>

From chris.gonnerman at usa.net  Tue Mar  6 21:31:40 2001
From: chris.gonnerman at usa.net (Chris Gonnerman)
Date: Tue, 6 Mar 2001 20:31:40 -0600
Subject: Overriding UserList methods
References: <b566atck0ruu4tae8n1srrkntbmme26gor@4ax.com> <aUFo6.14917$ML1.731870@e420r-atl2.usenetserver.com> <qqu9atsitacnuacu29k25rvtrieofeg6nk@4ax.com>
Message-ID: <009201c0a6ae$d59c8960$a100000a@local>

----- Original Message -----
From: "Daniel Klein" <danielk at aracnet.com>
Subject: Re: Overriding UserList methods
> On Mon, 5 Mar 2001 00:52:40 -0500, "Steve Holden" <sholden at holdenweb.com>
> wrote:
>
> >"Daniel Klein" <danielk at aracnet.com> wrote in message
> >news:b566atck0ruu4tae8n1srrkntbmme26gor at 4ax.com...
> >> I must be missing something, cos I can't figure out how to override a
> >method
> >> using the UserList module. Here is some code:
> >>
> >> import UserList
> >> class MyList(UserList.UserList):
> >>       __def __init__(self):
> >>         UserList.UserList.__init__(self)
> >>
> >While this would work, I don't see it's strictly necessary. Why not just
> >inherit the __init__ method from UserList?
>
> One of the reasons for subclassing UserList is to provide additional
instance
> variables. So I need my own __init__ method to initialize this, don't I?

Nope.  Only if you *actually* initialize them in the __init__ method, which
you aren't doing.

I wrote a module a while back that subclasses UserDict... not exactly the
same
problem but close enough to help.  If you are interested, visit

    http://newcenturycomputers.net/projects/python

and download the Chunk.py module.

Remember that when you call a method of an instance variable, i.e.:

    from UserList import UserList
    lst = UserList()
    lst.append(1)

what is (effectively) called is:

    UserList.append(lst, 1)

So:

    class MyList(UserList):
        def append(self, item):
            UserList.append(self, item)

is the "right" way to do it.  Of course, you can also just use the .append()
method of the underlying .data element, but I prefer to use the "proper"
encapsulation.






From tim.one at home.com  Mon Mar 12 14:45:43 2001
From: tim.one at home.com (Tim Peters)
Date: Mon, 12 Mar 2001 14:45:43 -0500
Subject: PyEval_CallObject and threads
In-Reply-To: <H58r6.9$697.25923@news.pacbell.net>
Message-ID: <LNBBLJKPBEHFEDALKOLCEECFJFAA.tim.one@home.com>

[Phlip]
> My colleague came over and asked me to write this:
>
> "We got multiple C++ pthreads calling PyEval_CallObject. How threadsafe
> is the Python API? A C++ module using Boost BPL creates multiple threads,
> which then call back into Python. When two threads simultaneously call
> PyEval_CallObject we go boom."
>
> I will now go examine his serialization primitives. But if anyone has any
> better ideas...

With very few exceptions, all calls to Python C API functions must be made
while holding the global interpreter lock.  The exceptions have mostly to do
with a handful of functions related to initialization and shutdown.  See the
section "Thread State and the Global Interpreter Lock" in the Python/C API
Reference Manual for details.




From gregj at pobox.com  Fri Mar  9 02:05:52 2001
From: gregj at pobox.com (Gregory Jorgensen)
Date: Fri, 09 Mar 2001 07:05:52 GMT
Subject: learning C
References: <mailman.983972829.23481.python-list@python.org> <983975751.18833@newsmaster-04.atnet.at> <3AA82422.53026D3D@cosc.canterbury.ac.nz> <mailman.984107772.3536.python-list@python.org>
Message-ID: <kh%p6.1339$54.1462@www.newsranger.com>

Stroustrup himself said it best: "C makes it easy to shoot yourself in the foot;
C++ makes it harder, but when you do it blows your whole leg off."

http://www.research.att.com/~bs/bs_faq.html#really-say-that


In article <mailman.984107772.3536.python-list at python.org>, D-Man says...
>
>On Fri, Mar 09, 2001 at 01:30:26PM +1300, Greg Ewing wrote:
>| Werner Schiendl wrote:
>| > 
>| > The fine thing is: C++ still HAS the scalpel, so what?
>| 
>| C++ is a Swiss army knife with both scissors and scalpel,
>| and a nasty tendency for the scalpel to shoot out and
>| stick you in the palm when all you were trying to do
>| was use the scissors...
>| 
>
>Nice.  I like this description. 
>
>yet-not-joining-the-war-ly y'rs  -D
>
>

Greg Jorgensen
Deschooling Society
Portland, Oregon, USA
gregj at pobox.com


From kens at sightreader.com  Wed Mar 21 20:05:47 2001
From: kens at sightreader.com (Ken Seehof)
Date: Wed, 21 Mar 2001 17:05:47 -0800
Subject: What is being compared?
References: <cfbu6.67$Hp.3494@typhoon.aracnet.com>
Message-ID: <014101c0b26c$3af6f890$04090a0a@upcast.com>

Documentation says:
"""
The operators <, >, ==, >=, <=, and != compare the values of two objects.
The
objects need not have the same type. If both are numbers, they are coverted
to
a common type. Otherwise, objects of different types always compare unequal,
and are ordered consistently but arbitrarily.
"""

I'm pretty sure that the default rules for mixed types are something like
this:
1. numbers are less than non-numbers
2. otherwise compare id(type(a)) with id(type(b))

Note that, rather weirdly, types are compared by their ids.  That's right.
In
other words, the address where the type object happened to get allocated.
Hence the behavior will be only be guaranteed to be consistent within an
execution of the python shell.  On some operating systems it will also
usually
to be mostly consistent from one execution to the next because the objects
tend to be allocated in the same order.

Actually, the documentation is slightly wrong.  Any type can overload
__cmp__
or the new rich comparisons.  Hence it is possible for instances and objects
to
define any inter-type behaviour you want.  In fact the only reason numbers
are
special is that they happen overload __cmp__ to do appropriate conversions.

>>> class God:
...  def __cmp__(self, x):
...     return 0
...
>>> c = God()
>>> c>42
0
>>> c<3
0
>>> c==3
1
>>> c=="spam"
1

----- Original Message -----
From: "Daniel Klein" <DanielK at jBASE.com>
Newsgroups: comp.lang.python
To: <python-list at python.org>
Sent: Wednesday, March 21, 2001 3:59 PM
Subject: What is being compared?


> Comparing a string to an integer...
>
> >>> 'a' < 1
> 0
>
> ..what is it actually comparing? Is it 'casting' anything?
>
> Just curious,
>
> Dan
>
>
> --
> http://mail.python.org/mailman/listinfo/python-list
>




From dalfes at itu.edu.tr  Tue Mar  6 19:08:45 2001
From: dalfes at itu.edu.tr (Nuzhet Dalfes)
Date: Wed, 07 Mar 2001 02:08:45 +0200
Subject: How to installing NumPy?
Message-ID: <3AA57C0D.BBC5E956@itu.edu.tr>

Hi, 

I am trying to install NumPy on a RedHat 7.0 running python2.0.
Everything is in the standard places (I think!) and I picked up and
unzipped http://download.sourceforge.net/numpy/NumPy1711_20.zip into
/usr/local/lib/python2.0/site-packages. And set PYTHONPATH to
/usr/local/lib/python2.0/site-packages.

But don't seem to be able to get 'import Numeric' going.

Help


Nuzhet
-------------- next part --------------
A non-text attachment was scrubbed...
Name: dalfes.vcf
Type: text/x-vcard
Size: 389 bytes
Desc: Card for Nuzhet Dalfes
URL: <http://mail.python.org/pipermail/python-list/attachments/20010307/21b2dd74/attachment.vcf>

From nospam at mega-nerd.com  Tue Mar 20 02:52:40 2001
From: nospam at mega-nerd.com (Erik de Castro Lopo)
Date: Tue, 20 Mar 2001 07:52:40 GMT
Subject: Tkinter : resizing Text widget with main window
References: <3AB11827.469F6C0C@mega-nerd.com> <slrn9bdls6.noo.rickp@tc.niof.net>
Message-ID: <3AB70C35.1B02410E@mega-nerd.com>

Rick Pasotto wrote:
> 
> On Thu, 15 Mar 2001 19:38:12 GMT in comp.lang.python, Erik de Castro Lopo wrote:
> > Hi all,
> >
> > How does one force a Tkinter.Text widget to be resized whenever
> > the main window is resized?
> 
> t=Text(width=30,height=12) # sets the initial size
> t.pack(expand=1,fill='both')

That may work for widgets which are pack()ed, but what about
widgets which are place in the parent window using grid ()?

This does work for grid ().

Erik
-- 
-----------------------------------------------------------------
Erik de Castro Lopo     nospam at mega-nerd.com      (Yes its valid)
-----------------------------------------------------------------
"I've got an idea! Sun, Oracle, and IBM hold down MS while Linux 
gets to kick them!" -- Lou Grinzo on LinuxToday.com


From kens at sightreader.com  Mon Mar 26 19:15:05 2001
From: kens at sightreader.com (Ken Seehof)
Date: Mon, 26 Mar 2001 16:15:05 -0800
Subject: Oatmeal (was warrior geeks)
References: <3ABF72F1.98CB78F@sage.att.com> <3ABFA209.CF27BC20@optushome.com.au>
Message-ID: <003101c0b652$fa514ee0$04090a0a@upcast.com>

I like oatmeal (porridge) too.

----- Original Message ----- 
From: "Tim Churches" <tchur at optushome.com.au>
To: "Garry Hodgson" <garry at sage.att.com>
Cc: <python-list at python.org>
Sent: Monday, March 26, 2001 12:09 PM
Subject: Re: warrior geeks


> Garry Hodgson wrote:
> > 
> > a funny thing happened at IPC9.  a bunch of were sitting in the
> > lounge (bar) talking, and one person mentioned that he was going
> > to study kendo while in japan this year.  i asked if he had previous
> > martial arts experience, and he said yes, he'd studied judo.
> > it then turned out that in this group of maybe 6-7 people, we had
> > this judo/kendo student, another who's done karate and now kickboxing,
> > another who did karate, myself (okinawan karate, aikijutsu and others),
> > and a pa qua practitioner.  at the next table sat another tae kwon
> > do/aikido
> > student.
> > 
> > it seemed like an odd coincidence, but it made me wonder: is there an
> > unusually high percentage of MA practitioners among the python
> > community?
> > or the open source community?  or software in general?
> 
> I'll use our time zone advantage to jump in early on this thread and
> opine that it is way off-topic. Not that I am opposed to
> marital...whoops...martial arts (although they are not my cup of tea),
> but I am concerned that if this threads continues, sooner or later
> someone will mention the whole "Geeks with Guns" thing that Eric S.
> Raymond promotes (see http://www.tuxedo.org/~esr/geeks-with-guns/ or for
> even worse IMHO see http://www.tuxedo.org/~esr/guns/ ), especially given
> ESR's apparent fondness for Python. That sort of thing gives me a fit of
> the shudders (playing with guns, not ESR's fondness for Python). I would
> have thought that if male hackers were looking for a hobby which
> involves phallic-shaped objects and chemically-propelled projectiles,
> then model rocketry would be just the ticket.
> 
> There, that should put an immediate end to this very off-topic thread...
> 
> Tim C
> Sydney, Australia
> PS Damn! I was cooking porridge (oatmeal) for breakfast and became so
> carried away composing this message that I completely forgot to stir the
> pot...
> 
> -- 
> http://mail.python.org/mailman/listinfo/python-list
> 




From fredrik at pythonware.com  Wed Mar 14 10:55:12 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 14 Mar 2001 15:55:12 GMT
Subject: String comparison
References: <3AAF9290.42BDAAD3@mailhost.stg.rt.daimlerchrysler.com>
Message-ID: <AvMr6.244$4d3.53135@newsb.telia.net>

Alexander Stirzel wrote:
> I printed processname just before this comparison and I got "netscape"
> on screen. (With no spaces there, that's not the problem) But when I
> >>>> print (processname=="netscape")
> I get a "0" on screen. And when I
> >>>> print (processname)
> I get "netscape" on screen.

maybe you have control characters (e.g. a carriage
return) left in the string?

try

>>> print repr(processname)

or just

>>> processname

(nit: print is a statement, not a function call.  don't use
parentheses around its argument)

Cheers /F

<!-- (the eff-bot guide to) the standard python library:
http://www.pythonware.com/people/fredrik/librarybook.htm
-->




From m.linnemann at addcom.de  Thu Mar 29 06:50:40 2001
From: m.linnemann at addcom.de (Michael Linnemann)
Date: Thu, 29 Mar 2001 13:50:40 +0200
Subject: Python scripts as Drop targets in Windows Explorer
References: <99qogo$o6g$1@troll.powertech.no> <99qqq3$qt7$0@216.39.170.247> <99r043$qqr$1@troll.powertech.no> <vdqs99.p16.ln@haegar.fqdn.th-h.de> <8kAw6.780$R6.22057@news1.oke.nextra.no>
Message-ID: <gi7v99.hp.ln@haegar.fqdn.th-h.de>

Am Thu, 29 Mar 2001 08:10:08 +0200 schrieb Syver Enstad:

>> Yeah, looks cool. What is it, perl? Czech? EBCDIC?

>>>default on Win2k is to save as unicode.

>> Come on, tune your W2k a little and make it talk plain :-)
>I don't know what you mean, isn't Unicode one of the most plain ways to talk
>nowadays? And I have also posted it as plain text so it shouldn't be too
>difficult for you.

It is not really difficult, especially for me in particular. I don ot
care about registries because I have none :-)

But you posted the code that I cited, and my newsreader (slrn) did _not_
trigger any mime decoding or something, that is, it showed your unicode
as plain ascii. And that looks like plain garbage.

So, either your newsreader told you it was ascii, but it wasn't; or it
knew it was _not_ ascii but did not declare it.

Regards
Michael


From freier at xtramind.com  Thu Mar 22 05:48:48 2001
From: freier at xtramind.com (Andreas Freier)
Date: Thu, 22 Mar 2001 11:48:48 +0100
Subject: Why Python/Jython?
Message-ID: <99cl9e$b2b$1@Merlin.et.bocholt.fh-gelsenkirchen.de>

In several articles I read that Python/Jython imporoves the programmers
performance many times. Why?
I am considering to use Jython for my job as Java programmer. How should I
use it to achieve this performance?

Thanks a lot.
Andreas Freier






From bsass at freenet.edmonton.ab.ca  Thu Mar  1 04:49:46 2001
From: bsass at freenet.edmonton.ab.ca (Bruce Sass)
Date: Thu, 1 Mar 2001 02:49:46 -0700 (MST)
Subject: CPAN functionality for python - requirements
In-Reply-To: <01022721090603.20069@branagan>
Message-ID: <Pine.LNX.4.33.0103010155520.386-100000@bms>

On Tue, 27 Feb 2001, Doug Hellmann wrote:
> On Tue, 27 Feb 2001, Bruce Sass wrote:
> > I'm a little concerned about the privacy aspect of this...
> >
> > > That decision is up to the client.  If the client has the smarts to turn
> > > a distutils package into an RPM, then the client would list it's preferred
> > > format as a distutils package and it would handle the rest.  If it can't,
> > > you can either select an RPM, or fall back to a distutils package.
> >
> > Why should the client need to "list" anything.
> >
> > client: what do you have?
> > server: this(deb,rh-5.rpm,rh-6.2.rpm,rh-7.rpm) that() other(hqx)
<...>
> I'm not sure why the "what do you have" question is needed.  The "send me
> that(mandrake.rpm)" interaction is what we want.

Automated tools need to query what is available so they can decide
where to satisfy a dependency from.  Packages will be available from a
variety of outlets, not just the python archive.

> The client requests packages in the formats prefered.  A standard source format
> should be available for all packages so that everybody can get every package in
> some form.  If the standard source format contains distutil-enabled scripts,
> platform specific distribution files could be generated from the source.

Yup.  What do you think of having the server look like a platform
specific distribution archive?  So, when the server sees requests for
(lets say) Debian style Packages and Release files (used to create the
available packages DB), it assembles them from the list of python
packages.  Requests for URLs that look like those of packages in RHs
archive whould result in a RH-rpm being served up.  This would allow
for complete integration with any client's native package archive
system the server knows about, and give the CPyAn archive full control
over package names, release numbers, etc. - the little bits that can
cause simple to fix, but annoying, dependency problems.
just a thought

> > This could work...
> > client: only show me (deb) and ()
> > client: what do you have?
>
> Right, that would be a bit more efficient than specifying a format for every
> package for every request.

There could also be an (all) format, why limit the possibilities.

> > The server should NOT be usable as a tool to track Python users and
> > their habits, and making it do so should require a conscious effort on
> > the operators part (so there is no opportunity for the operator to
> > say, "I don't track you, that's just how the software works").
>
> The server is likely to be a cgi, which by its nature means requests may be
> written to a log file.  Should that be disallowed?

Obviuosly there is nothing that can be done about normal system
logging, but the cgi/factory-object/whatever should not keep any
persistent user/connection information around (i.e., no log from the
CPyAN software which lists who got what in which format when).


- Bruce




From michael at stroeder.com  Sun Mar 18 05:50:38 2001
From: michael at stroeder.com (Michael =?iso-8859-1?Q?Str=F6der?=)
Date: Sun, 18 Mar 2001 11:50:38 +0100
Subject: Help with 500 error on cgi script
References: <quv8bto7r8s6f1tckc7qpmuir0kvufp1nb@4ax.com> <mailman.984909623.8814.python-list@python.org> <5739bt04e50ku01rn3dptgqch3l07p2cnf@4ax.com>
Message-ID: <3AB492FE.4D963C32@stroeder.com>

Sheila King wrote:
> 
> [Sun Mar 18 02:18:42 2001] [error] [client 4.3.200.226]     print "<H1>Good
> Result!</H1>
> 
> [Sun Mar 18 02:18:42 2001] [error] [client 4.3.200.226]
> ^
> 
> [Sun Mar 18 02:18:42 2001] [error] [client 4.3.200.226] SyntaxError: invalid
> token
> 
> But I'm at a loss to figure out what is wrong. :-(

What keeps you away from correcting the syntax error mentiond in the
error traceback?

Ciao, Michael.


From aahz at panix.com  Thu Mar 22 11:31:28 2001
From: aahz at panix.com (Aahz Maruch)
Date: 22 Mar 2001 08:31:28 -0800
Subject: Jython is cool?
References: <20010322103954.A18539@sync.nyct.net> <mailman.985276416.27994.python-list@python.org>
Message-ID: <99d9d0$n2a$1@panix6.panix.com>

In article <mailman.985276416.27994.python-list at python.org>,
Neil Schemenauer  <nas at arctrix.com> wrote:
>
>I'm hoping that the Mozilla guys catch the Python bug based on
>what Active State has been doing with Komoto and adds "<script
>language=python>" to Mozilla.

Google protection: "Komodo"
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

"I won't accept a model of the universe in which free will, omniscient
gods, and atheism are simultaneously true."  -- M


From bkline at rksystems.com  Wed Mar 28 12:17:40 2001
From: bkline at rksystems.com (Bob Kline)
Date: Wed, 28 Mar 2001 12:17:40 -0500 (EST)
Subject: Speed problems with Python vs. Perl
In-Reply-To: <tMnw6.77278$Xt3.10616558@news1.rdc1.az.home.com>
Message-ID: <Pine.LNX.4.10.10103281209290.29230-100000@rksystems.com>

On Wed, 28 Mar 2001, Tim Hochberg wrote:

> I haven't actually benchmarked (or even tried) any of these, but I'd
> be curious to see how the last one in particular stacks up against
> it's Perl equivalent. I don't have Perl installed at the moment,
> though so I leave that as an excercise for the reader....

$ cat tsplit.py
#!/usr/bin/python
import sys, string

i = 0
for line in sys.stdin.readlines():
    i = i + 1
    f = string.split(line)

print "Total lines read: " + `i`
$ cat tsplit.pl
#!/usr/bin/perl
$icount = 0;

while(<>) {
  $icount++;
  @f = split;
}
print "Total lines read: $icount\n";
$ time ./tsplit.py < manual.txt
Total lines read: 32542

real    0m4.865s
user    0m4.780s
sys     0m0.090s
$ time ./tsplit.pl < manual.txt
Total lines read: 32542

real    0m4.452s
user    0m3.140s
sys     0m0.060s

Fairly comparable.  Don't have a version with xreadlines() yet, so I
can't say how much improvement that would add for the python times.

-- 
Bob Kline
mailto:bkline at rksystems.com
http://www.rksystems.com





From dyoo at hkn.eecs.berkeley.edu  Fri Mar 16 07:47:36 2001
From: dyoo at hkn.eecs.berkeley.edu (Daniel Yoo)
Date: Fri, 16 Mar 2001 04:47:36 -0800 (PST)
Subject: Basic information requested
In-Reply-To: <E14drbv-0001Qd-00@mail.python.org>
Message-ID: <Pine.LNX.4.21.0103160445090.5412-100000@hkn.eecs.berkeley.edu>

Jamie MacIsaac <jamie at cctech.co.uk> wrote:
> Hi,
 
> I've just started looking at Python and would like to use it to perform
some
> simple admin tasks on a Windows network. However, I'm having difficulty
> finding out how to accomplish elementary tasks.
 
> Can someone please point me to sample code that shows how to map a
network
> share and copy some files? Is there a central resource of code samples?
 

Another source of sample Python is the Useless Python pages:

    http://www.lowerstandard.com/python/pythonsource.html

as well as the Vaults of Parnassus:

    http://www.vex.net/parnassus


At the moment, people are organizing to write the Python Cookbook, so you
might want to check with this resource too:

    http://pythoncookbook.activestate.com




From rob at hooft.net  Wed Mar 14 10:18:58 2001
From: rob at hooft.net (Rob Hooft)
Date: 14 Mar 2001 16:18:58 +0100
Subject: Speeding up glob.glob?
Message-ID: <m2elw0wf8d.fsf@hooft.net>

I have been playing around a bit with file name globbing, because 
I was getting performance problems in one of my programs that repeatedly
counts the number of files matching a certain pattern. In fact, I was
counting all files like "*.kcd", "*.kcd.Z", "*.kcd.bz2" and "*.kcd.gz".

To do that, I used to have a glob function in my "compress" module that
looked like this:

table={'.Z':'compress','.gz':'gzip','.bz2':'bzip2'}

def glob(pattern):
    """
    Same as glob.glob, but includes compressed files too.
    """
    import glob
    r=glob.glob(pattern)
    for ext in table.keys():
        r.extend(glob.glob(pattern+ext))
    return r

In a largish directory, this would take 0.3-0.5 seconds.
After some play, I ended up with:

def glob(pattern):
    """
    Same as glob.glob, but includes compressed files too.

    Contains optimization in case pattern does not contain a '/'
    """
    import fnmatch,re
    exp=fnmatch.translate(pattern)[:-1]+"("
    for ext in table.keys():
        exp=exp+"\\"+ext+"|"
    exp=exp[:-1]+")?$"
    pat=re.compile(exp)
    r=[]
    if os.sep in pattern:
        import glob
        fl=glob.glob(pattern+'*')
    else:
        fl=os.listdir('.')
    for filename in fl:
        if pat.match(filename):
            r.append(filename)
    return r

To my surprise, this version runs in 0.01-0.02 seconds in similar
largish directories.  I think a large fraction of this gain can also
be obtained in glob.glob itself. Currently the code reads:

-----------
        dirname, basename = os.path.split(pathname)
        if has_magic(dirname):
                list = glob(dirname)
        else:
                list = [dirname]
-----------

I think this could be something like:

-----------
        dirname, basename = os.path.split(pathname)
	if not dirname:
	        return glob1(os.curdir,basename)
        elif has_magic(dirname):
                list = glob(dirname)
        else:
                list = [dirname]
-----------

This saves lots of os.path.join's with empty directory paths.

Regards,

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


From nick at olsen.ch  Fri Mar  2 08:43:44 2001
From: nick at olsen.ch (Nick Jacobs)
Date: Fri, 02 Mar 2001 13:43:44 GMT
Subject: Splitting comp.lang.python
References: <mailman.983199848.8916.python-list@python.org> <Hgwm6.2969$8b1.507773@news4.rdc1.on.home.com> <79ym6.2288$TW.12847@tor-nn1.netcom.ca>
Message-ID: <3A9FA38B.1E5E03E2@olsen.ch>

Warren Postma wrote:

> Not again!?  We had this discussion ten times over. Anyone who feels like
> weighing in on this one should at least be polite and avoid beating the same
> dead horses, over and over again.
>

But comp.lang.python gets more traffic than it used to. The case for
splitting it is stronger than it was.  IMHO it is almost inevitable that
at some point in the not-very-distant future, it will be split. The
meaningful debate is about when and how, not about whether.





From bedge at troikanetworks.com  Thu Mar 22 11:14:01 2001
From: bedge at troikanetworks.com (Bruce Edge)
Date: Thu, 22 Mar 2001 08:14:01 -0800
Subject: Creating Python Data Server
References: <jQkk6.117367$Ch.20710042@newsrump.sjc.telocity.net> <uk86i5doi.fsf@ctwd0143.fitlinxx.com>
Message-ID: <20010322.081401.238962600.7262@mead.troikanetworks.com>

In article <uk86i5doi.fsf at ctwd0143.fitlinxx.com>, "David Bolen"
<db3l at fitlinxx.com> wrote:

> "dsavitsk" <dsavitsk at e-coli.net> writes:
> 
>> I am trying to set up client/server system like a database server with
>> stored procedures and what not where the procedures are written in
>> python
>> (or any other language), live on the server, and that isn't tied to any
>> particular database. So the user would press a button on the interface
>> sending a signal to the server which would instantiate whatever objects
>> were necessary on the server, process the data, and then return the
>> data to the client.
> 
> It sounds like a good fit for a straight forward RPC (remote procedure
> call) or remote component based sort of setup.

Or CORBA, look at OmniORB, the idl compiler generated Python bindings.

-Bruce.


From ben.hutchings at roundpoint.com  Wed Mar 28 17:46:33 2001
From: ben.hutchings at roundpoint.com (Ben Hutchings)
Date: 28 Mar 2001 14:46:33 -0800
Subject: Controling the execution of external programs on Win2K
References: <n6dw6.4720$45.24451@newsfeeds.bigpond.com>
Message-ID: <usnjx7bqu.fsf@roundpoint.com>

"Ben" <bencr at bigpond.com> writes:

> Hi,
> 
> I placed a post last night looking for a way to run exe's from a Python
> programme, where the Python programme waits for each instance of the exe to
> finish/return and exit code before proceding with it's next command.
> 
> As suggested I have looked at os.system( ) but on Win2K it seems every
> external program I execute spawns an external command shell and there
> doesn't seem a way to prevent this behaviour ... which is not good (atleast
> from what I am trying to do) ...  after a bit more investigation I believe
> it is possible to achieve this using the win32process module. I am just
> unsure of what args to pass to make it behave in this way.
<snip>

You can't do that directly with the CreateProcess function.  If you
were using the Win32 API directly you would call WaitForSingleObject
or WaitForMultipleObjects, passing the process handle it returns.
In Python you could use something like this:

    def CreateProcess(commandline, wait):
        import win32event, win32process
        process, thread, pid, tid = win32process.CreateProcess(
            None,
            commandline,
            None,
            None,
            0,
            win32process.NORMAL_PRIORITY_CLASS,
            None,
            None,
            win32process.STARTUPINFO())
        if wait:
            win32event.WaitForSingleObject(process, win32event.INFINITE)

-- 
Any opinions expressed are my own and not necessarily those of Roundpoint.


From hungjunglu at yahoo.com  Sun Mar 11 16:36:18 2001
From: hungjunglu at yahoo.com (hungjunglu at yahoo.com)
Date: Sun, 11 Mar 2001 21:36:18 -0000
Subject: Chinese text GIF file generation
Message-ID: <98gr4i+r42i@eGroups.com>

Hi,

Is there anything in Python (or C library) that take as input a text 
file in Chinese (BIG5/GB) and generates a GIF file? (Preferable in 
Windows.) I would imagine someone already wrote something for this in 
C...

If there is something similar for Vietnamese, I'd like to know, too.

regards,

Hung Jung




From robin at jessikat.fsnet.co.uk  Fri Mar  2 06:06:03 2001
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Fri, 2 Mar 2001 11:06:03 +0000
Subject: PEP 227 (was Re: Nested scopes resolution -- you can breathe again!)
References: <LNBBLJKPBEHFEDALKOLCMEEPJBAA.tim.one@home.com>
 <200102230259.VAA19238@cj20424-a.reston1.va.home.com>
 <mailman.983384008.21434.python-list@python.org>
 <97lqsr0311d@news2.newsguy.com>
 <mailman.983466367.2968.python-list@python.org>
Message-ID: <oz8bbLAb63n6EwfI@jessikat.demon.co.uk>

In article <mailman.983466367.2968.python-list at python.org>, Jeremy
Hylton <jeremy at alum.mit.edu> writes
>Or: global NAME in FUNCTION
>
>def eggs():
>    z = 2
>    def spam():
>        def inner():
>            global x in spam, z in eggs
>            x += 1
>            z += x
>        x = 23
>        inner()
>
>Jeremy
>
I suppose you could also add an 'as' clause eg

def eggs():
    x = 2
    def spam():
        def inner():
            global x in spam as spam_x, x in eggs as eggs_x
            spam_x += 1
            eggs_x += x
        x = 23
        inner()

-- 
Robin Becker


From fredrik at pythonware.com  Fri Mar  2 02:40:16 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 02 Mar 2001 07:40:16 GMT
Subject: Tkinter:  image doesn't exist.
References: <1iBn6.1691$Ry6.649664@typhoon2.ba-dsg.net>
Message-ID: <A7In6.16280$Qb7.2658868@newsb.telia.net>

David Allen wrote:
> What kind of a crappy error message is "image doesn't
> exist?"

it means exactly what it says: when created, the PhotoImage class
registers a photoimage object with Tk, but when you try to use it in
FooModule, it's no longer there.

are you perhaps using multiple root windows?

(each root has its own interpreter context, and images registered
in one context aren't visible in other contexts.  use Toplevel in-
stead).

Cheers /F

<!-- (the eff-bot guide to) the standard python library:
http://www.pythonware.com/people/fredrik/librarybook.htm
-->





From tim.one at home.com  Fri Mar 16 02:18:14 2001
From: tim.one at home.com (Tim Peters)
Date: Fri, 16 Mar 2001 02:18:14 -0500
Subject: PEP 6: Patch and Bug Fix Releases
In-Reply-To: <mailman.984662760.24168.clpa-moderators@python.org>
Message-ID: <LNBBLJKPBEHFEDALKOLCEEALJGAA.tim.one@home.com>

[Aahz Maruch]
> ...
>     At the same time, many of the core Python developers are
>     understandably reluctant to devote a significant fraction of their
>     time and energy to what they perceive as grunt work.

That's not really the hangup for PythonLabs.  For example, the overwhelming
bulk of my Python time is already consumed by grunt work.  I don't mind that;
I get paid for it, after all <$0.01 wink>.  The problem is that there's
simply no more time to give!

Bugfix releases are a great idea -- along with uncounted dozens of other
great ideas there's no time for either.  I have high hopes, though, that for
bugfix releases, the community can be effective in making up for our
scandalous sloth.  Thank you for pushing it along!

and-we'll-put-in-more-bugs-to-make-you-look-good<wink>-ly y'rs  - tim




From costas at meezon.com  Wed Mar 14 22:38:23 2001
From: costas at meezon.com (Costas Menico)
Date: Thu, 15 Mar 2001 03:38:23 GMT
Subject: Status of IDLE
Message-ID: <3ab038b5.9422105@News.CIS.DFN.DE>

Does anyone know the status of IDLE? Is anyone working on making
ehancements on this?

costas


From mickeyz at qwest.net  Sun Mar  4 01:00:52 2001
From: mickeyz at qwest.net (Randy A. Ynchausti)
Date: Sat, 3 Mar 2001 23:00:52 -0700
Subject: Collection interfaces
References: <3a90fa97.0@news1.mweb.co.za>   <CgJk6.515$v77.66568@newsread2.prod.itd.earthlink.net> <220220010813598042%cmh@bDistributed.com> <ud7cakvi8.fsf@roundpoint.com> <MPG.150312ceb68f1ada98969f@news.earthlink.net> <8Gem6.473353$U46.14105062@news1.sttls1.wa.home.com> <MPG.1503548620e632919896a4@news.earthlink.net> <MCjm6.474652$U46.14163288@news1.sttls1.wa.home.com> <MPG.150394195f7f04a19896a7@news.earthlink.net> <3A9A86B1.9A8C69A2@ix.netcom.com> <MPG.1504d14563dce6499896af@news.earthlink.net> <97fjla128ir@news2.newsguy.com> <MPG.1505a76831c34b749896b3@news.earthlink.net> <97h5s40vhc@news2.newsguy.com> <MPG.1506007591cf818b9896bc@news.earthlink.net> <slrn99opop.gr.amk@mira.erols.com> <MPG.1506265ebf6d934d9896be@news.earthlink.net> <97ibmc0k47@news1.newsguy.com> <MPG.1506cf3293b094b49896c1@news.earthlink.net> <97m0c7080c@news2.newsguy.com> <MPG.15085517debf4b649896ca@news.earthlink.net> <zb%n6.31033$o%6.586026@news1.rdc1.bc.home.com> <MPG.150b383cea226d409896d0@news.earthlink.net>
Message-ID: <kNko6.6986$eq3.707121@news.uswest.net>

Topmind,

> I don't dispute it's popularity, at least as a buzzword
> (actual decent usage of OO is even disputed by many
> OO fans.)

> You never know. Merit and proof is not even a prerequisite.

> It hurts one career to admit distaste for OO. I have a family to
> take care of.

Please implement code (from scratch) for the following:

Floating Point Numerical Precision
    default precision
    radix
    machine precision
    negative machine precision
    smallest number
    largest number
Solve Polynomials
Solve Error Function
Solve Gamma Function
Solve Beta Function
Bulirsch-Stoer interpolation algorithm
Newton interpolation algorithm
Spline interpolation algorithm
Langrange interpolation algorithm
Neville interpolation algorithm

Please keep track of the number of minutes you spend working on this.
Please keep track of the number of defects that you find and fix during your
work.  I mean, any thing that does not work the first time and be sure to
count each problem the compiler tells you about too.  Please keep track of
the number of statements of code you add, remove, and change during this
work.  If you supply me with these data and the code I will test the
accuracy of your solutions and compare your code with some OO code and post
my findings (as my spare time allows) for the entire newsgroup(s) to see.

<TongueInCheek>Oh... and a... no pair programming please!</TongueInCheek>

Randy




From flognat at flognat.myip.org  Mon Mar  5 16:28:30 2001
From: flognat at flognat.myip.org (Andrew Markebo)
Date: Mon, 05 Mar 2001 21:28:30 GMT
Subject: Little problem with the "+" operator
References: <983802310.155838463@news.t-online.de> <980a7i0rg0@news2.newsguy.com>
Message-ID: <m3lmqjvrd0.fsf@master.athome>

/ "Alex Martelli" <aleaxit at yahoo.com> wrote:
| [...]
| > Does someone know how to solve the problem?
| 
| Sure -- make both variables into strings, e.g.
| 
|     myVariable = str(variableOne) + "cool"

NONOno.. he wants to add 500 with the temperature when it is cool, 0
degrees celsius or uhm what it now is in fahrenheit ;-)

        /Andy


From Arthur_Siegel at rsmi.com  Fri Mar 23 15:23:30 2001
From: Arthur_Siegel at rsmi.com (Arthur_Siegel at rsmi.com)
Date: Fri, 23 Mar 2001 14:23:30 -0600
Subject: VPython/Tk app works on single processor but not dual
Message-ID: <005B2EEA.N22121@rsmi.com>

>I really want to build VPython from source myself, so that I can
>instrument and tweak the C code to check out some theories I've got
>about why it's hanging.  

Probably unrelated - but a VPython problem I had was solved by
turning off hardware acceleration on my display settings, which seems 
to have had the side effect of stopping a chronic
crashing problem I was having on my Windows2000 machine
totally unrelated to VPython - and driving me nuts. 

The same problem on a Linux machine would have been 
conclusive proof that it was not ready for real business 
critical desktop work.  But I doubt the issue would arise.

>If you've got a VC++ project file, I'd like
>to have that.  Is the VPython build process for Win32 documented
>anywhere?

The source at:

http://virtualphoton.pc.cc.cmu.edu/projects/visual/developers.htm

has a VC++ project file.  Build was straightforward. Point to your own
Python20 includes and libs and shoot.

>This problem is getting in the way of my general goal:  creating
>interactive programs that allow the user to play with graphic
>representations of mathematical concepts.  I'm hoping to create some
>stuff that K-12 students will find compelling and will help develop
>useful intuition about math.

As you may know from my edu-sig ramblings I am doing something
similar with VPython, concentrating on geometric rather than number concepts.

I very much like your phrase "building intiution" - that is also what
I am trying to accomplish.  Unfortunately it is difficult to test for a
developed intuition - so the focus on that kind of approach is shortchanged, 
IMO.    


Art



From paul at fxtech.com  Sun Mar 11 10:10:48 2001
From: paul at fxtech.com (Paul Miller)
Date: Sun, 11 Mar 2001 09:10:48 -0600
Subject: any interest in type-scoped static constants?
Message-ID: <3AAB9578.4C8754E4@fxtech.com>

If I write an extension type that mirrors the functionality of a C++
class, and the C++ has static constants within its scope, I can do this
in C++:

	Point p = Point::ZERO;

Where ZERO is defined as:

	class Point
	{
		static Point ZERO = Point(0, 0);
	};

If I create an extension type for Point, I can't do this:

	p = Point.ZERO

But I'd like to.

I'm seriously considering working on a patch to add type-specific
constants to provide OO access to extension type constants, rather than
putting all constants in the module scope. I wonder if anyone has
started this, and if this is considered a good idea. Do I need to write
a PEP?

My solution would be to add a dictionary to the extension type for
constants, that the extension code can initialize, and then add a
scoping test for Type.ATTR to add a check in the Type's attribute
dictionary.


From emile at fenx.com  Fri Mar 23 08:34:38 2001
From: emile at fenx.com (Emile van Sebille)
Date: Fri, 23 Mar 2001 05:34:38 -0800
Subject: newbe   how to read in a string?
References: <3ABB4E70.54885C9F@noos.fr>
Message-ID: <99fjjg$14gp7$1@ID-11957.news.dfncis.de>

Here, it looks like you're dealing with comma separated fields.

strFields = a.split(",")
thirdField = strFields[2]

HTH
--

Emile van Sebille
emile at fenx.com

---------
"William Famy" <william.famy at noos.fr> wrote in message
news:3ABB4E70.54885C9F at noos.fr...
> Hi.
>
> I am looking how to read in a string.
>
> Ex
> a='2000/12/28,24.3,87.1,65.4'
>
> I whant to extract the 87.1. In C i would code by a
> scanf("%s,%f,%f,%f\n"),...
>
> How can i do this in python?
>
> Thanks for answer.
>
> --
>        \\\|///
>                            \\  - -  //
>                             (  @ @  )
> +-------------------------oOOo-(_)-oOOo-------------------------+
> |  Famy William  36 avenue des barattes  74000 Annecy  France   |
> |  email:william.famy at noos.fr william.famy at mail-enseignant.com  |
> +---------------------------------Oooo--------------------------+
>                            oooO   (   )
>                           (   )    ) /
>                            \ (    (_/
>                             \_)




From jarober at mail.com  Sun Mar 25 11:57:38 2001
From: jarober at mail.com (James A. Robertson)
Date: Sun, 25 Mar 2001 16:57:38 GMT
Subject: Collection interfaces
References: <3a90fa97.0@news1.mweb.co.za>   <MPG.1504d14563dce6499896af@news.earthlink.net> <97fjla128ir@news2.newsguy.com> <97h5s40vhc@news2.newsguy.com> <MPG.1506007591cf818b9896bc@news.earthlink.net> <slrn99opop.gr.amk@mira.erols.com> <MPG.1506265ebf6d934d9896be@news.earthlink.net> <97ibmc0k47@news1.newsguy.com> <MPG.1506cf3293b094b49896c1@news.earthlink.net> <97m0c7080c@news2.newsguy.com> <MPG.15085517debf4b649896ca@news.earthlink.net> <zb%n6.31033$o%6.586026@news1.rdc1.bc.home.com> <MPG.150b383cea226d409896d0@news.earthlink.net> <kNko6.6986$eq3.707121@news.uswest.net> <MPG.150c46db9cec283e9896d4@news.earthlink.net> <3AA2A898.450B659E@mail.com> <gRCt6.12224$Im6.1308560@newsread1.prod.itd.earthlink.net> <3AB755F8.67E84126@mail.com> <%a%t6.17245$227.1601569@newsread2.prod.itd.earthlink.net> <3AB8A730.4BE13D22@mail.com> <61bv6.8693$P4.676800@newsread1.prod.itd.earthlink.net> <B6E2B64C.D8E3%t*b*r*u*m*l*e*y@f*u*s*e.n*e*t> <o1jv6.9635$P4.789083@newsread1.prod.itd.earthlink.net>
Message-ID: <3ABF74ED.BC09EC8@mail.com>

As to the LessonBrowser, it's part of the non-commercial product.  When
you open the system up, there it is



topmind wrote:
> 
> >
> >
> http://www.cincom.com:80/scripts/smalltalk.exe/gettingstarted/index.asp?cont
> > ent=trysmalltalknc
> >
> 
> It seems to be wacked out right now.
> 
> > Pardon the word wrapping my news client does. There's no spaces in the
> > above.
> >
> > That should get you to a place where you can download the lesson browser
> > (thanks again to Mark Windholtz, great job!) and the tutorial (thanks to
> Jim
> > F. as well).
> >
> 
> Lesson browser download?
> 
> The last thing I need is yet more foriegn software on my PC making it crash
> more and confusing other DLL's.
> 
> Please, HTML.
> 
> >...site runs on
> > Microsoft ASP VSscript instead of Smalltalk?
> >
> > Naw. Jim isn't afraid of anything. At least, nothing on Usenet!
> >
> > Me either :)
> >
> > The rest of the Cincom site, and the old Smalltalk portion of the site,
> does
> > indeed not run on Smalltalk. We'll get it all moved over eventually :)
> >
> > Heck, my own business unit, which is very into VisualWave, understands
> that
> > we have to use JavaScript on our pages to drive the browsers. There's
> > nothing wrong with being multi-lingual.
> >
> > >
> > > Do Chevy salesmen drive Fords?
> >
> > Some do. I work for Cincom and use both our Smalltalk products, I actually
> > prefer ObjectStudio when I'm doing Windows specific work, but I also have
> > Squeak and Dolphin around and have experimented with both.
> >
> > > -tmind-
> >
> >

-- 
James A. Robertson
Product Manager (Smalltalk), Cincom
jarober at mail.com
<Talk Small and Carry a Big Class Library>


From donod at home.com  Tue Mar 13 22:20:37 2001
From: donod at home.com (Don O'Donnell)
Date: Wed, 14 Mar 2001 03:20:37 GMT
Subject: What's wrong with me/Python ?
References: <3AAE0652.D68F253E@daimlerchrysler.com> <3AAE51FE.AAB1234C@alcyone.com>
Message-ID: <3AAEE248.25344A66@home.com>


Erik Max Francis wrote:
> 
> Alexander Stirzel wrote:
> 
> > I just wrote a small script for killing some hanging netscape
> > processes,
> > the source is listed below.
> > Now if I run this script I get this error message:
> >
> > 8<----------- error message -----------------------------
> > Traceback (most recent call last):
> >   File "/usr/sbin/netscape-killer.py", line 11, in ?
> >     splitted = string.split(line)
> > NameError: There is no variable named 'string'
> > 8<----------- error message -----------------------------
> >
> > Now here's the script:
> >
> > 8<------------ script -----------------------------------
> > #!/usr/local/bin/python
> > import os
> > import time
>         ...
> 
> > What did I do wrong? And why does Python want a "variable named
> > 'string'" here?
> 
> Because you're referencing `string' and it's never seen the variable
> before.  You're missing an `import string' statement at the top.
> 

Or better yet, with version 2.0, use the string method:

	splitted = line.split()

and you don't have to remember to import the string module. 
It should be faster as well, since the string functions are 
now implemented as calls to the string methods.

/Don


From chris.gonnerman at usa.net  Thu Mar 15 21:47:30 2001
From: chris.gonnerman at usa.net (Chris Gonnerman)
Date: Thu, 15 Mar 2001 20:47:30 -0600
Subject: reading binary data
References: <002f01c0acc7$301202d0$04090a0a@upcast.com>
Message-ID: <010101c0adc3$77911a40$a100000a@local>

YES!  Please release it!

----- Original Message -----
From: "Ken Seehof" <kens at sightreader.com>
To: <python-list at python.org>
Sent: Wednesday, March 14, 2001 2:41 PM
Subject: Fw: reading binary data


>
> ----- Original Message -----
> From: "Ken Seehof" <kens at sightreader.com>
> To: "Francois Forest" <francois_forest at yahoo.fr>
> Sent: Wednesday, March 14, 2001 2:37 PM
> Subject: Re: reading binary data
>
>
> > Take a look at the "struct" module.  It has all you need to mess around
> with
> > binary data.  I have an interactive tool (called filecracker) for
reverse
> > engineering binary files using the struct module, but I haven't found
time
> > to release it yet.  If you want, I can email it to you (if enough people
> > respond to this, I'll be motivated to post it in parnassus).
> >
> > BTW, there's a gotcha, that everyone learns the hard way early on.  Make
> > sure you use the 'b' flag when opening binary files.
> >
> > f = open('ni.wav', 'rb')
> >
> > If you don't, the file will be truncated at the first ascii end-of-file
> > marker that happens to be in the binary data.
> >
> > ----- Original Message -----
> > From: "Francois Forest" <francois_forest at yahoo.fr>
> > Newsgroups: comp.lang.python
> > To: <python-list at python.org>
> > Sent: Wednesday, March 14, 2001 3:46 PM
> > Subject: reading binary data
> >
> >
> > > I would like to read and write xxx.wav files,
> > > this assume to alternatively read characters,  integers 16 bits and
> > > integers 32 bits.
> > > I found that I can read a binary file and get the result in a string,
> > > Is there a way to convert a string slice into an int16, int32, or
> > > floatting point ?
> > > (what is usually done in C through a cast).
> > >
> > > More generally, how can I browse through standard libraries ?
> > >
> > > --
> > > http://mail.python.org/mailman/listinfo/python-list
> > >
> >
>
>
> --
> http://mail.python.org/mailman/listinfo/python-list
>




From whisper at oz.net  Mon Mar  5 14:51:26 2001
From: whisper at oz.net (Dave LeBlanc)
Date: 5 Mar 2001 19:51:26 GMT
Subject: python cgi syntax highlighter
References: <Pine.BSF.4.32.0103032234260.88338-100000@localhost>
Message-ID: <980qnu$u36$0@216.39.170.247>

check out the "markup" command in htmlgen:
http://starship.python.net/crew/friedrich/HTMLgen/html/main.html

Dave LeBlanc

On Sat, 3 Mar 2001 22:39:21 -0500, "Gregory (Grisha) Trubetskoy"
<grisha at ispol.com> wrote:

>
>Hello -
>
>I remember coming across a script that displayed python code in HTML with
>color syntax highlighting - does anyone know where I can find one? My
>searches don't turn up anything.
>
>Thanks,
>
>Grisha
>



From kendall at _drakealumni_.net  Fri Mar 30 14:21:29 2001
From: kendall at _drakealumni_.net (Kendall)
Date: Fri, 30 Mar 2001 19:21:29 GMT
Subject: restricted execution
Message-ID: <Z05x6.1894$bY4.6478@www.newsranger.com>

I'm building a system which allows users to submit Python code to be executed in
the context of a table search facility.  The user sends data files and Python
code and the system will execute the code, making the datafiles available as
tables with fast search methods.  The result is a customizable calculator able
to be extended by the users.

My question is about restricting what the user can do with Python.  I've
experimented with code like this

dict = {}
for i in AllowedBuiltinList :
exec "dict['" + i + "'] = " + i

__builtins__ = dict
del dict

Where the AllowedBuiltinList is missing things like __import__, open, and eval.
If I paste this in every user defined module, have I effectively prevented all
access to the file system and network?  I'm not letting them import sys or os in
their module either.

I also need to prevent infinite loops.  I've fiddled with the thread state
object in the interpreter and added a tick_limit.  In the main interpreter loop
I decrement this and raise a system error if its exhausted.  It works, but I'd
like to know if anyone else has done something similar.

Thanks!




From mmoon at apexmail.com  Wed Mar  7 01:13:13 2001
From: mmoon at apexmail.com (Mark)
Date: Tue, 6 Mar 2001 22:13:13 -0800
Subject: The in opperator -- How do I use it?
Message-ID: <983945437.96921@atlas.uniserve.ca>

How do I use the 'in' operator?
Currently I have having some difficulty in getting it to work
Mark
mmoon at apexmail.com




From jafo at tummy.com  Fri Mar  2 12:08:20 2001
From: jafo at tummy.com (Sean Reifschneider)
Date: Fri, 2 Mar 2001 10:08:20 -0700
Subject: PythonJournal.com becomes Interactive, new issue today
In-Reply-To: <blFn6.13318$__6.3143457@typhoon.southeast.rr.com>; from tuttledon@hotmail.com on Fri, Mar 02, 2001 at 04:29:59AM +0000
References: <mailman.983500743.14313.python-list@python.org> <hzEn6.1869$GD.875814@typhoon.snet.net> <blFn6.13318$__6.3143457@typhoon.southeast.rr.com>
Message-ID: <20010302100820.A9184@tummy.com>

On Fri, Mar 02, 2001 at 04:29:59AM +0000, Don Tuttle wrote:
>> Ouch, to be perfectly honest, if the above website was my first exposure
>to
>> Python's web application capabilities I'd run for the hills.  The layout
>of
>> the site is AWFUL.
>
>Gotta agree.  Especially if we have to wait 2 -3 years for Issue III.

I'm sure the editor would welcome specific suggestions as to the design
of the new site.  You should be able to reach her at editor at pythonjournal.com
or cognite at zianet.com.

We've had problems getting contributions to the site since the first
issue.  The idea was to make it more interactive and iterative, so that
instead of trying to collect updates for a specific "publication date",
it could just accept things as they came in, hopefully urging people to
participate more.

The site as it is is meant to be a starting place for reaching those
goals.  Participate.

Sean
--
 It's in that place where I put that thing that time.
                 -- _Hackers_
Sean Reifschneider, Inimitably Superfluous <jafo at tummy.com>
tummy.com - Linux Consulting since 1995. Qmail, KRUD, Firewalls, Python



From jacobus at geo.co.za  Wed Mar 28 05:12:22 2001
From: jacobus at geo.co.za (Jacobus van der Merwe)
Date: Wed, 28 Mar 2001 12:12:22 +0200
Subject: Example using ConfigParser
Message-ID: <3AC1B905.CECB7BE8@geo.co.za>

Hi,

Is there a piece of code somewhere where I can see how one
uses the ConfigParser module??

I have looked at the Library Reference, but clearly I'm not
enough of a Pythonese yet.  :-(

Thanks,

Jacobus





From costas at meezon.com  Mon Mar 12 20:43:22 2001
From: costas at meezon.com (Costas Menico)
Date: Tue, 13 Mar 2001 01:43:22 GMT
Subject: This math scares me
References: <3aad4062.1290021@News.CIS.DFN.DE> <slrn9aqkth.aki.scarblac@pino.selwerd.nl>
Message-ID: <3aad6ffe.3692710@News.CIS.DFN.DE>

I appreciate the lessons in floating point math. Unfortunately I had
to deal with this stuff coding in assembler. 

However I am new to Python and didnt realize that it displays numbers
like this unless you format them yourself. Some other languages I use
just do the rounding for you. 

Fortunately I discovered I can use "print 5.01 + 5.54" and it does the
job. It has the extra intelligence built into it to autoformat

Thanks for the input.

Costas

>costas at springmail.com <costas at springmail.com> wrote in comp.lang.python:
>> Ok, I can see maybe division having problems. But why does addition of
>> the two numbers below:
>> 
>> 5.01+5.54
>> 
>> give me this?
>> 
>> 10.550000000000001
>> 
>> I hope its some setting that can take care of this. I am using Python
>> under Windows 98.
>
>It's worse than that. Try:
>
>>>> 5.01
>5.0099999999999998
>
>The number is stored in binary. It's not possible to represent everything in
>a finite number of binary digits (just like you can't represent 1/3 in a
>finite number of decimal digits - you don't protest either when you see
>0.3333333333333333, do you? It's just as wrong).
>
>This is not a Python problem, this is due to the way computers store
>floating point numbers.
>
>If you can avoid them, don't use floats. If you're working with money, count
>cents instead (and use longs if you have numbers over a few hundred million
>cents).
>
>-- 
>Remco Gerlich



From stephen_purcell at yahoo.com  Thu Mar 22 12:46:31 2001
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Thu, 22 Mar 2001 18:46:31 +0100
Subject: Transparent GIFs in PIL
In-Reply-To: <Pine.LNX.4.30.0103222013420.31302-100000@fep132.fep.ru>; from phd@phd.pp.ru on Thu, Mar 22, 2001 at 08:14:17PM +0300
References: <20010322180110.A18476@freedom.puma-ag.com> <Pine.LNX.4.30.0103222013420.31302-100000@fep132.fep.ru>
Message-ID: <20010322184631.B18920@freedom.puma-ag.com>

Oleg Broytmann wrote:
> Hi!
> 
>    Tried to send the patch to PIL authors?

Can't remember. Looks like I originally wrote the patch on 10th June 1999,
but that was [earlier] in my Newbie period, so I doubt I had enough common
sense to mail the patch to Fredrik. (Sorry all...)

-Steve

-- 
Steve Purcell, Pythangelist
Get testing at http://pyunit.sourceforge.net/
Any opinions expressed herein are my own and not necessarily those of Yahoo



From aleaxit at yahoo.com  Sun Mar 11 08:44:48 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sun, 11 Mar 2001 14:44:48 +0100
Subject: map - lambda - problem
References: <3AAB6078.EE50AE8D@aon.at>
Message-ID: <98fvee0204u@news2.newsguy.com>

"Gregor Lingl" <aon.912502367 at aon.at> wrote in message
news:3AAB6078.EE50AE8D at aon.at...
    [snip]
> In direct mode I got:
>
> >>> r = []
> >>> for i in range(len(n[0])-1,-1,-1):
>         r.append(map(lambda x: x[i],n))

Here, i is in the global namespace, and therefore available to
functions (including lambdas).

> If i put this into a function:
>
> def rotate(piece):
>     r = []
>     for i in range(len(piece[0])-1,-1,-1):
>         r.append(map(lambda x: x[i], piece))

Here, i is a local variable and thus (in Python 2.0) not visible
in nested scopes (including lambdas).

> Why, exactly, this behaviour. Is i now taken from some other namespace?

Local scopes don't nest (in Python 2.0).  i must come from the
global namespace.

> If so, which one and why is it initialized with 0?
> How can I work around this?

Use lambda x, i=i: x[i] to inject i into the lambda's local namespace
(or, switch to Python 2.1, which does let local scopes nest).


Alex





From dsh8290 at rit.edu  Wed Mar  7 15:31:42 2001
From: dsh8290 at rit.edu (D-Man)
Date: Wed, 7 Mar 2001 15:31:42 -0500
Subject: I come to praise .join, not to bury it...
In-Reply-To: <000001c0a64c$497df250$a9807018@cr706570a>; from mcfletch@home.com on Tue, Mar 06, 2001 at 09:46:53AM -0500
References: <kuofvfjb77.fsf@lasipalatsi.fi> <000001c0a64c$497df250$a9807018@cr706570a>
Message-ID: <20010307153141.B29445@harmony.cs.rit.edu>

On Tue, Mar 06, 2001 at 09:46:53AM -0500, Mike C. Fletcher wrote:
| Here's a professional aesthetic's take:
| 
| 	mystring.join( )
| 
| This is a perfectly normal and good-looking construct.  It seems pythonic
| and beautiful, object orientation is part of Python, and this works nicely.
| 
| 	"'".join( )
| 	";".join( )
| 	".".join( )
| 	"!".join( )
| 	",".join( )
| 
| All look like executable line noise.  That is, the literal string's syntax
| makes the use of dotted attributes on such a string look jarring and
| off-putting.  You almost expect some weird function to jump out of this
| syntax.  It looks very Ruby/Perl-esque.
| 

Even with my newfound understanding, provided by the Martellibot, I
agree with this.  Calling a method on an object is fine.  Calling it
on a literal looks weird.  Take for example something like this

three = 1.__add__( 2 )

;-)


Anyways, after seeing part of the discussion here I tried a string
method on a string literal in Java and it worked as well.

I still sort of disagree with string methods like 'replace'.  After
all, it doesn't really replace anything.  It creates a new object with
the proper differences.  Perhaps if 'replace!' were also added to
it might be clearer.

-D




From p.magwene at snet.net  Sun Mar 11 00:43:28 2001
From: p.magwene at snet.net (Paul Magwene)
Date: Sun, 11 Mar 2001 00:43:28 -0500
Subject: Problems writting a wrapper around linux' ssh command
References: <31rd89.lki.ln@127.0.0.1> <fdwq6.1906$54.1822@www.newsranger.com> <p9ae89.imr.ln@127.0.0.1> <MzCq6.2024$54.2158@www.newsranger.com>
Message-ID: <2gEq6.2725$GD.1629798@typhoon.snet.net>

How about a wrapper around rsync (see http://rsync.samba.org/ )?  That would
take care of your site updating/mirroring.

rsync already provides a means of using ssh as the authentication method (by
setting
the environment variable RSYNC_RSH=/path/to/ssh ).  It looks like there's
already a REXX implementation and a perl wrapper ( see
http://rsync.samba.org/rsync/resources.html ), so you might have something
to work from in the initial design phase.

--Paul

"Gregory Jorgensen" <gregj at pobox.com> wrote in message
news:MzCq6.2024$54.2158 at www.newsranger.com...
: I'm not an ssh expert; I just know that it can authenticate without
prompting
: for a password, but only when the server and the client are configured
correctly
: with RSA keys. There are several versions of ssh, too. For more info I
refer you
: to:
:
: http://www.ssh.com/
:
: http://www.openssh.com/
:
: http://www.oreilly.com/catalog/sshtdg/
:
: I've seen quite a few web site updating/mirroring tools, but I don't know
any
: that use ssh/scp offhand. You may want to poke around freshmeat.
:
: http://freshmeat.net/
:
: Good luck.
:
:
: In article <p9ae89.imr.ln at 127.0.0.1>, gradha at iname.com says...
: >Er... I am new to ssh, so I don't even know what other methods there
are...
: >So far, I want to write a wrapper on top of sitecopy (a program which
allows
: >updating remote web sites through ftp) using ssh and scp to upload files
: >and make directory modifications. If there's any other way of doing this,
: >any help is very welcome. The only restriction is that the program I
write
: >from commandline or in background, accepts password only once, and logs
: >everything to files, so I thought wrapping around ssh and scp would be
the
: >easiest way.
:




From ken at lt.com  Mon Mar  5 16:40:10 2001
From: ken at lt.com (Kenneth Loafman)
Date: Mon, 05 Mar 2001 21:40:10 +0000
Subject: Revenge - Careful there
References: <97fbbm$v1o$1@snowmane.julianhaight.com> <97goaj$jj4$1@snowmane.julianhaight.com> <3A9BFE1D.B94F6AA5@lt.com> <97j7kn$p21$1@snowmane.julianhaight.com>
Message-ID: <3AA407BA.AA9C49A7@lt.com>

Not a problem.  No one person (client system) would ever call spammer-X
more than once.  The goal would be to have thousands of separate clients
calling spammer-X one time, adding up to thousands of protest calls.  We
would need to make sure the protest message was generic,
non-threatening, and that the person running the client computer agreed
to the goals of the protest.  Once that's done, then we can protest all
we want and no laws are broken.

Of course, one bad apple could spoil the whole project, but that person
would be the one with multiple calls to spammer-X and would be the one
in trouble.

...Ken

SPAMFIGHTER wrote:
> 
> Careful, again.
>  ANI is not like Caller-ID where the data is transmitted as an
> ascii 1200 bps modem tone between the first and second rings, in
> most cases.  With ANI, the identification is known BEFORE the
> phone at the other end rings.  A short call won't stop the trace
> if it rings the far end as it is grabbed (I forget how many
> milliseconds) prior to the application of the ring burst.
> Actually, even a lot of the more recent Caller ID implementations
> now connect to the called end and send the data while that phone
> is still on-hook, so the data gets there before that phone rings,
> too.  It depends on the Central Office and the version of
> software they are using and the CID box the subscriber has.  So,
> ground starts have always been covered, and now the residential
> loop starts are becoming covered, too, as the updated software
> becomes more common in digital (not analog) central offices.
> 
> Tom
> 
> "Kenneth Loafman" <ken at lt.com> wrote in message
> news:3A9BFE1D.B94F6AA5 at lt.com...
> | Matt Dragonfly Drury wrote:
> | >
> | > Many toll-free numbers use ANI for tracking inbound numbers,
> not Caller ID,
> | > which isn't affected by the star-code to disable the
> transmission of one's
> | > number. The argument being that since the company is
> effectively paying for
> | > a collect call, they've the right to know who is calling.
> | >
> | > For that matter, I recall a case a while back where someone
> phone-spammed a
> | > televangelist he didn't like, and was held liable for the
> costs of the
> | > hundreds of short calls his modem demon-dialed for him. Not
> the most
> | > discreet method, obviously.
> | >
> | > That being said, were such a tactic to be distributed - a
> telephone-like
> | > "DDoS attack" - I'm sure it'd be harder for accused spammers
> to pin it down
> | > to one person.
> | >
> | > I wonder what kind of tracing and identification is in place
> for
> | > Internet-telephony calls? I'm sure it depends on the
> provider.
> | >
> | > I've mused in the past on keeping a list of toll-free numbers
> handy whose
> | > companies I find distasteful, and giving them a piece of my
> mind on their
> | > dime when I happen to be near a phone some distance from my
> residence.
> |
> | All this discussion leads me to another interesting idea.  If I
> | *mistakenly* call a spammer and hang up, then no charge can be
> made.
> | What we need to do is apply that 1000's of times, all across
> the world.
> | A lot of us have multiple modems that we no longer use that
> could be put
> | to good use.  We could set up a Seti-like system where a work
> block is a
> | list of spammer's 800-numbers.  You would run a dialing program
> to dial
> | that number *exactly one* time, then step to the next number.
> Lots of
> | individual calls.  No single point of harassment, and the work
> block
> | would only be executed once per client system so that no single
> number
> | ever called more than once.
> |
> | Of course, we'd need to run the server somewhere outside the
> US, but
> | that's not hard to set up.  This could be fun.
> |
> | Think about it.  The ultimate revenge, especially if we could
> | incorportate telemarketers phone numbers into the mix as well.
> Those
> | would be limited to calls between 6 and 8pm to catch them at
> supper.
> | Totally distributed, just like they do to us.
> |
> | OK, some days I'm totally evil, but I'm getting tired of spam
> and
> | telemarketing, actually marketing in general.  Do unto others
> as they've
> | done unto you.
> |
> | ...Ken



From sdm7g at virginia.edu  Wed Mar 14 11:53:36 2001
From: sdm7g at virginia.edu (Steven D. Majewski)
Date: Wed, 14 Mar 2001 11:53:36 -0500 (EST)
Subject: main in mac?
In-Reply-To: <YKMr6.256$4d3.54207@newsb.telia.net>
Message-ID: <Pine.NXT.4.21.0103141145290.229-100000@localhost.virginia.edu>


On Wed, 14 Mar 2001, Fredrik Lundh wrote:

> Doug Hellmann wrote:
> > It's not that you're running on a Mac, it's that you're using an IDE.
> 
> footnote: not all IDE's have this problem, of course.
> 

I don't think I'ld define it as a problem. 

Any IDE needs to be able to do the equivalent.

The only questions are how you indicate which action you want. 
I can think of several alternates, but I'm not sure that any
other is more clear: 'Source' vs. 'Import' might mean something
to a unix shell user, but I wouldn't bet on it being clear to
a Mac user! 

-- Steve Majewski




From tjreedy at udel.edu  Mon Mar  5 11:58:30 2001
From: tjreedy at udel.edu (Terry Reedy)
Date: Mon, 5 Mar 2001 11:58:30 -0500
Subject: Questions from a beginner
References: <1103_983792888@breakdown>
Message-ID: <980gm5$gau$1@news.udel.edu>

"Alison Myers" <squirmy at freeuk.com> wrote in message
news:1103_983792888 at breakdown...
> Hi. I'm a beginning programmer at a fairly late age and have been looking
for a language to learn.

Welcome. Never too late.

> I have decided against starting with C or C++ because I think they will
be too complicated and they seem to be only really needed for writing
device drivers and operating systems, which is not what I wat to do.
> Is python a good language to learn first?

Most of us here think so.  Especially handy for learning is 1) the
avoidance of busywork such as declarations and unnecessary fences (such as
C braces) and 2) the interactive mode.  Whenever you wonder 'what happens
if ...', type in a few lines and find out.

> Will my knowledge help me with other languages later?

The basic ideas of sequence, alternation, looping, and bundling are common
to all.

> I will be writing for win98 and Linux and I know Python is available on
both of these, but would the code be transportable,

Unless you use a facility specific to a particular OS, yes.

> especially if I write things using tcl (this is probably a long way off
yet, but I'd like to know).

I'll let a tcler answer that (but I presume so).

> I have links to online tutorials, but where would I find good source
code?

A. Read some of the library modules.  B. Check out Parnassus site (URL in
another post today).

> Do I need to buy any books to help me and if so which do you recommend?
> Is there a list of programming projects anywhere or would I get these
from a textbook?

Pick something that motivates *you*.

> I probably need to brush up on my maths, so what specific topics should I
look at?
> TIA
>
> Alison


Terry J. Reedy




From shredwheat at mediaone.net  Mon Mar 26 22:31:55 2001
From: shredwheat at mediaone.net (Pete Shinners)
Date: Tue, 27 Mar 2001 03:31:55 GMT
Subject: compiling questions, a la lisp
Message-ID: <LQTv6.92647$Ch.13131089@typhoon.we.rr.com>

I've become quite comforable with python, and can understand why
simple compiling of python bytecode doesn't really create much
speedup.

I've been talking to some folks on the net (not sure how savvy
they are) that raised some questions i cannot answer.

these folks say that the LISP language (which i know nothing about),
can be compiled natively and provide a substantial speedup. these
people have said that after looking at python, it looks similar to
lisp, in that everything must be looked up in dictionarys before
use, etc. they wonder why compiling python natively doesn't do much
while compiling lisp does.

personally, i assume this is because parts of lisp are more static,
so less things need to be checked at runtime. is this the reason?
what is the biggest difference between lisp and python which makes
lisp more agreeable with native compiling?

second, i've heard other people mention that a python to lisp
converter would not be difficult, and then that code could take
advantage of the native compilers, and this get an increase in
speed. this seems a little odd to me, the same things that are
holding python code running native are not going to 'disappear'
when running under a lisp environment... are they? i assume
the answer to this question is the same as the previous, but
is there anything that would keep this from being beneficial?
python to lisp, and then to native code?






From JamesL at Lugoj.Com  Thu Mar  8 01:10:27 2001
From: JamesL at Lugoj.Com (James Logajan)
Date: Wed, 07 Mar 2001 22:10:27 -0800
Subject: "!=" is bad form. Re: sorry....never mind
References: <DZkp6.18811$pF4.3015652@news1.telusplanet.net> <fdlp6.16982$go.2157953@news0.telusplanet.net> <3AA6E308.5B63ECC6@cosc.canterbury.ac.nz>
Message-ID: <3AA72253.FB021815@Lugoj.Com>

Greg Ewing wrote:
> 
> Victor Louie wrote:
> >
> > <> is obsolescent.
> 
> Really? I hope it's not taken away, because that will
> break ALL of my code! I hate != (looks too much like
> C), so I always use <>. Nobody ever told me I was doing
> anything wrong!

Actually, for Python, "!=" IS totally bogus and should be considered the
unusual variant. Unless the logical negation operation has been changed from
"not" to "!" (shucks, version 1.5.2 still does all I need) then "!=" is not
suggestive of anything. In "C" the "!=" was suggestive of:

!(expression1 == expression2)

But Python doesn't use "!" for "not". The operations "<", ">", "<=", ">=",
"==", "<>" form a nice consistent and predictable set. The only objection
that has been made against "<>" is that it implies ordering when the
entities being compared may not have any ordering. Even granting this
dubious objection does not excuse the introduction of a form that is no
where else implied, hinted at, or suggested in any way by any other part of
the language. Guido made a mistake here and shows that even the brightest
can stumble once in a while.

Shucks, if one is going to borrow "!=" from "C", why stop there? Why not
borrow the keywords and concepts such as "register", "auto", "extern",
pointers, malloc, free, etc.


From cavvinas at terra.com.br  Fri Mar 23 13:03:14 2001
From: cavvinas at terra.com.br (=?iso-8859-1?Q?Estev=E3o_Cavinato?=)
Date: Fri, 23 Mar 2001 15:03:14 -0300
Subject: Digest Number 1936
References: <985212900.441.99052.l6@yahoogroups.com>
Message-ID: <000001c0b3cc$337cefa0$a438b0c8@testei>

Unsubscribe!!!
----- Original Message -----
From: <python-list at yahoogroups.com>
To: <python-list at yahoogroups.com>
Sent: Wednesday, March 21, 2001 7:15 PM
Subject: Digest Number 1936


> ------------------------ Yahoo! Groups Sponsor ---------------------~-~>
> We give away $70,000 a month! Come to iWin.com for
> your chance to win!
> http://us.click.yahoo.com/olMXHC/BJVCAA/4ihDAA/stAVlB/TM
> ---------------------------------------------------------------------_->
>
> There are 25 messages in this issue.
>
> Topics in this digest:
>
>       1. Re: learning python...
>            From: Simon Kesenci <tomega at earthlink.net>
>       2. Re: Private Business Relationship (Urgent)
>            From: "Terry Reedy" <tjreedy at udel.edu>
>       3. Substitute for join()
>            From: "Daniel Klein" <DanielK at jBASE.com>
>       4. Re: learning python...
>            From: "Bubba Brains" <chocolate at toblerone.yum>
>       5. Re: learning python...
>            From: Joshua Marshall <jmarshal at mathworks.com>
>       6. Re: learning python...
>            From: Timothy Grant <tjg at exceptionalminds.com>
>       7. RE: Substitute for join()
>            From: "Sean 'Shaleh' Perry" <shaleh at valinux.com>
>       8. Re: Substitute for join()
>            From: matt at mondoinfo.com (Matthew Dixon Cowles)
>       9. Re: Substitute for join()
>            From: Siggy Brentrup <bsb at winnegan.de>
>      10. Re: Substitute for join()
>            From: Steve Purcell <stephen_purcell at yahoo.com>
>      11. RE: Substitute for join()
>            From: "Sean 'Shaleh' Perry" <shaleh at valinux.com>
>      12. Re: learning python...
>            From: scarblac at pino.selwerd.nl (Remco Gerlich)
>      13. Re: Substitute for join()
>            From: "Daniel Klein" <DanielK at jBASE.com>
>      14. Useless spam reporting (was Re: Private Business Relationship
(Urgent))
>            From: aahz at panix.com (Aahz Maruch)
>      15. Re: learning python...
>            From: aahz at panix.com (Aahz Maruch)
>      16. Re: learning python...
>            From: Benoit Dupire <bdupire at seatech.fau.edu>
>      17. Re: learning python...
>            From: "Benjamin.Altman" <benjamin.altman at noaa.gov>
>      18. Re: learning python...
>            From: "Benjamin.Altman" <benjamin.altman at noaa.gov>
>      19. Re: New PEP: The directive statement
>            From: hzhu at users.sourceforge.net (Huaiyu Zhu)
>      20. Dynamically reloading extension modules
>            From: "Ken Seehof" <kens at sightreader.com>
>      21. Re: learning python...
>            From: Benoit Dupire <bdupire at seatech.fau.edu>
>      22. RE: What IDE are you'all using for Python 2.1 for Windows?
>            From: digitig at cix.co.uk (Tim Rowe)
>      23. Re: The most important modules
>            From: digitig at cix.co.uk (Tim Rowe)
>      24. Re: python suitability for large critical run forever apps
>            From: digitig at cix.co.uk (Tim Rowe)
>      25. I'm sure there's a good reason!
>            From: digitig at cix.co.uk (Tim Rowe)
>
>
> ________________________________________________________________________
> ________________________________________________________________________
>
> Message: 1
>    Date: Wed, 21 Mar 2001 19:20:25 GMT
>    From: Simon Kesenci <tomega at earthlink.net>
> Subject: Re: learning python...
>
> If you know programming, forget _Learning_ _Python_.  I tried
_Programming_
> _Python_ last year, but it covered version 1.3.  A good book if they
> updated it since then.  As a programmer, you should find the online
> tutorial easy and helpful <http://python.org/doc/current/tut/tut.html>.
>
> --S.
>
> Benjamin.Altman wrote:
>
> > Hello.  From the perspective of someone who already has programming
> > experience, would anyone know if O'reillys "Learning Python" book would
> > be better than say Prentice Hall's "Core Python Programming"?  Or would
> > it be better to go straight into something like "Programming Python"?
> >
> > Thanks,
> > Ben
> >
> >
>
> --
> http://mail.python.org/mailman/listinfo/python-list
>
>
> ________________________________________________________________________
> ________________________________________________________________________
>
> Message: 2
>    Date: Wed, 21 Mar 2001 14:26:19 -0500
>    From: "Terry Reedy" <tjreedy at udel.edu>
> Subject: Re: Private Business Relationship (Urgent)
>
> Crimanal scam reported to abuse at baymountain.net for appropriate action.
>
>
> --
> http://mail.python.org/mailman/listinfo/python-list
>
>
> ________________________________________________________________________
> ________________________________________________________________________
>
> Message: 3
>    Date: Wed, 21 Mar 2001 11:35:53 -0800
>    From: "Daniel Klein" <DanielK at jBASE.com>
> Subject: Substitute for join()
>
> I've got to make a script (written in version 2.0) to work on a machine
> running 1.5.2. The only incompatible line is one that uses the
> "".join(mylist) method. To 'get out of the woods' quickly, I substituted
it
> with
>
> s = ''
> for element in mylist: s = s + element
>
> Is this the most efficient way to code this?
>
> Thanks,
> Daniel Klein
>
>
> --
> http://mail.python.org/mailman/listinfo/python-list
>
>
> ________________________________________________________________________
> ________________________________________________________________________
>
> Message: 4
>    Date: Wed, 21 Mar 2001 14:35:02 -0500
>    From: "Bubba Brains" <chocolate at toblerone.yum>
> Subject: Re: learning python...
>
> Forget Learning Python?  I think it's the best reference/tutorial around.
I
> use python once every couple months, and sometimes forget the simple
python
> syntax.  The Learning Python book is great for getting up to speed fast,
and
> to look up common stuff.  The other OReilly book (Programming Python) is
> absolutely annoying to use.  It's not even a good reference book (good
luck
> finding the info you want quickly).  Don't know about the Prentice book.
>
> M
> http://www.mp3.com/maddogskullcap
>
> "Simon Kesenci" <tomega at earthlink.net> wrote in message
> news:Z97u6.16733$Im6.1761313 at newsread1.prod.itd.earthlink.net...
> > If you know programming, forget _Learning_ _Python_.  I tried
> _Programming_
> > _Python_ last year, but it covered version 1.3.  A good book if they
> > updated it since then.  As a programmer, you should find the online
> > tutorial easy and helpful <http://python.org/doc/current/tut/tut.html>.
> >
> > --S.
> >
> > Benjamin.Altman wrote:
> >
> > > Hello.  From the perspective of someone who already has programming
> > > experience, would anyone know if O'reillys "Learning Python" book
would
> > > be better than say Prentice Hall's "Core Python Programming"?  Or
would
> > > it be better to go straight into something like "Programming Python"?
> > >
> > > Thanks,
> > > Ben
> > >
> > >
> >
>
>
>
>
> -----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
> http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
> -----==  Over 80,000 Newsgroups - 16 Different Servers! =-----
> --
> http://mail.python.org/mailman/listinfo/python-list
>
>
> ________________________________________________________________________
> ________________________________________________________________________
>
> Message: 5
>    Date: 21 Mar 2001 19:46:46 GMT
>    From: Joshua Marshall <jmarshal at mathworks.com>
> Subject: Re: learning python...
>
> Bubba Brains <chocolate at toblerone.yum> wrote:
> > Forget Learning Python?  I think it's the best reference/tutorial
around.  I
> > use python once every couple months, and sometimes forget the simple
python
> > syntax.  The Learning Python book is great for getting up to speed fast,
and
> > to look up common stuff.  The other OReilly book (Programming Python) is
> > absolutely annoying to use.  It's not even a good reference book (good
luck
> > finding the info you want quickly).  Don't know about the Prentice book.
>
> I agree about Programming Python.  It _is_ a great book for
> programmers to use to learn Python, but it can be tedious to try look
> up something specific.
> --
> http://mail.python.org/mailman/listinfo/python-list
>
>
> ________________________________________________________________________
> ________________________________________________________________________
>
> Message: 6
>    Date: Wed, 21 Mar 2001 12:05:03 -0800
>    From: Timothy Grant <tjg at exceptionalminds.com>
> Subject: Re: learning python...
>
> > additional advanced concepts. From there, I would recommend "Programming
> > Python 2nd Ed." as well as the indispensible "Python Essential
Reference"
> > (altho I think a new edition is in the works so you might want to wait
for
> > that).
>
> I'm am searching for a review of PP2nd. Is it a significantly
> different book from PP1st?
>
> Also. I'd love to see a new version of Beazley's book. I know
> that Ivan isn't reading this list at the moment, but I found
> the colophon of PER to be a fascinating and enjoyable bit of
> reading. Hope he gets to do something in the update too.
>
>
> --
> Stand Fast,
>     tjg.
>
> Timothy Grant                         tjg at exceptionalminds.com
> Red Hat Certified Engineer            www.exceptionalminds.com
> Avalon Technology Group, Inc.         <><       (503) 246-3630
> >>>>>>>>>>>>>Linux, because rebooting is *NOT* normal<<<<<<<<<
> >>>>This machine was last rebooted:  64 days     1 hours ago<<
>
> --
> http://mail.python.org/mailman/listinfo/python-list
>
>
> ________________________________________________________________________
> ________________________________________________________________________
>
> Message: 7
>    Date: Wed, 21 Mar 2001 12:09:05 -0800 (PST)
>    From: "Sean 'Shaleh' Perry" <shaleh at valinux.com>
> Subject: RE: Substitute for join()
>
>
> On 21-Mar-2001 Daniel Klein wrote:
> > I've got to make a script (written in version 2.0) to work on a machine
> > running 1.5.2. The only incompatible line is one that uses the
> > "".join(mylist) method. To 'get out of the woods' quickly, I substituted
it
> > with
> >
> > s = ''
> > for element in mylist: s = s + element
> >
> > Is this the most efficient way to code this?
> >
>
> or use the syntax that works in both:
>
> import string # or from string import join
>
> string.join(mylist, "")
>
> --
> http://mail.python.org/mailman/listinfo/python-list
>
>
> ________________________________________________________________________
> ________________________________________________________________________
>
> Message: 8
>    Date: Wed, 21 Mar 2001 19:52:45 GMT
>    From: matt at mondoinfo.com (Matthew Dixon Cowles)
> Subject: Re: Substitute for join()
>
> On Wed, 21 Mar 2001 11:35:53 -0800, Daniel Klein <DanielK at jBASE.com>
> wrote:
>
> >I've got to make a script (written in version 2.0) to work on a machine
> >running 1.5.2. The only incompatible line is one that uses the
> >"".join(mylist) method.
>
> >>> import string
> >>> string.join(mylist,"")
>
> Will do what you want.
>
> Regards,
> Matt
> --
> http://mail.python.org/mailman/listinfo/python-list
>
>
> ________________________________________________________________________
> ________________________________________________________________________
>
> Message: 9
>    Date: 21 Mar 2001 21:08:18 +0100
>    From: Siggy Brentrup <bsb at winnegan.de>
> Subject: Re: Substitute for join()
>
> "Daniel Klein" <DanielK at jBASE.com> writes:
>
> > I've got to make a script (written in version 2.0) to work on a machine
> > running 1.5.2. The only incompatible line is one that uses the
> > "".join(mylist) method. To 'get out of the woods' quickly, I substituted
it
> > with
> >
> > s = ''
> > for element in mylist: s = s + element
> >
> > Is this the most efficient way to code this?
>
> Nope, use
>
> import string
> string.join(mylist,'')
>
> HIH
>   Siggy
> --
> Siggy Brentrup - bsb at winnegan.de - http://www.winnegan.de/
> ****** ceterum censeo javascriptum esse restrictam *******
>
>
> --
> http://mail.python.org/mailman/listinfo/python-list
>
>
> ________________________________________________________________________
> ________________________________________________________________________
>
> Message: 10
>    Date: Wed, 21 Mar 2001 21:12:46 +0100
>    From: Steve Purcell <stephen_purcell at yahoo.com>
> Subject: Re: Substitute for join()
>
> Daniel Klein wrote:
> > I've got to make a script (written in version 2.0) to work on a machine
> > running 1.5.2. The only incompatible line is one that uses the
> > "".join(mylist) method. To 'get out of the woods' quickly, I substituted
it
> > with
> >
> > s = ''
> > for element in mylist: s = s + element
> >
> > Is this the most efficient way to code this?
>
> No. Instead, try:
>
>     import string
>     s = string.join(mylist, '')
>
> -Steve
>
> --
> Steve Purcell, Pythangelist
> Get testing at http://pyunit.sourceforge.net/
> Any opinions expressed herein are my own and not necessarily those of
Yahoo
>
> --
> http://mail.python.org/mailman/listinfo/python-list
>
>
> ________________________________________________________________________
> ________________________________________________________________________
>
> Message: 11
>    Date: Wed, 21 Mar 2001 12:09:05 -0800 (PST)
>    From: "Sean 'Shaleh' Perry" <shaleh at valinux.com>
> Subject: RE: Substitute for join()
>
>
> On 21-Mar-2001 Daniel Klein wrote:
> > I've got to make a script (written in version 2.0) to work on a machine
> > running 1.5.2. The only incompatible line is one that uses the
> > "".join(mylist) method. To 'get out of the woods' quickly, I substituted
it
> > with
> >
> > s = ''
> > for element in mylist: s = s + element
> >
> > Is this the most efficient way to code this?
> >
>
> or use the syntax that works in both:
>
> import string # or from string import join
>
> string.join(mylist, "")
> --
> http://mail.python.org/mailman/listinfo/python-list
>
>
> ________________________________________________________________________
> ________________________________________________________________________
>
> Message: 12
>    Date: 21 Mar 2001 20:29:12 GMT
>    From: scarblac at pino.selwerd.nl (Remco Gerlich)
> Subject: Re: learning python...
>
> Benjamin.Altman <ben at .> wrote in comp.lang.python:
> > Hello.  From the perspective of someone who already has programming
> > experience, would anyone know if O'reillys "Learning Python" book would
> > be better than say Prentice Hall's "Core Python Programming"?  Or would
> > it be better to go straight into something like "Programming Python"?
>
> I thought Learning Python was a very good introduction, I was a CS student
> with experience in several languages. There are a few things to you need
to
> "get" once about Python, and the book explains them well. It doesn't go
very
> deep, but then, there's not that much to know, Python is simplicity.
>
> Programming Python I found a bit chaotic, never really used it much.
>
> Don't know Core Python Programming.
>
> --
> Remco Gerlich
> --
> http://mail.python.org/mailman/listinfo/python-list
>
>
> ________________________________________________________________________
> ________________________________________________________________________
>
> Message: 13
>    Date: Wed, 21 Mar 2001 12:30:22 -0800
>    From: "Daniel Klein" <DanielK at jBASE.com>
> Subject: Re: Substitute for join()
>
> Thanks everyone for your public and private answers. This is the greatest
> bunch of folks I have ever had the pleasure of being a part of!
>
> I have indeed recoded to use string.join().
>
> Again thanks,
> Dan
>
> "Daniel Klein" <DanielK at jBASE.com> wrote in message
> news:Mo7u6.58$Hp.2896 at typhoon.aracnet.com...
> > I've got to make a script (written in version 2.0) to work on a machine
> > running 1.5.2. The only incompatible line is one that uses the
> > "".join(mylist) method. To 'get out of the woods' quickly, I substituted
> it
> > with
> >
> > s = ''
> > for element in mylist: s = s + element
> >
> > Is this the most efficient way to code this?
> >
> > Thanks,
> > Daniel Klein
> >
> >
>
>
> --
> http://mail.python.org/mailman/listinfo/python-list
>
>
> ________________________________________________________________________
> ________________________________________________________________________
>
> Message: 14
>    Date: 21 Mar 2001 12:47:59 -0800
>    From: aahz at panix.com (Aahz Maruch)
> Subject: Useless spam reporting (was Re: Private Business Relationship
(Urgent))
>
> [posted & e-mailed]
>
> In article <99av6p$b87$1 at news.udel.edu>, Terry Reedy <tjreedy at udel.edu>
wrote:
> >
> >Crimanal scam reported to abuse at baymountain.net for appropriate action.
>
> That's useless.  baymountain.net is the netnews injection point for the
> news/mail gateway of python-list at python.org; the spam got sent to the
> e-mail list, not the newsgroup.  There's unfortunately no way to track
> that spammer down without access to the python.org mailserver.
> --
>                       --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)
>
> Androgynous poly kinky vanilla queer het Pythonista
http://www.rahul.net/aahz/
> Hugs and backrubs -- I break Rule 6
>
> "I won't accept a model of the universe in which free will, omniscient
> gods, and atheism are simultaneously true."  -- M
> --
> http://mail.python.org/mailman/listinfo/python-list
>
>
> ________________________________________________________________________
> ________________________________________________________________________
>
> Message: 15
>    Date: 21 Mar 2001 12:49:32 -0800
>    From: aahz at panix.com (Aahz Maruch)
> Subject: Re: learning python...
>
> In article <3ab90559_6 at goliath.newsfeeds.com>,
> Bubba Brains <chocolate at toblerone.yum> wrote:
> >
> >Forget Learning Python?  I think it's the best reference/tutorial around.
I
> >use python once every couple months, and sometimes forget the simple
python
> >syntax.  The Learning Python book is great for getting up to speed fast,
and
> >to look up common stuff.  The other OReilly book (Programming Python) is
> >absolutely annoying to use.  It's not even a good reference book (good
luck
> >finding the info you want quickly).  Don't know about the Prentice book.
>
> If you look at http://www.oreilly.com/catalog/python2/ you'll see that
> Programming Python just got a second edition; I've heard that there were
> a lot of improvements.
> --
>                       --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)
>
> Androgynous poly kinky vanilla queer het Pythonista
http://www.rahul.net/aahz/
> Hugs and backrubs -- I break Rule 6
>
> "I won't accept a model of the universe in which free will, omniscient
> gods, and atheism are simultaneously true."  -- M
> --
> http://mail.python.org/mailman/listinfo/python-list
>
>
> ________________________________________________________________________
> ________________________________________________________________________
>
> Message: 16
>    Date: Wed, 21 Mar 2001 15:33:10 -0500
>    From: Benoit Dupire <bdupire at seatech.fau.edu>
> Subject: Re: learning python...
>
>
>
> Simon Kesenci wrote:
>
> > If you know programming, forget _Learning_ _Python_.  I tried
_Programming_
> > _Python_ last year, but it covered version 1.3.  A good book if they
> > updated it since then.  As a programmer, you should find the online
> > tutorial easy and helpful <http://python.org/doc/current/tut/tut.html>.
>
> Programming Python was updated and now covers Python 2.0
> I am reading it, actually, and, till now, I find it great.. (but I am only
page
> 35! ;o) ).
>
>
> This edition is intended to be the 'sequel' to " Learning Python"... so
yes,
> you can try to read the Python tutorial and this book...
>
>
> --
> http://mail.python.org/mailman/listinfo/python-list
>
>
> ________________________________________________________________________
> ________________________________________________________________________
>
> Message: 17
>    Date: Wed, 21 Mar 2001 16:13:59 -0500
>    From: "Benjamin.Altman" <benjamin.altman at noaa.gov>
> Subject: Re: learning python...
>
> Do you know how it compares to the first edition in terms of presentation
and
> indexing?  I have heard from a few people that it is difficult to look up
things in
> the 1st edition.  The only review on amazon of the 2nd ed. claims it is a
complete
> rewrite despite his admitance that he has never read the book...
>
> Thanks,
> Ben
>
> Benoit Dupire wrote:
>
> > Simon Kesenci wrote:
> >
> > > If you know programming, forget _Learning_ _Python_.  I tried
_Programming_
> > > _Python_ last year, but it covered version 1.3.  A good book if they
> > > updated it since then.  As a programmer, you should find the online
> > > tutorial easy and helpful
<http://python.org/doc/current/tut/tut.html>.
> >
> > Programming Python was updated and now covers Python 2.0
> > I am reading it, actually, and, till now, I find it great.. (but I am
only page
> > 35! ;o) ).
> >
> > This edition is intended to be the 'sequel' to " Learning Python"... so
yes,
> > you can try to read the Python tutorial and this book...
>
> --
> http://mail.python.org/mailman/listinfo/python-list
>
>
> ________________________________________________________________________
> ________________________________________________________________________
>
> Message: 18
>    Date: Wed, 21 Mar 2001 16:07:54 -0500
>    From: "Benjamin.Altman" <benjamin.altman at noaa.gov>
> Subject: Re: learning python...
>
> It would be good to know if the new edition really is all that different
(other than
> the topics added) to the 1st ed.  The only review on Amazon says he never
read it...
>
> Ben
>
> Aahz Maruch wrote:
>
> > In article <3ab90559_6 at goliath.newsfeeds.com>,
> > Bubba Brains <chocolate at toblerone.yum> wrote:
> > >
> > >Forget Learning Python?  I think it's the best reference/tutorial
around.  I
> > >use python once every couple months, and sometimes forget the simple
python
> > >syntax.  The Learning Python book is great for getting up to speed
fast, and
> > >to look up common stuff.  The other OReilly book (Programming Python)
is
> > >absolutely annoying to use.  It's not even a good reference book (good
luck
> > >finding the info you want quickly).  Don't know about the Prentice
book.
> >
> > If you look at http://www.oreilly.com/catalog/python2/ you'll see that
> > Programming Python just got a second edition; I've heard that there were
> > a lot of improvements.
> > --
> >                       --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)
> >
> > Androgynous poly kinky vanilla queer het Pythonista
http://www.rahul.net/aahz/
> > Hugs and backrubs -- I break Rule 6
> >
> > "I won't accept a model of the universe in which free will, omniscient
> > gods, and atheism are simultaneously true."  -- M
>
> --
> http://mail.python.org/mailman/listinfo/python-list
>
>
> ________________________________________________________________________
> ________________________________________________________________________
>
> Message: 19
>    Date: Wed, 21 Mar 2001 21:37:16 GMT
>    From: hzhu at users.sourceforge.net (Huaiyu Zhu)
> Subject: Re: New PEP: The directive statement
>
> With the __future__ PEP, I envision there could also be things like
>
> from __future__ import nested_scope
> from __experimentatal__ import stackless
> from __optional__ import static_type
> from __mandatory__ import string_methods
>
> New feature would migrate from __experimental__, to __future__, then to
> either __optional__ or __mandatory__.  Things in __experimental__ are not
> guaranteed to be there in later versions.  Things in __future__ will
> definitely be there in a later version.  The __mandatory__ could be used
to
> make the intention about version requirement more explicit than
__version__
> would allow. (Someone asked about whether __future__ should be phased out
in
> later code.) The __optional__ could be used for permament optional
features
> that affect syntax.
>
> So with this directive PEP, the parallel would be
>
> directive transitional nested_scope
> directive experimentatal stackless
> directive optional static_type
> directive mandatory string_methods
>
> Is this correspondence correct?  (You may not agree on the need for other
> directives, but I want to know if the keyword translations are correct.)
>
> If so, I see two problems.
> 1. It introduces a new syntax while an existing syntax could be reused
>    without much cost.
> 2. There is no place for version and documentation info.
>
> So far as I can see, the impetus for directive keyword is that import is a
> run time action, while the __future__ is a compile time declaration.  But
is
> it so far a stretch?  All Python __magic__ words go beyond normal meaning.
> Defining __setattr__ will affect more than this method alone.  Having
> __init__.py will cause other modules to be imported.  And so on.
>
> If we look more broadly, in *nix, devices, processes, pipe and others are
> also represented as files.  Would it better to assign special syntax to
all
> of them?
>
> IMHO, reusing existing syntax is closer to Python than Perl, which
> introduces new syntax for many things that are only slightly different.
>
> Huaiyu
> --
> http://mail.python.org/mailman/listinfo/python-list
>
>
> ________________________________________________________________________
> ________________________________________________________________________
>
> Message: 20
>    Date: Wed, 21 Mar 2001 14:01:44 -0800
>    From: "Ken Seehof" <kens at sightreader.com>
> Subject: Dynamically reloading extension modules
>
> Has anyone found a technique for dynamically unloading and
> reloading an extension module?  I want to be able to modify,
> build, and reload an extension module from within my program.
>
> Unfortunately python apparently doesn't use FreeLibrary when
> a module object is released.  Maybe I can do some kind of
> hack to call FreeLibrary explicitly.  Is there a way to do this that
> is portable?
>
> ----------------------------------------------------
> Copyright (c) 2001 by Ken Seehof
> This document may not be distributed, copied,
> duplicated, or replicated in any form without
> express written permission by Ken Seehof.
> Express permission is hereby granted.
> kens at neuralintegrator.com
> ----------------------------------------------------
>
>
>
> --
> http://mail.python.org/mailman/listinfo/python-list
>
>
> ________________________________________________________________________
> ________________________________________________________________________
>
> Message: 21
>    Date: Wed, 21 Mar 2001 16:55:06 -0500
>    From: Benoit Dupire <bdupire at seatech.fau.edu>
> Subject: Re: learning python...
>
>
>
> Benjamin Altman wrote:
>
> > Do you know how it compares to the first edition in terms of
presentation and
> > indexing?  I have heard from a few people that it is difficult to look
up things in
> > the 1st edition.  The only review on amazon of the 2nd ed. claims it is
a complete
> > rewrite despite his admitance that he has never read the book...
> >
> > Thanks,
> > Ben
> >
>
> Yes, I also found "programming Python", 1st edition, from Mark Lutz,
difficult to
> read...
> it was the most complete book at the time it was written. I still think
it's the most
> complete and the more advanced book about Python (I did not see any
others.., tell me
> if I am wrong).  The 2nd chapter, which was intended to be a 'sneak
preview of Python'
> or 'a view from 10.000 feet'  was very long, and tedious to understand.. I
guess that
> 10000 feet, it's too high for my poor stomach. There were even backward
references to
> this chapter from many other chapters... The other pb, you mentionned it,
was the
> organization... but it was not intended to be a reference book...rather a
'linear
> book'.
>
> The second edition seems to be better organized, but once again, Ive just
read a small
> fraction of it.
> Anyway, it is stated in the introduction that it is still not a reference
book. The
> book presents Python's libraries, tools and programming techniques. Its
aim is "how to
> use the Core language in applications." So, there are some big examples...
>
> Anyway the book is  intended for those who already know the Core Language
and want to
> see some applications using it...
> the book was refocused, since the 1st edition, and it is not aimed at
beginners (it's
> why you have to read the 'leaning Python' book first).
> I was therefore happy to see that 'The view from 10000 feet' was removed,
and chapter 2
>
> is now fun to read, because it presents the libraries, so it's not so
steep,....so,
> good work, Mark!
> The book is going to take me a few weeks to read, though, as it is 1200
pages long, but
>
> seems very promising.. and yes, it was completely rewritten...
>
>
>
>
>
> --
> http://mail.python.org/mailman/listinfo/python-list
>
>
> ________________________________________________________________________
> ________________________________________________________________________
>
> Message: 22
>    Date: Wed, 21 Mar 2001 22:02 +0000 (GMT Standard Time)
>    From: digitig at cix.co.uk (Tim Rowe)
> Subject: RE: What IDE are you'all using for Python 2.1 for Windows?
>
> In article <mailman.985119746.9901.python-list at python.org>,
> moshez at zadka.site.co.il (Moshe Zadka) wrote:
>
> > On Tue, 20 Mar 2001, digitig at cix.co.uk (Tim Rowe) wrote:
> >
> > > triple-quoted strings (one of the things I consider unpythonic that
> > > seems to have been in Python from the start, FWIW!)
> >
> > It hasn't been in Python from the start. It was added when people found
> > it very very painful to emit large strings to files.
> >
> > Compare:
> >
> > message = '''\
> > From: %(origin)s
> > To: %(dest)s
> > Subject: triple quotes in Python
> >
> > Why are there triple quoted strings
> > in Python?
> > ''' % vars()
> > os.popen("/usr/lib/sendmail -oi -t", 'w').write(message)
> >
> > To
> >
> > message = (
> > "From: %(origin)s\n"
> > "To: %(dest)s\n"
> > "Subject: triple quotes in Python\n"
> > "\n"
> > "Python doesn't need no\n"
> > "stinking triple quotes\n" ) % vars()
> > os.popen("/usr/lib/sendmail -oi -t", 'w').write(message)
> >
> > Which version would you feel safer letting a non-programmer modify?
>
> So Joe Non-Programmer comes along and changes the first one to:
>
> message = '''\
> From: %(origin)s
> To: %(dest)s
> Subject: triple quotes in Python
>
> Why does Python have this "'''"
> string thing?
> ''' % vars()
> os.popen("/usr/lib/sendmail -oi -t", 'w').write(message)
>
> Moral: don't let a non-programmer modify your program code! (Even more
> important, don't let a malicious programmer /masquerading/ as a
> non-programmer modify your program code :-)
>
> But heck, I use the things, so clearly I can live with 'em. But emacs
> can't parse 'em!
>
> --
> http://mail.python.org/mailman/listinfo/python-list
>
>
> ________________________________________________________________________
> ________________________________________________________________________
>
> Message: 23
>    Date: Wed, 21 Mar 2001 22:02 +0000 (GMT Standard Time)
>    From: digitig at cix.co.uk (Tim Rowe)
> Subject: Re: The most important modules
>
> In article <Pine.GSO.4.21.0103191732210.22234-100000 at laplace>,
> mrq at for.mat.bham.ac.uk (Martyn Quick) wrote:
>
> > I've seen a few comments along the lines of "When you are a newbie it
> > will
> > take a while to realize which the important modules are."  As someone
> > who
> > is in the middle of wading through Chapter 3 of the Library Reference,
> > it
> > would be very helpful to know which ones are particularly worth
> > noting.  (Presumably  sys, (c)Pickle and shelve, from the ones I've so
> > far
> > in the chapter - but what other treats are out there for me?)
>
> <snip>
>
> For me it's been:
>
> sys
> xml.parsers.expat
> string
>
> But that probably just reflects the fact that I've been doing a fair bit
> with command line utilities (using sys.argv, sys.stdin, sys.stdout and
> sys.stderr) that use XML! As somebody else said, it depends on what
> /you're/ doing.
>
>
> --
> http://mail.python.org/mailman/listinfo/python-list
>
>
> ________________________________________________________________________
> ________________________________________________________________________
>
> Message: 24
>    Date: Wed, 21 Mar 2001 22:02 +0000 (GMT Standard Time)
>    From: digitig at cix.co.uk (Tim Rowe)
> Subject: Re: python suitability for large critical run forever apps
>
> In article <3AB77CE4.78378862 at troikanetworks.com>,
> bedge at troikanetworks.com (Bruce Edge) wrote:
>
> > I started using Python for come user CLI stuff. It has worked out so
> > well that we're thinking about extending it's use in our project.
> > I am not experienced enough to be able to answer some of the issues that
> > came up. Does anyone have any thoughts on these:
>
> <snip>
>
> It depends on what you mean by "large critical run forever apps". I
> routinely deal with systems that can kill a few hundred people if they go
> wrong, and that's what /I/ mean by "critical". But then I baulk at the
> suggestion that such a system might be "large", never mind the suggestion
> that it might be multithreaded! If you need something like IEC61508 SIL4
> or the military equivalents then forget Python (and C/C++, and Perl, and
> Ruby...). That doesn't make /any/ of those a "toy" language, just not
> right for that application. I've said before -- here, I think -- that I
> don't believe in general purpose languages; to misquote Michael Jackson,
> if a language offers to helps you with /every/ problem, it won't help you
> /much/ with /any/ problem.
> --
> http://mail.python.org/mailman/listinfo/python-list
>
>
> ________________________________________________________________________
> ________________________________________________________________________
>
> Message: 25
>    Date: Wed, 21 Mar 2001 22:02 +0000 (GMT Standard Time)
>    From: digitig at cix.co.uk (Tim Rowe)
> Subject: I'm sure there's a good reason!
>
> Can any of the assembled experts explain to me why:
> print r"\hello"
> is fine, but
> print r"hello\"
> isn't? In other words, why does the backslash escape quotes in a raw
> string? At one level I'm going to guess (without reading the spec) that
> it's because "that's what the spec says", but why? I thought the idea of
> raw strings was that they /didn't/ handle escapes!
>
> TIA,
>
> Tim
>
> --
> http://mail.python.org/mailman/listinfo/python-list
>
>
> ________________________________________________________________________
> ________________________________________________________________________
>
>
>
> Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
>
>




From shaleh at valinux.com  Thu Mar  1 17:32:39 2001
From: shaleh at valinux.com (Sean 'Shaleh' Perry)
Date: Thu, 01 Mar 2001 14:32:39 -0800 (PST)
Subject: gah! I hate the new string syntax
Message-ID: <XFMail.20010301143239.shaleh@valinux.com>

return ";".join(["%s=%s" % (k, params[k]) for k in params.keys()])

every day python seems to be moving closer to the line noise aspect of
coding.



From aahz at panix.com  Tue Mar 20 11:03:20 2001
From: aahz at panix.com (Aahz Maruch)
Date: 20 Mar 2001 08:03:20 -0800
Subject: Functionalism, aesthetics Was:(RE: I come to praise .join,
  not
References: <mailman.985052844.17993.python-list@python.org>
Message-ID: <997v08$s4o$1@panix3.panix.com>

In article <mailman.985052844.17993.python-list at python.org>,
Carlos Ribeiro  <carribeiro at yahoo.com> wrote:
>
>Its unfortunate that in many cases the non-expert user keeps quiet, 
>assuming that whatever opinion the "master" haves, its going to be better 
>than theirs - after all they are the so-called experts.

There's some truth to that -- but rarely on Usenet newsgroups.
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

There's a difference between a person who gets shit zie doesn't deserve
and a person who gets more shit than zie deserves.  --Aahz


From db3l at fitlinxx.com  Tue Mar 27 19:13:29 2001
From: db3l at fitlinxx.com (David Bolen)
Date: 27 Mar 2001 19:13:29 -0500
Subject: win32: Pipes opened with popen3 cant be closed!
References: <mailman.985707122.14017.python-list@python.org>
Message-ID: <ud7b2rbrq.fsf@ctwd0143.fitlinxx.com>

Nahuel Greco <ngreco at softhome.net> writes:

> Hi, im doing the following thing in Windows 98 / Activestate Python 2.0:
> 
> 	import win32pipe
> 
> 	cmd = "c:\\zebedee\\zebedee.exe -d 127.0.0.1"
> 	pipes = win32pipe.popen3(cmd,'t')
> 
> 	#Now.. if i do:
> 
> 	pipes[0].close()
> 	pipes[1].close()
> 	pipes[2].close()  # in this line, the program hangs.
> 
> 
> If i follow another closing order, the program stills hangs in the last
> close() call .. i think that when all the pipes are closed.. win32pipe
> try to close the child program.. and wait for that close.. but, the
> child program never exit.

Yes - the purpose of the popen# calls is to connect you to the child
process so you can interact with it.  Upon closing that final handle,
it'll wait for the process to exit so it can return any exit code.

If you don't actually want that interaction, you can use os.system()
to start off the process instead, but you won't be able to trap the
output, and you'll still have the long-lived problem you mention next.

Sometimes the order that you close the handles is important, in terms
of closing the right one so that the process decides it should exit
(say, when it's stdin goes away).

> If i do not close the pipes, when my python script exits, the child
> program is still running (i dont want that).

That's more a Windows thing than anything else.  There's no official
parent/child relationship like there is under Unix, so Windows will
let processes continue running even after the process that started
them exits, and even if they didn't want to disassociate themselves.
(This causes me no end of grief with respect to our systems management
tools and ensuring that no stray processes ever remain on failures).

It sounds like what you want to do is to ensure that you cause the
executable to exit before you actually exit.  That can be specific to
the executable you are using.

In looking at the product page you referenced (and the source), I
don't see any way via stdin to tell zebedee to exit - and with popen()
a message via stdin is really the only way you have available to you.
The main server receive loop - whether detached or not - actually
blocks on an accept() call, so it's not looking for any other source
of input.  It appears to just expect to be terminated when it's
supposed to exit.

You might try looking around for a process list/kill utility for 98,
and then using that to find out the pid for the started zebedee
process, and explicitly killing it when you want to terminate.

Alternatively, since it comes with source, my bet is that if you just
tweaked the ServerListener function in zebedee.c so that instead of
blocking in the accept(), it did a timeout select on the socket and on
a timeout, checked stdin for input as a reason for exiting, you could
easily shut it down from the Python script by just writing something
to its stdin.  (The Unix way to do this would be to include
fileno(stdin) in the select, but Windows can only select on sockets -
alternatively if you didn't mind being Windows specific you could use
native Win32 calls to wait on both the socket and stdin to the same
effect).

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


From dmaas at dcine.com  Sun Mar 11 15:46:00 2001
From: dmaas at dcine.com (Dan Maas)
Date: Sun, 11 Mar 2001 20:46:00 GMT
Subject: Poll'able Queue objects.
References: <txkr904gwu9.fsf@vortex.ukshells.co.uk>
Message-ID: <cuRq6.240807$o91.31035696@typhoon.nyroc.rr.com>

> As this didn't seem to exist i have coded a selectable Queue object
> based on pipes (which should thus work under unix and win32).  I have
> included the code below.  If you go to the following URL :

I hate to say it, but this code won't work on Win32 since you can't select()
or WaitForMultipleObjects() on anonymous pipes (thanks, MS!). To get the
expected non-blocking behavior you would have to use threads or a local
socket...

Regards,
Dan




From urner at alumni.princeton.edu  Sat Mar  3 11:14:04 2001
From: urner at alumni.princeton.edu (Kirby Urner)
Date: Sat, 03 Mar 2001 08:14:04 -0800
Subject: Adding a method to an instance: solution
References: <G2%n6.360714$JT5.11606234@news20.bellglobal.com> <Ua8o6.302710$Pm2.4401349@news20.bellglobal.com>
Message-ID: <1162at05tslee55208v1isl2bv7dmsctbj@4ax.com>

Bryan Mongeau <bryan at eevolved.com> wrote:

>In case anyone is wondering:


Yeah, Paul sent that to me too.  I had no idea about the 'new' 
module.

  >>> import new
  >>> dir(new)
  ['__doc__', '__name__', 'classobj', 'code', 'function', 
  'instance',   'instancemethod', 'module']
  >>> print new.__doc__
  Functions to create new objects used by the interpreter.

  You need to know a great deal about the interpreter to use this!

I'll have to see what else I can find out about it, play
with it some.

Thanks again for asking your question -- good opportunity 
to spin my wheels re this Python stuff!

Kirby



From william.famy at noos.fr  Fri Mar 23 19:34:05 2001
From: william.famy at noos.fr (William Famy)
Date: Sat, 24 Mar 2001 01:34:05 +0100
Subject: newbe how to create 2 __init__ in a class ?
Message-ID: <3ABBEB7D.6E75E5B8@noos.fr>

good morning or evening.

Is it possible te create a class with 2 __init__

exemple i whant to create a classe with
(year,month,day,price1,price2,price3) 
or*					(string) "2000/12/28,12.3,56.3,45.3"

in C++ I tell the type but how to do it in python.

thanks for help


-- 
 			     \\\|///
                           \\  - -  //
                            (  @ @  )
+-------------------------oOOo-(_)-oOOo-------------------------+
|  Famy William  36 avenue des barattes  74000 Annecy  France   |
|  email:william.famy at noos.fr william.famy at mail-enseignant.com  |
+---------------------------------Oooo--------------------------+
                           oooO   (   )
                          (   )    ) /
                           \ (    (_/
                            \_)


From jwbnews at scandaroon.com  Wed Mar 14 10:05:34 2001
From: jwbnews at scandaroon.com (John W. Baxter)
Date: Wed, 14 Mar 2001 07:05:34 -0800
Subject: This math scares me
References: <3aad4062.1290021@News.CIS.DFN.DE> <984518199.127140@newsmaster-04.atnet.at> <jwbnews-09D75F.13313313032001@corp.supernews.com> <yTwr6.1235$Tg.167429@ruti.visi.com> <98mu97$fd1$1@panix3.panix.com>
Message-ID: <jwbnews-EF2E58.07053314032001@corp.supernews.com>

In article <98mu97$fd1$1 at panix3.panix.com>,
 aahz at panix.com (Aahz Maruch) wrote:

> In article <yTwr6.1235$Tg.167429 at ruti.visi.com>,
> Grant Edwards <grante at visi.com> wrote:
> >
> >Of course we'd never imagined we'd ever see hardware FP and BCD
> >libraries were only marginally slower than binary ones. Now
> >that FP is usually in hardware, BCD FP (or fixed point) will be
> >pretty slow by comparison.
> 
> If anyone cares, it ought to be fairly easy to create an unbounded BCD
> module using Python's longs.

Indeed, although it might (or might not) be more useful to produce a 
Rational module as in SmallTalk, using Python longs for the two parts of 
the fraction.  Either or both would have a place.  But I have neither 
time nor need.  [And the SmallTalk idiom makes dealing with Rational 
easier than it would be in Python, I think.]

  --John

-- 
John W. Baxter   Port Ludlow, WA USA  jwbnews at scandaroon.com


From breakdown at btinternet.com  Sun Mar  4 13:52:27 2001
From: breakdown at btinternet.com (matt)
Date: Sun, 04 Mar 2001 18:52:27 GMT
Subject: should i or shouldnt i?
Message-ID: <1103_983731947@breakdown>

i'm doing a course which revolves around smalltalk, but the pace is a bit slow for me.
i was thinking of tackling another language to fill the hours and wondered about python.
am i mad or would it be a good idea? what can python offer me? if theres 1 book to buy (ive already downloaded eckels book), which should it be?
i run win98 and linux.
thanks



From michel at digicool.com  Sat Mar 24 15:45:29 2001
From: michel at digicool.com (Michel Pelletier)
Date: Sat, 24 Mar 2001 12:45:29 -0800
Subject: PEP 245: Python interfaces
In-Reply-To: <j4hf0jz36a.fsf@informatik.hu-berlin.de>
References: <Pine.LNX.4.32.0103222253510.26302-100000@localhost.localdomain> <j4hf0jz36a.fsf@informatik.hu-berlin.de>
Message-ID: <Pine.LNX.4.32.0103241227160.26302-100000@localhost.localdomain>

On 24 Mar 2001, Martin von Loewis wrote:

> The following message is a courtesy copy of an article
> that has been posted to comp.lang.python as well.
>
> Michel Pelletier <michel at digicool.com> writes:
>
> > I'm the author of PEP 245, a framework and syntax for interface objects in
> > Python.
> >
> > http://python.sourceforge.net/peps/pep-0245.html
> >
> > It has just recently been made a PEP, and I'd like to solicite some more
> > input from the community (you).
>
> Even though this is quite a long document, it is incomplete.

Yes, *sigh*.  I know I need more formalism, but should this be at the
expense of the section on goals?  I'm not sure.  I was raised in an XP
environment, the goals are the most important part to me.  This is
something I've been debating myself over.  I obviously want the PEP to be
accessable and not boring, but I also want it to be comprehensive.

> This is
> probably not surprising since it is the first draft, but to really
> evaluate, I think a number of details should be added. In order of my
> preference:
>
> - proposal for a concrete syntax. The PEP should provide a detailed
>   specification of the grammar changes, preferably in (A)EBNF or the
>   Grammar/Grammar format. Currently, only a "Syntax Overview" is
>   given; that doesn't answer all possible questions. In particular,
>   the role of the "implements" reserved(?) word is never specified.

Good idea, I'll try to emulate Guido's informal grammar description from
the Language Reference.

> - detailed description of the contents of the module Interface.  For
>   example, Interface.Exception.BrokenImplementation is used but not
>   defined. For the moment, I assume that the proposal has the same
>   definitions as the reference implementation.

Yes.  I'll try to be more complete for the next draft.

> - a reference implementation that works as a patch to Python. This is
>   probably not needed at the moment, but certainly prerequisite for
>   people to completely evaluate the PEP.

Yes, at the moment, I feel I'm somewhere near that point in the PEP
process as described by PEP 1:

"The PEP should be reviewed and accepted before a reference implementation
is begun, unless a reference implementation will aid people in studying
the PEP."

> I like many aspects of the draft, but a few points should be
> corrected, IMO:
>
> - naming of the Interface package. Few standard modules are in
>   uppercase, and I think this shouldn't be uppercase, either. Instead,
>   I'd prefer a module "interfaces", in analogy to the types module.

Good idea.  It's current spelling is an artifact of the reference
implementation.

> - packaging. I think the hierarchy is too deep, everything should be
>   contained in the interfaces module (there is not an awful lot of
>   stuff in it, is there?).

How do you mean?  I don't think the PEP proposes any depth beyond the
Interface module.

> - exceptions. I think the standard exception hierarchy should be used
>   better, e.g. inheriting from NotImplemented (or directly using that
>   exception) where appropriate. Also, BrokenImplementation seems a bad
>   choice of wording; Smalltalk uses subclassResponsibility to indicate
>   that an abstract method is not implemented.

I'll look into that, certainly reducing any amount of new exceptions is a
good idea.

> - nesting of interface assertions. Why is it necessary to support
>
>    implements FooInterface, (BarInterface, (BobInterface, MyClass.__implements__))
>
>   Wouldn't it be sufficient (and more consistent with classes) to write
>
>   implements (FooInterface, BarInterface, BobInterface)+MyClass.__implements__

I do not know the exact motivation behind Jim choosing a DAG to describe
an interface assertion.  I'll find out.

> - Access to fields. I think it is confusing that you won't write
>   Iface.feature, e.g. in
>
>   interface Iface:
>     def feature(self):
>       "does something"
>
>   Instead of writing Iface.feature, it appears that you write
>   Iface.getDescriptionFor("feature"). It would be better if attribute
>   access could be used.

Ah yes, this is an interesting point of debate.  The idea here is that the
interface of an interface is *not* the interface that it describes.  I was
not around when the types-sig (and Jim) came up with this; I cannot give
you the exact motivation, but it feels right to me.  If you'd like, I can
add a section to 'Dissenting Opinions' section on this point.

Thanks Martin!

-Michel



From tjg at exceptionalminds.com  Sun Mar 18 00:46:51 2001
From: tjg at exceptionalminds.com (Timothy Grant)
Date: Sat, 17 Mar 2001 21:46:51 -0800
Subject: installing 2.x over the top of 1.5.2
In-Reply-To: <m33dcbivr1.fsf@fred.local>; from hgg9140@seanet.com on Sat, Mar 17, 2001 at 07:47:14PM -0800
References: <mailman.984861502.27092.python-list@python.org> <m33dcbivr1.fsf@fred.local>
Message-ID: <20010317214651.A26597@trufflehunter.avalongroup.net>

On Sat, Mar 17, 2001 at 07:47:14PM -0800, Harry George wrote:
> Don't know if this qualifies as clean.  I have the rpm'd 1.5.2 under
> /usr.  Then I installed py 2.x under /usr/local.  During our
> (extended) transition period, we use shell scripts which set
> PYTHONHOME and PYTHONPATH before calling python1.5 or
> python2.0.  Once all is well in 2.x, we'll remove the rpm'd 1.5.2, and
> (conceivably) install 2.x into /usr.

Well, it sure ain't clean, but it's a start! I don't know why
this was made so difficult. I wonder if it will be fixed in
v2.1.

-- 
Stand Fast,
    tjg.

Timothy Grant                         tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.         <><       (503) 246-3630
>>>>>>>>>>>>>Linux, because rebooting is *NOT* normal<<<<<<<<<
>>>>This machine was last rebooted:  60 days 10:00 hours ago<<



From tjg at exceptionalminds.com  Mon Mar  5 16:58:18 2001
From: tjg at exceptionalminds.com (Timothy Grant)
Date: Mon, 5 Mar 2001 13:58:18 -0800
Subject: Can anyone recomend a good intoduction to C...
In-Reply-To: <3aa3a3f0.36440038@news.laplaza.org>; from xyzmats@laplaza.org on Mon, Mar 05, 2001 at 02:42:16PM +0000
References: <3aa366c6_2@nnrp1.news.uk.psi.net> <3aa3a3f0.36440038@news.laplaza.org>
Message-ID: <20010305135818.S14262@trufflehunter.avalongroup.net>

Hear, Hear.

Another vote for K&R. It's ancient, but I still use it when I
have to do C stuff.

I used it to learn C programming 14 years ago. When I first
picked it up, I said "I can't learn how to programme from this
book!" So I bought one of those 1000 page doorstops referenced
below. I quickly returned to K&R.

On Mon, Mar 05, 2001 at 02:42:16PM +0000, Mats Wichmann wrote:
> On Mon, 5 Mar 2001 10:14:35 -0000, "Simon Brunning"
> <sbrunning at trisystems.co.uk> wrote:
> 
> >... for a Python programmer?
> >
> >I'm going to have to bite the bullet and learn a bit of C. I'd at
> >least like to learn the *right* way to do it.
> 
> My advice is similar to the one that gets given to prospective Python
> programmers:  start with the source - K&R (Kernighan & Ritchie). It's
> kind of similar to the Python tutorial in ways: concise but completely
> to the point, from people who know why the language was written that
> way. This is what everyone learned the language from in the early
> days. Stay away from those 1000-page doorstops with CDroms included.

-- 
Stand Fast,
    tjg.

Timothy Grant                         tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.         <><       (503) 246-3630
>>>>>>>>>>>>>Linux, because rebooting is *NOT* normal<<<<<<<<<
>>>>This machine was last rebooted:  48 days  2:10 hours ago<<



From root at [127.0.0.1]  Tue Mar 27 10:34:03 2001
From: root at [127.0.0.1] (deadmeat)
Date: Tue, 27 Mar 2001 15:34:03 GMT
Subject: stripping cr/lf, lf, cr
Message-ID: <Lp2w6.705$p5.3213@news1.rivrw1.nsw.optushome.com.au>

Is there a platform independant function that will remove appropriate EOL
characters from a string?  I don't want to use strip() since it will also
remove other whitespace I want to keep.





From fredrik at pythonware.com  Wed Mar 21 02:56:50 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 21 Mar 2001 07:56:50 GMT
Subject: Base 2 to long integers and back
References: <mailman.985157126.16257.python-list@python.org> <5WYt6.846$sk3.282417@newsb.telia.net>
Message-ID: <69Zt6.850$sk3.282870@newsb.telia.net>

> def f2(val, joiner="".join, binmap=binmap):

and yes, feel free to rename the function ;-)

Cheers /F




From stephen_purcell at yahoo.com  Mon Mar 26 10:04:34 2001
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Mon, 26 Mar 2001 17:04:34 +0200
Subject: Lists and Files
In-Reply-To: <99njdo$oa3$1@news.lth.se>; from aaldv97@student.removethispart.vxu.se on Mon, Mar 26, 2001 at 04:23:24PM +0200
References: <99njdo$oa3$1@news.lth.se>
Message-ID: <20010326170434.A16310@freedom.puma-ag.com>

David Andreas Alderud wrote:
> I'm new to Python, started yesterday, is there a good way to handle lists
> and files?
> 

Try the 'pickle' module:

>>> import pickle
>>> stooges = ['val1', 'val2', 1, [['val2_1', 'val2_2', 0], ['val2_1',
... 'val2_2', 0]]]
>>> save_file = 'stooges.dat'
>>> output = open(save_file,'wb')
>>> pickle.dump(stooges, output)
>>> output.close()
>>> input = open(save_file,'rb')
>>> stooges2 = pickle.load(input)
>>> input.close()
>>> stooges2
['val1', 'val2', 1, [['val2_1', 'val2_2', 0], ['val2_1', 'val2_2', 0]]]

-Steve

-- 
Steve Purcell, Pythangelist
Get testing at http://pyunit.sourceforge.net/
Any opinions expressed herein are my own and not necessarily those of Yahoo



From aahz at panix.com  Sun Mar 25 12:34:56 2001
From: aahz at panix.com (Aahz Maruch)
Date: 25 Mar 2001 09:34:56 -0800
Subject: Newbie needs help to find bug in his first Perl program
References: <3ABCC757.6C6F17CA@earthlink.net>
Message-ID: <99la80$nvs$1@panix6.panix.com>

In article <3ABCC757.6C6F17CA at earthlink.net>,
Ron Stephens  <rdsteph at earthlink.net> wrote:
>
>sixtwo = raw_input ("where does %s rank for %y ?" % (option6,
>criteria2))
>
>Can anyone see what is wrong with this? I have several lines like this
>and they all give me error messages.

What error message are you getting?
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

"I won't accept a model of the universe in which free will, omniscient
gods, and atheism are simultaneously true."  -- M


From rs at richsomerfield.com  Thu Mar  1 15:11:34 2001
From: rs at richsomerfield.com (Richie Sum)
Date: Thu, 1 Mar 2001 20:11:34 -0000
Subject: [Q] Question from Python tutorial
References: <Lgxn6.6964$dL4.91439@vixen.cso.uiuc.edu> <3A9EA472.F15AF847@cybermesa.com> <hxxn6.6969$dL4.91744@vixen.cso.uiuc.edu>
Message-ID: <t9tbbpigh7gf33@corp.supernews.co.uk>

If u cut n paste your code in courier font then it is much easier to spot
possible problems.
Maybe a missing / extra space character?

RS


Young-Jin Lee <ylee12 at uiuc.edu> wrote in message
news:hxxn6.6969$dL4.91744 at vixen.cso.uiuc.edu...
> I typed them manually. I dind't copy and paste.
>
> "Jay O'Connor" <joconnor at cybermesa.com> wrote in message
> news:3A9EA472.F15AF847 at cybermesa.com...
> >
> >
> > Young-Jin Lee wrote:
> >
> > > Hi, I have a question from Python tutorial.
> > > The code segment in the page 60 (chapter 9. classes) is not executed.
> > > I don't know what I did wrong. Here is the code and the error I got.
> > >
> > > class MyClass:
> > >     "A simple example class"
> > >     i = 12345
> > >     def f( x ):
> > >         return 'hello world'
> > >
> > > x = MyClass()
> > >
> > > x.counter = 1
> > > while x.counter < 10:
> > >     x.counter = x.counter * 2
> > > print x.counter
> > >       ^
> > > SyntaxError: invalid syntax
> > >
> > > There is no typo, I have no idea of where this SyntaxError came from.
> > >
> > > Thanks in advance. Any comment would be greatly appreciated.
> >
> > You probably cut and pasted the code from the tutorial.  When I cut and
> > pasted your code into IDLE, I got the same error.  When I manyually
typed
> > the code, it works fine.  My guess is that the cut and paste didn't
paste
> > properly and the interactive interpreter didn't make the right break
> > between the 2 and the word print
> >
> > --
> > Jay O'Connor
> > joconnor at cybermesa.com
> > http://www.cybermesa.com/~joconnor
> >
> > Python Language Forum -
> > http://pub1.ezboard.com/fobjectorienteddevelopmentpython
> >
> >
>
>




From jmarshal at mathworks.com  Fri Mar 16 11:33:49 2001
From: jmarshal at mathworks.com (Joshua Marshall)
Date: 16 Mar 2001 16:33:49 GMT
Subject: list.sort(cmpfunc) question
References: <m1ae6lg5qt.fsf@pclabs188.labs.trema.com> <98t9dq$qvu$1@panix6.panix.com>
Message-ID: <98tf9d$g3v$1@news.mathworks.com>

Aahz Maruch <aahz at panix.com> wrote:
> In article <m1ae6lg5qt.fsf at pclabs188.labs.trema.com>,
> Harri Pasanen  <harri at trema.com> wrote:
>>
>>The list.sort(cmp) docs for Python 2.0 state:
>>
>>   The sort() method takes an optional argument specifying a
>>   comparison function of two arguments (list items) which should
>>   return -1, 0 or 1 depending on whether the first argument is
>>   considered smaller than, equal to, or larger than the second
>>   argument.

> Note that "should" carefully.  The current implementation of sort() only
> checks for a negative value, but that's not a behavior that should be
> relied on in your coding.  You can be sure that any anamolies in the
> behavior of cmp() will be fixed if the sort() algorithm ever changes.

Seems the word "should" should never appear in documentation.


From bedge at troikanetworks.com  Fri Mar 23 10:38:09 2001
From: bedge at troikanetworks.com (Bruce Edge)
Date: Fri, 23 Mar 2001 07:38:09 -0800
Subject: sharing data between modules
Message-ID: <20010323.073801.76065818.26070@mead.troikanetworks.com>

As a C++ guy I tend to put each class in a different file. In python this
gives me a different module for each class. My app uses this group of
classes, ie: modules, what's the best way to share data between them?

Currently I register the classes with each other using class static data, 
so they all have a reference to each other. Ideally, I'd like to be able
to have a few globals that every module can see.  

Am I doing this the hard way?

TIA, Bruce.


From grante at visi.com  Sat Mar 17 11:20:59 2001
From: grante at visi.com (Grant Edwards)
Date: Sat, 17 Mar 2001 16:20:59 GMT
Subject: This math scares me
References: <20010312170356.B3513@better.net> <mailman.984664520.26463.python-list@python.org> <HR4s6.877$lz1.73512@ruti.visi.com> <98qnq0$qc2$1@panix3.panix.com> <98umb1$f2i$1@panix3.panix.com>
Message-ID: <slrn9b7as9.10s.grante@tuxtop.visi.com>

On 16 Mar 2001 19:40:17 -0800, Aahz Maruch <aahz at panix.com> wrote:

>>Paging Ivan Van Laningham.....
>
>Ivan isn't reading c.l.py these days, but he sent me an e-mail response
>that he said I could post:
>
>We get base 60 from the Babylonians, who got it from the Sumerians. 
[...]
>Aztecs and Mayans used base 20, expanded from an earlier base 10 system,
>which appears to have grown from an earlier base 5.

It looks like I had them mixed up with the Babylonians.  I
swear, the denizens of c.l.p have the widest base of knowege of
any newsgroup I've read.

-- 
Grant Edwards                   grante             Yow!  I feel better about
                                  at               world problems now!
                               visi.com            


From ylee12 at uiuc.edu  Thu Mar  1 14:38:02 2001
From: ylee12 at uiuc.edu (Young-Jin Lee)
Date: Thu, 1 Mar 2001 13:38:02 -0600
Subject: [Q] Question from Python tutorial
References: <Lgxn6.6964$dL4.91439@vixen.cso.uiuc.edu> <3A9EA472.F15AF847@cybermesa.com>
Message-ID: <hxxn6.6969$dL4.91744@vixen.cso.uiuc.edu>

I typed them manually. I dind't copy and paste.

"Jay O'Connor" <joconnor at cybermesa.com> wrote in message
news:3A9EA472.F15AF847 at cybermesa.com...
>
>
> Young-Jin Lee wrote:
>
> > Hi, I have a question from Python tutorial.
> > The code segment in the page 60 (chapter 9. classes) is not executed.
> > I don't know what I did wrong. Here is the code and the error I got.
> >
> > class MyClass:
> >     "A simple example class"
> >     i = 12345
> >     def f( x ):
> >         return 'hello world'
> >
> > x = MyClass()
> >
> > x.counter = 1
> > while x.counter < 10:
> >     x.counter = x.counter * 2
> > print x.counter
> >       ^
> > SyntaxError: invalid syntax
> >
> > There is no typo, I have no idea of where this SyntaxError came from.
> >
> > Thanks in advance. Any comment would be greatly appreciated.
>
> You probably cut and pasted the code from the tutorial.  When I cut and
> pasted your code into IDLE, I got the same error.  When I manyually typed
> the code, it works fine.  My guess is that the cut and paste didn't paste
> properly and the interactive interpreter didn't make the right break
> between the 2 and the word print
>
> --
> Jay O'Connor
> joconnor at cybermesa.com
> http://www.cybermesa.com/~joconnor
>
> Python Language Forum -
> http://pub1.ezboard.com/fobjectorienteddevelopmentpython
>
>




From BgPorter at NOartlogicSPAM.com  Mon Mar  5 09:59:29 2001
From: BgPorter at NOartlogicSPAM.com (Brett g Porter)
Date: Mon, 05 Mar 2001 14:59:29 GMT
Subject: suggestions for more docs (was Re: Bug fix releases)
References: <LNBBLJKPBEHFEDALKOLCMEEPJBAA.tim.one@home.com> <mailman.982897324.9109.python-list@python.org>              <200103031821.NAA24060@panix3.panix.com> <mailman.983646726.27322.python-list@python.org> <3AA2EDBC.E933E78E@seebelow.org>
Message-ID: <lRNo6.343166$w35.55348697@news1.rdc1.nj.home.com>

"Grant Griffin" <not.this at seebelow.org> wrote in message
news:3AA2EDBC.E933E78E at seebelow.org...
All the comments I make are from the perspective of someone who spends 95%+
of his time writing C++, and came to Python about 2 years ago looking for a
'little language' as a diversion and a way to keep my brain fresh. I wasn't
expecting to fall for Python in the way that I have. However, I still don't
get to actually book that much time in the language (yet).

> - Tutorial, Part II: This would be directed at people who have practiced
> Python for, say, 3 months.  It would introduce them to the Python's more
> features that aren't covered in the original tutorial.  It would be
> written using examples (ala K&R), which would give a sense not just of
> "what and how" but more of "when and why".
Yes -- the big thing for me continues to be learning to Think in Python.
The hardest part for this task -- creating a body of useful sample code --
is already done. Reading through the code that comes under the Tools
directory has been the single best investment that I've made to thinking in
the Pythonic Way. An annotated guide to this would be golden.


> - A Tour of The Standard Library: This would help beginners learn the
> most important standard modules--in effect, clearing the forest of its
> useful (but less used) trees.  I really think that listing the most
> frequently used modules (and the most important functions _in_ those
> modules) would be a _great_ service to beginners.  Off the top of my
> head, they might be (in no order): string, os, os.path, sys, math, re,
> etc.  (Even just providing a _list_ of these things in order of
> importance would be a great service, though, admittedly, that wouldn't
> provide the PythonLabs team with more than a couple of hours of
> excitement. <wink>)

I haven't picked it up, but isn't this what the effbot guide is?
http://www.pythonware.com/people/fredrik/librarybook.htm
Hmm. I just checked out the TOC
http://www.pythonware.com/people/fredrik/librarybook-contents.htm
and really do need to get this.

...and let's not forget the tutorial pt. III -- The complete and utter guide
to hardcore Python.

--
// Brett g Porter * Senior Engineer, Application Development
// BgPorter @ artlogic . com * http://www.artlogic.com
// Art & Logic * Custom software solutions for hardware products
// Windows * MacOS * Embedded Systems





From bedge at troikanetworks.com  Fri Mar 23 20:04:45 2001
From: bedge at troikanetworks.com (Bruce Edge)
Date: Fri, 23 Mar 2001 17:04:45 -0800
Subject: syslog beeps and logs to console.... sometimes!?
Message-ID: <20010323.170444.1704365084.578@mead.troikanetworks.com>

I use syslog for debug info. It beeps and prints a:

  Message from syslogd at mead at Fri Mar 23 16:56:19 2001 ... mead cli:   
  CTREE: tokenize "cr"

type message to the console, both my xterm and text console 0 (Linux).

The behavior as to which messages beep changes depending on whether it's
running under pdb or not.

Here's my open log line, no console logging requested: 

	openlog( name, 0, LOG_USER )

Anyone else ever seen this?

This happens on 2.0 and 2.1b1.

Here's some output showing the normal syslog output interleaved with the
"Message from..." things that beep:


Mar 23 16:56:09 mead cli: INIT: loading topcmds...
Mar 23 16:56:09 mead cli: INIT: ...topcmds complete
Mar 23 16:56:10 mead cli: INIT: ...init complete

Message from syslogd at mead at Fri Mar 23 16:56:19 2001 ...
mead cli:    CTREE: tokenize "cr"

Message from syslogd at mead at Fri Mar 23 16:56:19 2001 ...
mead cli:    CTREE: ctree::tokenize tokens=['cr'], tokpos=[0], tokix=0
Mar 23 16:56:19 mead cli:    CTREE: tokenize "cr"
Mar 23 16:56:19 mead cli:    CTREE: ctree::tokenize 
Mar 23 16:56:20 mead cli:   CTREE: complete active_tok = 0

Message from syslogd at mead at Fri Mar 23 16:56:20 2001 ...
mead cli:    CTREE: parse tokix=0, active_tok=0
Mar 23 16:56:20 mead cli:    CTREE: parse tokix=0, active_tok=0
Mar 23 16:56:21 mead cli:     CTREE: getMatches self=root str=cr
Mar 23 16:56:21 mead cli:      CTREE: getLocalMatches self=root str=cr
Mar 23 16:56:22 mead cli:      CTREE: getRemoteMatches self=root str=cr
Mar 23 16:56:22 mead cli:     CTREE: parse type:0 Matches=['create']


From DSHIN at nc.rr.com  Thu Mar  1 16:07:48 2001
From: DSHIN at nc.rr.com (Daehyok Shin)
Date: Thu, 01 Mar 2001 21:07:48 GMT
Subject: [Q] Inspect variables during execution?
Message-ID: <ESyn6.12612$Z8.1615647@typhoon.southeast.rr.com>

Is there any function helping me to inspect variables for debugging during
execution?

--
Daehyok Shin (Peter)





From aleaxit at yahoo.com  Fri Mar 16 09:33:08 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 16 Mar 2001 15:33:08 +0100
Subject: Who's minister of propaganda this week?
References: <mailman.984595938.11127.python-list@python.org> <98otuq0nnf@news1.newsguy.com> <slrn9b0cc2.gsa.neelk@alum.mit.edu> <98q6ro2uoa@news2.newsguy.com> <slrn9b2u0r.j74.neelk@alum.mit.edu>
Message-ID: <98t8830nlb@news1.newsguy.com>

"Neelakantan Krishnaswami" <neelk at alum.mit.edu> wrote in message
news:slrn9b2u0r.j74.neelk at alum.mit.edu...
    [snip]
> > Keying a dict on class has the unfortunate side-effect of
> > not allowing polymorphism by inheritance -- objects of classes
> > _derived from_ the one used as the key will not match.
> >
> > You can preserve this basic idea by using as key, instead of the
> > class, some ad-hoc attribute which will be inherited (if it IS
    [snip]
> I actually have a function that walks the class tree checking it
> against the dictionary, which solves this problem.  You don't need to
> add a special attribute in Python since __class__ gives you what you
> need for free.

...except it doesn't, because it's _not_ inherited -- you have
to 'manually' walk the directed acyclic graph (tree? what tree?
we have _multiple_ inheritance:-) or iterate through the dictionary's
keys (assumed to be types) using isinstance for each on your
object.

> I only added it to my toolkit recently, though, sine I tend to use
> classes like algebraic data types in ML or Haskell -- as tagged
> variants rather than designed for user-extension. I find subclassing
> friendly designs are a lot harder to design in a way that doesn't
> reveal too much internal structure.

Inheritance (of implementation) is strong coupling, yes.  But,
it IS a very handy implementation technique -- just as, in
abstract algebra, you DON'T have to prove again for groups
all the theorems you've just proved for semigroups (they get
inherited, because so do the axioms), similarly you don't
have to redo any code that already works for the baseclass
(you do have to re-*test* if you do any overriding, but that
is another issue, and doesn't affect 'pure inheritance with
no overriding').

Here's a sketch of a general-by-type dispatcher module to
ensure one can dispatch on either type OR class (and in
the latter case gets inheritance by default, unless it's
explicitly overridden):


dispatch_keys = {}
next_key = 1

def type_or_class_of(anobj):
    try:
        return anobj.__class__
    except AttributeError:
        return type(anobj)

def get_dispatch_key_for_type(atype):
    try:
        return atype._dispatch_key_
    except AttributeError:
        return dispatch_keys.get(atype)

def set_dispatch_key_for_type(atype):
    global next_key
    try:
        atype._dispatch_key_ = next_key
    except AttributeError:
        dispatch_keys.setdefault(atype,next_key)
    next_key += 1

def ensure_type_has_dispatch_key(atype):
    if get_dispatch_key_for_type(atype) is not None:
        return
    set_dispatch_key_for_type(atype)


def get_dispatch_key_for_obj(anobj):
    atype = type_or_class_of(anobj)
    return get_dispatch_key_for_type(atype)

def set_dispatch_key_for_obj(anobj):
    atype = type_or_class_of(anobj)
    set_dispatch_key_for_type(atype)

def ensure_obj_has_dispatch_key(anobj):
    atype = type_or_class_of(anobj)
    ensure_type_has_dispatch_key(atype)


Alex





From donotreply at interbulletin.bogus  Thu Mar  8 17:12:00 2001
From: donotreply at interbulletin.bogus (Bill Tolbert)
Date: Thu, 08 Mar 2001 22:12:00 +0000
Subject: mxODBC on Windows 2000?
References: <3AA7B94A.23D5C849@interbulletin.com>
Message-ID: <3AA803B0.7E08E836@interbulletin.com>

It's not limited to ActivePython: I get the same (non) result under IDLE 
and the command line. I'm seeing it fail under Win2000 and Win98.

Bill

Rob Earl <donotreply at interbulletin.bogus> wrote in article 
<3AA7B94A.23D5C849 at interbulletin.com> : 
>
>Has anyone been able to get the mxODBC module for Python to work properly 
with ActivePython on Windows 2000?  
The reason I ask is because every time I try to execute a SQL command, 
the CPU usage goes up to 100% on my machine and the python shell is 
locked up.  Can anyone help?  Thanks.
_______________________________________________
Submitted via WebNewsReader of http://www.interbulletin.com



From whisper at oz.net  Wed Mar 21 12:34:50 2001
From: whisper at oz.net (Dave LeBlanc)
Date: 21 Mar 2001 17:34:50 GMT
Subject: Standalone executables
References: <998igr$4al$1@niagara.nonet> <iXWt6.156032$Z8.33024358@typhoon.southeast.rr.com> <99aldn$7dh$1@niagara.nonet>
Message-ID: <99aonq$a4c$0@216.39.170.247>

I found this on c.l.p.announce:
The Linux version of Installer for Python 2 is here:
 http://www.mcmillan-inc.com/installer_dnld.html

Background:
 The Installer is a sort of compilerless Freeze - a way 
of distributing Python apps with all the required support built 
in. The Linux version collects all dependencies and packages 
the pure Python into an archive attached to the executable.
Just tar, gzip and ship.
---------------------------------------
I believe this package also works with/for Windows.

Regards,

Dave LeBlanc

On Wed, 21 Mar 2001 16:38:25 GMT, mark at chaos.x-philes.com (Mark
Carroll) wrote:

>In article <iXWt6.156032$Z8.33024358 at typhoon.southeast.rr.com>,
>Don Tuttle <tuttledon at hotmail.com> wrote:
>(snip)
>>On the Windows side, check out
>>http://starship.python.net/crew/theller/py2exe/ to compile and
>>http://www.jrsoftware.org/isinfo.htm to install your program.  Can't help
>>you on the Linux side.
>
>Thanks very much!
>
>I wonder, for Linux, what if I wrote in Jython, compiled with jythonc,
>then used a Java native-code compiler? Would that be too insane? (-:
>
>-- Mark



From lsommerer at sewardweb.com  Thu Mar 22 09:50:48 2001
From: lsommerer at sewardweb.com (Lloyd Sommerer)
Date: Thu, 22 Mar 2001 08:50:48 -0600
Subject: learning python...
References: <3AB8F4E1.184DD9C@noaa.gov>
Message-ID: <3ABA1148.ABBC50C0@sewardweb.com>

About a month ago I was in about the same possition you describe
(programming experience, no python experience).  After looking at most of
the python books, I decided to go with _Python_Essential _Reference_.  It's
10 page introduction and 70 pages of more indepth material were just what I
needed to get a feel for python.  The rest of the book is about 200 pages
of material on various library modules.

The whole thing is well organized, and it was easy to vaguely recall where
some bit of information was when I needed to look something up.  You'll
also need the online documentation, and I found the way ActiveState
includes it in their python distribution to be quite helpful.

One final bit of advice:  When you're wondering how to use a specific
module, the best documentation is very often within the module itself.  I
didn't actually come across that advice anywhere, and it would have helped
me to have known it sooner.

Lloyd Sommerer

"Benjamin.Altman" wrote:

> Hello.  From the perspective of someone who already has programming
> experience, would anyone know if O'reillys "Learning Python" book would
> be better than say Prentice Hall's "Core Python Programming"?  Or would
> it be better to go straight into something like "Programming Python"?
>
> Thanks,
> Ben



From leo at iems.nwu.edu  Tue Mar 27 16:32:45 2001
From: leo at iems.nwu.edu (Leonardo B. Lopes)
Date: Tue, 27 Mar 2001 15:32:45 -0600 (CST)
Subject: order by instead of two queries (Apologies)
In-Reply-To: <Pine.LNX.4.10.10103271629140.27947-100000@rksystems.com>
Message-ID: <Pine.GSO.4.04.10103271531480.5279-100000@gomory.iems.nwu.edu>

Somehow I sent this to the wrong place. Sorry about that, and thanks for
the answer anyway...

Leo.

On Tue, 27 Mar 2001, Bob Kline wrote:

> On Tue, 27 Mar 2001, Leonardo B Lopes wrote:
> 
> > I have some data in a table that I would like to obtain order by a
> > specific instance of the data. ex:
> > 
> > Name:		Age:
> > 
> > Joe		50
> > Jill		60
> > Jack		65
> > 
> > What I would like to do is have a query that returns Jack first, then
> > the regular name order. Something like:
> > 
> > SELECT Name FROM clients ORDER BY Name='Jack',Name
> > 
> > Can this be done with one query?
> 
> This can be done in standard SQL (using a CASE operation) but this is a
> Python mailing list, not a SQL list.
> 
> -- 
> Bob Kline
> mailto:bkline at rksystems.com
> http://www.rksystems.com
> 
> 

========================================================================
Leonardo B. Lopes                                       leo at iems.nwu.edu 
Ph.D. Student                                              (847)491-8470
IEMS - Northwestern University              http://www.iems.nwu.edu/~leo




From tuttledon at hotmail.com  Thu Mar  1 20:26:35 2001
From: tuttledon at hotmail.com (Don Tuttle)
Date: Fri, 02 Mar 2001 01:26:35 GMT
Subject: [ANNOUNCE] Webware for Python 0.5
References: <mailman.983454182.31364.python-list@python.org>
Message-ID: <fFCn6.13480$Z8.1902263@typhoon.southeast.rr.com>

"Chuck Esterbrook"

>    * Python Server Pages (PSP, similar to ASP, PHP and JSP)

I 'm trying to get a handle on why I would want to use PSP on WinNt with
IIS.  What's the difference/advantage to using PSP compared to using ASP
with embedded Python script?

Don




From tim.hochberg at ieee.org  Fri Mar  9 17:12:03 2001
From: tim.hochberg at ieee.org (Tim Hochberg)
Date: Fri, 09 Mar 2001 22:12:03 GMT
Subject: Dictionary attributes and functions
References: <Pine.LNX.4.21.0103092014190.19640-100000@moroni.pp.asu.edu>
Message-ID: <Tycq6.11675$Xt3.1961859@news1.rdc1.az.home.com>

"iddwb" <iddwb at imap1.asu.edu> wrote in message
news:Pine.LNX.4.21.0103092014190.19640-100000 at moroni.pp.asu.edu...
> I'm looking to read a list of items where their may be duplicates.  I'd
> like to use a dictionary to do this but am way to new to python to know
> fully how to do it.  I'm hoping there is some set of fuctions or
> attributes that can be reference regarding dictionaries... specifically if
> key(a) is already in dictionsary x I'd like to know prior to
> adding.  rather than change the value of key(a) I'd like to add the value
> of key(`a) to the value of key(a) so I end up with

For some details of the dictionary methods try
http://www.python.org/doc/current/lib/typesmapping.html

> {keya: [123]}
> on insert of keya againg have something like
> {keya: [123, 234]}

For this particular problem, you probably want the oddly named setdefault
method :

    myDict.setdefault(aKey, []).append(aValue)

This will also work:

if myDict.has_key(aKey):
   myDict[key].append(aValue)
else:
   myDict[key] = [aValue]


-tim






From aleaxit at yahoo.com  Tue Mar 20 18:39:11 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Wed, 21 Mar 2001 00:39:11 +0100
Subject: Functionalism, aesthetics Was:(RE: I come to praise .join,  not
References: <mailman.985052844.17993.python-list@python.org> <996vss013mp@news2.newsguy.com> <998a11$q3e$1@nntp6.u.washington.edu>
Message-ID: <998psi02tl5@news2.newsguy.com>

"Donn Cave" <donn at u.washington.edu> wrote in message
news:998a11$q3e$1 at nntp6.u.washington.edu...
> Quoth "Alex Martelli" <aleaxit at yahoo.com>:
> ...
> | Buildings as such (and more generally dwelling-places) are old
> | enough that it's quite possible that we all share some basic
> | biological basis for sound judgment -- I'm not pronouncing
> | on that, but that's what Alexander is mostly writing about.
    [snip]
> Anyway, as long as we're talking about things that we spend our
> days doing, we certainly do have the capacity for sound judgement.
> It doesn't mean our judgement is therefore infallibly sound, but
> the collective judgement of an established culture brings a kind
> of sense to bear that can dwarf a brilliant appeal to abstract
> principles.

Do you perchance play contract bridge?  It's interesting, in this
regard, because you will find in it both stuff that is susceptible
of being settled once and for all (mostly in the card-play part)
AND stuff that is endlessly debatable (mostly in the bidding).

Indeed, it might have been conceived as a practical experiment
to test (among other things) these assertions of your about what
"we" are or aren't able to do regarding "things we spend out days
doing".  Lots of retirees spend their days playing contract bridge,
particularly today when they can do it online from the comfort of
their homes -- and online play has the advantage that every card
played by each participant is recorded forever, computer accessible,
susceptible to any kind of analysis one might dream of.

_Some_ of these guys and gals, merely 60 years after Emile Borel
first published in his "Theorie Mathematique du Bridge" some of
the "obvious" applications of Bayes' Theorem to card play, have
finally managed to pick the most elementary ones up _in a
practical sense_ -- in bridge, it goes by the name "Restricted
Choice", because that's how it was dubbed by a great champion
and writer more than a dozen years after Borel's book (and for
twenty more years, it was a *controversial* matter in _expert_
circles... and that's a mathematically *obvious* part of that
subset of the game which CAN be "settled once and for all"!!!).

This is an endeavour that (roughly in its present form) has been
around for over 70 years; for over a century, actually, other forms
of bridge have been practiced, with card-play aspects that are
not materially different from today's version.  Yet the most basic
and elementary aspects of the application of probability theory
to bridge card-play are STILL not in the "collective judgment of
... established culture" if one means by 'collective' the judgment
of the hoi polloi, as opposed to that of people who can be
called 'experts' of the game.  And don't get me started on the
applications of game theory -- it appears one needs a degree
in maths, as well as decent bridge expertise, to avoid making
the most incredible blunders in what should yet be a simple
indeed case of zero-sum, imperfect-information game (in the
game-theoretical sense; again, I refer to bridge _card-play_).

Memes do spread faster than genes -- but, as bridge shows,
we're still talking in terms of centuries, not of years, before a
meme has spread anywhere like enough to make 'collective
judgment' anything but an oxymoron.  Chemistry (cfr my
previous example on the 'paler pills') is an example of an
endeavour that has been around for centuries -- maybe we
don't spend our days "doing chemistry", but some of us
take lots of pills &c, yet that doesn't make our aesthetics
anywhere close to having any usefulness in their regard.

Economic activity IS what most of us spend our days at --
yet well-known, *obvious* fallacies which Ricardo was already
denouncing in ringing tones 200 years ago STILL rule the
economics opinion of most citizens (anything from "lump
of labour" to "their harbors are bad, so we should mine
ours", the latter being Ricardo's colorful but correct parallel
for the instinctive protectionist reaction to other nations'
protectionism).

Stored-program computers are quite a bit newer than
bridge, much newer than chemistry and even economics.
What makes you think 'collective judgment' has any real
worth in the former, any more than in the latter three?

"Dwelling in a place" is something that we have been
doing for FAR longer -- _building_ places to dwell in is
an activity which can be traced back for many millennia,
and the apperception of space runs very deep in our
genes (while that of, e.g., probability theory, does NOT:
indeed, we seem to be wired FOR certain fallacies, and
it takes lots of memes on top to correct that wiring!-).

Whatever holds in the realm of architecture, therefore,
has no necessary parallel in that of programming, where
"collective judgment" is concerned.


Alex





From garrett at cs.colostate.edu  Fri Mar  2 00:20:25 2001
From: garrett at cs.colostate.edu (Deon Garrett)
Date: Fri, 02 Mar 2001 05:20:25 GMT
Subject: To the point of defeat.....IDLE
References: <3A9AE62B.5070402@yahoo.com> <slrn99s6tu.8rl.garrett@falcon.cs.colostate.edu> <%ftn6.12742$zL.491362@e420r-atl1.usenetserver.com>
Message-ID: <slrn99ubbn.9ul.garrett@eagle.cs.colostate.edu>

Well, don't take this as the gospel truth -- I haven't actively tried to
break anything since upgrading my Python to 2.0.  But, it has been a while
and nothing has failed in my day-to-day use of the machine.  I'm actually
using Mandrake 7.2, but I use some of the RedHat stuff for administration
and it all seems to work.  Of course, I don't mind re-installing stuff from
scratch if I screw it up.  If you're talking about a mission critical system,
it would definitely be worth checking out before you bail right in to the
upgrade.  Just my two cents...

Deon


>
>Has anyone actually verified that the myriad pieces of RedHat Python aren't
>incompatible with 2.x? RedHat makes heavy use of Python for various system
>configuration tasks, and I've always been cautious about replacing the
>RPM-installed Python (so I installed newer versions
>with --prefix=/usr/local).
>
>regards
> Steve
>
>
>



From winson at mdsserv.mds.com.tw  Mon Mar 19 02:27:52 2001
From: winson at mdsserv.mds.com.tw (=?big5?B?sWm4zq9x?=)
Date: Mon, 19 Mar 2001 15:27:52 +0800
Subject: Tkinter Support
Message-ID: <001901c0b046$1f1a7200$340210ac@winson>

Hi,

I downloaded ActiveState's PythonWin, but cannot run Tkinter module. What's going on ??

--
Best Regards,

Winson Chang

3F, 2, 150 Lane, Sec. 5, Hsin I Rd., Taipei, Taiwan, R.O.C.

Tel: +886-2-27225333 ext 376
Fax: +886-2-27222330

Email: winson at mdsserv.mds.com.tw
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20010319/e452363d/attachment.html>

From bsb at winnegan.de  Thu Mar  8 07:56:42 2001
From: bsb at winnegan.de (Siggy Brentrup)
Date: 08 Mar 2001 13:56:42 +0100
Subject: urllib.urlopen and proxies
In-Reply-To: <XWKnOkoTSX9D5gaI8iSioP21b6Iv@4ax.com>
References: <vse4atgcgcfafth7c5tq3po7srd3l32f3h@4ax.com>
	<008301c0a57d$1331cde0$a100000a@local>
	<mailman.983804894.17415.python-list@python.org>
	<BfakOm9OYyn2V0Hg9h1SSS=a7t7Y@4ax.com>
	<mailman.983901970.11969.python-list@python.org>
	<vBKmOkB47iH8JfPSAvl1kJShzae+@4ax.com>
	<mailman.983969111.11625.python-list@python.org>
	<XWKnOkoTSX9D5gaI8iSioP21b6Iv@4ax.com>
Message-ID: <87r9081ks5.fsf@winnegan.de>

Paul Moore <paul.moore at uk.origin-it.com> writes:

> On Wed, 7 Mar 2001 13:44:11 +0100 , Siggy Brentrup <bsb at winnegan.de> wrote:
> 
> >Paul Moore <paul.moore at uk.origin-it.com> writes:
> >Well, obviously I should have tested before posting :-( There's one
> >omission in my code as well as three typos in urllib2.py (sample and
> >patch attached)
> 
> Hmm. I was using Python 2.0. I switched to Python 2.1, and applied your patch
> (did you send it to SourceForge as a bug report?

Sure, cf Bug #406683 "typos in urllib2"

> If not, would you like me to?) Running your script (with my proxy
> info) I still get an error...

> C:\Data>python21 proxy_auth.py
> Traceback (most recent call last):
>   File "proxy_auth.py", line 18, in ?
>     f = urllib2.urlopen('http://www.python.org/')
>   File "c:\applications\python21\lib\urllib2.py", line 135, in urlopen
>     return _opener.open(url, data)
>   File "c:\applications\python21\lib\urllib2.py", line 318, in open
>     '_open', req)
>   File "c:\applications\python21\lib\urllib2.py", line 297, in _call_chain
>     result = func(*args)
>   File "c:\applications\python21\lib\urllib2.py", line 822, in http_open
>     return self.do_open(httplib.HTTP, req)
>   File "c:\applications\python21\lib\urllib2.py", line 816, in do_open
>     return self.parent.error('http', req, fp, code, msg, hdrs)
>   File "c:\applications\python21\lib\urllib2.py", line 344, in error
>     return self._call_chain(*args)
>   File "c:\applications\python21\lib\urllib2.py", line 297, in _call_chain
>     result = func(*args)
>   File "c:\applications\python21\lib\urllib2.py", line 425, in
> http_error_default
>     raise HTTPError(req.get_full_url(), code, msg, hdrs, fp)
> urllib2.HTTPError: HTTP Error 407: Proxy Authentication Required

> I dumped out a few of the objects, and it looks like the request is
> sending a Proxy-Authorization header to the proxy, with the
> base64-encoded user ID and password in it. Is there any chance my
> proxy doesn't work that way, and requires a different form of
> authorisation? I know nothing much about proxies, but I seem to
> recall seeing "Squid" in some of the proxy messages - is Squid a
> type of proxy????

Yes, I'm running Squid as my proxy too and _I'm_in_control_of_it_.
Your proxy setup might use some other form of authentication, what
urllib2.ProxyHandler uses, looks like "Basic HTTP Authentication" to
me - no time to investigate further now.

> If this is some sort of bizarre peculiarity of my proxy setup, then
> feel free to drop the issue - I can live without getting this to
> work, and I don't want to take up too much of your time with this.

Don't worry, I'll take this as an opportunity to learn about proxy
authentication, I'll look into it this weekend.

> Thanks for all the help you've provided so far - I feel that I at least
> understand the problem much better now...!

You're welcome

  Siggy




From othello at javanet.com  Mon Mar 19 13:38:59 2001
From: othello at javanet.com (Raymond Hettinger)
Date: Mon, 19 Mar 2001 13:38:59 -0500
Subject: PEP 42 and Py2.1
References: <mailman.984978384.13645.python-list@python.org>
Message-ID: <3AB65243.833948E7@javanet.com>

Tim Peters wrote:

> [Raymond Hettinger]
> > I've checked out the Py2.1 beta and my most desired small feature didn't
> > make it.  To quote PEP 42:
> >
> >     Generalize eval() to accept any mapping objects for locals and
> >     globals.
> >
> > Is it too late to get this one in?
>
> Yes; it was too late the day after 2.1b1 was released (the first beta has the
> last new feature).

Bummer.

>
> > Is it difficult to implement?
>
> Afraid that's the wrong question:  resistance has been grounded in "go fast"
> reasons.  Using dicts directly avoids the runtime expense of indirecting thru
> an abstract mapping-object layer.

I'm suprised.  Keeping dictionaries as a special case gives:
       if type(localarg) = type({}):                        # one test in C can't
eat many cycles
            do it the way it's done now
       else:
            do it through mapping object indirection.

> > Is there anyway I can help get it done?
>
> Well, implementing it would be a help!  Then you can quantify the slowdown
> (if any -- who knows?).

I'll take a crack at it.

> personally-doubt-it-would-make-make-enough-difference-to-care-ly

>     y'rs  - tim

Raymond



From stephen_purcell at yahoo.com  Sun Mar  4 10:11:46 2001
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Sun, 4 Mar 2001 16:11:46 +0100
Subject: multipart/form-data submission with httplib + MimeWriter?
In-Reply-To: <3AA1A1E7.1360139A@cs.utwente.nl>; from stadt@cs.utwente.nl on Sun, Mar 04, 2001 at 03:01:11AM +0100
References: <mailman.983531043.10150.python-list@python.org> <3AA1A1E7.1360139A@cs.utwente.nl>
Message-ID: <20010304161146.B2779@freedom.puma-ag.com>

Richard van de Stadt wrote:
> Steve Purcell wrote:
> > Can anyone point me towards some example code for performing
> > 'multipart/form-data' submissions of form values and file contents
> > using httplib and MimeWriter?
> 
> I'm not sure what you mean.
> 
> Are you asking about 
>  - how to *fill in* such webforms automatically, or
>  - how to create a form with which a file can be uploaded, or
>  - how to deal with the contents of such a form when submitted?

None of the above. I meant exactly what I said; making multipart/form-data
POST submissions programmatically using httplib and MimeWriter.

But thanks for the reply anyway.

I figured it out in the end, after some experimentation. The two libraries
don't work well together for various reasons. For intance, MimeWriter can't
provide a content-length, which POST requests require. Fortunately it's
trivial to write multipart MIME messages 'the long way'.

(Background; I'm working on a client-side web testing library that sits on
PyUnit, and which will soon be available at webunit.sourceforge.net)

-Steve

-- 
Steve Purcell, Pythangelist
Get testing at http://pyunit.sourceforge.net/
Get servlets at http://pyserv.sourceforge.net/
"Even snakes are afraid of snakes." -- Steven Wright



From mlh at idi.ntnu.no  Sat Mar  3 17:18:04 2001
From: mlh at idi.ntnu.no (Magnus Lie Hetland)
Date: Sat, 3 Mar 2001 23:18:04 +0100
Subject: Missing the functional mind set
References: <mlb2at8ptrr7sh5dgq8qlk2201kquh6tfl@4ax.com> <uDao6.335972$ge4.119457077@news2.rdc2.tx.home.com> <t6f2at870pk82q96bnd5a2od4n8ns5kf83@4ax.com> <ntg2atcemm3hp9rb26hn7c5uin4dfn8lc9@4ax.com>
Message-ID: <97rqgb$lt2$1@tyfon.itea.ntnu.no>

"Kirby Urner" <urner at alumni.princeton.edu> wrote in message
news:ntg2atcemm3hp9rb26hn7c5uin4dfn8lc9 at 4ax.com...
> Daniel Klein <danielk at aracnet.com> wrote:
[snip]
> you could go:
>
>   >>> params = [None]*3 # list

This should be

    >>> params = [None]*4

>   >>> params[2] = uv1   # uservalue for param 2
>   >>> params[3] = uv2
>   >>> params[1] = uv3
>   >>> params[0] = uv4
>
[...]
> Kirby
>
--

  Magnus Lie Hetland          http://www.hetland.org

 "Reality is what refuses to disappear when you stop
  believing in it"                 -- Philip K. Dick





From thomas.heller at ion-tof.com  Fri Mar  2 08:02:32 2001
From: thomas.heller at ion-tof.com (Thomas Heller)
Date: Fri, 2 Mar 2001 14:02:32 +0100
Subject: Unresolved symbols in MSVCRT.DLL when running a py2exe packaged program.
References: <RKln6.6630$hn5.853203@news1.rdc1.mb.home.com> <hVBn6.7888$UZ4.3012833@news4.rdc1.on.home.com> <VNGn6.7720$hn5.948267@news1.rdc1.mb.home.com>
Message-ID: <97o5m8$q83kq$1@ID-59885.news.dfncis.de>

Here is a comment about a similar problem in
another py2exe-packed application, which
I received from Jeff Kunce:

>>I just tried runing a py2exe-packaged application on an older NT machine, and
>>it choked with an "Entry Point Not Found" error on MSVCRT.dll
>
>Well, I played with it some more and got it to work.
>
>I simply excluded all the DLLs that are normally found in NTs system32
>directory (MPR.dll, MSVCIRT.dll, MSVCRT.dll, NETAPI32.dll, NETRAP.dll,
>SAMLIB.dll) - and it works now on the older machine.
>
>It looks like there is no easy "good" solution to this problem. Here's a
>comment by Gordon McMillan on the topic:
>  http://www.wxpython.org/archive/2000-June/002469.html

Thomas




From tchur at optushome.com.au  Sat Mar 24 21:15:19 2001
From: tchur at optushome.com.au (Tim Churches)
Date: Sun, 25 Mar 2001 12:15:19 +1000
Subject: Referencing a container object's attributes
Message-ID: <3ABD54B7.8CDD2175@optushome.com.au>

Is there any way for a method of a class instance which is embedded in a
container class instance to look up attributes of that container class
instance (i.e. the one which contains it)? At the moment I am putting
the name of the container instance in an attribute of the containee (at
the time when the containee is instantiated). Is there a neater way? All
the Python texts cover referencing an object's superclass(es), but not
its container object(s).

Tim C



From jmarshal at mathworks.com  Thu Mar  1 22:35:15 2001
From: jmarshal at mathworks.com (Joshua Marshall)
Date: 2 Mar 2001 03:35:15 GMT
Subject: define a new func on the fly?
References: <3A9EE927.8744154E@troikanetworks.com> <o6Cn6.846$45.1115@www.newsranger.com> <3a9f0bc2.10059041@News.CIS.DFN.DE>
Message-ID: <97n4dj$nro$1@news.mathworks.com>

Costas Menico <costas at meezon.com> wrote:
> Gregory Jorgensen<gregj at pobox.com> wrote:

>>eval() evaluates an expression. exec() executes a statement. Try this:
>>
>>>>> exec("def f():\n\treturn 'hello'\t")

> When I try the above I get a syntax error... I am using PythonWin. Is
> this a bug?

> Costas


>>>>> f()
>>'hello'
>>
>>>>> eval("f()")
>>'hello'

exec seems to want a newline at the end:

  >>> exec("def f():\n\treturn 'hello'\t")
  Traceback (most recent call last):
    File "<stdin>", line 1, in ?
    File "<string>", line 2
      return 'hello'
                   ^
  SyntaxError: invalid syntax

  >>> exec("def f():\n\treturn 'hello'\n")
  >>> f()
  'hello'


From ransen_spam_me_not at nemo.it  Sun Mar  4 03:39:15 2001
From: ransen_spam_me_not at nemo.it (Owen F. Ransen)
Date: Sun, 04 Mar 2001 08:39:15 GMT
Subject: Python 2.1b1 released - debug versions?
References: <mailman.983573701.16010.clpa-moderators@python.org> <3aa1cb88.4366424@news.newsguy.com>
Message-ID: <3aa1fe2a.196614@news.newsguy.com>

On Sat, 03 Mar 2001 11:12:28 GMT, ransen_spam_me_not at nemo.it (Owen F.
Ransen) wrote:


https://sourceforge.net/project/showfiles.php?group_id=5470


>
>>Find out all about 2.1b1, including docs and downloads (Windows
>>installer and source tarball), at the 2.1 release page:
>>    http://www.python.org/2.1/
>
>I was hoping to try out Python 2.1 in order to see if
>it fixed the Py_Finalize() bug and the PyRun_SimpleFile()
>crash, but when I downloaded the windows installer
>Python-2.1b1.exe I can't find the debug library and
>dll, which I suppose would be called:
>
>  python21_d.lib
>  python21_d.dll
>
>Any pointers anyone? I probably do not yet know my way
>around Python releases, please bare with me!

--
Owen F. Ransen
http://www.ransen.com/
Home of Gliftic & Repligator Image Generators


From calcium at ozemail.com.au  Sat Mar 24 09:17:56 2001
From: calcium at ozemail.com.au (ccang)
Date: Sun, 25 Mar 2001 01:17:56 +1100
Subject: Python script to create HTML gallery, thumbnails
Message-ID: <3ABCAC94.B095CB9A@ozemail.com.au>

Hi there all,
I recently got a digital camera (Kodak DC4800) and wanted to
put all the crappy photos I shoot on the web.

I looked for software to create thumbnails and corresponding
HTML files but couldnt find any so I wrote my own using Python
and the PIL.

It creates a consolidated thumbnail and the corresponding HTML files.

No GUI yet so u'll have to edit the script to customize
your settings eg page headers, comments.

# Does only JPGs for now.
# Usage : python mkgallery.py [ filenames ]
#   Will produce 1 HTML file for the thumbnail index.
#                1 HTML file per JPG image.
#                1 consolidated thumbnail JPG.
#
# There will be GUI for this pretty soon.
#
# The HTML file for the thumbnail is generated in the same
# directory as where the script is run from.
# However, it assumes that it will reside
# in a subdirectory above where the other files (HTML, JPGS) reside.
# ie u'll have to move it.
#
# eg
#   myGallery/
#   myGallery/topic.html
#   myGallery/topic/topic_index.jpg
#   myGallery/topic/zpic1.jpg
#   myGallery/topic/zpic1.html
#       ...
#       ...
#   myGallery/topic/zpicN.jpg
#   myGallery/topic/zpicN.html
#

http://www.ozemail.com.au/~calcium/scripts.html

Sample output
http://www.ozemail.com.au/~calcium/rogaine.html

Feedback welcome,
Chai
--
| mailto:calcium at ozemail.com.au   | +61 401 688 408
|  http://www.ozemail.com.au/~calcium
|  I dont use the voice of Bart when I'm making love
|  to my husband, but Marge's voice turns him on a
|  little - Nancy Cartwright.



From gervaserules at yahoo.com  Wed Mar 14 19:06:59 2001
From: gervaserules at yahoo.com (Nathan Heagy)
Date: Thu, 15 Mar 2001 00:06:59 GMT
Subject: Console column Width
References: <hhAr6.3889$L7.27822@news1.sshe1.sk.home.com> <m3snkg7n54.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <DITr6.4002$L7.30409@news1.sshe1.sk.home.com>

Yes, this does work, except on MacOSX, which appears to not have TIOCGWINSZ
in its TERMIOS module. Is this true for FreeBSD/other BSDs?

nathan


"Michael Hudson" <mwh21 at cam.ac.uk> wrote in message
news:m3snkg7n54.fsf at atrus.jesus.cam.ac.uk...
> "Nathan Heagy" <gervaserules at yahoo.com> writes:
>
> > I can't figure out how to learn the (linux) console's column width from
> > within a script? That is, without using curses.
>
> This works for me:
>
> def getwidth():
>     return getheightwidth()[1]
>
> # Eep!  this next should be made rather more portable (ie. fallback to
> # env vars, then tigetstr("lines"), etc).
> def getheightwidth():
>     height, width = struct.unpack(
>         "hhhh", ioctl(0, TERMIOS.TIOCGWINSZ ,"\000"*8))[0:2]
>     return height, width
>
> Cheers,
> M.
>
> --
>   Not only does the English Language borrow words from other
>   languages, it sometimes chases them down dark alleys, hits
>   them over the head, and goes through their pockets.   -- Eddy Peters




From tdelaney at avaya.com  Tue Mar 20 19:06:59 2001
From: tdelaney at avaya.com (Delaney, Timothy)
Date: Wed, 21 Mar 2001 11:06:59 +1100
Subject: New PEP: The directive statement
Message-ID: <B43D149A9AB2D411971300B0D03D7E8B0FBFB8@natasha.auslabs.avaya.com>

There are however, two things against this PEP.

1. It allows for one incompatible language change (directive as keyword can
interfere with binding to name "directive" as the first line of a file).

2. It provides no method for retrieving version information about
transitional features.

(1) may in practice turn out to be a non-issue, but I think for people who
will actually be using this functionality, (2) will be vital.

Tim Delaney
Avaya Australia
+61 2 9352 9079

All recipient(s) of this email have permission to forward or reply to this
email, quoting this email in full or in part.


> -----Original Message-----
> From: Timothy Grant [mailto:tjg at exceptionalminds.com]
> Sent: Wednesday, 21 March 2001 3:53 AM
> To: python-list at python.org
> Subject: Re: New PEP: The directive statement
> 
> 
> I am not one of the gods of Python, but I'm not Python
> illiterate either. 
> 
> When I read PEP 236. I cringed. I understand the value, but the
> solution seemed far from elegant, and in Python, most things
> seem elegant (except for lambda<wink>.
> 
> When I read the below suggestion it made far more sense to me,
> and seemed to be a more elegant solution to the problem.
> 
> 
> On Tue, Mar 20, 2001 at 11:42:34AM +0100, Martin von Loewis wrote:
> > Please have a look at the PEP below. It is intended to replace the
> > __future__ import suggested in PEP 236.
> > 
> > Regards,
> > Martin
> > 
> > PEP: XXX
> > Title: The directive statement
> > Version: $Revision$
> > Author: Martin v. L?wis
> > Python-Version: 2.1
> > Status: Active
> > Type: Standards Track
> > Created: 20-Mar-2001
> 
> -- 
> Stand Fast,
>     tjg.
> 
> Timothy Grant                         tjg at exceptionalminds.com
> Red Hat Certified Engineer            www.exceptionalminds.com
> Avalon Technology Group, Inc.         <><       (503) 246-3630
> >>>>>>>>>>>>>Linux, because rebooting is *NOT* normal<<<<<<<<<
> >>>>This machine was last rebooted:  62 days 21:05 hours ago<<
> 
> -- 
> http://mail.python.org/mailman/listinfo/python-list
> 



From Barrett at STScI.Edu  Wed Mar 28 09:11:09 2001
From: Barrett at STScI.Edu (Paul Barrett)
Date: Wed, 28 Mar 2001 09:11:09 -0500
Subject: Biggest float number?
References: <3ABFBB88.E718B2D@nd.edu>
Message-ID: <3AC1F0FD.402EA573@STScI.Edu>

"C. S. Xu" wrote:
> 
> It seems the biggest floating number in Python can be:
>     2.**1024 - 1.
> which is about 1.8E308. Is there any way to get bigger value than this,
> just like long intergers almost have no limits?

Try using a log()-based algorithm, i.e. instead of multiplying two large
numbers together, take their log() and add them.

This approach has always worked for me, since I've found that precision
is usually less of an issue than range, when dealing with such large
numbers. 

-- 
Dr. Paul Barrett       Space Telescope Science Institute
Phone: 410-338-4475    ESS/Science Software Group
FAX:   410-338-4767    Baltimore, MD 21218


From donn at u.washington.edu  Tue Mar  6 14:36:38 2001
From: donn at u.washington.edu (Donn Cave)
Date: 6 Mar 2001 19:36:38 GMT
Subject: I come to praise .join, not to bury it...
References: <mailman.983890030.6303.python-list@python.org> <983894109.2077137603@news.silcom.com> <jwbnews-AA1BA7.09391406032001@corp.supernews.com>
Message-ID: <983e86$ln6$1@nntp6.u.washington.edu>

Quoth "John W. Baxter" <jwbnews at scandaroon.com>:
| With luck and good will, we ought not have so many of this sort of 
| discussion after the fact in the future.
|
| We now have the PEP process...and the discussion of the PEP is the right 
| place for alarms to be raised, both technical ("that fails when...") and 
| otherwise ("I'd rather see this than that.").

Actually I think that may be representative of a misunderstanding about
what's going on here.  I don't think I have seen a many posts in this
branch of threads that took the position, in so many words, that the
features in question shouldn't have been implemented in Python.  The
ones that have, have taken kind of surprising angles on it, like the
idea that string functions might profitably be omitted from a special
purpose small subset interpreter.  There have been plenty of criticism,
but speaking for myself, if people see a categoric rejection of the
feature there, they're reading something into my words that wasn't in
my head.  That is an easy mistake to make, it's just simplistic.

I would expect that if these same people were herded into a room and
subjected to a PEP discussion, they would neither raise technical
alarm nor suggest better alternatives, and not for lack of imagination
but simply because these features aren't that in kind of trouble.
They more or less make sense, and they're here for more or less valid
reasons.  Yet afterwards, you will read "this is line noise."  Not
because the PEP process failed, it just doesn't address the question
of how to program in Python.  We ought to recognize this landscape
pretty well, because we have seen the Perlites walking through it all
along.  They're acutely conscious of the difference between what the
language can be and what people make of it.  As Python gets richer,
the same forces come to bear on it.  I don't know if we have a choice
to not get richer, it's probably a categorical imperative that is as
silly to resist as the ensuing entropy.  But we can and do occasionally
turn away from a few features that come along, they don't all get into
Python and we always have the choice of how to write our own software.
We don't have a huge problem yet, but it's good to face up to what
little problems we do have.

The threads can get pretty tedious, and it sure is not obvious that
it's a good way to solve any problem.  There is an awful lot of talking
at cross purposes, particularly if you accept my premise that none of
us want to actually repeal the features in question.  But I think there's
some good in it anyway.  Only this morning, when I thought everything
must have been said several times over, someone made a point about
methods and literals, one of those insights that are so obvious in
retrospect I'm kind of embarrassed to admit, which is the sign of a
good point.  I don't want to suppress that kind of thing, because it
helps me think about what I can do to write well.

	Donn Cave, donn at u.washington.edu


From cce at clarkevans.com  Thu Mar  8 04:25:36 2001
From: cce at clarkevans.com (Clark C. Evans)
Date: Thu, 8 Mar 2001 04:25:36 -0500 (EST)
Subject: win32com:MakePy -- Early Binding without the size jitters?
In-Reply-To: <UPHp6.38461$1D5.1679137@e420r-atl1.usenetserver.com>
Message-ID: <Pine.LNX.4.21.0103080418530.13020-100000@clarkevans.com>

Hello.  I'd like to use early-binding with about three COM 
libraries, each one of the stub.pyc files over 100KB (with 
one in particular over 300KB).  However, I'm only using
about a dozen methods (including constructors, etc) from
each library.  I was wondering... is it safe to delete
methods/interfaces which I'm not using?  I've deleted
a few, and it still works for my purposes, so I was 
wondering if I'm shooting myself in the foot -- or if
there is a methodological approach to this pruning?
It'd be cool to get the .pyc alot smaller; no use
toossing away 1MB of memory when I only need a few K...

Thanks for any ideas you win32com experts may have.

Best,

Clark 

P.S.  One of the COM modules is MSXML3, and for other platforms,
is there a standard way to wrap XSLT processors?  Thanks!




From not.this at seebelow.org  Wed Mar  7 14:12:30 2001
From: not.this at seebelow.org (Grant Griffin)
Date: 7 Mar 2001 11:12:30 -0800
Subject: Can anyone recomend a good intoduction to C...
References: <mailman.983894288.19965.python-list@python.org> <983896344.701456@newsmaster-04.atnet.at> <m3lmqibw0a.fsf@atrus.jesus.cam.ac.uk> <983911749.217622@newsmaster-04.atnet.at> <UScp6.1097$y6.196347@ruti.visi.com> <983916572.286918@newsmaster-04.atnet.at> <9852b8015jm@news1.newsguy.com> <983967284.246700@newsmaster-04.atnet.at> <o9up6.1291$y6.235469@ruti.visi.com>
Message-ID: <98616u01pba@drn.newsguy.com>

In article <o9up6.1291$y6.235469 at ruti.visi.com>, grante at visi.com says...
>
>In article <983967284.246700 at newsmaster-04.atnet.at>, Werner Schiendl wrote:
>
>>> > That is of course a point, if you need to let the program
>>> > inspect you will try to keep it as simple as possible. But
>>> > given the same functionality, I think a C++ program will not be
>>> > more complex.
>>>
>>> The program will not be, the language is.  It's a hard-to-judge
>>> tradeoff.
>>
>>As to my knowledge, the application itself is inspected for
>>safety critical appliances. And you must not change anything
>>afterwards in the software or you need to re-inspect the
>>software.
>>
>>What use would it be to check the language?
>
>The point is that the more complex the language, the more
>difficult it is to inspect the application.  You can look at a
>few lines of C and have a pretty good chance at guessing what
>they do.  In C++ it's much harder to look at snippet of code
>and figure out what it's going to do.

To me, that's more an indictment of OO programming than C++.  OO programming is 
useful primarily for creating complex programs--which, by definition, are hard
to figure out in just a few lines.  Given familiarity with OO concepts, and
familiarity with one's favorite OO language, the exact syntax of the language
doesn't matter all that much.  Although Python's approach to OO certainly is
simpler and cleaner than C++'s, I don't find much difference in the difficulty
of reading OO code in the two; instead, most of the difficulty of reading OO
code boils down to understanding the underlying algorithms and data structures.

OTOH, the more semantic "punch" a language offers, the harder it can be to read,
per line.  My experience with Python has been that I generally write fewer lines
of code than I would in C or C++, but my code takes more effort per line to
write.  There certainly is a net gain, but it isn't an order of magnitude.  Or,
as Fred Brooks puts it, "There's no magic bullet": most of the hard work of
programming will remain, no matter what languages or tools we use.

though-python-makes-for-a-better-aesthetic-experience-ly y'rs,

=g2

_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation            http://www.iowegian.com



From wfocke at phoenixdsl.com  Tue Mar  6 20:30:20 2001
From: wfocke at phoenixdsl.com (Warren Focke)
Date: 6 Mar 2001 17:30:20 -0800
Subject: How to installing NumPy?
References: <3AA57C0D.BBC5E956@itu.edu.tr>
Message-ID: <9842vc$4bb$1@mothership.telocity.com>

In article <3AA57C0D.BBC5E956 at itu.edu.tr>,
Nuzhet Dalfes wrote:
>I am trying to install NumPy on a RedHat 7.0 running python2.0.
>Everything is in the standard places (I think!) and I picked up and
>unzipped http://download.sourceforge.net/numpy/NumPy1711_20.zip into

That's for windos.  You want to get Numeric-17.3.0.tar.gz, unpack it
(not in site-packages, maybe in /usr/local/src or ~/src), cd to the
directory it made, and follow the instructions there.  Probably
something like:

/usr/local/bin/python2.0 setup.py install
/usr/local/bin/python2.0 setup_all.py install

Warren Focke
-- 
Wrapping up her appearance with rubber, you can feel more comfortable.
 -- directions for a set of Chinese binoculars


From johngrayson at home.com  Wed Mar 28 14:52:15 2001
From: johngrayson at home.com (johngrayson at home.com)
Date: Wed, 28 Mar 2001 19:52:15 -0000
Subject: The Scale Widget
Message-ID: <99tfdf+slqc@eGroups.com>

Hans Kristian Ruud wrote:

> Is there a way to modify the scale widget so that the slider
> only can move in one direction?

Here's a programatic solution:

from Tkinter import *
import string

global scale, position

position = 100

def setHeight(canvas, heightStr):
    global position
    height = string.atoi(heightStr)

    if height < position:
        height = position
	if height > 250: height = 250
	position = height
	scale.set(position)

    height = height + 1
    position = height
    y2 = height - 30
    if y2 < 1:
        y2 = 1
    canvas.coords('poly',
                 15,20,35,20,35,y2,45,y2,25,height,5,y2,15,y2,15,20)
    canvas.coords('line',
                 15,20,35,20,35,y2,45,y2,25,height,5,y2,15,y2,15,20)

root = Tk()
root.title('Scale')

canvas = Canvas(root, width=50, height=50, bd=0, highlightthickness=0)
canvas.create_polygon(0,0,1,1,2,2, fill='cadetblue', tags='poly')
canvas.create_line(0,0,1,1,2,2,0,0, fill='black', tags='line')
scale = Scale(root, orient=VERTICAL, length=284, from_=0, to=250,
              tickinterval=50, command=lambda h,
              c=canvas:setHeight(c,h))
scale.grid(row=0, column=0, sticky='NE')
canvas.grid(row=0, column=1, sticky='NWSE')
scale.set(position)
root.mainloop()






From hamish_lawson at yahoo.co.uk  Sat Mar 10 16:35:20 2001
From: hamish_lawson at yahoo.co.uk (=?iso-8859-1?q?Hamish=20Lawson?=)
Date: Sat, 10 Mar 2001 21:35:20 +0000 (GMT)
Subject: Python In A Nutshell
Message-ID: <20010310213520.20774.qmail@web210.mail.yahoo.com>

I think "Perl in a Nutshell" serves as a good model: fast-track
tutorial on the language, reference on standard modules, and a
reference on the more common third-party modules.

In the standard modules section I'd hope to see nearly all the modules
other than perhaps those that are obsolescent, platform-specific, or
somewhat esoteric; one of the things I miss in David Beazley's "Python
Essential Reference" is coverage on htmllib, sgmllib and telnetlib. I'd
like to see some brief documentation of Tkinter and more extended
coverage of PyXML than is given in the online library reference.

For the third-party modules I'd nominate the following, grouped by
desirability of inclusion:

    ESSENTIAL
       DB-API
       mxDateTime
       Python Imaging Library
       pyunit
       win32all

    GOOD
       HTMLgen
       mxTextTools
       ReportLab
       wxPython
       xmlrpc
       soaplib

    NICE
       mod_python / mod_snake
       Numerical Python
       freeze/packaging


Hamish Lawson


____________________________________________________________
Do You Yahoo!?
Get your free @yahoo.co.uk address at http://mail.yahoo.co.uk
or your free @yahoo.ie address at http://mail.yahoo.ie



From robin at stop.spam.alldunn.com  Tue Mar 27 17:04:26 2001
From: robin at stop.spam.alldunn.com (Robin Dunn)
Date: Tue, 27 Mar 2001 14:04:26 -0800
Subject: wxPython: How to get a child by name?
References: <99qpce$ljk$1@newsreaderg1.core.theplanet.net>
Message-ID: <w88w6.971$NAC.9240731@news.randori.com>

"Franz GEIGER" <fgeiger at datec.at> wrote in message
news:99qpce$ljk$1 at newsreaderg1.core.theplanet.net...
> I currently try to polish some of my apps with a GUI and decided to go for
> wxPython.
>
> I use wxDesigner for the layout. As it generates code after every change,
I
> named all the text controls and checkboxes etc. of the dialog hosting all
> those controls by applying SetName() to the controls. The Dialog is small,
> so I do not use validators yet.
>
> To get a text control of the dialog I thought I could write
> theValue = self.FindWindowByName("tcSurname").GetValue().
> This raises an exception, because the return value is of type wxWindowPtr
> which does not have such a method.

...

> For a "quick" hack in a MSVC project I would downcast the returned pointer
> of GetDlgCtrl() to be able to call the subclassed object's methods. But
how
> do I proceed here?


You're on the right track, you're looking for wxPyTypeCast which takes a
shadow object and converts the swigified pointer within it to a different
type and wraps a new shadow object around it of the new type.  There are
some comments about it in wx.py.  (I'm making some changes soon that will
hopefully let me finally turn this hack into a no-op...)

But since you are using wxDesigner you could have saved yourself a bunch of
work and let it generate some "getters" for you.  Click on the Add Python
Source button to add a new sourcefile to your project, then double click the
filename under Python sources in the tree view to open the file.  Once there
you can add classes that create and use the specific forms you've designed,
and also add to the class event handlers and getters for the fields on the
form.  I usually end up changing a lot of the code that it generates for the
__init__ method but that's okay since it never regenerates this file, only
adds to it.  For the getters it makes code like this:

    def GetLoginBitmap(self):
        return wxPyTypeCast( self.FindWindowById(ID_LOGINBITMAP),
"wxStaticBitmap" )

    def GetPasswd(self):
        return wxPyTypeCast( self.FindWindowById(ID_PASSWD), "wxTextCtrl" )

    def GetUsername(self):
        return wxPyTypeCast( self.FindWindowById(ID_USERNAME),
"wxTextCtrl" )

    def GetURL(self):
        return wxPyTypeCast( self.FindWindowById(ID_URL), "wxComboBox" )

You just need to be sure that you give each control a unique ID name.  Once
you have the getters like this you can write code like:

    self.GetUsername().GetValue()


--
Robin Dunn
Software Craftsman
robin at AllDunn.com       Java give you jitters?
http://wxPython.org      Relax with wxPython!






From ransen_spam_me_not at nemo.it  Thu Mar  8 10:01:13 2001
From: ransen_spam_me_not at nemo.it (Owen F. Ransen)
Date: Thu, 08 Mar 2001 15:01:13 GMT
Subject: Memory leaks in VC6 2 line Python 2.1b1 apps?
Message-ID: <3aa88f52.5419185@news.newsguy.com>

Python 2.1b1

I tried to create a simple VC MFC app using the 
wizard as follows:

1) Multidoc shared MFC-DLL application.
2) Run time library debug multithreaded DLL
3) Added #include <python.h> to the main file
4) Added
        Py_Initialize()
        Py_Finalize()
   at the end of the InitInstance()
5) When I run I get memory leaks visible at
   exit.

Any clues anyone?

(I can't yet report this via Source Forge because my account
 problems have not be sorted out yet.)


--
Owen F. Ransen
http://www.ransen.com/
Home of Gliftic & Repligator Image Generators


From klin at palsoccer.com  Wed Mar 21 17:46:54 2001
From: klin at palsoccer.com (Kevin X Lin)
Date: Wed, 21 Mar 2001 17:46:54 -0500
Subject: socket programming
Message-ID: <99bb0g0nce@enews3.newsguy.com>

Hi there, I have the following problem with socket programming:

I'm trying to setup a tcp client to talk to a tcp server sitting at a
certain port. When I issue :

mysocket.setblocking(1)
mysocket.send('something')
mysocket.recv(1024)

I was expecting I can receive all the data sent by server. Unfortunately,
this is not necessary the case. Sometimes I can't get all of data, worse,
the application protocal doesn't define end tag of the data, which prevent
me from doing a loop until I receive all data.

But if I do this:

mysocket.setblocking(1)
mysocket.send('something')
sleep(0.1)
mysocket.recv(1024)

I can get all data, yet I believe this is not a good solution which might
casusing other problems in the future.

Any suggestion?

Thanks.

Kevin






From aleaxit at yahoo.com  Thu Mar 15 04:13:06 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 15 Mar 2001 10:13:06 +0100
Subject: Better string.translate?
References: <3aaf758f$1_2@news.zet.net> <mailman.984590438.506.python-list@python.org> <3ab0749f$1_3@news.zet.net>
Message-ID: <98q6rm0uoa@news2.newsguy.com>

"Stowasser Harald" <stowasser.h at idowa.de> wrote in message
news:3ab0749f$1_3 at news.zet.net...
> "Steve Purcell" <stephen_purcell at yahoo.com> schrieb :
> > Use the 'sub' function in the 're' module:
>
> Thank you, but i don't want do delete the characters. I need to
> translate them into Spaces.
>
> Anybody knows how?

This was Steve's suggestion for removal:

>> import re
>> re.sub('[^abcd]','', "string from which to remove all but a, b, c or d")
'cababcd'


And this will change-into-space instead:

>> import re
>> re.sub('[^abcd]',' ', "string from which to remove all but a, b, c or d")
'               c            a   b   a  b  c    d'

See the difference?  It's all in the second parameter to re.sub --
*what* you want to change each matched-substring to.  Change it
to an empty-string, it's deleted from the result; change it to a
space, it's changed into a space.


Alex







From bbollenbach at homenospam.com  Wed Mar  7 01:50:49 2001
From: bbollenbach at homenospam.com (Brad Bollenbach)
Date: Wed, 07 Mar 2001 06:50:49 GMT
Subject: The in opperator -- How do I use it?
References: <983945437.96921@atlas.uniserve.ca>
Message-ID: <dTkp6.13293$hn5.1934677@news1.rdc1.mb.home.com>

Think of the "in" operator as the list membership operator...because that's
what it is. :)

So if you want to know "if 3 is in [4, 5, 6]", then:

if 3 in [4, 5, 6]:
    print "it is."
else:
    print "it isn't."

it isn't.

In other words
"Mark" <mmoon at apexmail.com> wrote in message
news:983945437.96921 at atlas.uniserve.ca...
> How do I use the 'in' operator?
> Currently I have having some difficulty in getting it to work
> Mark
> mmoon at apexmail.com
>
>




From ws-news at gmx.at  Thu Mar  8 05:21:46 2001
From: ws-news at gmx.at (Werner Schiendl)
Date: Thu, 8 Mar 2001 11:21:46 +0100
Subject: "!=" is bad form. Re: sorry....never mind
References: <DZkp6.18811$pF4.3015652@news1.telusplanet.net> <fdlp6.16982$go.2157953@news0.telusplanet.net> <3AA6E308.5B63ECC6@cosc.canterbury.ac.nz> <3AA72253.FB021815@Lugoj.Com> <slrn9aehgr.33d.tim@vegeta.ath.cx>
Message-ID: <984047188.296438@newsmaster-04.atnet.at>

> I hate '<>' because it's too much like VB (and so many other BASIC
> variants).
>

I think there are at least the same number of programmers that hate the
other way because it looks to much like C.
Maybe we should invent some completely new operator, ending up with line
noise one day... :-]

regards
werner




From mlh at idi.ntnu.no  Sat Mar 24 17:26:44 2001
From: mlh at idi.ntnu.no (Magnus Lie Hetland)
Date: Sat, 24 Mar 2001 23:26:44 +0100
Subject: newbe   how to read in a string?
References: <3ABB4E70.54885C9F@noos.fr> <mailman.985355670.6600.python-list@python.org> <99g1ac02fug@news2.newsguy.com> <m366gz44g1.fsf@gondolin.beleriand>
Message-ID: <99j6u0$6cv$1@tyfon.itea.ntnu.no>

"Steffen Ries" <steffen.ries at sympatico.ca> wrote in message
news:m366gz44g1.fsf at gondolin.beleriand...
> "Alex Martelli" <aleaxit at yahoo.com> writes:
>
[snip]
> >
> > The 'old way' would be
> >     seq = map(float, string.split(a, ','))
> > and it still works fine of course, though I'd rather spell it
> >     seq = map(float, ','.split(a))
>
> You're (ab)using ",".join(seq) too much, it seems :-)
>
> What you wanted to say is: "seq = map(float, a.split(','))" (not
> counting the conversion problem of the first element...)
>

And I thought Alex found this syntax sooo logical and intuitive?

<wink, duck & cover>

:)

--

  Magnus Lie Hetland         http://www.hetland.org

 "Reality is that which, when you stop believing in
  it, doesn't go away."           -- Philip K. Dick





From sholden at holdenweb.com  Thu Mar  1 12:44:16 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 1 Mar 2001 12:44:16 -0500
Subject: Newbie createing .PYC for import
References: <3a9e83fa$1@news1.dbtech.net>
Message-ID: <kXvn6.10463$jv4.500125@e420r-atl2.usenetserver.com>

"Larry Bates" <lbates at syscon-computers.com> wrote in message
news:3a9e83fa$1 at news1.dbtech.net...
> I have spent the last couple of hours (unsuccessfully) attempting to
figure
> out how to generate a .PYC file for a module that I've defined.
Everything
> appears to imply that it's "automatic", but I'm not getting anywhere.  I
> have a module and I wish to generate a .PYC file so that I can import it
> into future projects.  Can anyone shed any light on how to accomplish
this?
>
> -Larry
>
When a module is imported for the first time (or when the source is more
recent than the current compiled file) a .pyc fileshould be created, in the
same directory as the .py file, containing the compiled code.

If your .pyc's aren't being created, this is probably a permissions problem
with the directory, or similar. This can happen, for example, if you develop
as one user but run as another, such as if you are testing with a web
server.

regards
 Steve





From urner at alumni.princeton.edu  Sat Mar  3 01:41:20 2001
From: urner at alumni.princeton.edu (Kirby Urner)
Date: Fri, 02 Mar 2001 22:41:20 -0800
Subject: Thinking Pythonically (was Re: gah! I hate the new string  syntax)
References: <mailman.983485923.1136.python-list@python.org> <B7In6.16281$Qb7.2658902@newsb.telia.net> <97ohh2$polt0$1@ID-77331.news.dfncis.de> <97opsb$nhu$1@nntp6.u.washington.edu> <97oupq$qijrh$1@ID-77331.news.dfncis.de> <97p3mh$bkm$1@tyfon.itea.ntnu.no> <B6C58CBD.212F%f8dy@diveintopython.org> <97q14l$jbi$0@216.39.151.169>
Message-ID: <7k31at0hbkuqgjtnsfts9nieskove1nli4@4ax.com>

Donn Cave <donn at oz.net> wrote:

>Quoth Mark Pilgrim <f8dy at diveintopython.org>:

>know, I guess it's not for everyone, but for myself, I want to
>challenge your notion of what it is to think Pythonically.
>
>	Donn Cave, donn at oz.net

Personally speaking, I found Pilgrim's defense of list 
comprehensions cogent, and I think his 'Dive Into Python' 
is pretty good.  So maybe he should have used {}.items() 
more often -- I bet he will in future.

I'm another one of those who did NOT come to Python from an 
extensive career in UNIX and related paraphernalia (awk yak 
gak...).  I've got Linux in a partition and visit it once in 
a while -- glad it's there, happy to see that world expanding 
by leaps and bounds.

My first exposure to computing was via the mainframe in college,
an IBM 360/370 VM with APL terminals scattered around campus.
In my psyche, Python is a lot like APL, but with the little
geeky greek-looking symbols replaced by more ordinary-looking
characterstrings.

I've studied Perl to a beginner level, but I think it's somewhat
ugly to look at.  Whereas I'm sure its crypto-compression 
attracts a lot of people, I find Python better because cleaner
somehow (simpler) -- dunno if that's what you were getting at
(your opinion sounds more expert/refined).  But that in no way 
means I think it's a "language war" with Perl -- I hope Perl 
sticks around forever, like Python will.

Pilgrim's use of list comprehensions wherever suitable doesn't
seem a dive into an arcane or cryptic style.  The 'map' thing
was cumbersome, because if your function had two or more 
parameters, but you wanted to hold one or more of them constant, 
you had do to things like:

  >>> def f(a,b): return a+b

  >>> f(1,1)
  2
  >>> g = lambda x: f(1,x)
  >>> map(g,[1,2,3,4])  # or collapse this + above into one line
  [2, 3, 4, 5]

or

  map(f,[1]*4,[1,2,3,4])
  [2, 3, 4, 5]

But with list comprehension, you just go:

  [f(1,x) for x in [1,2,3,4]]
  [2, 3, 4, 5]

That's a lot less "beating around the bush" to get that one
argument to hold at 1, letting the other vary.

Kirby






From paulp at ActiveState.com  Sun Mar 11 18:23:07 2001
From: paulp at ActiveState.com (Paul Prescod)
Date: Sun, 11 Mar 2001 15:23:07 -0800
Subject: Revive the types sig?
Message-ID: <3AAC08DB.9D4E96B4@ActiveState.com>

I have been involved with the types-sig for a long time and it has
consumed countless hours out of the lives of many brilliant people. I
strongly believe that it will only ever work if we change some of
fundamental assumptions, goals and procedures. At next year's
conference, I do not want to be at the same place in the discussion that
we were this year, and last year, and the year before. The last time I
thought I could make progress through sheer effort. All that did was
burn me out and stress out my wife. We've got to work smarter, not
harder.

The first thing we need to adjust is our terminology and goals. I think
that we should design a *parameter type annotation* system that will
lead directly to better error checking *at runtime*, better
documentation, better development environments an so forth. Checking
types *at compile time* should be considered a tools issue that can be
solved by separate tools. I'm not going to say that Python will NEVER
have a static type checking system but I would say that that shouldn't
be a primary goal.

I've reversed my opinion on this issue. Hey, even Guido makes mistakes.

I think that if the types-sig is going to come up with something
useful this time, we must observe a few principles that have proven
useful in developing Python:

1. Incremental development is okay. You do not need the end-goal in
mind before you begin work. Python today is very different than it was
when it was first developed (not as radically different than some
languages, but still different).

2. It is not necessary to get everything right. Python has some warts.
Some are easier to remove than others but they can all be removed
eventually. We have to get a type system done, test it out, and then
maybe we have to remove the warts. We may not design a perfect gem from
the start. Perfection is a goal, not a requirement.

3. Whatever feature you believe is absolutely necessary to a decent
type system probably is not. There are no right or wrong answers,
only features that work better or worse than other features.

It is important to understand that a dynamically-checked type
annotation system is just a replacement for assertions. Anything that
cannot be expressed in the type system CAN be expressed through
assertions.

For instance one person might claim that the type system needs to
differentiate between 32 bit integers and 64 bit integers. But if we
do not allow that differentiation directly in the type system, they
could do that in assertions. C'est la vie.

This is not unique to Python.  Languages like C++ and Java also have
type test and type assertion operators to "work around" the
limitations of their type systems. If people who have spent their
entire lives inventing static type checking systems cannot come up
with systems that are 100% "complete" then we in the Python world
should not even try. There is nothing wrong with using assertions for
advanced type checks. 

For instance, if you try to come up with a type system that can define
the type of "map" you will probably come up with something so
complicated that it will never be agreed upon or implemented.
(Python's map is much harder to type-declare than that of functional
languages because the function passed in must handle exactly as many
arguments as the unbounded number of sequences that are passed as
arguments to map.)

Even if we took an extreme position and ONLY allowed type annotations
for basic types like strings, numbers and sequences, Python would 
still be a better language. There are thousands of instances of these 
types in the standard library. If we can improve the error checking 
and documentation of these methods we have improved on the status 
quo. Adding type annotations for the other parameters could wait 
for another day.

----

In particular there are three features that have always exploded into
unending debates in the past. I claim that they should temporarily be
set aside while we work out the basics.

 A) Parameterized types (or templates): 

Parameterized types always cause the discussion to spin out of control
as we discuss levels and types of
parameterizability. A type system can be very useful with
parameterization. For instance, Python itself is written in C. C has no
parameterizability. Yet C is obviously still very useful (and simple!).
Java also does not yet have parameterized types and yet it is the most
rapidly growing statically typed programming language!

It is also important to note that parameterized types are much, much
more important in a language that "claims" to catch most or all type
errors at compile time. Python will probably never make that claim.
If you want to do a more sophisticated type check than Python allows,
you should do that in an assertion:

assert Btree.containsType(String)

Once the basic type system is in place, we can discuss the importance
of parameterized types separately later. Once we have attempted to use
Python without them, we will understand our needs better. The type
system should not prohibit the addition of parameterized types in the
future. 

A person could make a strong argument for allowing parameterization
only of basic types ("list of string", "tuple of integers") but I
think that we could even postpone this for the future.

 B) Static type checking: 

Static type warnings are important and we want to enable the development
of tools that will detect type errors before applications are shipped.
Nevertheless, we should not attempt to define a static type checking
system for Python at this point. That may happen in the future or never.

Unlike Java or C++, we should not require the Python interpreter
itself to ever reject code that "might be" type incorrect. Other tools
such as linters and IDEs should handle these forms of whole-program
type-checks.  Rather than defining the behavior of these tools in
advance, we should leave that as a quality of implementation issue for
now.

We might decide to add a formally-defined static type checking to
Python in the future. Dynamically checked annotations give us a
starting point. Once again, I think that the type system should be
defined so that annotations could be used as part of a static type
checking system in the future, should we decide that we want one.

 C) Attribute-value and variable declarations: 

In traditional static type checking systems, it is very important to
declare the type for attributes in a class and variables in a function. 

This feature is useful but it is fairly separable. I believe it should
wait because it brings up a bunch of issues such as read-only
attributes, cross-boundary assignment checks and so forth.

I propose that the first go-round of the types-sig should ONLY address
the issue of function signatures.

Let's discuss my proposal in the types-sig. Executive summary:

 * incremental development policy
 * syntax for parameter type declarations
 * syntax for return type declarations
 * optional runtime type checking
 * goals are better runtime error reporting and method documentation

Deferred for future versions (or never):

 * compile-time type checking
 * parameterized types
 * declarations for variables and attributes

http://www.python.org/sigs/types-sig/

-- 
Python:
    Programming the way
    Guido
    indented it.



From claird at starbase.neosoft.com  Mon Mar  5 01:42:19 2001
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 5 Mar 2001 00:42:19 -0600
Subject: Java guy interested in Python
References: <003801c0a459$a29c97f0$63da1542@wisdom> <mailman.983719509.18569.python-list@python.org>
Message-ID: <3B3577EAB7461A6F.6C3C0CBFFAFBFDD1.82AE68222D3770F0@lp.airnews.net>

In article <mailman.983719509.18569.python-list at python.org>,
Steve Purcell  <stephen_purcell at yahoo.com> wrote:
>Kyle Wayne Kelly wrote:
>> Can anyone tell me some of the things that Python offers which are not
>> available in Java?  I like Python's syntax, but what can Python do that
>> Java can not do?  Any comments are helpful.
>
>Just try it and see.
>
>The 2 hours that you might spend trying to solicit and evaluate responses
>from those on this list would already be enough to get a feel for Python!
>
>You can whiz through the tutorial, then write a toy program or browse through
>some existing code. (e.g. standard library modules)
>
>(Hint: don't expect to choose between Python and other languages based on
>'features' -- you won't find legibility, sanity and brevity mentioned
>on any Java feature list.)
			.
			.
			.
1.  'Notice how know one ever says, "Try Java for
    two hours, and see how you like it"?  'Happens
    all the time with Python.  A sufficiently ad-
    versarial person might try to make something
    of that.
2.  Python has a bunch of pleasant GUI bindings.
    Java has Swing.  PyGUI work tends to be fun,
    portable and quick.  Swing is none of those
    things.
3.  Why do you get to choose a GUI binding?  That's
    a consequence of a more general benefit:  it's
    easy to embed and extend Python.  Python plays
    well with all kinds of C APIs.  Java doesn't.
    Java has JNI, barely, but who wants to use it?
-- 

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


From Eugene.Leitl at lrz.uni-muenchen.de  Fri Mar 16 12:10:04 2001
From: Eugene.Leitl at lrz.uni-muenchen.de (Eugene Leitl)
Date: Fri, 16 Mar 2001 18:10:04 +0100 (MET)
Subject: OO-Browser
In-Reply-To: <_grs6.171$sk3.49961@newsb.telia.net>
Message-ID: <Pine.GSO.4.03.10103161808490.15226-100000@sun1.lrz-muenchen.de>

On Fri, 16 Mar 2001, Fredrik Lundh wrote:

> they're dead.
> 
> also see:
> http://hem.passagen.se/eff/2000_10_01_bot-archive.htm#1201078

Thanks. Luckily, OO-Browser is still alive. Woo!
	http://sourceforge.net/projects/oo-browser




From Kamlesh.Vazirani at jwt.com  Thu Mar 22 01:51:05 2001
From: Kamlesh.Vazirani at jwt.com (Kamlesh.Vazirani at jwt.com)
Date: Thu, 22 Mar 2001 12:21:05 +0530
Subject: DCOracle Error
Message-ID: <OFDAC158A1.88B78C02-ON65256A17.0025812E@jwt.com>

Connecting to Oracle database Using DCOracle python package gives following
error message,Pls help

Import DCOracle???????? (gives error message)

File c:\pyhton20\pythonwin\DCOracle\__init__.py
import sys,Buffer
ImportError : DLL load failed:one of the library file needed to run this
application cannot be found.

My email_id : kjv at cheerful.com

Regds
Kamy




From digitig at cix.co.uk  Wed Mar 21 17:02:00 2001
From: digitig at cix.co.uk (Tim Rowe)
Date: Wed, 21 Mar 2001 22:02 +0000 (GMT Standard Time)
Subject: What IDE are you'all using for Python 2.1 for Windows?
References: <mailman.985119746.9901.python-list@python.org>
Message-ID: <memo.20010321220220.57471A@digitig.compulink.co.uk>

In article <mailman.985119746.9901.python-list at python.org>, 
moshez at zadka.site.co.il (Moshe Zadka) wrote:

> On Tue, 20 Mar 2001, digitig at cix.co.uk (Tim Rowe) wrote:
> 
> > triple-quoted strings (one of the things I consider unpythonic that 
> > seems to have been in Python from the start, FWIW!)
> 
> It hasn't been in Python from the start. It was added when people found
> it very very painful to emit large strings to files.
> 
> Compare:
> 
> message = '''\
> From: %(origin)s
> To: %(dest)s
> Subject: triple quotes in Python
> 
> Why are there triple quoted strings
> in Python?
> ''' % vars()
> os.popen("/usr/lib/sendmail -oi -t", 'w').write(message)
> 
> To
> 
> message = (
> "From: %(origin)s\n"
> "To: %(dest)s\n"
> "Subject: triple quotes in Python\n"
> "\n"
> "Python doesn't need no\n"
> "stinking triple quotes\n" ) % vars()
> os.popen("/usr/lib/sendmail -oi -t", 'w').write(message)
> 
> Which version would you feel safer letting a non-programmer modify?

So Joe Non-Programmer comes along and changes the first one to:

message = '''\
From: %(origin)s
To: %(dest)s
Subject: triple quotes in Python

Why does Python have this "'''"
string thing?
''' % vars()
os.popen("/usr/lib/sendmail -oi -t", 'w').write(message)

Moral: don't let a non-programmer modify your program code! (Even more 
important, don't let a malicious programmer /masquerading/ as a 
non-programmer modify your program code :-)

But heck, I use the things, so clearly I can live with 'em. But emacs 
can't parse 'em!



From mwh21 at cam.ac.uk  Sun Mar 11 16:17:56 2001
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 11 Mar 2001 21:17:56 +0000
Subject: Static typing (was Re: Java guy interested in Python)
References: <mailman.984334635.27345.python-list@python.org> <98glhc02sga@news2.newsguy.com>
Message-ID: <m3hf1080or.fsf@atrus.jesus.cam.ac.uk>

"Alex Martelli" <aleaxit at yahoo.com> writes:

> "Tim Peters" <tim.one at home.com> wrote in message
> news:mailman.984334635.27345.python-list at python.org...
> > [Rainer Deyke]
> > > When I type 'int i;' in C++, what I really mean is "'i' behaves like an
> > > integer variable.".
> >
> > Since that's all "int i;" *can* mean in C++, seems tautological <wink>.
> 
> Surely you're joking, Mister Peters?
> 
>     class int {
>         public:
>             int(const char*=0);
>             ~int();
>             const char* data() const;
>     };
> 
>     int i;
> 
> Isn't this perfectly legal C++

No.

> , meaning something very different from "i behaves like an integer
> variable"...?  int is no "reserved word", after all.

Yes it is.  "identifier" and "keyword" are disjoint classes of token
in C++, at least by my reading of the 1997-12 working draft I've just
dredged up with google.

Unless I've missed some subtle joke...

Cheers,
M.

-- 
  About the use of language: it is impossible to sharpen a 
  pencil with a blunt axe. It is equally vain to try to do 
  it with ten blunt axes instead.
      -- E.W.Dijkstra, 18th June 1975. Perl did not exist at the time.


From mwh21 at cam.ac.uk  Thu Mar 29 07:07:38 2001
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: Thu, 29 Mar 2001 13:07:38 +0100 (BST)
Subject: python-dev summary, 2001-03-15 - 2001-03-29
Message-ID: <Pine.LNX.4.10.10103291306460.866-100000@localhost.localdomain>

 This is a summary of traffic on the python-dev mailing list between
 Mar 15 and Mar 28 (inclusive) 2001.  It is intended to inform the
 wider Python community of ongoing developments.  To comment, just
 post to python-list at python.org or comp.lang.python in the usual
 way. Give your posting a meaningful subject line, and if it's about a
 PEP, include the PEP number (e.g. Subject: PEP 201 - Lockstep
 iteration) All python-dev members are interested in seeing ideas
 discussed by the community, so don't hesitate to take a stance on a
 PEP if you have an opinion.

 This is the fourth summary written by Michael Hudson.
 Summaries are archived at:

  <http://starship.python.net/crew/mwh/summaries/>

   Posting distribution (with apologies to mbm)

   Number of articles in summary: 410

    50 |                 [|]                                    
       |                 [|]                                    
       |                 [|]                                    
       |                 [|]                                    
    40 |                 [|]                                    
       |                 [|] [|]                                
       | [|]             [|] [|]                                
       | [|]             [|] [|] [|]     [|]                    
    30 | [|]             [|] [|] [|]     [|]                    
       | [|]             [|] [|] [|]     [|]                    
       | [|]             [|] [|] [|]     [|] [|]                
       | [|]         [|] [|] [|] [|]     [|] [|]             [|]
    20 | [|] [|]     [|] [|] [|] [|]     [|] [|]             [|]
       | [|] [|]     [|] [|] [|] [|] [|] [|] [|]             [|]
       | [|] [|]     [|] [|] [|] [|] [|] [|] [|]     [|]     [|]
       | [|] [|]     [|] [|] [|] [|] [|] [|] [|]     [|]     [|]
    10 | [|] [|] [|] [|] [|] [|] [|] [|] [|] [|]     [|]     [|]
       | [|] [|] [|] [|] [|] [|] [|] [|] [|] [|]     [|]     [|]
       | [|] [|] [|] [|] [|] [|] [|] [|] [|] [|] [|] [|] [|] [|]
       | [|] [|] [|] [|] [|] [|] [|] [|] [|] [|] [|] [|] [|] [|]
     0 +-044-024-013-029-059-046-040-022-040-031-007-019-008-028
        Thu 15| Sat 17| Mon 19| Wed 21| Fri 23| Sun 25| Tue 27|
            Fri 16  Sun 18  Tue 20  Thu 22  Sat 24  Mon 26  Wed 28

 Bug-fixing for 2.1 remained a priority for python-dev this fortnight
 which saw the release of 2.1b2 last Friday.


    * Python 2.0.1 *

 Aahz posted his first draft of PEP 6, outlining the process by which
 maintenance releases of Python should be made.

  <http://python.sourceforge.net/peps/pep-0006.html>

 Moshe Zadka has volunteered to be the "Patch Czar" for Python 2.0.1.

  <http://mail.python.org/pipermail/python-dev/2001-March/013952.html>

 I'm sure we can all join in the thanks due to Moshe for taking up
 this tedious but valuable job!


    * Simple Generator implementations *

 Neil Schemenauer posted links to a couple of "simple" implementations
 of generators (a.k.a. resumable functions) that do not depend on the
 stackless changes going in.

  <http://mail.python.org/pipermail/python-dev/2001-March/013648.html>
  <http://mail.python.org/pipermail/python-dev/2001-March/013666.html>

 These implementations have the advantage that they might be
 applicable to Jython, something that sadly cannot be said of
 stackless.
 

    * portable file-system stuff *

 The longest thread of the summary period started off with a request
 for a portable way to find out free disk space:

  <http://mail.python.org/pipermail/python-dev/2001-March/013706.html>

 After a slightly acrimonious debate about the nature of Python
 development, /F produced a patch that implements partial support for
 os.statvfs on Windows:

  <http://sourceforge.net/tracker/index.php?func=detail&aid=410547&group_id=5470&atid=305470>

 which can be used to extract such information.

 A side-product of this discussion was the observation that although
 Python has a module that does some file manipulation, shutil, it is
 far from being as portable as it might be - in particular it fails
 miserably on the Mac where it ignores resource forks.  Greg Ward then
 pointed out that he had to implement cross-platform file copying for
 the distutils

  <http://mail.python.org/pipermail/python-dev/2001-March/013962.html>

 so perhaps all that needs to be done is for this stuff to be moved
 into the core.  It seems very unlikely there will be much movement
 here before 2.2.




From mwh21 at cam.ac.uk  Mon Mar 19 16:03:20 2001
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 19 Mar 2001 21:03:20 +0000
Subject: Python style questions
References: <98tb91$9t9$1@saltmine.radix.net> <m34rwpilxg.fsf@atrus.jesus.cam.ac.uk> <995o0u$itc$1@news.mathworks.com>
Message-ID: <m3vgp5h3on.fsf@atrus.jesus.cam.ac.uk>

Joshua Marshall <jmarshal at mathworks.com> writes:

> Michael Hudson <mwh21 at cam.ac.uk> wrote:
> > cobrien at Radix.Net (Cary O'Brien) writes:
> ...
> > The scoping can get a bit counter-intuitive
> >> 3. I really miss not having a "switch" or "case" statement.  Sniff.
> 
> > -- 
> > 3. Syntactic sugar causes cancer of the semicolon.
> >   -- Alan Perlis, http://www.cs.yale.edu/homes/perlis-alan/quotes.html
> 
> A switch statement may just be syntactic sugar, but not necessarily in
> a simple way.  A good compiler will compile switch statements into
> jump-tables or, at worst, binary searches.  Cluttery to code by hand.
> 
> Sure this is "just" an efficiency thing, but switches can get quite big.

Gah, don't news-servers respect cancels?  That post was only half
finished when I gave up on it, and then hit C-c C-c rather than C-c
C-d.

Obviously this isn't going to work for people on the mailing list...

Cheers,
M.

-- 
  Java is a WORA language! (Write Once, Run Away)
                	-- James Vandenberg (on progstone at egroups.com)
                           & quoted by David Rush on comp.lang.scheme


From stevewilliams at wwc.com  Sun Mar  4 14:35:43 2001
From: stevewilliams at wwc.com (Steve Williams)
Date: Sun, 04 Mar 2001 19:35:43 GMT
Subject: Missing the functional mind set
References: <mlb2at8ptrr7sh5dgq8qlk2201kquh6tfl@4ax.com> <uDao6.335972$ge4.119457077@news2.rdc2.tx.home.com> <t6f2at870pk82q96bnd5a2od4n8ns5kf83@4ax.com> <ntg2atcemm3hp9rb26hn7c5uin4dfn8lc9@4ax.com> <97rqgb$lt2$1@tyfon.itea.ntnu.no>
Message-ID: <3AA298F5.7742810E@wwc.com>

Magnus Lie Hetland wrote:

> "Kirby Urner" <urner at alumni.princeton.edu> wrote in message
> news:ntg2atcemm3hp9rb26hn7c5uin4dfn8lc9 at 4ax.com...
> > Daniel Klein <danielk at aracnet.com> wrote:
> [snip]
> >   >>> params = [None]*3 # list
>
> This should be
>
>     >>> params = [None]*4

[snip]

To be consistent with the "index == offset" mindset, counting should
start at zero.

As in:  "I have 2 children--0 boy and 1 girls".




From aahz at panix.com  Wed Mar 21 20:23:24 2001
From: aahz at panix.com (Aahz Maruch)
Date: 21 Mar 2001 17:23:24 -0800
Subject: urllib.urlopen doesn't accept Unicode strings.
References: <998tau$q6i$1@troll.powertech.no> <j4n1aeycsl.fsf@informatik.hu-berlin.de>
Message-ID: <99bk6c$hh7$1@panix6.panix.com>

In article <j4n1aeycsl.fsf at informatik.hu-berlin.de>,
Martin von Loewis  <loewis at informatik.hu-berlin.de> wrote:
>
>using types.UnicodeType is much better than type(u"")

Really?  I haven't seen any strong consensus on this issue.
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

"I won't accept a model of the universe in which free will, omniscient
gods, and atheism are simultaneously true."  -- M


From flognat at flognat.myip.org  Sat Mar 17 07:40:00 2001
From: flognat at flognat.myip.org (Andrew Markebo)
Date: 17 Mar 2001 13:40:00 +0100
Subject: Should I close after popen??
References: <XFMail.20010315142704.shaleh@valinux.com> <mailman.984697040.2458.python-list@python.org> <m3lmq6ppv0.fsf@master.athome> <98satl$1t8$0@216.39.151.169> <m3lmq6nk9a.fsf@master.athome> <98uf48$552$1@216.39.151.169>
Message-ID: <m3bsr0muvz.fsf@master.athome>

/ Donn Cave <donn at oz.net> wrote:
| [...]
| Depends on what you want to do.  For sure, select has the potential

Basically, run two or more popens in parallell, and when there is data
on one of them read it and do some work.. 

Basically what I do now is do two popen2, do a wait on the first, chew
some data, wait for the second, not critical at the moment, just
interested what to do if I had to chew the data from the second
process first.

| Anyway, it would be easier to say what you ought to use, if it were
| clear what problem you're trying to solve.  It's probably simpler

No problem at the moment :-) Only trying to sort out different methods
so I know how to handle a problem when it comes :-)

| Did I mention that you could find some relevant documentation in
| the pclose man page?

mmhm I had a problem associating pclose to pythin fileobject close.. I
was looking for pclose in the python docs :-)

| One unrelated advantage of popen2 is that if you supply a list of
| arguments, it will use them directly instead of a shell command
| line.  That is much safer.  A command line constructed at run

You mean like popen2(["command", "param", "param"]...)?

Thanx for the help!

        /Andy


From greg at cosc.canterbury.ac.nz  Thu Mar  8 19:30:26 2001
From: greg at cosc.canterbury.ac.nz (Greg Ewing)
Date: Fri, 09 Mar 2001 13:30:26 +1300
Subject: learning C
References: <mailman.983972829.23481.python-list@python.org> <983975751.18833@newsmaster-04.atnet.at>
Message-ID: <3AA82422.53026D3D@cosc.canterbury.ac.nz>

Werner Schiendl wrote:
> 
> The fine thing is: C++ still HAS the scalpel, so what?

C++ is a Swiss army knife with both scissors and scalpel,
and a nasty tendency for the scalpel to shoot out and
stick you in the palm when all you were trying to do
was use the scissors...

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


From pbrian at demon.net  Fri Mar 30 04:37:41 2001
From: pbrian at demon.net (Paul Brian)
Date: Fri, 30 Mar 2001 10:37:41 +0100
Subject: HTML Tidy and Python wrapper
Message-ID: <985945106.28359.0.nnrp-02.c1c3e154@news.demon.co.uk>

Dear all,

I was sure that I recently read that there was a python wrapper around the
HTML Tidy (http://www.w3.org/People/Raggett/tidy/). Unfortunately I cannot
find the reference. Does any one know if such a thing exists, as simple
searches through deja have come up nought.

thank you




From tomega at earthlink.net  Wed Mar 21 14:20:25 2001
From: tomega at earthlink.net (Simon Kesenci)
Date: Wed, 21 Mar 2001 19:20:25 GMT
Subject: learning python...
References: <3AB8F4E1.184DD9C@noaa.gov>
Message-ID: <Z97u6.16733$Im6.1761313@newsread1.prod.itd.earthlink.net>

If you know programming, forget _Learning_ _Python_.  I tried _Programming_ 
_Python_ last year, but it covered version 1.3.  A good book if they 
updated it since then.  As a programmer, you should find the online 
tutorial easy and helpful <http://python.org/doc/current/tut/tut.html>.

--S.

Benjamin.Altman wrote:

> Hello.  From the perspective of someone who already has programming
> experience, would anyone know if O'reillys "Learning Python" book would
> be better than say Prentice Hall's "Core Python Programming"?  Or would
> it be better to go straight into something like "Programming Python"?
> 
> Thanks,
> Ben
> 
> 



From phd at phd.russ.ru  Mon Mar 12 08:43:17 2001
From: phd at phd.russ.ru (Oleg Broytmann)
Date: 12 Mar 2001 07:43:17 -0600
Subject: Dr. Dobb's Python-URL! - weekly Python news and links (Mar 12)
Message-ID: <35C5925F88F04631.E73ECBD7DB304D6F.4C57555D2D73FB9F@lp.airnews.net>

    Guido van Rossum announces Python Software Foundation
        http://www.prnewswire.com/cgi-bin/stories.pl?ACCT=104&STORY=/www/story/03-06-2001/0001442061&EDATE=
    
    Reports from The Ninth International Python Conference
        http://groups.yahoo.com/group/python-list/message/127037
        http://www.amk.ca/python/writing/ipc9.html
        http://weblogs.userland.com/zopeNewbies/2001/03/06
        http://python.oreilly.com/news/pythonday1_0301.html
        http://python.oreilly.com/news/pythonday2_0301.html
        http://python.oreilly.com/news/pythonday3_0301.html
    
    wxPython = Python + wxWindows
        http://www-106.ibm.com/developerworks/library/l-wxpy/?dwzone=linux?open&l=252,t=gr,p=wxPython
    
    PythonWare PY20 distribution for Windows and Linux (rpm)
    Includes PythonWare Image Library 1.1.1, PythonWare Sound Toolkit,
    Tkinter 8.3.2, SOAPLIB, XML-RPC, and more
        http://www.pythonware.com/products/python/index.htm
    
    PyTREX is a Python implementation of TREX, an XML schema language
        http://pytrex.sourceforge.net/
    
    D'Arcy J.M. Cain merges PyGreSQL into PostgreSQL tree
        http://www.postgresql.org/mhonarc/pgsql-announce/2001-03/msg00001.html
    
    Andy Dustman moves MySQLdb to SourceForge
        http://sourceforge.net/forum/forum.php?forum_id=70801
    
    Rafal Smotrzyk posts python cgi syntax highlighter
        http://groups.yahoo.com/group/python-list/message/126756
    
    Jean-Luc Fontaine releases tclpython-2.0
        http://groups.yahoo.com/group/python-list/message/127517
    
    David Mertz annonces preliminary releases of sql2dtd.py and sql2xml
        http://groups.yahoo.com/group/python-list/message/127572
    
    
    Zope:
    
    Security alert and hotfix
        http://lists.zope.org/pipermail/zope-announce/2001-March/000385.html
    
    
    Hot threads of the week:
    
    The demise of 'from foo import *' and its implications
        http://groups.google.com/groups?lr=&group=comp.lang.python.*&safe=off&ic=1&th=239b31b698e5477f&seekd=918627400
    I come to praise .join, not to bury it
        http://groups.google.com/groups?num=25&lr=&group=comp.lang.python.*&safe=off&th=f5fa895f69b1b7c0&ic=1&seekd=918606249
    Can Python replace Visual Basic? Should it?
        http://groups.google.com/groups?lr=&group=comp.lang.python.*&safe=off&ic=1&th=68d5c86d6069b39f&seekd=917743230
    Can anyone recomend a good intoduction to C
        http://groups.google.com/groups?lr=&safe=off&th=555b54918c7bb1de&ic=1&seekd=918083102
    learning C
        http://groups.google.com/groups?lr=&group=comp.lang.python.*&safe=off&ic=1&th=8368cac8a98ecb9c&seekd=916271275
    Urgent Question about Python licensing
        http://groups.google.com/groups?lr=&group=comp.lang.python.*&safe=off&ic=1&th=1c57c3a40ae385a2&seekd=915908166


========================================================================

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

      Python.org's Python Language Website is the traditional
      center of Pythonia
          http://www.python.org
      Notice especially the master FAQ
          http://www.python.org/doc/FAQ.html

      PythonWare complements the digest you're reading with the
      daily python url
           http://www.pythonware.com/daily  

      comp.lang.python.announce announces new Python software.  Be
      sure to scan this newly-revitalized newsgroup at least weekly.
	  http://groups.google.com/groups?oi=djq&as_ugroup=comp.lang.python.announce

      Michael Hudson continues Andrew Kuchling's marvelous tradition
      of summarizing action on the python-dev mailing list once every
      other week.
	  http://starship.python.net/crew/mwh/summaries/
	  http://www.amk.ca/python/dev

      The Vaults of Parnassus ambitiously collect Python resources
          http://www.vex.net/~x/parnassus/   

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

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

      Cetus does much of the same
          http://www.cetus-links.de/oo_python.html

      Python FAQTS
          http://python.faqts.com/

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

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

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

      Archive probing trick of the trade:
	  http://groups.google.com/groups?oi=djq&as_ugroup=comp.lang.python&num=100


Previous - (U)se the (R)esource, (L)uke! - messages are listed here:
  http://purl.org/thecliff/python/url.html
or
  http://groups.google.com/groups?oi=djq&as_q=+Python-URL!&as_ugroup=comp.lang.python


Suggestions/corrections for next week's posting are always welcome.
[http://www.egroups.com/list/python-url-leads/ is hibernating.  Just
e-mail us ideas directly.]

To receive a new issue of this posting in e-mail each Monday morning,
ask <claird at neosoft.com> to subscribe.  Mention "Python-URL!".


-- The Python-URL! Team--

Dr. Dobb's Journal (http://www.ddj.com) is pleased to participate in and
sponsor the "Python-URL!" project.
-- 

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


From tino at wildenhain.de  Wed Mar 14 11:44:47 2001
From: tino at wildenhain.de (Tino Wildenhain)
Date: Wed, 14 Mar 2001 17:44:47 +0100
Subject: String comparison
References: <3AAF9290.42BDAAD3@mailhost.stg.rt.daimlerchrysler.com>
Message-ID: <3AAF9FFF.AF55B76B@wildenhain.de>

Hi Alexander,

try str(processname)=="netscape"

processname could be a class instance who looks
like a string but is not a string.
Also type(processname) and friends may help to
figure this out.

HTH
Tino Wildenhain

Alexander Stirzel wrote:
> 
> Hi!
> 
> I have a string processname with "netscape" in it and I want to compare
> it with the written word "netscape".
> 
> This are some lines of my source code:
> 
> (...)
> processname = splitted.pop()
> (... no the string netscape is in processname ...)
> if ((processname=="netscape") and ((string.atof(cputimeerg))>45) and
> ((string.atof(cpupercwert))>45)):
> (... the first comparison is false ... why? ...)
> 
> I printed processname just before this comparison and I got "netscape"
> on screen. (With no spaces there, that's not the problem) But when I
> >>>> print (processname=="netscape")
> I get a "0" on screen. And when I
> >>>> print (processname)
> I get "netscape" on screen.
> 
> This is strange...
> 
> Thanks for any help.
> 
> --
> Alexander Stirzel
> E-Mail: alexander.stirzel at daimlerchrysler.com
> *** welcome to hell, here's your copy of windows ***
> --
> http://mail.python.org/mailman/listinfo/python-list



From lsommerer at sewardweb.com  Thu Mar 15 14:50:53 2001
From: lsommerer at sewardweb.com (Lloyd Sommerer)
Date: Thu, 15 Mar 2001 13:50:53 -0600
Subject: Newbie: Is there a more pythonic approach?
References: <3AAE87E8.FE11B4F9@sewardweb.com> <HHMr6.253$4d3.54012@newsb.telia.net>
Message-ID: <3AB11D1D.3530C86C@sewardweb.com>

Thank you very much for the pointers, They were most helpful, except for the
one that didn't make sense to me.  I'm having trouble following how the change
below actually works:

Fredrik Lundh wrote:

> > if os.path.exists(base_directory+lowercase_url):
> >     new_url = base_url+lowercase_url
>
> if you like source code optimizations, you could prepend an empty
> extension to the extensions list and get rid of the if clause:
>
>       for extension in [''] + extension_list:
>
> > else:
> >     for extension in extension_list:
> >         if os.path.exists(base_directory+url_path+extension):
> >             new_url = base_url+url_path+extension
>

What I'm having trouble seeing is how this checks for the URL having the
correct extension, but the wrong case. The original script and the script as
it currently exists follow, in case refering to it proves helpful to anyone.

Lloyd Sommerer
Original Version:
==================================================
error_page = '/404.html'
base_url = 'http://churchsite.gatheringspot.com'
base_directory = '/home/gatheri/public_html/churchsite/'
extension_list = ['.html','.shtml','.htm','.txt','.php','.asp']

import string
import os

lowercase_url = string.lower(os.environ['REDIRECT_URL'])
url_path,url_extension = os.path.splitext(lowercase_url)
new_url = base_url+error_page
#
# Check to see if simply converting the URL to lowercase
# corrects the problem. If it doesn't, check to see if the
# extension is incorrect and fix it if it is.
#
if os.path.exists(base_directory+lowercase_url):
    new_url = base_url+lowercase_url
else:
    for extension in extension_list:
        if os.path.exists(base_directory+url_path+extension):
            new_url = base_url+url_path+extension
#
# Send HTTP header information to the browser so that it
# can request the "correct" document.
#
print "Status: 301 Permanently moved"
print "Location: " + new_url
print

Current Version:
==================================================
error_page = '404.html'
base_url = 'http://www.gatheringspot.com'
base_directory = '/home/gatheri/public_html'
extension_list = ['.html','.shtml','.htm','.txt','.php','.asp']

from string import lower
from os.path import exists, join, splitext
from urlparse import urljoin
from os import environ

lowercase_url = string.lower(environ['REDIRECT_URL'])
lowercase_url = lowercase_url[1:] # removes leading '/'
new_url = urlparse.urljoin(base_url,error_page)
#
# Check to see if simply converting the URL to lowercase
# corrects the problem. If it doesn't, check to see if the
# extension is incorrect and fix it if it is.
#
if exists(join(base_directory,lowercase_url)):
    new_url = urljoin(base_url,lowercase_url)
else:
    url_path,url_extension = splitext(lowercase_url)
    for extension in extension_list:
        if exists(join(base_directory,(url_path+extension))):
            new_url = urljoin(base_url,(url_path+extension))
            break # done when we find the first match.
#
# Send HTTP header information to the browser so that it
# can request the "correct" document.
#
print "Status: 301 Permanently moved"
print "Location: ", new_url
print



From fredrik at pythonware.com  Fri Mar 30 02:23:20 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 30 Mar 2001 07:23:20 GMT
Subject: Where are the examples/samples?
References: <20010330021322.08468.00001688@ng-fe1.aol.com>
Message-ID: <IvWw6.4263$4N4.893629@newsc.telia.net>

"LMHAMRIC" <lmhamric at aol.com> wrote:
> After looking at every Python book in the local bookstore and at every
> Python-related site I could find on the Internet I have to admit that I'm
> perplexed. I've seen a number of posts that have asked for examples of
> various things and when I follow the links - where are the examples?

http://www.oreilly.com/catalog/pythonsl/

first edition (electronic print-it-yourself edition only) here:
http://www.pythonware.com/people/fredrik/librarybook.htm

Cheers /F




From MarkH at ActiveState.com  Sun Mar 25 17:59:54 2001
From: MarkH at ActiveState.com (Mark Hammond)
Date: Sun, 25 Mar 2001 22:59:54 GMT
Subject: win32com, handling byref
References: <mailman.985526194.25003.python-list@python.org>
Message-ID: <3ABE7863.4060406@ActiveState.com>

Clark C. Evans wrote:

> I am trying to call the following function from an ActiveX 
> object.  It has a return value, plus three output parameters.
> Unfortunately, the type library doesn't mark the last three
> as "out", 

The type lib is clearly broken.  I would expect this control would also have problem with COM marshalling.

The best solution is clearly to fix the type lib.  This may not be possible tho :)

You are pretty much SOL with makepy as it stands - garbage in, garbage out :)

> Is this even remotely possible?  If not, is there a description
> of the low-level hoops that I could possibly jump through?

You will have to pass custom type-tuples to InvokeTypes().  This may be as simple as changing the sample makepy generated code you pased, changing the second integer in the OUT param tuples from 1 to 2 (win32com.PARAMFLAG_FOUT)

Mark.



From loewis at informatik.hu-berlin.de  Mon Mar 26 03:50:54 2001
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 26 Mar 2001 10:50:54 +0200
Subject: So what's wrong with __future__?  (Was Re: Why "from __future__" stinks ...)
References: <slrn9bdka2.3j6.hzhu@rocket.knowledgetrack.com> <LNBBLJKPBEHFEDALKOLCAEDKJIAA.tim.one@home.com> <mailman.985568441.15878.python-list@python.org> <vvmtbto161c3n1m3n7ub4vii2nta490alb@4ax.com>
Message-ID: <j41yrkc3rl.fsf@informatik.hu-berlin.de>

Courageous <jkraska1 at san.rr.com> writes:

> While I may be guilty of not paying attention, I just don't understand
> why it's needed at all? Those who don't want to play with new features
> should keep to an older Python revision. This is the standard way of
> doing things in the language business. This import from __future__
> thing seems not on like a glorious hack, but terribly unnecessary.
> What am I missing?

That it is often not feasible to avoid more recent versions. If you
want to give your program away, it may be that other people only have
more recent versions, and want to avoid the trouble of installing an
older one just to run a single program. Or, you are using a library
that turns out to have a bug, and there is a bug-fixed version of the
library, and that library version requires a newer Python version.

Regards,
Martin



From grante at visi.com  Mon Mar  5 09:19:48 2001
From: grante at visi.com (Grant Edwards)
Date: Mon, 05 Mar 2001 14:19:48 GMT
Subject: Can anyone recomend a good intoduction to C...
References: <3aa366c6_2@nnrp1.news.uk.psi.net>
Message-ID: <8gNo6.440$y6.128357@ruti.visi.com>

In article <3aa366c6_2 at nnrp1.news.uk.psi.net>, Simon Brunning wrote:
>... for a Python programmer?
>
>I'm going to have to bite the bullet and learn a bit of C. I'd at
>least like to learn the *right* way to do it.

The "right" way to do it is with a PDP-11, a Lear-Siegler ADM3a
and the v7 source code.  ;)  Some would argue in favor of an
ASR-33, but they tend to annoy the downstairs neighbors.

-- 
Grant Edwards                   grante             Yow!  Hmmm... A hash-singer
                                  at               and a cross-eyed guy were
                               visi.com            SLEEPING on a deserted
                                                   island, when...


From KatMouse at gmx.de  Mon Mar 19 16:07:15 2001
From: KatMouse at gmx.de (Eduard Hiti)
Date: Mon, 19 Mar 2001 22:07:15 +0100
Subject: How to increase PythonWin v2.0: COM browser limits?
References: <994psh$so5$1@news1.xs4all.nl>
Message-ID: <995sgi$7u0$04$1@news.t-online.com>

I just tried the COM browser here on my machine, and it gives me the same
error. I'm suspecting an error in Pythonwin...

Eduard

"Twan van der Schoot" <twanvds at xs4all.nl> wrote in message
news:994psh$so5$1 at news1.xs4all.nl...
| I'm trying to use the COM Browser in Pythonwin version 2.0.  But I run
into
| limitations of some kind.  As a Python newby I dutifully scanned FAQs and
| newsgroups, but alas!
|
| To reproduce the problem: On a heavily used NT 4.0 system with a lot of
| registered COM stuff, start Pythonwin version 2.0 go to the menu entry:
COM
| Browser to start the COM browser and expand the "Registered Categories"
| branch in the tree view.
|
| Pythonwin shows in the interactive window a backtrace on an exception.
The
| culprit raising this exception was the response:
|
| File "D:\Python20\win32com\client\util.py", line 53, in __GetIndex
|     result = self._oleobj_.Next(1)
| pywintypes.com_error: (-2147024882, 'Not enough storage is available to
| complete this operation.', None, None)
| win32ui: Exception in OnNotify() handler
|
| Something suggests that the number of registered objects is too large for
| the COM browser.  As I've -of course- the sources available, I was wonder
if
| there is some parameter which I can increase to solve the problem.  Or am
I
| running into limitations of the treeview control of MS Windows?
|
| TIA
|
| Twan van der Schoot
| twanvds at NOSPAM.xs4all.nl
|
|




From shaleh at valinux.com  Fri Mar 16 17:49:27 2001
From: shaleh at valinux.com (Sean 'Shaleh' Perry)
Date: Fri, 16 Mar 2001 14:49:27 -0800 (PST)
Subject: Type mismatchings and the rank newbie
In-Reply-To: <ylws6.2639$227.274111@newsread2.prod.itd.earthlink.net>
Message-ID: <XFMail.20010316144927.shaleh@valinux.com>

> The loop is not a problem; I just can't figure how to stick a different
> number form the user input total, into the second place of the list name
> (  Pn=[x, y, z] where 'n' is the number stuck in there by the loop) with
> every iteration of the loop.
> 
> The docs are confusing to me (I'm not a programmer and don't want to be, but
> there is no other way for me to do what I need to to see what this program
> will display for me).
> 


from the pyunit examples:

    for i in range(10000):
        exec("def test%i(self): pass" % i)



From ACIUFFO1 at irf.com  Thu Mar  8 21:23:02 2001
From: ACIUFFO1 at irf.com (Andrea Ciuffoli)
Date: Thu, 08 Mar 2001 18:23:02 -0800
Subject: I get lost with "signal"
Message-ID: <saa7ce17.026@irf.com>

Hi,
I'm trying to use "signal" library to control that an FTP connection is established within a specified time (e.g. 3 sec) but I got an Attribute error: SIGALRM.
Is that module supported by Windows NT and 98?
If not, is there another method to control that the connection is established without problem? How can I know if the server is down?
Thanks,
Andrea.




From moran at idx.com  Tue Mar 13 14:41:24 2001
From: moran at idx.com (Dan Moran)
Date: Tue, 13 Mar 2001 14:41:24 -0500
Subject: Missing libX11.a on AIX
Message-ID: <AEFF34C2E454D311993B00902745E4AB05498C6B@news.idx.com>

Hi Folks.

When running python on a AIX box (4.3.3), I get the following message:

exec(): 0509-036 Cannot load program ./python because of the following
errors:
        0509-150   Dependent module libX11.a(shr4.o) could not be loaded.
        0509-022 Cannot load module libX11.a(shr4.o).
        0509-026 System error: A file or directory in the path name does not
exist.

This box was not setup with the X11 libraries.
Can Python be compiled to exclude its dependency on X11?

::Dan Moran







From urner at alumni.princeton.edu  Sat Mar  3 10:54:43 2001
From: urner at alumni.princeton.edu (Kirby Urner)
Date: Sat, 03 Mar 2001 07:54:43 -0800
Subject: Hitting two targets:  OO + group theory (pedagogy)
References: <u2k0atkgkiatdp9ukj3mpd1ej0b648ldsm@4ax.com> <3aa1077a.1110803@nntp.sprynet.com>
Message-ID: <0932atgt9v8d8aboq90evoi5ipj5c0sk8s@4ax.com>

ullrich at math.okstate.edu (David C. Ullrich) wrote:

>On Fri, 02 Mar 2001 18:07:16 -0800, Kirby Urner
><urner at alumni.princeton.edu> wrote:
>
>>
>>============== OREGON CURRICULUM NETWORK =====================
>
>Does the Oregon Curriculum Network have any official status, ie
>any actual connection with the state of Oregon?
>

Other than the fact that I live and work in Oregon, as do
some of the people I collaborate with, no.

On the home page it says, in italics at the bottom:

      The OCN is privately sponsored and does not claim 
      to represent the official views of any government 
      agency

I visit schools sometimes, public and/or private.  Just
this week I guest-taught two segments of 1st grade re 
polyhedra.  Teacher workshops too -- was at the Math Summit
in 1997, which was about mapping a state-wide strategy
(Sir Roger Penrose was there to help us out).

>One comment would be that if we're talking about pedagogy and
>curricula and things we probably want to get the terminology
>straight. There's no such thing as a "relative prime" - you
>meant "the positive integers less than n which are
>relatively prime to n."

Yes.  All integers k, 0 < k < n, such that gcd(k,n)=1.

Thanks for the pointer re no such animal as a "relative
prime".  I'll need to tighten up my syntax.

>And more important there's no reason why the modulus n
>should be prime! This gizmo is a group under multiplication
>whether n is prime or not. (That's the point to specifying
>we're talking about the integers relatively prime to n...)

Very true.  In the referenced math-learn draft
http://www.mathforum.com/epigone/math-learn/shehlimpmimp
I do most of my example with modulus n = 20, showing how 
integers coprime with 20 (< 20) form a group.  I shouldn't
have stipulated a prime modulus in the above post.

>Seems more interesting to me to start with a class
>Group that includes stuff common to all groups, so
>that a subclass can represent any group at all. If we're
>going to implement what's above as part of a
>"curriculum" then at least we should make it clear
>that our sandbox is not "groups", it's a tiny part of
>the category of all groups.

Yes, just an example of a group.  In another essay, I 
go with permutations, showing how these might be associated
with "clubhouse codes" (exchanging letters for other 
letters as in a Julius Caesar code).  I'd also like 
to build on my OO + Polyhedra approach to get into 
symmetry groups -- getting topics to connect around to
other topics as much as feasible is what I mean by a
"curriculum" (stuff connecting around in all 
circumferential directions, to form a system -- the
notion of circling-back is built right into the 
etymology).

I try to be fairly general in my approach to the 
group of relatively prime integers under multiplication
vis-a-vis modulus n in my little essay for 8th graders:

   When we study group theory, we have a mnemonic 
   that's helpful if you've studied the book of 
   Genesis at least enough to know the story of 
   Cain and Abel (use a search engine maybe).  Use 
   the letters CAIN to stand for "Closure, Associative, 
   Inverse, Neutral" -- and there you have the 
   properties you need to call your set & operation 
   a "group".

   These properties are defined with respect to a set 
   of elements, and an operation for relating them 
   (we generally call the operation "multiplication", 
   no matter what it actually does). If multiplication 
   is not just associative, but commutative as well, 
   then that's where Abel comes in -- we call it an 
   Abelian Group.

Probably there's some syntax that could be tightened
up here too.

Thanks for your useful feedback.  Much appreciated.

Kirby



From stephen_purcell at yahoo.com  Mon Mar  5 10:01:56 2001
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Mon, 5 Mar 2001 16:01:56 +0100
Subject: Little problem with the "+" operator
In-Reply-To: <983802310.155838463@news.t-online.de>; from zamurai@gmx.net on Mon, Mar 05, 2001 at 03:25:10PM +0100
References: <983802310.155838463@news.t-online.de>
Message-ID: <20010305160156.A8303@freedom.puma-ag.com>

Zamurai wrote:
> variableOne = 500
> myVariable = variableOne + "cool"
> 
> But I get the following error: "unsupported operand types for +"
> 
> Does someone know how to solve the problem?

You can't add variables of different types together in Python and expect them
to magically agree on a sensible result.

So, you have to explicitly say:

>>> variableOne = 500
>>> myVariable = str(variableOne) + "cool"
>>> myVariable 
'500cool'
>>> 

Or,

>>> variableOne = 500
>>> myVariable = variableOne + int("cool")
Traceback (innermost last):
  File "<stdin>", line 1, in ?
ValueError: invalid literal for int(): cool
>>> 


This is infinitely more sane than:

% perl -e 'print 500 + "cool"'
500


-Steve

-- 
Steve Purcell, Pythangelist
Get testing at http://pyunit.sourceforge.net/
Get servlets at http://pyserv.sourceforge.net/
"Even snakes are afraid of snakes." -- Steven Wright



From cribeiro at mail.inet.com.br  Wed Mar 28 12:21:21 2001
From: cribeiro at mail.inet.com.br (Carlos Alberto Reis Ribeiro)
Date: Wed, 28 Mar 2001 14:21:21 -0300
Subject: diff lists
In-Reply-To: <3AC1DD4E.4BA7D421@aon.at>
Message-ID: <5.0.2.1.0.20010328142016.021b1c00@mail.inet.com.br>

Look at the kjBuckets package. It implements an efficient set type that is 
good for large operations such as set intersection, set union, etc.

At 14:47 28/03/01 +0200, Oliver Vecernik wrote:
>Hi,
>
>I've got following two lists:
>
>['a', 'b', 'c', 'd', 'e', 'f']
>['e', 'c', 'f']
>
>I'd like to have the result:
>
>['a', 'b', 'd']
>
>The list need not to be ordered. ['d', 'a', 'b'] will also be ok. What
>is the most effective way to achive that result?
>
>Oliver
>--
>http://mail.python.org/mailman/listinfo/python-list





From gregj at pobox.com  Sat Mar 10 15:10:32 2001
From: gregj at pobox.com (Gregory Jorgensen)
Date: Sat, 10 Mar 2001 20:10:32 GMT
Subject: How do I check return value of popen2.* ?
References: <5vpd89.4ei.ln@127.0.0.1>
Message-ID: <YSvq6.1899$54.2098@www.newsranger.com>

popen2.popen3 doesn't return the exit status from the child process. From the
documentation:

popen3 (cmd[, bufsize[, mode]]) 
Executes cmd as a sub-process. Returns the file
objects (child_stdout, child_stdin, child_stderr).

(see http://python.org/doc/current/lib/module-popen2.html)

However the popen2.Popen3 class (documented on the same page) does what you
need. This example works for me on a BSD system running Python 1.5.2
---
# t2.py
import sys, popen2

p = popen2.Popen3('./test.py')
p.wait()

# see os.wait() for description of error code
ec = p.poll()
print "exit code: %d, signal %d" % ( (ec >> 8) & 0xFF, ec & 0xFF )
print "---"
print p.fromchild.read()  # dump stdout from child process
print "---"


---
# test.py
import sys

print "hello, world!"
print "bye"
sys.exit(2)
---

Give that a try.

You will find the book "Python Essential Reference" by David Beazley worthwhile.


In article <5vpd89.4ei.ln at 127.0.0.1>, gradha at iname.com says...
>
>Hello.
>
>I have tried the following test. File t2.py:
>
>	#!/usr/bin/env python
>	
>	import sys, popen2
>	
>	sal, en, err = popen2.popen3 ("./test.py")
>	val1 = en.close()
>	print sal.readlines()
>	val2 = sal.close()
>	print err.readlines()
>	val3 = err.close()
>	
>	print val1, val2, val3
>	
>File test.py:
>
>	#!/usr/bin/env python
>	
>	import sys
>	
>	sys.exit(2)
>	
>Now, when I run t2, the output is:
>
>	[]
>	[]
>	None None None
>
>How am I supposed to check the return value of the opened pipe?


Greg Jorgensen
Deschooling Society
Portland, Oregon, USA
gregj at pobox.com


From gustav at morpheus.demon.co.uk  Mon Mar 26 15:22:16 2001
From: gustav at morpheus.demon.co.uk (Paul Moore)
Date: Mon, 26 Mar 2001 21:22:16 +0100
Subject: Building "non-standard" standard extensions
Message-ID: <719vbt8tlkb05h4n1mcvh65l0f0i62tgsj@4ax.com>

I'll explain...

The Windows distribution of Python does *not* come with the readline module
included, not surprisingly as readline is GPL, and not commonly available for
Windows.

However, I do have a binary version of the readline library available. Can I
build the readline module for Windows usingh this? Ideally, I'd like to do
this *without* having to rebuild the whole of Python for Windows (I'm happy
with just using the normal build for everything else). I've never built Python
for myself, so apologies if the answer to this is "obvious"...

(BTW, this doesn't just apply to readline - I have various other libraries
such as a curses port, which I wouldn't mind trying to integrate as well...)

Thanks,
Paul


From benjamin.altman at noaa.gov  Wed Mar 21 16:07:54 2001
From: benjamin.altman at noaa.gov (Benjamin.Altman)
Date: Wed, 21 Mar 2001 16:07:54 -0500
Subject: learning python...
References: <3AB8F4E1.184DD9C@noaa.gov> <Z97u6.16733$Im6.1761313@newsread1.prod.itd.earthlink.net> <3ab90559_6@goliath.newsfeeds.com> <99b44s$m5j$1@panix6.panix.com>
Message-ID: <3AB9182A.8FEBCF3E@noaa.gov>

It would be good to know if the new edition really is all that different (other than
the topics added) to the 1st ed.  The only review on Amazon says he never read it...

Ben

Aahz Maruch wrote:

> In article <3ab90559_6 at goliath.newsfeeds.com>,
> Bubba Brains <chocolate at toblerone.yum> wrote:
> >
> >Forget Learning Python?  I think it's the best reference/tutorial around.  I
> >use python once every couple months, and sometimes forget the simple python
> >syntax.  The Learning Python book is great for getting up to speed fast, and
> >to look up common stuff.  The other OReilly book (Programming Python) is
> >absolutely annoying to use.  It's not even a good reference book (good luck
> >finding the info you want quickly).  Don't know about the Prentice book.
>
> If you look at http://www.oreilly.com/catalog/python2/ you'll see that
> Programming Python just got a second edition; I've heard that there were
> a lot of improvements.
> --
>                       --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)
>
> Androgynous poly kinky vanilla queer het Pythonista   http://www.rahul.net/aahz/
> Hugs and backrubs -- I break Rule 6
>
> "I won't accept a model of the universe in which free will, omniscient
> gods, and atheism are simultaneously true."  -- M



From mwh21 at cam.ac.uk  Tue Mar  6 14:00:51 2001
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 06 Mar 2001 19:00:51 +0000
Subject: Long Shifty Bits
References: <kmap6.15$8Q.28605@news.pacbell.net>
Message-ID: <m3itlmbu3g.fsf@atrus.jesus.cam.ac.uk>

Phlip <phlip_cpp at my-deja.com> writes:

> Not hyp:
> 
> I want this expression to make sense:
> 
>         x = 1L << 200
> 
> If someone tells me "it will roll over at the 32nd bit", or mayby
> even "it will roll over when the CPU damn well feels like it",
> that's okay. I'l just construct a Bit Vector of some ilk and drop my
> bits into it. But the above notation is closest to our problem
> space, so it would be nice if it worked as expected.

Works for me.  Have you tried it?

It's not the solution to use if your bit vectors are sparse.  Also if
you mutate your bitvectors a lot, you might want to use an array or
list instead.

Cheers,
M.

-- 
  It's a measure of how much I love Python that I moved to VA, where
  if things don't work out Guido will buy a plantation and put us to
  work harvesting peanuts instead.     -- Tim Peters, comp.lang.python


From donn at oz.net  Fri Mar 16 20:12:48 2001
From: donn at oz.net (Donn Cave)
Date: 17 Mar 2001 01:12:48 GMT
Subject: Should I close after popen??
References: <XFMail.20010315142704.shaleh@valinux.com> <mailman.984697040.2458.python-list@python.org> <m3lmq6ppv0.fsf@master.athome> <98satl$1t8$0@216.39.151.169> <mailman.984774997.30315.python-list@python.org>
Message-ID: <98udmg$552$0@216.39.151.169>

Quoth "Ken Seehof" <kens at sightreader.com>:
[ re files close after last reference ]
| More specifically, they close when garbage collected, which
| is at some unknown time -after- they go out of scope and no
| reference to them ramains on the system.  Usually this distinction
| doesn't matter, but there are some cases where this can get you
| into trouble, and the result could be a nasty intermittent bug.
|
| In general, never rely on destructor timing in python.  It was ok
| in C++, but it won't always work in python.

That's the party line, since the Java implementation can't
finalize on reference counts and for that matter there's no
guarantee that the C implementation always will.

But at present it does.  In C Python, you may rely on destructor
timing if you want, it will happen exactly when it should.
The garbage collection option that was introduced in 2.0 does
not change that.  You have to understand that it's not official,
but it's so useful that it's hard to pretend it isn't there.

	Donn Cave, donn at oz.net


From cribeiro at mail.inet.com.br  Thu Mar 29 15:52:27 2001
From: cribeiro at mail.inet.com.br (Carlos Alberto Reis Ribeiro)
Date: Thu, 29 Mar 2001 17:52:27 -0300
Subject: PEP 245: Python interfaces
In-Reply-To: <99vbm002b8m@news2.newsguy.com>
References: <Pine.LNX.4.32.0103222253510.26302-100000@localhost.localdomain>
 <99o9g4$nh3$0@216.39.170.247>
 <Pine.LNX.4.21.0103261430360.25260-100000@localhost.localdomain>
 <j4wv99m503.fsf@informatik.hu-berlin.de>
Message-ID: <5.0.2.1.0.20010329172450.009f3ec0@mail.inet.com.br>

At 15:00 29/03/01 +0200, Alex Martelli wrote:
>Fortunately, Python eschews such subtleties (which breed their own
>complications); adopting the explicit and very clear distinction that
>Java draws between 'extending' (a class) and 'implementing' (an interface)
>seems much preferable to me than either totally blurring (in the Python
>source) this crucial distinction, or adopting C++-inspired half-remedies
>to the is-a/vs/implementation confusion.

I entirely agree with you. One of the points that keeps bugging me is that 
inheritance is not the only way to build OO systems. There are other 
techniques, such as composition, the decorator pattern, proxies, that lead 
to a more natural class hierarchy in several cases. By keeping interfaces 
distinct from the class hierarchies, we are in fact making the support of 
these approaches much simpler and straighforward.

For example, there is multiple inheritance. I know very few legitimate 
cases where it is a good idea. Composition usually attains the same 
results, but with less side effects and much better control over the 
process. Interfaces are exceptionally well suited for this kind of construct.


Carlos Ribeiro





From MarkH at ActiveState.com  Sat Mar 10 21:39:18 2001
From: MarkH at ActiveState.com (Mark Hammond)
Date: Sun, 11 Mar 2001 02:39:18 GMT
Subject: kbhit()  and  getch() question
References: <taecoqaa2mot62@corp.supernews.com>
Message-ID: <3AAAE499.80001@ActiveState.com>

Jeremy Reed wrote:

> In a previous question, I asked how to get non-blocking input from the user.
> The answer proposed is as follows:
> 
> import msvcrt
> 
> ch = 0
> while ch != 'x':
>     if msvcrt.kbhit():
>         ch = msvcrt.getch()
>         print ch
> 
> I have run into a few problems.  First of all, the kbhit() doesn't seem to
> work.  I am unable to find a way for the function to return true.
> 
> Second, the getch() function is listed in the documentation as a blocking
> call, but it doesn't block when i run it in the following script.
> 
> ch = 0
> while ch != 'x':
>     ch = mscvrt.getch()
>     print ch
> 
> Instead, it prints the equivalent to chr(255) in an infinite loop.
> 
> Is there some setup call or initialization function that I'm neglecting to
> call?  Why aren't these function working like the documentation says they
> should?

These functions will only work from a "console" program.  If you are in 
a GUI, keystrokes are delivered via the message loop, and life gets a 
little more difficult.  You will need to give more details about your 
program.

Mark.




From phd at phd.pp.ru  Thu Mar 22 13:08:11 2001
From: phd at phd.pp.ru (Oleg Broytmann)
Date: Thu, 22 Mar 2001 21:08:11 +0300 (MSK)
Subject: Transparent GIFs in PIL
In-Reply-To: <20010322185348.A19190@freedom.puma-ag.com>
Message-ID: <Pine.LNX.4.30.0103222107100.31302-100000@fep132.fep.ru>

On Thu, 22 Mar 2001, Steve Purcell wrote:
> I just checked; the latest version of PIL (1.1.1) would also need that patch
> to support GIF transparency, as far as I can tell.

   /F is there (I've just exchanged a pair of e-mail with him) :)))

Oleg.
----
     Oleg Broytmann            http://phd.pp.ru/            phd at phd.pp.ru
           Programmers don't die, they just GOSUB without RETURN.




From shaleh at valinux.com  Fri Mar  2 11:40:07 2001
From: shaleh at valinux.com (Sean 'Shaleh' Perry)
Date: Fri, 02 Mar 2001 08:40:07 -0800 (PST)
Subject: gah! I hate the new string syntax
In-Reply-To: <71ju9to3jlp6l6u1t2peihlc7gegku3a99@4ax.com>
Message-ID: <XFMail.20010302084007.shaleh@valinux.com>

On 02-Mar-2001 Tim Roberts wrote:
> "Sean 'Shaleh' Perry" <shaleh at valinux.com> wrote:
> 
>>return ";".join(["%s=%s" % (k, params[k]) for k in params.keys()])
>>
>>every day python seems to be moving closer to the line noise aspect of
>>coding.
> 
> But join is really the only one of the string methods that comes out so
> awkwardly.  The rest of them are quite sensible.  And, at least for now,
> you do still have the traditional string.join alternative.
> 
> Wouldn't your code be slightly quicker if you used items() instead of
> looking up the individual keys:
> 
> return ";".join(["%s=%s" % it for it in params.items()])

That snippet came from 'Dive into Python' a book for python programmers.  The
style of code in there is not anything I would want a fellow coder to follow.

The point of all of this was 'join' especially, but a few of the others (it
brother split) make for really ugly code.

" ".split(list) is another I have seen recently.



From cfelling at iae.nl  Mon Mar 12 09:57:44 2001
From: cfelling at iae.nl (Carel Fellinger)
Date: 12 Mar 2001 15:57:44 +0100
Subject: I have no idea
References: <g13r6.201$qt5.10920@news3.mia>
Message-ID: <98io58$462$1@animus.fel.iae.nl>

Brendhan Horne <brendhanhorne at bellsouth.net> wrote:
> In basic you use to be able to do this:
> 10 print "Hello World"
> 20 goto 10
> Run

> If you did that it would run up the screen in an endless loop until you hit
> the ESC key. If you added   ; it would cover the whole screen. How do you
> write the equivalent in python?

I'm not much of a BASIC-guy and you're much better of reading the
online tutorials, but...

startup the python interpreter and then:

    >>> while 1:
    ...     print "Hello World"
    ...
    Hello World
    Hello World
    Hello World
    Traceback (innermost last):
      File "<stdin>", line 1, in ?
    KeyboardInterrupt
    
[[ The KeyboardInterrupt is provoked by pressing
   the control-key and the C-key simultaniously. ]]

or add a ',' to the print statement, like:

    >>> while 1:
    ...     print "Hello World"
    ...
    Hello World Hello World Hello World Hello World Hello World Hello World
    Traceback (innermost last):
      File "<stdin>", line 1, in ?
    KeyboardInterrupt
    
-- 
groetjes, carel


From emile at fenx.com  Tue Mar 13 09:31:36 2001
From: emile at fenx.com (Emile van Sebille)
Date: Tue, 13 Mar 2001 06:31:36 -0800
Subject: Converting a text data file from positional to tab delimited.
References: <B6D37840.2124%lee@joramo.com>
Message-ID: <98lb61$2ec0j$1@ID-11957.news.dfncis.de>

The obvious change is to use a list to collect the fields, then string.join
to build the record.  If you're on 1.5.2, import string first.

See notes below.  HTH

Emile van Sebille
emile at fenx.com

"Lee Joramo" <lee at joramo.com> wrote in message
news:B6D37840.2124%lee at joramo.com...
> I am looking for suggestions to speed up the process of converting a large
<snip>
> inFile = open('rawdata.dat', 'r')
> outFile = open('delimted.dat', 'w')
> while 1:
>     lines = inFile.readlines(1000)
>     if not lines: break
>     for line in lines:

replace this:
>         delimitedLine = ""
with this:
        delimitedLine = []

>         delimit = ""
>         for field in layout:
>             #
>             #can this loop be improved??
>             #

replace this:
>             fieldValue = line[field[1]:field[2]]
>             delimitedLine = delimitedLine + delimit + fieldValue
>             delimit = "\t"
with this:
            outrec.append(line[field[1]:field[2]])

and then replace this:
>         outFile.write(delimitedLine+"\n")
with this:
           outfile.write("\t'.join(outrec)+"\n")
or with 1.5.2:
            outfile.write(string.join(outrec,'\t')+"\n")

> inFile.close()
> outFile.close()
> del lines
>




From emile at fenx.com  Sun Mar 11 18:57:49 2001
From: emile at fenx.com (Emile van Sebille)
Date: Sun, 11 Mar 2001 15:57:49 -0800
Subject: random number between -maxint and +maxint
References: <K2Sq6.360684$Pm2.5406628@news20.bellglobal.com>
Message-ID: <98h3j7$20qan$1@ID-11957.news.dfncis.de>

Don,

I think this may do it, but I'm not sure if it stays random or catches the
endpoint properly.  Hopefully someone with more theory will jump in with an
answer.

import random, sys
print random.randrange(-sys.maxint/2, sys.maxint/2)*2 + random.choice([0,1])

Emile van Sebille
emile at fenx.com

"Dan Parisien" <dan at eevolved.com> wrote in message
news:K2Sq6.360684$Pm2.5406628 at news20.bellglobal.com...
> I can't figure out a good way to do it, does anyone know a better solution
> than
>
> from whrandom import whrandom
> rand = whrandom()
> max = 2**30 + (2**30-1) #one less than 2**31
> randomnumber = rand.randrange(0, max) - rand.randrange(0, max)
>
> thank you for your time,
> Dan




From stephen_purcell at yahoo.com  Fri Mar 30 04:53:19 2001
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Fri, 30 Mar 2001 11:53:19 +0200
Subject: Creation d'un NG francophone sur Python
In-Reply-To: <9a1e6f011es@enews3.newsguy.com>; from spencerm@cruzio.com on Thu, Mar 29, 2001 at 11:54:01PM -0800
References: <1eqzyk6.gxd4jr1tkjykzN%francois.granger@free.fr> <9a1e6f011es@enews3.newsguy.com>
Message-ID: <20010330115319.A6765@freedom.puma-ag.com>

Michael Spencer wrote:
> Pardon, mais pourquoi un NG pour Python en fran?ais?
> 
> Python a cr?? par un nederlandais qui a ?crit en anglais - pas sa langue
> natif. Il a ?crit en anglais ? assurer la plus grande distribution de ses
> id?es. Pourquoi on marcherait en arri?re aux temps quand chaque groupe
> a maintenu sa <<independance>> par s'isoler des autres?

Oh, come on.  If a bunch of Francophones (or any other linguistic group) want
to be able to ask each other Python questions in their native language, how is
that 'moving backwards'?  They have work to do, they often write their
code in their native language, and they don't have time to waste on
carefully formulating postings in foreign languages such as English.

Besides, if they post in their own languages to *this* newsgroup, they get
unhelpful responses like your suggestion that they step into line and
speak English instead.

-Steve

-- 
Steve Purcell, Pythangelist
Get testing at http://pyunit.sourceforge.net/
Any opinions expressed herein are my own and not necessarily those of Yahoo



From aahz at panix.com  Fri Mar  2 11:54:51 2001
From: aahz at panix.com (Aahz Maruch)
Date: 2 Mar 2001 08:54:51 -0800
Subject: [Distutils] Re: CPAN functionality for python - requirements
References: <doughellmann@home.com> <20010228091351.C4A41371690@snelboot.oratrix.nl> <mailman.983404511.31754.python-list@python.org>
Message-ID: <97oj8r$ffm$1@panix3.panix.com>

In article <mailman.983404511.31754.python-list at python.org>,
Sean Reifschneider  <jafo at tummy.com> wrote:
>
>I anticipate having a web-based interface to the database as well.  I'm
>sure that we can get the CGI to let the user do that sort of search, but
>at the moment it's restricted so that the user has to specify the name
>of the package they're looking for.  I figured dumping gigabytes across
>some user's line wouldn't be optimal.  ;-)

If a search returns more than, say, a hundred items, deliver just a
hundred to the result list.  (E.g. "Your search found 5000 items,
listing the first hundred.")
-- 
                      --- Aahz (Copyright 2001 by aahz at pobox.com)

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

Nostalgia just ain't what it used to be


From cjw at sympatico.ca  Mon Mar 26 09:16:46 2001
From: cjw at sympatico.ca (Colin J. Williams)
Date: Mon, 26 Mar 2001 14:16:46 GMT
Subject: Best Python documentation ->HTML?
References: <pg0qbt0e598g1ecnfj5lj0lpa1q1pqvagq@4ax.com>
Message-ID: <3ABE4200.3B5BBCF@sympatico.ca>

Brian,

You might consider Scite - developed by Neil Hodgson
from his Scintilla.  It provides the ability to export
to either HTML or rtf.

Colin W.

Brian Glendenning wrote:

> I am a Python newbie and am confused about what the best way to turn
> Python documentation into HTML is (i.e., a "javadoc" equivalent). If
> it could also make good printed documentation (e.g., page numbers) as
> well as HTML, all the better.
>
> Pointers and suggestions welcome! -
>
> Cheers,
> Brian Glendenning, bglenden at nrao.edu




From jstraw at towhee.com  Wed Mar 28 14:00:28 2001
From: jstraw at towhee.com (Jason Straw.)
Date: 28 Mar 2001 14:00:28 -0500
Subject: KeyBoardInterrupt error
Message-ID: <985806059.1221.2.camel@jlaptop.towhee.com>

What can trigger a Keyboard Interrupt?

I need Ctrl-C to be ignored and Ctrl-D to map to a function.

gettting a single char at a time is a bad idea for my project


-- 
Jason Straw
jstraw at towhee.com OpenClassroom Project Developer
Yorktown High School Linux Users' Group Meetings Coordinator / yhslug.tux.org
What do IBM, $1,000,000,000 and Linux have to do with each other? 
Find out April 20th in Arlington, VA at the Yorktown High School Linux Users' Group





From jeremy at alum.mit.edu  Thu Mar  1 12:04:31 2001
From: jeremy at alum.mit.edu (Jeremy Hylton)
Date: Thu, 1 Mar 2001 12:04:31 -0500 (EST)
Subject: PEP 227 (was Re: Nested scopes resolution -- you can breathe again!)
In-Reply-To: <97lqsr0311d@news2.newsguy.com>
References: <LNBBLJKPBEHFEDALKOLCMEEPJBAA.tim.one@home.com>
	<200102230259.VAA19238@cj20424-a.reston1.va.home.com>
	<mailman.983384008.21434.python-list@python.org>
	<97lqsr0311d@news2.newsguy.com>
Message-ID: <15006.33055.663211.136994@w221.z064000254.bwi-md.dsl.cnc.net>

Or: global NAME in FUNCTION

def eggs():
    z = 2
    def spam():
        def inner():
            global x in spam, z in eggs
            x += 1
            z += x
        x = 23
        inner()

Jeremy



From aleaxit at yahoo.com  Wed Mar 28 14:29:30 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Wed, 28 Mar 2001 21:29:30 +0200
Subject: list vs tuple
References: <mailman.985800580.9546.python-list@python.org>
Message-ID: <99te7202o93@news1.newsguy.com>

"Brian Quinlan" <BrianQ at ActiveState.com> wrote in message
news:mailman.985800580.9546.python-list at python.org...
    [referring to the snippet:
> dir_list = ('/path', '/foo/bar', '/baz/bat/files')
> dir_list shouldn't change and this documents it.
    ]

> I'm not sure that this is a reasonable piece of self documentation since
> dir_list can change. It is the tuple ('/path', '/foo/bar',
'/baz/bat/files')
> that cannot change. Would you say that x = 5 documents that x cannot
change?

It's a common, though sometimes confusing, shorthand, to say
'variablename' meaning "the object (which really, _intrinsically_,
has no particular name, of course) to which the variable named
variablename is right now (temporarily) bound".  It does save
lots of ink^H^H^H bits -- as long as both parties to the given
conversation have enough Python knowledge to grasp that any
variable name can always be re-bound at any time, thus, any
assertions are no doubt meant to refer to the object that variable
is currently bound to.


Alex






From qrczak at knm.org.pl  Sat Mar 17 12:46:43 2001
From: qrczak at knm.org.pl (Marcin 'Qrczak' Kowalczyk)
Date: 17 Mar 2001 17:46:43 GMT
Subject: I come to praise .join, not to bury it...
References: <mailman.983485923.1136.python-list@python.org> <97p7iq$kk2$1@nntp6.u.washington.edu> <97rka002k9@news1.newsguy.com> <3AAF7E47.294617E1@sage.att.com> <98qeco02hn7@news1.newsguy.com> <3AB106CD.7D259EBC@san.rr.com> <mailman.984685357.7864.python-list@python.org>
Message-ID: <slrn9b78o3.3qv.qrczak@qrnik.zagroda>

Thu, 15 Mar 2001 11:47:10 -0600, Ken Seehof <kens at sightreader.com> pisze:

> Hmm ... I wonder what I'd do if I could overload __if__ ...

Or the newline operator which separates statements.
Haskell's monads are essentially that :-)

-- 
 __("<  Marcin Kowalczyk * qrczak at knm.org.pl http://qrczak.ids.net.pl/
 \__/
  ^^                      SYGNATURA ZAST?PCZA
QRCZAK


From whisper at oz.net  Tue Mar 20 22:49:22 2001
From: whisper at oz.net (Dave LeBlanc)
Date: 21 Mar 2001 03:49:22 GMT
Subject: Little red lines in PythonWin!
References: <mailman.985107113.4466.python-list@python.org> <uu24ojk5h.fsf@ctwd0143.fitlinxx.com>
Message-ID: <9998c2$a16$0@216.39.170.247>

>"... if spaces get mixed in it's bound to cause problems.  So in
>general that's nasty..."

Urban legend perhaps?

Seems like, so long as you actually indent dependent lines, such as
those comprising the body of an if statement, as little as one space
or tab character then you're legal.

So long as a text processor expands tabs to the same number of spaces,
then I don't see how any tab+space sequence that only varies by the
number of explicit space characters can ever possibly result in the
same or less whitespace to the left of text on a line. This makes it
hard to understand why one wouldn't want to use a mix of tabs and
spaces, especially "because of the nature of text editors on non-UNIX
platforms" (quote from 2.1.7).

\t\t\t\s\s\s and \s\s\s\t\t\t where \t --> \s\s\s\s (and \s represents
a single space character) will always create the same amount of space
in any text editor (as contrasted to a word processor) i've ever used
on a wide variety of OSs, including places where sp != 0x20 and tab !=
0x09 (EBCDIC et al.). Even text editors that allow the use of
proportional fonts seem to have a fixed idea of the screen area taken
up by a space character. (People who use word processors for text
editing, especially source code, deserve what they get.)

Frankly, using that many spaces (8) to indent seems like a waste of
screen real estate, not to mention making the code actually harder to
read. Perhaps 8 spaces/tab made sense when the primary output medium
was a decwriter or a tty, which was the last time I used so many
spaces to replace a tab, but not anymore.

Just my opinion...

Dave LeBlanc

On 20 Mar 2001 14:49:14 -0500, David Bolen <db3l at fitlinxx.com> wrote:

>In theory you really shouldn't get into trouble using tabs as long as
>you follow Python's parser definition of what a tab is (treat as
>moving to next multiple of 8 - Language Reference 2.1.7).  The trouble
>is mostly when an editor is using an explicit tab character but
>internally adjusting its dimension to something other than that rule.
>Even that might work as long as only tabs are used for indentation,
>but if spaces get mixed in it's bound to cause problems.  So in
>general that's nasty, but as long as you untabify in that same
>editor/IDE before distribution, since it knows its own rules, the
>untabified version should be fine.
>



From hzhu at mars.localdomain  Wed Mar 21 04:15:50 2001
From: hzhu at mars.localdomain (Huaiyu Zhu)
Date: Wed, 21 Mar 2001 09:15:50 GMT
Subject: Yet Another PEP:  Query Protocol Interface or __query__
References: <Pine.LNX.4.21.0103201837570.21343-100000@clarkevans.com> <998r4j02vmb@news2.newsguy.com> <Pine.LNX.4.21.0103202032080.14634-100000@clarkevans.com> <slrn9bgprm.rv.hzhu@mars.localdomain>
Message-ID: <slrn9bgrsn.rv.hzhu@mars.localdomain>

>I have an implementation of this idea at
>
>http://www.geocities.com/huaiyu_zhu/python/Interface/typecheck.txt

OK, let me just post them here.  The following is a description followed by
a pure python implementation.


			Type checking system in Python

Classes implement a hierarchy of inheritance of attributes (variables and
methods).  They are useful for implementation modularization.

There is another type of modularization that is similar in nature, but is
independent of classes.  This is the hierarchy of polymorphic interfaces.

Python supports polymorphism in a very liberal way: if it walks like a duck
and quacks like a duck, then it must be a duck.  The archetypical examples
is that one can usually substitute a writable file with a user defined class
object that happens to have a write() methods with appropriate property.

In many programs it is necessary to ensure that a certain objects has
certain properties.  But how can we express the idea that a function expect
a file-like object, while that can mean an endless list of things, many have
not yet been defined?

One solution promoted by some other languages is to define some abstract
classes for type checking purpose, and require them to be inherited by any
class that behaves like them.  This is like what interfaces do in Java.
This solution is not satisfactory because

- It breaks polymorphism.  One has to harness the property at the time the
  classes are defined, instead of when they are used.  For example, if an
  application requires a property of "ReadableFile" which is satisfied by
  sys.stdin, it is difficult to retrofit the definition of sys.stdin to be
  an instance of such.

- It mixes two incompatible inheritances.  For example, suppose that class A
  has method A.m() but its subclass B overrides it with B.m().  The property
  A satisfies is not necessarily satisfied by B.

>From the above it is clear that there is a useful conceptual hierarchy that
is independent of the class hierarchy.  For lack of a better name let us
just call them Types.  The overridability means that classes do not imply
Types, and the polymorphism means that Types do not imply classes.  Their
main conceptual difference is that

- classes embody implementations
- Types embody requirements

Another way to describe the difference is

- classes are inherited by specialization
- Types are inherited by abstraction

For example, if we specify that A is a superclass of B, this does not change
anything about A, but it does mean that B will have all the attributes of A.
In contrast, if we specify that A is a superType of B, this does not change
anything about B, but it does mean that objects accepted as B must now be
accepted as A as well.  As a concrete example, consider a function expecting
a Number as an argument.  If we declare that 2.3 is a Float, and Float is a
special case of Number, then the function should accept 2.3 as argument.

Therefore, Types should have their own hierarchy, possibly with its own
syntax, that is independent of classes.  

Since Python is a very dynamically typed system, it is also desirable to
make the type checking as flexible as possible, and completely voluntary,
while being able to deliver the same level of guarantees that a strongly and
statically typed language would give.

Type can also be used to express properties like Immutable, which could only
be guaranteed by a compiler.  These properties will also be very useful for
compiler optimization.

All said, there is a role for classes to play in this affair - It is
possible to fake such a Type system using a class, with some caveats:

- It is dynamic, and very inefficient at run time
- It is dynamic, and can be circumvented in various ways.

Nevertheless, this dynamical type-checking does provide a fine illustration
of the concepts discussed above.  If and when they can be coded in compiler
it will be of much greater use.

==================================================================

#!/usr/bin/env python
# $Id: test.py,v 1.1.1.1 2000/08/29 00:38:54 hzhu Exp $
"""
A simulated type checking system
Types are inherited by abstraction

"""

class Type:
	def __init__(self, name):
		"Define a new abstract Type"
		self.name = name
		self.types = []
		self.classes = []
		self.subs = []
		self.attrs = []
		self.objs = []
		
	def include(self, obj):
		"Check to see if object is of my Type"
		
		# If it is explicitly included objects
		for o in self.objs:
			if obj == o: return 1

		# If it is of the right type or class
		if type(obj) is type(Type):
			for c in self.classes:
				if isinstance(obj, c): return 1
		else:
			for t in self.types:
				if type(obj) == t: return 1

		# If it is one of our sub-Type
		for s in self.subs:
			if s.include(obj): return 1
		else:
			return 0

	def satisfiedby(self, obj):
		"""Check to see if object has all the required attributes
		Is this necessary? """
		for a in self.attrs:
			if not hasattr(obj, a):
				print "%s has no attr %s" %(`obj`, a)
				raise AssertionError

	def add_type(self, t):
		if t not in self.types: self.types.append(t)

	def add_class(self, c):
		if c not in self.classes: self.classes.append(c)

	def add_sub(self, s):
		if s not in self.subs: self.subs.append(s)

	def add_attr(self, m):
		if m not in self.attrs: self.attrs.append(m)

	def add_obj(self, o):
		if o not in self.objs: self.objs.append(o)

	def add_typeof(self, o): self.add_type(type(o))
	def add_classof(self, o): self.add_class(o.__class__)
	def add_attrof(self, o):
		for attr in dir(o):		self.add_attr(attr)

	def __repr__(self): return "Type(%s)" % `self.name`


def require(x, t):
	try:
		assert t.include(x)
	except AssertionError:
		print "%s is not of %s" % (`x`, t)
		raise
	t.satisfiedby(x)

def require_either(x, list):
	for t in list:
		if t.include(x):
			t.satisfiedby(x)
			return
	else:
		print "%s is not of any Type in %s" % (`x`, list)
		raise AssertionError

def require_all(x, list):
	for t in list: require(x,t)


#------------------------------------------------------------------
String = Type("String")
String.add_typeof("")

Integer = Type("Integer")
Integer.add_typeof(1)
Integer.add_typeof(1L)

Number = Type("Number")
Number.add_sub(Integer)

Float = Type("Float")
Float.add_typeof(0.0)
Number.add_sub(Float)

File = Type("File")
File.add_attr("close")

#------------------------------------------------------------------
if __name__ == "__main__":

	def f(x):
		require(x, Number)
		print x*2

	f(1)
	f(2.3)

	def f(x):
		require_either(x, (Number, String))
		print x*2
		
	f("2")

	import sys
	File.add_obj(sys.stdout)
	require(sys.stdout, File)

	require_either("2", (Number, String))
	#require_all("2", (Number, String))






From phd at phd.fep.ru  Fri Mar 30 05:43:50 2001
From: phd at phd.fep.ru (Oleg Broytmann)
Date: Fri, 30 Mar 2001 14:43:50 +0400 (MSD)
Subject: HELP urllib POST can't get it to work
In-Reply-To: <9a1m91$fdg$1@pheidippides.axion.bt.co.uk>
Message-ID: <Pine.LNX.4.33.0103301441080.29458-100000@phd.fep.ru>

On Fri, 30 Mar 2001, mike mcleod wrote:
> import urllib
> params = urllib.urlencode({'address': '07', 'identity': 'me',
>                            'resultpg':'html/sendresult.html', 'message':
> 'test',
>                            'fPage':'true',
>                            })
> f = urllib.urlopen("http://www.samemachine.uk/bin/afunc.pl?", params)
> print f.read()
>
> It looks like the order of the params come out different in the urlopen.

   This is because it uses dictionary, which may (and usually does) changes
the order of keys.

> But because the server will only accept in the order shown it fails.

   Wow! what a foolish server :)

> I have tried to do this using basic sockets but don't know how to get that
> work without lots of reading.

   I have a very little patch for urlencode that allows to pass lists, not
only dicts. My version is this:

# obj must be either dict or list of tuples [(k1, v1), (k2, v2), ...] or list of lists [[k1, v1], [k2, v2], ...]
def urlencode(obj):
     l = []
     if type(obj) == type({}):
         obj = obj.items()

     for k, v in obj:
         k = quote_plus(str(k))
         v = quote_plus(str(v))
         l.append(k + '=' + v)

     return string.join(l, '&')

Oleg.
----
     Oleg Broytmann            http://phd.pp.ru/            phd at phd.pp.ru
           Programmers don't die, they just GOSUB without RETURN.




From costas at meezon.com  Sat Mar 10 23:58:05 2001
From: costas at meezon.com (Costas Menico)
Date: Sun, 11 Mar 2001 04:58:05 GMT
Subject: Pythonwin default directory init
Message-ID: <3aab04bc.2359756@News.CIS.DFN.DE>

I use Pythonwin and I would like it if the File/Open or File/Save
would default to a directory of my choice when I start it up.

I have tried making a startup module to set the paths of sys but that
doesn't seem to work. I have looked thru the .py files trying to
figure out where I could fix this and can't find it.

Also setting it up as a shortcut on the Desktop defaults to the
Desktop. 

Does anyone know how to do this? 

TIA

Costas


From ws-news at gmx.at  Thu Mar  8 05:31:35 2001
From: ws-news at gmx.at (Werner Schiendl)
Date: Thu, 8 Mar 2001 11:31:35 +0100
Subject: learning C
References: <mailman.983972829.23481.python-list@python.org> <983975751.18833@newsmaster-04.atnet.at> <slrn9adklr.bv9.scarblac@pino.selwerd.nl>
Message-ID: <984047775.592522@newsmaster-04.atnet.at>

Remco Gerlich <scarblac at pino.selwerd.nl> wrote in message
news:slrn9adklr.bv9.scarblac at pino.selwerd.nl...
> Werner Schiendl <ws-news at gmx.at> wrote in comp.lang.python:
> > > I assumed the original poster wanted to become a good C code writer,
or at
> > > least a good C code reader.  You tell him to learn C++ instead, as if
> > learning
> > > C is obselete.
> >
> > If it is not obsolete, it should be at least considered deprecated for
new
> > projects.
>
> A baseless statement. There are many types of project for which C is still
> the best choice. For one thing, think embedded systems.
>

Would say, depends.

If you have some really simple thing to do (and those are much rarer breed
than some believe...) you are probably better of not using C++ extended
features, libraries and syntax. Especially if the code has to be inspected
externally for safety reasons.

But if you have a need for string handling, containers, dynamic object
creation and deletion, etc. you can of course do that in C, but doing it in
C++ will get you up and running much faster.

I personally use embedded platforms and C++ is readily available there
(VxWorks, to be precise).

> I can imagine something like Java deprecating C++, but C isn't going to go
> away soon.
>

I probably already would have done, if Sun did not insist on the
braindamaged approach of running a VM (even on a device that does not even
have connectors where you could download additional classes).
Considering for example the upcoming release of GNU Java compiler I think
this will become _much_ more realistic...

Consider embedded devices :-)
For safety issues I would clearly prefer Java to C++

regards
werner





From dbrueck at edgix.com  Mon Mar 19 12:31:08 2001
From: dbrueck at edgix.com (Dave Brueck)
Date: Mon, 19 Mar 2001 10:31:08 -0700
Subject: Killing threads
In-Reply-To: <995c4l$gn7$1@panix3.panix.com>
Message-ID: <BLEOJGEMBNBNEAKMIFDHOEBCPNAA.dbrueck@edgix.com>

> >> I'm confused.  Is there some reason you're not using Queue.Queue()?
> >
> >Queue is overkill for simple one-to-one workflows - it's built to be
> >multi-consumer and multi-producer so every get/put/check
> requires acquiring
> >and releasing the Queue's mutex (granted, in most cases that
> probably won't
> >make too big of a different, but...).
>
> That's partly true; there are still single producer/consumer cases where
> you need a mutex.  The only reason you can get away with it in Python is
> because the GIL (global interpreter lock) does a lot of the work for
> you.  The flip side is that the GIL means a mutex really adds little
> overhead in the single producer/consumer case.

Depending on your OS it still probably means an extra pair of kernel calls
for each Queue access. To me, in that particular case, it was worth it. To
you, it wasn't. No big deal.

> >Anyway, the point here is killing separate threads and my WorkQueue class
> >was that was the first example that came to mind. Note that
> whether or not
> >you use Queue.Queue makes little difference in solving this problem:
>
> Not overall, no, but it does mean you don't have to muck with mutexes.

Semaphores... and they are so simple to use I'd hardly consider it
"mucking".

> I'm a big believer in simple code.  One problem with the code you posted
> is that it has a polling loop (with time.sleep()) in it; that's not good
> for efficiency.  Use another queue!  ;-)

Huh? The poll isn't during the wait-for-work part of the class. This is a
polling loop during Stop(), when you're shutting down (most likely
destroying) the work queue, and is there to ensure a clean shutdown. The
sleep() call usually happens 0 or 1 times and no more. I too am a big
believer in simple (and intuitive code).

Using a queue (vs. sleep) there would arguably be bad for two reasons:
1 - it is NOT more efficient to use a queue (create/wait/signal/destroy a
queue vs. 0 or 1 sleep calls). Even if avoiding sleep() _was_ more
efficient, who cares? You're essentially optimizing a function that normally
is called once in the entire lifetime of a WorkQueue. For that matter,
there's plenty of other useless micro-optimizations to do as well.

2 (** much more important **) - it would NOT make the code simpler - you're
calling Stop to shutdown the worker thread so you set its flag to signal the
"stop working" state, wait a little for it to respond, and move on. Not only
is the sleep method simpler, it is more intuitive (or maybe I'm not seeing
your point.. can you elaborate on how using a queue is a simpler and/or more
intuitive approach?).

> I'm also not a big fan of dispatching threads on functions/methods.
> It's almost always clearer to just subclass from threading.Thread.

That's ok, I wasn't starting a fan club <0.6 wink> (to me it is at best
equally clear to subclass threading.Thread, and less so in this case).

-Dave




From ykingma at accessforall.nl  Fri Mar 30 14:19:47 2001
From: ykingma at accessforall.nl (Ype Kingma)
Date: Fri, 30 Mar 2001 20:19:47 +0100
Subject: Why not an __assign__ method?
References: <mailman.985969720.18484.python-list@python.org>
Message-ID: <3AC4DC4C.63383075@accessforall.nl>

Carlos,

<intermezzo>

> 
> p.s. Maybe __assign__ is not a good name; maybe something like __onassign__
> is better, but this does not change the heart of the proposal.
> 
> Carlos Ribeiro

Did you know that there is a programming language called Sartre that
has a special object that can be __unassign__'d from?

Hmmm, maybe it is possible to interpret Sartre in Python.

</intermezzo>

Regards,
Ype

-- 
email at xs4all.nl


From hamish_lawson at yahoo.co.uk  Fri Mar 30 17:31:36 2001
From: hamish_lawson at yahoo.co.uk (=?iso-8859-1?q?Hamish=20Lawson?=)
Date: Fri, 30 Mar 2001 23:31:36 +0100 (BST)
Subject: Know of Substantial Apps Written in Python?
Message-ID: <20010330223136.14277.qmail@web207.mail.yahoo.com>

Of course when we say 'substantial', we have to remember Python's claim
to reduce the amount of code and development time required (whether
because of the qualities of the language itself or its extensive
library). Perhaps the question ought to be rephrased "Know of apps
written in Python which in other languages would be substantial?" ;-)

Hamish Lawson


____________________________________________________________
Do You Yahoo!?
Get your free @yahoo.co.uk address at http://mail.yahoo.co.uk
or your free @yahoo.ie address at http://mail.yahoo.ie



From gabriel_ambuehl-py at buz.ch  Sat Mar  3 09:51:33 2001
From: gabriel_ambuehl-py at buz.ch (Gabriel Ambuehl)
Date: Sat, 3 Mar 2001 15:51:33 +0100
Subject: syslog broken on FreeBSD?
Message-ID: <10981466442.20010303155133@buz.ch>

-----BEGIN PGP SIGNED MESSAGE-----

Hello,
I've been trying to write to syslog using the following
import syslog
syslog("TEST)
but there won't ever appear any message in the syslog. Is this module
broken on FreeBSD 4.2 or am I doing something very wrong?

Any comments would be greatly appreciated.




TIA & best regards,
 Gabriel

-----BEGIN PGP SIGNATURE-----
Version: PGP 6.0.2i

iQEVAwUBOqD27MZa2WpymlDxAQGYsgf/aQ3M0Odgs/LDGwlDQ7wGWXPq+/8l8him
pTcNNsXPGzUDWCM4QhbeAs5fvktVxgh3JqIu23yre1G1VY+jzqrlKRDhQoTL/Bh2
7WINOLUO1H85lEWpQuIVNg9y9eJC6rjUzYVXOe+1cf8BOzMyjBDUaiYQP+lPWqNQ
yDlu+VDWGTke0pllruf7xNJVShtMjE4QINYTkkbXZndraIDdaQQl4YmirP83iomD
iFYzsO5roifCgzhaSd7yBTbxHx+qnm9fucc2E7157RvHpFNfyio5ANFkY5kRIfi9
JdceCnfDBFF9/wlHffg536Yjg2XEMCh2XcqBXxNqE7kAHwoDMzx7dw==
=czRi
-----END PGP SIGNATURE-----





From mysticsguy at yahoo.ca  Thu Mar 22 22:38:45 2001
From: mysticsguy at yahoo.ca (Yvon Boulianne)
Date: Fri, 23 Mar 2001 03:38:45 GMT
Subject: python, perl, lisp, tcl / got a decision to make, can you give me a  view point please ? and a joke for python users :)
References: <3aba3076.14395287@news.vif.com> <3ABAC36C.43FE03B7@SpamNOnotheworld.com>
Message-ID: <3abac4ce.19611504@news.vif.com>

In fact i took it there
http://www.my-opensource.org/lists/myoss/1999-11/msg00044.html
sorry to not having mention it, wasn't seeing the point about that,
but maybe it's important for some.  
after a dig,  it seem to come from a list of joke at
http://www.netfunny.com/rhf/   who come from the newsgroup
rec.humour.funny 

With Love & Counsciousness
Yvon


On Fri, 23 Mar 2001 03:30:45 GMT, David Lees
<deblNoNo at SpamNOnotheworld.com> wrote:

>You might want to attribute the source that you copied this joke from
>which is at:
>
>http://www.python.org/doc/Humor.html
>
>david lees

Yvon Boulianne ; Saphi
mysticsguy at yahoo.ca
www.monunivers.net
Humour,Paradox & Changements


From tjg at exceptionalminds.com  Fri Mar 16 16:47:30 2001
From: tjg at exceptionalminds.com (Timothy Grant)
Date: Fri, 16 Mar 2001 13:47:30 -0800
Subject: md5 passwords
Message-ID: <20010316134730.K3109@trufflehunter.avalongroup.net>

Can anyone explain to me how to use the md5 module to generate
passwords for the shadow password file?

I would like to be able to be able to os.system('useradd -p
md5/crypt username')

I can generate the required crypt hashes no problem, but can't
seem to figure out how to generate and md5 hash that looks
anything like what I see in /etc/shadow.

-- 
Stand Fast,
    tjg.

Timothy Grant                         tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.         <><       (503) 246-3630
>>>>>>>>>>>>>Linux, because rebooting is *NOT* normal<<<<<<<<<
>>>>This machine was last rebooted:  59 days  1:59 hours ago<<



From will at nospam.misconception.org.uk  Sun Mar 18 17:45:11 2001
From: will at nospam.misconception.org.uk (Will Newton)
Date: Sun, 18 Mar 2001 22:45:11 +0000
Subject: Subsetting a dictionary
Message-ID: <993dkm$ttr$1@news6.svr.pol.co.uk>

I need to get a subset of a dictionary:

D = Dictionary of objects (in this case, training examples for a decision 
tree)

D2 = {D | d in D, d under some constraint}

I need to be able to do this, but as it is temporary I do not want to 
destructively remove examples, so:

D.subset(c)
will return a new dictionary object that is equivalent to D under 
constraint c.

I have code that does this, but it's very slow, consuming typically 75% of 
program runtime - it is called many times, which seems to be unavoidable 
without modifying the algorithm (ID3) too far.

I currently have:

class TrainingData:
         data = {}
        def subset(self, attr, value):
                # Copy semantics
 
                td = TrainingData()
 
                td.data = self.data.copy()
 
                for item in td.data.keys():
                         if item[attr] != value:
                                   del td.data[item]

                return td

Can anyone see how I can speed this up?

I could conceivably do this with a list and use filter, I'll see how that 
works.


From tuttledon at hotmail.com  Sun Mar  4 11:24:14 2001
From: tuttledon at hotmail.com (Don Tuttle)
Date: Sun, 04 Mar 2001 16:24:14 GMT
Subject: [Python-Dev] Re: Bug fix releases
References: <mailman.983681106.14778.python-list@python.org>
Message-ID: <O_to6.21565$__6.5852201@typhoon.southeast.rr.com>

> [Don Tuttle]
> > But there is the CVS. And there are those like myself  who are CVS
> > impaired.  < list of desired features>
> >
>[Tim Peters]
> It's
> also curious that none of these have to do with bugfixes, they're all
> "shinier feature" things -- exactly what Aahz is telling us users *don't*
> want between major releases.  Good thing I don't believe anything you read
> <wink>.

Alright, I've betrayed myself as a features kind of guy, raised on "New and
Improved". But for me bug fixes count as "New and Improved".   I didn't
mention any specific bugs because I don't know where they exist.  But whever
they are, I want them fixed! (You know, typical end user behavior. ;-)

To me a truly new feature is something that either 1) breaks existing code
or 2) causes book publishers to spasm it fits of rage--"The damn book's not
even in the stores and it's already out of date!)

> [Don Tuttle]
> > I don't think it has to be as formal as an official  pep or separate
> > development track.  Just whenever our fearless dictator sees that
there's
> > enough good stuff in the CVS, let him decree that a new build
> > shall be made.
>
>[Tim Peters]
> If we could, barking a decree is easy but isn't enough to make a release
> *happen*.  I put in at least 80 hours during the week of a release, and
I'm
> only looking out for Windows (+ masses of general patches and bugfixes and
> email panics, etc).  The other PythonLabs guys aren't on vacation either.

As a Windows user I thank you.

>[Tim Peters]
> Somebody has to step up and do *work*.  But, historically, nobody
volunteers.
> If you look at Misc/HISTORY, you may be surprised to discover that the
most
> recent "pure bugfix release" was in November of 1994(!  and despite what
it
> was *called*, it snuck some new Tkinter features anyway).

Perhaps this is my ignorance of CVS, but what I had in mind was to take the
existing release source, replace offending code with update(s) from CVS,
compile, package, release. What's the offending code?  That's up to Guido.

As for who gets to compile, package, and release...well, he is your boss,
right? <wink>

Don




From stephen_purcell at yahoo.com  Thu Mar 15 03:44:37 2001
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Thu, 15 Mar 2001 09:44:37 +0100
Subject: Better string.translate?
In-Reply-To: <3ab0749f$1_3@news.zet.net>; from stowasser.h@idowa.de on Thu, Mar 15, 2001 at 08:53:02AM +0100
References: <3aaf758f$1_2@news.zet.net> <mailman.984590438.506.python-list@python.org> <3ab0749f$1_3@news.zet.net>
Message-ID: <20010315094437.B1581@freedom.puma-ag.com>

Stowasser Harald wrote:
> "Steve Purcell" <stephen_purcell at yahoo.com> schrieb :
> > Use the 'sub' function in the 're' module:
> 
> Thank you, but i don't want do delete the characters. I need to
> translate them into Spaces.

So change the '' replacement string into ' ' in the code I posted.

You could try (shock! horror!) reading the documentation for the 're.sub'
function. It'll take a long time to finish your code if you have to rely
on the mailing list to show you everything one character at a time. <wink>

   http://www.python.org/doc/lib/Contents_of_Module_re.html

-Steve

-- 
Steve Purcell, Pythangelist
Get testing at http://pyunit.sourceforge.net/
Any opinions expressed herein are my own and not necessarily those of Yahoo



From costas at meezon.com  Sat Mar 10 09:52:08 2001
From: costas at meezon.com (Costas Menico)
Date: Sat, 10 Mar 2001 14:52:08 GMT
Subject: import headaches/gripes
Message-ID: <3aaa3f94.15940553@News.CIS.DFN.DE>

I am using Pythonwin to develop apps.  I usually have a few modules
open. My main module imports all the necessary files at the start.

However when I change a module, I have to remember to reload() it
manually or quit Pythonwin. 

Why can't the import statement be smart enough to check if a file was
changed to import it anyway? All it has to do is compare .py to the
last import date/time.

Given my gripe is there some way I can force imports? It seems the
only way to do this is somehow to use reload(module name). Maybe some
logic like this would work:

if module_exists(utils):
	reload(utils)
else:
	import utils


where module_exists() is some imaginary function that can check if the
module is loaded.

Any ideas would be appreciated.

Costas



From sholden at holdenweb.com  Tue Mar 20 00:06:49 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Tue, 20 Mar 2001 00:06:49 -0500
Subject: [Very Long (11K)] Numeric PEPs, first public posts
References: <mailman.984748941.31248.python-list@python.org> <slrn9bconk.ns2.qrczak@qrnik.zagroda>
Message-ID: <1DBt6.120680$ML1.5626083@e420r-atl2.usenetserver.com>

"Marcin 'Qrczak' Kowalczyk" <qrczak at knm.org.pl> wrote in message
news:slrn9bconk.ns2.qrczak at qrnik.zagroda...
> Fri, 16 Mar 2001 15:18:43 +0200, Moshe Zadka <moshez at zadka.site.co.il>
pisze:
>
> > PEP: 237
> > Title: Unifying Long Integers and Integers
>
> I like it!
>
[enthusiastic commentary omitted].
>
> >     What to do about sys.maxint?
>
Why not return floating-point infinity? <0.25 wink>

> Deprecate like the long function etc., unless some important libraries
> would still be limited to that.
>
> >     What to do about PyInt_AS_LONG failures?
>
> A good question.
>
> >     What do do about %u, %o, %x formatting operators?
>
> If portability here is not that important, let them output negative
> numbers as negative. One can use "%x" % (x % 2**32) to get the old
> behavior.
>
So, longs become variable-length two's-complement integers?

> If it is, I don't know.
>
> >     How to warn about << not cutting integers?
>
> If there is a portability switch like from __future__ import foo,
> use that; it's bad that it doesn't happen at compile time, but maybe
> it's all that we can get. If not, I don't know.
>
How about:

    from __future__ import unified_field_theory

?

> >     Should the overflow warning be on a portable maximum size?
>
> I would drop it.
>
> >     Will unification of types and classes help with a more
> >     straightforward implementations?
>
> I don't know.
>
> > PEP: 238
> > Title: Non-integer Division
>
> I like it!
>
> > Open Issues
> >
> >     Should the // operator be renamed to "div"?
>
> If % is % and not "mod", I would probably spell it //. But I would
> call % "mod", except that it's too late.
>
No it isn't. It's never too late to _suggest_ a change :-) I seem to
remember suggesting that "print >>" could be omitted without making me cry.


> > PEP: 239
> > Title: Adding a Rational Type to Python
>
> I like it!
>
> I guess that sin(1.5) converts the argument to float?
>
Well, it shouldn't need to, unless the default numeric literal is a
rational: 1.5 already _is_ floating point, isn't it?

> How many digits does str(1/3) return, assuming it prints the number
> in decimal?
>
Forty-two. This might even have been the original question to which this is
the answer.

> Should str(5.0) return '5.0' or '5'? Note that it finally doesn't
> hurt if you have an integer instead of a rational, so there is more
> sense to let it return '5' than currently: integers can be treated
> as a special case of rationals (implemented as a separate type).
> But maybe accepting 4.0 as list indices etc. is too much.
>
This seems to conflict with dictionary semantics, where for all integer i,

    dict.has_key(i) == dict.has_key(float(i))

> What does repr(1/3) return? '1/3'?
>
Unless you think '20/60' or '12.7/38.1' might be better: it would seem that
the canonical representation of rationals would use remove a factor of
GCD(numerator, denominator). But, then, this is not my specialty.

> > PEP: 240
> > Title: Adding a Rational Literal to Python
>
> I like it!
>
> All PEPs together make arithmetic on literals exact, independent of
> the platform's integer size and float representation. Moreover one can
> always write 42 instead of 42.0 and get the same result (numerically).
>
I really don't want my languages deciding what representations to use for my
data: otherwise, what happens to guarantees that they can be persistent in
bounded storage?

> BTW. A nice thing about static typing is that it allowed Haskell's
> numeric literals and functions to have types deduced from the context
> of usage. 5+1/3 can be computed on either Rational or Double or
> Complex Float or any other (perhaps user defined) fractional type.
> Literals are internally converted from Integer (infinite precision)
> or Rational (= Ratio Integer). In Python it can't be done that way,
> but the conversion can be delayed such that the results are the same
> (modulo the *lack* of *loss* of precision if numbers are converted
> to float too late).
>
Surely rationals with infinite-length components will always be more
precises than your average floating-point number for strictly rational
values?

or-should-we-go-the-whole-hog-and-use-a-long-for-the-floating-
    point-characteristic-as-well-ly y'rs  - steve





From ransen_spam_me_not at nemo.it  Sat Mar 10 01:08:39 2001
From: ransen_spam_me_not at nemo.it (Owen F. Ransen)
Date: Sat, 10 Mar 2001 06:08:39 GMT
Subject: Memory leaks in VC6 2 line Python 2.1b1 apps?
References: <3aa88f52.5419185@news.newsguy.com> <hGPp6.53719$UZ4.13963947@news4.rdc1.on.home.com> <98b0a0$2g8i$1@shadow.skypoint.net>
Message-ID: <3aa91ea5.645940@news.newsguy.com>

On Fri, 9 Mar 2001 10:28:33 -0600, "Craig Holman"
<craig at patterncraft.com> wrote:

>Greetings,
>
>I'm afraid that both versions 2.0 and 2.1b1 leak memory for a single
>Py_Initialize, Py_Finalize sequence in a
>Visual C++ MFC application.

Phew! I thought it was just me!  ;)

I hope someone understands the rest of your post, I'm too new
to Python embedding to help. Also, for some reason, though I
did rebuild from source, VC gives me the leak "number" but not 
the line in the library which caused it.


--
Owen F. Ransen
http://www.ransen.com/
Home of Gliftic & Repligator Image Generators


From ACIUFFO1 at irf.com  Tue Mar  6 16:39:02 2001
From: ACIUFFO1 at irf.com (Andrea Ciuffoli)
Date: Tue, 06 Mar 2001 13:39:02 -0800
Subject: Run Scripts
Message-ID: <saa4e88f.040@irf.com>

Hello,
I am new in Python programming so my question is probably silly.

Starting from the Ftpmirror.py program I found in /Python20/Tools/Scripts, I've implemented a script to either get and send data from an FTP server (my OS is Windows NT). Now I would like to run the script automatically each hour, what is the best way to do that?
Consider that to call my script I need to add several arguments (e.g.  ftp.py -m -p 2164 /test c:/mytest).

Thanks in advance,
Andrea.




From loewis at informatik.hu-berlin.de  Fri Mar  2 11:02:52 2001
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 02 Mar 2001 17:02:52 +0100
Subject: Can SAX parser return <!-- COMMENTS -->  ???
References: <MWxn6.15038$x27.906254@news1.rdc2.on.home.com>
Message-ID: <j4d7c0jh03.fsf@informatik.hu-berlin.de>

"Bruce Fleming" <bfleming at bserv.com> writes:

> How can I expose comments in the content handler ?
> It does not return comments by default, and I cannot
> find where to hook into the class..?

A ContentHandler never gets comments reported. That is what the
LexicalHandler is for (which is only defined in PyXML at the moment).

It is a matter of ongoing debate whether applications should even deal
with comments; if an application needs XML comments for proper
operation, something is wrong with that application.

Regards,
Martin



From tim.one at home.com  Mon Mar 12 19:22:27 2001
From: tim.one at home.com (Tim Peters)
Date: Mon, 12 Mar 2001 19:22:27 -0500
Subject: PyEval_CallObject and threads
In-Reply-To: <aaar6.32$P26.59620@news.pacbell.net>
Message-ID: <LNBBLJKPBEHFEDALKOLCEEDFJFAA.tim.one@home.com>

[Phlip]
> My colleague came over and asked me to write this:
>
> "We got multiple C++ pthreads calling PyEval_CallObject. How threadsafe
> is the Python API? A C++ module using Boost BPL creates multiple threads,
> which then call back into Python. When two threads simultaneously call
> PyEval_CallObject we go boom."
>
> I will now go examine his serialization primitives. But if
> anyone has any better ideas...

[Tim]
> With very few exceptions, all calls to Python C API functions
> must be made while holding the global interpreter lock.  The exceptions
> have mostly to do with a handful of functions related to initialization
> and shutdown.  See the section "Thread State and the Global Interpreter
> Lock" in the Python/C API Reference Manual for details.

[Philip]
> Further inspection reveals our architect(s) think we need to call Python
> script asynchronously from multiple threads.

You started this talking about pthreads.  I don't know what "C++ pthreads"
might mean (C++ doesn't define any threading facilities last I knew), but do
know about POSIX pthreads.  Whatever, any thread calling a Python C API
function must acquire the global interpreter lock before making that call.

> By 'lock' we assume both you and the section you cite mean 'block'.

I don't know what you mean by "block".  The global interpreter lock is a
specific Python lock, implemented in various ways on various platforms using
the platform thread facilities (under POSIX pthreads it happens to be
implemented via a combination of a pthreads mutex and a pthreads condition
variable, for example).  It must be acquired and released according to the
protocols, and using the macros, explained in the referenced manual.

> Isn't this what Stackless Python is for?

Stackless has nothing to do with "real" threads.  If you're using real
threads, they must acquire the global interpreter lock before doing anything
with the C API.  Are you or are you not using platform OS threads (if you're
using POSIX pthreads, or using Python threads via the thread or threading
modules, the answer is yes)?  If yes, you must do what the manual says to do.

It is flatly illegal, for example, that "two threads simultaneously call
PyEval_CallObject".  The calls MUST be serialized via acquiring the global
interpreter lock.  Else "we go boom" is one of the more pleasant possible
outcomes.

If you're using microthreads within a single OS-level thread, then the latter
is always holding the global interpreter lock so this shouldn't be an issue.
But then your use of the word "pthreads" is a mystery.

you-already-know-the-answers-we're-just-trying-to-extract-them<wink>-ly
    y'rs  - tim




From dougfort at downright.com  Mon Mar  5 13:37:01 2001
From: dougfort at downright.com (Doug Fort)
Date: Mon, 5 Mar 2001 13:37:01 -0500
Subject: Python & mySQL
References: <DJQo6.27114$c7.7825529@news3.rdc1.on.home.com>
Message-ID: <i1Ro6.20298$Sl.898540@iad-read.news.verio.net>

Check out the mysqldb module:  http://dustman.net/andy/python/MySQLdb

I have some working code if you want to examine it.

Also see O'Reilly's http://www.onlamp.com

--
Doug Fort dougfort at downright.com
Senior Meat Manager
Downright Software LLC
http://www.dougfort.net

"Colin Meeks" <colinmeeks at home.com> wrote in message
news:DJQo6.27114$c7.7825529 at news3.rdc1.on.home.com...
> Does anybody have any pointers for integrating mySQL with Python.  I've
been
> using Python a while and am looking to develop a web backend using Python
> with mySQL.  My site is hosted by a Linux based ISP who has both Python
and
> mySQL but they don't currently have a link between the two.  I'm using
> Windows2000 as my development base, so I want something that is usuable on
> both, so it's easier for me to track any problems.
>
> Ideally I want something that doesn't involved recompiling Python if
that's
> possible, as i want the installation to be as straightforward for my ISP
and
> of course me :-)
>
> Many thanks in advance
>
> Colin
>
>




From aleaxit at yahoo.com  Fri Mar 16 11:27:10 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 16 Mar 2001 17:27:10 +0100
Subject: gah! I hate the new string syntax
References: <mailman.983485923.1136.python-list@python.org> <97sudi01i26@news1.newsguy.com> <98elrk$li8$1@nntp.Stanford.EDU> <98fc070thh@news1.newsguy.com> <98iqo2$56o$1@nntp.Stanford.EDU> <j4kr6.68739$ML1.3282814@e420r-atl2.usenetserver.com> <Pine.LNX.4.21.0103161339210.2809-100000@bast.jhb.equinox.co.za> <98t5bv0l8b@news1.newsguy.com> <Pine.LNX.4.21.0103161647280.2809-100000@bast.jhb.equinox.co.za>
Message-ID: <98tet00v03@news1.newsguy.com>

"Pieter Nagel" <pieter at nagel.co.za> wrote in message
news:Pine.LNX.4.21.0103161647280.2809-100000 at bast.jhb.equinox.co.za...
> On Fri, 16 Mar 2001, Alex Martelli wrote:
>
> > "Pieter Nagel" <pieter at nagel.co.za> wrote in message
> > news:Pine.LNX.4.21.0103161339210.2809-100000 at bast.jhb.equinox.co.za...
> >     [snip]
> > > Of course you can. The technique is to modify the String class itself
> >
> > It's not a class.  It's a non-class type, and does not support
> > modification.
>
> Yes. But the author I responded to said "assuming that String is a
> class, not a type". My answer was made within the context of that
> assumption.

All classes NOW allow you to change their methods -- but strings
NOW are not a class.  If we assume Python changes enough that
strings become class instances, we might also get a 'lock' feature
on dictionaries to let a dictionary using object avoid modifications
thereof by client-code -- and it might well be used to safeguard
certain crucial 'system-types'.  Thus, the ability to invasively
modify builtin types should not be assumed -- even if the type/class
split gets healed.

Non-invasive mods would be a better target -- and I think a
runtime variation of Haskell typeclasses might be a nice way
to present such mods.  E.g., suppose that, parallel to the
existing builtin 'isinstance(object, [class or type object])',
that just returns 0 or 1, we have in some future Python a
new builtin:
    getinstance(object, [typeclass object, or etc etc])
which returns either None (meaning the required 'cast' is
not feasible) OR a reference x such that the axiom:
    x = getinstance(y, z)
        =>
    x is None or isinstance(x, z)
is satisfied, and x 'represents y as a z' -- for example
if isinstance(y, z), then x is y should hold; if z was
type(1), then x==int(y), etc; the actual novelty would
come if z was a 'typeclass object', a new kind of things.

Typeclasses would be a new kind of classes, not directly
instantiable, such that: a class A might inherit from a
typeclass B, and then everything would work as if B was
a regular class; but also, a pre-existing class C could
be *post facto, non invasively on C* be made to belong to
B, contextually overriding some of B's methods (this
post-facto-instantiation would be recorded in B, not
in C).  This would, I believe, be implementable today
in pure Python.  Hmmm, I feel _close but not quite
there_ to thinking up that implementation -- maybe it
needs some metaclass black-magic (or resignation to a
rather slow example implementation, as it _would_ be
just an example anyway) -- I wouldn't want a class that
inherits from a typeclass to automatically be a typeclass
itself, for example.


Alex





From wesc at deirdre.org  Wed Mar 21 18:25:41 2001
From: wesc at deirdre.org (Wesley Chun)
Date: Wed, 21 Mar 2001 15:25:41 -0800 (PST)
Subject: learning python...
In-Reply-To: <3AB8F4E1.184DD9C@noaa.gov>
Message-ID: <Pine.LNX.4.31.0103211504000.5804-100000@emperor.deirdre.org>

On Wed, 21 Mar 2001, it was written:
> Hello.  From the perspective of someone who already has programming
> experience, would anyone know if O'reillys "Learning Python" book would
> be better than say Prentice Hall's "Core Python Programming"?  Or would
> it be better to go straight into something like "Programming Python"?

"Learning Python" was a welcomed 3rd main Python book to hit the market.
Some people complained that it over-expected its readers to know C, but
since it did, it suited me fine.  I found "Programming
 Python" to be too large for me when i was learning
Python.  It was difficult to learn from even though it was so complete.
I spoke with Mark Lutz about the 2nd edition, and it was indeed a com-
plete rewrite.  My copy is on order, so I haven't had a chance to go
thru it yet.

"Core Python Programming"'s target audience is a someone with previous
knowledge of some other high-level programming language, perhaps C or
Java, but does not require that knowledge as part of the reading.  in
the text, it will say something like, "if you are familiar with...,"
but otherwise will not penalize you if you don't.  the goal is to get
a programmer new to Python up-to-speed as quickly as possible, hence
sacrificing more of the advanced applications, as Daniel mentioned in his
post.  hopefully, we'll correct that when it is time to work on the 2nd
edition.  if you're a Python guru, it'll definitely be too boring for you!

i made a related post back in November:
http://groups.google.com/groups?q=books&hl=en&rnum=1&seld=963817644&ic=1

hope this helps!

-wesley

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Silicon Valley-SF Bay Area Python users group:  http://baypiggies.org

"Core Python Programming", Prentice Hall PTR, December 2000
    http://starship.python.net/crew/wesc/cpp/

wesley.j.chun :: wesc at baypiggies.org
cyberweb.consulting :: silicon.valley, ca
http://www.roadkill.com/~wesc/cyberweb/





From robin at jessikat.fsnet.co.uk  Mon Mar  5 11:56:54 2001
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Mon, 5 Mar 2001 16:56:54 +0000
Subject: the demise of 'from foo import * and its implications?
References: <3kVBAfATrDo6EwEM@jessikat.fsnet.co.uk>
 <mailman.983582584.8198.python-list@python.org>
 <z8dAGJAyqLo6EwlX@jessikat.fsnet.co.uk>
 <mailman.983746566.27513.python-list@python.org>
 <GqQAdIAXUuo6EwyY@jessikat.fsnet.co.uk>
 <mailman.983806269.21732.python-list@python.org>
Message-ID: <HoAgdIAWV8o6Ewyd@jessikat.demon.co.uk>

In article <mailman.983806269.21732.python-list at python.org>, Thomas
Wouters <thomas at xs4all.net> writes
>
....
>> 
>> doesn't from foo import * make things hard to locate at the global level
>> as well? exec/eval/import can always do this surely.
>
>The global namespace is a 'last call' namespace. The compiler doesn't (and
>can't, since you can modify global namespaces from outside the namespace)
>keep track of global names because it isn't necessary. Neither is true for
>function namespaces.
>
it isn't necessary in function namespaces either it's desirable to help
the compiler and apparently now required to enforce the constant scope
requirement. If the argument is that exec/eval/import can retarget names
then that argument applies to the global scope as well (and as you point
out) other things can modify globals as well. I would still like
regularity of expression to override efficiency as is currently the
case. 
-- 
Robin Becker


From bpurvy at packeteer.com  Tue Mar 20 13:12:09 2001
From: bpurvy at packeteer.com (Bob Purvy)
Date: Tue, 20 Mar 2001 10:12:09 -0800
Subject: Porting problem
Message-ID: <3AB79D79.FDAA15B7@packeteer.com>

I posted this on Friday with a different title but didn't get any
answers (it began with 'newbie' - maybe that was the problem!):

OK, I got this pSOS port sort of working for Python 2.0, and I thought
it would be a good idea to run the standard Python regression tests, to
verify that things are really correct.  But I'm getting a strange result

that I was hoping someone would understand:

>>> import string
>>> import regrtest
>>> dir(regrtest)
['__builtins__', '__doc__', '__file__', '__name__', 'sys']

why not any of the methods?

If I do the same thing on Linux on Python 1.5, dir(regrtest) shows all
the methods ('main', 'random', etc.), just as I expected.   So is there
some obvious reason why I'm not seeing those here?   Or is this some
obscure porting problem I just need to hunt down?


(since we don't have a real OS, I've embedded the whole thing within our

product and called Py_Main(), basically, so that's why I'm not just
running regrtest from the command line, as you'd normally do.)





From paulp at ActiveState.com  Tue Mar 13 17:52:14 2001
From: paulp at ActiveState.com (Paul Prescod)
Date: Tue, 13 Mar 2001 14:52:14 -0800
Subject: Yet another Python vs. Perl speed issue/question
References: <98m6ad+i25s@eGroups.com>
Message-ID: <3AAEA49E.399F9AAF@ActiveState.com>

nanotech at europa.com wrote:
> 
> All:
> 
> Python 2.0 on an HP (nice and big) takes ~30 seconds to lowercase all
> the lines in a 3meg file. Perl takes less than a second. What am I
> doing wrong?!?
> 
> Python
> ------
> import sys
> import string
> 
> stdin=sys.stdin
> lower=string.lower
> 
> print map(lower,stdin.readlines())

I'd suggest 

sys.stdin.read().lower()

-- 
Python:
    Programming the way
    Guido
    indented it.
       - (originated with Skip Montanaro?)



From dwig at advancedmp.net  Wed Mar 14 13:18:06 2001
From: dwig at advancedmp.net (Don Dwiggins)
Date: 14 Mar 2001 10:18:06 -0800
Subject: Who's minister of propaganda this week?
In-Reply-To: Paul Prescod's message of "Wed, 14 Mar 2001 06:49:34 -0800"
References: <B98041329020E007.7DA84CAD9D0562AB.817BBC572F480869@lp.airnews.net>
	<98nu1m$12e@dispatch.concentric.net> <3AAF84FE.F28A2C30@ActiveState.com>
Message-ID: <uelw09pup.fsf@advancedmp.net>

Paul Prescod writes:
B> The full depth of behaviors we should expect from a typeless language
>> have yet to be fathomed. Ruby trumps Python in the simple matter of
>> closures (imagine my shock when I discovered Python didn't have them). But
>> a language "like" those might win the decade.

> I don't know if you are serious (I mean closures are NOT that
> important!) but anyhow, Python 2.1 has nested scopes and I think that
> functions defined in those scopes are closures just like in any other
> language.

I'm relatively new to Python, and this is something I'm trying to get my
head around.  I know that closures (blocks) are extremely important in
Smalltalk (e.g., conditionals and looping are implemented entirely in terms
of blocks).  So, is this just an indication of Smalltalk's peculiarity, or
could Python benefit from the ability to exploit more robust closures in
something like the manner of Smalltalk?  For example, how about implementing
general iteration as something a container class does to a closure (as in
select:, detect:, inject:into:, etc.)?

-- 
Don Dwiggins                    "Solvitur Ambulando"
Advanced MP Technology
dwig at advancedmp.net




From yartz at imaginet.fr  Sat Mar  3 04:25:15 2001
From: yartz at imaginet.fr (Yann Schwartz)
Date: Sat, 03 Mar 2001 10:25:15 +0100
Subject: [ANNOUNCE] Webware for Python 0.5
References: <5.0.2.1.0.20010301205419.047bfd90@mail.mindspring.com> <mailman.983544603.4907.python-list@python.org>
Message-ID: <28800BFBA8329F65.A37ED22A938924F4.F61D6B84B77DFD40@lp.airnews.net>

On Fri, 02 Mar 2001 09:55:48 -0500, Geoff Talvola
<gtalvola at nameconnector.com> wrote:

>Chuck Esterbrook wrote:
>
>
>One more feature of WebKit/PSP that I _think_ doesn't exist in ASP, but I might be wrong:  The ability to forward a request to another Servlet or PSP purely within the app server (as opposed to redirecting the browser).  This allows you to do things like write a "controller" servlet that examines the request and then forwards to the appropriate servlet or PSP.  It's a nice model for certain types of problems that is really simple to implement in WebKit/PSP, and I think it would be difficult or maybe even 
impossible in ASP.

Actually ASP does. Server.Execute(url) and Server.Transfer(url)
permits just this. But it's an ASP 3.0 feature, available only on IIS5
(Win2k).

On a side note, i've began to play with ASP .NET, and it"s
breathtaking. I'm waiting for a really working release of Python .NET
to use ASP.NET with.
>
>On the flip side, one possible advantage of ASP with Python is that you might be able to take advantage of 3rd-party extensions that are designed to only work with ASP.  I don't know if such things exist.  I happen to be using a 3rd-party COM component in my web pages to draw dynamic graphs.  The component was designed for use with Visual Basic and Active Server Pages, but it works great with WebKit too.

3rd party extensions are usually COM components, so it wouldn't be a
problem to use them in "pure" Python with win32 extensions, provided
the COM components doesn't use ASP objects internally  (using native
ASP objects in COM components is not considered a Good Thing, though).


From loewis at informatik.hu-berlin.de  Wed Mar 14 06:20:25 2001
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 14 Mar 2001 12:20:25 +0100
Subject: Newbie: Is there a more pythonic approach?
References: <3AAE87E8.FE11B4F9@sewardweb.com>
Message-ID: <j4vgpclhqe.fsf@informatik.hu-berlin.de>

Lloyd Sommerer <lsommerer at sewardweb.com> writes:

> a) made use of python language features in a way that seems natural to
> "real" python programmers.

I could understand your script quite easily, and could not detect any
'misuse'.


> b) missed any language construction that makes the program "better"
> (faster, easier to read...)

To see whether a number of files exist, you can take two approaches:
1. construct the file names, and see which file exists
2. list the directory (os.listdir) and see whether you recognize any of
   the file names

Which one is more efficient much depends on the number of files in
each directory and on your operating system; if speed matters, I'd
benchmark them both. Your approach (option 1) is certainly easier to
read.

Regards,
Martin


From chrisw at nipltd.com  Mon Mar 19 11:45:26 2001
From: chrisw at nipltd.com (Chris Withers)
Date: Mon, 19 Mar 2001 16:45:26 +0000
Subject: Serial Port on Unix
Message-ID: <3AB637A6.FDEAE8FF@nipltd.com>

Hi,

I need to do some interaction with a serial port on Linux (actually getting an
SMS from a Mobile Phone attached via a PCMCIA card, if anyone has any better
ideas of how to setup an SMS target, please let me know ;-)

Anyway, I had a hunt on python.org without much luck. A few obscure postings and
a module from the vaults that turned out to be a broken link. What module should
I be using to do this and where can I get it from?

cheers,

Chris



From db3l at fitlinxx.com  Wed Mar 14 23:28:00 2001
From: db3l at fitlinxx.com (David Bolen)
Date: 14 Mar 2001 23:28:00 -0500
Subject: Help doing simple threading
References: <mailman.984574578.29018.python-list@python.org> <3AB041C6.FDDB84D3@theworld.com>
Message-ID: <ubsr38xm7.fsf@ctwd0143.fitlinxx.com>

David Lees <deblNonOspammy at theworld.com> writes:

> Thanks, these comment were quite helpful.  The most serious error was
> 'a)'.  when I just put the main to sleep after starting up the threads
> it worked fine.  The problem was that I was printing out the array
> values, before all the threads finished executing.

If by sleep, you just mean you added something like time.sleep() with
a fixed value, then you're still at risk, depending on the performance
of the other thread executions.

You will need to use some sort of synchronization primitive to ensure
that the main thread waits long enough for each of the worker threads
to complete.  If you know you want to wait until the threads actually
complete execution and terminate, and if you use the higher level
Threading module, you can "join" on the threads to block until they
complete.

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


From cribeiro at mail.inet.com.br  Wed Mar 28 11:57:43 2001
From: cribeiro at mail.inet.com.br (Carlos Alberto Reis Ribeiro)
Date: Wed, 28 Mar 2001 13:57:43 -0300
Subject: What *instance* called me & proxy methods
In-Reply-To: <3AC1CB08.CA169C59@tismer.com>
References: <3AC1BE92.ACE82741@ua.ac.be>
Message-ID: <5.0.2.1.0.20010328104602.021cfc40@mail.inet.com.br>

I've seen some talk on how to get the name of the caller. I got a related 
problem to solve, only slightly more complicated.

Can I know what *instance* called a method? I'm not talking about "self" 
here, because we *can* call a method using other techniques, such as 
copying the instance method pointer directly:

    >>> class myclass:
    ...   def hello(self):
    ...     print "hi! I'm ", self
    >>> class myproxy:
    ...   pass
    >>> a = myclass()
    >>> a.hello()
    hi! I'm  <__main__.myclass instance at 015093EC>
    >>> m = a.hello
    >>> p = myproxy()
    >>> p.m = m
    >>> p.m()
    hi! I'm  <__main__.myclass instance at 015093EC>

In both cases, hello() received the same value as the self parameter. This 
is exactly what I need; in fact, the im_self attribute of the instance 
method stores the pointer to self, which explains this behavior. However, I 
would also like to have some information to differentiate between these two 
calls. My goal is to implement some security checking as part of my 
proxy/adapter module [1]. I have already tried some simple minded 
approaches (too many to list here), so far with no success.

The best solution the I have thought so far can be implemented as a Python 
extension - if I manage to put my Linux box up and running (I'm gcc-less 
right now :-).

1) Implement a new object called 'ProxyMethod', derived of 'instance
    method', with the following differences:

    * It does not make the im_* attributes visible to the outside
      world;

    * It stores a read-only attribute to the proxy that 'owns' the
      ProxyMethod.

2) The 'ProxyMethod' constructor would take a instance method and
    build the proxy method from it.

3) The information about the proxy would need to be stored somewhere
    in the call stack. I dont know if it is possible without changing
    some of the Python internals.

4) The runtime semantics of the 'ProxyMethod' would be about the
    same as the ones of the instance method - just a direct call,
    without any extra intermediation.

I would like to know what the community thinks about turning this into a 
PEP. Maybe we can find a simpler solution, or it's better to keep this as a 
module outside the distribution.

For now I have a very simple Python implementation. It's far from perfect, 
and it has some undesirable side effects because it is not completely 
transparent, and it needs some intermediate processing. The only way that I 
found to signal to the proxied object about the existence of the proxy was 
to create a temporary "proxy" attribute (see point (3) above). However, it 
serves it purpose as a proof-of-concept:

----------------------------------------------------------------------
TEST RESULT

 >>> n = NuclearArsenal()
 >>> n.NuclearAttack("NY")
Only the proxy can order the attack
 >>> p = Proxy()
 >>> p.attack = ProxyMethod(n.NuclearAttack, p)
 >>> p.attack("NY")
Boooom! NY destroyed

----------------------------------------------------------------------
CODE

""" ProxyMethod proof-of-concept
     (c) 2001 Carlos Ribeiro - cribeiro at mail.inet.com.br
     Released in public domain for use "as-is"
"""

from types import *

class Proxy:
   pass

class ProxyMethod:

   def __init__(self, im, p):
     if type(im) != MethodType:
       raise "A proxy needs a method"
     if p.__class__ != Proxy:
       raise "A proxy method is bounded to a proxy"
     self.im = im
     self.p  = p

   def __call__(self, *a):
     # set a flag inside the target instance
     setattr(self.im.im_self, 'proxy', self.p)
     # do the call
     self.im(*a)
     # clear the flag
     delattr(self.im.im_self, 'proxy')

class NuclearArsenal:
   def NuclearAttack(self, city):
     if getattr(self, 'proxy', None):
       # called through the proxy
       print "Boooom! %s destroyed" % (city)
     else:
       print "Only the proxy can order the attack"

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

Carlos Ribeiro





From bbhaydon at bigpond.com  Sat Mar 24 21:25:01 2001
From: bbhaydon at bigpond.com (Brett Haydon)
Date: Sun, 25 Mar 2001 12:25:01 +1000
Subject: time to invest ???
References: <3abd3a17.7103358@news.vif.com>
Message-ID: <Yrcv6.2751$45.12829@newsfeeds.bigpond.com>

The beauty of Python is that it is very quick to learn how to use it as a
scripting language. Within days or even hours.

Mastery of the language may take a while (if thats possible), but Python
rewards the small steps on the path.

Brett

"Yvon Boulianne" <mysticsguy at yahoo.ca> wrote in message
news:3abd3a17.7103358 at news.vif.com...
> Hello, i start to learn Python some day ago, with lot of knowledge as
> a system admin but little as programmer and i wonder what is the
> average time/hours someone have to invest to be able to use a language
> like that easily.  i talk as well about devlopping the way of thinking
> of a programmer and the knowledge of the language. please don't tell
> me theres no average, at least take the idea of a normal guy, not a
> genius and not stupid  ;-)
>
> Thanks verry much and sorry for my english, my first language is
> french.
>
> Yvon Boulianne ; Saphi
> mysticsguy at yahoo.ca
> www.monunivers.net
> Humour,Paradox & Changements




From Peek at LVCM.comNOSPAM  Sat Mar 31 23:18:00 2001
From: Peek at LVCM.comNOSPAM (Ken Peek)
Date: Sat, 31 Mar 2001 20:18:00 -0800
Subject: Good GUI for Python
References: <mailman.985314033.5120.python-list@python.org>
Message-ID: <tcdavpati4c05b@corp.supernews.com>

Take a look at the "Fast Light Tool Kit" (FLTK).  I think the link is on the
python web site.  It does most of everything that most people need to do,
without the code bloat of Windoze.  The license is "BSD" style I think (which
means you can use it in a commercial product without GPL restrictions.)  This
GUI is embeddable in small hardware, and runs on many O/S's as well as stand
alone if you write the frame buffer and input device drivers.

If you would like a commercial product, I heard pyQT was very good...




From deblNoSpammy at theworld.com  Mon Mar 12 00:26:52 2001
From: deblNoSpammy at theworld.com (David Lees)
Date: Mon, 12 Mar 2001 05:26:52 GMT
Subject: looking for old python books
References: <eafoat4r8rs21v1h50239iqf9sktaetnjv@4ax.com>
Message-ID: <3AAC5E1B.28CE0825@theworld.com>

I think you might do better to browse the current on-line documentation
starting from www.python.org

david lees


dr0 wrote:
> 
> Hi all,
>  I was wondering if any of you have any old books on python you can
> sell to me?
> I'm a student and the prices for online stores's a bit high.
> Email me if you can help.
> 
> Any help is deeply appreciated,
> thanks.
> Kenny


From not.this at seebelow.org  Wed Mar  7 13:49:00 2001
From: not.this at seebelow.org (Grant Griffin)
Date: 7 Mar 2001 10:49:00 -0800
Subject: Can anyone recomend a good intoduction to C...
References: <mailman.983894288.19965.python-list@python.org> <3AA5EF59.507F1D91@olsen.ch> <983958878.976344@newsmaster-04.atnet.at> <77up6.1289$y6.235541@ruti.visi.com>
Message-ID: <985vqs01ipr@drn.newsguy.com>

In article <77up6.1289$y6.235541 at ruti.visi.com>, grante at visi.com says...
>
>In article <983958878.976344 at newsmaster-04.atnet.at>, Werner Schiendl wrote:
>>
>>> I don't think C deserves all the superlatives you apply to it,
>>> but I agree that it's instructive to study C++ as an example of
>>> a language that gets it wrong. Take a good language, C; add
>>> some good ideas; and end up with a baroque mess. Fascinating.
>>
>>To be completely honest, I can not see the obvius reason why
>>C++ should be such a bad programming language.
>
>It's complicated and unsafe.  Either one is tolerable without
>the other.  When you combine the two, you're dead.

I dunno...I think you've hit on exactly its _strength_: it's complicated and
unsafe.

As a virtually 100.0% backwards-compatible superset of C, it has little choice
but to be complicated and unsafe.  By "unsafe", maybe you are referring to its
backwards-compatibility with C's pointers: C++ is unsafe because C is unsafe. 
By "complicated", maybe you are refererring to the fact that it supports both
procedural programming (from C) and OO programming (which is its whole reason to
exist).

But those of us who like C++ don't regret its complication or its unsafety; we
lthink of that more as being "powerful" and "flexible".

I would agree that C++ is somewhat _unnecessarily_ compilicated (primarily in
terms of its silly protection features), but then again, nothing's perfect. 
(Heck, even Python has "warts". <wink>)

Still, in terms of its design goal (of grafting OO onto C while being 100.0%
backwards-compatible), I personally think C++ is a real Work of Art.  (In
contrast, I find Perl's approach to grafting on OO to be distinctly hokey.)  I
think that C++'s choice of design goal, in combination with its success in
meeting that goal, adequately explains why large numbers of lemmings use it.

>>It is widely available (on almost any platform, I would guess).
>
>If 1 billion people do a stupid thing, it's still a stupid thing.
>

(And you gotta admire the chutzpah of that one guy who thinks that the 1 billion
people who disagree with him must, ipso facto, be wrong. <wink>)

>>It supports for a real large number of different design approaches.
>
>That isn't always a good thing. ;)

isn't-it-nice-to-know-that-we-grants-don't-always
   -agree?-<wink>-ly y'rs

=g2

_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation            http://www.iowegian.com



From qrczak at knm.org.pl  Sat Mar 31 14:23:15 2001
From: qrczak at knm.org.pl (Marcin 'Qrczak' Kowalczyk)
Date: 31 Mar 2001 19:23:15 GMT
Subject: list vs tuple
References: <hGlw6.782$p5.3210@news1.rivrw1.nsw.optushome.com.au> <3AC20CD3.265A2D94@alcyone.com> <ru69cto61jtmrqsf809e38hb3pbm2lhi7g@4ax.com> <bC1x6.4319$4N4.905946@newsc.telia.net> <71q9ct4le63anefmiajj5cbh3mg04qu1ga@4ax.com>
Message-ID: <slrn9ccbl1.2ae.qrczak@qrnik.zagroda>

Fri, 30 Mar 2001 15:25:14 -0500, Michael Prager <Mike.Prager at noaa.gov> pisze:

> To the naive user, the EFFECT is different.  If the object is
> immutable, the statement appears to act just like it does in
> many other languages.  (It may not really act that way, but the
> difference is hidden.)

Indeed. Python's behavior is perfectly consistent. The problem is that
for immutable types there is no difference between passing by value
and passing by reference, so a naive user might think that Python
passes by value. They don't realize that in this case there would
be no difference and that the same behavior could be also explained
differently.

> That behavior is not in agreement with the "mental model" used
> by many other languages,

But most languages treat = like Python. In some of them it does not
really matter because almost all values are immutable (functional
programming), in others it matters as in Python for mutable objects
(Lisp, Scheme, Eiffel, Java, C#, Ruby, Smalltalk).

C, C++, Pascal, Perl and some older languages that I don't know pass
by value by default. All the rest pass by reference.

-- 
 __("<  Marcin Kowalczyk * qrczak at knm.org.pl http://qrczak.ids.net.pl/
 \__/
  ^^                      SYGNATURA ZAST?PCZA
QRCZAK


From stephen_purcell at yahoo.com  Thu Mar 22 12:01:10 2001
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Thu, 22 Mar 2001 18:01:10 +0100
Subject: Transparent GIFs in PIL
In-Reply-To: <newscache$m8xlag$gk9$1@proxy.bitmailer.com>; from rapto@arrakis.es on Thu, Mar 22, 2001 at 05:18:48PM +0100
References: <newscache$m8xlag$gk9$1@proxy.bitmailer.com>
Message-ID: <20010322180110.A18476@freedom.puma-ag.com>

Marcos S?nchez Provencio wrote:
> Is it possible to set a transparent color in a GIF file (originally not
> transparent) in PIL?
> 

With PIL 1.0, not directly. For later versions (if they exist), I don't know.

You can patch the PIL 1.0 version of 'GifImagePlugin.py' as follows:

############
--- GifImagePlugin.py   Thu May  4 11:15:49 2000
+++ /tmp/GifImagePlugin.py      Thu Mar 22 17:57:43 2001
@@ -303,6 +303,12 @@
        for i in range(256):
            s.append(chr(i) * 3)
 
+    # save transparency
+    if im.info.has_key('transparency'):
+        transparentIndex = im.info['transparency']
+        s.append('!' + chr(0xf9) + chr(4) + chr(1) + chr(0) + chr(0) +
+                 chr(transparentIndex) + chr(0))
+
     return s
 
 def getdata(im, offset = (0, 0), **params):
############


Then, if you know the index of the colour in the GIF's palette you can say:

   >> image = Image.open('my.gif')
   >> transparent_colour_index = 4
   >> image.info['transparency'] = transparent_colour_index
   >> image.save('my_transparent.gif')


-Steve


-- 
Steve Purcell, Pythangelist
Get testing at http://pyunit.sourceforge.net/
Any opinions expressed herein are my own and not necessarily those of Yahoo



From fredrik at pythonware.com  Wed Mar 14 17:12:02 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 14 Mar 2001 22:12:02 GMT
Subject: Package
References: <mailman.984607039.2853.python-list@python.org>
Message-ID: <S0Sr6.330$4d3.75074@newsb.telia.net>

Anbuselvam Palusamy wrote:

> Can any one explain how to create a packages in python?

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

Cheers /F

PS.  please don't post HTML mail to mailing lists/newsgroups.
in outlook express, use the "format" menu to change to plain
text before posting.




From aleaxit at yahoo.com  Sat Mar 10 15:16:33 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sat, 10 Mar 2001 21:16:33 +0100
Subject: How to efficently build a nested dictionary
References: <3AAA2042.E69A53FC@gmx.de> <98d9f1$1hopn$1@ID-11957.news.dfncis.de> <3AAA7188.266516E9@schlund.de>
Message-ID: <98e8vg12sqq@news1.newsguy.com>

"Carsten Gaebler" <cg at schlund.de> wrote in message
news:3AAA7188.266516E9 at schlund.de...
> I'd like to be able to print out the annual, monthly or daily traffic for
> each directory. With nested dictionaries I could easily accomplish this by
> using nested for loops. How would I do that with a compound key?
    [snip]
> > ky = (dir, year, month, day)
> > p_ttl = traffic.get(key,0)
> > traffic[ky] = p_ttl + bytes
    [snip]
> > > if not traffic.has_key(dir):
> > >     traffic[dir] = {year: {month: {day: bytes}}}
> > > elif not traffic[dir].has_key(year):
> > >     traffic[dir][year] = {month: {day: bytes}}
    [snip]

You can sort the tuple keys of the non-nested dictionary,
then loop over the sorted keys to accumulate traffic
totals -- it's not very different from working with the
nested version.


Alex





From ngps at madcap.dyndns.org  Mon Mar 12 10:16:04 2001
From: ngps at madcap.dyndns.org (Ng Pheng Siong)
Date: 12 Mar 2001 15:16:04 GMT
Subject: Writing docu
Message-ID: <98ip7k$sga$1@violet.singnet.com.sg>

Hi,

I'm writing a programming HOWTO using Vim. ("SSL programming in Python with
M2Crypto", for the curious.)

In the HOWTO, fragments of program code will be scattered thru-out the 
text, e.g.,

    Here's a simple example:

        def hello():
            print 'Hello, world'

    Another example:

        def hello2(there):
            print 'Hello,', there

    We can see that [...]


I want to prefix line numbers to the program code _only_, to facilitate
discussion, like this:

    Here's a simple example:

        1.1     def hello():
        1.2         print 'Hello, world'

    Another example:

        2.1     def hello2(there):
        2.2         print 'Hello,', there

    We can see that [...]


I'm currently using Gnu textutils 'nl' to generate line numbers, but 'nl'
doesn't do dotted numbers such as 1.1, 1.2, ...

What's the easiest way to do this in Vim? With a Python interpreter
embedded? Perl embedded?

What other Vim-compatible docu tools - LaTex, DocBook, whatever - would you
recommend for writing programming guides and manuals?

TIA. Cheers.


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



From fredrik at pythonware.com  Sat Mar 31 06:37:30 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Sat, 31 Mar 2001 11:37:30 GMT
Subject: Working with soaplib...
References: <mailman.985978481.7289.python-list@python.org>
Message-ID: <_jjx6.4438$4N4.956130@newsc.telia.net>

Gabriel Ambuehl wrote:
> AssertionError: argument must be tuple, method call/response, or fault
> instance
> - ----------------------------------------
>
> I think there's something wrong with my return statements of the
> not working functions but can't seem to figure out what to do about
> it.

the quickest fix is probably to change soapserver's do_POST
method to something like:

            ...

            # generate response
            try:
                response = self.call(method, params)
                # wrap response in a singleton tuple
                response = (response,)
            except:
                # report exception back to server

            ...

Cheers /F

PS. we're currently doing interop testing of soaplib 0.9.x
(a major overhaul).  eta mid-april.




From emile at fenx.com  Fri Mar 16 20:57:08 2001
From: emile at fenx.com (Emile van Sebille)
Date: Fri, 16 Mar 2001 17:57:08 -0800
Subject: Console printing
References: <3ab2c183.4460238@News.CIS.DFN.DE>
Message-ID: <98ugf9$3kojv$1@ID-11957.news.dfncis.de>

Invoke pythone with the -u option

Emile van Sebille
emile at fenx.com


"Costas Menico" <costas at meezon.com> wrote in message
news:3ab2c183.4460238 at News.CIS.DFN.DE...
> Why does the following program only prints the output (in the run
> method) when the while loop terminates? Please note that this is
> running in a thread. I though a sleep() is supposed to allow time for
> console to update itself. I am using Pythonwin.
>
> import threading
> import time
> import sys
>
> class wasteTime(threading.Thread):
>
>     def run(self):
>         # run forever
>         self.running=1
>         while self.running:
>             print time.time()
>             time.sleep(1)
>
>     def stop(self):
>         self.running=0
>
> th=wasteTime()
>
> print "Start thread"
> startTime=time.time()
> th.start()
> while time.time()<startTime+5:
>     time.sleep(1)
>
> # I want to kill the th thread right here.
> th.stop()
> print "Finished Thread"
> sys.exit()
>
> #Costas
>




From stevewilliams at wwc.com  Sat Mar 17 21:29:28 2001
From: stevewilliams at wwc.com (Steve Williams)
Date: Sun, 18 Mar 2001 02:29:28 GMT
Subject: Pep 238 Why new // operator (Non-integer Division)
References: <98tc2s$qir$1@news2.ottawa.cyberus.ca> <mailman.984822021.13503.python-list@python.org> <vag7bt09r54tv05nohtssd0ge891v3fa02@4ax.com>
Message-ID: <3AB41DA9.E9551984@wwc.com>

Courageous wrote:

> >>All these changes make my head spin.  There is more to life than learning
> >>all the fine details in a computer language.  The fewer number of ways to do
> >>something the better.
> >
> >Just because the changes are being made doesn't mean you have to USE them.
>
> This isn't true; the moment you encounter code written by someone else,
> you DO have to use them. This simple principle is the magic of Python
> minimalism, and the primary driving reason why new Python programmers
> can become so effective so fast (IMO).
>
> C//

Adding 'line noise' symbols to a language doesn't help at all.  It doesn't even
address the underlying problem.

The difference between fixed/floating point arithmetic resides in the operands,
not the operators.

For example, consider fixed point(precision,scale) arithmetic for variables like

    A decimal(5,2)

    B decimal(3,1)

    C decimal(9,6)

You want to detect overflows when A = B / C exceeds 999.99.  You want to round
'appropriately' to .01 without knowing what an LSB is.  And so on.

This is garden-variety, every-day commercial arithmetic.  Your paycheck and tax
witholding are calculated this way.

Python, actually, can handle 'decimal arithmetic' very well:  define variables as
{name:(integer value,precision,scale)} with functions for add, subtract, multiply,
divide, move, compare, and display.

Let's reserve the 'line noise' infix operators for enhancing Python's complex
arithmetic to handle the octonions.




From scarblac at pino.selwerd.nl  Tue Mar 20 03:06:33 2001
From: scarblac at pino.selwerd.nl (Remco Gerlich)
Date: 20 Mar 2001 08:06:33 GMT
Subject: HELP - beginner to programing/python
References: <Vpxt6.416511$Pm2.6478704@news20.bellglobal.com>
Message-ID: <slrn9be3n9.ap9.scarblac@pino.selwerd.nl>

EEN <iansworld at thevortex.com> wrote in comp.lang.python:
> Hi there. I am a newbie and am hearing that python is the way to go to start
> off. I am downloading it as i type this, but need to know how to use it :)
> of course :) Is there any site that has a good tutorial or beginners guides,
> or do you have to buy books? If so, waht do you recommend?
> 
> Thanks a lot for the help, i really appreciate any pointers!

One of the best tutorials for non-programmers is at
http://www.crosswinds.net/~agauld

It is a tutorial that teaches programming, using Python. It's also available
in book form.

Other tutorials are listed at www.python.org, under "Documentation" ->
"Introductions" -> "Introductions for non-programmers".

Also, you can join the Tutor list to ask beginner questions, see
http://mail.python.org/mailman/listinfo/tutor

-- 
Remco Gerlich


From insanc at cc.gatech.edu  Thu Mar 22 10:30:26 2001
From: insanc at cc.gatech.edu (Holland King)
Date: 22 Mar 2001 15:30:26 GMT
Subject: email address verification
Message-ID: <99d5qi$95l$1@solaria.cc.gatech.edu>

does python provide quick/simple way to verify that an email address exist?
i looked at the smtp library and the verify method but that does not seem
to be very standard anymore. is there anything else that is simalar?
thank you.


-- 
---
Joseph Holland King  | "God whispers to us in our pleasures, speaks in our
                     |  conscience, but shouts in our pains: it is His
                     |  megaphone to rouse a deaf world." C. S. Lewis


From spencerm at cruzio.com  Thu Mar 22 11:50:20 2001
From: spencerm at cruzio.com (Michael Spencer)
Date: Thu, 22 Mar 2001 08:50:20 -0800
Subject: Checking the type
References: <99a493$bhl$1@sun.rhrk.uni-kl.de>
Message-ID: <99daj70aeu@enews4.newsguy.com>


Markus Reitz <Markus_Reitz at yahoo.com> wrote in message
news:99a493$bhl$1 at sun.rhrk.uni-kl.de...
> Hi,
>
> is there a possibilty to check if a variable refers to an object of type
x?
> Does a typeof-command exist in Python:
>
>   if a typeof list:
>      #Commands only useable with list
>
> I read the Python Tutorial, but have not found a hint to this topic.
>
> Thanks
>
> Markus
>
>

Sample code:

     def Show (self, object):
          """ <object> can be number, string, list, tuple, or dictionary"""
          from types import *
          kind = type (object)
          if kind == StringType:
              self.Writeln (object)
          elif (kind == ListType) or (kind == TupleType):
               for item in object:
                    self.Writeln (item)
          elif kind == DictType:
               sKeys = object.keys()
               sKeys.sort()
               for key in sKeys:
                    self.Writeln ('%-16s   %s' % (key, `object[key]`))
          else:
              self.Writeln (`object`)

          self.Activate()

Mike







From mcfletch at home.com  Mon Mar 19 10:27:51 2001
From: mcfletch at home.com (Mike C. Fletcher)
Date: Mon, 19 Mar 2001 10:27:51 -0500
Subject: newbie - infinite loop 
In-Reply-To: <985012513.18971.0.nnrp-12.c1c3e154@news.demon.co.uk>
Message-ID: <000a01c0b089$29f0d0f0$a9807018@cr706570a>

Your algo is:

for each element in uniqueList:
	if the element equals the line:
		add the line to the list

for goes over the list in order, and you're adding a new element, you'll
find that the list grows each time. Try printing uniqueList on each
iteration to see what you get.  What you probably want is:

for line in file.readlines():
	if line not in uniqueList: # note the "not"
		uniqueList.append( line )

That's not going to be fast, but is the algo you seem to describe.
Enjoy,
Mike

-----Original Message-----
From: Paul Brian [mailto:pbrian at demon.net]
Sent: Monday, March 19, 2001 9:39 AM
To: python-list at python.org
Subject: newbie - infinite loop


All,

I have solved the opriginal problem but am left with a serious question.

Before I discovered the keywords in and not in, I tried to count unique
instances of names in a file.
I devcided to create a (neaarly) empty list, compare each name in the file
with that list. If the name was not in, append to list, otherwise leave it
alone.

After the following was run (and crashed) a couple of times I started using
my brain and realised there must be some "in" operator. However the problem
still stands - a for condition seems completely mutable.

So, line 3 sets a condition, 4 compares "q" to "q" and (wrongly for my
needs) adds "q" to the end of the list.
But then it seems that line 3 realises it has another element in its list
and goes back and compares
the newly appended "q" to my lineand so on ad infinitum (I have to kill the
process each time) !

Surely this should not happen.  Surely the line 3 condition "for every
element in this list", should not constantly expand but should be set solid
till the next pass?

What have I done wrong - is it the mutable list v the immutable tuple
problem (but it should work..surely?) Or is this just the way things are -
"live with it and do not try anything like this again?"

thanks in advance
==============
uniqueList = ['q']               # create an nearl empyty list (seems to
ignore line 3 if truly empty)
line = 'q'                          # fake the retrieval of line from file

for listElement in uniqueList:             #3
    if listElement == line:                    #4
        uniqueList.append(line)            #5 at this point it seems to
think that there is now another list
                                                        #element to compare
and so loops for ever...help!

print len(uniqueList)
==============


--
http://mail.python.org/mailman/listinfo/python-list




From jmarshal at mathworks.com  Fri Mar 16 13:09:32 2001
From: jmarshal at mathworks.com (Joshua Marshall)
Date: 16 Mar 2001 18:09:32 GMT
Subject: Killing threads
References: <3ab23ea7.328575630@News.CIS.DFN.DE>
Message-ID: <98tkss$juc$1@news.mathworks.com>

costas at meezon.com wrote:
> I have th program below. How does one go about kill a thread once
> started? Why is there no method for killing threads? I thought del
> would do it but obviously it doesn't.

I'd like to see this too.

There are the standard responses about how to go about killing threads
(for example having some variable like 'kill_yourself' which the
thread checks every so often), but there genuinely are times that
killing a thread is useful, like if a thread is running code that you
don't have control over.

One thing you can do on unix platforms is send a signal to the main
thread, if that happens to be the one you're interested in killing.


From daniel.dittmar at sap.com  Fri Mar 30 03:35:43 2001
From: daniel.dittmar at sap.com (Daniel Dittmar)
Date: Fri, 30 Mar 2001 10:35:43 +0200
Subject: Where are the examples/samples?
References: <20010330021322.08468.00001688@ng-fe1.aol.com>
Message-ID: <9a1gh0$ml5$1@news1.wdf.sap-ag.de>

The book 'Python Annotated Reference' seems to be along the lines you're
looking for.

Daniel

--
Daniel Dittmar
daniel.dittmar at sap.com
SAP DB, SAP Labs Berlin
http://www.sapdb.org/





From timr at probo.com  Mon Mar 19 14:24:39 2001
From: timr at probo.com (Tim Roberts)
Date: Mon, 19 Mar 2001 11:24:39 -0800
Subject: HTMLparsing abnormal html pages
Message-ID: <200103191924.LAA10010@probo.com>

You wrote:
>
>Tim Roberts wrote:
>> ... If not, is anybody aware of a fair HTML cleaner-upper?
>
>You could use the python-xml code to slurp the HTML into a DOM, and then
>format it using HtmlLineariser:
>
>>>> from xml.dom.writer import HtmlLineariser
>>>> from xml.dom.html_builder import HtmlBuilder
>>>> builder = HtmlBuilder()
>>>> builder.ignore_mismatched_end_tags = 1   # make less fussy 
>>>> html_text = open('public_html/index.html').read()
>>>> builder.feed(html_text)
>>>> pretty_printed = HtmlLineariser().linearise(builder.document)
>
>pretty_printed is now a nicely indented version of html_text.
>
>It's not the fastest thing in the world, but it might help you.

Thanks for taking the time to reply.  Maybe I'm a bonehead, but I can't find the imports you've mentioned.  I downloaded PyXML 0.6.4 (and 0.6.2 just to check), but HtmlLineariser, HtmlBuilder, and html_builder.py do not seem to exist.  The documentation refers to them, and one of the test routines (test_htmlb.py) calls them, but they aren't in the xml/dom tree anywhere.

Has this interface been completely replaced?  It looks to me like this:

   from xml.dom.ext.reader import HtmlLib
   from xml.dom.ext import PrettyPrint
   pretty_printed = PrettyPrint( HtmlLib.FromStream(...) )

performs the same function.  Have I missed a clue somewhere?
--
- Tim Roberts, timr at probo.com
  Providenza & Boekelheide, Inc.



From william.famy at noos.fr  Fri Mar 23 15:20:53 2001
From: william.famy at noos.fr (William Famy)
Date: Fri, 23 Mar 2001 21:20:53 +0100
Subject: newbe   how to read in a string?
References: <3ABB4E70.54885C9F@noos.fr> <mailman.985355670.6600.python-list@python.org> <99g1ac02fug@news2.newsguy.com> <mailman.985373910.14545.python-list@python.org>
Message-ID: <3ABBB025.C2A94026@noos.fr>

Thanks for the answer. I will use map now ;-)

I am using python 1.5.2 runing on a sid debian. I don't know python 2.0
or more.

William

Chris Gonnerman wrote:
> 
> BZZT!  The given string starts with a date which will cause float()
> to throw a ValueError.
> 
> I might say:
> 
>     seq = string.split(a, ",")
>     seq[2:] = map(float, seq[2:])
> 
> > List comprehensions are great, but when you DO want to get
> > map's primary job (applying an already-existing function
> > to parameter sequences) map is still best (with list
> > comprehension, you can avoid using _lambda_ just in order
> > to be able to use map -- I think eschewing lambda makes
> > your code clearer, but maybe that's just me).
_____________________________________________________________

I do not know lambda. I will look for it.


 
> No argument here.  I work in both 1.5.2 and 2.0 and rarely miss
> the 2.0 features, and I'm in the camp that finds ','.split(a)
> ugly.
___________________________________________________________
I do not understand the problem. perhaps later ;-)

  I don't like lambda and avoid it whenever possible.
> 

-- 
 			     \\\|///
                           \\  - -  //
                            (  @ @  )
+-------------------------oOOo-(_)-oOOo-------------------------+
|  Famy William  36 avenue des barattes  74000 Annecy  France   |
|  email:william.famy at noos.fr william.famy at mail-enseignant.com  |
+---------------------------------Oooo--------------------------+
                           oooO   (   )
                          (   )    ) /
                           \ (    (_/
                            \_)


From georg at isdgm.ve.cnr.it  Wed Mar 28 10:12:58 2001
From: georg at isdgm.ve.cnr.it (Georg Umgiesser)
Date: Wed, 28 Mar 2001 17:12:58 +0200
Subject: Speed problems with Python vs. Perl
Message-ID: <99sv33$8hk$1@news.cineca.it>

Last week I wrote a simple Python program and found out that it was 
terribly slow. Therefore I retried with Perl and found a much better 
performance. The programs simple read a file and split the lines on white 
space, something I have to do very often for data elaboration

Here are the two programs:

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

#!/usr/bin/python
 
import sys
import re
 
whitespace = re.compile("\s+")
 
def main():
 
    icount = 0
    for line in sys.stdin.readlines():
        icount = icount + 1
        f = whitespace.split(line)
    print "Total lines read: " + `icount`
 
if __name__ == '__main__':
    main()

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

#!/usr/bin/perl
 
$icount = 0;

while(<>) {
  $icount++;
  @f = split;
}
print "Total lines read: $icount\n";
                                                                            
---------------------------------------------------------

I ran the two programs with the line splitting and also commenting out the
line in which I split on whitespace. Here are the results:


Perl:
 
with line splitting
 
Total lines read: 12212
0.34user 0.00system 0:00.34elapsed 98%CPU (0avgtext+0avgdata 0maxresident)k
0inputs+0outputs (264major+34minor)pagefaults 0swaps
 
without line splitting
 
Total lines read: 12212
0.10user 0.00system 0:00.10elapsed 99%CPU (0avgtext+0avgdata 0maxresident)k
0inputs+0outputs (250major+34minor)pagefaults 0swaps
 
 
Python:
 
with line splitting
 
Total lines read: 12212
1.93user 0.01system 0:01.94elapsed 99%CPU (0avgtext+0avgdata 0maxresident)k
0inputs+0outputs (236major+311minor)pagefaults 0swaps
 
without line splitting
 
Total lines read: 12212
0.20user 0.00system 0:00.20elapsed 99%CPU (0avgtext+0avgdata 0maxresident)k
0inputs+0outputs (233major+311minor)pagefaults 0swaps
 
                                                                           
As you can see, without the splitting (only reading and counting the lines)
Perl is twice as fast, but once I split on whitespace, Python gets more 
than 5 times slower than Perl. 

These results have been achieved on a AMD with 650 MHerz. On my home 
machine, a 266 Celeron, the performance gap is worse. Without line 
splitting the programs
are about the same speed, but with splitting enabled, the Python version
becomes 10 times slower than the Perl version (5 secs against 0.5 secs)

I wonder why this happens? Both languages are interpreted. Is it the
fault of the implementation of the re module? Somebody has had a similar
experience. 

Georg Umgiesser


-- 

                                                             __
                                                           ^/..\^   
----------------------------------------------------------m( 00 )m---
Georg Umgiesser         | e-mail :              georg at isdgm.ve.cnr.it
Oceanography, ISDGM-CNR | web    : http://www.isdgm.ve.cnr.it/~georg/
1364 S. Polo            | tel    :              ++39 - 041 - 5216 875
30125 Venezia, Italia   | fax    :              ++39 - 041 - 2602 340
---------------------------------------------------------------------




From andreas3004 at hotmail.com  Sat Mar 17 14:57:51 2001
From: andreas3004 at hotmail.com (Andreas Kremer)
Date: Sat, 17 Mar 2001 20:57:51 +0100
Subject: text file to array and back
Message-ID: <990fgo$cdm$04$1@news.t-online.com>

Hi,

i am trying to write a python class to make access to a collection of
scientific analysis programs easier for me. In this case I need to read a
file which consists of a time series, collect it into an array (numpy
preferred) and after some operations write it back to disk as a text file.
My solutions are attached. These are my questions:
1. as I am new to python i suspect my joining and splitting not to be very
efficient, especially the joining. 2. Is there a way while reading the file
to store the data immediately into the array?
3. I am concerned about a memory leak. what happens to self.array if I read
the file again? It obviously doesnt append the data to the old, but are the
old data deleted from the memory?

Thanks in advance. Andreas Kremer.

    def __createarray(self):
        "creates an array of the outputfile"
        # open file, transfer content into list self.array
        file = open(self.outputfile, 'r')
        for line in file.readlines():
            dataline = map(float, string.split(line))
            self.array.append(dataline)
        file.close()
        # convert list self.array into an array (Numpy)
        self.array = array(self.array, Float64)

    def __writearray(self):
        "writes array to outputfile"
        # open file, format array and write it to disk
        file = open(self.outputfile, 'w')
        for line in self.array:
            dataline = string.join(str(line),'')
            dataline = dataline+'\n'
            dataline = string.replace(dataline,'[','')
            dataline = string.replace(dataline,']','')
            dataline = string.replace(dataline,',','')
            file.write(dataline)
        file.close()




From ykingma at accessforall.nl  Mon Mar  5 03:57:53 2001
From: ykingma at accessforall.nl (Ype Kingma)
Date: Mon, 05 Mar 2001 09:57:53 +0100
Subject: I come to praise .join, not to bury it...
References: <mailman.983756766.27347.python-list@python.org>
Message-ID: <3AA3550C.11C0EC7A@accessforall.nl>


"Delaney, Timothy" wrote:
> 
> > I have not heard ANY _technical_ arguments opposed to
> > this last time the discussion went around -- nothing but
> > vague aesthetic, "it should be the other way", "I find it
> > ugly" kind of complaints.  Unless and until further notice,
> > then, I class these together with the complaints of people
> > who dislike 0-based indexing, on similar vague bases -- as
> > 0-based indexing _works_ better than 1-based, then, for
> > me, it is a superior choice.  Aesthetics is in the eye of the
> 
> I truly have no complaints with the technical merits of ''.join().
> 
> However, I have *severe* complaints with the aesthetics and consistency.
> Quite simply, the way we look at things tends to be
> 
> "result" = "thing to be acted on to get the result" modified by "thing which
> does the action"
> 

Consider an alternate view:

"result" = "thing that will provide the result" modified by
                    "the action, ie. what thing should do"

> This is consistently followed in the standard python libraries in almost all
> cases ... but ''.join() is a glaring inconsistency. As a result, it seems
> unintuitive and ugly.

In the alternative view ''.join() reads like:

    O yee empty string, please join this list of things.

Would you call that inconsistent?

The difference between the two views is
one of the differences between "procedural" and OO.

After getting used to OO I even found it aesthetic.

Regards,
Ype

-- email at xs4all.nl


From kelley at bioreason.com  Sat Mar 10 16:52:48 2001
From: kelley at bioreason.com (Brian Kelley)
Date: Sat, 10 Mar 2001 21:52:48 +0000
Subject: ANNOUNCE: GladeBase
Message-ID: <3AAAA22F.F1B4B32B@bioreason.com>

GladeBase is an interface to the xml output of Glade (a gtk user
interface designer).

GladeBase has been used internally at Bioreason, Inc for at least 8
months and we have shipped several products using this interface.

GladeBase was designed to give easy access to the functionalities of
libglade through automatically binding a UI class to the Glade designed
window and a Controller class to the Glade defined callbacks.  In this
fashion it supplies a model-view-controller implementation.  A tutorial
was given at the Python 9 conference introducing Gtk/Glade and the MVC
implementation.  This tutorial, as well as the GladeBase sources, are
available at the following url's

http://www.bioreason.com/~brian/GladeBase.html
http://www.bioreason.com/~kelley/GladeBase.html (this will be available
early next week)

There are also some simple examples using Gtk and using Glade that are
currently available and more are on their way.  The tutorial audience
really was pressing the question, "how do you launch other programs
without locking the GUI?"

Enjoy!

--
Brian Kelley          Bioreason, Inc
505 995-8188 ext 207  150 Washington St Suite 303
kelley at bioreason.com  Santa Fe, NM, 87501




From fredrik at pythonware.com  Mon Mar 19 15:53:05 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Mon, 19 Mar 2001 20:53:05 GMT
Subject: Why "from __future__" stinks; a counter-offer
References: <mailman.984975803.9006.python-list@python.org> <jwbnews-3793B2.08394319032001@corp.supernews.com>
Message-ID: <Rkut6.2034$4N4.269099@newsc.telia.net>

John W. Baxter wrote:
> Which is just as well, because it means I can contentedly move on.

you mean there are people on comp.lang.python who have
better things to do than complaining about features they will
never use!?! ;-)

Amazing /F

    <insert favourite Jakob Nielsen quote here>




From fredrik at pythonware.com  Tue Mar 13 18:42:57 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Tue, 13 Mar 2001 23:42:57 GMT
Subject: syslog broken on FreeBSD?
References: <mailman.983631064.7213.python-list@python.org>
Message-ID: <5gyr6.161$4d3.34771@newsb.telia.net>

Gabriel Ambuehl wrote:
> I've been trying to write to syslog using the following
> import syslog
> syslog("TEST)
> but there won't ever appear any message in the syslog. Is this module
> broken on FreeBSD 4.2 or am I doing something very wrong?

did you look for exceptions?

    >>> import syslog
    >>> syslog("TEST")
    TypeError: call of non-function (type module)

try changing your call to

    >>> syslog.syslog("TEST")

(and before you go looking for the message, make sure you know
where LOG_INFO/LOG_USER messages are sent on your machine.
if you don't know, ask your local sysadm for help)

Cheers /F




From larskl at klassekampen.no  Tue Mar 27 09:28:24 2001
From: larskl at klassekampen.no (Lars Klęboe)
Date: Tue, 27 Mar 2001 16:28:24 +0200
Subject: Changing every other instance of <B> in a file
References: <8e41ctkk4l38jmj34070j15iotgv4is2cg@4ax.com> <99q6d7$2brvo$1@ID-11957.news.dfncis.de>
Message-ID: <3e71cto4blbb71qdpp7v3h1kn9lousaq78@4ax.com>

On Tue, 27 Mar 2001 05:56:26 -0800, "Emile van Sebille" <emile at fenx.com>
wrote:
>Here's one way (although I'm interested to see how re might do this):

[--- interresting code ---]

This is another (fancier ;)) implementation of the solution I've
been working on, but I still refuse to believe that the script has
to be this elaborate/big. *twisting my tired gray matter*

I guess re would do the trick (at least if combined with a simple
loop, I'll look into it when I'm really awake.

Thanks for your input !

-- 
Lars Kl?boe


From claird at starbase.neosoft.com  Thu Mar  8 10:43:33 2001
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 8 Mar 2001 09:43:33 -0600
Subject: grabbing return codes from os.system() call
References: <%wFp6.9481$dL4.126404@vixen.cso.uiuc.edu> <98795o$kq8$0@216.39.151.169>
Message-ID: <45BCEEDEDA54E206.15154B99753B2A0C.82A8F3BAA10D6E91@lp.airnews.net>

In article <98795o$kq8$0 at 216.39.151.169>, Donn Cave  <donn at oz.net> wrote:
>Quoth Damian Menscher <menscher+python at uiuc.edu>:
>| I'm new to Python (as in, my experience is essentially the tutorial),
>| but I've already come up with an interesting question:
>|
>| How do I get the return code from an os.system call?  I would have
>| expected I could do something like
>|
>| ---returncode---
>| #/bin/csh
>| echo do stuff
>| exit 3
>|
>| and then in my python program I could do
>|
>| print os.system('./returncode')
>|
>| But it prints out 768.  Not particularly useful, even after I recognize
>| the trick of dividing by 256 (byte-swapping going on?  No, because a
>| return code of 768 reports as 0).  Given that my real return codes
>| will be (possibly large) integers, this limitation will likely cause
>| some serious problems down the line.
>|
>| Even better would be a way of returning a string (the script I run
>| can be something other than csh, but it has to be a separate script).
>|
>| Ideas?  I'm trying to avoid writing the string out to a file and then
>| read the file back in to the python program....
>
>Your return codes will not be larger than 255, that's all the bits
>there are for this kind of thing (on UNIX, anyway.)  If you have a
>bigger number, you will have to put it somewhere else.  So writing
>it out to a file might not be such a bad idea.
>
>	Donn Cave, donn at oz.net

Guys, guys; you are making it too hard on Mr. Menscher.  My guess
is that he'd appreciate being told that, if he creates /tmp/other
with contents
  #!/bin/sh
  echo "This is a string from an external process."
he can then have fun with
  import os
  print os.popen("/tmp/other").read()
-- 

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


From aleaxit at yahoo.com  Tue Mar  6 04:20:06 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Tue, 6 Mar 2001 10:20:06 +0100
Subject: I come to praise .join, not to bury it...
References: <mailman.983485923.1136.python-list@python.org> <97p7iq$kk2$1@nntp6.u.washington.edu> <97rka002k9@news1.newsguy.com> <kuu257kb1n.fsf@lasipalatsi.fi>
Message-ID: <982a4f0su3@news1.newsguy.com>

"Erno Kuusela" <erno-news at erno.iki.fi> wrote in message
news:kuu257kb1n.fsf at lasipalatsi.fi...
> In article <97rka002k9 at news1.newsguy.com>, "Alex Martelli"
> <aleaxit at yahoo.com> writes:
>
> | I have not heard ANY _technical_ arguments opposed to
> | this last time the discussion went around -- nothing but
> | vague aesthetic, "it should be the other way", "I find it
> | ugly" kind of complaints.
>
> what's wrong with vague aesthetic?-) if lots of people share it,
> it's not to be dismissed so lightly.

Lots of people share the 'vague aesthetic' of desiring
to use braces instead of whitespace to delimit blocks
of statements, for example.

Python dismisses that lightly, and that is definitely
a part of what makes it such a wonderful language.


Alex





From nelson at crynwr.com  Tue Mar 27 13:16:53 2001
From: nelson at crynwr.com (Russell Nelson)
Date: Tue, 27 Mar 2001 13:16:53 -0500
Subject: Linux GPS program
References: <3AC01637.4EF3A6F2@crynwr.com> <slrn9c17k4.cfc.jra@dorothy.msas.net>
Message-ID: <3AC0D915.83C2C0C3@crynwr.com>

"Jay R. Ashworth" wrote:
> > Right now, it does moving maps (UTM bitmaps), track recording, and
> > playback. Will select the "best" of multiple maps.
> 
> No *schidt*.  Cool, Russ; thanks.  Any plans to do vector map data as
> well?  (Yeah, I know, I know; "it's never enough". :-)

I'm pretty ignorant here.  What sources of vector information are there,
how big is the dataset, and how do I render it so I can use lat/lon or
easting/northing to get my location?

> Did you use wxPython?

No.  I used pygtk, which is a set of Python bindings to the gtk
toolkit.  I also used libglade to load up an XML file of widget
definitions, and the python bindings to libglade.  The cool thing is
that somebody could write their own glade file, and use it with pygps on
a machine with a big screen.

I ended up with a little Python gps program, a big NMEA parsing module,
and a Lat/Lon to GPS converter module.

-- 
-russ nelson will be speaking at http://www.osdn.com/conferences/brie/
Crynwr sells support for free software  | PGPok | Watch out!  He's got
an
521 Pleasant Valley Rd. | +1 315 268 1925 voice | opinion, and he's not
Potsdam, NY 13676-3213  | +1 315 268 9201 FAX   | afraid to share it!


From m.hadfield at niwa.cri.nz  Sun Mar  4 15:55:29 2001
From: m.hadfield at niwa.cri.nz (Mark Hadfield)
Date: Mon, 5 Mar 2001 09:55:29 +1300
Subject: Performance when working directory is slow network drive
References: <983411179.898268@clam-ext> <mailman.983425921.14373.python-list@python.org> <983480535.813990@clam-ext>
Message-ID: <983739387.172624@clam-ext>

"Mark Hadfield" <m.hadfield at niwa.cri.nz> wrote in message
news:983480535.813990 at clam-ext...
> Now only two things puzzle me:

To save you all sleepless nights wondering about the answers to the
questions that I posed, here are the answers.

> 1. Why having the current directory in the search path affects the import
> time for the os module much more than it does for other modules I tested
> (like string). Is this because os is a package?

Basically, yes. Thanks to Fredrick Lundh for suggesting I look at the output
from "python -vv".

> 2. Why the number of empty-string entries at the head of sys.path varies
> between 0 and 2 on my system depending on how the interpreter is invoked
and
> what I ate for breakfast (or something).

One of the empty-string entries occurred because I had a sub-key with an
empty value by mistake under

HKEY_LOCAL_MACHINE\SOFTWARE\Python\PythonCore\2.0\PythonPath

The other is explained in the documentation for sys.path in the Library
Reference.

Thanks to Sean Reifschnedier and Fredrik Lundh for helping me with this and
to them & everyone else for not abusing me.

---
Mark Hadfield
m.hadfield at niwa.cri.nz  http://katipo.niwa.cri.nz/~hadfield
National Institute for Water and Atmospheric Research




From fig at monitor.net  Wed Mar 21 23:23:52 2001
From: fig at monitor.net (Stephen R. Figgins)
Date: Wed, 21 Mar 2001 20:23:52 -0800
Subject: Puzzling input problem
References: <3AB9449C.A1D14738@monitor.net> <mailman.985233449.28064.python-list@python.org>
Message-ID: <3AB97E58.7B01D485@monitor.net>

> Actually, raw_input isn't the culprit, exactly.  The print 
> statement's format handler doesn't know that raw_input() has 
> modified the screen contents, so it still behaves like the cursor is 
> just after the "prompt string: " on the screen, inserting a space 
> as the ref guide says it will.

Hmmm... very interesting.  The first print doesn't put the space in
right away, but waits for the next call to the print statement to add
the space.  That is not what I would have expected.  Interesting!

Stephen


From sholden at holdenweb.com  Thu Mar 22 23:26:58 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 22 Mar 2001 23:26:58 -0500
Subject: PEP 240 scares me
References: <Pine.LNX.4.21.0103222338530.29360-100000@kev.ioc.ee> <99e2h61aao@news2.newsguy.com>
Message-ID: <EjAu6.13052$_46.499367@e420r-atl2.usenetserver.com>

"Alex Martelli" <aleaxit at yahoo.com> wrote in message
news:99e2h61aao at news2.newsguy.com...
> "Pearu Peterson" <pearu at cens.ioc.ee> wrote in message
> news:Pine.LNX.4.21.0103222338530.29360-100000 at kev.ioc.ee...
>     [snip]
> > But making 1.2 to be a rational
> >                                 1
> >                                ---
> >                                 2
> > will most definitely surprise new (Python) programmers even more
*because*
>
> Nobody's proposing THAT, AFAIK -- 1.2 would be the rational 6/5 (and I
> gather the idea is that one uses 1.2e0 to get a float of similar value).
>
>
Maybe Pearu complains about a typo which looks unlike an error?

regards
 Steve





From cce at clarkevans.com  Tue Mar 20 14:48:38 2001
From: cce at clarkevans.com (Clark C. Evans)
Date: Tue, 20 Mar 2001 14:48:38 -0500 (EST)
Subject: New PEP: The directive statement
In-Reply-To: <200103201042.LAA20248@pandora.informatik.hu-berlin.de>
Message-ID: <Pine.LNX.4.21.0103201445410.21343-100000@clarkevans.com>

On Tue, 20 Mar 2001, Martin von Loewis wrote:
>     A directive_statement is a statement of the form
> 
>         directive_statement: 'directive' NAME [atom] [';'] NEWLINE
> 
...

>     Introducing 'directive' as a new keyword might cause
>     incompatibilities with existing code. Following the guideline in
>     [1], in the initial implementation of this specification,
>     directive is a new keyword only if it was used in a valid
>     directive_statement (i.e. if it appeared as the first non-string
>     token in a module).

I'm new to python, but why not have the module have a built-in
set of directives as a list?  As this could then be an attribute
of many different objects, including modules.

__directives__ = { 'transitional', 'nested-scopes' }





From sholden at holdenweb.com  Fri Mar  9 08:59:19 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Fri, 9 Mar 2001 08:59:19 -0500
Subject: BCD or fixed point numbers
References: <3AA8E272.7268DCFF@poczta.onet.pl>
Message-ID: <Go5q6.45192$1D5.2038486@e420r-atl1.usenetserver.com>

Sorry: wrong link. That should have been

    http://mail.python.org/pipermail/python-dev/1999-December/001771.html

regards
 Steve

"jacek czerwinski" <jacy at poczta.onet.pl> wrote in message
news:3AA8E272.7268DCFF at poczta.onet.pl...
> Has anybody some links for python with BCD or other fixed point numbers
> ?
>
>
> --
> Profesjonalny web-hosting? Tylko Serwis RUBIKON - http://rubikon.pl




From adjih-pam at crepuscule.com  Tue Mar 27 03:50:34 2001
From: adjih-pam at crepuscule.com (Cedric Adjih)
Date: 27 Mar 2001 08:50:34 GMT
Subject: Python and emacs
References: <mailman.985286728.17308.python-list@python.org>
Message-ID: <99pk8q$o1v$1@news.europeonline.net>

Don Dwiggins <dwig at advancedmp.net> wrote:
> A semi-idle question: has anyone looked into embedding Python as an
> extension language in Emacs, alongside Elisp?  I'm toying with the idea of
> doing some Emacs enhancements/modifications, and given the choice, I'd much
> rather do them in Python.  (One such enhancement would be to add
> Pythonwin-style "intellisense" popups to python-mode.)

A proof of concept is at <http://www.crepuscule.com/pyemacs/>,
running Python as a separate process.
(Independantly, there are also old patches for XEmacs, to link Python
into XEmacs ; another proof of concept)

-- Cedric.


From gunturu_neelima at rediffmail.com  Tue Mar  6 02:09:32 2001
From: gunturu_neelima at rediffmail.com (neelima  g)
Date: 6 Mar 2001 07:09:32 -0000
Subject: (no subject)
Message-ID: <20010306070932.23889.qmail@mailweb7.rediffmail.com>

sir,

     please inform me how to enter the data into a database which is having more than two columns,a text database.

                       thanking you,
                         neelima g.,
                        confirm 881077.

_____________________________________________________
Chat with your friends as soon as they come online. Get Rediff Bol at
http://bol.rediff.com







From jp_reed at yahoo.com  Thu Mar  8 05:27:08 2001
From: jp_reed at yahoo.com (Jeremy Reed)
Date: Thu, 8 Mar 2001 02:27:08 -0800
Subject: Who am I: can a class instance determine its own name?
References: <mailman.984034150.9635.python-list@python.org>
Message-ID: <taed32eml9qo9b@corp.supernews.com>

The easiest solution would be to give the name as a parameter at
instantiation and store it in the class as a member variable.

i.e.

class Foo:
    def whoami(self):
        print "Class: Foo   Name:",self.name
    def __init__(self, name):
        self.name = name

FooBar = Foo('FooBar')
FooBar.whoami()
#########
Class: Foo   Name: FooBar
#########
Offhand, I don't think there is a function call capable of doing what you
ask although I could be mistaken.


Tim CHURCHES wrote in message ...
This is probably an elementary question and the answer is probably writ
large in multiple places in the Python documentation, but...

...can an instance of a class determine the name of the variable to which it
is assigned? For example:

###########################
class Foo:
    def whoami(self):
         return "You are a Foo() but I do not know your name"

FooBar = Foo()

print FooBar.whoami()
###########################

How does one define the method whoami() so that it returns "FooBar"? This
sort of navel gazing is formally called introspection, I think (therefore I
am)?

Tim Churches
Sydney, Australia
(where, due to the Coriolis effect, the Python prompt does indeed look like
this: <<< - or maybe its because we are upside-down)









From jkraska1 at san.rr.com  Fri Mar 23 21:52:48 2001
From: jkraska1 at san.rr.com (Courageous)
Date: Sat, 24 Mar 2001 02:52:48 GMT
Subject: Good Editor? (What's the big deal with syntax colouring?)
References: <mailman.985291829.27633.python-list@python.org> <99dqqo$gpi$1@news.mathworks.com> <99e9sv$usmp$1@ID-11957.news.dfncis.de> <99emg2$qif@gap.cco.caltech.edu>
Message-ID: <ha3obts0kp05lonh3o25fsesr38bouo839@4ax.com>

>White on dark olive-green, baby!  Comments, of course, are in italicized 
>teal. I think the reason I find it so appealing is that looks like the color of 
>white chalk on a blackboard.  Very readable and low-strain.

*shrug*. I like black on parchment. Looks like print on paper. Very
readable and low-strain. :)

C//



From thomas.heller at ion-tof.com  Thu Mar  1 10:48:16 2001
From: thomas.heller at ion-tof.com (Thomas Heller)
Date: Thu, 1 Mar 2001 16:48:16 +0100
Subject: [Distutils] Re: CPAN functionality for python - requirements
References: <005d01c0a253$bc513dc0$f05aa8c0@lslp7o.int.lsl.co.uk>
Message-ID: <037301c0a267$07e60460$e000a8c0@thomasnotebook>

> Thomas Heller replied:
> > I wrote some docs. They are online already in Fred Drake's development
> > version:
> > http://python.sourceforge.net/devel-docs/
> 

Tony J Ibbs wrote:
> Erm - all I can see there that sound at all relevant are "Distributing
> Python Modules" and "Installing Python Modules", both credited to Greg
> Ward, and both in that <rant>infernally infuriating "one smidgeon of
> text per HTML page" format (which is worse than useless, so far as I'm
> concerned!)[1]</rant>.
> 
> Could you give a clearer reference, please?

This is the patch I posted:
http://mail.python.org/pipermail/distutils-sig/2001-February/001969.html

Note that I didn't say that the docs are complete...

Thomas





From mmealman at loki.thebisgroup.com  Thu Mar 22 15:55:28 2001
From: mmealman at loki.thebisgroup.com (Mark)
Date: 22 Mar 2001 14:55:28 -0600
Subject: Why Python/Jython?
References: <99cl9e$b2b$1@Merlin.et.bocholt.fh-gelsenkirchen.de> <mailman.985260089.28444.python-list@python.org> <1901D50DC30D30E4.16551C1AB0FB4363.C0D72E208759F8A7@lp.airnews.net> <99dkg301894@news1.newsguy.com> <3ABAB2F7.127D3872@webamused.com>
Message-ID: <slrn9bl0ka.9ck.mmealman@loki.thebisgroup.com>

In article <3ABAB2F7.127D3872 at webamused.com>, Joshua Macy wrote:
>Alex Martelli wrote:
>> 
>> "Cameron Laird" <claird at starbase.neosoft.com> wrote in message
>> news:1901D50DC30D30E4.16551C1AB0FB4363.C0D72E208759F8A7 at lp.airnews.net...
>>     [snip]
>> >     meaningful.  People do NOT start learning
>> >     Java lightly; it often takes hours just to
>> >     install and configure a minimal JDK, let
>> 
>> *blink* -- I'm hardly a Java-ite, myself, but
>> "install and configure a minimal JDK" I would
>> estimate as typically a 10-minute job (on, say,
>> NT -- is it an order of magnitude harder on,
>> say, Linux?!).  Maybe I misunderstand what
>> you mean...?
>>
>
>  It only takes about 10 minutes to download a Win32 installer and run
>it--but if by configure one means understand and implement workable
>solutions to the cryptic and bizarre interaction between Java package
>namespaces, classpaths, and the WinNT filesystem, I think an hour might
>be an underestimate.  Most programmers I know who art trying to "get
>into" Java end up either putting every jar in the world into their
>CLASSPATH environment variable, or downloading and installing something
>like SmartJ...but it takes them more than an hour to arrive at one of
>those solutions unless they have someone to help them.
>
>  Joshua

Yeah. Let's see, been programming for 15+ years, know about 10+ languages,
got into Java right after its beta(JDK 1.02) and spent two years screwing
around with it.

But I'm still real fuzzy on how their paths/jars/imports fully work and it's
a pain in the ass for me to install any app that wants to run off the JDK or
JRE, mostly because getting the stupid CLASSPATH right makes Quantum physics
look easy.

Java = good intentions, bad implementations.

-Mark



-----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
-----==  Over 80,000 Newsgroups - 16 Different Servers! =-----


From jmarshal at mathworks.com  Thu Mar 22 16:28:56 2001
From: jmarshal at mathworks.com (Joshua Marshall)
Date: 22 Mar 2001 21:28:56 GMT
Subject: Good Editor? (What's the big deal with syntax colouring?)
References: <mailman.985291829.27633.python-list@python.org>
Message-ID: <99dqqo$gpi$1@news.mathworks.com>

John Schmitt <jschmitt at vmlabs.com> wrote:
...
> The one-and-only colour combination that works best for source code is
> bright green on black.
...

Yes, you're right of course.  How silly of us.


From aleaxit at yahoo.com  Thu Mar 22 18:38:43 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 23 Mar 2001 00:38:43 +0100
Subject: PEP 240 scares me
References: <Pine.LNX.4.21.0103222338530.29360-100000@kev.ioc.ee>
Message-ID: <99e2h61aao@news2.newsguy.com>

"Pearu Peterson" <pearu at cens.ioc.ee> wrote in message
news:Pine.LNX.4.21.0103222338530.29360-100000 at kev.ioc.ee...
    [snip]
> But making 1.2 to be a rational
>                                 1
>                                ---
>                                 2
> will most definitely surprise new (Python) programmers even more *because*

Nobody's proposing THAT, AFAIK -- 1.2 would be the rational 6/5 (and I
gather the idea is that one uses 1.2e0 to get a float of similar value).


Alex





From stephen_purcell at yahoo.com  Fri Mar  2 14:13:00 2001
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Fri, 2 Mar 2001 20:13:00 +0100
Subject: create new statement?
In-Reply-To: <3A9FE7C2.1F5525D2@troikanetworks.com>; from bedge@troikanetworks.com on Fri, Mar 02, 2001 at 10:34:42AM -0800
References: <3A9FE7C2.1F5525D2@troikanetworks.com>
Message-ID: <20010302201300.A7718@freedom.puma-ag.com>

Bruce Edge wrote:
> I want to add my own command to the interp, but not have to use object
> calling syntax
> eg:
>     create obj opt
> not
>     create(obj, opt)
> 
> How do I define new statements, or, a way of calling functions using the
> former syntax?

You can get the effect you want by subclassing 'Cmd' in the 'cmd'
standard library module and using your subclass in place of the usual
interactive command interpreter.

-Steve

-- 
Steve Purcell, Pythangelist
Get testing at http://pyunit.sourceforge.net/
Get servlets at http://pyserv.sourceforge.net/
"Even snakes are afraid of snakes." -- Steven Wright



From cwc_NO at sound_NO.net  Sun Mar 25 18:46:49 2001
From: cwc_NO at sound_NO.net (Cary Coulter)
Date: Sun, 25 Mar 2001 23:46:49 GMT
Subject: ftpmirror.py problems
Message-ID: <vfhqbtgss3kotgac8gomnn0i9muh6j9ir7@4ax.com>

I am trying to use ftpmirror.py to duplicate various subdirectories.
When I test the replication between two hosts on my local network,
everything works like expected.

However, I am having problems replicating hosts residing on the
internet to my local system.  My local system is on a private network
(192.168...) behind a Linux firewall using IP Masq.  In this
environment, I am experiencing "hangs" in the ftpmirror.py
application.  I have to kill it with ^C.

I can normally download 10-40 files before this happens.  It seems to
be more prone if the files are smaller (can download in 3-5 seconds on
a cable modem) rather than the 10+ MB files.

Is anyone else experiencing this behavior?  Since it seem to work well
beween two local hosts, I am suspicious of the firewall and IP Masq.
I would be grateful for any pointers anyone can give to help remedy
this situation.

I am relatively new to Python. I really like what I see but I'm not
ready to debug this application.  I assume it really works and that
the problem is something else.

Thanks in advance for your input
Cary



From no at thank.you  Fri Mar 30 15:58:02 2001
From: no at thank.you (Owen Ap Owen)
Date: Fri, 30 Mar 2001 12:58:02 -0800
Subject: mxODBC chokes on dates! Any Help?
References: <MPG.152e4c579fa0baef9896a5@news.pacbell.net> <ulmpnnmte.fsf@ctwd0143.fitlinxx.com>
Message-ID: <MPG.152e9338351a7e1b9896a7@news.pacbell.net>

Thanks for the info. It is SQL Server that I am connecting to. But I got 
the same error with date fields when I set up a test connection to 
Access.

It's nice that a fix is in the works. But this seems like something of a 
show stopper (at least for the win32 side of things). 

Considering that the creator of mxODBC is trying to charge a laughably 
exorbitant license fee for commercial use he really needs to fix this 
kind of stuff very quickly. He also needs to add a lot more functionality 
and even then his licensing concept is pretty extreme.


In article <ulmpnnmte.fsf at ctwd0143.fitlinxx.com>, db3l at fitlinxx.com 
says...

> I ran into the same problem with some date fields in an MS SQL Server
> database, and Marc-Andre was able to revert back to an earlier way of
> handling date fields (he preferred a different C data type if defined
> in 2.0.0 than in earlier versions) to resolve it.  I was traveling and
> was a little tardy getting back to verify that the fix worked, but I
> expect it will show up in the next release of mxODBC.


From alf at leo.logilab.fr  Fri Mar  9 06:21:52 2001
From: alf at leo.logilab.fr (Alexandre Fayolle)
Date: Fri, 9 Mar 2001 11:21:52 +0000 (UTC)
Subject: OODB question
References: <WM1q6.7817$Z%5.6737@NewsReader>
Message-ID: <98aecg$13a9$1@norfair.nerim.net>

chris lamb <arinagour at yahoo.co.uk> wrote:
> I am sure that I read somewhere of an OODB implementation developed
> specifically with Python or having a Python interface.

> I am beginning to wonder if I dreamt it up since I cannot remember where I
> read about it!

> Can anyone help me ? (I doubt it!)


You may also want to check 4ODS, which is part of 4Suite 
(http://www.4Suite.org), and provides an ODMG compliant access to PostgreSQL,
Oracle or Dbm.

Alexandre Fayolle
-- 
http://www.logilab.com 
Narval is the first software agent available as free software (GPL).
LOGILAB, Paris (France).


From sholden at holdenweb.com  Thu Mar  8 03:53:25 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 8 Mar 2001 03:53:25 -0500
Subject: Who am I: can a class instance determine its own name?
References: <mailman.984034150.9635.python-list@python.org>
Message-ID: <UPHp6.38461$1D5.1679137@e420r-atl1.usenetserver.com>

This question recurs so frequently I would appreciate knowing of any
omprovement which could be made to FAQ entry 4.97: How can my code discover
the name of an object?

This tries to address the question in a helpful manner.

still-wishing-i-was-at-the-conference-ly y'rs  - steve

"Tim CHURCHES" <TCHUR at doh.health.nsw.gov.au> wrote in message
news:mailman.984034150.9635.python-list at python.org...
This is probably an elementary question and the answer is probably writ
large in multiple places in the Python documentation, but...

...can an instance of a class determine the name of the variable to which it
is assigned? For example:

###########################
class Foo:
    def whoami(self):
         return "You are a Foo() but I do not know your name"

FooBar = Foo()

print FooBar.whoami()
###########################

How does one define the method whoami() so that it returns "FooBar"? This
sort of navel gazing is formally called introspection, I think (therefore I
am)?

Tim Churches
Sydney, Australia
(where, due to the Coriolis effect, the Python prompt does indeed look like
this: <<< - or maybe its because we are upside-down)









From tino at wildenhain.de  Fri Mar  9 17:43:51 2001
From: tino at wildenhain.de (Tino Wildenhain)
Date: Fri, 09 Mar 2001 23:43:51 +0100
Subject: help needed with forms' parameters handling
References: <86FC34EA3C4F074AA6BBFFE52ED83FED27BE3E@METAEXCH.sfo.metamarkets.com>
Message-ID: <3AA95CA7.DCF14032@wildenhain.de>

Hi Petar,

I suppose after a couple of headaches you may want to try out
zope as well ;) http://www.zope.org/
It saves you from most of the work with forms and many more.
Hard to grasp the principe of operation at first but profitable
at the end.

HTH
Tino

Petar Karafezov wrote:
> 
> Hello all -
> 
> I am running Apache/1.3.14 (Unix) mod_python/2.7.2 Python/2.0. At this point
> I am not even sure that this is a mod_python issue, as I have seen this
> same behavior before I ported my cgi to mod_python. I am not sure what to do
> at this point..
> 
> Here's the issue:
> 
> When you do a POST to a python script you obtain the parameters using the
> util.FieldStorage, when using cgi it's cgi.FieldStorage. The returned result
> is a dictionary { key, value }. Everything is OK if you just what to work
> with this dictionary. However, if you try to pass this dictionary object to
> urllib.urlencode, we raise an exception:
> 
> Mod_python error: "PythonHandler parse2"
> 
> Traceback (most recent call last):
> 
>   File "/usr/local/lib/python2.0/site-packages/mod_python/apache.py", line
> 189, in Dispatch
>     result = object(self.req)
> 
>   File "/usr/local/apache/htdocs/test/parse2.py", line 147, in handler
>     params=urllib.urlencode(form)
> 
>   File "/usr/local/lib/python2.0/urllib.py", line 1086, in urlencode
>     for k, v in dict.items():
> 
> AttributeError: 'FieldStorage' instance has no attribute 'items'
> 
> Now, the funny part is that if I get this object 'form' - which was defined
> as form=util.FieldStorage(req) and just get the keys and values and form a
> new dictionary, I can pass that new one to urllib.urlencode with no
> problems.
> 
> ...
>     newparams={}
>     for key in form.keys():
>           newparams[key]=form[key]
>     params=urllib.urlencode(newparams)
> ...
> 
> Again, I run into the same issue with both util.FieldStorage when working
> with mod_python and with cgi.FieldStorage when using my cgi version.
> 
> I am quite happy with this workaround until I bumped into a different issue,
> which I am not exactly sure if it's related or not. When there're multiple
> checkboxes on a form:
> 
> <input type="checkbox"  name="frequencies" value="DEM" priority="6" checked>
> <input type="checkbox"  name="frequencies" value="HLF" priority="6">
> <input type="checkbox"  name="frequencies" value="DAY" priority="6">
> <input type="checkbox"  name="frequencies" value="WEK" priority="6">
> 
> and we have selected more than 1, then the form parameter value gets passed
> to the server as an array. For example, if I'd select DEM and DAY from the
> checkboxed above, my params would be:
> 
> 'frequences' : ['DEM','DAY']
> 
> This is all great, but when I pass the parameters to the server - my python
> script is playing the role of a proxy, then I am passing as value for
> 'frequences' the STRING "['DEM','DAY']" and not the ARRAY ['DEM','DAY'].
> 
> So, somewhere in what I do/use the array gets converted to a string. The way
> I see it is:
> 
> 1) It could be the python's FieldStorage when I originally get the keys and
> values
> 2) It could be that funky dictionary rewriting that I am doing before I pass
> the form parameters to urllib.urlencode.
> 3) I could be something in what urlencode does.
> 
> I am not sure where to look at this point.
> 
> Any ideas, anyone ?? Any help would be greatly appreciated!
> 
> Thanks for your help!
> 
> regards,
> 
> Petar Karafezov
> 
> MetaMarkets.com
> 415-575-3015
> -------------------------------------------
> Investing Out Loud at
> http://www.metamarkets.com
> -------------------------------------------
> 
> --
> http://mail.python.org/mailman/listinfo/python-list



From aahz at panix.com  Wed Mar 28 14:35:09 2001
From: aahz at panix.com (Aahz Maruch)
Date: 28 Mar 2001 11:35:09 -0800
Subject: list vs tuple
References: <hGlw6.782$p5.3210@news1.rivrw1.nsw.optushome.com.au> <99stcb$483$1@panix6.panix.com> <99t9om$2o2q3$1@ID-77331.news.dfncis.de> <buqw6.1999$sk3.675030@newsb.telia.net>
Message-ID: <99tedd$gjh$1@panix6.panix.com>

In article <buqw6.1999$sk3.675030 at newsb.telia.net>,
Fredrik Lundh <fredrik at effbot.org> wrote:
>Mark Pilgrim wrote
>>
>> It was my understanding that only tuples containing immutable objects
>> could be used as dictionary keys.  For instance, (1,2,3) can be, and
>> ("a","b","c") can be, but ([1,2,3], ["a","b","c"]) can not, because
>> the underlying objects could mutate and cause problems.  (Wow, that
>> sounds like a bad sci-fi movie.)  Is this true?
>
>no python interpreter within reach today?
>
>>>> d = {}
>>>> d[(1, 2, 3)] = 1
>>>> d[("a", "b", "c")] = 2
>>>> d[([1, 2, 3], ["a", "b", "c"])] = 3
>Traceback (most recent call last):
>  File "<stdin>", line 1, in ?
>TypeError: unhashable type

Sure, but someone could create a class with __hash__ and do some weird
stuff that way.  Not Recommended.
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

"Boost the stock market -- fire someone"


From ACIUFFO1 at irf.com  Thu Mar  8 14:42:09 2001
From: ACIUFFO1 at irf.com (Andrea Ciuffoli)
Date: Thu, 08 Mar 2001 11:42:09 -0800
Subject: signal
Message-ID: <saa77022.068@irf.com>

Hi,
I'm trying to use signal module to control that an FTP connection is established within a specified time but I got an Attribute error: SIGALRM.
Is that module supported by Windows NT and 98?
If not, is there another method to control that the connection is established without problem?
Thanks,
Andrea.




From whisper at oz.net  Mon Mar 19 15:00:22 2001
From: whisper at oz.net (Dave LeBlanc)
Date: 19 Mar 2001 20:00:22 GMT
Subject: Organizing Python
Message-ID: <995ogm$1u6$0@216.39.170.247>

What seems to be a reasonably trivial question to me, but i've not
been able to find an answer in any of the Python doc i've looked at.

Is there a convention for installing 3rd party add-on modules into a
Python installation so that:
	
	It's easy to update Python without reinstalling all the extras

	One doesn't have to have a mile-long PYTHONPATH

I've seen some mention of packages and local/packages directories wrt
to Linux installations, but I'm looking for a Windows solution.

I think part of why I have to ask this question is not quite
understanding how PYTHONPATH and import works. When you say "import
foo", then foo is taken to be a file and if you say "from foo import
bar", then foo is taken to be a directory? Would this mean that one
would have to have to have both foo and foo's parent directory in
PYTHONPATH?

Apologies if this seems like a trivial question; i'd just like to
learn how to do things pythonically.

CC to my email appreciated: whisper at oz.net

Regards,

Dave LeBlanc


From dsh8290 at rit.edu  Wed Mar  7 20:20:52 2001
From: dsh8290 at rit.edu (D-Man)
Date: Wed, 7 Mar 2001 20:20:52 -0500
Subject: learning C
In-Reply-To: <slrn9adklr.bv9.scarblac@pino.selwerd.nl>; from scarblac@pino.selwerd.nl on Thu, Mar 08, 2001 at 12:32:47AM +0000
References: <mailman.983972829.23481.python-list@python.org> <983975751.18833@newsmaster-04.atnet.at> <slrn9adklr.bv9.scarblac@pino.selwerd.nl>
Message-ID: <20010307202051.B277@harmony.cs.rit.edu>

On Thu, Mar 08, 2001 at 12:32:47AM +0000, Remco Gerlich wrote:
| Werner Schiendl <ws-news at gmx.at> wrote in comp.lang.python:
| > If it is not obsolete, it should be at least considered deprecated for new
| > projects.
| 
| A baseless statement. There are many types of project for which C is still
| the best choice. For one thing, think embedded systems.

C does have it strong points.  After all, python is essentially a C
program.

| I can imagine something like Java deprecating C++, but C isn't going to go
| away soon.

Java?  I thought Python took care of that.

;-)

-D




From tjreedy at udel.edu  Sun Mar 18 21:56:36 2001
From: tjreedy at udel.edu (Terry Reedy)
Date: Sun, 18 Mar 2001 21:56:36 -0500
Subject: Subsetting a dictionary
References: <993dkm$ttr$1@news6.svr.pol.co.uk>
Message-ID: <993sif$92d$1@news.udel.edu>

"Will Newton" <will at nospam.misconception.org.uk> wrote in message
news:993dkm$ttr$1 at news6.svr.pol.co.uk...
>
> I need to get a subset of a dictionary:
>
> D = Dictionary of objects (in this case, training examples for a decision
> tree)
>
> D2 = {D | d in D, d under some constraint}

Hmmm.  What you need is a currently imaginary feature that we might call a
dict comprehension,
with syntax paralleling that for list comprehensions:

D2 = {key:val for (key,val) in D if constraint(key,val)}

This idea might be an argument for (or against?) making the phrase 'key,val
in dict' meaningful.

Anyone for a PEP?

Terry J. Reedy





From jmorris2 at twcny.rr.com  Tue Mar 27 23:10:30 2001
From: jmorris2 at twcny.rr.com (Morrisett)
Date: Wed, 28 Mar 2001 04:10:30 GMT
Subject: CfP: Babel01 Workshop on Multi-Language Infrastructure and Interoperability
References: <3abf4ae1@msrnews.research.microsoft.com> <99o2n2$9on$1@216.39.170.247> <99or9g$6tg$1@mulga.cs.mu.OZ.AU>
Message-ID: <Wudw6.3242$8R5.469322@typhoon.nyroc.rr.com>

"Fergus Henderson" <fjh at cs.mu.oz.au> wrote in message
news:99or9g$6tg$1 at mulga.cs.mu.OZ.AU...
> In summary, the program committee is very strongly committed to
> adhering to the highest academic standards, particularly with
> regard to integrity and fairness in the peer review process.
> We will do all that we can to avoid bias or the perception of bias.

Fergus said this quite well, but I'll reiterate:  The goal of the
workshop is to bring together any and all who are interested in issues
of language interoperability.  Researchers at Microsoft are obviously
interested in this in the context of .NET, but many others of us
are as well.

Oh yes, and my research has not been heavily funded by Microsoft :-)

-Greg




From donod at home.com  Sun Mar 18 23:42:04 2001
From: donod at home.com (Don O'Donnell)
Date: Mon, 19 Mar 2001 04:42:04 GMT
Subject: Python style questions
References: <98tb91$9t9$1@saltmine.radix.net> <98tclh$ap6$2@lahti.tasking.nl> <98th1k$ka2$1@saltmine.radix.net>
Message-ID: <3AB5893C.39BB26D@home.com>


Cary O'Brien wrote:
> 
> In article <98tclh$ap6$2 at lahti.tasking.nl>,
> Dick Streefland <dick_streefland at tasking.com> wrote:
> >cobrien at Radix.Net (Cary O'Brien) wrote:
> >| 4. I need to change things like "88aa99bb" (hex string) into integers.  Should I
...
> >c. use string.atoi(s, 16), or for Python >= 2.0, the builtin int(s, 16).
...
> >
> >>> import string
> >>> string.atoi('abcd',16)
> 43981
> >>>
> 
> Ok, that works.  Missing in my "Python essential reference" though.
> 
hmmm, that's funny it's in my PER, under "string" in the Library
section, page 116:

atoi(s, [, base])
    Converts string s to an integer base is an optional base.

/Don


From tim.one at home.com  Sun Mar 18 23:21:24 2001
From: tim.one at home.com (Tim Peters)
Date: Sun, 18 Mar 2001 23:21:24 -0500
Subject: Why "from __future__" stinks; a counter-offer
In-Reply-To: <5.0.2.1.0.20010318214456.022db7f0@mail.inet.com.br>
Message-ID: <LNBBLJKPBEHFEDALKOLCGEIOJGAA.tim.one@home.com>

[Carlos Ribeiro]
> Have you ever seen this [deleting obsolete statements] happen in *any*
> sufficiently complex system? I'm afraid we'll be stuck with this
> solution" for a long, long time.

Because __future__.py contains reliable info about the state of various
incompatible changes, it will be easy to automate removing obsolete
future_statements; future_statements were deliberately designed so that
extraordinarily simple tools can be written to do this.  If you don't want
to, fine, but you can't sell me the idea that there will be anything other
than sloth preventing it.  This one was designed from the start to be easy to
get rid of.

[Tim]
>> ... in the absence of a competing PEP it's certain  to remain as
>> it is, and the last 2.1 beta release is scheduled for this coming
>> Friday (as>noted in PEP 226)

> So thats the matter? Forgive me, I'm pretty new on this list and I may be
> missing something. The lack of a PEP... this sound too much like
> design-by-comitee to me.

A PEP is "merely" a spec.  Without a spec, people don't really know what's
under discussion, and implementers don't know what to do.  You can rail
against the future_statement because nothing about is hidden:  there's a PEP
that describes it in detail.  The alternatives remain amorphous for lack of a
PEP, and into that lack of detail some people project the fulfillment of
their wishes.  Spell *anything* out in full and complaints zoom.

> I agree with the *intent* of the PEP - to keep  everything nice and
> documented, and to put some order into the process.  However I feel that
> in this case it sounds as a lazy excuse for not thinking about a better
> to solve the problem.

Sorry, but we don't have unbounded time.

> It's something like "the PEP is everything, lets follow the PEP as-is".

Yes indeed!  The spec *is* everything, and even Guido has to write PEPs.

> Why is it so difficult to write or change the existing PEP?

Don't look at me:  I gave up a night of sleep to write the future_statement
PEP, and I'm happy with it.  Since it's "good enough" in my eyes, I'm not
going to give up another night of sleep to dream up alternatives; the PEP I
wrote was already my best stab at solving the specific problem I wanted to
solve.  If somebody else wants to, great.

> ...
> You are right - the message is important, and this is one of the most
> compelling reason to abandon "import from __future__". It does send the
> wrong message - in fact it does not make any sense (if you were talking
> about "import quantum computing", that's in the future :-).

Then I suggest you either haven't read the PEP, or the PEP is unclear.  When
I do

    from __future__ import nested_scopes

at the top of a file today, what I'm saying in full is:

    I want to use nested scopes in this module, and I'm writing it
    using a compiler earlier than 2.2.  2.2 doesn't exist yet, and
    nested scopes do not become the rule until that (future) release.
    So now-- under 2.1 --I have to ask for it.  When 2.2 is released, I
    can get rid of this.

> And don't let me get started on the relativity aspects, because the
> "future" is not an  absolute, defined state - it's going to be
> differente for every release.

Of course.  What of it?  When 2.2 comes out, I'll run a cmdline tool to nuke
that subset of my future_statements that are no longer needed.  For people
who take copies of my code before that point and use them under pre-2.2
releases, the future_statements will correctly point out that they need a
future release of the *compiler*.  "future" here is wrt the version of the
language in use, not with respect to your birthdate <0.6 wink>.

> I think that some people got used to the proposed syntax, and as such
> they  dont see anything wrong with it.

I doubt that anyone has had time enough to "get used to" this syntax yet.
Because I wrote the PEP, you can be sure I've seen the most email about it.
Favorable comments outweighed negative about 2 to 1.  That makes it one of
the most popular changes in language history <0.9 wink>.

> ...
> And unfortunately I dont have the resources to grab the source and
> actually *do* something about it, this rant aside.

You don't need to be an implementer to write a PEP.

> Part of my rants are about the name of the module - __future__.py is
> a bad and misleading name in my opinion.

We disagree.

> One alternative would be to call this  module __compiler__.py. It's a
> better name because:
>
> - it is related to the *function* carried out by this module (it
> is handled at compile time, and carries compile-time information)

I agree __future__ doesn't say that, but __compiler__ misses the message that
an incompatible language change is being addressed (and, sorry, but
"incompatible language change" *is* about "before and after" -> time, wrt the
language definition).

> - it has nothing to do with the perception of time

A point against it; see above.

> - it could export other symbols and mappings useful for the script about
> the compile-time environment.

As I said in the msg to which you're replying, I actively want not to
conflate incompatible language changes with general directives.  A
future_statement signals a sea change, it's not just about minor fiddling.

> I dont think that this is a real solution to the problem, but it makes
> it look a little better.

Except for the first point, I think it's worse.

> It is also more consistent and has some prospect of being extensible.

I actively want future_statements not to be extensible.  This mechanism is
directed at a specific task, and one that's quite unlike any other in that it
addresses incompatibility across releases, a prospect so dreadful that it
deserves a unique mechanism.

> Also it's a easy change to make now without breaking any PEPs or the
> schedule.

Since the future_statement is already implemented, and the assorted people
who did the implementation have a full plate of work already lined up for
this week, there is no bandwidth to change it now unless somebody else
volunteers.  Such is life!  Hard as it may be to believe now, I predict
everyone will survive this.

> OTHER ISSUES & WEIRD IDEAS
>
> 1) A different way to handle the problem is to make the Python compiler
> itself scriptable. However, this is also dangerous, and poses a
> much bigger set of problems. The basic concept is to have a __compiler__
> symbol, pretty much like the __main__ symbol, except that this one is set
> only inside the  compiler. All code inside an if __compiler__: block
> would be executed immediately by the compiler.

Sorry, but I don't see how that addresses incompatible language changes.

> 2) There is something else bugging me. It is not clear what is the effect
> that "from __future__" has on the compiler.

Nor will it ever be:

    from __future__ import xyz

can have *any effect whatsoever* at compile time, depending on xyz.  It
could, for example, treat the rest of the module as Perl6 code, if Guido
decides that this Python experiment was a silly waste of time, and that
Python 3.0 should be identical to Perl6.  It's up to the PEP for xyz to spell
out xyz's compile-time semantics.

> PEP-236 states that this is handled at compile time.

The PEP says that it has both compile-time and runtime semantics.

> However the __future__.py module does exist and has some information in
> it.

Yes, and at runtime the future_statement imports the feature name from
__future__.py (this is explained in the PEP).

> For python applications that get distributed, this has a lot of potential
> side effects. Someone may have the right version of the interpreter, but
> the wrong version of the __future__ module.

Since __future__.py is distributed along with the compiler, you may as well
ask what will happen if someone screws up and gets a wrong version of any
library module.  However, the module __future__.py is a passive repository of
release information, so the worst that can happen in this case is that code
doing introspection on that info will get some wrong info.  That has no
effect on the compile-time semantics, though (they're hardwired into the
compiler).

> ...
> 3) In some time it may become be impossible to tell which subset of the
> language you are using - for instance, from a code snippet.

I expect you mean which *version* of the language is being used.  For
example,

    i = 3/4

*may* not return 0 in some future version of Python, and there is indeed no
way to tell which version is intended from looking at that line in isolation.

> This is *very* important to understand, because we are changing the
> semantics of the language, not the syntax, and it will not be obvious
> from the context of a snippet which "switch" has to be used.

Note that in the case of nested scopes, as soon as 2.2 comes out there is
*no* switch to restore 2.1 behavior.  This characteristic is common to all
uses of a future_statement.

> (We have to concede that this is going to be a problem whenever there
> is any semantic change - it is just impossible to avoid).

Sure, and it's just as true of changes to what assorted library modules do.
Not unique to incompatible language changes, or to Python, or to 2.1 or 2.2.

BTW, incompatible language changes are rare.

i-spent-more-time-on-this-reply-than-on-recovering-from-all-incompatible-
    python-changes-since-'91-ly y'rs  - tim




From embed at geocities.com  Tue Mar  6 12:52:09 2001
From: embed at geocities.com (Warren Postma)
Date: Tue, 6 Mar 2001 12:52:09 -0500
Subject: absolute namespace for callbacks?
References: <3AA50AFB.83C7459F@troikanetworks.com> <983895797.296667@newsmaster-04.atnet.at> <3AA51144.3DD3FC3F@troikanetworks.com>
Message-ID: <tu9p6.4987$TW.25313@tor-nn1.netcom.ca>

"Bruce Edge" <bedge at troikanetworks.com> wrote in message
> My callback func is defined in one module, and the callback is executed
> from another module which does not know in advance what module the
> callback func resides in.

If you passed the STRING 'my_function' to someone, then namespace lookup
would be necessary. if you passed a <function object>, no namespaces enter
into it.

Try this:

def foo(bar):
    pass
print type(foo)
print type('foo')

You are used to TCL it seems, where type(anything) is ultimately String.

Warren





From aleaxit at yahoo.com  Tue Mar  6 12:32:29 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Tue, 6 Mar 2001 18:32:29 +0100
Subject: absolute namespace for callbacks?
References: <3AA50AFB.83C7459F@troikanetworks.com> <983895797.296667@newsmaster-04.atnet.at> <3AA51144.3DD3FC3F@troikanetworks.com>
Message-ID: <9836vg0261u@news1.newsguy.com>

"Bruce Edge" <bedge at troikanetworks.com> wrote in message
news:3AA51144.3DD3FC3F at troikanetworks.com...
> My callback func is defined in one module, and the callback is executed
> from another module which does not know in advance what module the
> callback func resides in. So, by default, the caller cannot find the
> callback func in it's own namespace.
> The callback registration procedure needs to specify the namespace where
> the callback func exists.
>
> OTOH, maybe I'm missing something here :-)

Maybe -- why use names, when you can refer to the
callable function-object directly?


Alex





From fredrik at pythonware.com  Thu Mar  1 13:22:10 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Thu, 01 Mar 2001 18:22:10 GMT
Subject: A counter-proposal to __future__ in PEP 236
References: <j47l29lpo8.fsf@informatik.hu-berlin.de>
Message-ID: <mrwn6.16173$Qb7.2637930@newsb.telia.net>

Martin von Loewis wrote:
> To selectively activate nested scopes in Python 2.1, PEP 236 proposes
> the "future" statement. That means that you have to write
>
> from __future__ import nested_scopes
>
> Even though this looks like an import statement, it is not: instead,
> it changes the meaning of variable binding in the presence of nested
> functions.

Really?  I'd say it imports an alternate compiler, and activates it
for all modules that happens to contain a from __future__ state-
ment.

> I personally consider this a confusing notation, and propose a
> different syntax, which would read
>
> directive nested_scopes

Note that we're not talking optional standard features here, we're
talking untested experimental stuff, on a scale too large to handle
in an ordinary prerelease cycle.

__future__ stuff may go away or change, if they turn out to be
a bad idea.  If you're easily confused, you shouldn't use them at
all.

But with your proposal, you make them look like optional features.

That isn't just a bad idea, it makes Python look bad too.

> The directive statement currently only supports activating nested
> scopes, but it may get further use later, e.g. to activate
> case-insensitivity or set the source file character set.

Smells like hypergeneralization to me (which is a worse sin than
premature optimization).  Source code encodings and experimental
compiler features are two entirely different things.

> In this implementation, directive is only considered as a keyword if
> it appears at the beginning of the module

And you think from __future__ is a hack?  Gimme a break.

Cheers /F




From richard at starfighter.freeuk.com  Thu Mar 22 13:55:46 2001
From: richard at starfighter.freeuk.com (Richard Townsend)
Date: Thu, 22 Mar 2001 18:55:46 -0000
Subject: Catching exceptions in Tkinter's mainloop
References: <%dws6.9947$Q47.2723350@news1.rdc1.tn.home.com> <drMs6.32482$HR6.3694604@nnrp4.clara.net> <owzt6.21061$Q47.5935326@news1.rdc1.tn.home.com> <6QNt6.37618$g63.4509118@nnrp3.clara.net> <Qydu6.26214$Q47.7466405@news1.rdc1.tn.home.com>
Message-ID: <oWru6.48063$HR6.5435948@nnrp4.clara.net>

"Mike Callahan" <mcalla at home.com> wrote in message
news:Qydu6.26214$Q47.7466405 at news1.rdc1.tn.home.com...
> Richard,
>
> On my system, the exception is captured and printed in test.log but
mainloop
> is not exited. Is this what happens on your system?
>
> Mike C.
>

I've tested it on Win98 and Linux (both using Python2.0 & Tk8.3) and it
definitely exits the mainloop in both cases. I have no idea why it doesn't
do the same for you...

Richard





From akuchlin at mems-exchange.org  Mon Mar 19 11:00:03 2001
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 19 Mar 2001 11:00:03 -0500
Subject: Python-related standards
Message-ID: <3dae6h21h8.fsf@ute.cnri.reston.va.us>

It struck me today that it might be a good idea to create a series of
Python-related standard documents that would be a counterpart to PEPs.
We already have some documents that would fit into this series:

        * The DB-API.
        * The DO-SIG's CORBA mapping.
        * The SAX and DOM mappings.

One future topic might be standard interfaces for cryptographic
functions (hashes, block ciphers, &c.)  Other topics might be metadata
for Python packages (currently being done in PEP 241), or defining a
structured text format for docstrings.  IMHO such topics aren't really
suitable for a PEP, since they don't necessarily have anything to do
with the core interpreter or language.

What does everyone think?

--amk


From Peter.Dintelmann at dresdner-bank.com  Tue Mar 27 09:16:36 2001
From: Peter.Dintelmann at dresdner-bank.com (Dr. Peter Dintelmann)
Date: Tue, 27 Mar 2001 16:16:36 +0200
Subject: dynamic web pages?
Message-ID: <99q701$jlr10@news-1.bank.dresdner.net>

Hi,

is there any module to create dynamic web pages
using Python?

I am looking for something like Perl's Mason (www.masonhq.com)
(or PHP). The business logic of my web app should be implemented
in Python classes which are accessed through a Python/HTML
mixture.

Thanks in advance,

    Peter





From sholden at holdenweb.com  Wed Mar 28 19:57:57 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Wed, 28 Mar 2001 19:57:57 -0500
Subject: setvbuf() ??
References: <slrn9c50mp.me7.wcuddy@lserv.ja10629.home>
Message-ID: <FPvw6.13939$5H6.443849@e420r-atl2.usenetserver.com>

"Wayne Cuddy" <wcuddy at lserv.ja10629.home> wrote in message
news:slrn9c50mp.me7.wcuddy at lserv.ja10629.home...
> I need to print a period on an xterm every few seconds to provide status
to the
> user, I don't want to force a newline.  How do I turn off line buffering
for
> standard out?
>
If you're writing the xterm as a file I would have thought the file.flush()
function would do this for you.

regards
 Steve





From daniel.orlowski at bg-sys.de  Thu Mar  1 08:19:43 2001
From: daniel.orlowski at bg-sys.de (Daniel Orlowski)
Date: Thu, 01 Mar 2001 14:19:43 +0100
Subject: how can i copy a sublist
Message-ID: <3A9E4C6F.E2DD63C3@bg-sys.de>

Hi,

i try to copy a sublist from list "a" to list "b". After that i alter a
value in "b" and oops! value in "a" is also altered.
so it looks like that:
>>> a=[[1,2],[3,4]]
>>> b=[]
>>> b.append(a[0])
>>> b
[[1, 2]]
>>> b[0][0]=3
>>> a
[[3, 2], [3, 4]]

how can i copy a sublist without
"get_a_value_from_a_list_and_put_it_into_another"

Regards,
Daniel




From robin.thomas at starmedia.net  Thu Mar  1 02:03:53 2001
From: robin.thomas at starmedia.net (Robin Thomas)
Date: Thu, 01 Mar 2001 02:03:53 -0500
Subject: "import X" and "from X import y" produce different
  results, Why?
In-Reply-To: <ifmn6.6635$hn5.857862@news1.rdc1.mb.home.com>
Message-ID: <4.3.1.2.20010301020014.00d33b50@exchange.starmedia.net>

At 06:46 AM 3/1/01 +0000, Parzival Herzog wrote:

>So, apparently the imported X is a copy of xx.X, not the same object by
>another name.
>Can anyone explain this result? Is this result justified by a Python
>language specifcation?

You "re-bound" the local name X to 99. When you did that, the object 
previously bound to the local name X -- 35, which was bound to X by "from 
xx import X" -- was unbound. A clearer example without the import to 
confuse you:

X = 35
print X
X = 99
print X


--
Robin Thomas
Engineering
StarMedia Network, Inc.
robin.thomas at starmedia.net




From jafo at tummy.com  Sat Mar 24 01:55:01 2001
From: jafo at tummy.com (Sean Reifschneider)
Date: Fri, 23 Mar 2001 23:55:01 -0700
Subject: Python T-shirts
In-Reply-To: <99fr5e$sp4$1@panix6.panix.com>; from aahz@panix.com on Fri, Mar 23, 2001 at 07:46:54AM -0800
References: <mailman.985019605.8781.python-list@python.org> <99fr5e$sp4$1@panix6.panix.com>
Message-ID: <20010323235501.C30668@tummy.com>

On Fri, Mar 23, 2001 at 07:46:54AM -0800, Aahz Maruch wrote:
>If you're in the USA, S&H is $3.50, for a total cost of $13.50.  Also,

Yeah, apparently the secret with T-shirts is to send them Express Mail in
the kind of "courier envelope", which is flat-rate...

Sean
-- 
 "Isn't having a smoking section in a restaurant kind of like having a
 peeing section in a swimming pool?"  -- David Broadfoot
Sean Reifschneider, Inimitably Superfluous <jafo at tummy.com>
tummy.com - Linux Consulting since 1995. Qmail, KRUD, Firewalls, Python



From rol9999 at attglobal.net  Wed Mar 28 21:43:29 2001
From: rol9999 at attglobal.net (Roland Schlenker)
Date: Wed, 28 Mar 2001 21:43:29 -0500
Subject: setvbuf() ??
References: <slrn9c50mp.me7.wcuddy@lserv.ja10629.home>
Message-ID: <3AC2A151.6513795F@attglobal.net>

Wayne Cuddy wrote:
> 
> I need to print a period on an xterm every few seconds to provide status to the
> user, I don't want to force a newline.  How do I turn off line buffering for
> standard out?

How about:

import sys

sys.stdout.write(".")
sys.stdout.flush()

Roland Schlenker


From ykingma at accessforall.nl  Wed Mar 28 17:03:56 2001
From: ykingma at accessforall.nl (Ype Kingma)
Date: Wed, 28 Mar 2001 23:03:56 +0100
Subject: Is inheritance broken?
References: <mailman.985808317.25055.python-list@python.org>
Message-ID: <3AC25FB7.7FD8759D@accessforall.nl>

Jonathan,

<snip>

> 
> If I really wanted this to work, guess one thing I could do is to cheat by
> creating a 'flat' list of classes and then manually creating my class tree.
> Something like this:
> 
> <code>
> 
> # 'flat' classes
> class parallelogram_flat:
>         def angle(self): return "variable"
>         def side(self): return "variable"
> 
> class rhombus_flat:
>         def side(self): return "same"
> 
> class rectangle_flat:
>         def angle(self): return "90 degree"
> 
> class square_flat:
>         pass
> 
> # 'real' classes
> class parallelogram(parallelogram_flat):
>         pass
> class rhombus(rhombus_flat, parallelogram_flat):
>         pass
> class rectangle(rectangle_flat, parallelogram_flat):
>         pass
> class square(square_flat, rhombus_flat, rectangle_flat, parallelogram_flat):
>         pass
> 
> # usage
> 
> s = square()
> print "squares have", s.angle(), "angles"
> print "squares have", s.side(), "sides"
> 
> </code>
> 
> Ugly and error prone, no?
>

This code follows a rule: never instantiate a class that
is inherited from. (In other languages your 'flat' classes
might be called 'abstract' in order to indicate that they cannot
be instantiated.)
IIRC this is a design pattern in class hierarchies.

In this particular case you avoid having a real square that
inherits from a real rectangle. In geometry a square
is a special case of rectangle. However this special case
from geometry is not implemented well by using OO inheritance.

The 'flat' and 'real' approach in the above class hierarchy
feels a lot better. It allows more choice of where to put
eventual attributes (eg. side length).

So, no it's not ugly at all, quite the contrary.
(Being non native in English, I don't know what a rhombus is,
so I might change my mind...)

It's also not really error prone: you only need to check
right to left, there is no need to go deeper into
the inheritance hierarchy.


Regards,
Ype Kingma


-- 
email at xs4all.nl


From sholden at holdenweb.com  Thu Mar  8 21:48:37 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 8 Mar 2001 21:48:37 -0500
Subject: signal
References: <mailman.984080591.17942.python-list@python.org>
Message-ID: <UzXp6.41938$1D5.1931646@e420r-atl1.usenetserver.com>

"Andrea Ciuffoli" <ACIUFFO1 at irf.com> wrote in message
news:mailman.984080591.17942.python-list at python.org...
Hi,
I'm trying to use signal module to control that an FTP connection is
established within a specified time but I got an Attribute error: SIGALRM.
Is that module supported by Windows NT and 98?
If not, is there another method to control that the connection is
established without problem?
Thanks,
Andrea.

Look for the timeoutsocket module. I think it's in the Vaults of Parnassus,
which I believe you can search from www.python.org

regards
 Steve




From fredrik at pythonware.com  Sat Mar 31 04:21:56 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Sat, 31 Mar 2001 09:21:56 GMT
Subject: list vs tuple
References: <hGlw6.782$p5.3210@news1.rivrw1.nsw.optushome.com.au> <3AC20CD3.265A2D94@alcyone.com> <ru69cto61jtmrqsf809e38hb3pbm2lhi7g@4ax.com> <bC1x6.4319$4N4.905946@newsc.telia.net> <71q9ct4le63anefmiajj5cbh3mg04qu1ga@4ax.com> <mailman.985991202.5578.python-list@python.org> <7Bfx6.989$p5.3791@news1.rivrw1.nsw.optushome.com.au>
Message-ID: <Ukhx6.4421$4N4.950603@newsc.telia.net>

"deadmeat" <root@[127.0.0.1]> wrote:
> It's dependant on datatype, which most people would think as inconsistant.

since you don't know what python does, why don't
you just shut up?




From tim.one at home.com  Sun Mar 18 00:55:22 2001
From: tim.one at home.com (Tim Peters)
Date: Sun, 18 Mar 2001 00:55:22 -0500
Subject: Meta-class inheritance problem (Python2.0 bug?)
In-Reply-To: <VnWs6.17199$3x.1641111@typhoon.ne.mediaone.net>
Message-ID: <LNBBLJKPBEHFEDALKOLCOEGGJGAA.tim.one@home.com>

[posted & mailed]

[David Abrahams, on Lorien Dunn's boost + metaclass mystery]

Note that Lorien opened a bug report on this (and thank you!):

http://sourceforge.net/tracker/index.php?func=detail&
    aid=409355&group_id=5470&atid=105470

> Just to be absolutely clear, I never said it was a bug in Python.

And we never said it wasn't <wink>:  let's fight about it in comments
attached to the bug report.  Say anything interesting on c.l.py wrt this
instead, and it will vanish most likely forgotten.  Ditto for email.

sourceforge-has-a-memory-because-we-don't-ly y'rs  - tim




From pearu at cens.ioc.ee  Wed Mar 21 01:57:00 2001
From: pearu at cens.ioc.ee (Pearu Peterson)
Date: Wed, 21 Mar 2001 08:57:00 +0200
Subject: Base 2 to long integers and back
In-Reply-To: <mailman.985145066.15092.python-list@python.org>
References: <mailman.985145066.15092.python-list@python.org>
Message-ID: <Pine.LNX.4.21.0103210851330.30869-100000@kev.ioc.ee>

On Wed, 21 Mar 2001, Tim CHURCHES wrote:

> Can anyone suggest a fast method of converting base2 strings
> (e.g. '0000101110') to Python long integers, and back? I couldn't locate
> any functions in the standard library to do this, but no doubt I am
> overlooking something obvious.

Try GMPY:
	http://gmpy.sourceforge.net/

(though it is not in the standard library, it is extremely fast)

Here is an example session:
>>> from gmpy import mpz,digits
>>> mpz('0000101110',2) 
mpz(46)
>>> digits(46,2)
'101110'

Regards,
	Pearu



From dwig at advancedmp.net  Tue Mar 20 16:42:46 2001
From: dwig at advancedmp.net (Don Dwiggins)
Date: 20 Mar 2001 13:42:46 -0800
Subject: windows script host and python
In-Reply-To: "Alex Martelli"'s message of "Tue, 20 Mar 2001 00:15:21 +0100"
References: <%tKr6.2570$pX.26174@news1.onlynews.com>
	<98oq3r$1e0$1@troll.powertech.no> <8Aps6.104$pl.1515@newsfeed.slurp.net>
	<98tc4d0rmk@news1.newsguy.com> <lvf8bt4p6e6naq0vimgnkhs175eoqbajfr@4ax.com>
	<mailman.985028069.29930.python-list@python.org>
	<9963te04s9@news2.newsguy.com>
Message-ID: <uwv9k6rs9.fsf@advancedmp.net>

Alex Martelli writes:
>> Actually, it fails on my NT workstation also:
>     [snip]
>> File "e:\python20\win32com\client\__init__.py", line 87, in Moniker
>> moniker, i, bindCtx = pythoncom.MkParseDisplayName(Pathname)
>> pywintypes.com_error: (-2147221020, 'Invalid syntax', None, None)

> ...but do you have WMI installed?  AFAIK, it doesn't come as default
> with either 98 or NT (perhaps 2000 and/or ME -- not sure) -- it has
> to be downloaded from the MS site and installed.

You're right, I don't (and don't really know much about it) -- just out of
curiosity, I snipped the code from the message and tried it out.  Oh well,
post in haste, repent at leisure...

-- 
Don Dwiggins, Beta Version 0.9 -- Release 1.0 due out Real Soon Now
Advanced MP Technology
dwig at advancedmp.net




From LLewin at oreilly.com  Wed Mar 21 17:55:10 2001
From: LLewin at oreilly.com (LLewin at oreilly.com)
Date: 21 Mar 2001 22:55:10 GMT
Subject: learning python...
References: <3AB9182A.8FEBCF3E@noaa.gov>
Message-ID: <99bbge21ni0@news1.newsguy.com>

Hi,
The 2nd edition of Programming Python is in fact significantly different (80% I would say).  Mark talks about the differences between editions in his Preface (which is why folks that have read the new one, and not the old one, know about how the book has evolved).  If you go to the link Aahz has provided for the book on our site, you can check out the table of contents.  If you compare it to the 1st edition's, you'll see that the approach is new.  The book moves quickly into tools and applications.  
Laura
LLewin at oreilly.com

> It would be good to know if the new edition really is all that different (other than
> the topics added) to the 1st ed.  The only review on Amazon says he never read it...
> 
> Ben
> 
> Aahz Maruch wrote:
> 
> > In article <3ab90559_6 at goliath.newsfeeds.com>,
> > Bubba Brains <chocolate at toblerone.yum> wrote:
> > >
> > >Forget Learning Python?  I think it's the best reference/tutorial around.  I
> > >use python once every couple months, and sometimes forget the simple python
> > >syntax.  The Learning Python book is great for getting up to speed fast, and
> > >to look up common stuff.  The other OReilly book (Programming Python) is
> > >absolutely annoying to use.  It's not even a good reference book (good luck
> > >finding the info you want quickly).  Don't know about the Prentice book.
> >
> > If you look at http://www.oreilly.com/catalog/python2/ you'll see that
> > Programming Python just got a second edition; I've heard that there were
> > a lot of improvements.
> > --
> >                       --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)
> >
> > Androgynous poly kinky vanilla queer het Pythonista   http://www.rahul.net/aahz/
> > Hugs and backrubs -- I break Rule 6
> >
> > "I won't accept a model of the universe in which free will, omniscient
> > gods, and atheism are simultaneously true."  -- M
> 


==================================
Posted via http://nodevice.com
Linux Programmer's Site


From lluang at northwestern.edu  Sat Mar 31 20:14:22 2001
From: lluang at northwestern.edu (Louis Luangkesorn)
Date: Sat, 31 Mar 2001 19:14:22 -0600
Subject: Know of Substantial Apps Written in Python?
References: <3ac4e4a8_1@news4.newsfeeds.com> <KE6x6.415$SR6.235983@typhoon1.ba-dsg.net> <3ac4ff12$1_2@news4.newsfeeds.com>
Message-ID: <3AC680EE.997108AB@northwestern.edu>

Why don't you look at the Python Conference Proceedings pages off of the
www.python.org?

I'm not sure why the prejudice against using a language to drive C (or
C++ or Fortran) modules.  Paul Dubois at LLNL and others note that using
Python to drive C modules changes the way that these modules are used,
since you can code the functionality in C then use it to do real work
via Python.  It seems similar to the way the financial community will
often code algorithms C, then run them as DLL's from a spreadsheet such
as Excel or 1-2-3.  If using Python instead of directly working in C or
Fortran changes the way a module is used, wouldn't this give you insight
into the character of the language?

A few applications that have not been mentioned yet in this thread are
Reportlab, climate analysis, scientific simulations, driving astronomy
software, bioinfomatics, all of these are found in various proceedings
of the Python and Open Source conferences.  Reportlab is downloadable,
some of the others probably need something more than just your average
PC though :-)

[you can tell what I use Python for by my choice of examples :-) ]

Louis

Dry Ice wrote:

> > > Know of substantial apps written in Python?
>
> [...]
>
> > What's the definition of "substantial"?
>
> Good question.  To sharpen the point of my inquiry:
>
> As programs grow in size and complexity, both
> programmers and the results begin to show fatigue.
> Slick as Java is, for example, I have found the
> real-world results of which I'm aware so far to
> be troubling in the Windows environment.  Still
> evaluating, but it does not look good.
>
> What "substantial" is might be defined in terms
> of project length.  Perhaps something which would
> take a good programmer several months.
>
> Something as complex as a web browser might do,
> if the language in question does not merely act as
> glue for modules written in C or whatever.
>
> In the end, this is like evaluating people.  We can
> listen to what others say about a person, or what a
> person says about his or her self- or we can simply
> see how they BEHAVE.

--
K Louis Luangkesorn
lluang at northwestern.edu  http://pubweb.nwu.edu/~kll560  PGP:0xF3E2D362
Whatsoever things are true, ... honest, ... just, ... pure, ... lovely,
... of good report; if there be any virtue, and if there be any praise,
think on these things.- motto - Northwestern University




From py.list at localhost.localdomain  Sun Mar 25 02:18:56 2001
From: py.list at localhost.localdomain (py.list at localhost.localdomain)
Date: Sun, 25 Mar 2001 16:48:56 +0930 (CST)
Subject: time to invest ???
In-Reply-To: <3abd3a17.7103358@news.vif.com>
Message-ID: <Pine.LNX.4.10.10103251638220.23294-100000@localhost.localdomain>

I've started hanging around the edges of Python and this list after
spending about 12 months learning pretty much everything practical there
is to know about PHP and MySQL (and various other bits and pieces
including shell scripting and some Javascript, etc, etc) ... my experience
is just identify a task you want to get done and then hunt down examples,
models, something close and just hack it til it works. Approaching
programming from a point of view where you want to learn/understand
everything before you use it will prove to be an overwhelming and even
impossible task.

For example, I'm looking at Python now to do non-web jobs that PHP isn't
so great for, and which shell scripts can't quite handle. Perl is the
other option, but I tend to believe what I read here about Perl ...
Basically, I'd like to build a PHP code manager in Python which will
automate a great deal of the hack work in setting up a web application.

So anyway, I reckon you learn to think like a programmer by doing
programming ... the more you do it, the quicker it happens I suppose.

Mick


> Hello, i start to learn Python some day ago, with lot of knowledge as
> a system admin but little as programmer and i wonder what is the
> average time/hours someone have to invest to be able to use a language
> like that easily.  i talk as well about devlopping the way of thinking
> of a programmer and the knowledge of the language. please don't tell
> me theres no average, at least take the idea of a normal guy, not a
> genius and not stupid  ;-)




From hyakugei at nospam.theorganization.net  Sat Mar  3 18:00:23 2001
From: hyakugei at nospam.theorganization.net (Digital Hyakugei)
Date: Sat, 03 Mar 2001 23:00:23 GMT
Subject: Thanks, will check it out (WDDX in Python)
References: <3a9faea1$1@news1> <j47l28jgt1.fsf@informatik.hu-berlin.de>
Message-ID: <bIeo6.332916$f36.11764485@news20.bellglobal.com>

Just downloaded it from sourceforge... Thanks again!
jos


"Martin von Loewis" <loewis at informatik.hu-berlin.de> wrote in message
news:j47l28jgt1.fsf at informatik.hu-berlin.de...
> "Digital Hyakugei" <jos at trapeze.removespam.com> writes:
>
> > Hello again. I've scoured the python.org site, used Google and have only
had
> > hints that there _might_ be a WDDX parser/encoder for Python... If
anyone
> > has a more concrete example, source code, link, anything, i'd really
> > appreciate it!
>
> PyXML 0.6.4 (and earlier) has a wddx library, see pyxml.sourceforge.net.
>
> Regards,
> Martin




From tuttledon at hotmail.com  Wed Mar  7 22:21:17 2001
From: tuttledon at hotmail.com (Don Tuttle)
Date: Thu, 08 Mar 2001 03:21:17 GMT
Subject: News from Python9!!!
Message-ID: <NUCp6.65206$Z8.11697516@typhoon.southeast.rr.com>

For those of you, like myself, who are NOT at the Python bash in Long Beach,
some very good news from there  has been posted at ActiveState.

1) The "Python Software Foundation" has been created
http://www.activestate.com/Corporate/Communications/Releases/Press983863928.
html

2)The Programmers' Choice for Python award winner is Christian Tismer
http://www.activestate.com/Corporate/Communications/Releases/Press983917761.
html

3) A "Python Cookbook" project has started.
http://www.activestate.com/Corporate/Communications/Releases/Press983583358.
html

4)And a bugfix release of Python 2.0 has been released. It includes PPM
(Programmer's Package Manager ) to make finding and installing Python
modules simple.
http://www.activestate.com/Corporate/Communications/Releases/Press983916629.
html

Don




From alex at shindich.com  Sun Mar  4 16:46:26 2001
From: alex at shindich.com (Alex Shindich)
Date: Sun, 4 Mar 2001 13:46:26 -0800
Subject: shortcut for 'self' wish
References: <3aa050ad.35416419@News.CIS.DFN.DE> <97pnpf$99$1@news.udel.edu> <ot11atghgm5cccc9oc8tm2pq453c1n8ukh@4ax.com>
Message-ID: <XHyo6.151$gc4.71724@news.pacbell.net>

"Kirby Urner" <urner at alumni.princeton.edu> wrote in message
news:ot11atghgm5cccc9oc8tm2pq453c1n8ukh at 4ax.com...
> "Terry Reedy" <tjreedy at udel.edu> wrote:
> >
> >This is entirely a social convention for publicly displayed code.  For my
> >own work, I use s for self (and o for other in __add__(), etc.).  Python
is
> >not a straightjacket.
> >
>
> I agree with Pilgrim in 'Dive Into Python' that we should
> honor the 'self' convention.  These standards make source
> more readable across the board and innovating some local
> idiosyncratic convention just to save keystrokes (or
> whatever) strikes me as self-indulgent.
>
> Kirby
I completely agree! The "self" convention aids readability. In fact, I would
not mind if Python were to enforce that the first parameter is always called
self. But then again, I have been known for my extremist views.. ;)




From sdm7g at Virginia.EDU  Mon Mar 26 16:14:36 2001
From: sdm7g at Virginia.EDU (Steven D. Majewski)
Date: Mon, 26 Mar 2001 16:14:36 -0500 (EST)
Subject: python, OS X, and fcntl
In-Reply-To: <mthf0g719g.fsf@astron.berkeley.edu>
Message-ID: <Pine.NXT.4.21.0103261603470.345-100000@localhost.virginia.edu>


On 26 Mar 2001, Johann Hibschman wrote:

> Hi folks,
> 
> I'm having trouble with the fcntl module under Mac OS X.  It
> sounds like this is an old problem that has resurfaced, since I can
> only find references to it in 1999 archives about OS X Server.
> Is there a known workaround?
> 
> I get, while running test_fcntl:
> 
> Status from fnctl with O_NONBLOCK:  0
> struct.pack:  '\x00\x03\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00'
> Traceback (most recent call last):
>   File "test_fcntl.py", line 31, in ?
>     rv = fcntl.fcntl(f.fileno(), FCNTL.F_SETLKW, lockdata)
> 
> Any hints?
> 

Somebody mentioned this in one of the mailing lists.
This was working under public beta. 

The problem line (in test/test_fcntl.py) is:

if sys.platform in ('netbsd1', 'Darwin1.2',
                    'freebsd2', 'freebsd3', 'freebsd4', 'freebsd5',
                    'bsdos2', 'bsdos3', 'bsdos4',
                    'openbsd', 'openbsd2'):


The final release (and the early developer release candidate) are
bumped up to 1.3:

[localhost:~] sdm7g% uname -a
Darwin localhost 1.3 Darwin Kernel Version 1.3: Thu Mar  1 06:56:40 PST
2001; root:xnu/xnu-123.5.obj~1/RELEASE_PPC  Power Macintosh powerpc


*AND* something else has changed in Python 2.1 beta in parsing the 
system info in sys.platform: 

Python 2.1b2 (#1, 03/23/01, 20:07:45) 
[GCC Apple DevKit-based CPP 6.0alpha] on darwin1
Type "copyright", "credits" or "license" for more information.
>>> import sys  
>>> sys.platform
'darwin1'


So, the wrong branch gets taken and test_fcntl uses the wrong
format for the lock data. 


My guess is that we *don't* want the minor number in the sys.platform
string (the current behaviour), because it will break again on a new
release, so the test in test_fcntl.py ought to be changed to 'darwin1'.
 

-- Steve Majewski




From mrq at for.mat.bham.ac.uk  Thu Mar 29 07:03:15 2001
From: mrq at for.mat.bham.ac.uk (Martyn Quick)
Date: Thu, 29 Mar 2001 13:03:15 +0100
Subject: BLT extensions with BeOpen Python 2.0
Message-ID: <Pine.GSO.4.21.0103291300160.11408-100000@babbage>

I've been trying to install the BLT extensions for Tk into my BeOpen
distribution of Python 2.0.  However, it seems that the directory
structure that the BLT distribution expects (see www.tcltk.com/blt) is
different to that provided by the python distribution.

Has anyone succeeded in installing BLT with the distribution, or do I need
to switch distributions?  (and to which one?)

Thanks,

Martyn

--------------------------------------------------------
Dr. Martyn Quick  (Research Fellow in Pure Mathematics)
University of Birmingham, Edgbaston, Birmingham, UK.
http://www.mat.bham.ac.uk/M.R.Quick



From bos at hack.org  Tue Mar 27 10:35:37 2001
From: bos at hack.org (Rikard Bosnjakovic)
Date: Tue, 27 Mar 2001 17:35:37 +0200
Subject: Eat this one
Message-ID: <3AC0B349.EE0AA9B2@hack.org>

I feel ashamed to show the code, but I didn't find any other solution :)

txt = list(cgi.escape(open("somefile").read()))
for i in range(len(txt)):
    try:
        txt.remove(" ")
    except:
        try:
            txt.remove("\n")
        except:
            try:
                txt.remove("<")
            except:
                try:
                    txt.remove(">")
                except:
                    try:
                        txt.remove("&")
                    except:
                        pass
                    pass
                pass
            pass
        pass



What way would be better and _faster_ to wipe out the five chars from
the list?

-- 
Rikard Bosnjakovic - http://a214.ryd.student.liu.se/cv/ - ICQ: 1158217

Anyone sending unwanted advertising e-mail to my address will be
charged $250 for network traffic and computing time. By extracting my
address from this message or its header, you agree to these terms.


From scarblac at pino.selwerd.nl  Tue Mar 27 05:09:23 2001
From: scarblac at pino.selwerd.nl (Remco Gerlich)
Date: 27 Mar 2001 10:09:23 GMT
Subject: unique items in lists
References: <_nZv6.2878$yU4.468236@news.infostrada.it>
Message-ID: <slrn9c0ph1.o5h.scarblac@pino.selwerd.nl>

comp.lang.python <micampe at f2s.com> wrote in comp.lang.python:
> Which is the best (read: fastest) way to remove duplicated items from a list
> of strings?

The standard idiom is to put the items in a dictionary, then get the
dictionary keys:

def unique_strings(l):
   dict = {}
   for s in l:
      dict[s] = 1
   return dict.keys()

I think this is also very fast, at least lots faster than comparing all the
strings in a double for loop or something.

-- 
Remco Gerlich


From jmarshal at mathworks.com  Thu Mar  8 19:32:21 2001
From: jmarshal at mathworks.com (Joshua Marshall)
Date: 9 Mar 2001 00:32:21 GMT
Subject: class fields vs __getattr__
Message-ID: <9898al$j3j$1@news.mathworks.com>

If you try to get a field of an object, it seems __getattr__ is called
only if no such field exists on the class.  For exmaple:

  >>> class T:
  ...   x = 1
  ...   def __getattr__(self, name):
  ...     return 2
  ... 
  >>> ob = T() 
  >>> ob.foo
  2
  >>> ob.x
  1

I would have expected ob.x to return 2.  Can anyone explain the
rationale behind this behavior?


From timr at probo.com  Fri Mar  2 02:27:14 2001
From: timr at probo.com (Tim Roberts)
Date: Thu, 01 Mar 2001 23:27:14 -0800
Subject: "import X" and "from X import y" produce different results, Why?
References: <ifmn6.6635$hn5.857862@news1.rdc1.mb.home.com>
Message-ID: <l4iu9t06fjd6bsdlih2n7aiuabkqkt1i10@4ax.com>

"Parzival Herzog" <parz at home.com> wrote:

>In the Essential Python Reference, it states (p. 72):
>
>"The from statement is identical to import, e3xcept that instead of creating
>a name referring to the newly created module namespace, references to one or
>more of the objects defined in the module are placed into the current
>namespace."
>
>So, apparently the imported X is a copy of xx.X, not the same object by
>another name.
>Can anyone explain this result? Is this result justified by a Python
>language specifcation?

Alex gave an excellent and detailed description of the issue here, but I
wonder if I could volunteer the mental model that helps me grasp this
concept.

>----------
># xx.py
>X = 35
>----------
># yy.py
>from xx import X
>import xx

At this point, there exists an object with the value "35".  The names
"xx.X" and "X" are both bound to that object.

>print "BEFORE, X =", X
>print "BEFORE, xx.X =", xx.X
>X = 99

At this point, there exists an object with the value "35".  xx.X is still
bound to that object.  There is also a NEW object with the value "99".  The
name "X" is bound to that object.

Assigning 99 to X did not change the value of the object X was bound to.
Rather, it created a new object and bound X to it.

>print "AFTER, X =", X
>print "AFTER, xx.X =", xx.X
>----------
--
- Tim Roberts, timr at probo.com
  Providenza & Boekelheide, Inc.


From matthias at rice.edu  Mon Mar 19 10:45:36 2001
From: matthias at rice.edu (Matthias Felleisen)
Date: Mon, 19 Mar 2001 09:45:36 -0600
Subject: PLT is moving
Message-ID: <3AB629A0.75234DE5@rice.edu>

Dear friends and colleagues in the programming languages
 and functional programming community --

The Rice portion of the Programming Languages Team (PLT) is relocating to

 NORTHEASTERN UNIVERSITY (BOSTON)

where I have accepted a new position.

I will form a programming languages and software systems lab with
Will Clinger, Karl Lieberherr, David Lorenz, Paul Steckler, Mitch Wand.
Together, we will create a new PhD concentration in this research area with
a full spectrum of courses and research topics, including

 programming language design and types
 compilers and run-time environments
 programming environments, program analyses, software tools
 OO design, analysis; aspect-oriented program design
 software systems in high-level languages

Please re-direct promising PhD applicants to the new lab in Boston.

The move will happen over the summer 2001. DrScheme and friends will move
with us though the Web will make this move (mostly) transparent.

Regards -- Matthias Felleisen




From mix77 at usa.net  Tue Mar 27 09:31:06 2001
From: mix77 at usa.net (Mix)
Date: Tue, 27 Mar 2001 16:31:06 +0200
Subject: Combining Lists
Message-ID: <99q887$hl9$1@ctb-nnrp2.saix.net>

How does one take two or more lists and combined them?
For example, suppose one has the following 2 lists :
 list1=['1','2']
  and
 list2=['3','9,'15']

Now how  does one get the following list
 list3=['1','2','3','9','15']
from a combing the two lists (i.e list1 and list2)?

Thanks.
Mix




From im_cne at hotmail.com  Mon Mar  5 09:27:33 2001
From: im_cne at hotmail.com (tech)
Date: Mon, 5 Mar 2001 08:27:33 -0600
Subject: iis and python
Message-ID: <F5622749FED6DA5D.AF4594816A4865C9.16CD98F1C1FAC3F6@lp.airnews.net>

I am going to do a project building a web front end  while my friend is
going to build the back end.

The backend will be MS SQL. The front end will be a web interface, IIS 5.0
or whatever comes with win2k .

I have a choice of doing php or python.

In order to use python it would have to entail cgi or ASP. Is this correct.
Or is there an alternative. I dont know if Zope is the right thing. I hate
DTML.

Is using python in ASP a pain in the but? Am I going to have to learn a
whole technology just to implelment python well into it?

I hear troubleshooting ASP pages is not fun.

The only time I have used iis is with php and it was a total snap. Painless.




From martin.fraeulin at t-online.de  Tue Mar 27 06:47:15 2001
From: martin.fraeulin at t-online.de (Martin =?iso-8859-1?Q?Fr=E4ulin?=)
Date: Tue, 27 Mar 2001 13:47:15 +0200
Subject: sitemap and ftp
References: <3AC04D0B.5A44A4D5@aon.at>
Message-ID: <3AC07DC3.28190071@t-online.de>

Hi Oliver,

try the htmllib and the urllib module. You could use the first one to
parse your HTML files. The second one is for retrieving files from
URLs. To work with URLs it is also helpful to use urlparse module.
All these modules and their methods are described in the Python
documentation.

Once, I tried to write a program that grabs whole sites from URL to
the local disk but I never finished it.

Erich Seifert


Oliver Vecernik wrote:
> 
> Hi,
> 
> I'm relatively new to Python, but I felt immediately in love because of
> it's clear structure and ease to read the code. Unfortenately I still
> lack a lot of knowledge. Maybe some can help me or point me to the right
> spot in the docs.
> 
> I'd like to parse through my website tree and generate an automated
> sitemap from the <title> tags. Is there a module that helps me achieving
> this task?
> 
> Furthermore I'd like to copy all *.html files and all *.jpg and *.gif
> files to my ISP preserving the directory structure. Before copying the
> *.html files have to be filtered (adding a header and a footer). Has
> anybody written such routines already?
> 
> Regards,
> Oliver


From sheila at spamcop.net  Sun Mar 18 01:01:22 2001
From: sheila at spamcop.net (Sheila King)
Date: Sun, 18 Mar 2001 06:01:22 GMT
Subject: Pep 238 Why new // operator (Non-integer Division)
References: <98tc2s$qir$1@news2.ottawa.cyberus.ca> <mailman.984822021.13503.python-list@python.org> <vag7bt09r54tv05nohtssd0ge891v3fa02@4ax.com> <3AB42890.AD9E9FF0@wwc.com> <mqi8bt8q6oadhucrot1fcdp38782di0dao@4ax.com>
Message-ID: <boj8bt0js4u7qa74hrrap4sdmqkvk71t6l@4ax.com>

On Sun, 18 Mar 2001 05:45:16 GMT, Sheila King <sheila at spamcop.net> wrote in
comp.lang.python in article <mqi8bt8q6oadhucrot1fcdp38782di0dao at 4ax.com>:

:
:Um, in base pi, wouldn't pi = 1   ???

Oh, crud, never mind. You're right. pi = 10. Duh.

--
Sheila King
http://www.thinkspot.net/sheila/
http://www.k12groups.org/



From sheila at spamcop.net  Sun Mar 25 12:08:28 2001
From: sheila at spamcop.net (Sheila King)
Date: Sun, 25 Mar 2001 17:08:28 GMT
Subject: Comparing efficiency in sending mail
References: <3A7FA5BA.82FB7593@esec.com.au> <mailman.981444758.4937.python-list@python.org>
Message-ID: <dm8sbt84f8jrvcu085a8pt8hvhdfvvlv9k@4ax.com>

Some time back, in response to a request about how to send mail using Python,
the following article was posted:

On Mon, 5 Feb 2001 23:24:35 -0800 (PST), <chris at onca.catsden.net> wrote in
comp.lang.python in article <mailman.981444758.4937.python-list at python.org>:

:import os
:
:f = os.popen ( "/usr/sbin/sendmail", "w" )
:
:f.write ( """\
:From: your address
:To: recipient address
:Subject: spam
:
:spam spam wonderful spam spam spam
:""" )
:
:
:If you dont have sendmail on your system, or some other MTA, then you can
:use smtplib to send it directly to an off-site MTA. I can get you some
:simple code for that, but the above is the preferred way.

My question is this:
Why is using os.popen and calling sendmail preferred over using the smtplib
module? Is it less overhead for the server? If so, why?

I have found the smtplib module very easy to use and understand. I was
telnetting into my webhost just now, and experimenting with some of the advice
given above, but first of all, exactly as typed above didn't work on my host.
I had to add the -t switch to the call to sendmail, like this:

f = os.popen("/usr/lib/sendmail -t", "w")

I know that the -t switch tells sendmail to get the recipient for the SMTP
envelop from the message headers. I've been experimenting over how to have
more control over the SMTP envelope. I was looking at some sendmail docs here:
http://www-wsg.cso.uiuc.edu/sendmail/tutorial/run/commandline.html
and tried using the -bs switch at a command line prompt, but had trouble with
"bare line feeds", since my host is really using qmail, and just has a
sendmail wrapper around qmail.

Anyhow, that last part is really off-topic here, as it is an issue with how to
use sendmail/qmail. I just wanted to point out, that using the smtplib module
is easier and gives me more control over the messages I send, and therefore
I'm curious as to why I would want to use popen to send to sendmail instead.

--
Sheila King
http://www.thinkspot.net/sheila/
http://www.k12groups.org/





From mwh21 at cam.ac.uk  Sun Mar 11 10:56:08 2001
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 11 Mar 2001 15:56:08 +0000
Subject: Static typing (was Re: Java guy interested in Python)
References: <mailman.984160454.14131.python-list@python.org>
Message-ID: <m3r9048fl3.fsf@atrus.jesus.cam.ac.uk>

=?iso-8859-1?q?Hamish=20Lawson?= <hamish_lawson at yahoo.co.uk> writes:

> Michael Hudson:
> 
>     I suspect that you'd basically end up listing the methods required,
>     eg:
> 
>     def func(file : <has 'write', 'read', 'seek'>, ...): pass
> 
>     I'm not sure that makes a "type system".  Hmm.  More thinking
>     required.
> 
> Not sure that makes a type system? Isn't that what the modern
> understanding of a type is - a list of operations that are expected to
> be supported, as with Java's idea of interfaces being distinct from
> classes? 

Well, what I really meant was, I'm not sure that going down this road
would gain anyone anything.  I agree it might; but I certainly don't
think it's a given.

> I actually thought your example showed an interesting idea -
> anonymous, defined-on-the-fly interfaces. (That is, if static typing
> is wanted.)

Of course, a problem is that an object having attributes called
"write", "read" and "seek" is one thing, but these attributes in fact
being methods that interact in the desired manner is something else.

Cheers,
M.

-- 
 Very clever implementation techniques are required to implement this
 insanity correctly and usefully, not to mention that code written
 with this feature used and abused east and west is exceptionally
 exciting to debug.       -- Erik Naggum on Algol-style "call-by-name"


From cfelling at iae.nl  Thu Mar 29 09:18:48 2001
From: cfelling at iae.nl (Carel Fellinger)
Date: 29 Mar 2001 16:18:48 +0200
Subject: Defining class files
References: <3ac33343$0$12248$ed9e5944@reading.news.pipex.net>
Message-ID: <99vg88$5eq$1@animus.fel.iae.nl>

Neil Benn <neil.benn at cambridgeantibody.com> wrote:
> Hello,

>             Sorry to ask such a basic question but -

> I'm wokring through the tutorial and have reached the section on classes in
> Python.  Working through the example on classes I have entered the class
> using the interpreter:-
...
>     I then imported the class, tried to assign the class and ivoke a method

You didn't import the class, you imported the module:)

Modules, classes, packages are python's way to group things.

A module is a file which name ends with ".py".  A module contains
Python statements, e.g. a class statement or an assignment statement.

A package is a directory with a possible empty file named "__init__.py".
Any sensible package contains modules.

The import statement basicly looks for a file with the extension ".py"
[[if there are dots in the import statement like in "import A.B", then
it first looks for a package named "A", and inside that package it
looks for a module file named "B.py"]], it then loads that file,
compiles it and provide access to the goodies of that module through
an module-object.  That module object is normally bound to a var with
the name of that module (without the ".py" ofcourse).

e.g.:

Module file A:

var = 'I am a var'

def f():
   print 'spam'

def class C:
    def f(self):
        print 'more spam'

Some other module, or just from the interactive prompt:

>>> import A
>>> print A.var
I am a var
>>> A.f()
spam
>>> c = A.C()
>>> c.f()
more spam

this will look for a file named "A.py" in your local directory, and if
it can't be found there, iit will search for it in some python specific
directories.  After the impost all goodies are accessible by prefixing
them with "A."

You could also do:

>>> from A import f, C
>>> print var
I am a var
>>> f()
spam
>>> c = C()
>>> c.f()
more spam


Pittfalls to look out for:

In python an assignment is the binding of an object (the right-hand
side) to a name (the left-hand side).  One of the implications of this
is that after a "from import var" var only refers to the value of
A.var at the time of the import. Assigning to A.var or to var won't
change the value of the other.

>>> import A
>>> from A import var
>>> print A.var, var
I am a var I am a var

>>> var = "me too"
>>> print A.var, var
I am a var me too

>>> A.var = "all's swell"
>>> print A.var, var
all's swell me too
-- 
groetjes, carel


From h-karls at telia.com  Sun Mar 25 16:22:25 2001
From: h-karls at telia.com (Hakan Karlsson)
Date: Sun, 25 Mar 2001 21:22:25 GMT
Subject: Good GUI for Python
References: <mailman.985314033.5120.python-list@python.org>
Message-ID: <lktv6.1599$sk3.549470@newsb.telia.net>

> First, ActiveStates' Python tool cannot support Tkinter and
> I confirmed with their stuff. So Tkinter maybe good, but not
> compatible with PythonWin.

> Is there any suggestions ? Change my develop environment
> or is there another good gui for me ??

If you want to run/tryout different GUI-toolkits then PythonWorks
may be something for you. As user programs run using a run environment
in a separate process you may set it up for any available toolkit. Besides
that a GUI-editor for Tkinter and for the Tkinter extension uiToolkit is
also
included.

/Hakan





From brian at sweetapp.com  Thu Mar 15 02:29:43 2001
From: brian at sweetapp.com (Brian Quinlan)
Date: Wed, 14 Mar 2001 23:29:43 -0800
Subject: [Q] Bug in Python?
In-Reply-To: <t6Zr6.92712$__6.18256422@typhoon.southeast.rr.com>
Message-ID: <HPEJKHHCMJHPMNLDAPLMIEDLCEAA.brian@sweetapp.com>

Sorry, but you missed a pretty subtle Pythonism :-)

When Python examines the code block for the function 'a', it sees that 'var'
is being assigned a value without having been declared global. It therefore
considers it to be a local variable and then throws an exception when you
print 'var' because you are printing a variable that has not (yet) been
assigned a value. The relevant part of the python language reference
follows:

"The global statement is a declaration which holds for the entire current
code block. It means that the listed identifiers are to be interpreted as
globals. While using global names is automatic if they are not defined in
the local scope, assigning to global names would be impossible without
global."

So, you might want to change your code to:

############# test.py ###############
var = None

def a():
    global var
    print var # No exception anymore
    if var != None:
        var = None  # >> pass isn't useful anymore

def b():
     print var
     a()

b()
#################################

Cheers,
Brian

-----Original Message-----
From: python-list-admin at python.org
[mailto:python-list-admin at python.org]On Behalf Of Daehyok Shin
Sent: Wednesday, March 14, 2001 10:16 PM
To: python-list at python.org
Subject: [Q] Bug in Python?


Recently, I met a strange behavior of Python 6.1.
When I tried the following module,
I got an error message in "print var":
    UnboundLocalError: var

But, when I changed "var=None" to "pass",
the error message was gone and I got an expected result:--printing None and
None.
What happens here? Is this a bug in Python or I missed something?

############# test.py ###############
var = None

def a():
     print var #UnboundLocalError
     if var != None:
          var = None  # >> pass

def b():
     print var
     a()

b()
#################################

--
Daehyok Shin (Peter)
Terrestrial Hydrological Ecosystem Modelers
Geography Department
UNC-CH


--
http://mail.python.org/mailman/listinfo/python-list




From vsbabu at erols.com  Fri Mar 23 19:18:02 2001
From: vsbabu at erols.com (Satheesh Babu)
Date: Fri, 23 Mar 2001 19:18:02 -0500
Subject: Pretty-printing ugly blocks of SQL?
References: <mailman.985274369.22340.python-list@python.org>
Message-ID: <99gp1c$ld5$1@bob.news.rcn.net>

I had made one long time ago, in lex and C. I'm and was a LEX newbie - so
you can try this with caution :-)

http://vsbabu.csoft.net/software/lsqlb.html

--
v.s.babu
vsbabu at erols.com
http://vsbabu.csoft.net
"Steve Purcell" <stephen_purcell at yahoo.com> wrote in message
news:mailman.985274369.22340.python-list at python.org...
> Does anyone have some code for parsing an ugly block of SQL and printing
it
> out in a more easily readable form?
>
> -Steve
>
> --
> Steve Purcell, Pythangelist
> Get testing at http://pyunit.sourceforge.net/
> Any opinions expressed herein are my own and not necessarily those of
Yahoo
>




From ws-news at gmx.at  Fri Mar  9 05:49:00 2001
From: ws-news at gmx.at (Werner Schiendl)
Date: Fri, 9 Mar 2001 11:49:00 +0100
Subject: learning C
References: <mailman.983972829.23481.python-list@python.org> <983975751.18833@newsmaster-04.atnet.at> <3AA82422.53026D3D@cosc.canterbury.ac.nz> <20010308221519.A4367@harmony.cs.rit.edu> <mailman.984113052.19927.python-list@python.org>
Message-ID: <984135226.858948@newsmaster-04.atnet.at>

> I have written more lines of C than I care to remember.  C is for CONTROL.
> Nowadays I use what some believe is the best C++ compiler in the world,
> GCC, and I still just write C.
>
> Frankly, if C++ works for Werner that's his business, but I feel he has
> effectively called the C lovers on this list stupid.  Rudeness is for that
> other comp.lang.p* list.

This was not at all intended.

If you read it that way, I apologize for that.

>
> IMHO being told that your favorite compiled language should be deprecated
> is rude.  It's not like it's InterCal, after all.  I don't like assembler
> but I would never say it should be deprecated; I positively hate LISP but
> you LISP fans shouldn't take that personally.

The compiler vendors already did it on a lot of platforms.
C++ will swallow C, because C is nearly a subset of C++.

If you have a full cupboard of tools installed, what sense does it make to
ignore 90% of it??
Why not have a look at all of it and use what's suited for the application
best?


regards
werner




From loewis at informatik.hu-berlin.de  Thu Mar  8 05:29:03 2001
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 08 Mar 2001 11:29:03 +0100
Subject: Unicode getting in the way...
References: <RDEp6.199688$v.18718576@typhoon.nyroc.rr.com>
Message-ID: <j4snkomu4w.fsf@informatik.hu-berlin.de>

"Dan Maas" <dmaas at dcine.com> writes:

> At some random point in my code, foo changes from being a
> 'string' object to a  'unicode' object. Subsequent attempts
> to manipulate foo fail with a UnicodeError (ASCII decoding
> error, ordinal not in range(128)).
> 
> How can I stop this from happening? I just need raw 8-bit
> strings everywhere!

Apparently, you add a Unicode object to the string at some point. You
need to find the place where that happens, and convert the Unicode
object into a byte string. Python cannot do that for you, since you
need to specify what encoding you want to use.

Regards,
Martin



From pj at sgi.com  Fri Mar  2 23:52:32 2001
From: pj at sgi.com (Paul Jackson)
Date: 3 Mar 2001 04:52:32 GMT
Subject: Thinking Pythonically (was Re: gah! I hate the new string syntax)
References: <mailman.983485923.1136.python-list@python.org> <97opsb$nhu$1@nntp6.u.washington.edu> <97oupq$qijrh$1@ID-77331.news.dfncis.de> <97pcu2$n6n$1@animus.fel.iae.nl>
Message-ID: <97ptag$81s65$1@fido.engr.sgi.com>

Sorry - couldn't resist the following ...

Carel wrote:
> $ cleartool lock `cleartool lsvob -l | grep ^Tag: data | cut -c 6-`

I think the 'data' (2nd grep arg) should be removed.

And (groan) yet another way to write this would be:

    cleartool lsvob -l | awk '$1 == "Tag:" { print $2; }' | xargs cleartool lock

This way avoids assumptions of the exact number of spaces in
the lsvob output, and works if more vob's need locking than fit
in the system's MAXARGS (God help you if you have a ClearCase
installation _that_ big!).

... I write this trusting that everyone else on this thread has
more sense than to follow this up with yet more ways to write
shell one-liners ... clearly I lack such sense ...

-- 
                          I won't rest till it's the best ...
                          Manager, Linux System Software
                          Paul Jackson <pj at sgi.com> 1.650.933.1373


From nas at python.ca  Tue Mar 27 13:05:36 2001
From: nas at python.ca (Neil Schemenauer)
Date: Tue, 27 Mar 2001 10:05:36 -0800
Subject: large arbitrary binary data in python
In-Reply-To: <20010327194343.A7081@freedom.puma-ag.com>; from stephen_purcell@yahoo.com on Tue, Mar 27, 2001 at 07:43:43PM +0200
References: <slrn9c1hf7.m5q.wcuddy@lserv.ja10629.home> <20010327192520.F1260@freedom.puma-ag.com> <20010327194343.A7081@freedom.puma-ag.com>
Message-ID: <20010327100536.B18517@glacier.fnational.com>

Steve Purcell wrote:
> Python is a 'yes' language.

Depends on the question. :-)

  destroys-your-filesystem-ly y'rs Neil



From wware at world.std.com  Mon Mar  5 12:38:14 2001
From: wware at world.std.com (Will Ware)
Date: Mon, 5 Mar 2001 17:38:14 GMT
Subject: Coroutines, and memory leaks
References: <8oh089.7it.ln@127.0.0.1>
Message-ID: <G9qJnq.1M4@world.std.com>

David Given (dg at pearl.tao.co.uk) wrote:
> 1. Does Python do coroutines?

Chris Tismer has modified the Python virtual machine to
enable coroutines and other artifacts of modified control
flow. This is described at www.stackless.com and you may
also be interested in the resulting thread package at
http://world.std.com/~wware/uthread.html

-- 
-----------------------------------+---------------------
 22nd century: Esperanto, geodesic | Will Ware
 domes, hovercrafts, metric system | wware at world.std.com


From gradha at iname.com  Sat Mar  3 05:11:26 2001
From: gradha at iname.com (gradha at iname.com)
Date: Sat, 3 Mar 2001 11:11:26 +0100
Subject: Any documentation about Python daemons?
Message-ID: <e0gq79.8og.ln@127.0.0.1>

Hello.

I would like to write now a program which acts as a server on the same
machine for other clients in a Linux environment, something in the spirit
of chronyd, if you know the program.

However, I am new both to Python and process' intercommunication. Could
you suggest me ways of doing this and/or point me to documentation
explaining such communication methods? I first thought about nice pipes,
but then I realized that the server program should be running always, not
being allowed to fork, and clients could mantain multiple communication
channels at the same time (ej: daemon + 2 clients). How could I do this?

-- 

 Grzegorz Adam Hankiewicz   gradha at iname.com - http://gradha.infierno.org
 Other web pages:  http://glub.ehu.es/  -  http://welcome.to/gogosoftware/


From root at rainerdeyke.com  Sun Mar  4 16:05:27 2001
From: root at rainerdeyke.com (Rainer Deyke)
Date: Sun, 04 Mar 2001 21:05:27 GMT
Subject: I come to praise .join, not to bury it...
References: <mailman.983485923.1136.python-list@python.org> <97p7iq$kk2$1@nntp6.u.washington.edu> <97rka002k9@news1.newsguy.com> <jwbnews-665CC3.11122604032001@corp.supernews.com>
Message-ID: <r6yo6.344939$ge4.122193733@news2.rdc2.tx.home.com>

"John W. Baxter" <jwbnews at scandaroon.com> wrote in message
news:jwbnews-665CC3.11122604032001 at corp.supernews.com...
> OK, thanks.  I'm convinced.  For the first time.  (Which won't
> necessarily keep me from writing non-polymorphic things with the string
> module, but it might make me *try* to remember.)

Actually, since the introduction of string methods, the string module has
started to behave polymorphically.  Here's the implementation of
'string.join' as of 2.0:

def join(words, sep = ' '):
    return sep.join(words)

String method make the string module more powerful.


--
Rainer Deyke (root at rainerdeyke.com)
Shareware computer games           -           http://rainerdeyke.com
"In ihren Reihen zu stehen heisst unter Feinden zu kaempfen" - Abigor




From kalle at gnupung.net  Thu Mar 22 16:33:49 2001
From: kalle at gnupung.net (Kalle Svensson)
Date: Thu, 22 Mar 2001 22:33:49 +0100
Subject: expanding a variable
In-Reply-To: <B7C74E1A6FBAD411BFC60002B309DDF4E435@mailbox.juilliard.edu>; from jsoons@juilliard.edu on Thu, Mar 22, 2001 at 04:05:13PM -0500
References: <B7C74E1A6FBAD411BFC60002B309DDF4E435@mailbox.juilliard.edu>
Message-ID: <20010322223348.C984@apone.network.loc>

Sez Jonathan Soons:
> 
> I am trying to give my cv group (who all have logins like cv???.)
> all the same .profile. The variable in the last line of the script
> doesn't seem to be expanding, even though this is very similar to
> an example in "Core Python Progr..."

Note: I haven't read the book, so I don't know the example.

> d = commands.getoutput('ls /u')
> dirs = string.split(d, '\n')
> for dir in dirs:
>     if re.search('^cv', dir):

This could be expressed as
      if dir[:2] == "cv":
or (with Python 1.6 and up)
      if dir.search("cv") == 0:
I don't know which of the three is best, but I would use the slice and
string compare (my first).

>         os.system('cp  /u/test/.profile /u/dir/')

Try
          os.system('cp /u/test/.profile /u/%s/' % dir)
or
          os.system('cp /u/test/.profile /u/' + dir + '/')
perhaps?

> 
> Should I stick to bash for this kind of thing?

Nah, I don't think so.  I use python for almost all my scripting.
BTW, check out the standard module "shutil".
http://www.python.org/doc/current/lib/module-shutil.html

Peace,
  Kalle
-- 
Email: kalle at gnupung.net     | You can tune a filesystem, but you
Web: http://www.gnupung.net/ | can't tune a fish. -- man tunefs(8)
PGP fingerprint: 0C56 B171 8159 327F 1824 F5DE 74D7 80D7 BF3B B1DD
 [ Not signed due to lossage.  Blame Microsoft Outlook Express. ]



From anbu at adsc.com  Fri Mar 30 13:01:09 2001
From: anbu at adsc.com (Anbuselvam Palusamy)
Date: Fri, 30 Mar 2001 10:01:09 -0800
Subject: Application distribution with python
Message-ID: <001001c0b943$6c86b9d0$9071fea5@ANBU>

All,

I need to install python for my client machine with my application.

Can any one knows what are files to ship to client machine or
list of distribution files.

any links?


Thanks
Anbu






From fredrik at pythonware.com  Wed Mar 14 16:42:31 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 14 Mar 2001 21:42:31 GMT
Subject: This math scares me
References: <3aad4062.1290021@News.CIS.DFN.DE> <20010312170356.B3513@better.net> <mailman.984480856.18097.python-list@python.org> <984486395.407379@newsmaster-04.atnet.at> <slrn9asf1d.ja3.qrczak@qrnik.zagroda> <984518199.127140@newsmaster-04.atnet.at> <jwbnews-09D75F.13313313032001@corp.supernews.com> <984521930.27710@newsmaster-04.atnet.at> <Vyxr6.1238$Tg.168802@ruti.visi.com>
Message-ID: <bBRr6.324$4d3.73310@newsb.telia.net>

Grant Edwards wrote:
> Yes.  All FP calculations took place in BCD.  You got exact results
> as long as your values didn't exceed the representations limits.

the first computer I did any serious programming on (the ABC80,
which together with it's successors dominated the swedish per-
sonal computer scene before the IBM PC arrived) offered "string
arithmetics":

    sum = add$(term, term, precision) ! sub$(), mul$(), div$()

where the arguments were ordinary strings, and the precision
argument was an integer giving the number of decimals if positive,
or number of digits if negative.  no other limitations, except for
available memory (16k total) and user patience (3.58 MHz Z80).

(the ABC80 also offered BCD-based floating point, with 6 decimal
digits in the mantissa and one byte for sign and and exponent...)

Cheers /F




From robin at illusionsexeculink.com  Tue Mar 20 11:27:08 2001
From: robin at illusionsexeculink.com (robin at illusionsexeculink.com)
Date: Tue, 20 Mar 2001 11:27:08 -0500
Subject: The most important modules
References: <Pine.GSO.4.21.0103191732210.22234-100000@laplace>
Message-ID: <agqebtgvf3b2mhklqqle0f0f67uqto6cro@4ax.com>

Martyn Quick <mrq at for.mat.bham.ac.uk> wrote:

>I've seen a few comments along the lines of "When you are a newbie it will
>take a while to realize which the important modules are."

These are them for me:
__builtin__ 
string
re
os
sys

Followed by:
time
glob 
traceback 
getopt 
types 

>I notice that as I become aware of
>more of Python's capabilities, I find that previous bits of code can be
>greatly simplified.  Presumably this is a common experience?

I believe so. It depends on how much you are used to the naughty ways
of other languages. <grin>

As an example, try/except/else can simplify an awful lot of manual
bounds checking, and make your code more robust to boot.

> Lutz & Ascher's "Learning Python"

This is *the* book to get when starting out.

>(becoming less of a newbie day-by-day)

As are we all.

-----
robin                          robin at illusionsexeculink.com
media artist /                 remove illusions to reply
information architect          www.execulink.com/~robin/

Isolation tank paves way to execute an attack. 


From fredrik at pythonware.com  Thu Mar 29 06:21:01 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Thu, 29 Mar 2001 11:21:01 GMT
Subject: A newbie question on ftplib.
References: <87Ew6.5454$45.29758@newsfeeds.bigpond.com>
Message-ID: <xUEw6.4147$4N4.842877@newsc.telia.net>

Srihari Vijayaraghavan wrote:
> When I execute the following statement under Linux it uploads the binary
> file properly, but under Windows NT it creates only 512 bytes file at the
> ftp server.
>
> ftp.storbinary('STOR ' + xyz, open(xyz), 1024)
>
> I tried various values (1, 128, 256, 515) for the "number of blocks"
> parameter.

did you try reading from the file?

to make sure you get all data from a binary file, you
need to open it in binary mode.  try this:

    ftp.storbinary('STOR ' + xyz, open(xyz, 'rb'), 1024)

see the "builtin functions" section in the library reference for
more info on open.

Cheers /F




From skip at pobox.com  Wed Mar 28 13:21:19 2001
From: skip at pobox.com (Skip Montanaro)
Date: Wed, 28 Mar 2001 12:21:19 -0600 (CST)
Subject: Is inheritance broken?
In-Reply-To: <NDBBLAKACLGJJBDNGGCMKEINCCAA.jcc.ugm@ix.netcom.com>
References: <NDBBLAKACLGJJBDNGGCMKEINCCAA.jcc.ugm@ix.netcom.com>
Message-ID: <15042.11167.848530.564353@beluga.mojam.com>

    >>>> class square(rhombus, rectangle):
    ... 	pass
    ...
    >>> test_square = square()
    >>> print test_square.angle()
    variable
    >>> 

    Jonathan> It looks like the inherited angle method from rectangle is
    Jonathan> being ignored.

Yup.

    Jonathan> Isn't this wrong?

Nope.  It's a feature of multiple inheritance, not a bug.  The search
through base classes is left-to-right, depth first, thus parallelogram.angle
is found before rectangle.angle.  If you reverse the order of the base
classes in the definition of the square class, angle() will return the
correct value, but side() will return "variable".  This points out a general
gotcha with multiple inheritance.  You can pick the search order (depth
first or breadth first), but you wind up living with your choice.

Here's a version that's perhaps clear(er):

    class parallelogram:
	def angle(self): return "variable"
	def side(self): return "variable"

    class rhombus(parallelogram):
	def side(self): return "same"

    class rectangle(parallelogram):
	def angle(self): return "90 degrees"

    class square1(rhombus, rectangle):
	pass

    class square2(rectangle, rhombus):
	pass

    s1 = square1()
    print "squares have", s1.angle(), "angles"
    print "squares have", s1.side(), "sides"

    s2 = square2()
    print "squares have", s2.angle(), "angles"
    print "squares have", s2.side(), "sides"

Someone with more Python years under their belt will have to answer why
Guido chose depth first instead of breadth first search for methods when
multiple inheritance is involved.  I'm pretty sure it's a choice that can't
be changed now, however.  There's probably too much code that relies on the
current semantics.

-- 
Skip Montanaro (skip at pobox.com)
(847)971-7098



From dfb at mrao.cam.ac.uk  Fri Mar  9 07:46:29 2001
From: dfb at mrao.cam.ac.uk (David Buscher)
Date: Fri, 9 Mar 2001 12:46:29 +0000
Subject: Pmw.Blt and python 2.0/2.1
Message-ID: <Pine.SOL.4.30.0103091241460.28488-100000@mraos>

I've used Pmw.Blt with Python 1.5.2 with no problems, but I can't get it
to work with Python 2.0 or 2.1. I can get Pmw to work fine, but not
Pmw.Blt. Is there a known problem with this? Here is the session
transcript:

Python 2.1b1 (#1, Mar  8 2001, 20:37:37)
[GCC egcs-2.91.66 19990314/Linux (egcs-1.1.2 release)]
on linux2
Type "copyright", "credits" or "license" for more
information.
>>> from Tkinter import *
>>> root=Tk()
>>> import Pmw
>>> Pmw.Blt.haveblt(root)
0
>>> root.tk.call('package', 'require', 'BLT')
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
TclError: couldn't load file "/usr/lib/libBLT.so.2.4":
/usr/lib/libBLT.so.2.4: undefined symbol:
Tk_CanvasTagsParseProc
>

The same dialogue on python1.5 has no problems

Python 1.5.2 (#1, Feb  1 2000, 16:32:16)  [GCC
egcs-2.91.66 19990314/Linux (egcs- on linux-i386
Copyright 1991-1995 Stichting Mathematisch Centrum,
Amsterdam
>>> from Tkinter import *
>>> root=Tk()
>>> import Pmw
>>> Pmw.Blt.haveblt(root)
1
>>> root.tk.call('package', 'require', 'BLT')
'2.4'

I'm running Redhat 6.2, and the 1.5.2 python is 'out of the box', while
the 2.1 is compiled from source - could this cause problems?

My deduction from this output is that this is a problem with the way
tkinter loads shared libraries - has this changed from 1.5 to 2.0/2.1? I'm
using the same shared libraries for tk and blt in all cases.

thanks
David

--------------------------------------+----------------------------------
 Cavendish Laboratory                 |  Phone +44 1223 337302
 Madingley Road                       |  Fax   +44 1223 354599
 Cambridge CB3 0HE, UK                |  http://www.mrao.cam.ac.uk/~dfb
--------------------------------------+----------------------------------



From deblNoNo at SpamNOnotheworld.com  Thu Mar 22 22:30:45 2001
From: deblNoNo at SpamNOnotheworld.com (David Lees)
Date: Fri, 23 Mar 2001 03:30:45 GMT
Subject: python, perl, lisp, tcl / got a decision to make, can you give me a 
 view point please ? and a joke for python users :)
References: <3aba3076.14395287@news.vif.com>
Message-ID: <3ABAC36C.43FE03B7@SpamNOnotheworld.com>

You might want to attribute the source that you copied this joke from
which is at:

http://www.python.org/doc/Humor.html

david lees


From ransen_spam_me_not at nemo.it  Sat Mar  3 01:12:44 2001
From: ransen_spam_me_not at nemo.it (Owen F. Ransen)
Date: Sat, 03 Mar 2001 06:12:44 GMT
Subject: Bug in Py_Finalize() ?
References: <mailman.983565673.18104.python-list@python.org>
Message-ID: <3aa1889e.2213634@news.newsguy.com>

On Fri, 2 Mar 2001 15:40:41 -0500, "Tim Peters" <tim.one at home.com>
wrote:

>Please try this again with the lastest CVS version, or pick up 2.1b1 (we'll
>be releasing that later today).  Something "very much like this" got fixed
>last month; here's Guido's checkin msg:

Thanks for the pointer. 

Now I am going to reveal my ignorance, what does CVS stand for?


--
Owen F. Ransen
http://www.ransen.com/
Home of Gliftic & Repligator Image Generators


From MarkH at ActiveState.com  Fri Mar 16 17:53:30 2001
From: MarkH at ActiveState.com (Mark Hammond)
Date: Fri, 16 Mar 2001 22:53:30 GMT
Subject: PythonCOM connectableserver
References: <3AB24DEE.ED103E37@nortelnetworks.com>
Message-ID: <3AB29AA5.7050602@ActiveState.com>

Masog, David wrote:

> I've used win32com\demos\connect.py as a starting point, but I'm trying
> to write the connectable client portion in VC++ rather than Python.
> 
> When I comment out the line containing the self._BroadcastNotify the
> script executes successfully(I made it return a value to the client).
> 
> When I leave the _BroadcastNotify line in I successfully catch the fired
> event, but immedately afterward receive:
> 	Fatal Python error: PyEval_ReleaseThread: wrong thread state
> 
> I have seen some postings that refer to similar thread issues calling
> Python callbacks from C but don't know how this would apply in this
> case.
> Could someone please shed some light on this issue or direct me towards
> a different approach?

Do you get this error with the sample, or only once you have modified it.

My guess is that you are implementing the other side yourself, and that 
then tries to use Python.  This code is probably not managing the thread 
state correctly.

I can't be more specific as I dont have more specific information about 
the problem.

Mark.



From pieter at nagel.co.za  Fri Mar 16 06:47:19 2001
From: pieter at nagel.co.za (Pieter Nagel)
Date: Fri, 16 Mar 2001 13:47:19 +0200
Subject: gah! I hate the new string syntax
In-Reply-To: <j4kr6.68739$ML1.3282814@e420r-atl2.usenetserver.com>
References: <mailman.983485923.1136.python-list@python.org> <97sudi01i26@news1.newsguy.com> <98elrk$li8$1@nntp.Stanford.EDU> <98fc070thh@news1.newsguy.com> <98iqo2$56o$1@nntp.Stanford.EDU> <j4kr6.68739$ML1.3282814@e420r-atl2.usenetserver.com>
Message-ID: <Pine.LNX.4.21.0103161339210.2809-100000@bast.jhb.equinox.co.za>

"Amit Patel" <amitp at Xenon.Stanford.EDU> wrote:
>
> Even if strings were a class, and I could extend it, it is NOT the
> thing to do to add new functionality like capitalize().  Suppose
> strings do not have a capitalize method, and I want to add one.  And
> suppose strings are classes and can be extended.  If I wrote:
>
>   class MyString(string):
>      def capitalize(self):
>         ...
>
>
> then I can only capitalize strings I created!

Of course you can. The technique is to modify the String class itself
at runtime so that it understands new methods. Then all instances
will support that, whether you created them or not.

Suppose this is the String module you have to work with, and you
can't change it:

	class String:
	    def __init__(self, str):
	        self.str = str

	    def __str__(self):
	        return self.str

	globalString = String('foo')

And here is your extenstion:

	class MyString:
	    def capitalize(self):
	        self.str = self.str.capitalize()

	String.__bases__ = String.__bases__ + (MyString,)
	globalString.capitalize()
	print globalString

Note that you can now capitalize the String instance which you did
not create.

-- 
     ,_
     /_)              /| /
    /   i e t e r    / |/ a g e l



From daniel at dittmar.net  Thu Mar 29 14:23:40 2001
From: daniel at dittmar.net (Daniel Dittmar)
Date: Thu, 29 Mar 2001 21:23:40 +0200
Subject: forcing dictionary order
References: <3AC380AE.B225C07F@uniserve.com>
Message-ID: <3AC38BBC.2D6782E0@dittmar.net>

> I'd like (not a must!) to walk a dictionary in the same order in which
> it was created. Example:

Subclass UserDict and keep track of additions yourself.

Daniel


From pearu at cens.ioc.ee  Tue Mar 27 01:09:04 2001
From: pearu at cens.ioc.ee (Pearu Peterson)
Date: Tue, 27 Mar 2001 08:09:04 +0200
Subject: Biggest float number?
In-Reply-To: <3ABFBB88.E718B2D@nd.edu>
References: <3ABFBB88.E718B2D@nd.edu>
Message-ID: <Pine.LNX.4.21.0103270805001.645-100000@cens.ioc.ee>


On Mon, 26 Mar 2001, C. S. Xu wrote:

> Hi all,
> 
> It seems the biggest floating number in Python can be:
>     2.**1024 - 1.
> which is about 1.8E308. Is there any way to get bigger value than this,
> just like long intergers almost have no limits?

Try GMPY

  http://gmpy.sourceforge.net/

For example,
>>> from gmpy import mpf
>>> mpf('1e2000000000')
mpf('1.e2000000000')
>>> mpf('1.1e200000000')+mpf('3.1e200000000')
mpf('4.2e200000000')

Regards,
	Pearu



From fredrik at pythonware.com  Thu Mar 29 17:10:23 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Thu, 29 Mar 2001 22:10:23 GMT
Subject: htmllib samples
References: <3AC2F48E.4EF77C71@aon.at> <gKCw6.4130$4N4.837469@newsc.telia.net> <9a05uv$png$0@216.39.170.247>
Message-ID: <jpOw6.4239$4N4.877816@newsc.telia.net>

Dave LeBlanc wrote:
> Ok, i'll bite - why would you want to use the significantly greater
> complexity of SGML if you think you want to do HTML?

htmllib is based on sgmllib, but the classes in htmlib are designed
for HTML formatting rather than HTML parsing.  in my experience,
most real-life problems are parser related (e.g. extract all images,
look for title tags, etc), not formatting related.

Cheers /F




From sholden at holdenweb.com  Mon Mar 19 23:51:22 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Mon, 19 Mar 2001 23:51:22 -0500
Subject: Cookies
References: <mailman.985050745.11337.python-list@python.org>
Message-ID: <xoBt6.120591$ML1.5619667@e420r-atl2.usenetserver.com>

Outlook Express. Aarrgghh. I have *no* idea why it isn't mail-quoting this
content, and can't be bothered messing with settings any more.

Sorry ...

"Ken Seehof" <kens at sightreader.com> wrote in message
news:mailman.985050745.11337.python-list at python.org...
Thanks for your help and persistence.  I have come to an adequate
compromise solution. . . .

"David Bolen" <db3l at fitlinxx.com> writes:
> "Ken Seehof" <kens at sightreader.com> writes:
>
> > > Or, if it isn't going to connect to the server, how about just letting
> > > the Python application open up it's own web page internally to run
> > > appropriate code on your web server to generate the identifier (which
> > > could run as a server script), and then Python could parse it right
> > > from the resulting HTML information.
> >
> > Again, it's the same catch-22.  I need the server to generate the same
> > id that the javascipt generated earlier, but I need each client to have
> > a unique id.
>
> I'm still somewhat confused as to the requirement that your javascript
> needs the identifier - is there something else it gets used for as
> part of the web page other than to generate it on behalf of the
> application?

Yes, the page where the user downloads the application also contains
a form (to be filled out while the user is waiting for the download).
The data in the form is written to a database on the server keyed by
the newly generated identifier.  When the application is downloaded,
it uses the same identifier to query the server database for the form
data.  Of course since there are multiple clients, the server needs to
know which client is making the request.

[SH] Erm, is the download still valid if the user fails to submit the form?
This sounds a bit like yer classical modal operation (but I'm guessing).
Might the response rate increase if they had to fill the form out before
they got the download? In which case, might the response to the form be
available tp condition the download process? (Guessing again).


The solution I've ended up with is for the server to generate the id,
the web page to display it, and the application to prompt the user to
manually type it in.

[SH] They had better really *want* this software ...

It would be nice to skip the step where the application prompts the
user (hence the search for a cookie oriented solution), but now that
I see it in action it's not really a big deal (i.e. problem solved, unless
there's a really clever idea that doesn't involve any other compromise).

> If you support generating the id the first time via a script running
> from a normal browser, couldn't you do the same thing as I suggested
> above by letting your application itself be that browser - you could
> just generate the id server side and parse the results right within
> your application, so no need to transfer the id at all.
>
> In other words, why not just use the web page to download the app, and
> that's it.  Don't depend on the web page and browser access for
> anything else.  Let your application connect to your server (since you
> mention it'll do it anyway) and during its first execution get the
> server to generate the id and then use it from there.

I want the browser to be the user interface for the form because the
the application must be very lightweight (i.e. not be a GUI).  There are
other reasons as well to do it this way.  The executable comes from a
possibly overburdened central server, but the data entry form (which
may be customized) is on any of several custom servers (which don't
have enough security to offer an executable).

[SH] Reminder: web clients need to have GUI's: anything you can do sitting
at your browser, a Python web client program can do too (well, not look at
the pages, but I'm sure you understand what I mean).

> If you need to have the first generation take place in the context of
> an HTTP session, that can still be encapsulated by your Python
> application rather than depending on a separate Javascript execution
> in a different browser.

Thanks again everyone,
- Ken

hoping-i'm-not-flogging-this-to-death-ly y'rs  - steve





From moshez at zadka.site.co.il  Tue Mar 27 17:16:26 2001
From: moshez at zadka.site.co.il (Moshe Zadka)
Date: Wed, 28 Mar 2001 00:16:26 +0200
Subject: PEP 240 scares me
In-Reply-To: <15040.53203.949420.930912@beluga.mojam.com>
References: <15040.53203.949420.930912@beluga.mojam.com>, <Pine.LNX.4.21.0103222338530.29360-100000@kev.ioc.ee>
	<99e2h61aao@news2.newsguy.com>
Message-ID: <E14i1lK-0007Qv-00@darjeeling>

On Tue, 27 Mar 2001 11:37:23 -0600 (CST), Skip Montanaro <skip at pobox.com> wrote:
 
> Okay, dumb question time.  Suppose the compiler sees the string "1.2" in the
> input stream.  How does it translate that into a rational number (6/5 or
> 12/10 or 24/20 or ...)?  Does it simply move the decimal point over
> sufficiently and create the corresponding rational number that contains the
> appropriate power-of-ten denominator?

That's what Rational.py (in CVS nondist/sandbox/rational/) does.

> What about rationals that can't be
> represented correctly using decimal notation (e.g. how does one get from
> 0.3333 to 1/3, or is it simply approximated by 3333/10000?)

The later. If you want 1/3, you need to ask for it:

third = rational(1)/3

I guess I could be bribed to have

rational("0.(3)") == rational(1)/3

I just need to recall the formula for the sum of a geometric series and
apply it. There is no rational which cannot be written as

\d*.\d*(\d*)

Where () means "repeat ad infinitum".
-- 
"I'll be ex-DPL soon anyway so I'm        |LUKE: Is Perl better than Python?
looking for someplace else to grab power."|YODA: No...no... no. Quicker,
   -- Wichert Akkerman (on debian-private)|      easier, more seductive.
For public key, finger moshez at debian.org  |http://www.{python,debian,gnu}.org



From grante at visi.com  Thu Mar 22 10:27:43 2001
From: grante at visi.com (Grant Edwards)
Date: Thu, 22 Mar 2001 15:27:43 GMT
Subject: Cross-platform serial port access
References: <99d274$ogv$1@news.netpower.no>
Message-ID: <PRou6.1592$SB2.181656@ruti.visi.com>

In article <99d274$ogv$1 at news.netpower.no>, David Brown wrote:

>I am relatively new to Python, but am planning to use it for a
>number of small programs that should be able to run both on
>Windows (for initial development and testing) and Linux (once
>we get a server running properly). Some of these programs will
>need to communicate with external devices using a serial port -
>is this going to work in a portable fashion using Python?

Currently, serial port stuff in Python is not portable between
Mac/MS/Unix.

>Does anyone know if Linux supports RTS toggle handshaking
>(i.e., the RTS line is high while transmitting, for convenient
>connection to RS-485 convertors),

Not with the standard PC serial port driver.  I hacked the
16x50 driver to do RTS toggle several years ago, but it never
worked reliably.  It's very difficult to do it reliably in
software at high baud rates.

Your best bet is to buy a board that supports RTS toggle in
hardware or use a 232-485 converter that handles it
automatically by detecting the presence of data.  The only
board I've used under Linux that supported hardware RTS toggle
was Cyclades Y series 8-port board.  I believe that RTS toggle
support doesn't work in their newer boards, so you might have
to look around.

>and is it easy to control from Python?  Are there any Python
>modules designed to work with modems?

-- 
Grant Edwards                   grante             Yow!  Could I have a drug
                                  at               overdose?
                               visi.com            


From aahz at panix.com  Thu Mar  8 23:07:42 2001
From: aahz at panix.com (Aahz Maruch)
Date: 8 Mar 2001 20:07:42 -0800
Subject: Coroutines, and memory leaks
References: <8oh089.7it.ln@127.0.0.1>
Message-ID: <989kue$64j$1@panix3.panix.com>

In article <8oh089.7it.ln at 127.0.0.1>, David Given <dg at pearl.tao.co.uk> wrote:
>
>1. Does Python do coroutines?
>
>Reason: at one part of the program I want to run a command-line-like
>routine, that processes and emits log messages at intervals. I want to
>display the result in a window. The most elegant way of connecting this
>routine to GTK is to put them in different threads; but since I don't need
>true concurrency, coroutines would be the simplest solution.

Actually, in Python, threads probably *are* the simplest solution.  One
problem with coroutines is that blocking I/O causes the coroutines to
block.  Anyway, I just finished putting up links to my IPC9 slideshow,
which you may find useful:

http://starship.python.net/crew/aahz/
-- 
                      --- Aahz (Copyright 2001 by aahz at pobox.com)

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

"..., and some of you may regard all women as evil traps that exist only
to tease, torture, and suck out your very soul."  --DrMax


From aahz at panix.com  Fri Mar 16 11:10:41 2001
From: aahz at panix.com (Aahz Maruch)
Date: 16 Mar 2001 08:10:41 -0800
Subject: Python style questions
References: <98tb91$9t9$1@saltmine.radix.net>
Message-ID: <98tdu1$cfk$1@panix2.panix.com>

In article <98tb91$9t9$1 at saltmine.radix.net>,
Cary O'Brien <cobrien at Radix.Net> wrote:
>
>1. I have a class that handles communications with a remote service.
>   It is fairly complex, having to deal with timeouts and retransmits
>   and the like.  The class definition is getting bigger than I'd like.
>   Should I
>
>    a. Live with it
>    b. Factor into mix-in classes
>    c. Define some methods outside of the class definition.
>    d. ???

Separate out the data handling from the communication.  I'd suggest
creating a separate class that handles the low-level communication and
putting an instance of it into your main class.

>2. One thing I've done is to define classes used only by another class
>   inside that class.  For example, if a class A needs to keep track of a
>   bunch of "things", then the class definition for the "things" is inside
>   of the class definition of A.  This is
>
>   a. SOP
>   b. Brilliant
>   c. Idiotic

Depends.  I'm not fond of that trick, though.  It makes more sense to
divide this stuff into modules.

>3. I really miss not having a "switch" or "case" statement.  Sniff.

Use a dictionary.
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

Three sins: BJ, B&J, B&J


From max at alcyone.com  Sat Mar 31 21:04:40 2001
From: max at alcyone.com (Erik Max Francis)
Date: Sat, 31 Mar 2001 18:04:40 -0800
Subject: Conditional operator in Python?
Message-ID: <3AC68CB8.22AAF7B8@alcyone.com>

Okay, okay, but I still have to ask.

Why isn't there a conditional operator (a ? b : c) in Python?  2.0 now
has augmented assignments, and the FAQ even acknowledges that the
conditional operator is sometimes quite convenient and gives some (very
awkward) Python alternatives.

I've been reading comp.lang.python for a while (since just before 2.0
shipped) and I can't really remember much discussion about it.  Is it a
perennial request and I'm just missing it?

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ I sidestep the either
\__/ or choices of logic and choose both. | Ken Feit
    Product's Quake III Arena Tips / http://www.bosskey.net/
 Tips and tricks from the absolute beginner to the Arena Master.


From aleaxit at yahoo.com  Sun Mar  4 12:56:45 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sun, 4 Mar 2001 18:56:45 +0100
Subject: Subclassing builtin types
References: <i6u4atslg4euv732pvaujk6u036kemcidf@4ax.com>
Message-ID: <97tvkt0t0o@news2.newsguy.com>

"Daniel Klein" <danielk at aracnet.com> wrote in message
news:i6u4atslg4euv732pvaujk6u036kemcidf at 4ax.com...
> One of the classes I have to write  is merely a wrapper around a 'list'
object.

That's what UserList is for -- a class that wraps the list builtin object,
so you can inherit and override whatever you need.



Alex





From paulp at ActiveState.com  Sun Mar 11 10:45:17 2001
From: paulp at ActiveState.com (Paul Prescod)
Date: Sun, 11 Mar 2001 07:45:17 -0800
Subject: question , Does pytyon excist on a pda, like psion?
References: <xDJq6.28348$KE1.1127955@zwoll1.home.nl> <98frub$1mhg0$1@fu-berlin.de>
Message-ID: <3AAB9D8D.944591D7@ActiveState.com>

Python is also available on Windows CE-based and PalmOS-based machines:

http://starship.python.net/crew/mhammond/ce/
http://www.endeavors.com/pippy/

 Paul Prescod
-- 
Python:
    Programming the way
    Guido
    indented it.



From aleaxit at yahoo.com  Thu Mar  1 09:07:23 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 1 Mar 2001 15:07:23 +0100
Subject: A counter-proposal to __future__ in PEP 236
References: <j47l29lpo8.fsf@informatik.hu-berlin.de>
Message-ID: <97ll2u02qmp@news2.newsguy.com>

"Martin von Loewis" <loewis at informatik.hu-berlin.de> wrote in message
news:j47l29lpo8.fsf at informatik.hu-berlin.de...
    [snip]
> In this implementation, directive is only considered as a keyword if
> it appears at the beginning of the module. Therefore, code that
> happens to use directive as an identifier will not break (unless it
> assigns to directive as the first thing in the module).
>
> I'd appreciate any comments as to whether this directive statement is
> better or worse for the purpose of indicating that nested scopes are
> used in a module.

It's indisputably much more explicit and direct.  Recycling
'from/import' to mean something very different than other uses
of from/import is subtle and cute.  This implies a huge plus
for the 'directive' patch.

Yes, 'directive' does theoretically imply possible code
breakage -- although, if it were possible to further
special-case things so that the directive statement
always took TWO or more directive-words, as in, e.g.

directive nested scopes

or

directive simple scopes

and/or took a colon, e.g.

directive: scopes nest

or

directive: no nesting

it might perhaps be possible to avoid breaking ANY code
that currently might be working by any stretch of the
imagination (I have not checked how hard it might be to
look-forward in the lexer so that only directive-colon
triggered the new keyword-token -- just reasoning in
the abstract).  [It seems to me the colon would not
damage readability at all here -- on the contrary].

Even if the theoretical breakage of some hypothetical
old module starting with
    directive = 23
or
    directive, floopres = 23,42
was to be the price of avoiding the cuteness of
'from/import' "overloading", I would still vote for
directive (or, rather, against the overloading thing)
if I did have a vote in the matter.  Such special cases
and cuteness are overdone in other languages (C++ and
its different uses of 'static' come to mind...), and
they're a bear to teach to newbies.  The _semantics_
of PEP 236 are a great idea; I do earnestly hope there
is some way, ANY way, to avoid its syntax, and this
'directive' idea gives me a glimmer of hope...!


Alex





From michael+gnus at trollope.org  Thu Mar  8 21:23:08 2001
From: michael+gnus at trollope.org (Michael Powe)
Date: 08 Mar 2001 18:23:08 -0800
Subject: Can anyone recomend a good intoduction to C...
References: <3aa366c6_2@nnrp1.news.uk.psi.net>
Message-ID: <87r907wuib.fsf@cecilia.trollope.org>

>>>>> "Simon" == Simon Brunning <sbrunning at trisystems.co.uk> writes:

    Simon> ... for a Python programmer?  I'm going to have to bite the
    Simon> bullet and learn a bit of C. I'd at least like to learn the
    Simon> *right* way to do it.

Along with K&R, I think you will find the C FAQ valuable.  An
"abridged" online version is available at 

http://www.eskimo.com/~scs/C-faq/top.html

But, the FAQ is available in book form, as well; I recommend its
purchase. 

Steve Summit, <C Programming FAQs:  Frequently Asked Questions>

mp

-- 
    -----------Michael Powe         Portland, Oregon USA----------
"There are 100,000 total marijuana smokers in the U.S., and most are
Negroes, Hispanics, Filipinos and entertainers.  Their Satanic music,
jazz and swing, result from marijuana use.  This marijuana can cause
white women to seek sexual relations with Negroes, entertainers and
any others."  -- Harry J. Anslinger, Dir. Fed. Bureau of Narcotics,
Congressional Testimony, 1937



From sholden at holdenweb.com  Thu Mar 22 06:55:35 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 22 Mar 2001 06:55:35 -0500
Subject: re.match
References: <mailman.985169080.17724.python-list@python.org> <BN3u6.894$sk3.291185@newsb.telia.net> <Wmgu6.4572$WC2.157632@e420r-atl1.usenetserver.com> <Tdiu6.2956$4N4.403069@newsc.telia.net>
Message-ID: <eOlu6.6360$_46.230654@e420r-atl2.usenetserver.com>

"Fredrik Lundh" <fredrik at pythonware.com> wrote in message
news:Tdiu6.2956$4N4.403069 at newsc.telia.net...
> Steve wrote:
> > I thought Unix was Ken's attempt to extend the reach of
> > regular expressions to the user base?
>
> could be, could be.
>
> or maybe it was an attempt to introduce more humor into
> computing ("UNIX", "g/re/p", "dd", "biff", "awk", etc).  given
> the recent "__future__ isn't serious" brouhaha, I'm afraid
> they might have failed...
>
> Cheers /F
>
Aha! Retro Python!

    from __past__ import sense_of_humor
    sense_of_humor.install(martelli.screwdriver)

regards
 Steve





From michael at stroeder.com  Fri Mar 30 19:22:54 2001
From: michael at stroeder.com (Michael =?iso-8859-1?Q?Str=F6der?=)
Date: Sat, 31 Mar 2001 02:22:54 +0200
Subject: Speedup hint for users of viewcvs package
References: <m14j45Y-000CnEC@artcom0.artcom-gmbh.de> <190551356909.20010330211141@buz.ch> <mailman.985990600.4176.python-list@python.org>
Message-ID: <3AC5235E.7FB6118@stroeder.com>

Neil Schemenauer wrote:
> 
> I have no idea why more people don't know about FastCGI.

I once tried it but without success. Can you point us to recent
version of Python modules and examples?

Ciao, Michael.


From be at mek.dtu.dk  Wed Mar 28 06:29:37 2001
From: be at mek.dtu.dk (Brian Elmegaard)
Date: Wed, 28 Mar 2001 13:29:37 +0200
Subject: Making entrys work with tkinter.
Message-ID: <3AC1CB21.4BD10EE8@mek.dtu.dk>

Hi group,

I have been trying to find out how to use the an entry and grab the
result. I am sorry, that I have not been able to figure out from
different sources how bind the entry to an event and make it put back
the result to the calling function. 
What I try is:

class Sheetlabel:
    def __init__(self,canvas,x,y):
        self.frame=Toplevel(width=10,height=20)
        self.frame.title('Insert label')
        Label(self.frame,text='Label:').pack(side='left')
        self.entry=Entry(self.frame)
        #FIX THIS
        self.frame.bind('<Return>',self.ok(x,y,canvas))
        self.entry.pack(side='right')
        self.frame.focus_set()
        
    def ok(self,x,y,canvas,event=None):
        self.text=self.entry.get()
        self.frame.destroy()
        canvas.create_text(20+x,20+y,text=self.text)
        return self.text

which should put the label text on the canvas. With a create_text in
__init__ there is no problem. So somehow I think I am struggling with
scopes. Is that right?
-- 
Brian
http://www.et.dtu.dk/energysystems
http://www.rk-speed.dk http://fiduso.dk http://sunsite.auc.dk/dk-tug
\TeX, tak


From fredrik at pythonware.com  Wed Mar 14 18:22:36 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 14 Mar 2001 23:22:36 GMT
Subject: Thoughts about extensions to the Python language
References: <5.0.2.1.0.20010314181331.0213eac0@mail.inet.com.br> <mailman.984609740.11496.python-list@python.org>
Message-ID: <03Tr6.349$4d3.79695@newsb.telia.net>

Jeremy Hylton wrote:
> Guido did a poll at developers day last week.  The question was: "Is
> Python development moving too fast or at just the right speed?"

as someone pointed out at that meeting, the group of people staying
an extra day at the conference to talk about future development might
be somewhat biased...

(and given that the rest of the "python's future" session was spent
discussing forward iterators, I'm not sure I trust that congregation
at all ;-)

Cheers /F




From pearu at cens.ioc.ee  Wed Mar 21 04:14:29 2001
From: pearu at cens.ioc.ee (Pearu Peterson)
Date: Wed, 21 Mar 2001 11:14:29 +0200
Subject: Help: Rich comparison in derived class, but not in base (boost.python)
Message-ID: <Pine.LNX.4.21.0103211112040.1131-100000@kev.ioc.ee>

Hi!

I am using boost.python for wrapping C++ classes and since boost.python 
does not yet support rich-comparison (any ideas, where I should send
such a request? I am not able to subscribe to boost list:( ). 

I tried to do the wrapping in Python. But that seems to not work either: 
I define a new class `expr', derived from a base `ex' (exposed with
boost.python) that does not have rich-comparisons, with __lt__, etc
methods:

class expr(ex):
    def __init__(self,*args):  
        ex.__init__(self,*args)
    def __cmp__(self,other):
        raise NotImplementedError,"expr.__cmp__() should not be called"
    def __lt__(self, other):
        return relational(self, other, relational.less)
    # also other richcomp methods are defined but not shown here

then in Python (2.1b1) I get:

>>> expr(2) < expr(3)
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
  File "<stdin>", line 5, in __cmp__
NotImplementedError: expr.__cmp__() should not be called

I checked also the following approach:
class expr:
    def __init__(self,*args):
        self.e = ex(*args)
    def __cmp__(self,other):
        raise NotImplementedError,"expr.__cmp__() should not be called"
    def __lt__(self, other):
        return relational(self.e, other, '<')
    # also other richcomp methods are defined but not shown here

that works fine but then I would have to re-define also all `ex' methods
in `expr' --- there are more than 30 of them, so I would not want to do
that. Especially, if in future boost.python will support rich-comparisons.

I appreciate any hints how to wrap classes with no rich-comparisons with
classes that add such fuctionality.

Thanks,
	Pearu




From slinkp23 at yahoo.com  Wed Mar 21 23:01:08 2001
From: slinkp23 at yahoo.com (Paul Winkler)
Date: Thu, 22 Mar 2001 04:01:08 GMT
Subject: Is there python code highlighting in emacs ?
References: <mailman.985223430.22832.python-list@python.org>
Message-ID: <3AB978A9.353A0E4B@yahoo.com>

John McMonagle wrote:
> 
> The hilit19 package in emacs provides keyword highlighting when editing
> code.  For example, if you are writing in C, comments are italicised in
> red, strings are in grey, import statements are in lavender, functions
> are in bold-blue, etc.
> 
> Does anyone know if there is python support in the hilit19 emacs lisp
> package ?

I don't know about hilit19, but I use python-mode which highlights
using font-lock-mode. Comes with GNU Emacs 20.whatever. Try M-x
python-mode and see if anything happens.

python-mode does many other useful things, too. I can't imagine
working without it at this point.

-- 
...................    paul winkler   ....................
custom calendars & printing: http://www.calendargalaxy.com
       A member of ARMS:   http://www.reacharms.com
            home page:  http://www.slinkp.com


From loewis at informatik.hu-berlin.de  Fri Mar 16 16:19:15 2001
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 16 Mar 2001 22:19:15 +0100
Subject: X properties
References: <200103161745.MAA07707@smtp2.andrew.cmu.edu>
Message-ID: <j4bsr1h0oc.fsf@informatik.hu-berlin.de>

Brian E Gallew <geek+ at cmu.edu> writes:

> In any of the GUI toolkits, is there a way to add random window
> manager properties?  After the window has been created?

Good question. There is the python-xlib project,
http://sourceforge.net/projects/python-xlib, which does support X11
properties (obviously), you just need to get the window id out of your
toolkit - I know Tk can give provide you with the window ID.

Regards,
Martin



From root at rainerdeyke.com  Wed Mar  7 13:11:24 2001
From: root at rainerdeyke.com (Rainer Deyke)
Date: Wed, 07 Mar 2001 18:11:24 GMT
Subject: I come to praise .join, not to bury it...
References: <mailman.983485923.1136.python-list@python.org> <97p7iq$kk2$1@nntp6.u.washington.edu> <97rka002k9@news1.newsguy.com> <3AA2F4B3.A7496E76@cosc.canterbury.ac.nz> <97vmoh061p@news2.newsguy.com> <3AA46134.D385208B@cosc.canterbury.ac.nz> <98292h0s0g@news1.newsguy.com> <3AA58BD0.3324C709@cosc.canterbury.ac.nz> <9850ge013ns@news1.newsguy.com>
Message-ID: <gRup6.364652$ge4.128679630@news2.rdc2.tx.home.com>

"Alex Martelli" <aleaxit at yahoo.com> wrote in message
news:9850ge013ns at news1.newsguy.com...
> in fact, just about any test for 'IS object X of exact type
> Y' you can find in the Python sources, unless they are
> shortcut-like 'accelerators' for a special case _before_
> the more-general request 'please o mr X try to behave like
> an Y, are you able to?', can be seen as implementation
> defects, in my personal opinion.

I'm curious: how do you feel about the automagic conversions of functions
(and only true function, not function-like objects) into unbound methods,
and from unbound methods into bound methods?  I'm thinking that the former
could be eliminated entirely, and the latter replaced by a __bind__ magic
method that allows the mechanism to apply to function-like objects.


--
Rainer Deyke (root at rainerdeyke.com)
Shareware computer games           -           http://rainerdeyke.com
"In ihren Reihen zu stehen heisst unter Feinden zu kaempfen" - Abigor




From urner at alumni.princeton.edu  Sat Mar  3 14:26:51 2001
From: urner at alumni.princeton.edu (Kirby Urner)
Date: Sat, 03 Mar 2001 11:26:51 -0800
Subject: Missing the functional mind set
References: <mlb2at8ptrr7sh5dgq8qlk2201kquh6tfl@4ax.com> <uDao6.335972$ge4.119457077@news2.rdc2.tx.home.com> <t6f2at870pk82q96bnd5a2od4n8ns5kf83@4ax.com>
Message-ID: <ntg2atcemm3hp9rb26hn7c5uin4dfn8lc9@4ax.com>

Daniel Klein <danielk at aracnet.com> wrote:

>Is there a better approach?
>
>Daniel Klein

You might poke values into a list out of order if 
you create the default placeholders in advance.

Instead of:

  >>> params = {}       # dictionary
  >>> params[2] = uv1   # uservalue for param 2
  >>> params[3] = uv2
  >>> params[0] = uv3
  >>> params[1] = uv4

you could go:

  >>> params = [None]*3 # list
  >>> params[2] = uv1   # uservalue for param 2
  >>> params[3] = uv2
  >>> params[1] = uv3
  >>> params[0] = uv4

You'd end up with an ordered list, same as if you'd
stuffed a dictionary and splurted out a list ordered
by key.

Kirby



From rikbo716 at student.liu.se  Tue Mar 20 03:36:15 2001
From: rikbo716 at student.liu.se (Rikard Bosnjakovic)
Date: Tue, 20 Mar 2001 09:36:15 +0100
Subject: Sequence-spreading
Message-ID: <3AB7167F.2816AA21@student.liu.se>

Hi.

I made a small spreading-sequence with the following code:

------
arg = "some string"
pm = 3
offsets = []
for i in xrange(pm):
    everyX = filter(lambda y: not divmod(y, pm)[1], range(len(arg)))
    added = map(lambda x: x+i, everyX)
    offsets.append(filter(lambda x: x<len(arg), added))

print offsets
------

which gives the correct result:

>>> ## working on region in file /usr/tmp/python-3115DCp...
[[0, 3, 6, 9], [1, 4, 7, 10], [2, 5, 8]]

"pm" is the number of sequences, and then the sequences are increasing.
0 in first, 1 in second, 2 in third, 3 in first and so on, until the
length of the string 'arg' is reached.  However, the loop-code is pretty
unreadable and awkward.

Anyone got a better solution for it?



-- 
Rikard Bosnjakovic - http://a214.ryd.student.liu.se/cv/ - ICQ: 1158217

Anyone sending unwanted advertising e-mail to my address will be
charged $250 for network traffic and computing time. By extracting my
address from this message or its header, you agree to these terms.


From aleaxit at yahoo.com  Thu Mar 22 18:36:48 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 23 Mar 2001 00:36:48 +0100
Subject: python, perl, lisp, tcl / got a decision to make, can you give me a view point please ? and a joke for python users :)
References: <mailman.985296601.15357.python-list@python.org>
Message-ID: <99e2h40aao@news2.newsguy.com>

"Mike C. Fletcher" <mcfletch at home.com> wrote in message
news:mailman.985296601.15357.python-list at python.org...
    [snip]
> Lisp is fun, but once you get past the program-is-data stuff it's not
really
> mind-expanding either.  Maybe that's just 'cause I only ever used
AutoLISP,
> could be there's some consciousness-building stuff somewhere in the more
> exotic variants...

Actually, any plain old Scheme *plus the book "Structure and Interpretation
of Computer Programs"* is a great mind-expander, IMHO (SICP is _the_
reason to learn Scheme!-).


Alex





From dsh8290 at rit.edu  Tue Mar 13 11:45:39 2001
From: dsh8290 at rit.edu (D-Man)
Date: Tue, 13 Mar 2001 11:45:39 -0500
Subject: Static typing (was Re: Java guy interested in Python)
In-Reply-To: <m3elw39a7m.fsf@atrus.jesus.cam.ac.uk>; from mwh21@cam.ac.uk on Sun, Mar 11, 2001 at 11:06:53PM +0000
References: <mailman.984334635.27345.python-list@python.org> <98glhc02sga@news2.newsguy.com> <m3hf1080or.fsf@atrus.jesus.cam.ac.uk> <98gt3f064l@news2.newsguy.com> <m3elw39a7m.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <20010313114539.C10298@harmony.cs.rit.edu>

On Sun, Mar 11, 2001 at 11:06:53PM +0000, Michael Hudson wrote:
| "Alex Martelli" <aleaxit at yahoo.com> writes:
| 
| > Sorry, too subtle -- the point was meant to be that a #define int
| > foo _would_ make it legal (and incomprehensible to anybody, if the
| > #define is well-hidden enough) -- sort of trying to see who'd fail
| > to spot this.  But on re-reading it falls pretty flat -- sorry!
| 
| Actually, I thought
| 
| #define int foo
| 
| was pedantically illegal C++ that was almost certain to work, but on
| checking I see that it's probably legal.  I also hadn't taken this
| possibility into account when replying above; catching people out with
| the preprocessor isn't really fair...
| 

If you use g++ 2.96, 

#define or  anything
#define not anything

is illegal unless you give it the -fno-operator-keyword flag.  I found
this out when I tried to compile a particular C++ program available on
the 'net.

I don't know if this is the GNU people trying to help C++ programmers
out, or if it is a stricter adherence to the standard, or a deviation
from the standard.

-D




From ransen_spam_me_not at nemo.it  Wed Mar  7 01:31:47 2001
From: ransen_spam_me_not at nemo.it (Owen F. Ransen)
Date: Wed, 07 Mar 2001 06:31:47 GMT
Subject: Hang in embedded Python (Windows) app
References: <9822i5$bkn$1@thoth.cts.com>
Message-ID: <3aa6d52b.6074276@news.newsguy.com>

In simpler terms: does PyRun_SimpleFile crash?

On Mon, 5 Mar 2001 23:08:11 -0800, "Dave Fassett" <dfassett at cts.com>
wrote:

>I'm writing a Windows app with an embedded Python interpreter. I have the
>stdio output going to my own window and I've made a few working custom
>object in C that I can use from scripts. My problem is the system will hang
>when running the script, and if I break in the debugger it's inside of NT,
>with no real stack trace present. This happens when I run my script
>repeatedly - and always when I sit on the accelerator key that runs the
>current script (i.e., repeating the script as fast as possible). Even the
>simplest script that uses no custom objects will hang (such as "for p in
>range(1,10):
> print p" - which doesn't hang if I repeat it slowly), so it's not anything
>about reference counts or using custom objects incorrectly. I've compiled
>Python 2.0 under Windows using VC++ 6.0 using threads. This is a real
>showstopper, so any help would be greatly appreciated!
>
>Dave
>
>

--
Owen F. Ransen
http://www.ransen.com/
Home of Gliftic & Repligator Image Generators


From dnew at san.rr.com  Tue Mar  6 12:31:49 2001
From: dnew at san.rr.com (Darren New)
Date: Tue, 06 Mar 2001 17:31:49 GMT
Subject: absolute namespace for callbacks?
References: <3AA50AFB.83C7459F@troikanetworks.com>
Message-ID: <3AA51F06.F37E87E@san.rr.com>

> How does one register a callback in Python, when the callback func
> exists in a different namespace?

Knowing both Tcl and a little Python, I think I understand your confusion.

In Tcl, you pass the name of the function to be called, as a string. The
interpreter does not know, at the time of creating the callback, that it's
actually the name of a function. Hence, it needs to be "fully qualify"
namespace-wise in order for the interpreter to map from the function name to
the function body when the callback invokes it.

In Python, you pass (essentially) a pointer to the function itself. That is,
the widget is storing the address of the function to be invoked in its local
variables, rather than the name of the function. The function is an object
(in the OO sense) so it knows its own namespace and can get to its locals
thru pointers embedded in the function object itself. 

(Modulo oversimplification, of course.)

-- 
Darren New / Senior MTS & Free Radical / Invisible Worlds Inc.
San Diego, CA, USA (PST).  Cryptokeys on demand.


From schorsch at schorsch.com  Fri Mar 23 14:23:06 2001
From: schorsch at schorsch.com (Georg Mischler)
Date: Fri, 23 Mar 2001 20:23:06 +0100
Subject: Commercial Products in Python?
References: <3abb86ee.142862695@localhost>
Message-ID: <99g4oe$3b1$05$1@news.t-online.com>

Victor Muslin wrote:

> A friend of mine is considering developing a shrink-wrapped commercial
> product. Python seems like a possible choice, but it is not clear
> whether there are any precedents (though probably similar concerns
> apply to a Perl-based product).
> 
> I would very much appreciate an advice in this area and possibly
> answers to some of the following questions:

I won't claim to answer your "what is the best way" type of
questions. Instead, I'll just tell you what I have found
reasonable and practical for my own stuff.


> 1. Are there any existing shirnk-wrap commercial products written in
> Python?
>
> 2. If so, what are they?

Not sure if this fits a narrow definition of "shrink wrapped",
but it's definitively a commercial product, shipped to customers
on CD-ROM in a cardboard box: http://www.schorsch.com/rayfront/
Rayfront consists of (among many other things) 30'000 lines
of custom Python code, not counting various included toolkits.


> 3. What is the best way to go about it? Is it better to have customers
> install Python and give them bytecode files or use one of the programs
> that creates an executable?

My installer brings its own (renamed) interpreter, the standard
library in sources, and the custom modules as bytecode files.
A full multiplatform installation of Rayfront consists of about
2'500 files anyway, so that packing 150 of those into a frozen
binary looked rather pointless. For smaller packages, the math
will be different, with other reasons for or against freezing.
Installing the plain byte code files also has the advantage
that any bugfix updates just need to replace a few usually
relatively small files.

Relying on the users to have/install a working Python on
their machine means to build your application on very shaky
foundations. If all your customers were python hackers
anyway, then it might work, but since you're talking about
"shrink wrapped" stuff, it probably won't.


> 4. There is an easy shrink-wrap install for Python (at least on
> Windows NT). Are there such on Unix/Linux platforms? (Typically
> customers don't like having to compile software from source).

I use InstallBuilder on Windows, and custom shell/python
scripts on unix. This may depend on the nature of your software
and your target audience, but you should be careful not to
depend on a GUI for installing on unix. This is especially
true for software that may be installed on server systems
across a local network.

Also check the recent activities of the distutils-SIG.
There are new tools around now that support building your
installable data sets and more, that weren't available yet
when I started my project.


> 5. If the answer to question #1 is "no", why not?

I'll leave that to those who didn't try... ;)


> 6. Are there any licensing issues to be aware of?

I think you will have to include the text of the Python
license somewhere in your installation/documentation, but
I may be wrong. At least that's what I have done.


> Anything else that comes to mind in this direction would be very
> helpful.

Well, my "real" problems were not Python related at all,
so mentioning them here wouldn't help you much.


Have fun!

-schorsch

-- 
Georg Mischler  --  simulations developer  --  schorsch at schorsch.com
+schorsch.com+  --  Lighting Design Tools  --  http://www.schorsch.com/


From aleaxit at yahoo.com  Wed Mar 14 18:09:30 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 15 Mar 2001 00:09:30 +0100
Subject: Who's minister of propaganda this week?
References: <mailman.984595938.11127.python-list@python.org>
Message-ID: <98otuq0nnf@news1.newsguy.com>

"Steven D. Majewski" <sdm7g at virginia.edu> wrote in message
news:mailman.984595938.11127.python-list at python.org...
    [snip]
> ( OK: not having multiple dispatch in Python means that some manual
> dispatching is sometimes required, but usually this is burried deep
> in some classes __coerce__ method. )

"Usually" in arithmetic, perhaps.  For fancier stuff, lack of
multimethods sometimes cramps (not that it cramps any
less in C++, mind you), though some patterns (such as
'dynamic visitor') can help a bit in certain (reasonably
frequent) special cases.


>  If think you're expecting a file, what you are probably *really*
> expecting is a file-like object with a read or write method. Testing
> if it is in fact a file is probably the wrong thing to do. ( This
> has been termed "pedantic typechecking" in c.l.python ). If you try

Right.  But this holds in well-designed C++ as well -- one
programs (across components) via abstract interfaces (aka
pure abstract classes, aka protocol classes -- see Lakos' opus
on large-scale C++ design for a deep defense of this, as well
as the origin of the 'protocol-classes' terminology for it).

The only real issue is whether this can be checked at compile
time, or needs to be verified at runtime (in C++, you get
static checking by default, but dynamic_cast<> exists as well,
so the language explicitly recognizes the need for dynamic
typechecking of the does-this-object-implement-this-interface
variety -- it does require you to 'package' methods up into
interfaces, but that's not a big issue [the interface segregation
principle dictates that an interface must have _few_ methods;
by-method typechecks, as in Python, push that to the extreme;
and note that Martin expounded the IS principle _before_ he
became Python-interested, back when he was C++ all the
way down:-)].


Alex





From mklein at alumni.caltech.edu  Thu Mar 22 17:34:17 2001
From: mklein at alumni.caltech.edu (Michael Klein)
Date: Thu, 22 Mar 2001 14:34:17 -0800
Subject: Einstein's Riddle
References: <oNPq6.396644$ge4.138479566@news2.rdc2.tx.home.com> <98glrd$1si1a$1@ID-75892.news.dfncis.de> <3AABE712.499CCF4C@home.com> <98gp5b$1u0oi$1@ID-75892.news.dfncis.de> <98i5880pl2@news2.newsguy.com> <98kfrl$pug$1@sshuraaa-i-1.production.compuserve.com> <98kntc0o5l@news2.newsguy.com>
Message-ID: <3ABA7DE9.8C7A2AEE@alumni.caltech.edu>

> [ eating emu ]

You guys are talking about eating the poor German's pet!  Have you no heart.



From fredrik at pythonware.com  Tue Mar 20 02:20:31 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Tue, 20 Mar 2001 07:20:31 GMT
Subject: Why "from __future__" stinks; a counter-offer
References: <mailman.985029808.7088.python-list@python.org> <memo.20010319213204.8771C@digitig.compulink.co.uk> <9965t7$s57$1@panix6.panix.com> <9969ta$2is$1@newsg3.svr.pol.co.uk>
Message-ID: <3xDt6.2393$4N4.290910@newsc.telia.net>

Will Newton wrote:
> How often is this "from __future__" expected to be used?

*never*, by programmers writing production code (or any kind of
portable Python code, for that matter).

> Just experimental code written for bleeding edge versions of Python or a
> significant portion of code (that will eventually become legacy code)?

the former.  the whiners don't get it, but the whole idea here is to
extend the beta period, not to introduce optional features into the
language.

Cheers /F




From nomail at nomail.com  Sat Mar 31 21:39:40 2001
From: nomail at nomail.com (Dry Ice)
Date: Sat, 31 Mar 2001 18:39:40 -0800
Subject: Know of Substantial Apps Written in Python?
References: <3ac4e4a8_1@news4.newsfeeds.com> <KE6x6.415$SR6.235983@typhoon1.ba-dsg.net> <3ac4ff12$1_2@news4.newsfeeds.com> <3AC680EE.997108AB@northwestern.edu>
Message-ID: <3ac69340_2@news4.newsfeeds.com>

> I'm not sure why the prejudice against using a language to drive C (or
> C++ or Fortran) modules.  [...]

No prejudice implied.  Such a use is entirely reasonable.

I am evaluating, however, Python-as-development-language,
not as glue for modules which offer the bulk of the
functionality in a project.

For years many accused Visual Basic of being a "toy"
language (like "real" coders only use C, dude) because
to do FAST stuff one really had to link to something).

(Then the OO people called it a toy for other reasons.
Now that it does lots of OO stuff, it's a toy because it's
not cross-platform.  Truth is, I see some Visual Basic
everywhere I look.  Particularly, with considerable
amusement, when I'm accused of prejudice against
RAD languages as glue.)






From max at alcyone.com  Sat Mar 10 21:19:40 2001
From: max at alcyone.com (Erik Max Francis)
Date: Sat, 10 Mar 2001 18:19:40 -0800
Subject: Can anyone recomend a good intoduction to C...
References: <mailman.983894288.19965.python-list@python.org> <ewwp6.1466$y6.241005@ruti.visi.com> <slrn9aeblb.cg1.scarblac@pino.selwerd.nl> <djagat4kk3cs9hg0c3k39cudk6goblhkg6@4ax.com> <989mbk$24nuv$1@fido.engr.sgi.com> <slrn9ah66m.k09.qrczak@qrnik.zagroda> <98auhn0270j@news2.newsguy.com> <slrn9aiaq7.rmu.qrczak@qrnik.zagroda> <98dpoe12ckn@news1.newsguy.com> <3AAA85C1.81FB90AA@shore.net> <98eb1f02uqq@news1.newsguy.com>
Message-ID: <3AAAE0BB.CBCC99EC@alcyone.com>

Alex Martelli wrote:

> It's no doubt possible that gcc has significantly worsened in the
> last 5 or 6 years, but I'll confess that would really surprise me.

No, it's actually gotten far better.  The pre-egcs versions of g++ had
some really glaringly seriously problems with modern draft Standard (now
actual Standard) compliance like exceptions, templates and the like.  If
you weren't using these, though, g++ was just fine (which was probably
the case for you).  Note that modern version of egcs have all the
Standard language _features_ all settled out, but there are a few
Standard compliance problems in the library.  These are scheduled to be
fixed in gcc 3.0.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ It is impossible to love and be wise.
\__/ Francis Bacon
    Computer science / http://www.alcyone.com/max/reference/compsci/
 A computer science reference.


From lluang at northwestern.edu  Mon Mar 12 17:36:33 2001
From: lluang at northwestern.edu (Louis Luangkesorn)
Date: Mon, 12 Mar 2001 16:36:33 -0600
Subject: This math scares me
References: <3aad4062.1290021@News.CIS.DFN.DE>
Message-ID: <3AAD4F71.B6BD980F@northwestern.edu>

In addition to the comments you have gotten regarding floating point
arithmetic being inexact, try not to give those final digits in your
answer to anyone or do calculations that rely on those last digits. So if
you ever print this out, if you use the formatted output [i.e. print("
%6.4f" % (x))] that last bit will not appear, so you would be reasonably
safe.

Louis

costas at springmail.com wrote:

> Ok, I can see maybe division having problems. But why does addition of
> the two numbers below:
>
> 5.01+5.54
>
> give me this?
>
> 10.550000000000001
>
> I hope its some setting that can take care of this. I am using Python
> under Windows 98.
>
> TIA
>
> Costas

--
K Louis Luangkesorn
lluang at northwestern.edu  http://pubweb.nwu.edu/~kll560  PGP:0xF3E2D362
Whatsoever things are true, ... honest, ... just, ... pure, ... lovely,
... of good report; if there be any virtue, and if there be any praise,
think on these things.- motto - Northwestern University



From aleaxit at yahoo.com  Fri Mar  9 07:10:46 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 9 Mar 2001 13:10:46 +0100
Subject: Static typing (was Re: Java guy interested in Python)
References: <mailman.984131592.17401.python-list@python.org>
Message-ID: <98ah8301ndm@news2.newsguy.com>

"Hamish Lawson" <hamish_lawson at yahoo.co.uk> wrote in message
news:mailman.984131592.17401.python-list at python.org...
    [snip]
> Wouldn't there be a hierarchy of interface types for file-like objects,
> with some interfaces specifying more methods than others? You'd then

It probably wouldn't be a hierarchy -- there is no special sense
in which any of .read, .readline, .readlines, .xreadlines is "more
fundamental" than any of the others (Haskell's typeclasses are the
best way I know to express such common situations -- methods can be
defined in a mutually recursive way in the typeclass, the types that
are its specific instances may break the recursion any way they like,
and client-code gets all the desirable usage convenience).

> choose the interface type that specified the collection of methods that
> you require of your passed object.

Any of the 2**4-1 non-empty subsets of the above 4 may make
sense.  Why invent 15 names for the subsets?  The collection
of methods that you require is exactly the set of methods
you will actually call.  Interfaces may have their uses (I'd
LOVE a way to get Haskell typeclasses in Python!-) but this
doesn't seem to be a particularly useful one.


Alex





From ksaranat at yahoo.com  Sun Mar  4 01:05:38 2001
From: ksaranat at yahoo.com (Kirk S)
Date: Sun, 04 Mar 2001 06:05:38 GMT
Subject: Help with finding how much memory a variable is using
References: <JAwl6.17037$5n4.338201@news6-win.server.ntlworld.com>
Message-ID: <87n1b2vzke.fsf@pulsar.wlfdle1.on.wave.home.com>

"Rich Somerfield" <rich_somerfield at tertio.com> writes:

> Hi,
> 
> I am generating a huge, huge list of data (cant think of a decent way to
> reduce the required storage size and have it in a usable form).  Everytime i
> try to keep this list for future derivations I get a memory problem from
> Windows (effectively terminating my python script).
> 
> I presume this is because of the huge list.  Is it possible to find out the
> amount of memory a variable [not a type of the variable, the actual data
> contained within the variable] is taking up?  This variable is local to a
> class and it would appear that it is when i go out of scope of that class
> that the memory error is being issued.  Would I be better making this
> variable global (across all classes)? e.g. would this problem still happen
> if the variable was not just local to 1 class?
> 
> Would a different python datatype solve this problem for me ?
> 
> If I cant handle this amount of data then I will have to think of a way of
> better representing it, this will be unfortunate though as and compression /
> reduction of data will [as far as i can think] vastly reduce performance -
> which is already a concern.
> 
> Regards
> Rich
> 
> 
> 

Hello Rich.  I'm new to python myself but with 75 gig hard drives
available why wouldn't you just settle for a huge swap file on your hard
drive?  If 75 gigs isn't enough I know there is a way (in freebsd/linux at
least) to make multiple drives look like 1 to the OS.  

Then again improving the code for the program is a better first step.
Anyways, good luck.

kirk

-- 
Public key @ http://www.geocities.com/ksaranat/pub-key.txt


From vsoftsmith at aol.com  Wed Mar  7 21:14:24 2001
From: vsoftsmith at aol.com (VSOFTSMITH)
Date: 08 Mar 2001 02:14:24 GMT
Subject: parallel port
Message-ID: <20010307211424.26826.00000202@ng-mm1.aol.com>

how can i send a character stream to the parallel port to drive a printer?
python 2.0, win me

thank you


From fredp at mygale.org.nospam  Tue Mar 20 12:14:47 2001
From: fredp at mygale.org.nospam (Fred Pacquier)
Date: 20 Mar 2001 17:14:47 GMT
Subject: Accessing Oracle from Python remotely ?
Message-ID: <Xns906AB9C901807PaCmAnRDLM@161.48.76.20>

Hi,

I have an Oracle (7) database running on a Windows NT 4 server. Is it 
possible (easy, difficult ?) to read the data in those tables from a python 
script running on a remote Unix (Solaris) box ?

Sorry if the question is silly -- I do know a bit of python but I'm 
inexperienced in databases and their connectivity. Any pointers welcome 
though.

Thanks,
fp

-- 
YAFAP : http://www.multimania.com/fredp/


From digitig at cix.co.uk  Tue Mar  6 13:17:00 2001
From: digitig at cix.co.uk (Tim Rowe)
Date: Tue, 6 Mar 2001 18:17 +0000 (GMT Standard Time)
Subject: XML Module problems
References: <mailman.982853044.8344.python-list@python.org>
Message-ID: <memo.20010306181753.34635C@digitig.compulink.co.uk>

In article <mailman.982853044.8344.python-list at python.org>, 
vonWedel at lfpt.rwth-aachen.de (Lars von Wedel) wrote:

> Hello Martin,
> 
> > > First I tried Python on a Solaris machine, finding out that the
> > > python distribution installed does not contain _any_ XML parsers!?
> > 
> > It sure does include pyexpat - you have to have the expat library to
> > compile it, though. Of course, I don't know how you had installed
> > Python, so it is certainly possible that there is no XML parser in
> > your installation.
> 
> As far as I know we installed a binary package from some site, don't
> know
> from where exactly, though. However, in xml/parsers there is expat.py
> but no pyexpat.

<snip>

That's what I have with my (ActiveState) 2.0 installation. 
xml.parsers.expat is very like pyexpat but with some subtle differences 
that make the name change appropriate (for example, IIRC expat returns 
attributes as a dictionary rather that pyexpat's list -- expat's version 
makes much more sense to me). You might also want to look at pyxie which 
will niftily dynamically bind members calls to element names -- I'm sure 
Sean will turn up with a URL in a moment, but a Google search finds it 
easily enough.


From jurberg at my-deja.com  Thu Mar 22 17:12:26 2001
From: jurberg at my-deja.com (jurberg Last Name)
Date: Thu, 22 Mar 2001 22:12:26 +0000 (UTC)
Subject: Accessing Python stuff thru OLE...
Message-ID: <200103222212.OAA03445@mail20.bigmailbox.com>

Hi all,

We are looking for ways to move away from PowerBuilder.  I have identified ways to use Java thru OLE but I would also like to try using Python (cause it's much better than Java IMHO).  Here is what I am trying to do (and have already done with a Java example):

1) Call functions on a Python object 
-- works great

2) Get events back to PowerBuilder from Python 
-- Can sort of do on Java by using ActiveX bridge.  Don't know how in Python.

3) Open a window from a Python object thru OLE.
-- Can sort of do with wxPython.  How to do with Tkinter?

I used the example code at the end of this post.  I took the MyAboutBox from the wxPython demo and tried opening it thru my OLE object.  Works great the first time, but I get the only 1 instance of wxApp allowed per thread exception afterwords.  Anyway to fix this?  

I also tried using tk.  It worked when the OLE object was created from PythonWin but not from PowerBuilder.  Is there a python path thing I am missing?

Thanks,
John Urberg

# PowerBuilderGuiTest.py
from   wxPython.wx       import *
from   wxPython.html     import *
import wxPython.lib.wxpTag

class MyAboutBox(wxDialog):
    <... wxPython Demo About Box code ...>

class MyApp(wxApp):
    def OnInit(self):
        frame = MyAboutBox(None)
        frame.Show(true)
        self.SetTopWindow(frame)
        self.ShowTip(frame)
        return true

class PowerBuilderGuiTest:
   _public_methods_ = ['openFrame', 'openTkFrame', 'throwException']
   _reg_progid_ = "Python.PowerBuilder.Gui.Test"
   _reg_clsid_ = "{7934BD91-1E00-11D5-848F-0050DAC2AA7E}"

   def openFrame(self):
        app = MyApp(0)
        app.MainLoop()

   def openTkFrame(self):
        from Tkinter import *
        root = Tk()
        w = Label(root, text="Hello, world!")
        w.pack()
        root.mainloop()
        
   def throwException(self):
        raise Exception, 'Hello from Python'
       
if __name__ == '__main__':
   import win32com.server.register
   win32com.server.register.UseCommandLine(PowerBuilderGuiTest)


------------------------------------------------------------
--== Sent via Deja.com ==--
http://www.deja.com/



-- 
Posted from [12.24.248.70] by way of mail20.bigmailbox.com [209.132.220.51] 
via Mailgate.ORG Server - http://www.Mailgate.ORG


From t*b*r*u*m*l*e*y at f*u*s*e.n*e*t  Mon Mar 26 07:53:43 2001
From: t*b*r*u*m*l*e*y at f*u*s*e.n*e*t (Troy Brumley)
Date: Mon, 26 Mar 2001 07:53:43 -0500
Subject: [VW] lesson browser (was Re: Collection interfaces)
References: <3a90fa97.0@news1.mweb.co.za>   <3A9A86B1.9A8C69A2@ix.netcom.com> <MPG.1504d14563dce6499896af@news.earthlink.net> <97fjla128ir@news2.newsguy.com> <97h5s40vhc@news2.newsguy.com> <slrn99opop.gr.amk@mira.erols.com> <MPG.1506265ebf6d934d9896be@news.earthlink.net> <97ibmc0k47@news1.newsguy.com> <MPG.1506cf3293b094b49896c1@news.earthlink.net> <97m0c7080c@news2.newsguy.com> <MPG.15085517debf4b649896ca@news.earthlink.net> <zb%n6.31033$o%6.586026@news1.rdc1.bc.home.com> <MPG.150b383cea226d409896d0@news.earthlink.net> <kNko6.6986$eq3.707121@news.uswest.net> <MPG.150c46db9cec283e9896d4@news.earthlink.net> <3AA2A898.450B659E@mail.com> <gRCt6.12224$Im6.1308560@newsread1.prod.itd.earthlink.net> <3AB755F8.67E84126@mail.com> <%a%t6.17245$227.1601569@newsread2.prod.itd.earthlink.net> <3AB8A730.4BE13D22@mail.com> <61bv6.8693$P4.676800@newsread1.prod.itd.earthlink.net> <B6E2B64C.D8E3%t*b*r*u*m*l*e*y@f*u*s*e.n*e*t> <o1jv6.9635$P4.789083@newsread1.prod.itd.earthlink.net>
Message-ID: <B6E4A607.DCF2%t*b*r*u*m*l*e*y@f*u*s*e.n*e*t>

in article o1jv6.9635$P4.789083 at newsread1.prod.itd.earthlink.net, topmind at
topmind at technologist.com wrote on 3/25/01 4:39 AM:

>> 
>> 
> 
http://www.cincom.com:80/scripts/smalltalk.exe/gettingstarted/index.asp?cont
ent=trysmalltalknc
>> 
> 
> It seems to be wacked out right now.

OK, we just converted the site and we're still shaking it down.

> 
>> Pardon the word wrapping my news client does. There's no spaces in the
>> above.
>> 
>> That should get you to a place where you can download the lesson browser
>> (thanks again to Mark Windholtz, great job!) and the tutorial (thanks to
> Jim
>> F. as well).
>> 
> 
> Lesson browser download?
> 
> The last thing I need is yet more foriegn software on my PC making it crash
> more and confusing other DLL's.
> 
> Please, HTML.

The lesson browser is a VisualWorks parcel that introduces new Smalltalkers
to both the environment and the language in a controlled browser in a
tutorial mode so that newbies don't have to wrestle with multiple browser
windows until they are ready to.

No DLLs, all Smalltalk.

> 
>> ...site runs on
>> Microsoft ASP VSscript instead of Smalltalk?

The Smalltalk specific part of the site is Smalltalk powered.




From amitp at Xenon.Stanford.EDU  Sat Mar 10 20:53:56 2001
From: amitp at Xenon.Stanford.EDU (Amit Patel)
Date: 11 Mar 2001 01:53:56 GMT
Subject: gah! I hate the new string syntax
References: <mailman.983485923.1136.python-list@python.org> <97p7iq$kk2$1@nntp6.u.washington.edu> <97soak$qh9$1@nntp.Stanford.EDU> <97sudi01i26@news1.newsguy.com>
Message-ID: <98elrk$li8$1@nntp.Stanford.EDU>

 Alex Martelli <aleaxit at yahoo.com> wrote:
| "Amit Patel" <amitp at Xenon.Stanford.EDU> wrote in message
| news:97soak$qh9$1 at nntp.Stanford.EDU...
|     [snip]
| > When I see split(a,b), I think, "Oh, this isn't an inherent part of a
| > string-- *I* could write it, even though I didn't write the string
| > object."
| >
| > When I see b.split(a), I think, "This is a basic part of a string, and
| > I *cannot* write it."
| 
| Reasonable, but the latest assertion is a bit too strong -- tone down
| to "cannot write it without typeswitching or other potential inefficiencies
| since it doesn't do a general dispatch on b".

What I mean to say is that I cannot write new methods on strings.  So
my function mysplit(a,b) is written mysplit(a,b), because I am not
Guido.  But if I were Guido, then I could just declare that mysplit is
now a method on string objects, and we'd write b.mysplit(a).  So the
syntax that everyone has to use depends on who wrote mysplit and who
has control over the string class.  Since I don't control the string
class, I have to use a different syntax than someone who does have
control over the string class gets to use.  That seems okay if the
function is some essential part of that class.  But it seems "wrong"
if the function is there merely for convenience.

   - Amit





-- 
--
Amit J Patel, Computer Science Department, Stanford University
http://www-cs-students.stanford.edu/~amitp/


From xu.20 at nd.edu  Fri Mar  9 23:34:10 2001
From: xu.20 at nd.edu (Changsen Xu)
Date: Fri, 09 Mar 2001 23:34:10 -0500
Subject: class func & var name overlap
Message-ID: <3AA9AEC2.12BA1E95@nd.edu>

hi all,

I just now tried function name and variable name overlaping
in Python's script (1) and (2):

(1)
class x:
        x = 3
        def x(self): return 5
y=x()
print y.x


(2)
class x:
        x = 3
        def x(self): return 5
        x = 6
y=x()
print y.x


Then if I run 1), I got screen echo as
        <method x.x of x instance at 80e17d4>
while for 2), I got
        6

I'm here curious why Python doesn't even report an error
for this obvious mistake ?



From paulp at ActiveState.com  Thu Mar 15 08:09:22 2001
From: paulp at ActiveState.com (Paul Prescod)
Date: Thu, 15 Mar 2001 05:09:22 -0800
Subject: Raise exception in caller's stack frame
Message-ID: <3AB0BF02.BA0DB5E8@ActiveState.com>

Can someone remind me how to raise an exception in my caller's stack
frame? I'll add it as a recipe to the Python cookbook (or you can do it
if you want the fame and...well, just fame).
-- 
Take a recipe. Leave a recipe.  
Python Cookbook!  http://www.activestate.com/pythoncookbook



From dnew at san.rr.com  Fri Mar 23 18:35:28 2001
From: dnew at san.rr.com (Darren New)
Date: Fri, 23 Mar 2001 23:35:28 GMT
Subject: Python Script to dial a VPN connection, automate VPN routing table updates.
References: <FCNu6.66711$A6.13912607@news4.rdc1.on.home.com>
Message-ID: <3ABBDDC0.96162AA3@san.rr.com>

Warren Postma wrote:
> So after solving #2 above in Python, I wondered, does anyone know how, in
> Python win32 extensions, one might dial an RAS connection, and if it's
> possible to do so  through the win32 API somewhere or through COM? Then I
> could solve #1 above as well.

I had to do the same thing, but I use Tcl and NT4. I saved a shortcut to the
dialup, and I just execute "start yadda.rnk" to make the connect, giving a
delay of 5 seconds afterwards.


-- 
Darren New / Senior MTS & Free Radical / Invisible Worlds Inc.
San Diego, CA, USA (PST).  Cryptokeys on demand.
A million monkeys in a room with a million typewriters 
              will only yield half a million pregnant monkeys.


From bkline at rksystems.com  Tue Mar 27 16:34:20 2001
From: bkline at rksystems.com (Bob Kline)
Date: Tue, 27 Mar 2001 16:34:20 -0500 (EST)
Subject: order by instead of two queries
In-Reply-To: <3AC0FB19.7291F7DF@iems.nwu.edu>
Message-ID: <Pine.LNX.4.10.10103271629140.27947-100000@rksystems.com>

On Tue, 27 Mar 2001, Leonardo B Lopes wrote:

> I have some data in a table that I would like to obtain order by a
> specific instance of the data. ex:
> 
> Name:		Age:
> 
> Joe		50
> Jill		60
> Jack		65
> 
> What I would like to do is have a query that returns Jack first, then
> the regular name order. Something like:
> 
> SELECT Name FROM clients ORDER BY Name='Jack',Name
> 
> Can this be done with one query?

This can be done in standard SQL (using a CASE operation) but this is a
Python mailing list, not a SQL list.

-- 
Bob Kline
mailto:bkline at rksystems.com
http://www.rksystems.com




From sulfugor at rocketmail.cum  Mon Mar 12 11:20:19 2001
From: sulfugor at rocketmail.cum (sulfugor)
Date: Mon, 12 Mar 2001 16:20:19 GMT
Subject: Einstein's Riddle
References: <oNPq6.396644$ge4.138479566@news2.rdc2.tx.home.com> <98glrd$1si1a$1@ID-75892.news.dfncis.de> <3AABE712.499CCF4C@home.com> <98gp5b$1u0oi$1@ID-75892.news.dfncis.de> <98i5880pl2@news2.newsguy.com>
Message-ID: <984414019.468780@news.aei.ca>

Alex Martelli wrote in message <98i5880pl2 at news2.newsguy.com>...
>"Boris Popov" <no-spam-popov at bigserver.com> wrote in message
>news:98gp5b$1u0oi$1 at ID-75892.news.dfncis.de...
>> "Bruce Elrick" <belrick at home.com> wrote in message
>> news:3AABE712.499CCF4C at home.com...

>> P.S. What's emu ?
>
>A large flightless bird, whose meat is appreciated by more and
>more connaisseurs (it's red meat, reasonably tasty, quite soft,
>light in cholesterol & fats).  You can get a LOT of information
>about it from the American Emu Association, http://www.aea-emu.org/.
>(Particularly with the many recent scares about "mad cow's disease"
>aka BSE, emu meat has received a huge market boost -- it's now
>quite common for supermarkets around here to offer it, restaurants
>to feature it in several dishes, etc).
>
>Any association with Europe's "Economic and Monetary Union" is,
>one can hope, totally coincidental (some people WOULD argue that
>_that_ EMU _is_ 'owned' "by the German", but that's a base and
>unsubstantiated canard...).
>
>
I think we 're dealing with another kind of fish here : the lame duck
presidency of the EMU by germany.

sulfugor , pundit





>Alex
>
>
>




From root at rainerdeyke.com  Thu Mar  8 22:59:53 2001
From: root at rainerdeyke.com (Rainer Deyke)
Date: Fri, 09 Mar 2001 03:59:53 GMT
Subject: learning C
References: <mailman.983972829.23481.python-list@python.org> <983975751.18833@newsmaster-04.atnet.at> <3AA82422.53026D3D@cosc.canterbury.ac.nz>
Message-ID: <ZyYp6.376000$ge4.132139072@news2.rdc2.tx.home.com>

"Greg Ewing" <greg at cosc.canterbury.ac.nz> wrote in message
news:3AA82422.53026D3D at cosc.canterbury.ac.nz...
> Werner Schiendl wrote:
> >
> > The fine thing is: C++ still HAS the scalpel, so what?
>
> C++ is a Swiss army knife with both scissors and scalpel,
> and a nasty tendency for the scalpel to shoot out and
> stick you in the palm when all you were trying to do
> was use the scissors...

That's pretty much accurate.


--
Rainer Deyke (root at rainerdeyke.com)
Shareware computer games           -           http://rainerdeyke.com
"In ihren Reihen zu stehen heisst unter Feinden zu kaempfen" - Abigor




From dsavitsk at e-coli.net  Thu Mar  8 13:08:47 2001
From: dsavitsk at e-coli.net (dsavitsk)
Date: Thu, 8 Mar 2001 12:08:47 -0600
Subject: Another Tkinter question
Message-ID: <PUPp6.14231$tV.1448780@newsrump.sjc.telocity.net>

I have a small tkinter app that contains a (PMW) ScrolledFrame which in turn
has a grid of entry widgets (it sort of emulates a small spread sheet).
What I would like to do is add a new row of entry widgets while the thing is
running, and I'm not sure how to go about this as I've already .packed them
in.  (Also, if anyone knows a better way to do this than tons of entry
widgets that would be great.)

I can put some of my code up if need be.

thanks,

doug




From jmarshal at mathworks.com  Wed Mar 28 14:58:47 2001
From: jmarshal at mathworks.com (Joshua Marshall)
Date: 28 Mar 2001 19:58:47 GMT
Subject: Is inheritance broken?
References: <NDBBLAKACLGJJBDNGGCMKEINCCAA.jcc.ugm@ix.netcom.com> <mailman.985803830.15914.python-list@python.org>
Message-ID: <99tfpn$qn7$1@news.mathworks.com>

Skip Montanaro <skip at pobox.com> wrote:

> Someone with more Python years under their belt will have to answer why
> Guido chose depth first instead of breadth first search for methods when
> multiple inheritance is involved.  I'm pretty sure it's a choice that can't
> be changed now, however.  There's probably too much code that relies on the
> current semantics.

A breadth-first search would be too arbitrary.  Additions of
intermediate [super]classes in the inheritance hierarchy could
drastically change the result of a method lookup.

There are other solutions to multiple-inheritance method lookup
besides depth-first searching, but DFS is easy to explain/understand/
analyse.


From chris.gonnerman at usa.net  Fri Mar  9 08:44:33 2001
From: chris.gonnerman at usa.net (Chris Gonnerman)
Date: Fri, 9 Mar 2001 07:44:33 -0600
Subject: kbhit()  and  getch() question
References: <20010309071702.4626.qmail@web114.yahoomail.com>
Message-ID: <001901c0a89f$17d252a0$a100000a@local>

Dunno.  Check the version of your MSVCRT.DLL (mine is 6.10.8637.0).

Do Start, Find, Files or Folders.  Enter MSVCRT.DLL in Named: and your C:
drive
in Look In and do Find Now.  You may find more than one, but likely the one
in
c:\windows\system is the one being used.  (Do the find just in case there is
a
different version in your path... this could be your problem).

Right click on the file and select Properties, then click the Version tab.

You could also use SFC (the System File Checker) to install a "stock"
version
from your original master set.  This is not recommended unless you *know*
the
installed version is bad; however I might try it anyway since the previous
installed copy is backed up by SFC (so I could just put it back if I had
problems).

If this doesn't work we may need to try other voodoo.

----- Original Message -----
From: "Jeremy Reed" <jp_reed at yahoo.com>
To: "Chris Gonnerman" <chris.gonnerman at usa.net>
Sent: Friday, March 09, 2001 1:17 AM
Subject: Re: kbhit() and getch() question


> I am using python 2.0 and windows 98.  I checked the version just to be
> sure.  Do you have any idea what is causing the problem?
>
> --- Chris Gonnerman <chris.gonnerman at usa.net> wrote:
> > ----- Original Message -----
> > From: "Jeremy Reed" <jp_reed at yahoo.com>
> > Subject: kbhit() and getch() question
> >
> > > In a previous question, I asked how to get non-blocking input from
> > the
> > user.
> > > The answer proposed is as follows:
> > >
> > > import msvcrt
> > >
> > > ch = 0
> > > while ch != 'x':
> > >     if msvcrt.kbhit():
> > >         ch = msvcrt.getch()
> > >         print ch
> > >
> > > I have run into a few problems.  First of all, the kbhit() doesn't
> > seem to
> > > work.  I am unable to find a way for the function to return true.
> > [brutal hack]
> >
> > This code as given works for me on Win98SE, Python 2.0.  What is your
> > platform
> > and Python version?
> >
> >
>
>
> =====
>      ///////////  /////   /////  /////////////
>        /////      ///   ////    //  /////  //
>      //   //     ///////           /////
>      /////      ///   ////        /////
>  ///////////  /////    /////     /////
>
>
>
> __________________________________________________
> Do You Yahoo!?
> Get email at your own domain with Yahoo! Mail.
> http://personal.mail.yahoo.com/
>




From mwh21 at cam.ac.uk  Fri Mar  9 03:57:44 2001
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 09 Mar 2001 08:57:44 +0000
Subject: Static typing (was Re: Java guy interested in Python)
References: <003801c0a459$a29c97f0$63da1542@wisdom> <mailman.983688785.707.python-list@python.org> <slrn9a7gtl.ca6.qrczak@qrnik.zagroda> <mailman.984009683.19473.python-list@python.org>
Message-ID: <m3wv9z9v5j.fsf@atrus.jesus.cam.ac.uk>

Don Dwiggins <dwig at advancedmp.net> writes:

> So, are there any efforts in the Python community to provide
> something in this area?  I'm thinking of things like type inference,
> and of something like "permissive typing" (compiler/interpreter to
> programmer: "you don't have to tell me the type of a variable or
> argument, but if you do, I'll use that information to help you
> wherever I can").

Well, you could try ploughing through the archives of the types-sig.
Python's relentless dynamism makes it hard though, and I don't think
DC have the funds or the expertise the Lisp people did in the
seventies.

There are two sides to static typing, which on the types-sig got
dubbed "OPT" (make things quicker) and "ERR" (better compile-time
error checking), which are somewhat independent.  Oh, and there's
"DOC" (makes the meaning of code clearer) too, which is rather related
to "ERR".

I'm personally more interested in "OPT", but that's certainly not the
only reasonable view.

Also it's not clear that tighter type checking is necessarily a good
thing.  Consider the following function:

def write16(file, int):
    file.write(struct.pack("h",int))

If you type-annotate this, you might write (using entirely made up
syntax):

def write16(file : <file>, int : <int>):
    file.write(struct.pack("h",int))

but then what about passing in a StringIO object, or some user wrapper
of a file object?  What makes a "<file>"?  At the moment, it's just a
loose collection of methods, some of which are optional in some
circumstances.

now-if-you'll-excuse-me-i'm-late-for-lectures-ly y'rs
m.

-- 
  This makes it possible to pass complex object hierarchies to 
  a C coder who thinks computer science has made no worthwhile
  advancements since the invention of the pointer.
                                       -- Gordon McMillan, 30 Jul 1998


From dmasog at nortelnetworks.com  Fri Mar 16 12:31:26 2001
From: dmasog at nortelnetworks.com (Masog, David)
Date: Fri, 16 Mar 2001 12:31:26 -0500
Subject: PythonCOM connectableserver
Message-ID: <3AB24DEE.ED103E37@nortelnetworks.com>

I've used win32com\demos\connect.py as a starting point, but I'm trying
to write the connectable client portion in VC++ rather than Python.

When I comment out the line containing the self._BroadcastNotify the
script executes successfully(I made it return a value to the client).

When I leave the _BroadcastNotify line in I successfully catch the fired
event, but immedately afterward receive:
	Fatal Python error: PyEval_ReleaseThread: wrong thread state

I have seen some postings that refer to similar thread issues calling
Python callbacks from C but don't know how this would apply in this
case.
Could someone please shed some light on this issue or direct me towards
a different approach?

David Masog
dmasog at yahoo.com


From carribeiro at yahoo.com  Sun Mar 18 20:55:18 2001
From: carribeiro at yahoo.com (Carlos Ribeiro)
Date: Sun, 18 Mar 2001 22:55:18 -0300
Subject: Why "from __future__" stinks; a counter-offer
In-Reply-To: <LNBBLJKPBEHFEDALKOLCMEICJGAA.tim_one@email.msn.com>
References: <AD70ED184AFED311ADFD009027AC917F83617A@dmc1nt.zoho.com>
Message-ID: <5.0.2.1.0.20010318214456.022db7f0@mail.inet.com.br>

At 16:04 18/03/01 -0500, you wrote:
>Good -- then it will stick out <0.5 wink>.  Double double-underscore names
>are intended to "stick out", btw, and future_statements are intended to be
>temporary.  In that sense, the uglier they are, the more incentive to get rid
>of them ASAP.

Have you ever seen this happen in *any* sufficiently complex system? I'm 
afraid we'll be stuck with this "solution" for a long, long time.

> > Lastly and most importantly, it's not too late to do
> > something saner.
>
>Unclear, although in the absence of a competing PEP it's certain to remain as
>it is, and the last 2.1 beta release is scheduled for this coming Friday (as
>noted in PEP 226)

So thats the matter? Forgive me, I'm pretty new on this list and I may be 
missing something. The lack of a PEP... this sound too much like 
design-by-comitee to me. I agree with the *intent* of the PEP - to keep 
everything nice and documented, and to put some order into the process. 
However I feel that in this case it sounds as a lazy excuse for not 
thinking about a better to solve the problem. It's something like "the PEP 
is everything, lets follow the PEP as-is". Why is it so difficult to write 
or change the existing PEP?

>(...) future_statements are a temporary aid for
>use during a transition period.  "require", "using", "this_code_uses" and
>"feature" all send a wrong message in that respect. "pragma" is
>inappropriate because future_statements are about syntax and semantics, not
>pragmatics.

You are right - the message is important, and this is one of the most 
compelling reason to abandon "import from __future__". It does send the 
wrong message - in fact it does not make any sense (if you were talking 
about "import quantum computing", that's in the future :-). And don't let 
me get started on the relativity aspects, because the "future" is not an 
absolute, defined state - it's going to be differente for every release.

I think that some people got used to the proposed syntax, and as such they 
dont see anything wrong with it. On the other hand, words as "pragma", 
"directive" are not so "fashionable" - they sound too much like C or 
similar stuff. But *lots* of people are not comfortable with the "import 
from __future__ " construction. I'm one of them. And unfortunately I dont 
have the resources to grab the source and actually *do* something about it, 
this rant aside.

Part of my rants are about the name of the module - __future__.py is a bad 
and misleading name in my opinion. One alternative would be to call this 
module __compiler__.py. It's a better name because:

- it is related to the *function* carried out by this module (it is handled 
at compile time, and carries compile-time information)
- it has nothing to do with the perception of time
- it could export other symbols and mappings useful for the script about 
the compile-time environment.

I dont think that this is a real solution to the problem, but it makes it 
look a little better. It is also more consistent and has some prospect of 
being extensible. Also it's a easy change to make now without breaking any 
PEPs or the schedule.


OTHER ISSUES & WEIRD IDEAS

1) A different way to handle the problem is to make the Python compiler 
itself scriptable. However, this is also dangerous, and poses a much bigger 
set of problems. The basic concept is to have a __compiler__ symbol, pretty 
much like the __main__ symbol, except that this one is set only inside the 
compiler. All code inside an if __compiler__: block would be executed 
immediately by the compiler.

2) There is something else bugging me. It is not clear what is the effect 
that "from __future__" has on the compiler. PEP-236 states that this is 
handled at compile time. However the __future__.py module does exist and 
has some information in it. For python applications that get distributed, 
this has a lot of potential side effects. Someone may have the right 
version of the interpreter, but the wrong version of the __future__ module. 
(this is even worse if the compiler is scriptable...) What is going to happen?

3) In some time it may become be impossible to tell which subset of the 
language you are using - for instance, from a code snippet. This is *very* 
important to understand, because we are changing the semantics of the 
language, not the syntax, and it will not be obvious from the context of a 
snippet which "switch" has to be used. (We have to concede that this is 
going to be a problem whenever there is any semantic change - it is just 
impossible to avoid).


Ouch! That was a long rant... going to sleep now :-)



_________________________________________________________
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com




From aleaxit at yahoo.com  Sun Mar 18 15:17:25 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sun, 18 Mar 2001 21:17:25 +0100
Subject: Who's minister of propaganda this week?
References: <B98041329020E007.7DA84CAD9D0562AB.817BBC572F480869@lp.airnews.net> <98nu1m$12e@dispatch.concentric.net> <660ACCA1E4816F7E.C359CDF4168EF3FB.3B6B05538D619770@lp.airnews.net> <3AAFA13D.76E32C3F@divalsim.it> <98oa4o01i4v@news2.newsguy.com> <3AB0176D.8070903@yahoo.com> <3AB0C70A.C6A55DF5@ActiveState.com> <3AB40BE3.3080001@yahoo.com> <3AB417FD.1905AC3D@ActiveState.com> <mailman.984938364.6537.python-list@python.org>
Message-ID: <99351o189f@news2.newsguy.com>

"Michael Chermside" <mcherm_python at yahoo.com> wrote in message
news:mailman.984938364.6537.python-list at python.org...
    [snip]
> Well, most of my experience has been in staticly typed languages. I
> suspect that even in dynamically typed languages I tend to program
> in a "static typed fashion". Do you have any suggestions of examples
> of good code to read (or good books to read!) which would help me to
> realize what kinds of advantages I can get from dynamic typing?

I think the most instructive (parallel) reading might be "Design
Patterns" side by side with the "Design Patterns Smalltalk Companion".

I wish we had a similar 'companion' book for Python...


Alex





From dnew at san.rr.com  Wed Mar 14 12:20:32 2001
From: dnew at san.rr.com (Darren New)
Date: Wed, 14 Mar 2001 17:20:32 GMT
Subject: Who's minister of propaganda this week?
References: <B98041329020E007.7DA84CAD9D0562AB.817BBC572F480869@lp.airnews.net> <98nu1m$12e@dispatch.concentric.net>
Message-ID: <3AAFA861.C773D977@san.rr.com>

Phlip wrote:
> Ruby trumps Python in the simple matter of
> closures (imagine my shock when I discovered Python didn't have them). 

Can someone summarize the important difference between closures and object
instances? Other than convenience, I mean, it would seem that something like

class yadda:
  def __init__(self, a, b):
    self.a = a ; self.b = b
  def __call__(self, x, y):
    return (self.a + x * self.b * y)

would seem to be quite a lot like a closure with a and b bound and x and y
free?

(Pardon the python syntax errors.)

Now, coroutines, that's a bit different. :-)

-- 
Darren New / Senior MTS & Free Radical / Invisible Worlds Inc.
San Diego, CA, USA (PST).  Cryptokeys on demand.


From whisper at oz.net  Mon Mar 26 13:51:32 2001
From: whisper at oz.net (Dave LeBlanc)
Date: 26 Mar 2001 18:51:32 GMT
Subject: Is it possible to sort a dictionary?
References: <GEtv6.59583$Lm2.6378825@news0.telusplanet.net> <3ABE771B.7B9E1631@student.liu.se>
Message-ID: <99o33k$aq4$0@216.39.170.247>

How about:

# prints the elements of myDictionary in sort order
# error handling needed (myKeys != none for example)
myKeys = myDictionary.keys()
myKeys.sort()
for key in myKeys:
	print myDictionary[key]

Dave LeBlanc

On Mon, 26 Mar 2001 00:54:19 +0200, Rikard Bosnjakovic
<rikbo716 at student.liu.se> wrote:

>Victor Louie wrote:
>
>> I was just wondering if anyone knows how to sort a dictionary by
>> value...since the dictionary is <key: value> and where my value is a list.
>
>You can't sort the actual dictionary since they're mapping. You can,
>however, put the values in a list and sort the list. Something like
>this:
>
>    for key in thedict.keys():
>        sortedlist.append( (key, thedict[key]) )
>    sortedlist.sort()
>
>
>Test-run:
>
>>>> foo = {"foo":[1,2,3], "bar":[2,3,4], "gazonk":[3,4,5]}
>>>> foo
>{'foo': [1, 2, 3], 'gazonk': [3, 4, 5], 'bar': [2, 3, 4]}
>>>> sorted = []
>>>> for key in foo.keys():
>...   sorted.append( (foo[key], key) )
>... 
>>>> sorted.sort()
>>>> sorted
>[([1, 2, 3], 'foo'), ([2, 3, 4], 'bar'), ([3, 4, 5], 'gazonk')]
>
>
>
>-- 
>Rikard Bosnjakovic - http://a214.ryd.student.liu.se/cv/ - ICQ: 1158217
>
>Anyone sending unwanted advertising e-mail to my address will be
>charged $250 for network traffic and computing time. By extracting my
>address from this message or its header, you agree to these terms.



From howard at brazee.net  Wed Mar 14 09:59:13 2001
From: howard at brazee.net (Howard Brazee)
Date: Wed, 14 Mar 2001 07:59:13 -0700
Subject: Einstein's Riddle
References: <oNPq6.396644$ge4.138479566@news2.rdc2.tx.home.com> <wi6elw1pg5y.fsf@merlin.irisa.fr> <3AAE9730.5A7A4249@brazee.net> <slrn9atjsc.5j.scarblac@pino.selwerd.nl>
Message-ID: <3AAF8740.907533CC@brazee.net>

The original poster must have used a Spammer tool.  This thread is cross-posted
elsewhere as well, and I have seen one other coded solution, in Forth.

It would be fun to see a bunch of language solutions, first to solve this problem,
and then a more generic version to solve this class of problems.

Remco Gerlich wrote:

> Howard Brazee <howard at brazee.net> wrote in comp.lang.python:
> > This is cross-posted.  Please tell us which language you are using.
> >
> > Yoann Padioleau wrote:
> >
> > > "Steve Graham" <js.graham at home.com> writes:
> > >
> > > > Who would be interested in using his/her brain (and his computer) to solve
> > > > the following?
> > > here it is:
> > >
> > > %----------------------------------------------------------------
> > > % a sicstus program
>
>    hint ^^^^^^^
>
>
> Sicstus Prolog. Seems a nice choice for this sort of problem.
>
> --
> Remco Gerlich



From chris.gonnerman at usa.net  Thu Mar  8 23:49:38 2001
From: chris.gonnerman at usa.net (Chris Gonnerman)
Date: Thu, 8 Mar 2001 22:49:38 -0600
Subject: learning C
References: <mailman.983972829.23481.python-list@python.org> <983975751.18833@newsmaster-04.atnet.at> <3AA82422.53026D3D@cosc.canterbury.ac.nz> <20010308221519.A4367@harmony.cs.rit.edu>
Message-ID: <016f01c0a854$602aa2a0$a100000a@local>

Werner Schiendl wrote:
> 
> The fine thing is: C++ still HAS the scalpel, so what?

And that's all I use.

(I've BEEN trying to avoid this argument, but...)

Sad but true, the first OO language I tried.  It ruined me for OOP for
probably five years.  The next was Visual Basic, which at least was 
easier.

I have written more lines of C than I care to remember.  C is for CONTROL.
Nowadays I use what some believe is the best C++ compiler in the world,
GCC, and I still just write C.

Frankly, if C++ works for Werner that's his business, but I feel he has 
effectively called the C lovers on this list stupid.  Rudeness is for that
other comp.lang.p* list.

IMHO being told that your favorite compiled language should be deprecated
is rude.  It's not like it's InterCal, after all.  I don't like assembler
but I would never say it should be deprecated; I positively hate LISP but
you LISP fans shouldn't take that personally.

Do what works for you.





From fredrik at pythonware.com  Fri Mar  2 03:04:18 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 02 Mar 2001 08:04:18 GMT
Subject: How can I get the logged in users username on Win32?
References: <3a9f50fd.58862977@news.is.co.za>
Message-ID: <6uIn6.16285$Qb7.2659273@newsb.telia.net>

"nomad" wrote:
> After searching through the Python Docs, I found that on Un*x I could
> use getpass.getuser(), but there is no equivalent for the Win32
> platform.
>
> In other words, how can I get the username in Windows?

getpass.getuser()

(it works as long as USERNAME is set, which it usually is)

Cheers /F




From gmcm at hypernet.com  Sat Mar 17 13:59:19 2001
From: gmcm at hypernet.com (Gordon McMillan)
Date: 17 Mar 2001 18:59:19 GMT
Subject: Meta-class inheritance problem (Python2.0 bug?)
References: <6Kzs6.35062$v5.73895@newsfeeds.bigpond.com> <5nDs6.35198$v5.74638@newsfeeds.bigpond.com>
Message-ID: <90678B1B1gmcmhypernetcom@199.171.54.194>

[posted and mailed]

loriendNOSPAM at bigpond.com (Lorien Dunn) wrote: 

>I've located the point in the python 2.0 code where this occurs, so my 
>question is now: 
>
>        What needs to be changed to make this wotk with meta-classes as
>        well?: 
>
>ceval.c, line 1816 (reformatted for clarity):

[snip]

That's the same code as Python 1.5. Did this work before?

>From your previous post, it does not appear to be using the Don Beaudry 
hook, but then again, I don't know what type(MyBoostClass) is.

- Gordon


From steven at haryan.to  Fri Mar  2 23:51:42 2001
From: steven at haryan.to (Steven Haryanto)
Date: Sat, 03 Mar 2001 11:51:42 +0700
Subject: Editing Python source as XML?
In-Reply-To: <j44rxcjgnw.fsf@informatik.hu-berlin.de>
References: <mailman.983546704.10789.python-list@python.org>
Message-ID: <4.3.2.7.2.20010303115013.02c28410@bdg.centrin.net.id>

I am resending this to the list. Also thanks for the other
responses.

>Date: Sat, 03 Mar 2001 01:23:38 +0700
>To: Martin von Loewis <loewis at informatik.hu-berlin.de>
>From: Steven Haryanto <steven at haryan.to>
>Subject: Re: Editing Python source as XML?
>
>At 11:10 PM 3/2/2001, you wrote:
>>Steven Haryanto <steven at haryan.to> writes:
>>
>> > A recent experience of having to modify a long C source file has
>> > led me to wonder whether there are any tool/effort towards editing
>> > Python source code as XML or other form of structured document.
>>
>>Not sure *why* you'd want to do that, but it is certainly possible to
>>convert any script automatically to an XML document with the help of
>>the parser module.
>
>Creating a DTD that matches the Python grammar and editing the
>XML source code in a validating XML editor gives, among others,
>the comfort of on-the-fly syntax checking and eliminate the
>necessity to indent manually :-)
>
>> > What I would like is an editor that presents the source file as a
>> > tree outline, where I can collapse selected class or function
>> > definition body, certain long if's/for's, some section of code,
>> > view/hide the asserts, etc.
>>
>>PythonWin, and the Emacs Python mode, both can do that today.
>
><Slap in the forehead> Never knew that before! Guess I will
>be using Pythonwin a lot from now on :-)
>
>What about collapsing/expanding some logical section of code,
>is there another way beside:
>
>  # the following several pages of code does this & that
>  if 1:
>      # code
>
>      # code
>
>      # code
>
>Thanks,
>Steve
>
>>Regards,
>>Martin




From jmcmonagle at velseis.com.au  Wed Mar 21 20:13:31 2001
From: jmcmonagle at velseis.com.au (John McMonagle)
Date: Thu, 22 Mar 2001 11:13:31 +1000
Subject: Is there python code highlighting in emacs ?
Message-ID: <3AB951BB.F03055E7@velseis.com.au>

The hilit19 package in emacs provides keyword highlighting when editing
code.  For example, if you are writing in C, comments are italicised in
red, strings are in grey, import statements are in lavender, functions
are in bold-blue, etc.

Does anyone know if there is python support in the hilit19 emacs lisp
package ?




From ws-news at gmx.at  Wed Mar  7 07:10:05 2001
From: ws-news at gmx.at (Werner Schiendl)
Date: Wed, 7 Mar 2001 13:10:05 +0100
Subject: Can anyone recomend a good intoduction to C...
References: <mailman.983894288.19965.python-list@python.org> <983896344.701456@newsmaster-04.atnet.at> <m3lmqibw0a.fsf@atrus.jesus.cam.ac.uk> <983911749.217622@newsmaster-04.atnet.at> <UScp6.1097$y6.196347@ruti.visi.com> <983916572.286918@newsmaster-04.atnet.at> <9852b8015jm@news1.newsguy.com>
Message-ID: <983967284.246700@newsmaster-04.atnet.at>

> > That is of course a point, if you need to let the program inspect you
will
> > try to keep it as simple as possible.
> > But given the same functionality, I think a C++ program will not be more
> > complex.
>
> The program will not be, the language is.  It's a hard-to-judge tradeoff.
>

As to my knowledge, the application itself is inspected for safety critical
appliances.
And you must not change anything afterwards in the software or you need to
re-inspect the software.

What use would it be to check the language?

> > ...as long as it is well designed of course, nobody forces you to use
any
> > number of features at all possible
> >
> > don't you think so?
>
> I do not: C++ *forces* "any number of features" on you -- you can't
> easily subset it and ENSURE features outside of the subset are not
> intruding.
>

Why this?

You can use templates. But you can leave them at the side with no ill
effects.
You can use multiple inheritance. But, as well, you can leave it out if you
don't really need or like it.
You can use function/method overloading. But you can as well select a
distinct name.
You can use the new streams syntax introduced with C++, and it may have
additional benefits.
But nobody hinders you using the traditional APIs read, write, and friends.

> > Think about string operations: I think about 50 % of all security issues
I
> > read about are somehow related with a (string) buffer overflow. The
string
> > family of classes does not have that kind of problem.
>
> No, but it does have other issues -- e.g.,
>     std::string oneletter('A', 1);
> does NOT do what one might expect, nor does it give any warning
> in the implementations I have at hand (oneletter becomes a string
> of 65 characters all equal to '\001', not one of 1 character
> equal to 'A' -- oops, arguments the wrong way around...!-).
>

Yes, trading characters as integral types is maybe not a very good decision
from today's view on things.
However, I guess that would not do that much bad and at least not introduce
a security problem that turns out after years the software is in use.

> The trade-off is no doubt a win for std::string here -- it
> removes many more issues than it introduces.  But it's not
> so clear overall -- considering that you can't subset, not
> really.  I think that C++ (used properly, particularly with
> STRONG reliance on GOOD libraries) is a definite net win
> over C, but I can well see somebody might disagree, most
> particularly when security-crucial applications are being
> targeted.  (If I _could_ throw a switch to subset C++,
> _ensuring_ at least warnings for a zillion potential errors
> and complexities, the picture would of course change).
>

I think security is not a problem of C++...
You can easily write C or even Python code that does introduce security
problems.

- werner





From bobbo at milwpc.con  Fri Mar  9 23:14:40 2001
From: bobbo at milwpc.con (Elaine Sweeny)
Date: Fri, 9 Mar 2001 22:14:40 -0600
Subject: true newbie
Message-ID: <3aa9a9e9$0$65277$38c90ffe@news1.twtelecom.net>

can anyone help me find a tutorial for python thats for people just starting
out in programming. The one one that comes with it is ok but its still too
difficult.
        thanks




From akuchlin at mems-exchange.org  Thu Mar 22 16:24:26 2001
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 22 Mar 2001 16:24:26 -0500
Subject: Why Python/Jython?
References: <99cl9e$b2b$1@Merlin.et.bocholt.fh-gelsenkirchen.de> <mailman.985260089.28444.python-list@python.org> <1901D50DC30D30E4.16551C1AB0FB4363.C0D72E208759F8A7@lp.airnews.net> <99dkg301894@news1.newsguy.com>
Message-ID: <3dy9txzed1.fsf@ute.cnri.reston.va.us>

"Alex Martelli" <aleaxit at yahoo.com> writes:
> *blink* -- I'm hardly a Java-ite, myself, but
> "install and configure a minimal JDK" I would
> estimate as typically a 10-minute job (on, say,
> NT -- is it an order of magnitude harder on,
> say, Linux?!).  Maybe I misunderstand what
> you mean...?

A co-worker of mine once took an hour to install the JDK on Solaris;
most of this time was spent lost on Sun's horribly organized Web site
trying to find the exact version required.

--amk


From erikd at zip.com.au  Sat Mar  3 05:55:23 2001
From: erikd at zip.com.au (Erik de Castro Lopo)
Date: Sat, 03 Mar 2001 10:55:23 GMT
Subject: Any documentation about Python daemons?
References: <e0gq79.8og.ln@127.0.0.1>
Message-ID: <3AA0CBA5.30F81111@zip.com.au>

gradha at iname.com wrote:
> 
> Hello.
> 
> I would like to write now a program which acts as a server on the same
> machine for other clients in a Linux environment, something in the spirit
> of chronyd, if you know the program.

Sorry, i don't.

> However, I am new both to Python and process' intercommunication. Could
> you suggest me ways of doing this and/or point me to documentation
> explaining such communication methods? I first thought about nice pipes,
> but then I realized that the server program should be running always, not
> being allowed to fork, and clients could mantain multiple communication
> channels at the same time (ej: daemon + 2 clients). How could I do this?

This sounds like a job for sockets. The Python documentation has a very 
good section on the sockets module which I found pretty easy to follow.
There is also a more advanced module called SocketServer but I haven't
used that.

As for having multiple clients connect to one server you have a number
of options:

   - fork - only really of much use if the different server
     processes do not need to share very much data
   - threads - this overcomes the data sharing problems of
     forking but you will need to use locking to protect 
     shared data.
   - one process which uses select to manipulate multiple
     input and output connections

My application was a Tk based logging daemon which displays the 
output of a number of clients graphically. Because the clients 
all had to plot to the same Tk widgets I chode the multi-threaded 
model. I managed to get something working from scratch in about 
10 hours spread over three days.

Hope this helps,
Erik
-- 
-----------------------------------------------------------------
Erik de Castro Lopo     nospam at mega-nerd.com      (Yes its valid)
-----------------------------------------------------------------
"I think there is a world market for maybe five computers." 
 -- Thomas Watson, Chairman of IBM, 1943


From garyl at directhit.com  Tue Mar 13 17:09:47 2001
From: garyl at directhit.com (Gary)
Date: Tue, 13 Mar 2001 22:09:47 GMT
Subject: deleting IMAP messages using imaplib
Message-ID: <LUwr6.30$YP3.125512@typhoon.ne.mediaone.net>

Can anyone tell me how to delete messages in an IMAP mailbox? The doc says
delete is used to delete a mailbox, while expunge is used to clear out
deleted items. How do I delete a message itself?

Thanks for any advice,

-Gary




From gustav at morpheus.demon.co.uk  Tue Mar 13 16:31:47 2001
From: gustav at morpheus.demon.co.uk (Paul Moore)
Date: Tue, 13 Mar 2001 21:31:47 +0000
Subject: [Python-Dev] Revive the types sig?
References: <jeremy@alum.mit.edu> <15020.9404.557943.164934@w221.z064000254.bwi-md.dsl.cnc.net> <200103120711.AAA09711@localhost.localdomain> <mailman.984439040.24419.python-list@python.org>
Message-ID: <s44tatolisnipdfm4r4p1aqt2e09e458hk@4ax.com>

On Mon, 12 Mar 2001 18:15:15 -0500, barry at digicool.com (Barry A. Warsaw) wrote:
>>>>>> "UO" == Uche Ogbuji <uche.ogbuji at fourthought.com> writes:
>
>    UO> I know this isn't the types SIG and all, but since it has come
>    UO> up here, I'd like to (once again) express my violent
>    UO> disagreement with the efforts to add static typing to Python.
>    UO> After this, I won't pursue the thread further here.
>
>Thank you Uche!  I couldn't agree more, and will also try to follow
>your example, at least until we see much more concrete proposals from
>the types-sig.

My view on all this is that type annotations are likely to be counterproductive,
in the face of Python's high degree of polymorphism. For example, what is the
type of

def fac(n):
    if n <= 1: return 1
    return n * fac(n-1)

?

Obvious answer, integer -> integer. And that's likely what the programmer would
say.

But why can't this function be used with *any* type which supports comparison
with the number 1, multiplication, and subtraction of 1?

In this case, any type signature (at least, any one which is specified by the
programmer) is likely to *limit* the usability of the function, instead of
helping.

Avoiding this sort of unintended restrictiveness is likely to be hard without
type deduction technology, which I believe is a way beyond what is being
proposed...

Paul


From cobrien at Radix.Net  Fri Mar 16 12:03:48 2001
From: cobrien at Radix.Net (Cary O'Brien)
Date: 16 Mar 2001 12:03:48 -0500
Subject: Python style questions
References: <98tb91$9t9$1@saltmine.radix.net> <98tclh$ap6$2@lahti.tasking.nl>
Message-ID: <98th1k$ka2$1@saltmine.radix.net>

In article <98tclh$ap6$2 at lahti.tasking.nl>,
Dick Streefland <dick_streefland at tasking.com> wrote:
>cobrien at Radix.Net (Cary O'Brien) wrote:
>| 4. I need to change things like "88aa99bb" (hex string) into integers.  Should I
>| 
>|    a. use expr("0x"+s) 
>|    b. not use such things
>|    c. ???
>
>c. use string.atoi(s, 16), or for Python >= 2.0, the builtin int(s, 16).
>

Hmm...

>>> int('abcd',16)
Traceback (innermost last):
  File "<stdin>", line 1, in ?
TypeError: int requires exactly 1 argument; 2 given
>>> 

Is this a 2.0 feature?

>>> import string
>>> string.atoi('abcd',16)
43981
>>> 

Ok, that works.  Missing in my "Python essential reference" though.

Thanks...

-- cary


From donn at oz.net  Tue Mar 13 18:09:03 2001
From: donn at oz.net (Donn Cave)
Date: 13 Mar 2001 23:09:03 GMT
Subject: deleting IMAP messages using imaplib
References: <LUwr6.30$YP3.125512@typhoon.ne.mediaone.net>
Message-ID: <98m9af$ivm$0@216.39.151.169>

Quoth "Gary" <garyl at directhit.com>:
| Can anyone tell me how to delete messages in an IMAP mailbox? The doc says
| delete is used to delete a mailbox, while expunge is used to clear out
| deleted items. How do I delete a message itself?

It's only a flag, to identify the message that as you noted
will be actually deleted later by an expunge.

 ok, error = imapobject.store(number, 'FLAGS', '(\Deleted)')

	Donn Cave, donn at oz.net


From tim.one at home.com  Mon Mar 19 00:05:01 2001
From: tim.one at home.com (Tim Peters)
Date: Mon, 19 Mar 2001 00:05:01 -0500
Subject: PEP 42 and Py2.1
In-Reply-To: <3AB45C34.C8D1E82D@javanet.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCAEJCJGAA.tim.one@home.com>

[Raymond Hettinger]
> I've checked out the Py2.1 beta and my most desired small feature didn't
> make it.  To quote PEP 42:
>
>     Generalize eval() to accept any mapping objects for locals and
>     globals.
>
> Is it too late to get this one in?

Yes; it was too late the day after 2.1b1 was released (the first beta has the
last new feature).

> Is it difficult to implement?

Afraid that's the wrong question:  resistance has been grounded in "go fast"
reasons.  Using dicts directly avoids the runtime expense of indirecting thru
an abstract mapping-object layer.

> Is there anyway I can help get it done?

Well, implementing it would be a help!  Then you can quantify the slowdown
(if any -- who knows?).

personally-doubt-it-would-make-make-enough-difference-to-care-ly
    y'rs  - tim




From Marc.Poinot at onera.fr  Mon Mar 19 11:52:18 2001
From: Marc.Poinot at onera.fr (Marc Poinot)
Date: Mon, 19 Mar 2001 17:52:18 +0100
Subject: [ANNOUNCE] More Python tamers
Message-ID: <3AB63942.4D880593@onera.fr>

Hi you all,

please heck again http://lsm.abul.org
where we have now at least three more Python
tamers for our European Python Meeting:

Andy Robinson (UK) will talk about ReportLab
Christian Tismer, Dinu Gherman (DE) will talk 
about stackless python.
And maybe two other Python tamers soon...

Marcvs [alias 

Hey! Thou honest winegrower,
I was about to take this leaf,
But, aren't there indigestible
Reptiles amongst thy Wineyards ?
Let me hurriedly drop it,
Before the foe bites thy collarbone.

Oh no, please thou honest softgrower, 
Do not mistake my tree-frog catcher
With an undesirable neibourgh.
Let it walk amongst grapes, cipiou
and snares, tamers are surrounding us.
You can take your french leaf.

La balade of the reptile -
Michel de Tontaigne (1328)

]



From aleaxit at yahoo.com  Sun Mar 25 11:53:25 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sun, 25 Mar 2001 18:53:25 +0200
Subject: Referencing a container object's attributes
References: <200103250239.SAA10240@mail.novus-tele.net> <mailman.985488572.31554.python-list@python.org> <m3ae6azeuc.fsf@vole.object-craft.com.au>
Message-ID: <99l80p01enf@news1.newsguy.com>

"Dave Cole" <djc at object-craft.com.au> wrote in message
news:m3ae6azeuc.fsf at vole.object-craft.com.au...
    [snip]
> Tim> One question: does the circularity inherent in the above (in
> Tim> which a reference to the container object is placed in an
> Tim> attribute of each containee) have any potential for causing
> Tim> problems - particularly since I want to pickle the container
> Tim> objects?
>
> You should be able to pickle the container, but you will have to
> manually remove all of the circular references yourself when you want
> to free the object(s).
>
> In Python 2.x I understand that the garbage collector can detect and
> handle circular references.

...and in 2.1 you can have WEAK references in one of the two
directions (typically, for the child->parent backlink) to avoid
needing garbage collection, too.


Alex





From Bruce at EckelObjects.com  Thu Mar  1 16:30:10 2001
From: Bruce at EckelObjects.com (Bruce Eckel)
Date: Thu, 01 Mar 2001 13:30:10 -0800
Subject: Has anyone been able to convince Bruce Eckel to write
  "Thinking in Python"
In-Reply-To: <BD674CDC0A68C642B60038986FBB56DF015C47@mailsrv.ASPATECH.COM.BR>
References: <BD674CDC0A68C642B60038986FBB56DF015C47@mailsrv.ASPATECH.COM.BR>
Message-ID: <200103011330100430.0A95B767@mail.rdc1.sdca.home.com>

See:
http://www.mindview.net/Python/ThinkingInPython.html

Most current information can be found at:
http://www.mindview.net/notes.html
===================
Bruce Eckel    http://www.BruceEckel.com
Contains free electronic books: "Thinking in Java 2e" &
"Thinking in C++ 2e"
Please subscribe to my free newsletter -- just send any
email to:
join-eckel-oo-programming at earth.lyris.net
My schedule can be found at:
http://www.mindview.net/Calendar/Calendar.html
===================


*********** REPLY SEPARATOR  ***********

On 3/1/01 at 6:06 PM Chris Richard Adams wrote:

>If anyone has worked with Java - they would know how
Bruce's bible
>'Thinking in Java' has helped thousands of developers.
Could we be as
>fortunate with him considering a Python book?? Can only
pray and beg.






From Kilgallen at eisner.decus.org.nospam  Thu Mar  1 10:08:42 2001
From: Kilgallen at eisner.decus.org.nospam (Larry Kilgallen)
Date: 1 Mar 2001 10:08:42 -0500
Subject: [ANNOUNCE] pyAda, first public release
References: <3A9DB9E1.8DD0E36D@bigfoot.de> <Xvsn6.7$R4.940@news.clarkson.edu>
Message-ID: <UVZcPj21AeC9@eisner.encompasserve.org>

In article <Xvsn6.7$R4.940 at news.clarkson.edu>, "Brad Clements" <bkc at Murkworks.com> writes:
> I thought one of the big points of using Ada was its verifiability -- making
> it more suitable for complex, "large-scale" (i.e, expensive) projects, like
> missile defense.
> 
> Plugging in a "non-verifiable" script plug-in kinda defeats this purpose..
> doesn't it?

Indeed that is "one of the big points of using Ada", but there is no
rule preventing people form using Ada for other purposes too, and many
of us do so.

While Ada has many capabilities, one does not have to use them all in
every program.  I believe that in 13 years of using Ada I have never
used floating point.


From fredrik at pythonware.com  Tue Mar 13 18:42:56 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Tue, 13 Mar 2001 23:42:56 GMT
Subject: Python2C
References: <3AAE98EA.2754A358@lindrum.de>
Message-ID: <4gyr6.160$4d3.34825@newsb.telia.net>

Michael Lindrum wrote:
> I'm searching for the lib that is necessary to compile C-Code generated
> by Python2C.
> OS: SunOS
> Who knows where I can find it?

assuming you mean libpython, all you have to do is to
download and build python itself:

    http://www.python.org

(if that's not it, consider telling us what library you think
you're missing)

Cheers /F




From flognat at flognat.myip.org  Sat Mar 17 16:24:29 2001
From: flognat at flognat.myip.org (Andrew Markebo)
Date: 17 Mar 2001 22:24:29 +0100
Subject: Alternative documentation formats?? (chm and info)
Message-ID: <m3elvw2inm.fsf@master.athome>

Hi!

Uhm anyone out there thrown the Python docs through some magic filters
and got them chm and/or info-formatted??

If my mind doesn't fail me, chm comes with activestate, but it is a
small hassle installing, uninstalling it.. Maybe I don't need to
worry.. it is 'ok' to use? :-)

        /Andy

p.s. The main reason for info format is that the browser for this
formats are part of my fingers.. CHM because I have found it
convenient ;-)


From mal at lemburg.com  Wed Mar 14 15:57:00 2001
From: mal at lemburg.com (M.-A. Lemburg)
Date: Wed, 14 Mar 2001 21:57:00 +0100
Subject: Update: New mxODBC 2.0.0 Windows Installer Build
Message-ID: <3AAFDB1C.8DDE58EC@lemburg.com>

The Windows installer I announced two weeks ago (just before 
heading off to IPC9) had a bug which caused the .execute()
statements to hang.

The bug seems to have been caused by a problem in the distutils
build process on Windows. A clean recompile solved the problem.

Please redownload the egenix-mx-commercial-2.0.0.win32-py2.0.exe
installer from my Python Pages. I have deliberately not altered the
version number since nothing in the code itself has changed.

Sorry for any inconvenience this may have caused -- the distutils
build process is still very new (especially the distutils Windows 
installer written by Thomas Heller; thanks Thomas :). Feedback
about the Windows installer should be directed to the Python
distutils SIG mailing list (see python.org).

Thanks,
-- 
Marc-Andre Lemburg
______________________________________________________________________
Company & Consulting:                           http://www.egenix.com/
Python Pages:                           http://www.lemburg.com/python/



From jsoons at juilliard.edu  Mon Mar 19 12:13:50 2001
From: jsoons at juilliard.edu (Jonathan Soons)
Date: Mon, 19 Mar 2001 12:13:50 -0500
Subject: name of tmpfile
Message-ID: <B7C74E1A6FBAD411BFC60002B309DDF4E429@mailbox.juilliard.edu>

I am trying to find out the real name of

t = tempfile.TemporaryFile()

so that I can do

os.system('llp -d CHECKS -ofmt=checks <real_tempfile_name>')   # this is how
I print to a printer

According to the book, os.tmpfile() never gets a real name, so I can't use
that.

And p = tempfile.gettempprefix(t)

is supposed to put a reference to the real name in p. So

how do I pass the real name to os.system()?





From aleaxit at yahoo.com  Sat Mar 10 17:05:37 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sat, 10 Mar 2001 23:05:37 +0100
Subject: Python In A Nutshell
References: <nG5q6.54910$__6.11376601@typhoon.southeast.rr.com> <Ufcq6.140786$47.2078615@news.bc.tac.net> <Blfq6.60716$__6.11886688@typhoon.southeast.rr.com> <98c4ok$fhd$1@panix2.panix.com> <slrn9aj98s.ock.grante@isis.visi.com>
Message-ID: <98e8vi22sqq@news1.newsguy.com>

"Grant Edwards" <grante at visi.com> wrote in message
news:slrn9aj98s.ock.grante at isis.visi.com...
    [snip]
> "Sorry it's so long, I didn't have time to write it short."
>
> No idea where that quote is from, but I noticed in grad school that

Pascal (as in, Blaise, the mathematician and philosopher -- not the
programming language later named after him!-), in one of his letters.


Alex





From jkahanpa at pcu.helsinki.fi.invalid  Mon Mar  5 06:17:31 2001
From: jkahanpa at pcu.helsinki.fi.invalid (Jere Kahanpaa)
Date: 5 Mar 2001 11:17:31 GMT
Subject: gah! I hate the new string syntax
References: <mailman.983485923.1136.python-list@python.org>
Message-ID: <97vskb$ctd$1@oravannahka.helsinki.fi>

Sean 'Shaleh' Perry <shaleh at valinux.com> wrote:
: return ";".join(["%s=%s" % (k, params[k]) for k in params.keys()])

: every day python seems to be moving closer to the line noise aspect of
: coding.

Agreed. Luckily anough I found Python in the clean, glorious days of 1.5.2 
- the Unicode support in >1.6 was the *only* reason for me to upgrade as 
XML processing in a multi-OS, multi-coding environment is really tricky 
without Unicode.

Jere Kahanp??
-- 
Lord, make my words as sweet as honey, for one day I may have to eat them


From tchur at optushome.com.au  Mon Mar 26 15:09:45 2001
From: tchur at optushome.com.au (Tim Churches)
Date: Tue, 27 Mar 2001 06:09:45 +1000
Subject: warrior geeks
References: <3ABF72F1.98CB78F@sage.att.com>
Message-ID: <3ABFA209.CF27BC20@optushome.com.au>

Garry Hodgson wrote:
> 
> a funny thing happened at IPC9.  a bunch of were sitting in the
> lounge (bar) talking, and one person mentioned that he was going
> to study kendo while in japan this year.  i asked if he had previous
> martial arts experience, and he said yes, he'd studied judo.
> it then turned out that in this group of maybe 6-7 people, we had
> this judo/kendo student, another who's done karate and now kickboxing,
> another who did karate, myself (okinawan karate, aikijutsu and others),
> and a pa qua practitioner.  at the next table sat another tae kwon
> do/aikido
> student.
> 
> it seemed like an odd coincidence, but it made me wonder: is there an
> unusually high percentage of MA practitioners among the python
> community?
> or the open source community?  or software in general?

I'll use our time zone advantage to jump in early on this thread and
opine that it is way off-topic. Not that I am opposed to
marital...whoops...martial arts (although they are not my cup of tea),
but I am concerned that if this threads continues, sooner or later
someone will mention the whole "Geeks with Guns" thing that Eric S.
Raymond promotes (see http://www.tuxedo.org/~esr/geeks-with-guns/ or for
even worse IMHO see http://www.tuxedo.org/~esr/guns/ ), especially given
ESR's apparent fondness for Python. That sort of thing gives me a fit of
the shudders (playing with guns, not ESR's fondness for Python). I would
have thought that if male hackers were looking for a hobby which
involves phallic-shaped objects and chemically-propelled projectiles,
then model rocketry would be just the ticket.

There, that should put an immediate end to this very off-topic thread...

Tim C
Sydney, Australia
PS Damn! I was cooking porridge (oatmeal) for breakfast and became so
carried away composing this message that I completely forgot to stir the
pot...



From sholden at holdenweb.com  Wed Mar 28 22:40:27 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Wed, 28 Mar 2001 22:40:27 -0500
Subject: Which Python version is better on Win32
References: <20010329.3014838@mis.configured.host>
Message-ID: <rcyw6.8737$m04.667866@e420r-atl1.usenetserver.com>

"TM" <tm at kernelconsult.com> wrote in message
news:20010329.3014838 at mis.configured.host...

> I've seen recently that yhere were several releases of Python for the MS
> Windows platform:
> ActivePython is one of them there's also PythonWin which is not branded,
> beOpenPython and maybe others.
>
> Which one would you advise me to use and why?

BeOpen Python is the standard 2.0 release, as built by the development team
while they were working at BeOpen (which looks, by the way, as though it
might now BeClosed).

PythonWin is actually just one component of ActiveState's ActivePython
distribution. Unless anyone can tell you any better, it bundles BeOpen
Python with a huge set of add-on goodies devised and produced mostly by Mark
Hammond.

IMHO you would be best advised to go for ActivePython unless you
specifically want a small installation with just the Python language core on
it.

regards
 Steve





From gunturu_neelima at rediffmail.com  Tue Mar  6 02:34:17 2001
From: gunturu_neelima at rediffmail.com (neelima  g)
Date: 6 Mar 2001 07:34:17 -0000
Subject: (no subject)
Message-ID: <20010306073417.2546.qmail@mailweb32.rediffmail.com>

after creating the empty database,how to enter the values into it  which is having morethan two columns.
                       confirm 881077


_____________________________________________________
Chat with your friends as soon as they come online. Get Rediff Bol at
http://bol.rediff.com







From jmarshal at mathworks.com  Fri Mar 23 09:45:35 2001
From: jmarshal at mathworks.com (Joshua Marshall)
Date: 23 Mar 2001 14:45:35 GMT
Subject: New PEP: The directive statement
References: <mailman.985085006.24236.python-list@python.org> <slrn9bk7fk.fnm.scarblac@pino.selwerd.nl> <j4k85gkiye.fsf@informatik.hu-berlin.de>
Message-ID: <99fnif$gsq$1@news.mathworks.com>

Martin von Loewis <loewis at informatik.hu-berlin.de> wrote:
> scarblac at pino.selwerd.nl (Remco Gerlich) writes:

>> First the inconsistency: the syntax definition in (C) claims that some
>> restrictions on the placement of directive may be in place, depending on
>> the directive. However, (D) notes that directives must appear as the first
>> non-string token in a file, for backwards compatibility. So in practice,
>> directive can't be used anywhere in a file.

> That's not inconsistent. There is only one directive defined so far,
> to activate nested scopes, and it must appear at the top of the file.
> In some later release of Python, directive will become a proper
> keyword, at which time introducing directives with less strict
> placement rules is possible.

Please, no more keywords.  It's unnecessary in this case.
How about an interpreter switch:

  python -directive nested_scopes fu.py


From tjg at exceptionalminds.com  Fri Mar 16 17:49:26 2001
From: tjg at exceptionalminds.com (Timothy Grant)
Date: Fri, 16 Mar 2001 14:49:26 -0800
Subject: Type mismatchings and the rank newbie
In-Reply-To: <ylws6.2639$227.274111@newsread2.prod.itd.earthlink.net>; from jgbrawley@earthlink.net on Fri, Mar 16, 2001 at 10:21:18PM +0000
References: <ylws6.2639$227.274111@newsread2.prod.itd.earthlink.net>
Message-ID: <20010316144926.P3109@trufflehunter.avalongroup.net>

On Fri, Mar 16, 2001 at 10:21:18PM +0000, John Brawley wrote:
> Rank newbie here, trying to do something specific.
> I need to use a set of numbers (say, 1 to 25), and create 25 lists from the
> numbers (1 thru 25), which lists have a character "P" and a number (one of
> the 25).
> 
> The lists I want look like :
> P1=[x, y, z]
> P2=[x, y, z]
> P3=[x, y, z]
> (etcetera down to P25)

I'm not completely sure I understand the question, but you
might try something like this...

exec 'P%s=[%s,%s,%s]' %(1,'x','y','z')

This will create and execute a python statement that looks like
this...

P1=[x,y,z]

However, I do believe that you best bet would be to use a
dictionary...

P = {}
P[1] = [x,y,z]
P[2] = [x,y,z]
P[3] = [x,y,z]

-- 
Stand Fast,
    tjg.

Timothy Grant                         tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.         <><       (503) 246-3630
>>>>>>>>>>>>>Linux, because rebooting is *NOT* normal<<<<<<<<<
>>>>This machine was last rebooted:  59 days  3:00 hours ago<<



From aahz at panix.com  Mon Mar 26 10:26:48 2001
From: aahz at panix.com (Aahz Maruch)
Date: 26 Mar 2001 07:26:48 -0800
Subject: So what's wrong with __future__?  (Was Re: Why "from __future__" stinks ...)
References: <slrn9bdka2.3j6.hzhu@rocket.knowledgetrack.com> <LNBBLJKPBEHFEDALKOLCAEDKJIAA.tim.one@home.com> <mailman.985568441.15878.python-list@python.org>
Message-ID: <99nn3o$p0v$1@panix6.panix.com>

In article <mailman.985568441.15878.python-list at python.org>,
Timothy Grant  <tjg at exceptionalminds.com> wrote:
>
>OK, here's where I show my ignorance. Why did it have to be 100%
>backward compatible? Things in this industry get changed every day
>that are not 100% backward compatible. I tend to think that this was
>one of Microsoft's biggest problems with their OS. They insisted on
>backward compatability. I often wonder how much long-term pain (crappy
>OS stability) could have been avoided if MS had convinced their users
>to accept a bit of short term pain when Win95 was introduced.

Note that we're talking about two layers of compatibility here:

* The hack of announcing an incompatible language change

* The incompatible language change itself

If you say that the hack should itself be incompatible, there's no point
in doing the hack at all.  The hack was developed to postpone (not
eliminate) the need to deal with the incompatible language change.

The problem is that part of the point of Python is to be able to pick up
a random piece of source code and use it.  *BUT*, it can be hard to tell
what versions of Python the code can run under.

Overall, this is an ugly solution, but given the axioms that Tim was
working with, probably the best possible solution in a short period of
time.  Change the axioms and you get a different solution....
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

"I won't accept a model of the universe in which free will, omniscient
gods, and atheism are simultaneously true."  -- M


From bwalker at earth1.net  Sat Mar 31 16:10:28 2001
From: bwalker at earth1.net (Bill Walker)
Date: Sat, 31 Mar 2001 16:10:28 -0500
Subject: very basic, how do you find square roots?
References: <9a408a$1mpp$1@raewyn.paradise.net.nz>
Message-ID: <tcchks47dikbf6@corp.supernews.com>

Remember that Python has the Complex math type and that the general solution
of a quadratic can involve complex quantities. You've chosen a good language
to develop that program. Import the math module and use the complex square
root.
Good luck.


"johnston112" <gordonj at paradise.net.nz> wrote in message
news:9a408a$1mpp$1 at raewyn.paradise.net.nz...
> Yes, that is my question :)
> I am trying to make a program that dose the quadratic equation.
>
> johnston112
>
>




From cfelling at iae.nl  Fri Mar  9 15:30:54 2001
From: cfelling at iae.nl (Carel Fellinger)
Date: 9 Mar 2001 21:30:54 +0100
Subject: cdrecord output capture
References: <Pine.LNX.4.30.0103081022410.23372-100000@alpha.cerm.unifi.it>
Message-ID: <98behu$3ol$1@animus.fel.iae.nl>

Morelli Enrico <morelli at alpha.cerm.unifi.it> wrote:
> Dear all,

> I am a new user of python.

> I trying to write a small python program under linux to create and
> save ISO files using mkisofs, cdrecord and dialog.

> I'm able to capture mkisofs output using popen2, but I'm not able to do
> the same thing with cdrecord. I had see the source code of eroaster
> (graphic python program that use mkisofs and cdrecord) but this program
> use a difficult way for me, I don't understand all code.

Can you show some code?

doing the obvious seems to work here:

>>> import os
>>> os.popen('cdrecord -scanbus').readlines()
['Cdrecord 1.9 (i586-pc-linux-gnu) Copyright (C) 1995-2000 J\366rg Schilling\012', 'Linux sg driver version: 2.1.39\012', "Using libscg version 'schily-0.1'\012", 'scsibus0:\012', "\0110,0,0\011  0) 'PLEXTOR ' 'CD-ROM PX-40TS  ' '1.11' Removable CD-ROM\012", '\0110,1,0\011  1) *\012', '\0110,2,0\011  2) *\012', '\0110,3,0\011  3) *\012', '\0110,4,0\011  4) *\012', '\0110,5,0\011  5) *\012', '\0110,6,0\011  6) *\012', '\0110,7,0\011  7) *\012', 'scsibus1:\012', "\0111,0,0\011100) 'PLEXTOR ' 'CD-R   PX-W8432T' '1.07' Removable CD-ROM\012", '\0111,1,0\011101) *\012', '\0111,2,0\011102) *\012', '\0111,3,0\011103) *\012', '\0111,4,0\011104) *\012', '\0111,5,0\011105) *\012', '\0111,6,0\011106) *\012', '\0111,7,0\011107) *\012']
>>> os.popen2('cdrecord -scanbus')[1].readlines()
same output

-- 
groetjes, carel


From aahz at panix.com  Fri Mar 16 18:15:09 2001
From: aahz at panix.com (Aahz Maruch)
Date: 16 Mar 2001 15:15:09 -0800
Subject: Killing threads
References: <mailman.984769881.20374.python-list@python.org>
Message-ID: <98u6pt$6j$1@panix6.panix.com>

In article <mailman.984769881.20374.python-list at python.org>,
Dave Brueck <dbrueck at edgix.com> wrote:
>
>Or just a simple flag. Here's a class I use in designs that use work flows
>(work objects pulled off one queue, processed, and moved onto the next
>queue). If you just need an always-processing background thread you can
>eliminate all the stuff with the semaphore.

I'm confused.  Is there some reason you're not using Queue.Queue()?
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

Three sins: BJ, B&J, B&J


From brian at brian.cbs.umn.edu  Thu Mar 22 10:54:40 2001
From: brian at brian.cbs.umn.edu (Brian Langenberger)
Date: Thu, 22 Mar 2001 15:54:40 +0000 (UTC)
Subject: Good Editor?
References: <3ab9a32b$1_1@news2.one.net> <3AB9FAC9.97EA420E@student.liu.se> <slrn9bjv7m.4qv.grey@teleute.dmiyu.org>
Message-ID: <99d780$l2g$1@laurel.tc.umn.edu>

Steve Lamb <grey at despair.rpglink.com> wrote:
: On Thu, 22 Mar 2001 14:14:49 +0100, Rikard Bosnjakovic
: <rikbo716 at student.liu.se> wrote:
:>I prefer Emacs and python-mode, it's really neat. But I know a lot of
:>guys who won't touch Emacs even with a long stick :)

: <ObEMACSJoke>
: That's because Emacs won't fit on a long stick of RAM.  Everyone knows it
: takes 3 long sticks of RAM before Emacs will run.
: </ObEMACSJoke>

(E)ventually (M)alloc()s (A)ll (C)omputer (S)torage

I've spent countless hours editing Python code in Emacs, really.
Between that and the interactive interpreter, I simply don't
use anything else.



From greybria at direct.ca  Wed Mar 21 23:27:18 2001
From: greybria at direct.ca (Brian & Colleen)
Date: Wed, 21 Mar 2001 20:27:18 -0800
Subject: Stupid String Question
References: <3AB8327C.493C70A5@direct.ca>
Message-ID: <3AB97F25.A61F4688@direct.ca>

Thanks guys. I knew it was way too easy. As soon as I looked back in my
1.5.2 Quick Ref, there it was. It makes a big difference when you know
what you're looking for.

--
 Brian Smith
 greybria at direct.ca
 http://mypage.direct.ca/g/greybria





From Dan.Rolander at marriott.com  Fri Mar 23 08:57:42 2001
From: Dan.Rolander at marriott.com (Rolander, Dan)
Date: Fri, 23 Mar 2001 08:57:42 -0500
Subject: Standalone executables
Message-ID: <6176E3D8E36FD111B58900805FA7E0F80CCF6472@mcnc-mdm1-ex01>

Yes, I also use Gordon's Installer with wxPython and it works fine. I see
this misconception posted all the time, and I suppose it's because Installer
has some built-in support for Tkinter, which isn't required for wxPython.

Although py2exe is easier to use, the last time I tried it it still didn't
support pyXML which I need, so I'll continue to use Installer.

Dan

-----Original Message-----
From: gmcm at hypernet.com [mailto:gmcm at hypernet.com]
Sent: Thursday, March 22, 2001 3:22 PM
To: python-list at python.org
Subject: Re: Standalone executables


[posted and mailed]

Don Tuttle wrote: 

>"Dave LeBlanc"
>> I found this on c.l.p.announce:
>> The Linux version of Installer for Python 2 is here:
>>  http://www.mcmillan-inc.com/installer_dnld.html
>
>Mark, judging from comments posted here,  your main issue of which
>installer/pseudo compiler to use will depend on you choice of GUI. 
>py2exe = wxPython.  Tkinter = Gordon's Installer.  So as tools stand
>now, the question is which is more important,  using wxPython, or
>distributing to both Windows & Linux. (Or maybe you could port py2exe to
>Linux for Thomas ;-)

I use my installer with wxPython all the time (I rarely use TK). 
It doesn't need any special support (well, there was a time when
what Robin was doing and what I was doing conflicted, but it
seems to be OK again now).

On Linux, the packager may have to take care of wxWindows on the
target machine (if it's not statically linked); it's only on Windows 
that the intaller will trace the (system) dependencies of extension 
modules.

- Gordon
-- 
http://mail.python.org/mailman/listinfo/python-list



From dmaas at dcine.com  Sat Mar 10 15:34:45 2001
From: dmaas at dcine.com (Dan Maas)
Date: Sat, 10 Mar 2001 20:34:45 GMT
Subject: Problems writting a wrapper around linux' ssh command
References: <31rd89.lki.ln@127.0.0.1>
Message-ID: <Fdwq6.236804$o91.30298800@typhoon.nyroc.rr.com>

> I am trying to write a python wrapper around ssh. The
> last and only problem I have is password verification.
> I try to popen2.popen3 the ssh command, but no matter
> what I write to the input file, the ssh command ignores it
> and keeps requesting *keyboard* input.

Right, ssh tries to bypass stdin and read the password from its controlling
terminal. You can avoid this by creating a virtual terminal and attaching to
it before executing ssh. The standard 'pty' modules does it for you:

import pty, os

[ssh_pid, ssh_fd] = pty.fork()

if ssh_pid == 0: # child
     os.execvp('ssh', ['ssh', hostname])

... and now you can talk to the ssh process via ssh_fd.

Be aware that the pty module is somewhat of a hack - the process of setting
up a virtual terminal is very platform-dependent, and Linux 2.4 introduced a
new method that pty doesn't use yet. It should work fine on most Linux
systems though.

Regards,
Dan




From phlip_cpp at my-deja.com  Wed Mar 14 23:39:34 2001
From: phlip_cpp at my-deja.com (Phlip)
Date: 15 Mar 2001 04:39:34 GMT
Subject: Who's minister of propaganda this week?
References: <B98041329020E007.7DA84CAD9D0562AB.817BBC572F480869@lp.airnews.net> <98nu1m$12e@dispatch.concentric.net>
Message-ID: <98ph26$5a5@dispatch.concentric.net>

Someone said this today, but something seems to have happened to 
it:

> an object is data with code attached
> a closure is code with data attached

Correct. But a closure declares implicitly, on the spot. That's why you 
don't have to interrupt your own flow to orchestrate one from scratch, as 
an object, the way you would in Python or (gods help us) C++.

-- 
  Phlip                          phlip_cpp at my-deja.com
============== http://phlip.webjump.com ==============
  --  "Probably one of the toughest times in anyone's life is when
       you have to kill someone you love because they'r the Devil"
                                     --Emo Phillips  --


From tim.one at home.com  Fri Mar  2 15:40:41 2001
From: tim.one at home.com (Tim Peters)
Date: Fri, 2 Mar 2001 15:40:41 -0500
Subject: Bug in Py_Finalize() ?
In-Reply-To: <3aa0c6e5.809467@news.newsguy.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCEEIDJDAA.tim.one@home.com>

[Owen F. Ransen]
> I tried to report this bug to sorgeforge, signed up, but
> it kept on giving me page not found errors after I logged
> in....

Bless you for trying, anyway.

> Trying to find a bug in my C extension I ended up writing
> this loop:
>
>     for (UINT i = 0 ; i < 10 ; i++) {
>         Py_Initialize();
>         Py_Finalize() ;
>     }
>
> which crashes Python when i=8 with the message:
> Windows 32, VC++ program.
>
> Fatal Python error: UNREF invalid object
>
> My understanding is that I can use Py_Initialize() and
> Py_Finalize() in pairs to free up memory and to start
> with a clean interpreter each time I need it. And that
> is why I'd like to use Py_Finalise....

Please try this again with the lastest CVS version, or pick up 2.1b1 (we'll
be releasing that later today).  Something "very much like this" got fixed
last month; here's Guido's checkin msg:

----------------------------
2001/02/20 21:43:24 gvanrossum Python/import.c
The code in PyImport_Import() tried to save itself a bit of work and
save the __builtin__ module in a static variable.  But this doesn't
work across Py_Finalise()/Py_Initialize()!  It also doesn't work when
using multiple interpreter states created with PyInterpreterState_New().

So I'm ripping out this small optimization.

This was probably broken since PyImport_Import() was introduced in
1997!  We really need a better test suite for multiple interpreter
states and repeatedly initializing.

This fixes the problems Barry reported in Demo/embed/loop.c.
----------------------------

I'd try it for you right now, but I'm in the process of building the 2.1b1
Windows installer and won't have the bandwidth until that's done ...

just-enough-spare-cycles-to-write-email-between-reboots<wink>-ly y'rs  - tim




From dale at out-think.NOSPAMco.uk  Thu Mar 22 17:16:29 2001
From: dale at out-think.NOSPAMco.uk (Dale Strickland-Clark)
Date: Thu, 22 Mar 2001 22:16:29 +0000
Subject: Automatic install of ActivePython
References: <elikbtker40ll94076q82tugve0jaqh264@4ax.com> <mailman.985296990.15952.python-list@python.org>
Message-ID: <vcukbt0pqrb24rtk71b5bukq02tps8hgpt@4ax.com>

Paul Prescod <paulp at ActiveState.com> wrote:

>ActivePython has a pretty sophisticated ability to do command-line
>installs. This is documented in the release notes which should be in
>your "start menu". Yes, you can control the target directory.
>
>Remember: the install can download things off of the web. You should
>turn those features off if you are going to be doing batch installs in
>places without net connectivity.

Got it.

That's just what I was looking for.

Thanks.
--
Dale Strickland-Clark
Out-Think Ltd
Business Technology Consultants


From Eugene.Leitl at lrz.uni-muenchen.de  Fri Mar  9 10:05:25 2001
From: Eugene.Leitl at lrz.uni-muenchen.de (Eugene Leitl)
Date: Fri, 9 Mar 2001 16:05:25 +0100 (MET)
Subject: architectural issues cot'd; which http server to embed?
Message-ID: <Pine.GSO.4.03.10103091554060.26053-100000@sun1.lrz-muenchen.de>

I haven't received much feedback on the last architectural query yet,
as most of you people are probably on the conference. Still, here's 
goes.

I've been fooling around with diverse http servers written in plain (i.e.
containing no components written in other languages) Python (specifically,
CPython. Perhaps threaded Python, if this doesn't limit my options in
regard to portability).
 
I have (in my head, at least) a multi-component (XML-talking server-client
(sockets)) architecture driven via a web front (browser plugin or Java
client) end by user queries.  This will work transparently, whether run
locally, or on another machine on the network. I don't need the
multicomponent part so badly right now, but this is needed for a future
framework, of which this will be an integrated component.

It has to be able to run on every target which has GNU tools and TCP/IP
running on it. In practice, this will mean diverse flavors of *nix, and
Redmondware. The ideal delivery form would be a bunch of binaries bundled
in an archive, and a (frozen) GUI-based (with a shell fallback option for
the hard cases) installer written in Python. I can't install my own Python
on the client machine, and there's not enough time to do it properly (I
have certainly found enough interesting options), so I'm going to tolerate
some amount of bloat in the distribution, at least for time being.

Right now I'm researching which http server to embed (using freeze) into
the front-end. The performance requirements are quite negligable, but it
should be better non-blocking (Medusa alone won't do, will Medusa/bobo
combo? Zope is way too fat, CGIHTTPServer.py would do nicely, but it
doesn't seem to be able to support server-side Python embedding into HTML
when being frozen), as queries can take a while, and it might see more
than one user, or several queries from the same user. Otoh, I'd rather not
call multi-MByte frozen binaries via cgi-bin. It would be nice to have a
solution which allows one to embed Python code into HTML, which can get
interpreted by the Python instance which is present in the webserver, or
Python packaged in a shared library.

Another alternative I haven't researched yet is extending Python with a
minimal, well-debugged off-shelf webserver written in C. This would be
probably more palatable with passing Python scripts from cgi-bin and/or
Python embedding into server-side HTML. Discussions and pointers to code
are very welcome.

Resistancy to memory leaks is another design facet. If I can solve the
problem of long-running Python instances slowly draining memory
(considering my limited amount of clue a very real factor) by paying with
startup overhead by launching them at every query, that appears tolerable.
Even if it involves calling multi-MByte monster binaries, I'd have to bite
the bullet (I'd rather not, of course). How bad is the memory draining
issue, given debugging tool and Python2.0 cycle-elimination during gc?
Should I contemplate using a better gc?

Does this make sense, or I am missing some gotcha, or obvious alternative?
Sorry if I sound confused, I probably am. This project is rather ambitious
for my current skills, but I'd rather not tackle it in C. I'd certainly be
unable to handle this in C on the time schedule (2..4 months), so there's
not much choice, anyway.

TIA,
-- Eugene




From deblNonOspammy at theworld.com  Mon Mar  5 14:13:50 2001
From: deblNonOspammy at theworld.com (David Lees)
Date: Mon, 05 Mar 2001 19:13:50 GMT
Subject: Trouble running a Grayson Example
Message-ID: <3AA3E573.6DBE6D2E@theworld.com>

I am trying to run example 8.9 (Example_8_9.py) from Graysons book,
under Python 2.0 and get the following errors.  Can someone tell me what
is going on?  Perhaps this has to do with changes between 1.5.2 and 2.0
Python?

Thanks in advance.

david lees


Traceback (most recent call last):  File
"d:\Python20\Pythonwin\pywin\framework\scriptutils.py", line 301, in
RunScript
    exec codeObject in __main__.__dict__
  File "E:\Grayson1\Chapter08\Example_8_9.py", line 79, in ?
   ddnotebook = DDNotebook()
  File "E:\Grayson1\Chapter08\AppShell.py", line 55, in __init__
  self.__createInterface()
  File "E:\Grayson1\Chapter08\AppShell.py", line 282, in
__createInterface
   self.createInterface()
  File "E:\Grayson1\Chapter08\Example_8_9.py", line 75, in
createInterface
    self.createNotebook()
  File "E:\Grayson1\Chapter08\Example_8_9.py", line 26, in
createNotebook  Pmw.NoteBookR,(self.interior(),),)
  File "E:\Downloads\pmw\Pmw\Pmw_0_8_4\lib\PmwLoader.py", line 174, in
__getattr__
AttributeError: NoteBookR


From aleaxit at yahoo.com  Thu Mar 22 14:36:41 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 22 Mar 2001 20:36:41 +0100
Subject: Python and emacs
References: <mailman.985287273.17896.python-list@python.org>
Message-ID: <99dkg711894@news1.newsguy.com>

"Sean 'Shaleh' Perry" <shaleh at valinux.com> wrote in message
news:mailman.985287273.17896.python-list at python.org...
>
> On 22-Mar-2001 Don Dwiggins wrote:
> > A semi-idle question: has anyone looked into embedding Python as an
> > extension language in Emacs, alongside Elisp?  I'm toying with the idea
of
> > doing some Emacs enhancements/modifications, and given the choice, I'd
much
> > rather do them in Python.  (One such enhancement would be to add
> > Pythonwin-style "intellisense" popups to python-mode.)
>
> An editor that spoke python would not be a bad thing.  However, trying to
get

Whaddya mean, 'would'?  VIM does that right now, and has for years
(see www.vim.org)...


Alex





From grante at visi.com  Wed Mar 14 10:31:26 2001
From: grante at visi.com (Grant Edwards)
Date: Wed, 14 Mar 2001 15:31:26 GMT
Subject: This math scares me
References: <3aad4062.1290021@News.CIS.DFN.DE> <20010312170356.B3513@better.net> <mailman.984480856.18097.python-list@python.org> <984486395.407379@newsmaster-04.atnet.at> <slrn9asf1d.ja3.qrczak@qrnik.zagroda> <984518199.127140@newsmaster-04.atnet.at> <jwbnews-09D75F.13313313032001@corp.supernews.com> <984521930.27710@newsmaster-04.atnet.at> <Vyxr6.1238$Tg.168802@ruti.visi.com> <984561044.512222@newsmaster-04.atnet.at>
Message-ID: <i9Mr6.180$lz1.7188@ruti.visi.com>

In article <984561044.512222 at newsmaster-04.atnet.at>, Werner Schiendl wrote:

>> >Does this make a difference, as soon as arithmetics takes place?
>>
>> Yes.  All FP calculations took place in BCD.  You got exact
>> results as long as your values didn't exceed the
>> representations limits.
>>
>Exactly, and this is for example the case for 1/3, 1/7 and so on...
>Doesn't sound like an real improvement...

BCD floating point was primarily intended for monetary
calculations.

There is no such thing as 1/3 or 1/7 of a dollar.  

If you go to a bank and ask to withdraw 1/3 of a dollar,
they'll tell you you can have 0.33 or 0.34 dollars, but they
can't give you 1/3 dollars.

If you want to use rational numbers, you should use rational
numbers and not floating point.

-- 
Grant Edwards                   grante             Yow!  Now KEN and BARBIE
                                  at               are PERMANENTLY ADDICTED to
                               visi.com            MIND-ALTERING DRUGS...


From costas at meezon.com  Fri Mar 16 20:47:05 2001
From: costas at meezon.com (Costas Menico)
Date: Sat, 17 Mar 2001 01:47:05 GMT
Subject: Console printing
Message-ID: <3ab2c183.4460238@News.CIS.DFN.DE>

Why does the following program only prints the output (in the run
method) when the while loop terminates? Please note that this is
running in a thread. I though a sleep() is supposed to allow time for
console to update itself. I am using Pythonwin.

import threading
import time
import sys

class wasteTime(threading.Thread):

    def run(self):    
        # run forever
        self.running=1
        while self.running:
            print time.time()
            time.sleep(1)

    def stop(self):
        self.running=0

th=wasteTime()

print "Start thread"
startTime=time.time()
th.start()
while time.time()<startTime+5:
    time.sleep(1)

# I want to kill the th thread right here.
th.stop()
print "Finished Thread"
sys.exit()

#Costas



From wayne.izatt at _nospam_myself.com  Tue Mar  6 02:42:47 2001
From: wayne.izatt at _nospam_myself.com (Wayne Izatt)
Date: Tue, 06 Mar 2001 07:42:47 GMT
Subject: Can anyone recomend a good intoduction to C...
References: <3aa366c6_2@nnrp1.news.uk.psi.net>
Message-ID: <Xx0p6.183448$df5.5139683@news1.crdva1.bc.home.com>

Kerninghan and Ritchie - "The C Programming Language" is a beautiful text.
Everything you want , nothing you don't. Plus, it weighs about as much as a
magazine. Not many languages can teach you all you need to know in that kind
of space. The Python Tutorial, of course, is in the same league!

cheers



"Simon Brunning" <sbrunning at trisystems.co.uk> wrote in message
news:3aa366c6_2 at nnrp1.news.uk.psi.net...
> ... for a Python programmer?
>
> I'm going to have to bite the bullet and learn a bit of C. I'd at
> least like to learn the *right* way to do it.
>
> --
> Cheers,
> Simon Brunning
> sbrunning at bigfoot.com
>
>
>




From kca17 at dial.pipex.com  Wed Mar 28 11:08:23 2001
From: kca17 at dial.pipex.com (Eddie and Babs)
Date: Wed, 28 Mar 2001 17:08:23 +0100
Subject: Python assembler
Message-ID: <3ac20dcc$0$15028$cc9e4d1f@news.dial.pipex.com>

The Python library reference says of the 'dis' module: "Since there is no 
Python assembler, this module defines the Python assembly language."


Is the lack of a Python bytecode assembler a deliberate policy decision or
is it just that nobody has written one yet? Or have many tried, but none
lived to tell the tale?


I would consider such an assembler if there were no obvious reasons for it
not to exist.

&.


From bkline at rksystems.com  Fri Mar 30 09:21:06 2001
From: bkline at rksystems.com (Bob Kline)
Date: Fri, 30 Mar 2001 09:21:06 -0500 (EST)
Subject: Where are the examples/samples?
In-Reply-To: <20010330021322.08468.00001688@ng-fe1.aol.com>
Message-ID: <Pine.LNX.4.10.10103300910190.32588-100000@rksystems.com>

On 30 Mar 2001, LMHAMRIC wrote:

> After looking at every Python book in the local bookstore and at
> every Python-related site I could find on the Internet I have to
> admit that I'm perplexed. I've seen a number of posts that have
> asked for examples of various things and when I follow the links -
> where are the examples? Lots of theory, lots computer-science
> indulgence, but no examples. 

No examples?

Here's some laziness indulgence (brought to you courtesy of Google and 
www.python.org):

http://python.org/doc/current/tut/tut.html
http://cvs.sourceforge.net/cgi-bin/cvsweb.cgi/python/dist/src/Demo/?cvsroot=python
http://starship.python.net/crew/wesc/cpp/book/

Also, sit down and read the FAQ, focusing on the "How do I ..."
questions.

-- 
Bob Kline
mailto:bkline at rksystems.com
http://www.rksystems.com




From aleaxit at yahoo.com  Thu Mar 15 06:39:27 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 15 Mar 2001 12:39:27 +0100
Subject: Magnitude of the wx* market (was: Python In A Nutshell - suggestions)
References: <A909AD29CBD32AFC.4F1641C135749F87.7149617C7C166896@lp.airnews.net> <memo.20010315004230.1527B@digitig.compulink.co.uk>
Message-ID: <98qacl02c1e@news1.newsguy.com>

"Tim Rowe" <digitig at cix.co.uk> wrote in message
news:memo.20010315004230.1527B at digitig.compulink.co.uk...
    [snip]
> > I happen to know, though, that several publishers have
> > rejected wxWindows proposals.
> >
> > I doubt, though, that there'd be much trouble making the
> > case that the subject deserves a chapter in a larger work.
>
> And the larger work in question is "Python Programming on Win32" by Mark

Which is a great book, but, of course, *IS* Win32-specific; pity,
because the very reason for being of wxWindows is CROSS-platform,
portable development...!!!


Alex





From tdelaney at avaya.com  Wed Mar 21 19:03:44 2001
From: tdelaney at avaya.com (Delaney, Timothy)
Date: Thu, 22 Mar 2001 11:03:44 +1100
Subject: Yet Another PEP:  Query Protocol Interface or __query__
Message-ID: <B43D149A9AB2D411971300B0D03D7E8B0FBFC0@natasha.auslabs.avaya.com>

A couple of definite flaws (one major). Code changed below.

class __requirement:
	""""""
	
	def __init__(self):
		""""""

	def verify (self, adaptor):
		""""""

		#	For now, we will just verify that the object has the
		#	named attribute. Later this can be more complex
		#	(parameters, signatures, etc)

		attr = getattr(adaptor, self.getName(), None)

		if attr:
			return adaptor

		#	Not required, but explicit is better than implicit
;)
		return None

class interface:
	""""""
	
	def __init__(self):
		""""""

	def requirements (self):
		""""""
		#	return sequence of __requirement - requirements are
		#	generated automatically by factory methods in the
		#	interface class
		return ()

	def __adapt__(self, descriptor):
		""""""

		#	For now, only interface and its subclasses will be
		#	directly adapted. Later we will add a verification
		#	process directly here as the requirements() method
		#	should return an empty sequence unless requirements
		#	have been added by subclasses or instances

		if self == descriptor or isinstance(descriptor, interface):
			return self

__interface = interface()

class implements:
	""""""
	
	def __init__(self):
		""""""

	def __query__(self, requirement):
		""""""
		

def adapt(descriptor, object):

	#	Verify that the interface object actually meets
	#	the requirements for the interface descriptor interface

	if descriptor:
		assert adapt(None, descriptor)
	else:
		descriptor = __interface

	adapted = None
	__adapt__ = getattr(object, '__adapt__', None)

	if __adapt__:
		adapted = __adapt__(descriptor)

	if not adapted:

		adapted = object
		__query__ = getattr(object, '__query__', None)
		
		for r in descriptor.requirements():
			
			if __query__:

				query = adaptor.__query__(self)

				if query == None:
					adapted = r.verify(adaptor,
__query__)
				elif not query:
					adapted = None

				if not adapted:
					break

	if adapted:
		return adapted
	else:
		raise AttributeError('Does not implement interface')

if __name__ == '__main__':
	print adapt(None, implements())

Tim Delaney
Avaya Australia
+61 2 9352 9079



From qrczak at knm.org.pl  Sat Mar 24 16:34:22 2001
From: qrczak at knm.org.pl (Marcin 'Qrczak' Kowalczyk)
Date: 24 Mar 2001 21:34:22 GMT
Subject: static methods
References: <mailman.985383828.2942.python-list@python.org> <mailman.985457492.29908.python-list@python.org>
Message-ID: <slrn9bq4mu.rth.qrczak@qrnik.zagroda>

Sat, 24 Mar 2001 13:11:43 -0500, Robin Thomas <robin.thomas at starmedia.net> pisze:

> In C, "static" has been overexploited such that it has many different
> meanings in different contexts that have nothing to do with being
> "static".

A recent example in C99:

       [#7] A declaration of a parameter as ``array of type'' shall
       be adjusted to ``qualified pointer to type'', where the type
       qualifiers (if any) are those specified within the [  and  ]
       of  the  array  type derivation.  If the keyword static also
       appears within the [ and ] of  the  array  type  derivation,
       then  for  each  call  to  the  function,  the  value of the
       corresponding actual argument shall provide  access  to  the
       first  element of an array with at least as many elements as
       specified by the size expression.

[...]

       [#21] EXAMPLE 5 The following are  all  compatible  function
       prototype declarators.

               double maximum(int n, int m, double a[n][m]);
               double maximum(int n, int m, double a[*][*]);
               double maximum(int n, int m, double a[ ][*]);
               double maximum(int n, int m, double a[ ][m]);

       as are:

               void f(double (* restrict a)[5]);
               void f(double a[restrict][5]);
               void f(double a[restrict 3][5]);
               void f(double a[restrict static 3][5]);

       (Note  that  the  last  declaration  also specifies that the
       argument corresponding to a in any call to f must be a  non-
       null  pointer  to  the  first  of at least three arrays of 5
       doubles, which the others do not.)

-- 
 __("<  Marcin Kowalczyk * qrczak at knm.org.pl http://qrczak.ids.net.pl/
 \__/
  ^^                      SYGNATURA ZAST?PCZA
QRCZAK


From topmind at technologist.com  Sun Mar 25 22:01:23 2001
From: topmind at technologist.com (topmind)
Date: Mon, 26 Mar 2001 03:01:23 GMT
Subject: Collection interfaces
References: <3a90fa97.0@news1.mweb.co.za>   <97fjla128ir@news2.newsguy.com> <97h5s40vhc@news2.newsguy.com> <MPG.1506007591cf818b9896bc@news.earthlink.net> <slrn99opop.gr.amk@mira.erols.com> <MPG.1506265ebf6d934d9896be@news.earthlink.net> <97ibmc0k47@news1.newsguy.com> <MPG.1506cf3293b094b49896c1@news.earthlink.net> <97m0c7080c@news2.newsguy.com> <MPG.15085517debf4b649896ca@news.earthlink.net> <zb%n6.31033$o%6.586026@news1.rdc1.bc.home.com> <MPG.150b383cea226d409896d0@news.earthlink.net> <kNko6.6986$eq3.707121@news.uswest.net> <MPG.150c46db9cec283e9896d4@news.earthlink.net> <3AA2A898.450B659E@mail.com> <gRCt6.12224$Im6.1308560@newsread1.prod.itd.earthlink.net> <3AB755F8.67E84126@mail.com> <%a%t6.17245$227.1601569@newsread2.prod.itd.earthlink.net> <3AB8A730.4BE13D22@mail.com> <61bv6.8693$P4.676800@newsread1.prod.itd.earthlink.net> <B6E2B64C.D8E3%t*b*r*u*m*l*e*y@f*u*s*e.n*e*t> <o1jv6.9635$P4.789083@newsread1.prod.itd.earthlink.net> <3ABF74ED.BC09EC8@mail.com>
Message-ID: <7iyv6.11924$ue1.1104528@newsread2.prod.itd.earthlink.net>

"James A. Robertson" <jarober at mail.com> wrote in message
news:3ABF74ED.BC09EC8 at mail.com...
> As to the LessonBrowser, it's part of the non-commercial product.  When
> you open the system up, there it is
>
>

What exactly would I be "downloading"?

Is it a Java applet or what?

What do you mean by "open the system up"?

Being non-commercial does not necessarily make it
safer (less DLL's etc.)

Is there an HTML version?

I clicked on the "visualworks online tutorial" link, it asked for
my email ( foo at bar.com, of course :-), and then it brought me
back to the front page.

The second try gave me an HTTP error.

-tmind-





From jcm at bigskytel.com  Thu Mar  1 09:02:32 2001
From: jcm at bigskytel.com (David Porter)
Date: Thu, 1 Mar 2001 07:02:32 -0700
Subject: how can i copy a sublist
In-Reply-To: <3A9E4C6F.E2DD63C3@bg-sys.de>; from daniel.orlowski@bg-sys.de on Thu, Mar 01, 2001 at 02:19:43PM +0100
References: <3A9E4C6F.E2DD63C3@bg-sys.de>
Message-ID: <20010301070232.A7006@bigskytel.com>

* Daniel Orlowski <daniel.orlowski at bg-sys.de>:

> i try to copy a sublist from list "a" to list "b". After that i alter a
> value in "b" and oops! value in "a" is also altered.
> so it looks like that:
> >>> a=[[1,2],[3,4]]
> >>> b=[]
> >>> b.append(a[0])

b.append(a[0][:])

This copies the list items and creates a new object. OTOH, the original 
references the same object that a[0] references. Therefore, when b[0] is
changed, so is a[0] (because it is really the object that they both
reference that is changed). 

hth, 
David



From MarkH at ActiveState.com  Fri Mar 16 17:57:28 2001
From: MarkH at ActiveState.com (Mark Hammond)
Date: Fri, 16 Mar 2001 22:57:28 GMT
Subject: KeyboardInterrupt in IIS?
References: <mailman.984765389.11361.python-list@python.org>
Message-ID: <3AB29B94.808@ActiveState.com>

fluxent at yahoo.com wrote:

> I'm running Python via ASP pages in IIS (win2k).
> 
> Occasionally I start getting errors when I ask for any page that look 
> like
> HTTP 500.100 - Internal Server Error - ASP error
> Internet Information Services
> 
> ----------------------------------------------------------------------
> ----------
> 
> Technical Information (for support personnel)
> 
> Error Type:
> Python ActiveX Scripting Engine (0x80020009)
> Traceback (innermost last): KeyboardInterrupt: 
> /test/biztrack/admin/timechargesReview.asp
> 
> It eventually seems to go away on its own, or else I reboot (gee, 
> thought I didn't have to do that anymore...)

I have no idea what causes this, but do seem to understand what Python 
is doing.

Python installs a Ctrl+C handler.  For some reason, it appears this 
handler fires at strange times, even when Python is _not_ running. 
However, Python still sets an internal flag.  Next time Python is 
entered, it immediately dies with this exception as the flag is set.

I think I need to change the code to ensure that Python ignores this and 
any other pending Python exception that is set whenever a COM call is 
entered.  Can anyone see a reason _not_ to do this?  We did it for xpcom 
and it solved similar issues.  It does however mean that Ctrl+C will 
generally only kill Python code if Python code is actually running as it 
is pressed.  This sounds OK to me...

Mark.



From just at letterror.com  Sat Mar 24 17:45:39 2001
From: just at letterror.com (Just van Rossum)
Date: Sat, 24 Mar 2001 23:45:39 +0100
Subject: "object" vs "dynamic" programming styles
References: <3ABD117E.1E5492C4@erols.com>
Message-ID: <3ABD2390.A2A28A12@letterror.com>

"Edward C. Jones" wrote:
 
> Here are three methods for doing the same thing. When is it better to use
> one of the approaches rather than another?

> # Program 3
> 
> # The function New_X returns an instance of either class A or class B.
> def New_X(c):
>     if c == 'a':
>         return A()
>     if c == 'b':
>         return B()
 
My personal preference would definitely be version 3, the factory function.
Less magic, clearer code.

When I first started writing OO code (which was with Python) I was often
tempted to do lots of smart things in the __init__ method. But it only
makes things harder in the long run. I recently had to extend a module I
had written a few years back: all the smart things in the __init__ were
of no use to me now: it assumed a quite different usage.
So I changed the __init__ to be as simple as possible, and added a couple
of factory functions. It's all nice and clean now...

Just


From Norman_Shelley-RRDN60 at email.sps.mot.com  Wed Mar  7 12:04:40 2001
From: Norman_Shelley-RRDN60 at email.sps.mot.com (Norman Shelley)
Date: Wed, 07 Mar 2001 10:04:40 -0700
Subject: Python 2.1b1 setup.py bug for HPUX11
References: <3AA41658.AAA154F1@email.sps.mot.com> <j41ys9oq71.fsf@informatik.hu-berlin.de>
Message-ID: <3AA66A28.E476A494@email.sps.mot.com>

Martin von Loewis wrote:

> Norman Shelley <Norman_Shelley-RRDN60 at email.sps.mot.com> writes:
>
> > In order to build python 2.1b1 for
> > % uname -a
> > HP-UX wshelley B.11.00 A 9000/785 2014123772 two-user license
> >
> > I had to modify the setup.py script as follows. BTW, I did not request
> > curses or termios in Modules/Setup either.
>
> Obviously not - if you had, setup.py would not attempt to build it
>
> > % diff setup.py setup.py.orig
> > 385c385
> > <             ###exts.append( Extension('termios', ['termios.c']) )
> > ---
> > >             exts.append( Extension('termios', ['termios.c']) )
> > 418,419c418,419
> > <             ###exts.append( Extension('_curses', ['_cursesmodule.c'],
> > <             ###                       libraries = curses_libs) )
> > ---
> > >             exts.append( Extension('_curses', ['_cursesmodule.c'],
> > >                                    libraries = curses_libs) )
>
> What is the problem with building the termios and _curses modules? Is
> it generally impossible to do so on HP/UX? Are some libraries or
> header files necessary that are not generally available, but require
> additional packages being installed?



> Are there additional #defines or
> other magic needed to compile these modules?

Apparently so  /usr/include/curses.h is available.


>
>
> It would be desirable to find a patch to setup.py that
>
> a) automatically detects whether it can build these modules on HP/UX,
> b) automatically skips them when building them is not possible, and
> c) has no impact on other systems.
>
> Your patch won't fulfil requirement c), as it drops termios
> altogether. If you can come up with a patch that solves the problem
> automatically, please submit it to sourceforge.net/projects/python.
>
> Thanks,
> Martin

gcc -g -O2 -Wall -Wstrict-prototypes -fpic -I.
-I/tmp_mnt/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/./Include
-IInclude/ -I/usr/local/include -c
/tmp_mnt/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c
-o build/temp.hp-ux-B.10.20-9000/782-2.1/_cursesmodule.o


/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:115:
conflicting types for `setupterm'
/usr/local/lib/gcc-lib/hppa2.0-hp-hpux10.20/2.95.1/include/curses.h:227:
previous declaration of `setupterm'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCursesWindow_untouchwin':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:285:
warning: implicit declaration of function `untouchwin'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCursesWindow_redrawwin':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:287:
warning: implicit declaration of function `redrawwin'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCursesWindow_is_wintouched':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:292:
warning: implicit declaration of function `is_wintouched'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCursesWindow_wsyncup':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:294:
warning: implicit declaration of function `wsyncup'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCursesWindow_wsyncdown':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:295:
warning: implicit declaration of function `wsyncdown'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCursesWindow_wcursyncup':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:298:
warning: implicit declaration of function `wcursyncup'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCursesWindow_idcok':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:303:
warning: implicit declaration of function `idcok'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCursesWindow_immedok':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:304:
warning: implicit declaration of function `immedok'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCursesWindow_wtimeout':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:305:
warning: implicit declaration of function `wtimeout'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCursesWindow_getbegyx':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:308:
warning: implicit declaration of function `getbegyx'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:308:
warning: `arg1' might be used uninitialized in this function
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:308:
warning: `arg2' might be used uninitialized in this function
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCursesWindow_getmaxyx':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:309:
warning: implicit declaration of function `getmaxyx'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:309:
warning: `arg1' might be used uninitialized in this function
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:309:
warning: `arg2' might be used uninitialized in this function
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCursesWindow_getparyx':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:310:
warning: implicit declaration of function `getparyx'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:310:
warning: `arg1' might be used uninitialized in this function
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:310:
warning: `arg2' might be used uninitialized in this function
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCursesWindow_wattron':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:312:
`attr_t' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:312:
(Each undeclared identifier is reported only once
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:312:
for each function it appears in.)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:312:
parse error before `arg1'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:312:
`arg1' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCursesWindow_wattroff':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:313:
`attr_t' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:313:
parse error before `arg1'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:313:
`arg1' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCursesWindow_wattrset':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:314:
`attr_t' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:314:
parse error before `arg1'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:314:
`arg1' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCursesWindow_notimeout':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:328:
warning: implicit declaration of function `notimeout'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCursesWindow_winsdelln':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:330:
warning: implicit declaration of function `winsdelln'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCursesWindow_syncok':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:331:
warning: implicit declaration of function `syncok'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCursesWindow_mvderwin':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:334:
warning: implicit declaration of function `mvderwin'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCursesWindow_wresize':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:337:
warning: implicit declaration of function `wresize'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCursesWindow_AddCh':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:368:
`attr_t' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:368:
parse error before `attr'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:376:
`attr' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCursesWindow_AddStr':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:414:
`attr_t' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:414:
parse error before `attr'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:423:
`attr' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:443:
`attr_old' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:443:
warning: implicit declaration of function `getattrs'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCursesWindow_AddNStr':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:460:
`attr_t' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:460:
parse error before `attr'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:469:
`attr' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:489:
`attr_old' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:493:
warning: implicit declaration of function `mvwaddnstr'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:495:
warning: implicit declaration of function `waddnstr'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCursesWindow_Bkgd':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:506:
`attr_t' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:506:
parse error before `attr'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:514:
`attr' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:527:
warning: implicit declaration of function `wbkgd'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCursesWindow_BkgdSet':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:535:
`attr_t' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:535:
parse error before `attr'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:543:
`attr' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:556:
warning: implicit declaration of function `wbkgdset'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCursesWindow_Border':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:568:
warning: implicit declaration of function `wborder'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCursesWindow_DerWin':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:632:
warning: implicit declaration of function `derwin'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:632:
warning: assignment makes pointer from integer without a cast
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCursesWindow_EchoChar':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:647:
`attr_t' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:647:
parse error before `attr'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:655:
`attr' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:672:
warning: implicit declaration of function `pechochar'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:676:
warning: implicit declaration of function `wechochar'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCursesWindow_GetBkgd':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:697:
warning: implicit declaration of function `getbkgd'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCursesWindow_GetKey':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:755:
warning: implicit declaration of function `keyname'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCursesWindow_GetStr':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:776:
warning: implicit declaration of function `wgetnstr'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:797:
warning: implicit declaration of function `mvwgetnstr'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCursesWindow_Hline':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:816:
`attr_t' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:816:
parse error before `attr'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:824:
`attr' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:848:
warning: implicit declaration of function `whline'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCursesWindow_InsCh':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:859:
`attr_t' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:859:
parse error before `attr'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:867:
`attr' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCursesWindow_InStr':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:929:
warning: implicit declaration of function `winstr'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:934:
warning: implicit declaration of function `winnstr'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:939:
warning: implicit declaration of function `mvwinstr'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:944:
warning: implicit declaration of function `mvwinnstr'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCursesWindow_InsStr':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:961:
`attr_t' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:961:
parse error before `attr'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:970:
`attr' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:990:
`attr_old' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:994:
warning: implicit declaration of function `mvwinsstr'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:996:
warning: implicit declaration of function `winsstr'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCursesWindow_InsNStr':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1007:
`attr_t' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1007:
parse error before `attr'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1016:
`attr' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1036:
`attr_old' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1040:
warning: implicit declaration of function `mvwinsnstr'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1042:
warning: implicit declaration of function `winsnstr'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCursesWindow_Is_LineTouched':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1054:
warning: implicit declaration of function `is_linetouched'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCursesWindow_Overlay':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1139:
warning: implicit declaration of function `copywin'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCursesWindow_PutWin':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1198:
warning: implicit declaration of function `putwin'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCursesWindow_RedrawLine':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1208:
warning: implicit declaration of function `wredrawln'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCursesWindow_SubWin':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1287:
warning: implicit declaration of function `subpad'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1287:
warning: assignment makes pointer from integer without a cast
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCursesWindow_Scroll':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1311:
warning: implicit declaration of function `wscrl'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCursesWindow_TouchLine':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1326:
warning: implicit declaration of function `touchline'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1331:
warning: implicit declaration of function `wtouchln'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCursesWindow_Vline':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1344:
`attr_t' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1344:
parse error before `attr'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1352:
`attr' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1376:
warning: implicit declaration of function `wvline'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCurses_termattrs':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1515:
warning: implicit declaration of function `termattrs'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCurses_termname':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1517:
warning: implicit declaration of function `termname'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1517:
warning: passing arg 1 of `PyString_FromString' makes pointer from integer
without a cast
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCurses_can_change_color':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1520:
warning: implicit declaration of function `can_change_color'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCurses_has_colors':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1521:
warning: implicit declaration of function `has_colors'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCurses_isendwin':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1524:
warning: implicit declaration of function `isendwin'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCurses_filter':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1525:
warning: implicit declaration of function `filter'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCurses_noqiflush':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1527:
warning: implicit declaration of function `noqiflush'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCurses_Color_Content':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1545:
warning: implicit declaration of function `color_content'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCurses_Curs_Set':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1584:
warning: implicit declaration of function `curs_set'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCurses_getsyx':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1629:
warning: implicit declaration of function `getsyx'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1623:
warning: `x' might be used uninitialized in this function
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1623:
warning: `y' might be used uninitialized in this function
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCurses_GetWin':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1686:
warning: implicit declaration of function `getwin'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1686:
warning: assignment makes pointer from integer without a cast
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCurses_HalfDelay':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1712:
warning: implicit declaration of function `halfdelay'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCurses_has_key':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1725:
warning: implicit declaration of function `has_key'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCurses_Init_Color':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1751:
warning: implicit declaration of function `init_color'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCurses_Init_Pair':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1769:
warning: implicit declaration of function `init_pair'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCurses_InitScr':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1802:
`ACS_ULCORNER' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1803:
`ACS_LLCORNER' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1804:
`ACS_URCORNER' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1805:
`ACS_LRCORNER' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1806:
`ACS_LTEE' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1807:
`ACS_RTEE' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1808:
`ACS_BTEE' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1809:
`ACS_TTEE' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1810:
`ACS_HLINE' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1811:
`ACS_VLINE' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1812:
`ACS_PLUS' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1813:
`ACS_S1' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1814:
`ACS_S9' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1815:
`ACS_DIAMOND' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1816:
`ACS_CKBOARD' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1817:
`ACS_DEGREE' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1818:
`ACS_PLMINUS' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1819:
`ACS_BULLET' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1820:
`ACS_LARROW' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1821:
`ACS_RARROW' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1822:
`ACS_DARROW' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1823:
`ACS_UARROW' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1824:
`ACS_BOARD' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1825:
`ACS_LANTERN' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1826:
`ACS_BLOCK' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCurses_KeyName':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:1954:
warning: assignment makes pointer from integer without a cast
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCurses_Napms':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:2025:
warning: implicit declaration of function `napms'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCurses_Pair_Content':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:2099:
warning: implicit declaration of function `pair_content'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCurses_pair_number':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:2126:
`A_COLOR' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCurses_QiFlush':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:2147:
warning: implicit declaration of function `qiflush'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCurses_setsyx':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:2176:
warning: implicit declaration of function `setsyx'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCurses_Start_Color':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:2192:
warning: implicit declaration of function `start_color'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:2195:
`COLORS' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:2198:
`COLOR_PAIRS' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCurses_tigetflag':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:2219:
warning: implicit declaration of function `tigetflag'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCurses_tigetnum':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:2232:
warning: implicit declaration of function `tigetnum'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCurses_tigetstr':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:2245:
warning: implicit declaration of function `tigetstr'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:2245:
warning: assignment makes pointer from integer without a cast
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCurses_UngetCh':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:2359:
warning: implicit declaration of function `ungetch'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `PyCurses_Use_Env':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:2378:
warning: implicit declaration of function `use_env'
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:
In function `init_curses':
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:2522:
`A_COLOR' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:2544:
`COLOR_BLACK' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:2545:
`COLOR_RED' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:2546:
`COLOR_GREEN' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:2547:
`COLOR_YELLOW' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:2548:
`COLOR_BLUE' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:2549:
`COLOR_MAGENTA' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:2550:
`COLOR_CYAN' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:2551:
`COLOR_WHITE' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:2592:
`KEY_MIN' undeclared (first use in this function)
/net/roti/wssg/hpux10x/usrlocal/packages/python2.1/Python-2.1b1/Modules/_cursesmodule.c:2592:
`KEY_MAX' undeclared (first use in this function)
WARNING: building of extension "_curses" failed: command 'gcc' failed with
exit status 1




From guido at python.org  Sat Mar 31 19:54:59 2001
From: guido at python.org (Guido van Rossum)
Date: Sat, 31 Mar 2001 19:54:59 -0500
Subject: PythonLabs Team Moves to ActiveState
Message-ID: <E14jW8x-0001Rj-00@mail.python.org>

PythonLabs Team Moves to ActiveState

To all Python users and developers:

   Less than half a year ago, I moved with my team to Digital Crations, in the
   hope of finding a new permanent home for Python development. At
   Digital Creations we've done several good things for Python, such as forming
   the Python Software Foundation and the management of the Python 2.1 release.
   
   Unfortunately, Digital Creations's original plans for PythonLabs didn't 
   work out as hoped, and we weren't able to reach mutual agreement on
   workable alternative plans - despite trying for months.
   
   I am proud to have found a new home for my entire team: starting
   today, Tim Peters, Barry Warsaw, Jeremy Hylton, Fred Drake and myself
   are working for ActiveState. We will be spending part of our
   time on core Python development (including Jython and Mailman) and
   part of our time on Python infrastructure improvements that also
   benefit ActiveState's efforts such as Komodo and Python.NET.
   
   Python will remain Open Source; ActiveState has no desire to
   monetize or brand the Python language or specific Python
   distributions. All future work we do on Python as ActiveState
   employees will be owned by the PSF.
   
   We're excited to be working for ActiveState: they are one of the companies
   most committed to Python, and they have great products! Plus, we know
   they have deep financial backing. We trust that ActiveState will
   provide a stable home for Python for many years.
   
   ActiveState has also offered to take over hosting of the
   python.org and starship sites. On behalf of the Python community,
   we're grateful for this support of the two prime community sites for
   Python, and we expect to be implementing the transitions shortly.
   
   These are exciting times for the PythonLabs team - and also for Python
   and its community. Mainstream successes for Python are showing up
   everywhere, and we're proud to be a part of such a smart and friendly
   community. A great year lies ahead!
   
   Note, however, that moving to ActiveState will have some minor effects
   on the future of Python development. ActiveState is concerned about
   the seeming rift between the Perl and Python communities and wants
   to do its best to help the Open Source scripting community work together.
   In order to make it easier for Perl programmers to transition to Python,
   we will add optional curly braces for Python. While we are fully aware
   of the effect this will have on Python usability, we have finally decided
   to bite the bullet and accept that until this happens, Python will never
   be popular.

   Barry has already approved PEP 357, Adding Block Delimiters to Python.
   Since we are heavily commited to backwards compatability, a transition
   period of three months has been mandated. For those who want the changes
   to start sooner, we added "from __future__ import block_delimiters" in
   Python 2.1, which will be released about two weeks from now.

   Larry Wall and I agreed that now it makes sense for Perl to have 
   the Perl compiler enforce correct indentation in Perl sources. This
   will probably be implemented in Perl 5.7, and will certainly be in Perl 6.
   Also, after some talks with Larry, I've decided to adopt the Perl 
   numerical model for Python. The Perl model is simple, predictable and
   easy to explain. This is a small enough change that it does not need
   a PEP, and Moshe will implement it before 2.1rc1 is out.
   
   Now that the catalog-sig finally has a fully working product, ActiveState
   decided they will base a CPAN competitor on it. This server will be able
   to deal with both Perl and Python modules. While the client will remain
   completely Open Source, it was decided that the server will remain
   ActiveState proprietary. While it is true that the original server was
   written by the Python community, ActiveState has put a lot of effort
   into making it general enough, and we need to justify the costs somehow.

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




From grante at visi.com  Fri Mar  2 18:03:38 2001
From: grante at visi.com (Grant Edwards)
Date: Fri, 02 Mar 2001 23:03:38 GMT
Subject: cataloging words in text file
References: <3AA0124E.74A39783@comm.mot.com> <v9Vn6.396$z6.33544@ruti.visi.com>
Message-ID: <eFVn6.400$z6.34905@ruti.visi.com>

In article <v9Vn6.396$z6.33544 at ruti.visi.com>, Grant Edwards wrote:

>>I remember this homework assignment for my data structures (c++)
>>class: read in a large file, and create a data structure containing
>>every word in the file and the number of times it appears.

>#!/usr/local/bin/python2.1
>import sys
>d={}
>for w in sys.stdin.read().split():
>    if d.has_key(w):
>        d[w] += 1
>    else:
>        d[w] = 1
>print d    
>
>You may want something a little more sophisticated than
>split(), since this solution is case sensitive and treats
>"foo" and "foo." as two different words.

A call to translate() with an appropriate translation table
before the split() solves those issues. ;)

-- 
Grant Edwards                   grante             Yow!  I am NOT a nut....
                                  at               
                               visi.com            


From tariq.rashid at iname.com  Thu Mar 29 15:16:58 2001
From: tariq.rashid at iname.com (Taz)
Date: Thu, 29 Mar 2001 21:16:58 +0100
Subject: wavlets?
References: <tc27anovquqm79@corp.supernews.co.uk> <aXaw6.22728$qv3.9950577@nnrp5-w.sbc.net>
Message-ID: <tc75k57pvgfu5f@corp.supernews.co.uk>

linux only at the moment ... i'll make sure as far as i can that the whole
thing compiles and runs on other systems as far as poss...

taz

"Grant Munsey" <gmunsey at adobe.com> wrote in message
news:aXaw6.22728$qv3.9950577 at nnrp5-w.sbc.net...
> What platform are you using. For Windows there is a pretty nice tuned
> library called IPP
> that does wavelet stuff and a lot more. I have a pretty complete set of
> cover routines for
> the Windows 1.0b2 version. Intel is about to release version 1.0 and when
> they do I was
> going to put the Python module source up on the Adobe Open Source site.
>
> http://developer.intel.com/software/products/ipp/index.htm
>
>
> "Taz" <tariq.rashid at iname.com> wrote in message
> news:tc27anovquqm79 at corp.supernews.co.uk...
> > i'm about to write a wavelets module for python - does one already
> exist...
> > couldn't find one?
> >
> > taz
> >
> >
> >
>
>




From charlie at gabrielinteractive.com  Fri Mar 16 15:52:53 2001
From: charlie at gabrielinteractive.com (Charlie Barrows)
Date: Fri, 16 Mar 2001 15:52:53 -0500
Subject: PyImport_ImportModule
References: <98tcks$7bl$1@allthetime.news.cais.net> <Zgrs6.170$sk3.49958@newsb.telia.net>
Message-ID: <98tug5$i3f$1@allnight.news.cais.net>

Right, so if I wanted to import TestScript into __main__ (ala "from module
import *"), is that possible using PyImport_ImportModuleEx?  If so, what
parameters am I providing incorrectly, it seems I've tried them all.


"Fredrik Lundh" <fredrik at pythonware.com> wrote in message
news:Zgrs6.170$sk3.49958 at newsb.telia.net...
> Charlie Barrows wrote:
> > I'm just starting to tinker with embedding Python in a C app and I'm
trying
> > to get a handle on where modules end up when they're imported.
> >
> > If I run this code:
> >
> > PyObject *module = PyImport_ImportModule("TestScript");
> >
> > ... which namespace is TestScript imported into?
>
> it's own namespace, of course.
>
> ImportModule is the same thing as "import module", not
> "from module import *".
>
> Cheers /F
>
>




From Richard.Philips at ua.ac.be  Wed Mar 28 05:36:02 2001
From: Richard.Philips at ua.ac.be (Richard Philips)
Date: Wed, 28 Mar 2001 12:36:02 +0200
Subject: Name of a method of an instance
Message-ID: <3AC1BE92.ACE82741@ua.ac.be>

A question on introspection:

class Log:

    def __init__(self):
        self.previous = None

    def methodA(self):
        self.previous = "methodA"

    def methodB(self):
        self.previous = "methodB"


myinstance = Log()
myinstance.methodA()
myinstance.methodB()

How can I calculate the name of the method from within the method itself
?

Thanks,

Richard Philips
University of Antwerp





From bsass at freenet.edmonton.ab.ca  Fri Mar 16 19:13:43 2001
From: bsass at freenet.edmonton.ab.ca (Bruce Sass)
Date: Fri, 16 Mar 2001 17:13:43 -0700 (MST)
Subject: Newbie confusion about 'return'
In-Reply-To: <3AB2A2BE.A467DE00@hotmail.com>
Message-ID: <Pine.LNX.4.33.0103161700410.13729-100000@bms>

On Fri, 16 Mar 2001, scott wrote:

<...>
> IDLE:
<...>
> >>> find ("abcde", "d")
> 3 (in blue indicating stdout)
>
> The above is what I expected.
> ------------------------------------
>
> Running python on the file (find.py):
<...>
> find ("abcde", "d")
>
> I ran it as: $ python find.py
> *Nothing* was returned (just gave me the command prompt again), instead
> of the '3' that I expected.
> ---------------------------------------
>
> Why the difference? I thought one of the points of the interactive mode
> was to test code to see if it works before putting it in a file...
> If I add a couple of print statements in the file, I get the expected
> result (3) from running 'python find.py':

The interactive interpreter displays the value of the line when you
hit ENTER, a program will not produce output unless you ask for it
(errors excepted, of course).

> def find(str, ch):
>       index = 0
>       while index < len(str):
>              if str[index] == ch:
>                    print index
>                    return
>              index = index + 1
>       print -1
>       return
>
> find ("abcde", "d")

or do this...
---8< find.py ---
def find(str, ch):
    index = 0
    while index < len(str):
        if str[index] == ch:
             return index
        index = index + 1
    return -1

print find("abcde", 'd')
--->8---

i.e., what you had for the second piece of code, except print out
the returned value.


- Bruce




From jp_reed at yahoo.com  Thu Mar  8 05:21:37 2001
From: jp_reed at yahoo.com (Jeremy Reed)
Date: Thu, 8 Mar 2001 02:21:37 -0800
Subject: kbhit()  and  getch() question
Message-ID: <taecoqaa2mot62@corp.supernews.com>

In a previous question, I asked how to get non-blocking input from the user.
The answer proposed is as follows:

import msvcrt

ch = 0
while ch != 'x':
    if msvcrt.kbhit():
        ch = msvcrt.getch()
        print ch

I have run into a few problems.  First of all, the kbhit() doesn't seem to
work.  I am unable to find a way for the function to return true.

Second, the getch() function is listed in the documentation as a blocking
call, but it doesn't block when i run it in the following script.

ch = 0
while ch != 'x':
    ch = mscvrt.getch()
    print ch

Instead, it prints the equivalent to chr(255) in an infinite loop.

Is there some setup call or initialization function that I'm neglecting to
call?  Why aren't these function working like the documentation says they
should?




From jkraska1 at san.rr.com  Fri Mar  9 02:41:06 2001
From: jkraska1 at san.rr.com (Courageous)
Date: Fri, 09 Mar 2001 07:41:06 GMT
Subject: Coroutines, and memory leaks
References: <8oh089.7it.ln@127.0.0.1> <989kue$64j$1@panix3.panix.com>
Message-ID: <qj2hatspp3fdg09lrjsale1t6c692eg2ba@4ax.com>

>>1. Does Python do coroutines?

http://www.stackless.com



C//



From qrczak at knm.org.pl  Fri Mar  9 14:13:12 2001
From: qrczak at knm.org.pl (Marcin 'Qrczak' Kowalczyk)
Date: 9 Mar 2001 19:13:12 GMT
Subject: Can anyone recomend a good intoduction to C...
References: <mailman.983894288.19965.python-list@python.org> <ewwp6.1466$y6.241005@ruti.visi.com> <slrn9aeblb.cg1.scarblac@pino.selwerd.nl> <djagat4kk3cs9hg0c3k39cudk6goblhkg6@4ax.com> <989mbk$24nuv$1@fido.engr.sgi.com> <slrn9ah66m.k09.qrczak@qrnik.zagroda> <98auhn0270j@news2.newsguy.com>
Message-ID: <slrn9aiaq7.rmu.qrczak@qrnik.zagroda>

Fri, 9 Mar 2001 16:57:47 +0100, Alex Martelli <aleaxit at yahoo.com> pisze:

> > choice: low-level programming, i.e. OS interface, OS itself, runtime
> > systems for other languages, time-critical modules for other languages.
> > C++ doesn't fit there, is unnecessarily too complicated for these
> > tasks, and it isn't a high level language either.
> 
> C++ fits *perfectly well* for these tasks.

Maybe under Windows, but it's rarely used at all under Linux.
Kernel is in C, system interface is in C, system-level libraries are
almost exclusively in C. C++ is used in some projects (KDE, nsgmls)
but lower level libraries are generally expected to be available in C.
On my machine 536 packages require glibc and only 38 require libstdc++
(the latter include 12 of Qt+KDE and 6 of ClanLib).

-- 
 __("<  Marcin Kowalczyk * qrczak at knm.org.pl http://qrczak.ids.net.pl/
 \__/
  ^^                      SYGNATURA ZAST?PCZA
QRCZAK


From thomas at xs4all.net  Sat Mar 10 05:02:28 2001
From: thomas at xs4all.net (Thomas Wouters)
Date: Sat, 10 Mar 2001 11:02:28 +0100
Subject: Urgent Question about Python licensing
In-Reply-To: <98cpfe$5oj$0@216.39.170.247>; from whisper@oz.net on Sat, Mar 10, 2001 at 08:43:26AM +0000
References: <98cpfe$5oj$0@216.39.170.247>
Message-ID: <20010310110228.L404@xs4all.nl>

On Sat, Mar 10, 2001 at 08:43:26AM +0000, Dave LeBlanc wrote:
> Hi;

> While reading up on the recent Python conference on O'Reilly's site
> (http://python.oreilly.com/news/pythonday1_0301.html - actually on the
> 'day 2' page), I cam across the following:

[ The python licence hassles.]

> My questions are these:
> 1. What prevents Python and GPL code from being distributed together?
> If I write some Python code that drives a GPL'd app (but  which does
> not depend on that app), what prevents me from offering both my python
> code and the GPL'd app on the same medium (disk/CD/web)?

'derives' is a bit of a general and weird term. If your code does not depend
on a GPL library or binary, you can just distribute it. You can also
distribute both a GPL application and a Python-licenced application on the
same medium -- there is no restriction to that, as long as you make clear
what licence applies to what part. What you cannot do is link a
Python-licenced piece of software to a GPL-licenced part, *and* distribute
that resulting binary. Not even using dynamic linking, for some reason.
(Don't ask me, I'm not a lawyer.) I am not sure if you can attempt to import
a module, which might be a C module that is covered, possibly indirectly, by
the GPL. My gut says that that should be just fine, but my gut usually tries
to empty itself when it comes to lawyerese.

> 2. Is it currently so, or will Python end up with, some license that
> requires that python scripts be covered by the GPL?

No. Python is used in too many non-GPL-compatible environments for it to
ever become GPL. It might fork into a GPL and a non-GPL fork, but since the
non-GPL fork can actually be used in all places a GPL fork could be, that
would never actually work. The one exception is currently the
GPL-incompatibility, but it's a rare case and it is being solved.

Most Python developers, if not all, consider the BSD-ish licence a good
thing. And more importantly, Guido does. And absolutely most importantly, at
least since last monday, the Python Software Foundation does. The PSF holds
the rights of all future Python enhancements, and hopefully a good chunk of
the previous ones. And the PSF is a mix of corporations and individual
Bruces.

> 3. Is Python 2.0 covered by that evil CNRI 1.6 license? I avoided 1.6
> mostly because of the license, and thought that 2.0 had "skipped" 1.6
> and used a far more equitable license then the 1.6 one. (1.6's license
> permits the copyright holder to suddenly decide that anyone using the
> covered software must pay a fee (or any other thing they so desire -
> this is only an example), and by virtue of your even so much as having
> read the license are implicitly bound to their unilateral declaration
> - lovely Virginia laws).

Because 2.0 is a work derived from 1.6, it *has* to be covered by the 1.6
licence. The 1.6 licence clearly states that it applies to all derived works
as well. I'm not a lawyer, so I cannot answer your doubts about the fee
clause, but the intent of the core Python development team is to give Python
as free a licence as technically possible. If Python was not already licence
burdened (by the CWI, the CNRI and BeOpen), the Python Software Foundation
would own the copyright to all of it, and would probably opt for a free and
uncluttered a licence as possible.

One of the goals of the PSF is to *fix* this damned licencing issue once and
for all. That means holding the IP of as much of Python as we can possibly
get, and safeguard it against this kind of confusion forever more. That
might mean changing the licence again (once the PSF owns all the copyright,
which might take a long time), it might even mean changing it a lot, but it
means the PSF will do everything it can to keep Python free to use in any
circumstance, without people having to worry about the licence at all.

I'd like to note, though, that the PSF can only succeed in all that if it
has the community's support. It seems to have the support of the attendees
of the Python Conference last week, but I'd like to hope we have the support
of most of the rest of the python community as well :) We (actually the
officers of the board we assigned) are working on the infrastructure for the
PSF where you can actually send questions and suggestions, but until then,
feel free to bring issues up here, or address one of the board members or
executive officers directly. Like me, for example ;)

For more info on the PSF: http://www.python.org/psf.html

If-it-were-up-to-me-alone-the-Python-licence-would-read-"Use-it-or-not--I-
  -don't-care"-ly y'rs,
-- 
Thomas Wouters <thomas at xs4all.net>

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



From donn at oz.net  Thu Mar  8 01:34:32 2001
From: donn at oz.net (Donn Cave)
Date: 8 Mar 2001 06:34:32 GMT
Subject: grabbing return codes from os.system() call
References: <%wFp6.9481$dL4.126404@vixen.cso.uiuc.edu>
Message-ID: <98795o$kq8$0@216.39.151.169>

Quoth Damian Menscher <menscher+python at uiuc.edu>:
| I'm new to Python (as in, my experience is essentially the tutorial),
| but I've already come up with an interesting question:
|
| How do I get the return code from an os.system call?  I would have
| expected I could do something like
|
| ---returncode---
| #/bin/csh
| echo do stuff
| exit 3
|
| and then in my python program I could do
|
| print os.system('./returncode')
|
| But it prints out 768.  Not particularly useful, even after I recognize
| the trick of dividing by 256 (byte-swapping going on?  No, because a
| return code of 768 reports as 0).  Given that my real return codes
| will be (possibly large) integers, this limitation will likely cause
| some serious problems down the line.
|
| Even better would be a way of returning a string (the script I run
| can be something other than csh, but it has to be a separate script).
|
| Ideas?  I'm trying to avoid writing the string out to a file and then
| read the file back in to the python program....

Your return codes will not be larger than 255, that's all the bits
there are for this kind of thing (on UNIX, anyway.)  If you have a
bigger number, you will have to put it somewhere else.  So writing
it out to a file might not be such a bad idea.

	Donn Cave, donn at oz.net


From martin.skott at propylon.com  Fri Mar 16 06:18:08 2001
From: martin.skott at propylon.com (Martin Skott)
Date: 16 Mar 2001 11:18:08 +0000
Subject: Cookies
References: <mailman.984707839.22839.python-list@python.org>
Message-ID: <lmq6m07j.fsf@propylon.com>

"Ken Seehof" <kens at sightreader.com> writes:
> Alternate question:
> 
> Is there a technique common to both javascript and python that will
> obtain a unique identifier for a computer (e.g. IP address).  It must
> not be browser specific.

The other guys seems to have answered your first question so I will
try to answer this.
You should first of all be aware that there is no common unique
identifier for a computer available from the net today[1]. IP
addresses as you suggest aren't useable because of of multiple users
being placed behind the same firewall/proxy server whos IP all their
requests originate from. If your application is for an intranet only
without firewalls or proxy servers then this might be different but
generally it fits. 
A far better solution is to generate something unique for the user to
carry around. This is easily done using something like the choice
function of the python random interface. I know nothing about
Javascript so how you would come around this thing using it I don't
know but the fact that using IPs is bad is universal.

[1]Why would somebody use cookies if there was?
-- 
Martin Skott
Software Developer
Propylon - Enabling Universal Mobility
http://www.propylon.com
Tel: +353 096 37245
Mobile: +353 087 9680370


From neilh at scintilla.org  Fri Mar 23 02:06:10 2001
From: neilh at scintilla.org (Neil Hodgson)
Date: Fri, 23 Mar 2001 07:06:10 GMT
Subject: Ruby will overtake Python within four years.
References: <mailman.985324111.22754.python-list@python.org>
Message-ID: <CBCu6.19022$992.115376@news-server.bigpond.net.au>

   There are a few things I like about Ruby but not enough to convert over.
Some of the advantages to me over Python are that there is no separation
between types and classes and you can define iterators.

   The main thing I don't like in Ruby is the syntax. I have written lexers
for Python but Ruby was too much trouble being about half way between Perl
and Python in syntactical complexity. Using indentation for block structure
is, after some initial disorientation, a massive improvement.

   The Python community is putting in a large amount of effort into making
the language, library and tools better. Ruby has Matz who does an enormous
amount of excellent work but Python has many more people working just as
hard often leading to features with a higher degree of quality. The area
that meant the most to me when initially using Python was its support for
COM and related technologies on Windows. This support is better than for any
non-Microsoft language and even beats Microsoft's implementations in some
ways.

   As well as several good open source development environments for Python,
there are 4 commercial IDEs although 2 of these are still in beta. These are
not generic IDEs with the editor changed to style Python - they are new
projects with most of the code written in Python by people who believe in
Python.

   A comparison from a Rubyists point of view is at
http://www.ruby-lang.org/en/compar.html

   You could also look at Lua at www.lua.org. Its main advantage is its
small size which means its easier to understand how it works than either of
Python or Ruby.

   Neil




From scarblac at pino.selwerd.nl  Tue Mar 27 08:09:12 2001
From: scarblac at pino.selwerd.nl (Remco Gerlich)
Date: 27 Mar 2001 13:09:12 GMT
Subject: R: unique items in lists
References: <_nZv6.2878$yU4.468236@news.infostrada.it> <slrn9c0ph1.o5h.scarblac@pino.selwerd.nl> <t5_v6.3466$yU4.477532@news.infostrada.it>
Message-ID: <slrn9c1424.o5h.scarblac@pino.selwerd.nl>

comp.lang.python <micampe at f2s.com> wrote in comp.lang.python:
> Here is what I'm doing now:
> 
>         for w in words:
>             while words.count(w) > 1:
>                 words.remove(w)
> 
> comments?

First, you modify a list that you are looping over. That can give
unpredictable results, but in this case it seems to work.

It will take a long time. 'words.remove(w)' takes time proportional to the
length of the list; words.count(w) takes time proportional to the length of
the list itself, and you repeat that for every duplicate of a word, and the
total is repeated for every unique word. In total it takes time proportional
to N^3 (for a list of length N), I think.

The dictionary method does one dictionary action for each word, and then it
does a dict.keys() which takes time proportional to each unique word. In total
it takes time proportional to N.

For a longer list, N^3 will become much much bigger than N, so your solution
is slower, unfortunately.

My function returns a new list and doesn't change the old list, you can do
that with

def make_unique_list(l):
   dict = {}
   for word in l:
      dict[word] = 1
   l[:] = dict.keys() # Slice assignment changes the old list

Try them both out, I'd say :). For smaller lists they're both quick, but
when you get to lengths over a few thousand, the difference is obvious.

-- 
Remco Gerlich


From paulp at ActiveState.com  Sun Mar 18 13:14:18 2001
From: paulp at ActiveState.com (Paul Prescod)
Date: Sun, 18 Mar 2001 10:14:18 -0800
Subject: Who's minister of propaganda this week?
References: <B98041329020E007.7DA84CAD9D0562AB.817BBC572F480869@lp.airnews.net> <98nu1m$12e@dispatch.concentric.net> <660ACCA1E4816F7E.C359CDF4168EF3FB.3B6B05538D619770@lp.airnews.net> <3AAFA13D.76E32C3F@divalsim.it> <98oa4o01i4v@news2.newsguy.com> <mailman.984619039.7856.python-list@python.org> <98q9h60122v@news2.newsguy.com> <3AB4F520.40104@yahoo.com>
Message-ID: <3AB4FAFA.22DDD3AD@ActiveState.com>

Michael Chermside wrote:
> 
>...
> 
> But I'm quite surprised that you think you pay a 10% penalty
> for going through contortions to adhere to static typechecking.
> You see... in my own experience, I hardly ever want to violate
> it. I generally design type hierarchys and use objects in places
> where I expect an object of that type or of a direct ancestor.

One issue is that you often do not have control over the design of type
hierarchies. You need to glue together two or more large frameworks and
you have no opportunity to set up interfaces etc. as you want.

Another issue is that there are certain kinds of advanced programming
that are are exceedingly difficult in statically typed languages. For
instance let's say you are implementing some kind of object request
broker and you are handed a new interface at runtime. In Java, the way
you handle new interfaces is basically to generate Java "wrapper code".
For instance "JIDL". If you didn't ALREADY have JIDL, you would have to
count the cost of developing it as part of the cost of impleenting CORBA
in Java.

In Python, you don't need anything like that. You could either ignore
the interface declaration or you could write code that reads and
enforces it at runtime. Or you could generate wrapper code as you do in
Java. You have a choice of three different levels of safety versus
performance versus effort. And even if you do decide to generate the
wrapper code, I claim it would be a simpler task than generating
strongly-typed Java wrappers.

-- 
Take a recipe. Leave a recipe.  
Python Cookbook!  http://www.activestate.com/pythoncookbook



From skip at pobox.com  Tue Mar 27 12:37:23 2001
From: skip at pobox.com (Skip Montanaro)
Date: Tue, 27 Mar 2001 11:37:23 -0600 (CST)
Subject: PEP 240 scares me
In-Reply-To: <99e2h61aao@news2.newsguy.com>
References: <Pine.LNX.4.21.0103222338530.29360-100000@kev.ioc.ee>
	<99e2h61aao@news2.newsguy.com>
Message-ID: <15040.53203.949420.930912@beluga.mojam.com>

    Alex> AFAIK -- 1.2 would be the rational 6/5

Okay, dumb question time.  Suppose the compiler sees the string "1.2" in the
input stream.  How does it translate that into a rational number (6/5 or
12/10 or 24/20 or ...)?  Does it simply move the decimal point over
sufficiently and create the corresponding rational number that contains the
appropriate power-of-ten denominator?  What about rationals that can't be
represented correctly using decimal notation (e.g. how does one get from
0.3333 to 1/3, or is it simply approximated by 3333/10000?)

-- 
Skip Montanaro (skip at pobox.com)
(847)971-7098



From bh at intevation.de  Fri Mar 30 04:00:15 2001
From: bh at intevation.de (Bernhard Herzog)
Date: 30 Mar 2001 11:00:15 +0200
Subject: [Q] Python wrapper for ArcView Shapelib?
References: <U5Tw6.209744$__6.36574947@typhoon.southeast.rr.com>
Message-ID: <6qk857skbk.fsf@abnoba.intevation.de>

"Daehyok Shin" <DSHIN at nc.rr.com> writes:

> Is there a python wrapper for ArcView Shapelib?

Yes. I've written one with SWIG. It's available at
ftp://intevation.de/users/bh/pyshapelib-0.1.tar.gz. Last time I checked,
there was a link to it on the shapelib's homepage, but I have prblems
accessing that page at the moment.

  Bernhard

-- 
Intevation GmbH                                 http://intevation.de/
Sketch                                 http://sketch.sourceforge.net/
MapIt!                                               http://mapit.de/


From not.this at seebelow.org  Thu Mar  1 23:01:16 2001
From: not.this at seebelow.org (Grant Griffin)
Date: Thu, 01 Mar 2001 22:01:16 -0600
Subject: is and == (is not other_thread)
References: <3A9DD10F.C8ECF803@seebelow.org> <SYln6.16081$Qb7.2628916@newsb.telia.net>
Message-ID: <3A9F1B0C.8E6530C8@seebelow.org>

Fredrik Lundh wrote:
> 
> Grant Griffin wrote:
> > 1) Are two strings that are equal in the "==" sense always
> > guaranteed to be equal in the "is" sense?
> 
> no.
> 
> > 2) Is "is" faster than "==" for strings?
> 
> marginally.  "==" executes a bit more C code before it
> gets to the identity test.
> 
> > 3) Internally, does "==" for strings always begin with an "is"
> > operation?
> 
> yes
> 
> for the details, see cmp_outcome in Python/ceval.c, and
> PyObject_Compare in Objects/Object.c.
> 
> Cheers /F

Thanks /F (and others) for some great answers!

you-bots-know-even-more-obscure-facts-than-kurt-russell-did-in-
   'the-computer-wore-tennis-shoes'-<wink>-ly y'rs,

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation            http://www.iowegian.com


From york at ficom.net  Wed Mar 28 11:49:34 2001
From: york at ficom.net (york at ficom.net)
Date: 28 Mar 2001 16:49:34 GMT
Subject: OO Windows Registry API
Message-ID: <99t4mu42c57@news1.newsguy.com>

After using the _winreg module I was pining for the fjords -- oh, sorry -- I was pining for a more Pythonesc API.  I noticed that the Python documentation mentions that a better API may be written in the future, but none has been forthcoming.  So -- while up at 2AM with my 6-week-old son last night -- I wrote a wrapper that exposes the registry as a dictionary-like object.  It seems to work really well, and more importantly it feels more like a Python module should.

I decided to search Python-Dev for mentions of a higher-level API and found a squabble over various proposals.  Did the idea of an OO interface die last August, or is there any interest in what I've done?

(Untested) Example code:

from winreg import Registry, NewKey, HKEY_CURRENT_USER
reg = Registry()

# Create a new "My App" key under HKEY_CURRENT_USER/Software.
reg[HKEY_CURRENT_USER]['Software']['My App'] = NewKey()

# Add some values to the newly created key.
# Note: the module only supports strings right now, but that's
#  easy to change.
reg[HKEY_CURRENT_USER]['Software']['My App']['foo'] = 'value 1'
reg[HKEY_CURRENT_USER]['Software']['My App']['bar'] = 'value 2'

# Create a new subkey.
reg[HKEY_CURRENT_USER]['Software']['My App']['subkey'] = NewKey()

# Delete a value.
del reg[HKEY_CURRENT_USER]['Software']['My App']['foo']

# Delete a key.
del reg[HKEY_CURRENT_USER]['Software']['My App']


Any comments or suggestions?  If anyone is interested in the code, I'll post it on the web.

--
Benji York
york at ficom.net


==================================
Posted via http://nodevice.com
Linux Programmer's Site


From DanielK at jBASE.com  Wed Mar 21 18:53:17 2001
From: DanielK at jBASE.com (Daniel Klein)
Date: Wed, 21 Mar 2001 15:53:17 -0800
Subject: learning python...
References: <3AB8F4E1.184DD9C@noaa.gov> <Z97u6.16733$Im6.1761313@newsread1.prod.itd.earthlink.net> <3ab90559_6@goliath.newsfeeds.com> <99b44s$m5j$1@panix6.panix.com> <3AB9182A.8FEBCF3E@noaa.gov>
Message-ID: <N9bu6.66$Hp.3347@typhoon.aracnet.com>

Pay no attention to Amazon reviews. You are ALWAYS going to get people who
love it and those who hate and everything inbetween. If you want to get
information, this is the best place to get it.

The 2nd edition of Programming Python is a great tutorial and has an
enormous store of example code but is self proclaimed as 'not being for the
beginner as this material is covered in Learning Python'.

Dan

"Benjamin.Altman" <benjamin.altman at noaa.gov> wrote in message
news:3AB9182A.8FEBCF3E at noaa.gov...
> It would be good to know if the new edition really is all that different
(other than
> the topics added) to the 1st ed.  The only review on Amazon says he never
read it...
>
> Ben
>
> Aahz Maruch wrote:
>
> > In article <3ab90559_6 at goliath.newsfeeds.com>,
> > Bubba Brains <chocolate at toblerone.yum> wrote:
> > >
> > >Forget Learning Python?  I think it's the best reference/tutorial
around.  I
> > >use python once every couple months, and sometimes forget the simple
python
> > >syntax.  The Learning Python book is great for getting up to speed
fast, and
> > >to look up common stuff.  The other OReilly book (Programming Python)
is
> > >absolutely annoying to use.  It's not even a good reference book (good
luck
> > >finding the info you want quickly).  Don't know about the Prentice
book.
> >
> > If you look at http://www.oreilly.com/catalog/python2/ you'll see that
> > Programming Python just got a second edition; I've heard that there were
> > a lot of improvements.
> > --
> >                       --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)
> >
> > Androgynous poly kinky vanilla queer het Pythonista
http://www.rahul.net/aahz/
> > Hugs and backrubs -- I break Rule 6
> >
> > "I won't accept a model of the universe in which free will, omniscient
> > gods, and atheism are simultaneously true."  -- M
>




From phd at phd.russ.ru  Mon Mar 19 10:22:36 2001
From: phd at phd.russ.ru (Oleg Broytmann)
Date: 19 Mar 2001 09:22:36 -0600
Subject: Dr. Dobb's Python-URL! - weekly Python news and links (Mar 19)
Message-ID: <F2D07C44A22A4BCE.5028A746960341BF.087428AC3B7E59F4@lp.airnews.net>

Listen to Uche Ogbuji:  "Type errors are not even close to the majority
of those I make while programming in Python, and I'm quite certain that
the code I've written in Python is much less buggy than code I've written
in strongly-typed languages.  Expressiveness, IMO, is a far better aid to
correctness than artificial restrictions (see Java for the example of
school-marm programming gone amok)."
    
    Finn Bock announces Jython 2.1 alpha 1
        http://mail.python.org/pipermail/python-announce-list/2001-March/000713.html
    
    Barry Warsaw releases Mailman 2.0.3
        http://mail.python.org/pipermail/python-announce-list/2001-March/000708.html
    
    Aahz Maruch makes PEP 6: Patch and Bug Fix Releases
        http://mail.python.org/pipermail/python-announce-list/2001-March/000714.html
    
    Moshe Zadka releases Numeric PEPs
        http://mail.python.org/pipermail/python-list/2001-March/033712.html
    
    Ka-Ping Yee: pydoc.org: Python Documentation Online
        http://www.pydoc.org/
    
    Python Asynconronous HTTP Client
        http://sourceforge.net/projects/asynchttp/
    
    LUCI is a Universal Configuration Interface
        http://jsautret.free.fr/luci/
    
    American students are painfully unprepared for life after graduation (joke)
        http://www.bbspot.com/News/2001/03/perl_test.html
    
    The Python Web Services Developer, Part 2
        http://www-106.ibm.com/developerworks/webservices/library/ws-pyth2/
    
    Python Web Application Tools: Zope, Webware, Quixote, Jython
        http://www.oreillynet.com/pub/a/python/2001/03/14/pythonnews.html
    
    Scott Hathaway announces sql2xml
        http://groups.yahoo.com/group/python-list/message/127724
    
    Chuck Esterbrook: Using Mix-ins with Python
        http://noframes.linuxjournal.com/lj-issues/issue84/4540.html
    
    Robin Dunn makes _socket.pyd for Python 2.0 that includes the OpenSSL support
        http://mail.python.org/pipermail/python-list/2001-March/033801.html
    
    
    Zope:
    Zope 2.3.1 beta 2 released
        http://lists.zope.org/pipermail/zope-announce/2001-March/000388.html
    A lot of bug fixes and new Btree data structures; ZCatalog upgrade recommended
    
    Hamish Lawson annonunces ZPT, a next-generation template technology
        http://mail.python.org/pipermail/python-list/2001-March/033834.html
    
    
    Hot threads of the week:
    This math scares me
        http://groups.google.com/groups?lr=&group=comp.lang.python.*&safe=off&ic=1&th=26a09eb0f1b8ea4a&seekd=914722604
    Einstein's Riddle
        http://groups.google.com/groups?lr=&group=comp.lang.python.*&safe=off&ic=1&th=86109686c16ab447&seekd=915376443
    Who's minister of propaganda this week?
        http://groups.google.com/groups?lr=&group=comp.lang.python.*&safe=off&ic=1&th=13481eebd2054e26&seekd=914450748
    


========================================================================

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

      Python.org's Python Language Website is the traditional
      center of Pythonia
          http://www.python.org
      Notice especially the master FAQ
          http://www.python.org/doc/FAQ.html

      PythonWare complements the digest you're reading with the
      daily python url
           http://www.pythonware.com/daily  

      comp.lang.python.announce announces new Python software.  Be
      sure to scan this newly-revitalized newsgroup at least weekly.
	  http://groups.google.com/groups?oi=djq&as_ugroup=comp.lang.python.announce

      Michael Hudson continues Andrew Kuchling's marvelous tradition
      of summarizing action on the python-dev mailing list once every
      other week.
	  http://starship.python.net/crew/mwh/summaries/
	  http://www.amk.ca/python/dev

      The Vaults of Parnassus ambitiously collect Python resources
          http://www.vex.net/~x/parnassus/   

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

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

      Cetus does much of the same
          http://www.cetus-links.de/oo_python.html

      Python FAQTS
          http://python.faqts.com/

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

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

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

      Archive probing trick of the trade:
	  http://groups.google.com/groups?oi=djq&as_ugroup=comp.lang.python&num=100


Previous - (U)se the (R)esource, (L)uke! - messages are listed here:
  http://purl.org/thecliff/python/url.html
or
  http://groups.google.com/groups?oi=djq&as_q=+Python-URL!&as_ugroup=comp.lang.python


Suggestions/corrections for next week's posting are always welcome.
[http://www.egroups.com/list/python-url-leads/ is hibernating.  Just
e-mail us ideas directly.]

To receive a new issue of this posting in e-mail each Monday morning,
ask <claird at neosoft.com> to subscribe.  Mention "Python-URL!".


-- The Python-URL! Team--

Dr. Dobb's Journal (http://www.ddj.com) is pleased to participate in and
sponsor the "Python-URL!" project.
-- 

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


From donn at u.washington.edu  Thu Mar  1 19:59:52 2001
From: donn at u.washington.edu (Donn Cave)
Date: 2 Mar 2001 00:59:52 GMT
Subject: define a new func on the fly?
References: <3A9EE927.8744154E@troikanetworks.com>
Message-ID: <97mra8$p30$1@nntp6.u.washington.edu>

Quoth Bruce Edge <bedge at troikanetworks.com>:
| I'm trying to define a new function on the at runtime,
|
| this fails:
|
| >>> eval( "def global_func_name():\n\tpass" )
| Traceback (most recent call last):
|   File "<stdin>", line 1, in ?
|   File "<string>", line 1
|     def global_func_name():
|       ^
| SyntaxError: invalid syntax
|
| any idea how I can do this?


   >>> exec "def global_func_name():\n\tpass\n"
   >>> global_func_name
   <function global_func_name at 0x8000ca34>
   >>>

	Donn Cave, donn at u.washington.edu


From support at internetdiscovery.com  Thu Mar 29 00:17:09 2001
From: support at internetdiscovery.com (Mike Clarkson)
Date: Thu, 29 Mar 2001 05:17:09 GMT
Subject: Jython and JPython Questions...How similar to Python, Tkinter?
References: <3ABFF14A.30D0096D@earthlink.net> <20010328114505.E8325@freedom.puma-ag.com> <mailman.985773639.18235.python-list@python.org>
Message-ID: <3ac2c36d.1544880@24.0.228.33>

On Wed, 28 Mar 2001 12:00:51 +0200, Steve Purcell
<stephen_purcell at yahoo.com> wrote:

>Steve Purcell wrote:
>> Now if someone ever comes up with a library that gives Swing/AWT a Python
>> interface, that would be another story...
>
>That should read 'a Tkinter' interface, obviously. Need more coffee...

Usually that kind of thing is very difficult, even with more coffee:
the event models are too different (allthough i don't know the
specifics of AWT/Swing). The only was to get really independent is an
XML UIML type of abstraction, like on mozilla.org, or inside
pythonworks. But that's higher up, and I'm not convinced myself that
that isn't too high to be practical as a UI coding method. We'll see.

But in the meantime, having Tkinter run on jython, even using JNI,
will be valuable to help test the CPython/jpython pair. Finn says it's
not too much work to bring jtkinter up to date with jython 2.x, and I
hope he finds the time to do so.

Mike.




From DavidA at ActiveState.com  Sat Mar  3 00:38:05 2001
From: DavidA at ActiveState.com (David Ascher)
Date: Fri, 2 Mar 2001 21:38:05 -0800
Subject: Editing Python source as XML?
In-Reply-To: <b_Tn6.57540$o85.322150@news-server.bigpond.net.au>
Message-ID: <GBEFIDBHIMEFEFACBMKFAEDMCAAA.DavidA@ActiveState.com>

Neil H:

> Adam Logghe:
> > www.activestate.com has a project called "Komodo" which is
> written with a
> > combination of the Mozilla codebase and Python (my understanding is that
> > originally this was also to have a substantial portion written
> in Perl but
> > the developers could code so much faster in Python that Perl
> ended up as a
> > minor part).
>
>     It'd be excellent to hear from ActiveState about their use of Python
> versus Perl here. Just a company statement that "Perl sucks"
> would do fine.
> ;)

Sorry, Neil, no such luck. =)

I'll say this:

 1) I'm the tech lead for Komodo
 2) Python is my favorite language
 3) Mark Hammond works on Komodo
 4) Python is a great component language
 5) This is not an official ActiveState statement.

As for Perl, well, Perl has its uses too.  =)

I'll give the long version of this argument next week at IPC9, if anyone's
interested.  See some of you there!

--David Ascher

Vote for Your Favorite Python & Perl Programming
Accomplishments in the first Active Awards!
http://www.ActiveState.com/Awards





From garry at sage.att.com  Tue Mar 13 14:50:36 2001
From: garry at sage.att.com (Garry Hodgson)
Date: Tue, 13 Mar 2001 19:50:36 GMT
Subject: IPC9 quotes wanted
References: <slrn9ai9ln.10q.amk@mira.erols.com>
Message-ID: <3AAE7A0C.DB72EF4E@sage.att.com>

"A.M. Kuchling" wrote:

> I intended to put a note on the message board about this, but
> forgot...  If you were at IPC9 and heard someone say something
> quotable, please let me know who it was and what they said, so that it
> can be added to the quotation file.  Thanks!

greg wilson, in his excellent "so you want to write about python?" 
tutorial, was speaking about the importance of source control.
i don't have the exact quote (my notes are elsewhere), but he said
something along the lines of:

	given the choice between a good text editor and a good source
	control system, i'll take the source control, and use "cat" to
	write my code.

excellent advice.

-- 
Garry Hodgson                   Once in a while
Senior Hacker                   you can get shown the light
Software Innovation Services    in the strangest of places
AT&T Labs                       if you look at it right


From immanuel at yours.com  Wed Mar 21 11:56:07 2001
From: immanuel at yours.com (Immanuel)
Date: Wed, 21 Mar 2001 08:56:07 -0800
Subject: pdb with multithreads & PyEval_CallObject
Message-ID: <Zo6u6.56$I77.29627@news.pacbell.net>

I apologize if this is a repost...but my previous posting did make it to the
newsgroup.

I have a Linux multi-threaded application which both extend & embed Python.
The threads are created in C++, and the entry point of the application is
started via
Python; one of these threads, however, have a PyEval_CallObject in it. Since
I need
to acquire the interpreter lock before calling PyEval_CallObject, this means
that pdb
is unable to break at the callback function. Does anyone know how to get
around this?
The alternative to debugging this is, of course, good old prints.

Thanks.
  Immanuel

Code snippet:

=======================
"main" python thread:

def cbfunc(arg):
    # callback func
    ...

app.start()         # threads created
app.installCallback(cbfunc)
...

=======================
"embedded" thread:
tstate = PyThreadState_New(interp)
Py_AcquireThread(tstate)
...
PyEval_CallObject(cbfunc)
....
PyThreadState_Clear(tstate)
Py_ReleaseThread(tstate)
PyThreadState_Delete(tstate)





From jmcmonagle at velseis.com.au  Wed Mar 21 23:27:11 2001
From: jmcmonagle at velseis.com.au (John McMonagle)
Date: Thu, 22 Mar 2001 14:27:11 +1000
Subject: Is there python code highlighting in emacs ?
References: <mailman.985223430.22832.python-list@python.org> <3AB978A9.353A0E4B@yahoo.com>
Message-ID: <3AB97F1F.70FF97DD@velseis.com.au>

> >
> > The hilit19 package in emacs provides keyword highlighting when editing
> > code.  For example, if you are writing in C, comments are italicised in
> > red, strings are in grey, import statements are in lavender, functions
> > are in bold-blue, etc.
> >
> > Does anyone know if there is python support in the hilit19 emacs lisp
> > package ?
> 
> I don't know about hilit19, but I use python-mode which highlights
> using font-lock-mode. Comes with GNU Emacs 20.whatever. Try M-x
> python-mode and see if anything happens.
> 
> python-mode does many other useful things, too. I can't imagine
> working without it at this point.
> 

I already use python-mode which works fine for indenting, etc.  How do I
configure this font-lock-mode for python  ? What do I add to my .emacs
file ?




From fredrik at pythonware.com  Wed Mar 14 03:14:01 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 14 Mar 2001 08:14:01 GMT
Subject: Unpickling from strings
References: <qHxr6.19658$pR3.3830844@news2-win.server.ntlworld.com> <3gyr6.159$4d3.34762@newsb.telia.net> <3AAEB597.4010007@ntlworld.com> <vjFr6.180$4d3.41807@newsb.telia.net>
Message-ID: <dLFr6.185$4d3.42620@newsb.telia.net>

Marcin Chady wrote:
> Both pickling and unpickling is done by the same script, and the classes
> are defined in a separate module imported by that script.

btw, looking at the first part of the pickled instance might
help you figure out what's wrong:

    (i<module>\012<class>\012...

make sure <module> and <class> is what they're supposed
to be.

Cheers /F




From lisowski.tomasz at sssa.NOSPAM.com.pl  Wed Mar 28 06:59:13 2001
From: lisowski.tomasz at sssa.NOSPAM.com.pl (Tomasz Lisowski)
Date: Wed, 28 Mar 2001 13:59:13 +0200
Subject: Making entrys work with tkinter.
References: <3AC1CB21.4BD10EE8@mek.dtu.dk>
Message-ID: <99sjii$3uu$1@news.tpi.pl>

Uzytkownik "Brian Elmegaard" <be at mek.dtu.dk> napisal w wiadomosci
news:3AC1CB21.4BD10EE8 at mek.dtu.dk...
> Hi group,
>
> I have been trying to find out how to use the an entry and grab the
> result. I am sorry, that I have not been able to figure out from
> different sources how bind the entry to an event and make it put back
> the result to the calling function.
> What I try is:
>
> class Sheetlabel:
>     def __init__(self,canvas,x,y):
>         self.frame=Toplevel(width=10,height=20)
>         self.frame.title('Insert label')
>         Label(self.frame,text='Label:').pack(side='left')
>         self.entry=Entry(self.frame)
>         #FIX THIS
>         self.frame.bind('<Return>',self.ok(x,y,canvas))
>         self.entry.pack(side='right')
>         self.frame.focus_set()
>
>     def ok(self,x,y,canvas,event=None):
>         self.text=self.entry.get()
>         self.frame.destroy()
>         canvas.create_text(20+x,20+y,text=self.text)
>         return self.text
>
> which should put the label text on the canvas. With a create_text in
> __init__ there is no problem. So somehow I think I am struggling with
> scopes. Is that right?

If I am not mistaken, even callbacks can take only one argument - the event
object. All other arguments must have default values. In your case, you want
to call a callback with x, y, and canvas, as arguments. This is not
supported.
Better store x, y, and canvas as instance attributes, and use them in the
ok() method.

Tomasz Lisowski




From tim.one at home.com  Fri Mar  2 14:01:23 2001
From: tim.one at home.com (Tim Peters)
Date: Fri, 2 Mar 2001 14:01:23 -0500
Subject: python-dev summary 2001-02-15 - 2001-03-01
In-Reply-To: <Pine.LNX.4.10.10103021258550.18596-100000@localhost.localdomain>
Message-ID: <LNBBLJKPBEHFEDALKOLCOEHHJDAA.tim.one@home.com>

[Michael Hudson]
>  ...
>  The iteraators discussion seems to have disappeared.  At least, your
>  author can't find it!

Greg Wilson set up an egroups (now Yahoo) discussion group for it, where
debate has continued:

    http://groups.yahoo.com/group/python-iter

Marc-Andre Lemburg set up another (or agitated toward this end) mailing list
for it on SourceForge, for reasons that escape me at the moment.  I never got
around to subscribing to that too; the link on SF:

    http://www.geocrawler.com/redir-sf.php3?list=python-iterators

appears broken, and the archive link:

    http://lists.sourceforge.net/archives//python-iterators/

leads to a page that doesn't make sense.

hoping-that's-not-wholly-prophetic-ly y'rs  - tim




From tuttledon at hotmail.com  Wed Mar 21 23:28:38 2001
From: tuttledon at hotmail.com (Don Tuttle)
Date: Thu, 22 Mar 2001 04:28:38 GMT
Subject: Standalone executables
References: <998igr$4al$1@niagara.nonet> <iXWt6.156032$Z8.33024358@typhoon.southeast.rr.com> <99aldn$7dh$1@niagara.nonet> <99aonq$a4c$0@216.39.170.247>
Message-ID: <Wbfu6.145012$__6.26818921@typhoon.southeast.rr.com>

"Dave LeBlanc"
> I found this on c.l.p.announce:
> The Linux version of Installer for Python 2 is here:
>  http://www.mcmillan-inc.com/installer_dnld.html

Mark, judging from comments posted here,  your main issue of which
installer/pseudo compiler to use will depend on you choice of GUI.  py2exe =
wxPython.  Tkinter = Gordon's Installer.  So as tools stand now, the
question is which is more important,  using wxPython, or distributing to
both Windows & Linux. (Or maybe you could port py2exe to Linux for Thomas
;-)

Don




From rikbo716 at student.liu.se  Wed Mar 21 07:39:30 2001
From: rikbo716 at student.liu.se (Rikard Bosnjakovic)
Date: Wed, 21 Mar 2001 13:39:30 +0100
Subject: Bug or feature?
References: <99a433$b8b$1@sun.rhrk.uni-kl.de>
Message-ID: <3AB8A102.108AA8B3@student.liu.se>

Markus Reitz wrote:

> I expected a TypeError message, because a String is not a list and so it
> can't be used to extend an existing list.

No, but a String is a sequence.  Try [].extend(4711) and you will get
the following error:

TypeError: list.extend() argument must be a sequence

And a sequence could be parsed like what you wrote, "a list of
characters".


-- 
Rikard Bosnjakovic - http://a214.ryd.student.liu.se/cv/ - ICQ: 1158217

Anyone sending unwanted advertising e-mail to my address will be
charged $250 for network traffic and computing time. By extracting my
address from this message or its header, you agree to these terms.


From ws-news at gmx.at  Fri Mar  9 05:21:09 2001
From: ws-news at gmx.at (Werner Schiendl)
Date: Fri, 9 Mar 2001 11:21:09 +0100
Subject: Can Python replace Visual Basic? Should it?
References: <5DZo6.11126$hn5.1578410@news1.rdc1.mb.home.com> <mailman.984010931.26257.python-list@python.org> <iSBp6.14005$hn5.2035325@news1.rdc1.mb.home.com> <mailman.984094514.29514.python-list@python.org>
Message-ID: <984133550.452523@newsmaster-04.atnet.at>

Well, I know Visual Basic pretty well and used it for some projects in the
past.
Mainly to develop custom ActiveX controls with a proprietary data interface
and special requirements.

I _really_ enjoyed the ease of that process in Visual Basic compared to, e.
g. Visual C++'s MFC and ATL.
Of course on is limited in some ways, but for us it did a perfect job.

It's a bad idea to do rather low level stuff in Visual Basic, because its
really painful to juggle bits and bytes in VB.
But you can code that part in a DLL and use those functions from VB (does
not need to be a COM server, just a plain DLL is sufficient, although a COM
server integrates much better with VB...)

What I am still missing in all the alternatives for Windows programming is
the easy integration with the native Windows component model (DCOM). Doing
it manually is really a pain and completely unnecessary for so many
programming tasks.

If you are programming for purity in code, if OO is theoretical stuff for
you consisting of inheritance, polymorphism and encapsulation, or if a
really wonderful intuitiv IDE with far the best autocomplete and library
browsing features I have seen so far is just pain for you because you prefer
coding the GUI in plain text -> do not consider Visual Basic

Otherwise its a neat tool for the purposes mentioned.

I did not do much COM stuff in Python right now, but I think you cannot
compare it to Visual Basic at this time.
Of course, Mark Hammond does a really impressing job on that issue, but
Microsoft is still somewhat in advance with VB I think.

hth
werner


D-Man <dsh8290 at rit.edu> wrote in message
news:mailman.984094514.29514.python-list at python.org...
> On Thu, Mar 08, 2001 at 02:10:22AM +0000, Brad Bollenbach wrote:
> [snip]
> |
> | Maybe the fact that only one person who saw my original post has
> | actually migrated from VB to Python is an answer in itself.
>
> Maybe there are other reasons entirely.  For example, I am not a VB
> programmer, nor have I ever used VB.  Maybe there just aren't many
> people on this list who have used VB, or more specifically used VB
> before using Python.  And again, maybe not.  I come from an Eiffel,
> C++, Java, C background (pretty much in that order) mostly on *nix
> systems.
>
> -D
>
>




From aleaxit at yahoo.com  Fri Mar  9 07:24:55 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 9 Mar 2001 13:24:55 +0100
Subject: COM support (was Re: Can Python replace Visual Basic? Should it?)
References: <5DZo6.11126$hn5.1578410@news1.rdc1.mb.home.com> <mailman.984010931.26257.python-list@python.org> <iSBp6.14005$hn5.2035325@news1.rdc1.mb.home.com> <mailman.984094514.29514.python-list@python.org> <984133550.452523@newsmaster-04.atnet.at>
Message-ID: <98ai2j01o8j@news2.newsguy.com>

"Werner Schiendl" <ws-news at gmx.at> wrote in message
news:984133550.452523 at newsmaster-04.atnet.at...
    [snip]
> I did not do much COM stuff in Python right now, but I think you cannot
> compare it to Visual Basic at this time.
> Of course, Mark Hammond does a really impressing job on that issue, but
> Microsoft is still somewhat in advance with VB I think.

Speaking as somebody whose main bill-paying-techolological-expertise
right now is COM -- I disagree.  VB may well be better for those who
_lack_ significant COM expertise and still need to handle a lot of COM
stuff -- if nothing else, because it does a good job of handling the
visual parts of Active/X, and comes with a zillion wizards which (not
always successfully) strive to hide the complexities from you.  But
if one DOES know COM (or needs to do complex-enough stuff that the
knowledge of COM itself is not optional anyway), Python is better --
because it does _NOT_ strive to hide things behind your back, you are
in far better control -- essentially the same potential for "control
of what's going on" that you get with VC++ and ATL, but without the
horrid complexities thereof.

Like with ATL, though, you do have to study the sources you're supplied;
the docs are not bad, but they never go deep enough into the system's
bowels -- only study of sources can take you there.
[[However, this also applies to theoretically "simple" stuff, as soon as
you need to answer questions that are deep enough in themselves, e.g.,
"if I malloc a zillion byte then free them, will the memory be given
back to the system or will my process keep it to itself for possible
future needs" -- this can be crucial in practice, but only studying
the sources of your C system's runtime libraries will help you answer.]]


Alex





From tim.one at home.com  Tue Mar 13 04:25:11 2001
From: tim.one at home.com (Tim Peters)
Date: Tue, 13 Mar 2001 04:25:11 -0500
Subject: Einstein's Riddle
In-Reply-To: <NI9r6.2583$54.2812@www.newsranger.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCCEEMJFAA.tim.one@home.com>

[Gregory Jorgensen]
> So I'm stupid, is that it?

Hmm.  If you're convinced that assuming that someone owned a fish is an
unjustified leap of irrational faith, I'm not sure you should trust yourself
to read between the lines in real life either <wink>.

> ...
> Actually I did figure out that the German owned the fish. Furthermore
> I was able to determine that the fish is named Eric.

Indeed, not only that, but Eric is also the "Sweed" living in the white house
too.  The puzzle didn't say that the houses were distinct, or that each was
only of a single color, or that the people living in the houses weren't also
pets with respect to the other people <ahem>.

poor-eric-the-swedish-fish-owned-by-herman-the-german-ly y'rs  - tim




From kwoeltje at mail.mcg.edu  Thu Mar 29 07:20:06 2001
From: kwoeltje at mail.mcg.edu (Keith F. Woeltje)
Date: Thu, 29 Mar 2001 12:20:06 GMT
Subject: Recommendations for Python Books
References: <3ac2582c$1@news.starhub.net.sg>
Message-ID: <3AC32878.EFC2248E@mail.mcg.edu>

If you want free, look at the Python web site page listing introductions
( http://www.python.org/doc/Intros.html ). The tutorial that comes with
Python is frequently recommended as a starting point, but I found it too
terse.

For someone completely new to programming, /Learn to Program Using
Python/ by Gauld would be a good first book, but you would need to
follow it up with the tutorial or another book. My favorite book so far
is /The Quick Python Book/ by McDonald, but I also liked /Learning
Python/ by Lutz and Ascher. Both books cover the basics, but their
organization and emphasis is different. I've looked at /Sams Teach
Yourself Python in 24 Hours/ by Van Laningham at Borders. It's
organization and content seem fine, but they use relatively small screen
shots to display example program output, and I didn't like that. /Core
Python/ by Chun looks like it might also be an OK starting point, but
I've only glanced through it.

O'Reilly has released a second edition of /Programming Python/ covering
Python 2.0, but it seems to be aimed at the intermediate to advanced
Python programmer, and probably isn't the best choice for starting out.

HTH, and good luck.
-KFW

Wong Da Qiang wrote:
> 
> Just a newbie to programming and python, need some recommendations on books
> to buy


From winson at mdsserv.mds.com.tw  Mon Mar 19 09:17:41 2001
From: winson at mdsserv.mds.com.tw (???)
Date: Mon, 19 Mar 2001 22:17:41 +0800
Subject: Tkinter Support
References: <001901c0b046$1f1a7200$340210ac@winson> <3AB5C475.AF89FD43@ActiveState.com>
Message-ID: <000901c0b07f$603b2d40$340210ac@winson>

Sir,

I follow your procedure but I got following message:

"Connectint to SOAP server ActiveState Package Repository..." and

Error(10060): Operation Timeout...

--
Best Regards,

Winson Chang

3F, 2, 150 Lane, Sec. 5, Hsin I Rd., Taipei, Taiwan, R.O.C.

Tel: +886-2-27225333 ext 376
Fax: +886-2-27222330

Email: winson at mdsserv.mds.com.tw
----- Original Message -----
From: "Paul Prescod" <paulp at ActiveState.com>
To: "?i???q" <winson at mdsserv.mds.com.tw>; <python-list at python.org>
Sent: Monday, March 19, 2001 4:33 PM
Subject: Re: Tkinter Support


> ActivePython comes with TKinter but not Tk. You can use the package
> manager utility to install it (see the ActivePython start menu). Just
> type "install tk" from the PPM command line.
> --
> Take a recipe. Leave a recipe.
> Python Cookbook!  http://www.activestate.com/pythoncookbook
>
> --
> http://mail.python.org/mailman/listinfo/python-list
>




From aleaxit at yahoo.com  Fri Mar  9 10:33:47 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 9 Mar 2001 16:33:47 +0100
Subject: Can anyone recomend a good intoduction to C...
References: <mailman.983894288.19965.python-list@python.org> <3AA5EF59.507F1D91@olsen.ch> <983958878.976344@newsmaster-04.atnet.at> <77up6.1289$y6.235541@ruti.visi.com>
Message-ID: <98at5o02538@news2.newsguy.com>

"Grant Edwards" <grante at visi.com> wrote in message
news:77up6.1289$y6.235541 at ruti.visi.com...
    [snip]
> >To be completely honest, I can not see the obvius reason why
> >C++ should be such a bad programming language.
>
> It's complicated and unsafe.  Either one is tolerable without
> the other.  When you combine the two, you're dead.

Not necessarily.

> >It supports for a real large number of different design approaches.
>
> That isn't always a good thing. ;)

Let's say that being an excellent designer IS part of the
survival-necessary conditions for C++ (but that does extend
to C, for non-trivial applications)...:-).

Assuming an excellent designer, it IS a very good thing
that he or she can fully exercise his or her design flair.

"Multi-Paradigm DESIGN for C++", by Coplien, is a small
book, but well worth reading.  (As it happens, the breadth
of design approaches that Python affords isn't really any
narrower than those afforded by C++, so, while you surely
can survive without such excellent design skills in Python,
including ability to choose the ideal mixture of paradigms
and interfacing them seamlessly, having such skills is
anything but a waste -- it will let you do an insanely great
job rather than merely a superb one!-)


Alex





From mark at chaos.x-philes.com  Wed Mar 21 11:38:25 2001
From: mark at chaos.x-philes.com (Mark Carroll)
Date: Wed, 21 Mar 2001 16:38:25 GMT
Subject: Standalone executables
References: <998igr$4al$1@niagara.nonet> <iXWt6.156032$Z8.33024358@typhoon.southeast.rr.com>
Message-ID: <99aldn$7dh$1@niagara.nonet>

In article <iXWt6.156032$Z8.33024358 at typhoon.southeast.rr.com>,
Don Tuttle <tuttledon at hotmail.com> wrote:
(snip)
>On the Windows side, check out
>http://starship.python.net/crew/theller/py2exe/ to compile and
>http://www.jrsoftware.org/isinfo.htm to install your program.  Can't help
>you on the Linux side.

Thanks very much!

I wonder, for Linux, what if I wrote in Jython, compiled with jythonc,
then used a Java native-code compiler? Would that be too insane? (-:

-- Mark


From wware at world.std.com  Thu Mar  8 08:03:59 2001
From: wware at world.std.com (Will Ware)
Date: Thu, 8 Mar 2001 13:03:59 GMT
Subject: News from Python9!!!
References: <NUCp6.65206$Z8.11697516@typhoon.southeast.rr.com> <00ac01c0a785$efa17a00$a100000a@local> <mailman.984039492.28445.python-list@python.org>
Message-ID: <G9vqyn.2M3@world.std.com>

Chris Gonnerman wrote:
> I've just got one problem...  When I go to the links given, I get this:
> <html><head><title>Site Not Found</title></head>
> <body>No web site is configured at this address.</body></html>

Martin Franklin (martin.franklin at westerngeco.com) wrote:
> If your mail client is anything like mine it wraped the URL so that you 
> lost the .html at the end...

Nope, that's not it. Even when I manually add the ".html" I get the
same thing. At least, I did last night, but this morning it appears
to be fixed.

-- 
-----------------------------------+---------------------
 22nd century: Esperanto, geodesic | Will Ware
 domes, hovercrafts, metric system | wware at world.std.com


From danielk at aracnet.com  Mon Mar  5 01:35:27 2001
From: danielk at aracnet.com (Daniel Klein)
Date: Sun, 04 Mar 2001 22:35:27 -0800
Subject: Overriding UserList methods
References: <b566atck0ruu4tae8n1srrkntbmme26gor@4ax.com>
Message-ID: <2oc6atgkn0ror77rj3b6j8d8d2s0gaa5k7@4ax.com>

I found the answer in the Beazley book. It's

def append(self, object):
        self.data.append(object)

using the 'data' instance variable automatically set up in the UserList class.

But I'll try the other suggestions anyway just to get a better understanding.

Thanks,
Dan


From dsh8290 at rit.edu  Wed Mar  7 17:51:17 2001
From: dsh8290 at rit.edu (D-Man)
Date: Wed, 7 Mar 2001 17:51:17 -0500
Subject: Java guy interested in Python
In-Reply-To: <003801c0a459$a29c97f0$63da1542@wisdom>; from kknbk2@bellsouth.net on Sat, Mar 03, 2001 at 09:17:24PM -0600
References: <003801c0a459$a29c97f0$63da1542@wisdom>
Message-ID: <20010307175117.B8@harmony.cs.rit.edu>

On Sat, Mar 03, 2001 at 09:17:24PM -0600, Kyle Wayne Kelly wrote:
| 
|    Can anyone tell me some of the things that Python offers which are not
|    available in Java?  I like Python's syntax, but what can Python do
|    that Java can not do?  Any comments are helpful.

In addition to other non-Java related advantages, Python (through
using the Jython interpreter) can access the Python Standard Library
and the Java Standard Library simultaneously and painlessly all while
using the nice clean and comfortable syntax of Python.  

:-)

-D

(see http://jython.sourceforge.net)



From tim.one at home.com  Fri Mar  2 13:30:54 2001
From: tim.one at home.com (Tim Peters)
Date: Fri, 2 Mar 2001 13:30:54 -0500
Subject: gah! I hate the new string syntax
In-Reply-To: <XFMail.20010302084209.shaleh@valinux.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCIEHFJDAA.tim.one@home.com>

[Sean 'Shaleh' Perry]
> good ole:
>
> from string import join
>
> l = join(';', ['a','b','c']) # thanks, this is clear and obvious
>
> Please o' gods of python, do not deprecate the clean and obvious
> syntax that is found in the string module.

Sean, that claim of "obvious" would have been a lot more convincing if you
had managed to remember the correct order of string.join's arguments <wink>.




From hgg9140 at cola.ca.boeing.com  Fri Mar  9 17:21:41 2001
From: hgg9140 at cola.ca.boeing.com (Harry George)
Date: Fri, 9 Mar 2001 22:21:41 GMT
Subject: how to make a function thread-safe ?
References: <3AA94A6A.50138001@noaa.gov>
Message-ID: <xqxsnkm1t3e.fsf@cola.ca.boeing.com>

A true function with no global side-effects should be thread-safe.
[Or at least that's true in any language I've multi-threaded so far.]

If there are global side effects, you need to capture them in some
structure (with a class being a convenient format), and pass that (one
per thread) to the function.

Your suggestion is one of several syntactic alternative to passing a
structure: make the function a method of the class which has the
per-thread data.

j vickroy <jim.vickroy at noaa.gov> writes:

> If a function may be applied (simultaneously) by multiple threads is the
> safe (correct) procedure to wrap the function inside a class and have
> each thread create its own instance of the class?
> 
> Thanks

-- 
Harry George                E-mail:  harry.g.george at boeing.com
The Boeing Company          Renton:  (425) 237-6915
P. O. Box 3707  02-CA       Everett: (425) 266-3868
Seattle, WA 98124-2207      Page:    (425) 631-8803  


From ark at research.att.com  Mon Mar 12 22:25:39 2001
From: ark at research.att.com (Andrew Koenig)
Date: Tue, 13 Mar 2001 03:25:39 GMT
Subject: This math scares me
References: <3aad4062.1290021@News.CIS.DFN.DE> <Hwbr6.1139$Tg.127571@ruti.visi.com> <3aad4485.2348499@News.CIS.DFN.DE> <jwbnews-7CC03F.19050312032001@corp.supernews.com>
Message-ID: <yu993dcimjt8.fsf@europa.research.att.com>

John> Many of us feel that the present truthful display is correct, and the 
John> former (before Python 2) display of 10.55 was wrong because it was a lie 
John> (some don't feel quite *that* strongly about it but still prefer the new 
John> form).

I'm curious -- Does Python now follow either the IEEE or the more
restrictive Scheme rules for floating-point conversion?

Briefly, IEEE requires that input and output conversion have no more
than 0.47 LSB error; Scheme requires that input conversion be exact
(that is, that the internal representation always be the closest
(according to the rounding rules currently in effect) internal number
to the infinite-precision value read, and that output yield the
smallest number of significant digits necessary for idempotence).



-- 
Andrew Koenig, ark at research.att.com, http://www.research.att.com/info/ark


From schaffer at optonline.net  Mon Mar 19 20:33:07 2001
From: schaffer at optonline.net (Les Schaffer)
Date: Tue, 20 Mar 2001 01:33:07 GMT
Subject: VPython/Tk app works on single processor but not dual
References: <995gi1$9ce$1@bob.news.rcn.net>
Message-ID: <8fcdbt486b73nftq3re889ldsc9s32ndlk@4ax.com>

On 19 Mar 2001 17:44:33 GMT, fcy at enteract.com (Fred Yankowski) wrote:

>This program
>works fine on WinNT if I "set affinity" of the python.exe process so
>that it's forced to run on a single processor, but if I let it run
>normally on my dual-processor system, the application hangs in short
>order, refusing to respond to mouse actions in either the Tk or
>VPython display windows.

works fine for me on Win2K and dual PIII 800 machine. 

>The program itself is available, for now, at
><http://ontosys.com/src/residue.py>.  I'm using Python 2.0

ditto...

> and VPython 2001-02-27 

the top of __init__.py in visual subdir says: "Visual-2000-11-26",
though i could have sworn i just built this last month with fresh
source. i'll try your version if indeed there is something newer.

>on NT4SP6, 

win2k

> on dual Pentium-III 800 processors.  

ditto

>I have version 8.3.2 of tk83.dll and tcl83.dll.

ditto (it actually says 8.3.2.2).

video card = Geforce II GTS, Tyan motherboard

les schaffer


From tjg at exceptionalminds.com  Wed Mar 21 17:59:01 2001
From: tjg at exceptionalminds.com (Timothy Grant)
Date: Wed, 21 Mar 2001 14:59:01 -0800
Subject: Dynamically reloading extension modules
In-Reply-To: <007f01c0b252$852bb370$04090a0a@upcast.com>; from kens@sightreader.com on Wed, Mar 21, 2001 at 02:01:44PM -0800
References: <007f01c0b252$852bb370$04090a0a@upcast.com>
Message-ID: <20010321145901.M24341@trufflehunter.avalongroup.net>

On Wed, Mar 21, 2001 at 02:01:44PM -0800, Ken Seehof wrote:
> Has anyone found a technique for dynamically unloading and
> reloading an extension module?  I want to be able to modify,
> build, and reload an extension module from within my program.
> 
> Unfortunately python apparently doesn't use FreeLibrary when
> a module object is released.  Maybe I can do some kind of
> hack to call FreeLibrary explicitly.  Is there a way to do this that
> is portable?

I'm not sure I know what your asking, but does

reload(modulename) fit the bill?


-- 
Stand Fast,
    tjg.

Timothy Grant                         tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.         <><       (503) 246-3630
>>>>>>>>>>>>>Linux, because rebooting is *NOT* normal<<<<<<<<<
>>>>This machine was last rebooted:  64 days  3:14 hours ago<<



From phlip_cpp at my-deja.com  Tue Mar  6 14:17:21 2001
From: phlip_cpp at my-deja.com (Phlip)
Date: Tue, 6 Mar 2001 11:17:21 -0800
Subject: Long Shifty Bits
References: <kmap6.15$8Q.28605@news.pacbell.net> <m3itlmbu3g.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <MLap6.18$8Q.33391@news.pacbell.net>

Michael Hudson wrote:

> Works for me.  Have you tried it?

>>> x = 1L << 200
>>> x
1606938044258990275541962092341162602522202993782792835301376L

Ouch. Just trying it did not occur to me. The expression's deep inside a 
blob of logic, and I just assumed when I hit a bug up around 32 elements 
that I had misrepresented <<'s prowess.

-- 
  Phlip                          phlip_cpp at my-deja.com
============== http://phlip.webjump.com ==============
  --  My opinions are those of your employer  --


From aleaxit at yahoo.com  Wed Mar 28 11:25:30 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Wed, 28 Mar 2001 18:25:30 +0200
Subject: Semi-final code as text...
References: <3ABCC757.6C6F17CA@earthlink.net> <3ABE73B2.CAFCA177@student.liu.se> <3ABFEC14.BD3B0709@earthlink.net> <3ABFF28E.EE28A322@earthlink.net>
Message-ID: <99t39q02t0h@news2.newsguy.com>

"Ron Stephens" <rdsteph at earthlink.net> wrote in message
news:3ABFF28E.EE28A322 at earthlink.net...
    [snip]
Using 'arrays' (a generic term; in Python, they are called
lists) and loops will save you a lot of repetitive code:

> x = 1
> option1 = raw_input ("what is option1:")
> x = x + 1
> option2 = raw_input ("what is option2:")
> x = x + 1
> while x:
>  if x > number: break
>  option3 = raw_input ("what is option3:")
>  x = x + 1
>  if x > number: break
>  option4 = raw_input ("what is option4:")
>  x = x + 1
>  if x > number: break
>  option5 = raw_input ("what is option5:")
>  if x > number: break
>  option6 = raw_input ("what is option6:")
>  break

Change into:

options = []
for x in range(0, number):
    options.append(raw_input("what is option%s:"%(x+1)))

[and similarly for other 'arrays' that you're simulating
 with bunches of similarly-named variables: criteria1 etc,
 rank1 etc, oneone and onetwo and ... and twoone and ...];

whenever you need to access the i-th option (for example),
you'll write options[i] (where you'd now repeat the code
with option1, option2, etc) -- just remember that i goes
from 0 to N-1 for some N, not from 1 to N.


This will make your code drastically smaller and simpler
and remove artificial limits on number of criteria and
options.


Alex





From claird at starbase.neosoft.com  Mon Mar  5 07:38:00 2001
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 5 Mar 2001 06:38:00 -0600
Subject: Questions from a beginner
References: <1103_983792888@breakdown>
Message-ID: <7D924B83F563035A.78ADE854BCA8F2C6.DA35B8F7D112D810@lp.airnews.net>

In article <1103_983792888 at breakdown>,
Alison Myers  <squirmy at freeuk.com> wrote:
>Hi. I'm a beginning programmer at a fairly late age and have been looking for a language to learn.
>I have decided against starting with C or C++ because I think they will be too complicated and they seem to be only really needed for writing device drivers and operating systems, which is not what I wat to do.
>Is python a good language to learn first? Will my knowledge help me with other languages later?
>I will be writing for win98 and Linux and I know Python is available on both of these, but would the code be transportable, especially if I write things using tcl (this is probably a long way off yet, but I'd like to know).
>I have links to online tutorials, but where would I find good source code?
>Do I need to buy any books to help me and if so which do you recommend?
>Is there a list of programming projects anywhere or would I get these from a textbook?
>I probably need to brush up on my maths, so what specific topics should I look at?
			.
			.
			.
Most of the answers are, "Yes."  I've argued before that
Python is the best first language
<URL: http://www.oreillynet.com/pub/a/network/2000/06/02/magazine/python_first_language.html >.
-- 

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


From kens at sightreader.com  Wed Mar 21 13:53:45 2001
From: kens at sightreader.com (Ken Seehof)
Date: Wed, 21 Mar 2001 10:53:45 -0800
Subject: learning python...
References: <3AB8ECB5.2197E940@noaa.gov>
Message-ID: <003001c0b238$474fd1e0$04090a0a@upcast.com>

Count me as one vote for "Learning Python".

"Learning Python" is somewhat easier to read than "Programming
Python" while being more compact too.

The only material that I noticed in "Programming Python" that is
missing from "Learning Python" is the extending and embedding
stuff, which is an advanced topic that you probably don't need to
worry about quite yet.

Mark Lutz applied his experience from writing "Programming
Python" to "Learning Python", so it's simply a better book.

I haven't read "Core Python Programming".

----- Original Message ----- 
From: "Benjamin.Altman" <benjamin.altman at noaa.gov>
Newsgroups: comp.lang.python
To: <python-list at python.org>
Sent: Wednesday, March 21, 2001 10:02 AM
Subject: learning python...


> Hello.  From the perspective of someone who already has programming
> experience, would anyone know if O'reillys "Learning Python" book would
> be better than say Prentice Hall's "Core Python Programming"?  Or would
> it be better to go straight into something like "Programming Python"?
> 
> Thanks,
> Ben
> 




From paulp at ActiveState.com  Sun Mar 18 12:48:20 2001
From: paulp at ActiveState.com (Paul Prescod)
Date: Sun, 18 Mar 2001 09:48:20 -0800
Subject: [Python-Dev] Revive the types sig?
References: <jeremy@alum.mit.edu> <15020.9404.557943.164934@w221.z064000254.bwi-md.dsl.cnc.net> <200103120711.AAA09711@localhost.localdomain> <mailman.984439040.24419.python-list@python.org> <s44tatolisnipdfm4r4p1aqt2e09e458hk@4ax.com> <mailman.984581571.10100.python-list@python.org> <1pWvOlsMJ9Zxk0d4s9BhFmMCSusy@4ax.com>
Message-ID: <3AB4F4E4.6C6E83AC@ActiveState.com>

Paul Moore wrote:
> 
> ....
> 
> I wasn't suggesting "total" polymorphism (although I did overstate my case a
> bit). What I was pointing out was essentially that it is possible to define
> totally "plug-compatible" replacements for builtin types, and have them work
> everywhere the built-in type does. With the exception of code that does "if
> type(x) != type(1)".

I think that making plug-compatible subclasses for the built-in types is
extremely difficult and rarely accomplished. The built-in types aren't
even plug-compatible with each other: integers can't be used everywhere
floats can nor vice versa. Sometimes you get lucky. Sometimes you don't.
For instance can you today make a non-integer object that can be used as
an index for a standard Python list? I would be surprised.

It's important to remember that Python objects do not typically go very
long before being passed into the C layer. And the C layer already has a
strict dynamic type annotation system.

Another idea: type annotations move you CLOSER to the world of total
type interoperability because it provides an obvious place to hook in
conversion code from your "int-like" object to a real integer that
really could be used to index a list.

>...
> Of course, if type annotations will *actually* formalise and support
> interfaces in some form, that's entirely different - I'm mostly neutral, about
> something like this.
> 
>     def f(n):
>         assert <n supports the "integer" interface>

Yes, type annotations will certainly support interfaces/protocols or
whatever you want to call them. On the other hand, after a first stab at
protocols for basic numeric types I gave up. Python is not quite that
flexible and polymorphic.

I'm not sure if any language is that flexible. One of the defining
characteristics of the basic types such as integer and long is that
Python knows how to manipulate their bit patterns in memory. The only
way you could make a type that behaves "exactly like an integer" is to
have a conversion to an integer -- which isn't quite the same.

>...
> My concern is that over-use of assertions like this limit "creative use" of
> functions, in contexts which the author may never have envisioned.

There are two different issues here. One is "creative use" where a
sophisticated programmer understands all of the code being used and is
confident that an object will "just work" despite the fact that the code
was not designed to be polymorphic. Most often, under a more strongly
checked type system, you would use inheritance to make this work in the
future.

The second issue are documented functions or methods that work with a
variety of objects that "look the same." In this case, the type system
uses interfaces to formalize what you've already being doing.

> The best example to date in Python is "file-like" objects. There are *lots* of
> library functions which take such objects. Not just files, but also
> user-defined objects which act like files. And many such functions explicitly
> state that they only require a subset of the full file interface. (Often not
> requiring seekability, sometimes requiring nothing more than readline).

Yes, this will use interfaces. Remember that a ton of modules that work
on file-like objects are *already type-checked*. They are C extensions
that pass their data through PyArg_ParseTuple.

-- 
Take a recipe. Leave a recipe.  
Python Cookbook!  http://www.activestate.com/pythoncookbook



From donn at oz.net  Thu Mar 22 23:10:49 2001
From: donn at oz.net (Donn Cave)
Date: 23 Mar 2001 04:10:49 GMT
Subject: Multiple Processes on Unix with Python
References: <3ABA8919.835D8BED@centropolisfx.com>
Message-ID: <99eic9$i06$0@216.39.151.169>

Quoth Nickson Fong <nickson at centropolisfx.com>:
| Can anyone show me a quick example on how to fork multiple processes on
| a multiple cpu Irix machine. I know I need os.fork() and os.execvp() but
| can't figure out how they work together.

The reason it's so hard, is that there are so many different things
you might want to do.  One thing you might be missing is os.waitpid(),
but it's hard to say.

Just for something completely different, take a look at this:

    import os

    p = []
    p.append(os.spawnv(os.P_NOWAIT, '/bin/date', ('date',)))
    p.append(os.spawnv(os.P_NOWAIT, '/bin/date', ('date',)))
    p.append(os.spawnv(os.P_NOWAIT, '/bin/date', ('date',)))

    while p:
        for i in range(len(p)):
            pi, st = os.waitpid(p[i], os.WNOHANG)
            if pi == p[i]:
                del p[i]
                print 'pid', pi, 'status', st
                break

	Donn Cave, donn at oz.net


From whisper at oz.net  Thu Mar 29 00:36:49 2001
From: whisper at oz.net (Dave LeBlanc)
Date: 29 Mar 2001 05:36:49 GMT
Subject: newbie question
References: <73125560B5E4EF16.CC3FC0DB3FD693E3.4AF0ACCBD2EFF29B@lp.airnews.net> <99fll8$996$1@rex.ip-plus.net> <mailman.985367791.3511.python-list@python.org> <mailman.985375195.17238.python-list@python.org> <84EF7ECA6B21860E.2BACA551CEE41AA7.AE240BE7DBA69E93@lp.airnews.net>
Message-ID: <99uhlh$lrt$0@216.39.170.247>

Hi;

I've installed both Zope and Apache, and found that Apache was harder
to install then Zope due to poor configuration documentation (a
listing of all the configuration directives without any description of
their use or any sort of "big picture" tutorial doesn't help).

However, Apache is considerably easier to use then Zope, since Zope
lacks (currently - i've heard a book is on the way) any real sort of
tutorial that explains simple things like it's architecture very well
or how to do basic tasks, how the control panel is organized etc.  I
even tried a pretty specific tutorial and it didn't work.

As long as I have anybody's attention, can anybody recommend a _good_
tutorial for Apache? All i've seen are  rehashs of the configuration
instructions that come with Apache and that's just not good enough!

Of course, it's hard to beat an infinite price/performance ratio (_if_
you can get it to perform!) :-)

Regards,

Dave LeBlanc

On 23 Mar 2001 14:42:33 -0600, claird at starbase.neosoft.com (Cameron
Laird) wrote:

>In article <mailman.985375195.17238.python-list at python.org>,
>Moshe Zadka  <moshez at zadka.site.co.il> wrote:
>>On 23 Mar 2001, claird at starbase.neosoft.com (Cameron Laird) wrote:
>>
>><snipped a lot of wise things about CGI and Apache>
>>> That's a lot more exciting
>>> for them than hearing that, with a mere week or two of
>>> study, they can install Zope on their own.
>>
>>That's pure FUD. Zope takes <depending on your internet connection> time
>>to install, about 5 minutes to compile and 30 seconds to startup. And
>>using it is not necessarily harder then learning CGI. I'm only mentioning
>>it because "Zope is hard" seems to be a commonly uttered phrase, one that
>>I believed until I actually had to get a Zope server up and running.
>>
>>Hooking up Zope to Apache (say, with reverse proxying) is very easy, though
>>it does use some Apache configuration knowledge. This can all be learned
>>in a day, and anyone seriously using Apache needs to know about this
>>stuff.
>			.
>			.
>			.
>There are a lot of people who are frivolously using Apache.
>
>That is, I run into a lot of people who aren't equipped to
>install Apache on their own, let alone Zope.  They can *buy*
>Apache, though (more precisely, its service equivalent), and
>they want to know how far that'll take them, without any more
>"programming" or system administration on their part.  What
>use could someone uncomfortable with Zope installation pos-
>sibly have with Python as a programming language?  Well, it's
>a surprise to me, too, but often the answer is, "quite a lot."
>
>Moshe, you're right in all you write here.  I've seen quite a
>few folks, though, for whom it's all irrelevant.  They imagine
>themselves to have better things to do with a day than learn
>reverse proxying (for example).
>-- 
>
>Cameron Laird <claird at NeoSoft.com>
>Business:  http://www.Phaseit.net
>Personal:  http://starbase.neosoft.com/~claird/home.html



From apardon at trout.vub.ac.be  Fri Mar 23 10:07:41 2001
From: apardon at trout.vub.ac.be (apardon at trout.vub.ac.be)
Date: 23 Mar 2001 15:07:41 GMT
Subject: How to use map
Message-ID: <slrn9bmq1g.pav.apardon@trout.vub.ac.be>

If f is a function and k, l and m are lists of the same length
then the following two are IMO more or less equivallent. (*)

  for i in range(len(m)):
    m[i] = f( k[i] , l[i] )

and

  m = map(f,k,l)


I was wondering should k, l and m now be lists of lists
of the same length and all sublists having the same
length too, if it would be possible to eliminate the
use of any loop and get the same result as

  for i in range(len(m)):
    for j in range(len(m[0])):
      m[i][j] = f( k[i][j] , l[i][j] )


(*) I know they are not entirely equivallent but for my
purpose they are equivallent enough

-- 
Antoon Pardon


From phqktrcmuhul at kbxlexeniiq.com  Tue Mar 27 04:27:28 2001
From: phqktrcmuhul at kbxlexeniiq.com (phqktrcmuhul at kbxlexeniiq.com)
Date: Tue, 27 Mar 2001 09:27:28 GMT
Subject: BUILD YOUR OWN CABLE BOX DESCRAMBLER ((LEGAL))
Message-ID: <42Zv6.34027$kF1.142972@brie.direct.ca>

sssleafgbunur
 
L.EGAL C.A.B.L.E TV DE-S.C.R.A.M.B.L.E.R

Want to watch Sporting Events?--Movies?--Pay-Per-View??

*This is the Famous R-O Shack TV D.e.s.c.r.a.m.b.l.e.r 
You can assemble it from R,a,d,i,o S,h,a,c,k parts for about 
$12 to $15.

We Send You: 
1, E-Z To follow Assembly Instructions. 
2. E-Z To read Original Drawings. 
3. Total Parts List.

**** PLUS SOMETHING NEW YOU MUST HAVE! ****
Something you can't do without.

THE UP-TO-DATE 6 PAGE REPORT: 
USING A D-E-S-C-R-A-M-B-L-E-R LEGALLY

Warning: You should not build a T.V D.e.s.c.r.a.m.b.l.e.r
without reading this report first.

You get the complete 6 page report "Using a D-e-s-c-r-a-m-b-
l-e-r LEGALLY",  
the instruction plans and theeasy to follow diagram all for just--
$10.00 

Fill out form below and send it, along
with your $10.00 payment to:

C-a-b-l-e-F-R-E-E-TV
1342 East Vine Str #218
Kissimmee Fl 34744
 
(Cash, Check or Money Order.) 

(Florida residents include 7% Florida State Sales Tax) 

(All orders outside the U.S.A. add $5.00)


PRINT YOUR:

NAME___________________________________________
___________

ADDRESS_______________________________________
____________

CITY/STATE/ZIP__________________________________
__________

E-MAIl 
ADDRESS_______________________________________
_____ 

We are NOT ASSOCIATED in any way with R-A-D-I-O S-H-A-
C-K. 
Neither the design nor instructions were developed 
by, are sold by, or are endorsed by R-a-d-i-o S-h-a-c-k. 
Parts for this fine-tuning device are available 
at many electronics stores (including R-a-d-i-o S-h-a-c-k) 
This is not a R-a-d-i-o S-h-a-c-k product.
  
 
>>  ``P`o`st`ed` `usi`n`g` `a `tri`al` `c`opy` of `S`m`a`rt `Pos`t `Pro
>>  D`o`w`n`l`oad` a copy at` `http`:`/`/`g`o.to`/`d`ef`i`s`of
 
 
 
 
 
 
 
 
 
ufqlavifax
 
 
 
rjvavigmgwuv


From timr at probo.com  Sat Mar 17 23:54:28 2001
From: timr at probo.com (Tim Roberts)
Date: Sat, 17 Mar 2001 20:54:28 -0800
Subject: How do I get Win32 to locate DLLs elsewhere (not really a Python thing)?
References: <mailman.984506597.12469.python-list@python.org> <3AB1F697.485A776D@yahoo.fr>
Message-ID: <jof8bt8potvlvunbqfsircn0re6l8sra3o@4ax.com>

Francois Forest <francois_forest at yahoo.fr> wrote:

>I got the same kind of problem on WinNT and Tcl80.dll,
>I did the same tinkering you did and saw no improvement either.
>Finally, I cleaned up the registry of all useless path (a poor install
>program had set some C:\RECYCLER\123456xxx\... in many paths).

If you're running Windows 2000, this might not be entirely the
application's problem.  If you MOVE an application that is known to the
Windows installer, it will search certain parts of the registry and adjust
occurrances of the old directory name to the new directory name.  Usually,
this is a good thing, but if you happen to drag an application to the
recycle bin intending to get rid of it, you can end up with bogus paths
like this.
--
- Tim Roberts, timr at probo.com
  Providenza & Boekelheide, Inc.


From bobhicks at adelphia.net  Sun Mar 11 18:45:50 2001
From: bobhicks at adelphia.net (Robert L Hicks)
Date: Sun, 11 Mar 2001 23:45:50 GMT
Subject: Python In A Nutshell
References: <nG5q6.54910$__6.11376601@typhoon.southeast.rr.com> <Ufcq6.140786$47.2078615@news.bc.tac.net> <3AABB165.959ABD5@sympatico.ca>
Message-ID: <B6D1784E.216D%bobhicks@adelphia.net>

I would use a chapter for each widget set maybe (Pmw, wxPython, PyKDE/Qt,
Py/GTK etc.). I would stick primarily with Tkinter though and give lots of
good examples. I don't have a nutshell book so I am not familiar with the
type of content.

From: "Colin J. Williams" <cjw at sympatico.ca>
Organization: Sympatico
Newsgroups: comp.lang.python
Date: Sun, 11 Mar 2001 17:28:04 GMT
Subject: Re: Python In A Nutshell


Laura, 

Two suggestions: 
Produce the book with a binding so that it will lie flat on a desk when open
Document the classes in wxPython
Colin W. 

LLewin at oreilly.com wrote:
Hi, 
Yep--Python Nutshell is "forthcoming"...we're just in the planning stages at
this point--very early on in the process.  Any suggestions are welcome!
Laura 
LLewin at oreilly.com 

> Hey Alex,  congrats on landing "Python In A Nutshell" book deal. One
> problem.  I went down to the book store and it's not there. What gives?
> <wink> 
> 
> Don 
> 
> 

==================================
Posted via http://nodevice.com
Linux Programmer's Site





From aleaxit at yahoo.com  Fri Mar  9 08:59:37 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 9 Mar 2001 14:59:37 +0100
Subject: Remote script execution with Python ...
References: <7Ybp6.7729$r4.7336@news.indigo.ie> <ulmqil8qu.fsf@roundpoint.com> <Lpwp6.8059$r4.7712@news.indigo.ie>
Message-ID: <98ank601u1b@news2.newsguy.com>

"Robert Gahan" <gahanr at gofree.indigo.ie> wrote in message
news:Lpwp6.8059$r4.7712 at news.indigo.ie...
> Thanks Ben.
>
> os.popen2 is not available to windows programmers but

Works fine for me (NT4, Python 2.0 standard build).


Alex





From grey at despair.rpglink.com  Sat Mar  3 04:23:35 2001
From: grey at despair.rpglink.com (Steve Lamb)
Date: Sat, 03 Mar 2001 09:23:35 -0000
Subject: Can I capture console output from an os.system() call?
References: <97p7nt$a2s@news.or.intel.com>
Message-ID: <slrn9a1e0m.p7f.grey@teleute.dmiyu.org>

On Fri, 2 Mar 2001 14:46:05 -0800, Brian Geddes <brian.j.geddes at intel.com>
wrote:
>When I call a foreign program using an os.system("program.exe") call, is
>possible to get at the console output of "program.exe" from my python
>script?

    I believe you're looking for os.popen()

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


From tjg at exceptionalminds.com  Mon Mar 19 19:50:01 2001
From: tjg at exceptionalminds.com (Timothy Grant)
Date: Mon, 19 Mar 2001 16:50:01 -0800
Subject: HELP - beginner to programing/python
In-Reply-To: <Vpxt6.416511$Pm2.6478704@news20.bellglobal.com>; from iansworld@thevortex.com on Tue, Mar 20, 2001 at 12:23:17AM +0000
References: <Vpxt6.416511$Pm2.6478704@news20.bellglobal.com>
Message-ID: <20010319165001.L3109@trufflehunter.avalongroup.net>

On Tue, Mar 20, 2001 at 12:23:17AM +0000, EEN wrote:
> Hi there. I am a newbie and am hearing that python is the way to go to start
> off. I am downloading it as i type this, but need to know how to use it :)
> of course :) Is there any site that has a good tutorial or beginners guides,
> or do you have to buy books? If so, waht do you recommend?
> 
> Thanks a lot for the help, i really appreciate any pointers!

the Python Tutorial is one of the best available you may find
it at http://http:www.python.org and follow the links under 
documentation.

There are also several good books. The best of which for a
newbie would probably be O'Reilly's Learning Python book.

--
Stand Fast,
	tjg.

Timothy Grant                         tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.         <><       (503) 246-3630
>>>>>>>>>>>>>Linux, because rebooting is *NOT* normal<<<<<<<<<
>>>>This machine was last rebooted:  62 days  5:02 hours ago<<



From pae at dim.com  Wed Mar 28 01:49:08 2001
From: pae at dim.com (Phil Earnhardt)
Date: Tue, 27 Mar 2001 23:49:08 -0700
Subject: Collection interfaces
References: <3a90fa97.0@news1.mweb.co.za> <MPG.15085517debf4b649896ca@news.earthlink.net> <zb%n6.31033$o%6.586026@news1.rdc1.bc.home.com> <o1jv6.9635$P4.789083@newsread1.prod.itd.earthlink.net> <99l6s8$gpu$1@nyheter.chalmers.se> <m38zlq4w34.fsf@chaos.resnet.gatech.edu>
Message-ID: <lv13ct8qbni7f3h7q9cujj6d3iequ8to2s@4ax.com>

On 27 Mar 2001 18:42:49 -0500, Lex Spoon <lex at cc.gatech.edu> wrote:

>matso at dtek.chalmers.se (Mats Olsson) writes:
>
>>     Yea, use a Java applet instead. Guaranted not to fiddle with your
>> DLL's :-)
>
>Oh great, so we get to fiddle with JAR's instead of DLL's!  At least
>JAR is easier to pronounce.  :)

Can you explain your comment, Lex? How is downloading and running
applets from a jarfile have anything to do with fiddling with DLLs?

--phil



From Eugene.Leitl at lrz.uni-muenchen.de  Sat Mar 10 11:17:52 2001
From: Eugene.Leitl at lrz.uni-muenchen.de (Eugene.Leitl at lrz.uni-muenchen.de)
Date: Sat, 10 Mar 2001 17:17:52 +0100
Subject: architectural issues cot'd; which http server to embed?
References: <mailman.984150374.638.python-list@python.org> <5Hrq6.12053$v5.51041@newsfeeds.bigpond.com>
Message-ID: <3AAA53B0.530C4627@lrz.uni-muenchen.de>

Brett Haydon wrote:
> 
> Not quite sure what you've got happening there..
> 
> but look at www.modpython.org for an apache/python solution.

I've thought about that, but it looks like a total overkill,
considered that a-5-page-of-Python-http-server would suffice.

Apache is big, impossible (for me) to understand to modify,
has tons of options, and puts many files in different places.



From pearu at cens.ioc.ee  Wed Mar  7 04:26:06 2001
From: pearu at cens.ioc.ee (Pearu Peterson)
Date: Wed, 7 Mar 2001 11:26:06 +0200 (EET)
Subject: [Distutils] Using boost with distutils (fwd)
Message-ID: <Pine.LNX.4.21.0103071125120.7222-100000@kev.ioc.ee>

Hi!

I am trying to use distutils with boost (www.boost.org) on MD 7.0 Linux
and with gcc 2.95.2.

When compiling/linking C++ codes, distutils uses gcc compiler. But then
importing shared module fails (it cannot find `cerr').
When I manually used g++ for compiling/linking, import succeeded.

Now I use the following code fragment
++++++++++++++++++++++++++++++++++++++++++++++
from distutils import sysconfig
save_init_posix = sysconfig._init_posix
def my_init_posix():
    print 'my_init_posix: changing gcc to g++'
    save_init_posix()
    g = sysconfig._config_vars
    g['CC'] = 'g++'
    g['LDSHARED'] = 'g++ -shared'
sysconfig._init_posix = my_init_posix
+++++++++++++++++++++++++++++++++++++++++++++
in my setup.py file.

I am wondering if there is a "nicer" way to change compiler and
linker program in setup.py?

If a package contains both C and C++ codes, then the above will not
work. Any ideas how to build such mixed extension modules?

I'll appreciate an example setup.py that uses boost.

Pearu


_______________________________________________
Distutils-SIG maillist  -  Distutils-SIG at python.org
http://mail.python.org/mailman/listinfo/distutils-sig




From aleaxit at yahoo.com  Fri Mar  9 08:43:19 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 9 Mar 2001 14:43:19 +0100
Subject: gah! I hate the new string syntax
References: <mailman.983615170.23313.python-list@python.org> <tN5o6.2746$eH3.14913@NewsReader> <Koco6.7518$1D5.244602@e420r-atl1.usenetserver.com> <mailman.983997252.5628.python-list@python.org> <meAp6.29961$ML1.1514835@e420r-atl2.usenetserver.com> <mailman.984014353.5058.python-list@python.org> <BcCp6.37817$1D5.1582482@e420r-atl1.usenetserver.com>
Message-ID: <98amlj01sp6@news2.newsguy.com>

"Steve Holden" <sholden at holdenweb.com> wrote in message
news:BcCp6.37817$1D5.1582482 at e420r-atl1.usenetserver.com...
    [snip]
> > | > What about one who is both English and Scots, yet never set foot on
> > | > that island ;-)?
    [snip]
> Either you were born in England, or you were born in Scotland, or you are
> neither English nor a Scot. Part-either-or-both, maybe, but not fully

Wrong!  You're talking about people whose seafaring skills and
traditions are (deservedly) legendary.  It's perfectly acceptable
for either an English or a Scot to be born aboard a ship...!!!


Alex





From whisper at oz.net  Mon Mar 26 14:22:32 2001
From: whisper at oz.net (Dave LeBlanc)
Date: 26 Mar 2001 19:22:32 GMT
Subject: RFC: PEP243: Module Repository Upload Mechanism
References: <mailman.985475916.3683.python-list@python.org> <99n57o$dkq@gap.cco.caltech.edu>
Message-ID: <99o4to$e8n$0@216.39.170.247>

After reading the PEP, that was my thought too: there should be a
platform-independent setting for os.

Also, How would one distinguish between (for example), NT Workstation
vs NT Server (ditto for Windows 2000) - there are a few differences
between them that can control whether or not an app will run on one
but not the other I believe. How about the patch level of the OS -
Currently i'm running Windows NT Worstation 4.0sp6 for example. Patch
level does make a difference (sp4 screwed up tcp/ip - only Microsoft
<g>).

How about a generic "Linux" for those cases where the app is set up to
build/run on _any_ linux distro (if that's possible)? Ditto for
Windows meaning any windows... Could this be "Windows-any..." ? What
about a case where the app runs on some few of these os's, but not
all? (i.e "linux-only" for example (bad programmer, BAD!)).

One solution might be to have a webpage form where one could click on
the os's etc. that apply and also has an upload button... Not being
familiar with distutils, this might be redundant, but maybe this form
could encourge people to put some descriptive text to go along with
their submission.

A minor niggle: any reason it's ix86 and not just x86? Is someone
besides intel and clones going to build x86 architecture? Come to
think of it, x86 isn't just Intel anymore what with all the clone
makers like AMD, National, Transmeta...

Sorry, another minor niggle: "swalowpost.cgi" is cute, but not nearly
as meaningful (or frankly, as credible) as something like
"submitmodule.cgi". Some management types just have no sense of humor
;-)

Um... btw, it's "swallow", not swalow :-)

Regards,

Dave LeBlanc

P.S Would or could your server be smart enough to do
tar.gz->tgz->zip->bzip repackaging for different users?
While not really necessary, it might tend to increase comfort level
for some ("where the @#$%@ do I find bzip for windows?" someone might
say...). (I do know where to find bzip for windows :-).)

On Mon, 26 Mar 2001 02:15:37 -0800, Nathaniel Gray
<n8gray at caltech.edu.is.my.e-mail.address> wrote:

><posted & mailed>
>
>Sean Reifschneider wrote:
>>     The following are valid os names:
>> 
>>         aix beos debian dos freebsd hpux mac macos mandrake netbsd
>>         openbsd qnx redhat solaris suse windows yellowdog
>> 
>> [snip]
>>
>>     Version is the official version string specified by the vendor for
>>     the particular release.  For example, "2000" and "nt" (Windows),
>>     "9.04" (HP-UX), "7.0" (RedHat, Mandrake).
>> 
>>     The following are valid architectures:
>> 
>>         alpha hppa ix86 powerpc sparc ultrasparc
>
>Can any or all of these be set to None, or some equivalent?  Python code is 
>often cross-platform, after all.
>
>-n8
>
>-- 
>_.~'^`~._.~'^`~._.~'^`~._.~'^`~._.~'^`~._
>             Nathaniel Gray
>   California Institute of Technology
>     Computation and Neural Systems
>     n8gray <at> caltech <dot> edu
>_.~'^`~._.~'^`~._.~'^`~._.~'^`~._.~'^`~._
>



From sholden at holdenweb.com  Tue Mar 13 19:18:53 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Tue, 13 Mar 2001 19:18:53 -0500
Subject: This math scares me
References: <3aad4062.1290021@News.CIS.DFN.DE> <20010312170356.B3513@better.net> <mailman.984480856.18097.python-list@python.org> <984486395.407379@newsmaster-04.atnet.at> <slrn9asf1d.ja3.qrczak@qrnik.zagroda> <984518199.127140@newsmaster-04.atnet.at> <jwbnews-09D75F.13313313032001@corp.supernews.com> <yTwr6.1235$Tg.167429@ruti.visi.com>
Message-ID: <vRyr6.68185$1D5.3491592@e420r-atl1.usenetserver.com>

"Grant Edwards" <grante at visi.com> wrote in message
news:yTwr6.1235$Tg.167429 at ruti.visi.com...
> In article <jwbnews-09D75F.13313313032001 at corp.supernews.com>, John W.
Baxter wrote:
> >In article <984518199.127140 at newsmaster-04.atnet.at>,
> > "Werner Schiendl" <ws-news at gmx.at> wrote:
> >
> >> True, but even 0.1 and 0.2 need not necessarily be exact values.
> >
> >Of the single decimal digit numbers, 20% are exact (x.0 and
> >x.5, for not huge x).  For two decimal places, it's down to 4%
> >(x.00, x.25, x.50, x.75) and it spirals quickly downward from
> >there.
> >
> >Bill Gates got this right in the early days, delivering two
> >Basic interpreters: binary floating point for speed, and
> >decimal floating point for money and such.  (I never looked at
> >how well his interpreters did...I just used them.)
>
> I don't know if you can credit Gates with doing anything other
> than following the pack.  Back when I first started doing
> software (CP/M 1.7), all language packages for "personal"
> computers came with both BCD and binary FP math libraries so
> you could do exact monetary calculations if you wanted.
>
> Of course we'd never imagined we'd ever see hardware FP and BCD
> libraries were only marginally slower than binary ones. Now
> that FP is usually in hardware, BCD FP (or fixed point) will be
> pretty slow by comparison.
>
For monetary calculations none of this really matters. As long as you round
to 2 decimal places after each calculation the rounding error never gets bif
enough to bother about. And, of course, BCD is pretty fast on machines (such
as mainframes) which have hardware support for it built in. Maybe we need a
library we can use for BCD when running Linux under MVS.

or-is-that-VMS-ly y'rs  - steve





From oliphant at ee.byu.edu  Sun Mar 18 23:56:13 2001
From: oliphant at ee.byu.edu (Travis Oliphant)
Date: Sun, 18 Mar 2001 21:56:13 -0700
Subject: newbie needs help installing NumPy, reference material.
In-Reply-To: <xeAq6.3544$R_6.353426@newsread2.prod.itd.earthlink.net>
References: <xeAq6.3544$R_6.353426@newsread2.prod.itd.earthlink.net>
Message-ID: <Pine.HPX.4.33.0103182149420.10898-100000@sesame.ee.byu.edu>

> Hey everyone,

Hi Noah,

>
> Just trying to get numeric python installed on a win2000 machine with python
> 2.0, Idle, etc installed (but without a c compiler). I've downloaded
> http://download.sourceforge.net/numpy/NumPy1711_20.zip, but am unclear about
> whether it's the right version (there's an 18.0 version) and how to install
> it.  The documentation at www.python.org was a little outdated.
>

If I were just getting started as you are, I would probably download the
Python 2.b1 binary and the new Numeric 18.0.0-Python-2.1.zip that Paul
just made from the Numeric CVS tree.  It has support for rich comparisons
and inplace operators which will be very useful for you.

Get the Numeric zip at sourceforge.

Good luck,

Travis Oliphant




From jwbaxter at olympus.net  Sat Mar 31 22:54:38 2001
From: jwbaxter at olympus.net (John W. Baxter)
Date: Sat, 31 Mar 2001 19:54:38 -0800
Subject: PythonLabs Team Moves to ActiveState
References: <mailman.986086501.5576.python-list@python.org> <v4wx6.121194$Ch.16134973@typhoon.we.rr.com> <tcd4hd17gg2vbb@corp.supernews.com>
Message-ID: <310320011954383675%jwbaxter@olympus.net>

In article <tcd4hd17gg2vbb at corp.supernews.com>, Scottie <me at nospam.net>
wrote:

> I agree completely.
> Imagine having to indent Perl!
> the horror, the horror....

First, one must reach agreement on which of several
OnlyRightWayToIndent is to be enforced.  That should save us at least
until next March 31.

  --John


From stephen_purcell at yahoo.com  Fri Mar 30 03:03:19 2001
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Fri, 30 Mar 2001 10:03:19 +0200
Subject: Python GUI with Swing
In-Reply-To: <001e01c0b8ea$16a14660$340210ac@winson>; from winson@mdsserv.mds.com.tw on Fri, Mar 30, 2001 at 03:21:45PM +0800
References: <001e01c0b812$43f00760$340210ac@winson> <mailman.985889320.22860.python-list@python.org> <3AC3BE89.3F95C8DC@ou.edu> <001e01c0b8ea$16a14660$340210ac@winson>
Message-ID: <20010330100319.C5773@freedom.puma-ag.com>

??? wrote:
> [snip] Swing maybe slow but my customer can accept.
> If I use Swing thru Jython, it will be slower then directly using Java to
> develop the system. So I need a solution to use Swing in Python, that's all.


So you want to use C Python because it's faster than Jython, but you want
the Swing GUI? I think you're making things difficult for yourself.

If Swing is fast enough for your customers, no Java code could be too
slow for them.  :-)

Therefore, why not write the application in Jython, and re-write
speed-critical parts in Java? For most parts of your application, Jython won't
be any slower than Java.

(Personally, I would just write everything in Jython and add 100MHz to the
minimum recommended system requirements to make up for the difference between
Jython and straight Java.)

-Steve

-- 
Steve Purcell, Pythangelist
Get testing at http://pyunit.sourceforge.net/
Any opinions expressed herein are my own and not necessarily those of Yahoo



From fredrik at pythonware.com  Fri Mar 16 11:35:05 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 16 Mar 2001 16:35:05 GMT
Subject: PyImport_ImportModule
References: <98tcks$7bl$1@allthetime.news.cais.net>
Message-ID: <Zgrs6.170$sk3.49958@newsb.telia.net>

Charlie Barrows wrote:
> I'm just starting to tinker with embedding Python in a C app and I'm trying
> to get a handle on where modules end up when they're imported.
>
> If I run this code:
>
> PyObject *module = PyImport_ImportModule("TestScript");
>
> ... which namespace is TestScript imported into?

it's own namespace, of course.

ImportModule is the same thing as "import module", not
"from module import *".

Cheers /F




From tim.one at home.com  Sat Mar 31 18:08:40 2001
From: tim.one at home.com (Tim Peters)
Date: Sat, 31 Mar 2001 18:08:40 -0500
Subject: Know of Substantial Apps Written in Python?
In-Reply-To: <E14jSWh-0003L4-00@darjeeling>
Message-ID: <LNBBLJKPBEHFEDALKOLCKEIHJJAA.tim.one@home.com>

[Moshe Zadka]
> ...
> Python would never fit on a floppy.

Sure it can!  Python + the entire base std library (top level of Lib/) still
fits on a floppy with ease, if you use any common compression scheme.  Just
for fun, I just built a zip file with that stuff for the Windows 2.1b2
distribution, and it came in at just over a tad above 1Mb (so about 70% of
one floppy).  What that leaves out from the *std* Windows distribution is,
primarily, 7Mb (unpacked) of HTML docs, Tcl/Tk, the C header files and link
libraries, and the Tools/ directory (incl. IDLE).

the-batteries-are-bulkier-than-the-toy-ly y'rs  - tim




From norm at cybersurf.net  Sun Mar  4 23:18:29 2001
From: norm at cybersurf.net (Norm Lyons)
Date: Sun, 4 Mar 2001 21:18:29 -0700
Subject: Newbie question about nested lists and constructor functions...
References: <W1Do6.4683$ke1.121750@jekyl.ab.tac.net> <mailman.983764626.21888.python-list@python.org>
Message-ID: <csEo6.4686$ke1.122169@jekyl.ab.tac.net>

Sorry about that Chris.

I am trying to create instantiated objects with the following properties:

list_var = [(key, value), left, right]

where the "key, value" pair are a tuple, and the variables "left" and
"right" are empty lists [ ] which will be manipulated to hold values later
in the program after they have already been initialized to empty lists.  The
purpose of this object is to hold the necessary data for nodes in a binary
search tree.  I know how to implement this type of data structure in C++ but
I seem to be getting hung up on the syntax differences in the Python class
and member variable declarations.

An example of one of my attempts would be:

class Node:
     def __init__(self, key = ''):
          left = []
          right = []
          data = (key, 'value')
          newnode = [data, left, right]
     def info(self):
          return self.data

The "info" function is only intended to return the values of the variables
"key and value".  The other values are used as pointers to child nodes in
the tree.  As I mentioned earlier, I have been through the online tutorial a
few times and have not found any information that clears this issue up for
me.  Obviously, the code above is incomplete as it should also contain the
appropriate "if" and "elif" statements to properly branch according to the
information passed to the constructor function.  Unfortunately, I do not
know what I am doing wrong.  Any suggestions?

Norm.

P.S. Sorry about the contact information on the last message.  Someone was
fooling around with my newsgroup account settings and I didn't notice it
until the last message was already posted.




Chris Gonnerman wrote in message ...
>Could you please post example(s) so we know exactly what is giving you
>trouble?  I at least don't quite know what you are asking.
>
>The constructor function (Python calls it "__init__") is created thus:
>
>class Spam:
>    def __init__(self):
>        ...
>
>and of course __init__ might have additional arguments.  If you are
creating
>a nested list with known initializers, you might say:
>
>myvar = [ [ "list", 2, 3, 4 ], [ 'A', 'B', 'C' ] ]
>
>so if you want to create a nested list in an __init__ function:
>
>class Spam:
>    def __init__(self):
>        self.eggbasket = [ [ "list", 2, 3, 4 ], [ 'A', 'B', 'C' ] ]
>
>to assign a nested list to a class attribute.  If you are passing in
>information
>which you wish to assign that way, you would embed it in the given syntax.
>For
>instance:
>
>class Spam:
>    def __init__(self, eggs):
>        self.eggbasket = [ eggs, [ "spam", "spam", "spam" ] ]
>
>my_spam = Spam([ "egg", "egg", "egg"])
>
>which would nest the given list of eggs within the eggbasket member of
>my_spam.
>You might alternately want to make a copy of the initializer instead of
>storing
>a reference:
>
>class Spam:
>    def __init__(self, eggs):
>        self.eggbasket = [ eggs[:], [ "spam", "spam", "spam" ] ]
>
>[:] (slice of all) is the Pythonic "shallow" list copier.
>
>Of course this is all potentially nonsense as, again, I don't exactly know
>what
>you are wanting to do.
>
>----- Original Message -----
>From: "Bartok" <JollyRoger at hotmail.com>
>Subject: Newbie question about nested lists and constructor functions...
>> I need some help with nested lists when using class constructors to
>> initialize their contents.  I found the tutorial to be of not much help
in
>> this area and have tried everything that I can think of.
>>
>> I understand the process as I am an experienced C++ programmer but I seem
>to
>> be having some difficulty using the correct syntax.  Can anyone please
>help?
>
>
>




From thomas at xs4all.net  Mon Mar 12 09:07:46 2001
From: thomas at xs4all.net (Thomas Wouters)
Date: Mon, 12 Mar 2001 15:07:46 +0100
Subject: finding and listing the differences between two files
In-Reply-To: <qJsq6.209137$v.20440644@typhoon.nyroc.rr.com>; from cjc26@nospam.cornell.edu on Sat, Mar 10, 2001 at 04:35:34PM +0000
References: <mailman.983966710.3462.python-list@python.org> <qJsq6.209137$v.20440644@typhoon.nyroc.rr.com>
Message-ID: <20010312150746.R404@xs4all.nl>

On Sat, Mar 10, 2001 at 04:35:34PM +0000, Cliff Crawford wrote:
> * Chris Richard Adams <chrisa at ASPATECH.COM.BR> menulis:
> | 
> | Anyone know the equivalent in Python or am I better off sticking to
> | shell.

> 2.1 has a module called "difflib" which you can use.  Also, a diff
> program written in Python comes with the Python source
> (Tools/scripts/ndiff.py).

Note that ndiff is 'not	quite normal diff' (hence the 'n'.) It is quite
superior if you want 'readable' diffs, and if you want to spot subtle
changes, though.

Note that ndiff is 'not quite normal diff' (hence the 'n',) It is quite
superior if you want 'readable' diffs, and if you want to spot sublte
changes, though. 

(And here's the ndiff of the above two almost identical pieces ;)

- Note that ndiff is 'not quite normal diff' (hence the 'n'.) It is quite
?                                                          ^              
+ Note that ndiff is 'not quite normal diff' (hence the 'n',) It is quite
?                                                          ^              
- superior if you want 'readable' diffs, and if you want to spot subtle
?                                                                    -
+ superior if you want 'readable' diffs, and if you want to spot sublte
?                                                                   +
- changes, though.
+ changes, though. 
?                 +

-- 
Thomas Wouters <thomas at xs4all.net>

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



From rob at ZOOstation.cc  Tue Mar 13 03:12:37 2001
From: rob at ZOOstation.cc (Rob Brown-Bayliss)
Date: 13 Mar 2001 21:12:37 +1300
Subject: Einstein's Riddle
In-Reply-To: <oNPq6.396644$ge4.138479566@news2.rdc2.tx.home.com>
Message-ID: <200103130812.f2D8CbF03138@ZOOstation.cc>

> The question is: "Who owns the fish?"
> 
> Hints:

First: As a child I was neither oasweed nor a brit, a german, dane or
norwegian.   I had as pets a cat, a dog, a budgie (from which I never
bit the head off of), some gold fish and at last three ocassions
multiple tanks of tropical fish.

Second:  The sweed keeps dogs, but are they his?

Third: Any one who smokes is a fool, even if they keep fish or belive no
one else can solve trivial problems

Fourth:  When is his more important "I cant do this!" question going to
be answered (the one uniting realativity and gravity with the McDonalads
money making theory?)

Fifth:  I drink beer, but don't own a horse.

Sixth: All my neighbours drink water.

Taking all that in to account - Why would a known smart fella ask the
world (how else did he know that most of you could not answer it?) an
obviously unimportant question? - unless his wife had an unexplainable
reciept for fish food...





From gherman at darwin.in-berlin.de  Tue Mar 20 07:15:55 2001
From: gherman at darwin.in-berlin.de (Dinu Gherman)
Date: Tue, 20 Mar 2001 13:15:55 +0100
Subject: How to find out about free space on a given drive?
References: <3AB5C19E.FA40DDA9@darwin.in-berlin.de> <mailman.984991890.14766.python-list@python.org> <CDE3F34783E0C6CB.0DCEA469EF336A5B.984705F2492598CF@lp.airnews.net>
Message-ID: <3AB749FB.3C4B7345@darwin.in-berlin.de>

Cameron Laird wrote:
> 
> When it matters enough to me, maybe I'll write up (partial)
> os.statvfs() implementations for Win*, or even MacOS, perhaps
> starting from <URL: http://mini.net/cgi-bin/wikit/526.html >
> (see "Disk free capacity") and submit the sources for inclu-
> sion in the distribution library.  It surprises me a teensy
> bit when I read, "I don't think you'll find a cross-platform
> solution", without the "yet" which belongs at the end.

I kicked-off a little debate on this y'day on python-dev, 
where, in the end, it didn't seem so impossible anymore:

  http://mail.python.org/pipermail/python-dev/2001-March/013706.html

I also found a good reference for Mac folks that might be 
helpful in implementing such a function in the os module: 

  http://developer.apple.com/techpubs/mac/Files/Files-96.html

Marc-Andre posted a sample Python implementation with some
white spots to fill in that can also be found in the thread
on python-dev.

Cameron or Jack (if he is listening): I'd be really happpy
to see that before Python 3000... but can't offer much more
then a few pints on the next conference. ;-)

Best regards,

Dinu

-- 
Dinu C. Gherman
ReportLab Consultant - http://www.reportlab.com
................................................................
"The only possible values [for quality] are 'excellent' and 'in-
sanely excellent', depending on whether lives are at stake or 
not. Otherwise you don't enjoy your work, you don't work well, 
and the project goes down the drain." 
                    (Kent Beck, "Extreme Programming Explained")


From larskl at klassekampen.no  Tue Mar 27 08:21:00 2001
From: larskl at klassekampen.no (Lars Klęboe)
Date: Tue, 27 Mar 2001 15:21:00 +0200
Subject: Changing every other instance of <B> in a file
Message-ID: <8e41ctkk4l38jmj34070j15iotgv4is2cg@4ax.com>

I have quite a few (~70000) quark tagged files I'd like to HTMLize, and
as my tool of choice has been python for every other script, I thought
I'd stick to what I know. Or thought I knew.

The original file.txt (quark tagged) :

	Blablabla <B> talktalk <B> blabla blabla balbalblabla

The resulting file.html (html)

	Blablabla <B> talktalk </B> blabla blabla balbalblabla

As you can tell, every other instance of <B> is to be changed into </B>.

How on earth can I do this as easy as I did in sed ? I'm stuck
mindwise, locked on a useless solution I'm too embarrased
to show you ;)

All help is appreciated !

-- 
Lars Kl?boe



From shaleh at valinux.com  Fri Mar  2 13:36:45 2001
From: shaleh at valinux.com (Sean 'Shaleh' Perry)
Date: Fri, 02 Mar 2001 10:36:45 -0800 (PST)
Subject: gah! I hate the new string syntax
In-Reply-To: <LNBBLJKPBEHFEDALKOLCIEHFJDAA.tim.one@home.com>
Message-ID: <XFMail.20010302103645.shaleh@valinux.com>

On 02-Mar-2001 Tim Peters wrote:
> [Sean 'Shaleh' Perry]
>> good ole:
>>
>> from string import join
>>
>> l = join(';', ['a','b','c']) # thanks, this is clear and obvious
>>
>> Please o' gods of python, do not deprecate the clean and obvious
>> syntax that is found in the string module.
> 
> Sean, that claim of "obvious" would have been a lot more convincing if you
> had managed to remember the correct order of string.join's arguments <wink>.
> 

As I said in other mails, forgive me, brain dead from reading WAY too much perl
this week.



From dwndgg at opti.cgi.net  Wed Mar  7 13:46:09 2001
From: dwndgg at opti.cgi.net (Dan)
Date: Wed, 07 Mar 2001 18:46:09 GMT
Subject: Python 2.1b1 dumps core
Message-ID: <3aa6808b.962524627@news.okc1.ok.home.com>

I have a pure Python based web server app that dumps core when run
under Python 2.1b1.  It runs fine under Python 2.0.  I can send a core
dump to someone if that is useful to anyone.  If not then I might be
able to narrow it down to a certain part of the Python code and post a
sample that breaks..

-Dan



From danielk at aracnet.com  Sun Mar 18 17:24:54 2001
From: danielk at aracnet.com (Daniel Klein)
Date: Sun, 18 Mar 2001 14:24:54 -0800
Subject: Who's minister of propaganda this week?
References: <B98041329020E007.7DA84CAD9D0562AB.817BBC572F480869@lp.airnews.net> <98nu1m$12e@dispatch.concentric.net> <660ACCA1E4816F7E.C359CDF4168EF3FB.3B6B05538D619770@lp.airnews.net> <3AAFA13D.76E32C3F@divalsim.it> <98oa4o01i4v@news2.newsguy.com> <mailman.984619039.7856.python-list@python.org> <98q9h60122v@news2.newsguy.com> <slrn9b52dj.8it.qrczak@qrnik.zagroda>
Message-ID: <jqcabtsuvmc3io4pbn3ujeuaibg5b2hsbg@4ax.com>

On 16 Mar 2001 21:46:29 GMT, qrczak at knm.org.pl (Marcin 'Qrczak' Kowalczyk)
wrote:

> In statically typed languages I use most of my errors
>(over 80%) are detected at compile time.
>
>Most of them are type errors (e.g. mismatching the number of
>arguments) or name binding errors (e.g. forgotten module imports or
>using a parameter which is not present in this function after some
>reorganization).

Yes, but  any reasonably tested program will catch these kinds of errors
anyway...or perhaps you don't test things. :-) It's a mistake to rely on static
type checknig as a substitute for unit testing.

Python (and Smalltalk) go one step further by 'hiding' the compilation step
thus making for a shorter development cycle and legendary productivity gains.

Having dealt with both Java and Python, I can confirm that once you are use to
a dynamically typed language, variable declarations become like training
wheels; once you learn how to 'ride the bike' they just get in the way.

Daniel Klein


From chris.gonnerman at usa.net  Fri Mar 23 14:01:32 2001
From: chris.gonnerman at usa.net (Chris Gonnerman)
Date: Fri, 23 Mar 2001 13:01:32 -0600
Subject: newbe   how to read in a string?
References: <3ABB4E70.54885C9F@noos.fr> <mailman.985355670.6600.python-list@python.org> <99g1ac02fug@news2.newsguy.com>
Message-ID: <009701c0b3cb$b4ec6c60$a100000a@local>

----- Original Message ----- 
From: "Alex Martelli" <aleaxit at yahoo.com>
Subject: Re: newbe how to read in a string?


> "Chris Gonnerman" <chris.gonnerman at usa.net> wrote in message
> news:mailman.985355670.6600.python-list at python.org...
>     [snip]
> >     import string
> >     seq = string.split(a, ",")
> >     for i in range(1, len(seq)): # this skips the 0 index,
> >         seq[i] = float(seq)      # which needs to stay a string.
> >
> > In Python 2.0 the code might be shorter, using string methods and list
> > comprehension, but I personally prefer the "old way" myself.
> 
> The 'old way' would be
>     seq = map(float, string.split(a, ','))
> and it still works fine of course, though I'd rather spell it
>     seq = map(float, ','.split(a))

BZZT!  The given string starts with a date which will cause float()
to throw a ValueError.

I might say:

    seq = string.split(a, ",")
    seq[2:] = map(float, seq[2:])

> List comprehensions are great, but when you DO want to get
> map's primary job (applying an already-existing function
> to parameter sequences) map is still best (with list
> comprehension, you can avoid using _lambda_ just in order
> to be able to use map -- I think eschewing lambda makes
> your code clearer, but maybe that's just me).
 
No argument here.  I work in both 1.5.2 and 2.0 and rarely miss
the 2.0 features, and I'm in the camp that finds ','.split(a) 
ugly.  I don't like lambda and avoid it whenever possible.
 





From emile at fenx.com  Thu Mar 22 20:51:11 2001
From: emile at fenx.com (Emile van Sebille)
Date: Thu, 22 Mar 2001 17:51:11 -0800
Subject: Einstein's Riddle
References: <oNPq6.396644$ge4.138479566@news2.rdc2.tx.home.com> <98glrd$1si1a$1@ID-75892.news.dfncis.de> <3AABE712.499CCF4C@home.com> <98gp5b$1u0oi$1@ID-75892.news.dfncis.de> <98i5880pl2@news2.newsguy.com> <98kfrl$pug$1@sshuraaa-i-1.production.compuserve.com> <98kntc0o5l@news2.newsguy.com> <3ABA7DE9.8C7A2AEE@alumni.caltech.edu>
Message-ID: <99eace$to04$1@ID-11957.news.dfncis.de>

I thought the german had the fish that didn't exist?

if-you-eat-it-and-it-never-was-can-you-still-ride-a-bicycle?-ly y'rs


Emile van Sebille
emile at fenx.com

-----


"Michael Klein" <mklein at alumni.caltech.edu> wrote in message
news:3ABA7DE9.8C7A2AEE at alumni.caltech.edu...
> > [ eating emu ]
>
> You guys are talking about eating the poor German's pet!  Have you no
heart.
>




From nas at python.ca  Thu Mar 22 00:54:13 2001
From: nas at python.ca (Neil Schemenauer)
Date: Wed, 21 Mar 2001 21:54:13 -0800
Subject: New PEP: The directive statement
In-Reply-To: <Pine.LNX.4.33.0103212154070.6632-100000@bms>; from bsass@freenet.edmonton.ab.ca on Wed, Mar 21, 2001 at 10:21:50PM -0700
References: <mailman.985085006.24236.python-list@python.org> <slrn9bi7q3.2sa.hzhu@rocket.knowledgetrack.com> <Pine.LNX.4.33.0103212154070.6632-100000@bms>
Message-ID: <20010321215413.B25810@glacier.fnational.com>

On Wed, Mar 21, 2001 at 10:21:50PM -0700, Bruce Sass wrote:
> Why was (something along the lines of)...
...
> use:
>     __future__.nested_scopes
>     # code that relies on nested scopes
> except FutureError:
>     # code that does not use nested scopes
> 
> ...rejected?  [It looks like a natural to me <shrug>...

"from __future__ import x" is a compile time switch.  May I
suggest reading the PEP:

    http://python.sourceforge.net/peps/pep-0236.html

Its not long and reading it will make you better in bed.

  about-ready-to-try-anything-ly y'rs Neil



From aahz at panix.com  Sun Mar 18 13:04:20 2001
From: aahz at panix.com (Aahz Maruch)
Date: 18 Mar 2001 10:04:20 -0800
Subject: What has become of PEP224 ?
References: <3ab4f320@nntp.server.uni-frankfurt.de>
Message-ID: <992tb4$qf5$1@panix2.panix.com>

[cc'd to Barry Warsaw in case he wants to comment]

In article <3ab4f320 at nntp.server.uni-frankfurt.de>,
Michael 'Mickey' Lauer  <mickey at Vanille.de> wrote:
>
>Hi. If I remember correctly PEP224 (the famous "attribute docstrings")
>has only been postponed because Python 2.0 was in feature freeze
>in August 2000. Will it be in 2.1 ? If not, what's the reason ? What
>is needed for it to be included in 2.1 ?

I believe it has been essentially superseded by PEP 232; I thought
function attributes were going to be in 2.1, but I don't see any clear
indication.
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

Three sins: BJ, B&J, B&J


From jmg at ecs.soton.ac.uk  Tue Mar 13 04:29:03 2001
From: jmg at ecs.soton.ac.uk (Jacek Generowicz)
Date: 13 Mar 2001 09:29:03 +0000
Subject: Extending Python: rewriting a single method in C
References: <g0wv9vxms0.fsf@scumbag.ecs.soton.ac.uk> <slrn9apn8f.382.jepler@potty.housenet>
Message-ID: <g0ofv6xbj4.fsf@scumbag.ecs.soton.ac.uk>

jepler at inetnebr.com (Jeff Epler) writes:

> On 12 Mar 2001 11:13:51 +0000, Jacek Generowicz
>  <jmg at ecs.soton.ac.uk> wrote:
> >I have an object-oriented numerical code written in Python. It has
> >come to the stage where the speed of the code is seriously impeding
> >progress. Profiling reveals (surprise, surprise) that most of the time
> >is spent in only 2 methods.
> >
> >Is there any way of re-writing just these methods in C ?
> 
> It's easy to write a function in C.  Other posters have mentioned references 
> to documentation about doing this.

As I mentioned in my replies to Paul and Alex (which I inadvertently
sent by mail rather than through the NG) I know that this can be done,
but I specifically want to rewrite single class/instance methods . . .

> However, you cannot make a builtin function be a method of a class/instance.

Bummer.

> Instead, you could write:
> 
> from cmodule import fast_a, fast_b
> 
> class fastKlass(Klass):
> 	def a(self, int1, int2):
> 		fast_a(self, int1, int2)
> 	def b(self, int1, int2):
> 		fast_b(self, int1, int2)
> In fast_a and fast_b, you can get arguments by using the ParseTuple string
> "(Oii)" and then use PyObject_SetAttrString and PyObject_GetAttrString
> to store or retrieve attributes (including, if necessary, instance
> methods) from the instance by their name.

Perhaps it is appropriate to insert here the exchange between Alex and
myself:

JG> AM> JG> Could something along following lines work ?
JG> AM> JG> 
JG> AM> JG> Write a module called my_extensions containing a function called
JG> AM> JG> foo_bar.
JG> AM> JG> 
JG> AM> JG> import my_extensions
JG> AM> JG> 
JG> AM> JG> class foo:
JG> AM> JG>    bar = my_extensions.foo_bar
JG> AM> 
JG> AM> No, this does not work, because only python-function-objects, NOT
JG> AM> builtin-function-objects, are mutated into unbound methods when
JG> AM> they are attributes of a class-object.
JG> 
JG> Yes, I've noticed this after a bit of experimenting.
JG> 
JG> AM>  Note that the name
JG> AM> 'builtin function object' is a bit misleading, as it actually
JG> AM> describes any function _written in C_ (part of an extension).
JG> AM> 
JG> AM> What CAN work is more explicit delegation:
JG> AM> 
JG> AM> class foo:
JG> AM>     def bar(self, whatever, blahblah):
JG> AM>         return my_extensions.foo_bar(self, whatever, blahblah)
JG> 
JG> Yes, but I was trying to avoid the function call overhead
JG> (particularly as one of the functions I need to rewrite is very short
JG> indeed).

> Note that if there are *many* calls to the method, rather than each
> invocation taking a large amount of time,

Bingo, this is exactly the case for one of the methods.

> then this approach is not going to gain you much, since it actually
> *increases* the function-call overhead.

And here is what Alex had to say about this:

AM> > Yes, but I was trying to avoid the function call overhead
AM> > (particularly as one of the functions I need to rewrite is very short
AM> > indeed).
AM> 
AM> No way to do that -- the function call overhead is there even
AM> if you call directly from Python to the C extension. 

But isn't the function call overhead greater for calling python
functions than it is for calling the C extensions ?

AM> If the function/method is called from a few places, you might
AM> manually inline in there, maybe.

Not sure I understand what you mean exactly. Are you suggesting that I
call an extension function wherever I would previously have called and
instance method ?

How do I get at the members of a class in C if that class was defined
in python ?

JE> In fast_a and fast_b, you can get arguments by using the ParseTuple string
JE> "(Oii)" and then use PyObject_SetAttrString and PyObject_GetAttrString
JE> to store or retrieve attributes (including, if necessary, instance
JE> methods) from the instance by their name.

Is there an organized reference where I can find the specifications of
all these functions ?  All I have come across shows example code, but
no full descriptions of what can/must be done.

Maybe some kind soul could show me what the correct code for a minimal
extension type is ?

For example, I tried to write an extension type equivalent to

class foo:
    def __init__ ( self, a ):
	self.a = a
    def show ( self ):
	print a

My various attempts, almost work, but there is always something not
quite right.

Incidentally, why does the xxobject.c file in the Objects directory of the
Python distribution contain no initxxobject function ?

Jacek


From dsh8290 at rit.edu  Tue Mar 13 11:35:41 2001
From: dsh8290 at rit.edu (D-Man)
Date: Tue, 13 Mar 2001 11:35:41 -0500
Subject: gah! I hate the new string syntax
In-Reply-To: <1eq0wri.16iujikxa9nloN%francois.granger@free.fr>; from francois.granger@free.fr on Fri, Mar 09, 2001 at 09:59:12PM +0000
References: <97opsb$nhu$1@nntp6.u.washington.edu> <mailman.983615170.23313.python-list@python.org> <tN5o6.2746$eH3.14913@NewsReader> <Koco6.7518$1D5.244602@e420r-atl1.usenetserver.com> <mailman.983997252.5628.python-list@python.org> <m2vgpkyfjo.fsf@mycroft.actrix.gen.nz> <mailman.984063199.18779.python-list@python.org> <98an7i01tgp@news2.newsguy.com> <mailman.984154996.23254.python-list@python.org> <1eq0wri.16iujikxa9nloN%francois.granger@free.fr>
Message-ID: <20010313113541.B10298@harmony.cs.rit.edu>

On Fri, Mar 09, 2001 at 09:59:12PM +0000, Fran?ois Granger wrote:
| D-Man <dsh8290 at rit.edu> wrote:
| 
| > So we think, but no one has such records.  Actually, we are all
| > "people" and came from the Garden of Eden
| 
| Sorry, I can't find it on a map ?

You won't be able to find it on a map.  Adam and Eve (and all their
descendants, aka everybody) were banished from the Garden a long time
ago, thus no one has been able to find it since.  

-D


PS.  If you are unfamiliar with the reference, read the beginning of
     the Bible, it explains everything.




From lance.hill at olsinc.net  Thu Mar 22 19:50:55 2001
From: lance.hill at olsinc.net (lance.hill at olsinc.net)
Date: Fri, 23 Mar 2001 00:50:55 GMT
Subject: CGI for generating xml pages
Message-ID: <3aba9bc5.431603914@nnrp.wiznet.ca>

I have a python script which creates xml. If I redirect the output to
a file, when I launch those pages from my browser all is well.
Unfortunately, I can't seem to get it so I can write directly to my
browser from the script. Instead, I get an internal server error.

Can someone kindly give me a clue? Do I need to do some processing
prior to printing to standard output?

Thanks in advance,
L. Hill


From AlfMarion at hotmail.com  Sat Mar 31 15:20:15 2001
From: AlfMarion at hotmail.com (AlfMarion)
Date: Sat, 31 Mar 2001 20:20:15 GMT
Subject: very basic, how do you find square roots?
References: <9a408a$1mpp$1@raewyn.paradise.net.nz>
Message-ID: <3ac63be0.119712073@news.rcn.com>

On Sat, 31 Mar 2001 19:05:43 +1200, "johnston112"
<gordonj at paradise.net.nz> wrote:

>Yes, that is my question :)
>I am trying to make a program that dose the quadratic equation.
>
>johnston112
>
>


>>> 2**.5
1.41421356237


Alf Marion


From mbarnett at ualberta.ca  Sun Mar 11 19:01:28 2001
From: mbarnett at ualberta.ca (Matt)
Date: Sun, 11 Mar 2001 17:01:28 -0700
Subject: Questions about no more nested imports
References: <98gi4l$n9n$0@216.39.170.247>
	 <m3n1as850y.fsf@atrus.jesus.cam.ac.uk> <98h1k4$ai4$0@216.39.170.247>
Message-ID: <mlUq6.3517$Dp6.602755@news-rep.ab.videon.ca>

In article <98h1k4$ai4$0 at 216.39.170.247>, whisper at oz.net (Dave LeBlanc) 
wrote:


> Ok, read the PEP (more or less again). It's not all that clear, but
> does answer my original questions.
>
> Could someone please clarify the scoping wrt to classes? I would
> expect that the following code would refer to the variable at the
> class scope, but it's not clear to me if it would do so:
>
> class foo
>  bar = 1
>  def fee
>   print bar

AFAIK that won't work in the new scoping rules. Specifically see the
discussion section at http://python.sourceforge.net/peps/pep-0227.html , but
it sounds like bar is unresolved:

    The
    name resolution rules are typical for statically scoped languages,
    with three primary exceptions:

        - Names in class scope are not accessible.

    <snip>

    If a name binding
    operation occurs in a class definition, it creates an attribute on
    the resulting class object.  To access this variable in a method,
    or in a function nested within a method, an attribute reference
    must be used, either via self or via the class name.


In your example fee sounds like it needs to be defined along the lines of:

class foo:
    bar = 1
    def fee(self):
        print self.bar

I could be wrong about this (I don't have access to a 2.1 binary), but
that's how it reads to me.



From aleaxit at yahoo.com  Wed Mar  7 05:43:11 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Wed, 7 Mar 2001 11:43:11 +0100
Subject: Can Python replace Visual Basic? Should it?
References: <mailman.983910669.7085.python-list@python.org>
Message-ID: <9853cc016ks@news1.newsguy.com>

"J.J. Dukarm" <jdukarm at hydracen.com> wrote in message
news:mailman.983910669.7085.python-list at python.org...
    [snip]
> The three biggest problems for me were: (a) finding a suitable GUI (I
chose
> wxPython); (b) finding a robust way to communicate with commercial
database
> products (I am experimenting with various ways of doing ODBC, but am still
> nervous); and (c) packaging the app for distribution (just started
figuring
> it out).

For (c), look at both py2exe (simpler) and McMillan's Installer (richer).

For (b), do have a look at ADO.  Yes, I know you want to get free from
Microsoft, but COM is what they generally do _right_, and ADO/OLEDb is
really a very good architecture; you generally get a choice of OLEDb
providers when talking via ADO to any database (e.g., for Oracle, there
are providers made by Oracle and others made by Microsoft, plus, you
CAN fall back to the generic provider which dresses up ODBC as OLEDb
as a last-ditch measure, but you generally won't need to) and get good
performance AND flexible programming.

The only real minus is that it will still tie you to Windows, where
all of the rest of your choices might let you sell a Linux version
too.  You might work around that by doing a DB-API compliant wrapper
on top of ADO (something I've long thought about).  But I guess that
if portability is crucial, then Lemburg's mxODBC is probably your
best bet, particularly now that it's "gone commercial".


Alex





From kens at sightreader.com  Thu Mar 15 19:01:59 2001
From: kens at sightreader.com (Ken Seehof)
Date: Thu, 15 Mar 2001 18:01:59 -0600
Subject: Cookies
Message-ID: <008301c0adac$52fb1260$04090a0a@upcast.com>

Short version:

Is it possible for a stand-alone python application to read a cookie without the help of a browser?"

Long version: 

I have a web page W and a client side python script P.  I want W to generate a GUID (perhaps using javascript) which is then sent to a server S, where it is to be used as a key in a database.  Now I run P (on the client, and not in a browser) and I want P to somehow find the value of the GUID, so that it can query the database on S.


Alternate question:

Is there a technique common to both javascript and python that will obtain a unique identifier for a computer (e.g. IP address).  It must not be browser specific.


Motivation:

I need to write a record of data from an html form, keyed by some kind of unique client id, to the server and then be able to query that record from a stand alone python script running on the client.  The missing link is getting the html form and the python script to agree on a value for the id; the rest is easy.


- Ken

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20010315/58c1b896/attachment.html>

From n8gray at caltech.edu.is.my.e-mail.address  Fri Mar  2 20:15:18 2001
From: n8gray at caltech.edu.is.my.e-mail.address (Nathan Gray)
Date: Fri, 02 Mar 2001 17:15:18 -0800
Subject: python-dev summary 2001-02-15 - 2001-03-01
References: <mailman.983559723.29891.python-list@python.org>
Message-ID: <3AA045A6.3060408@caltech.edu.is.my.e-mail.address>

Tim Peters wrote:

> [Michael Hudson]
> 
>>  ...
>>  The iteraators discussion seems to have disappeared.  At least, your
>>  author can't find it!
> 
> 
> Greg Wilson set up an egroups (now Yahoo) discussion group for it, where
> debate has continued:
> 
>     http://groups.yahoo.com/group/python-iter

Any reason why non-members are excluded from reading the archives on this list?

-n8

-- 
_.~'^`~._.~'^`~._.~'^`~._.~'^`~._.~'^`~._
               Nathaniel Gray
     California Institute of Technology
      Computation and Neural Systems
       n8gray <at> caltech <dot> edu
_.~'^`~._.~'^`~._.~'^`~._.~'^`~._.~'^`~._



From tjg at exceptionalminds.com  Thu Mar 15 12:40:38 2001
From: tjg at exceptionalminds.com (Timothy Grant)
Date: Thu, 15 Mar 2001 09:40:38 -0800
Subject: Process details
In-Reply-To: <slrn9b14cv.fae.quinn@riyal.ugcs.caltech.edu>; from quinn@riyal.ugcs.caltech.edu on Thu, Mar 15, 2001 at 09:55:43AM +0000
References: <mailman.984439220.26863.python-list@python.org> <slrn9b14cv.fae.quinn@riyal.ugcs.caltech.edu>
Message-ID: <20010315094037.J3109@trufflehunter.avalongroup.net>

On Thu, Mar 15, 2001 at 09:55:43AM +0000, Quinn Dunkan wrote:
> On Mon, 12 Mar 2001 15:22:16 -0800, Timothy Grant <tjg at exceptionalminds.com>
> wrote:
> >Hi all,
> >
> >I hope everyone is enjoying there first week back after the
> >events of P9.
> >
> >I need to create a list of processes and process information. I
> >have created a class that contains much of the information in
> >/proc/'pid'/status.
> >
> >It works quite well, however, it takes about four seconds to
> >populate a list of 90 procs. This seems a bit long. Is there a
> >faster way to get ahold of the information than reading /proc?
> 
> You don't say what OS you're using, but from the sig I'm guessing Linux.  In
> Linux, /proc is the official interface.  /bin/ps and friends use it.  Try
> 'cat /proc/[0-9]*/status >/dev/null'.  If that takes less than four seconds,
> then the slowdown is likely in your code, not /proc.  On this linux machine,
> it takes a few hundredths of a second for several hundred procs.

Oh, I'm sure the slowdown is in my code. the cat mentioned
above is extremely fast...

...I just went and looked at my code and discovered the major
bottleneck--I think. The problem comes due to the fact that not
all process list all entries in the /status file. This means
that a sequential read assigning to variables won't work, so I
am actually looking for entry names in a loop.

I just added a break statement and shaved three seconds of the
time! I'm sure there is more room to optimize, but exiting my
loop after I found what I was looking for sure helps.

Sometimes I just need a second set of eyes even if they don't
even see my code. Thanks.

-- 
Stand Fast,
    tjg.

Timothy Grant                         tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.         <><       (503) 246-3630
>>>>>>>>>>>>>Linux, because rebooting is *NOT* normal<<<<<<<<<
>>>>This machine was last rebooted:  57 days 21:47 hours ago<<



From dfassett at cts.com  Tue Mar  6 02:08:11 2001
From: dfassett at cts.com (Dave Fassett)
Date: Mon, 5 Mar 2001 23:08:11 -0800
Subject: Hang in embedded Python (Windows) app
Message-ID: <9822i5$bkn$1@thoth.cts.com>

I'm writing a Windows app with an embedded Python interpreter. I have the
stdio output going to my own window and I've made a few working custom
object in C that I can use from scripts. My problem is the system will hang
when running the script, and if I break in the debugger it's inside of NT,
with no real stack trace present. This happens when I run my script
repeatedly - and always when I sit on the accelerator key that runs the
current script (i.e., repeating the script as fast as possible). Even the
simplest script that uses no custom objects will hang (such as "for p in
range(1,10):
 print p" - which doesn't hang if I repeat it slowly), so it's not anything
about reference counts or using custom objects incorrectly. I've compiled
Python 2.0 under Windows using VC++ 6.0 using threads. This is a real
showstopper, so any help would be greatly appreciated!

Dave





From sk+10 at cs.brown.edu  Sat Mar 10 23:55:53 2001
From: sk+10 at cs.brown.edu (Shriram Krishnamurthi)
Date: 10 Mar 2001 23:55:53 -0500
Subject: Questions from a beginner
References: <1103_983792888@breakdown> <7D924B83F563035A.78ADE854BCA8F2C6.DA35B8F7D112D810@lp.airnews.net> <BAC83A82CDF10F60.77025EDCDF82B08B.EF81409F68BD8FE6@lp.airnews.net>
Message-ID: <w7dwv9wgazq.fsf@cs.brown.edu>

claird at starbase.neosoft.com (Cameron Laird) writes:

> Well, Scheme's my second choice for a first language <URL:
> http://www.sunworld.com/sunworldonline/swol-10-2000/swol-1013-regex.html >
> Why only second?  For me, it's 'cause Python remains easier
> to embed-and-extend, and has a wider range of what I regard
> as significant GUI toolkit bindings, and Win* conveniences.
> Also, I still have mixed feelings about Scheme in large-team
> projects.

Oh please, Cameron.

None of the topics you list have been a real issue in teaching
beginners to program, in my considerable experience in this arena.

Furthermore, if they really are issues in *your* educational context,
pick DrScheme.  It provides most of them, and more.

Finally, there are many significant educational benefits to Scheme:

1. Much more sophisticated pedagogy.

2. Much more sophisticated programming environments.

What large team projects have to do with the choice of a first
programming language is slightly beyond me, but even in that
department I'm still waiting for the day Python catches up with
DrScheme's units and mixins approach.

Shriram


From emile at fenx.com  Sat Mar 24 11:23:56 2001
From: emile at fenx.com (Emile van Sebille)
Date: Sat, 24 Mar 2001 08:23:56 -0800
Subject: difference between script and prompt ! looking for the Problem
References: <3ABCCA03.90871133@noos.fr>
Message-ID: <99iht4$1c7hg$1@ID-11957.news.dfncis.de>

You need to print from within a script.  Interactive mode assumes it.

ie:

print c.r, c.i

HTH
--

Emile van Sebille
emile at fenx.com

---------
"William Famy" <william.famy at noos.fr> wrote in message
news:3ABCCA03.90871133 at noos.fr...
> Hi.
>
> I this is my problem.
>
> I have a file fonc.py where i write a class.
>
> class comp:
> def __init__(re,im):
> self.r=re
> self.i=im
>
>
>
> when i use a python shell i see this
>
>
> >>> import fonction
> >>> c=fonction.comp(3.0,4.5)
> >>> c.r,c.i
> (3.0, 4.5)
> >>>
>
> but when i use the following script
>
> test.py
> #!/usr/bin/python
> import fonction
> c=fonction.comp(3.0,-4.5)
> c.r, c.i
>
>
> i have no result
>
> bash-2.04$ ./test.py
> bash-2.04$
>
> Where is the problem? where is my error?
>
> Thanks for answer.
> --
>        \\\|///
>                            \\  - -  //
>                             (  @ @  )
> +-------------------------oOOo-(_)-oOOo-------------------------+
> |  Famy William  36 avenue des barattes  74000 Annecy  France   |
> |  email:william.famy at noos.fr william.famy at mail-enseignant.com  |
> +---------------------------------Oooo--------------------------+
>                            oooO   (   )
>                           (   )    ) /
>                            \ (    (_/
>                             \_)




From stephen_purcell at yahoo.com  Sun Mar  4 09:36:35 2001
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Sun, 4 Mar 2001 15:36:35 +0100
Subject: Missing the functional mind set
In-Reply-To: <mlb2at8ptrr7sh5dgq8qlk2201kquh6tfl@4ax.com>; from danielk@aracnet.com on Sat, Mar 03, 2001 at 10:08:48AM -0800
References: <mlb2at8ptrr7sh5dgq8qlk2201kquh6tfl@4ax.com>
Message-ID: <20010304153635.A2779@freedom.puma-ag.com>

Daniel Klein wrote:
> The purpose of this code is to return a list of dictionary values sorted by
> key, resulting in 'templist'.
> 
> >>> d = {3: 'd', 2: 'b', 1: 'a', 0: 'z'}
> >>> templist = []
> >>> for n in range(len(d)):
> 	templist.append(d[n])
> 
> >>> templist
> ['z', 'a', 'b', 'd']

The problem is that sorting a list can't be done in a functional way.

Either of the following is vaguely functional, though:

	>>> d = {3: 'd', 2: 'b', 1: 'a', 0: 'z'}
	>>> keys = d.keys()
	>>> keys.sort()
	>>> map(d.get, keys)
	['z', 'a', 'b', 'd']


or

	>>> d = {3: 'd', 2: 'b', 1: 'a', 0: 'z'}
	>>> stuff = d.items()
	>>> stuff.sort()
	>>> map(lambda i: i[1], stuff)
	['z', 'a', 'b', 'd']

The second is probably more efficient.

-Steve

-- 
Steve Purcell, Pythangelist
Get testing at http://pyunit.sourceforge.net/
Get servlets at http://pyserv.sourceforge.net/
"Even snakes are afraid of snakes." -- Steven Wright



From jwbnews at scandaroon.com  Thu Mar 15 03:03:53 2001
From: jwbnews at scandaroon.com (John W. Baxter)
Date: Thu, 15 Mar 2001 00:03:53 -0800
Subject: This math scares me
References: <3aad4062.1290021@News.CIS.DFN.DE> <20010312170356.B3513@better.net> <mailman.984480856.18097.python-list@python.org> <984486395.407379@newsmaster-04.atnet.at> <slrn9asf1d.ja3.qrczak@qrnik.zagroda> <984518199.127140@newsmaster-04.atnet.at> <jwbnews-09D75F.13313313032001@corp.supernews.com> <98ohuc$du9$1@news.udel.edu> <7bQr6.644$lz1.17686@ruti.visi.com> <98oku1$f6a$1@news.udel.edu> <erRr6.659$lz1.20662@ruti.visi.com>
Message-ID: <jwbnews-9FD55E.00035215032001@corp.supernews.com>

In article <erRr6.659$lz1.20662 at ruti.visi.com>,
 grante at visi.com (Grant Edwards) wrote:

> I believe that there are plans (on some exchanges) to switch
> pricing from base-2 fractions to decimal fractions.

There is a mixture right now on NYSE (not sure about NASDAQ).  Some 
issues fractions; some issues cents.

(I haven't followed government bonds for a while...are they still 
reported as dollars and 32nds (per $100 face), with the 32nds 
represented like
   99.16
(meaning 16/32, or 50 cents)?  That one is (or was) a little hard to get 
used to

By mid April, the conversion on NYSE and I think NASDAQ is supposed to 
be complete...a bunch more issues convert any moment now.

John...whose old line brokerage firm in Ann Arbor arranged a floor visit 
to  the ASE.  I had to tag along with the firm's floor broker.  It 
was--by accident--the perfect day:  they were (almost) all dressed in 
1920s outfits rather than the required smocks, in honor of a major 
anniversary of the move of the American Stock Exchange to their indoor 
quarters from outside (on "the Curb" literally) under a buttonwillow 
tree.  Fun!!!  [My visit was in 1970 or so.]

-- 
John W. Baxter   Port Ludlow, WA USA  jwbnews at scandaroon.com


From topmind at technologist.com  Sat Mar  3 20:10:28 2001
From: topmind at technologist.com (Topmind)
Date: Sun, 04 Mar 2001 01:10:28 GMT
Subject: Collection interfaces
References: <3a90fa97.0@news1.mweb.co.za> <3A933830.7F98D118@ix.netcom.com> <CgJk6.515$v77.66568@newsread2.prod.itd.earthlink.net> <220220010813598042%cmh@bDistributed.com> <ud7cakvi8.fsf@roundpoint.com> <MPG.150312ceb68f1ada98969f@news.earthlink.net> <8Gem6.473353$U46.14105062@news1.sttls1.wa.home.com> <MPG.1503548620e632919896a4@news.earthlink.net> <MCjm6.474652$U46.14163288@news1.sttls1.wa.home.com> <MPG.150394195f7f04a19896a7@news.earthlink.net> <3A9A86B1.9A8C69A2@ix.netcom.com> <MPG.1504d14563dce6499896af@news.earthlink.net> <97fjla128ir@news2.newsguy.com> <MPG.1505a76831c34b749896b3@news.earthlink.net> <97h5s40vhc@news2.newsguy.com> <MPG.1506007591cf818b9896bc@news.earthlink.net> <slrn99opop.gr.amk@mira.erols.com> <MPG.1506265ebf6d934d9896be@news.earthlink.net> <97ibmc0k47@news1.newsguy.com> <MPG.1506cf3293b094b49896c1@news.earthlink.net> <97m0c7080c@news2.newsguy.com> <MPG.15085517debf4b649896ca@news.earthlink.net> <zb%n6.31033$o%6.586026@news1.rdc1.bc.home.com>
Message-ID: <MPG.150b383cea226d409896d0@news.earthlink.net>

> I notice comp.lang.java has been missing from your incessant tirades.
> 
> How about this: OO has widespread market acceptance.
> 
> You want metrics? Look here: http://www.monster.com.
> 

I don't dispute it's popularity, at least as a buzzword
(actual decent usage of OO is even disputed by many
OO fans.)


> Understanding OO is a good career move; science is irrelevant. I doubt that
> anyone can "prove" anything about this industry that doesn't have a dollar
> sign associated with it. Most OO people I know do it because they can make a
> good living at it and they find it intellectually satisfying.
> 
> Please take your discussion to comp.table.programming (I hear it's the next
> big thing).
> 

You never know. Merit and proof is not even a prerequisite.

>
> Maybe if you identify yourself you'll be taken more seriously. Better yet,
> go soak your head.
> 


It hurts one career to admit distaste for OO. I have a family to
take care of. Where does Salmon Rushdie hang out, btw?


> I apologize to all affected for responding to an anonymous troll.
> 
> Ugh.
> 
> Steve
> 
> "Topmind" <topmind at technologist.com> wrote in message
> news:MPG.15085517debf4b649896ca at news.earthlink.net...
> >
> > >
> > >I've often noticed that demanding metrics is the last
> > >line of defense of the conservatives in our field against
> > >innovation
> >
> > So science be damned? At least it is official now.
> >
> > I am *not* against innovation. I am against the overhyping
> > of fads without proof.
> >

[snip]




From digitig at cix.co.uk  Tue Mar  6 13:17:00 2001
From: digitig at cix.co.uk (Tim Rowe)
Date: Tue, 6 Mar 2001 18:17 +0000 (GMT Standard Time)
Subject: Proposal: allow '?' and '!' in identifiers
References: <uwvakgkx5.fsf@cs.uu.nl>
Message-ID: <memo.20010306181751.34635B@digitig.compulink.co.uk>

In article <uwvakgkx5.fsf at cs.uu.nl>, piet at cs.uu.nl () wrote:

> >>>>> Joshua Marshall <jmarshal at mathworks.com> (JM) writes:
> 
> JM> Christoph Horst <rc-bashar at gmx.net> wrote:
> >> Nathaniel Gray <n8gray at caltech.edu.is.my.email.address> wrote:
> >>> IMHO it's not worth adding "!", but if "?" doesn't break anything 
> > then it's >>a net gain.  I assert that it _would_ make Python more 
> > expressive to add an >>elegant way of indicating that a function 
> > returns only boolean values.
> 
> >> What's wrong with prefixing the function name with 'is'?
> 
> JM> I wouldn't say anything is wrong with that approach.  But then I 
> > could
> JM> also ask "Why do we need underscores as valid identifier characters?
> JM> You can just capitalize the next letter."
> JM> The question mark is an unused character, and carries some
> JM> natural-language meaning.  It's useful.
> 
> Please don't do it. There are more useful things to do with ? like the
> ternary operator for conditional expressions.
> -- 
> Piet van Oostrum <piet at cs.uu.nl>
> URL: http://www.cs.uu.nl/~piet [PGP]
> Private email: P.van.Oostrum at hccnet.nl

But please not in Python. One of the things that makes Python so clean, 
AFAICS is that it has /very/ few notations that are not remarkably close 
to their natural language meanings -- even (), [] and {} evoke some notion 
of grouping. ".", "!=", "==", sadly with 2.x "+=" etc, but still 
remarkably few. Anyway, the ternary operator is deprecated in every C/C++ 
style guide I have ever seen, so it's hardly a good reason to muddy the 
clarity of the language.


From cribeiro at mail.inet.com.br  Fri Mar 30 23:39:59 2001
From: cribeiro at mail.inet.com.br (Carlos Alberto Reis Ribeiro)
Date: Sat, 31 Mar 2001 01:39:59 -0300
Subject: Why not an __assign__ method?
In-Reply-To: <4.3.1.2.20010330115856.0256bb00@exchange.starmedia.net>
References: <5.0.2.1.0.20010330125432.0216c4b0@mail.inet.com.br>
Message-ID: <5.0.2.1.0.20010331004906.021bc9b0@mail.inet.com.br>

Robin, please have some patience, and read my entire answer before 
replying. I think I have a good point here.

At 14:41 30/03/01 -0500, Robin Thomas wrote:
>At 01:26 PM 3/30/01 -0300, Carlos Alberto Reis Ribeiro wrote:
>
>>The question is: Why not have a __assign__ method, that gets called on 
>>assignment? It should be called on the *right side* of the expression (at 
>>the left side it does not make sense - I leave the proof as an exercise 
>>for the reader <wink>). The return value would be then assigned to the 
>>left side of the expression.
>
>You should really get the source to CPython and start reading it.

I'll do it, I promise :-)

>>z = a + b + c
>
>The "no changes to Python" implementation of your example is:
>
>z = a + b
>z += c

I've pointed out at such alternative in an earlier email (this morning, 
with other subject). The concern then is that this leads to code that's 
harder to understand. It would be better to have the expression in it's 
original format, and leave the optimizations for the compiler. Nobody said 
that it was easy :-)

I also have previously pointed out that I believed that the intermediate 
object was actually instantiated inside the __add__ method of NumPy. I have 
some reasons to believe so, even before reading the code (more on this later).

As you suggested, I checked the bytecode that is generated. It's simply a 
stack-based calculator:

 >>> import dis
 >>> p = compile('z = a + b + c', '<string>', 'single')
 >>> dis.disassemble(p)
           0 SET_LINENO               0
           3 SET_LINENO               1
           6 LOAD_NAME                0 (a)
           9 LOAD_NAME                1 (b)
          12 BINARY_ADD
          13 LOAD_NAME                2 (c)
          16 BINARY_ADD
          17 STORE_NAME               3 (z)
          20 LOAD_CONST               0 (None)
          23 RETURN_VALUE
 >>>

After reading this, I still think that it is possible to implement my 
proposal. Please note that I'm talking about two related things here: the 
__assign__ proposal (that only involves a change on STORE_NAME behavior, as 
it is shown on step 5 below); and the NumPy optimization, made possible by 
the __assign__ feature, by using a temporary flag on the intermediate objects.

1) The Python compiler itself didn't create any new object. The bytecode 
only asks for a BINARY_ADD. This is going to call some C code, that in turn 
must create an intermediate variable to store the result before returning. 
In the case of the NumPy array, the code for this must be on the numerical 
extensions itself.

(I'll check this, I promise. But please keep on reading...)

2) Now imagine that, while creating the object to store the intermediate 
value, we set a 'temp' attribute on the new object. The operation is done 
(in this case, BINARY_ADD at pos 12), and the resulting object is returned, 
still with the 'temp' flag set.

3) Then, at pos 16, other BINARY_ADD is issued. Some C code is called 
again. Inside this code, we check the temp attribute. If the temp attribute 
is true, *and* the operation can be done in-place, then we do it this way. 
The object is again returned, still with the temp flag set.

4) Note that not all operations can be done inplace. However, for arrays, 
any operation that is done element-by-element can be done this way. Add is 
an obvious example.

5) At line 17, we have the STORE_NAME opcode. Again, inside C code, we 
would call the __assign__ method of the object that is on the top of the 
stack. The __assign__ method would then return the object to be assigned. 
By default, that is just a "stub" function, that returns the object itself.

6) By intercepting __assign__, the NumPy code could check the temp flag on 
the array object. The flag is still there; so we clear the flag, to point 
out that in the future it is not safe to do the operation inplace anymore.


>Think about what your "temp flag" would do under the current behavior of 
>threads in Python.

Why threads are not a problem?

- Because temporary objects have meaning only in the context of the 
mathematical expression where they are create/evaluated/returned;
- Because each thread have their own stack for mathematical operations 
(even in Stackless Python; the word 'stackless' has nothing to do with the 
mathematical stack);
- Because we don't do multithreaded evaluation of mathematical expressions; 
or, to put it other way: any mathematical expression is always evaluated in 
the context of a single thread. If we did otherwise, then the code above 
could fail; but this is not the case.


>into the syntax-illegal but perhaps (and perhaps not) bytecode-legal:
>
>z = ((a + b) += c)

I have some reasons to believe that the hack above is bytecode-legal. I'll 
try it with NumPy this weekend (I'll need to grap the bytecode assembler 
first).

>z = a + b + c % (d * e) / f ^ (g or 42)

The apparent complexity of the expression changes nothing. In the end 
everything is just turned into a very simple stack calculator. Anyone who 
ever programmed in Forth, PostScript, or HP calculators will recognize it 
<wink>.



Carlos Ribeiro





From paul at fxtech.com  Thu Mar  1 09:48:23 2001
From: paul at fxtech.com (Paul Miller)
Date: Thu, 01 Mar 2001 08:48:23 -0600
Subject: python performance
References: <J8fn6.756$me1.378895@news2.cableinet.net>
Message-ID: <3A9E6137.974AE84F@fxtech.com>

> Can  anyone comment on the speed of Python programs under Windows?
> I am interested in using Python as a embedded game scripting language.
> Amongst other things, I want to doing AI, but not the low level routines
> like path finding.


Go out and buy Kingdom Under Fire. It uses Python for its AI and
scripted events.


From ws-news at gmx.at  Tue Mar  6 16:19:15 2001
From: ws-news at gmx.at (Werner Schiendl)
Date: Tue, 6 Mar 2001 22:19:15 +0100
Subject: Can Python replace Visual Basic? Should it?
References: <5DZo6.11126$hn5.1578410@news1.rdc1.mb.home.com> <98266k0p5h@news1.newsguy.com> <3q8p6.12338$hn5.1798569@news1.rdc1.mb.home.com> <983898657.328387@newsmaster-04.atnet.at> <pF9p6.12685$hn5.1803263@news1.rdc1.mb.home.com>
Message-ID: <983913833.132829@newsmaster-04.atnet.at>

Brad,

Of course the things you mention are annoying and maybe really nasty some
time.
I do not consider restrictive IDE good IDEs.

In fact I think best is to have the _possibility_ to see and change whats
really in the box.
But not the _need_ to, that a lot of command line toolkits have.

Also I find it more convenient to draw the dialogs than to code something,
try it, see its ugly, rearange things, try.......

I personally like the way visual basic handles this visual parts of the
program.
I cannot find it very restrictive, at least in current versions (6 namely)

What I hate about VB is that it costs real headaches if a project grows to
big for the IDE.
It seems to get better, when the machine has more RAM.

But I have had a project with about 3 MB of sources (everything VB can do,
classes, ocx, forms,...)
and it repeatedly lost settings, could not open - _this_ is what annoys me
about an IDE

> then. As for VB, it can be said simply that I love playing with OO (yes, I
> call it "playing" even though it's what's paying the bills :) and of
course,
> VB doesn't offer that yet.

I a pure sense, of course you may be right.
VB does not support source code reuse by inheritance.
But it supports polymorphisms (lookup the Implements keyword, if you dont
believe...)
Of course it supports encapsulation.

regards
werner





From not.this at seebelow.org  Wed Mar 28 09:21:49 2001
From: not.this at seebelow.org (Grant Griffin)
Date: Wed, 28 Mar 2001 08:21:49 -0600
Subject: NumPY: Reading/Writing Arrays To/From Files
References: <3aae3bd6$0$42882$272ea4a1@news.execpc.com>
Message-ID: <3AC1F37D.E8ADA48@seebelow.org>

Ben Held wrote:
> 
> I would like to be able to read/write NumPY array objects to/from ascii
> files.  I could not find support for this in the NumPY code.  Any ideas?
> 
> Ben

I forget exactly, but I think there's something out there called
"arrayio".  Look for Travis Oliphant's page: I think you can find it
there.

Alternatively, you can do what I did once, which is to code it up
yourself using a Python list, then convert the list into a NumPy array. 
(This is probably slower than arrayio, but it works.)

necessity-is-the-mother-of-invention-ly y'rs,

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation            http://www.iowegian.com


From no-unsolicited-mail-to-swart at bigserver.com  Sat Mar  3 00:21:35 2001
From: no-unsolicited-mail-to-swart at bigserver.com (Steve Wart)
Date: Sat, 03 Mar 2001 05:21:35 GMT
Subject: Collection interfaces
References: <3a90fa97.0@news1.mweb.co.za> <3A933830.7F98D118@ix.netcom.com> <CgJk6.515$v77.66568@newsread2.prod.itd.earthlink.net> <220220010813598042%cmh@bDistributed.com> <ud7cakvi8.fsf@roundpoint.com> <MPG.150312ceb68f1ada98969f@news.earthlink.net> <8Gem6.473353$U46.14105062@news1.sttls1.wa.home.com> <MPG.1503548620e632919896a4@news.earthlink.net> <MCjm6.474652$U46.14163288@news1.sttls1.wa.home.com> <MPG.150394195f7f04a19896a7@news.earthlink.net> <3A9A86B1.9A8C69A2@ix.netcom.com> <MPG.1504d14563dce6499896af@news.earthlink.net> <97fjla128ir@news2.newsguy.com> <MPG.1505a76831c34b749896b3@news.earthlink.net> <97h5s40vhc@news2.newsguy.com> <MPG.1506007591cf818b9896bc@news.earthlink.net> <slrn99opop.gr.amk@mira.erols.com> <MPG.1506265ebf6d934d9896be@news.earthlink.net> <97ibmc0k47@news1.newsguy.com> <MPG.1506cf3293b094b49896c1@news.earthlink.net> <97m0c7080c@news2.newsguy.com> <MPG.15085517debf4b649896ca@news.earthlink.net>
Message-ID: <zb%n6.31033$o%6.586026@news1.rdc1.bc.home.com>

I notice comp.lang.java has been missing from your incessant tirades.

How about this: OO has widespread market acceptance.

You want metrics? Look here: http://www.monster.com.

Understanding OO is a good career move; science is irrelevant. I doubt that
anyone can "prove" anything about this industry that doesn't have a dollar
sign associated with it. Most OO people I know do it because they can make a
good living at it and they find it intellectually satisfying.

Please take your discussion to comp.table.programming (I hear it's the next
big thing).

Maybe if you identify yourself you'll be taken more seriously. Better yet,
go soak your head.

I apologize to all affected for responding to an anonymous troll.

Ugh.

Steve

"Topmind" <topmind at technologist.com> wrote in message
news:MPG.15085517debf4b649896ca at news.earthlink.net...
>
> >
> >I've often noticed that demanding metrics is the last
> >line of defense of the conservatives in our field against
> >innovation
>
> So science be damned? At least it is official now.
>
> I am *not* against innovation. I am against the overhyping
> of fads without proof.
>
> Relational DB's are newer than OO, does that mean that
> OO defenders are "holdouts"?
>
> If OO is an "experiment", then fans of it should admit
> it as such. The road is littered with paradigms and
> ideas that were over-hyped, and then shrank back
> to where they belong. Heavy top-down is an example
> that ran about a decade in the limelight. Expert
> systems, CASE, etc, are others.
>
> I hope Table Oriented Programming hits hard and does
> the same. All the OO celebs will squirm. Call me
> cruel, but they deserve it. It deserves as much
> of a chance for experimentation in the mainstream
> as anything out there now.
>
>
> > -- since nobody's really funded to run serious
> >(and hugely costly) controlled experiments, such demanders
> >can rest reasonably tranquil that they won't be faced with
> >such metrics.  Unfortunately, the world has this habit of
> >passing them by anyway.
>
> No, just learning by trial and error. 50 years from
> now we will have a better perspective on this OO
> experiment.
>
>
> > Been there, done that, more or less whatever you mean by
> > "table-oriented programming"
>
> It may be purely subjective. I had an affinity for grids
> and tables even in 2nd grade. Born to Table if you will.
>
>
> > ...then your resulting sources will never be as good,
> > clear, clean, maintainable, and high-quality, as those
> > of a better programmer, who DOES refactor every time
> > the occasion affords.
>
> My favorite approaches are *not* orthogonal to refactoring,
> but simply *less* fragile or messed up if you refactor
> regularly.
>
>
> > And, depending on the axis of change, any paradigm
> > can result in code that is hard to change to meet
> > the new/changed needs at equally high quality
>
> Well, my experience tells me that the "task axis"
> is the most "business friendly" axis to build
> along (being that multi-axis paradigms still need
> a lot of work and thus we are forced to emphasize
> one at the expense of others for now.)
>
>
> >You still need design ability, flair, and a decent
> >amount of domain-specific experience somewhere in
> >the team, to enjoy a substantial chance of delivering
> >LARGE software-systems that present all the desired
> >quality attributes
>
> The SIZE thing again?
>
> 1. What is good for large size may not be optimal for
> other sizes.
>
> 2. OO fans disagree about at what size OO shines in.
>
> 3. The task-centric approach is size friendly because
> the task is mostly impervious to the rest of the
> system (unless the task changes it's requirements, which
> is an issue in any paradigm).
>
> OO's noun-centricity seems more scale-poor in that
> regard because schemas change when growth appears.
> The 3-graph approach puts a layer between schema
> changes and the task more or less.
>
>
> > Fortunately, given the current market for IT professionals (and
> > this comment applies to the last 25 years), no competent worker
> > in the field ever needs to be stuck for long under management
> > that is totally impervious to common sense and reason
>
> For one, I don't buy the labor shortage crap. For example, try
> to get a telecommuting job. If there was an IT labor shortage, then
> bosses would fold on their demand to see a physical butt in a
> physical chair.
>
> Second, the percentage of PHB's is too great to avoid.
> The supply of non-PHB's is too low to satisfy the
> desire of tech workers to escape them.
>
>
> >> I have shown in great detail why procedural seems better able
> >> to deal with non-boundaried differences, and I have yet to
> >
> > You have done no such thing (at least in your posts in
> > this thread, and the URL you mentioned here).
>
> I don't know what you have not seen yet, but my collection
> of examples is far greater than what the other side
> offers as far as comparative examples. All I get from
> them are the typical shape, animal, and stack examples
> (and a few other "classics").
>
> Regarding an internal* component view, I will believe it when I
> actually see it work. Like much in OO lore, it sounds
> wonderful on paper, but rubs rough with the nitty
> gritty of reality. I guess that is what happens when
> you design an entire paradigm around idealized shape,
> animal, and stack examples.
>
> * External components are quite successful, but nobody
> has figured out how to apply it to internal systems
> very well. BTW, OO is not a prerequisite for external
> components.
>
>
> > maybe because you lack the experience to do it well.
>
> Well, youses all seem to lack the articulation skills/experience
> to describe why granularity is not a problem. You still
> have not found an easy way to override 1/3 of a method.
> You just hand-wave that "experience will magically fix
> it all". I seems like you just "get used to" refactoring
> when it happens. IOW, tolerate bad weather because you
> lived their for so long.
>
> > But -- without calling it that way, without the productivity
> > that reusable patterns or good tools can give me
>
> Patterns are not native to OO and "good tools" can be built
> to lessen the shortcomings of *any* paradigm. Name a p/r
> fault and I could describe a potential tool to help.
>
> -tmind-
>




From paulp at ActiveState.com  Tue Mar 20 17:12:00 2001
From: paulp at ActiveState.com (Paul Prescod)
Date: Tue, 20 Mar 2001 14:12:00 -0800
Subject: ASCII encoding error: ordinal not in range(128)
References: <91o5ih$2kv$1@news.wrc.xerox.com> <20001219172933.B29727@ActiveState.com> <006201c0b184$a60b9430$5aa4af0d@dssc.mc.xerox.com>
Message-ID: <3AB7D5B0.AB7C7B5D@ActiveState.com>

Jose Isaias Cabrera wrote:
> 
> ...
>
>   File "C:\myprograms\qna0.py", line 1156, in FindIfAttachments
> 
>     name = code.encode()
> 
> UnicodeError: ASCII encoding error: ordinal not in range(128)

I'm going to presume that "code" is a Unicode string. In that case, you
have to say HOW you want it encoded. What do you want to do with the
non-ASCII characters? Probably what you want is:

"abc".encode("Latin-1")

That means that characters between 128 and 255 will be encoded as the
equivalent byte-codes. Characters above 255 will simply not be
representable. I'm presuming that you won't run into any of those...if
you will be, then you need to think about representation some more.

-- 
Take a recipe. Leave a recipe.  
Python Cookbook!  http://www.activestate.com/pythoncookbook



From cfelling at iae.nl  Sun Mar 18 22:08:12 2001
From: cfelling at iae.nl (Carel Fellinger)
Date: 19 Mar 2001 04:08:12 +0100
Subject: Help with new scope problem
References: <993bar$7h5$1@news.chatlink.com>
Message-ID: <993t6s$el4$1@animus.fel.iae.nl>

Luis Cortes <lcortes at aodinc.com> wrote:
> Hello,

> I have a piece of code that I think should work fine, but I believe that
> I've hit a scope problem (Python 2.1b1).  Does anyone out there have a hint
> as to how to fix it??

> code:
...sniped

if I use the exact same code, and add a call to getarguments it seems to
work here, but...

> def getarguments():
...
>         print box, gifname, newgifname

Surely *this* print statement gave the expected outcome?
It does here.

>         sys.exit(0)
>         return ( box, gifname, newgifname )

> THE PROBLEM:  when I have more than 3 arguments, the global variables do not
> change to the correct variables, but instead print their default values.


-- 
groetjes, carel


From ransen_spam_me_not at nemo.it  Fri Mar  9 01:06:27 2001
From: ransen_spam_me_not at nemo.it (Owen F. Ransen)
Date: Fri, 09 Mar 2001 06:06:27 GMT
Subject: Memory leaks in VC6 2 line Python 2.1b1 apps?
References: <3aa88f52.5419185@news.newsguy.com> <hGPp6.53719$UZ4.13963947@news4.rdc1.on.home.com>
Message-ID: <3aa972cb.2633764@news.newsguy.com>

On Thu, 08 Mar 2001 17:53:17 GMT, "Tom" <NoSpam at NoSpam.com> wrote:

>"Owen F. Ransen" <ransen_spam_me_not at nemo.it> wrote in message
>news:3aa88f52.5419185 at news.newsguy.com...
>> Python 2.1b1
>>
>> I tried to create a simple VC MFC app using the
>> wizard as follows:
>>
>> 1) Multidoc shared MFC-DLL application.
>> 2) Run time library debug multithreaded DLL
>> 3) Added #include <python.h> to the main file
>> 4) Added
>>         Py_Initialize()
>>         Py_Finalize()
>>    at the end of the InitInstance()
>> 5) When I run I get memory leaks visible at
>>    exit.
>>
>> Any clues anyone?
>
>For comparison sake.
>I use VC6 and Python 2.0.  I create an EXE that calls the same same
>functions that you mention, and Python itself doesn't leak (I get various
>wierd leaks from other sources, but not Python so far as I can tell).
>
>Tom.


Thanks Tom, Aaargh! Back to 2? 


--
Owen F. Ransen
http://www.ransen.com/
Home of Gliftic & Repligator Image Generators


From kens at sightreader.com  Mon Mar 26 19:46:17 2001
From: kens at sightreader.com (Ken Seehof)
Date: Mon, 26 Mar 2001 16:46:17 -0800
Subject: Some Basic Questions about Python
References: <tbv8g06eg66tec@corp.supernews.com> <E14hfrj-0005CB-00@darjeeling>
Message-ID: <006401c0b657$5699e4b0$04090a0a@upcast.com>

Moshe Zadka <moshez at zadka.site.co.il> says:

> On Mon, 26 Mar 2001 15:14:05 -0500, "Bill Walker" <bwalker at earth1.net>
wrote:
>
> > 1) If I write an application and want to pass it on to others who may
not
> > have Python installed, how do I do this? I understand there is no
compiler
> > for
> > Python and Tk yet that generates .exe files.
>
> There is, and you don't want to use it.

Sure you do.  It's http://starship.python.net/crew/gmcm/installer_dnld.html

It makes executables for Windows and Linux.  Typical results on Windows
are around 500K depending on what modules you import.  I haven't used
it on Linux.

> > 2) Can I distribute the Python install file along with my app without
> > violating a copyright or license?
>
> Yes, and that is a much better way.
> But you don't want to trust some stranger on usenet --- read the license
> yourself.

I think the licensing is the same whether you use Gordons installer or
include python.  (I'm not a lawyer, so ignore this).

> > 3) Better yet, can I include only the base files needed to run my
> > application with the source code? Say Pythonw.exe and only the
appropriate
> > DLLs, libs, etc.
>
> Why the hell? Any windows app comes with 10mb of useless crap anyway,
> why break tradition? ;-)
> More seriously, disk space is cheap, your time isn't.

Yeah, never mind that there are several reasons to make a very small
distributable standalone application (fast download, put it on a floppy,
no install step, etc.)

On the other hand, since we can assume that whoever is running your
application is a programmer anyway, they obviously need python for
other reasons even if they don't know it yet.  Therefore you should
force them to install python too.  It's for their own good.  :-)

- Ken





From sholden at holdenweb.com  Tue Mar  6 18:47:59 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Tue, 6 Mar 2001 18:47:59 -0500
Subject: Run Scripts
References: <mailman.983914809.20356.python-list@python.org>
Message-ID: <gKep6.23838$ML1.1220160@e420r-atl2.usenetserver.com>

"Andrea Ciuffoli" <ACIUFFO1 at irf.com> wrote in message
news:mailman.983914809.20356.python-list at python.org...
Hello,
I am new in Python programming so my question is probably silly.

Starting from the Ftpmirror.py program I found in /Python20/Tools/Scripts,
I've implemented a script to either get and send data from an FTP server (my
OS is Windows NT). Now I would like to run the script automatically each
hour, what is the best way to do that?
Consider that to call my script I need to add several arguments (e.g.
ftp.py -m -p 2164 /test c:/mytest).

Thanks in advance,
Andrea.
+++++

In batch mpde, look at the "at" command, and have the batch job resubmit
itself each time it runs. Or if you have "Scheduled Events" on your system
(look in Windows Explorer), use the graphical interface.

regadrs
 Steve





From qwuhlw at safe.com  Thu Mar 29 14:33:29 2001
From: qwuhlw at safe.com (qwuhlw at safe.com)
Date: Thu, 29 Mar 2001 19:33:29 GMT
Subject: --------- MUST READ-----------------  3562
Message-ID: <d6Mw6.963$b43.2150@NewsReader>

EVIDENCE ELIMINATOR:
 The unique program which professionally cleans and speeds up your
PC..........

DELETING CACHE AND HISTORY WILL NOT PROTECT YOU!!!
You dont want to be caught out!!...... make your PC safe...

THIS PROGRAM IS A MUST!!! CHECK IT OUT...............

http://www.evidence-eliminator.com/go.shtml?A654845

THE NETS HOTTEST SOFTWARE!!!!!!!!!!!!!!!!!!!

nbihnzubusjgtidpzxylkdjqguzwzyfmuydztcljtyihpvmsfqmtcjqrduskqrbuppbt



From max at alcyone.com  Sat Mar  3 14:22:30 2001
From: max at alcyone.com (Erik Max Francis)
Date: Sat, 03 Mar 2001 11:22:30 -0800
Subject: Missing the functional mind set
References: <mlb2at8ptrr7sh5dgq8qlk2201kquh6tfl@4ax.com>
Message-ID: <3AA14476.32679BD5@alcyone.com>

Daniel Klein wrote:
> 
> The purpose of this code is to return a list of dictionary values sorted by
> key, resulting in 'templist'.
> 
> >>> d = {3: 'd', 2: 'b', 1: 'a', 0: 'z'}
> >>> templist = []
> >>> for n in range(len(d)):
>         templist.append(d[n])

Why not just:

>>> d = {3: 'd', 2: 'b', 1: 'a', 0: 'z'}
>>> templist = d.values()
>>> templist
['d', 'b', 'a', 'z']

> My question is, is there a 'functional' (or list comprehension)
> solution to
> this? I don't seem to have a sufficient grasp of functional
> programming as yet.

I don't think so.  The sort method sorts the list in place; it returns
None.  That being said, you can get what you want in two lines:

>>> d = {3: 'd', 2: 'b', 1: 'a', 0: 'z'}
>>> result = d.values()     
>>> result.sort()
>>> result
['a', 'b', 'd', 'z']

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ One completely overcomes only what one assimilates.
\__/ Andre Gide
    The laws list / http://www.alcyone.com/max/physics/laws/
 Laws, rules, principles, effects, paradoxes, etc. in physics.


From guido at digicool.com  Fri Mar 23 17:29:02 2001
From: guido at digicool.com (Guido van Rossum)
Date: Fri, 23 Mar 2001 17:29:02 -0500
Subject: Python 2.1b2 released
Message-ID: <200103232229.RAA19777@cj20424-a.reston1.va.home.com>

On time, and with a minimum of fuss, we've released Python 2.1b2.
Thanks again to the many developers who contributed!

Check it out on the Python website:

    http://www.python.org/2.1/

or on SourceForge:

    http://sourceforge.net/project/showfiles.php?group_id=5470&release_id=28334

As it behooves a second beta release, there's no really big news since
2.1b1 was released on March 2:

- Bugs fixed and documentation added. There's now an appendix of the
  Reference Manual documenting nested scopes:

    http://python.sourceforge.net/devel-docs/ref/futures.html

- When nested scopes are enabled by "from __future__ import
  nested_scopes", this also applies to exec, eval() and execfile(),
  and into the interactive interpreter (when using -i).

- Assignment to the internal global variable __debug__ is now illegal.

- unittest.py, a unit testing framework by Steve Purcell (PyUNIT,
  inspired by JUnit), is now part of the standard library.  See the
  PyUnit webpage for documentation:

    http://pyunit.sourceforge.net/

Andrew Kuchling has written (and is continuously updating) an
extensive overview: What's New in Python 2.1:

    http://www.amk.ca/python/2.1/

See also the Release notes posted on SourceForge:

    http://sourceforge.net/project/shownotes.php?release_id=28334

We are planning to make the final release of Python 2.1 on April 13;
we may release a release candidate a week earlier.

We're also planning a bugfix release for Python 2.0, dubbed 2.0.1; we
don't have a release schedule for this yet.  We could use a volunteer
to act as the bug release manager!

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



From stevewilliams at wwc.com  Sun Mar  4 18:51:46 2001
From: stevewilliams at wwc.com (Steve Williams)
Date: Sun, 04 Mar 2001 23:51:46 GMT
Subject: A counter-proposal to __future__ in PEP 236
References: <j47l29lpo8.fsf@informatik.hu-berlin.de> <97ll2u02qmp@news2.newsguy.com> <3A9FBE5C.95615241@sewardweb.com> <97rqq209c1@news1.newsguy.com>
Message-ID: <3AA2D515.65AC8F54@wwc.com>

Alex Martelli wrote:

> "Lloyd Sommerer" <lsommerer at sewardweb.com> wrote in message
> news:3A9FBE5C.95615241 at sewardweb.com...
> [snip]
> Actually, colons are used, for readability, in various situations where
> it's quite rare for the suite to follow on the same line -- who ever
> does that with 'class Name:', for example?
>
> Alex

In 'if' statements I whisper a quiet 'then' and the fingers type ':'.  A
quiet 'is defined as' also works.




From mgjv at tradingpost.com.au  Fri Mar 16 21:32:02 2001
From: mgjv at tradingpost.com.au (Martien Verbruggen)
Date: Sat, 17 Mar 2001 13:32:02 +1100
Subject: ZPT, a next-generation template technology
References: <98ufbu$kkh$1@uranium.btinternet.com>
Message-ID: <slrn9b5j52.i70.mgjv@martien.heliotrope.home>

On Sat, 17 Mar 2001 01:43:50 -0000,
	Hamish Lawson <hamish_lawson at btinternet.com> wrote:
> 
> * Zope is a leading open-source web application server and content-
> management framework, written in Python by Digital Creations.

So why did you post it to cll.perl.misc and cl.java? At least alt.php is
slightly relevant to this, although you might piss them off as well. I
don't know what their policy on offtopic posts is.

However, cl.perl.misc gets enough offtopic crap as it is. Don't post
future announcements here.

Martien
-- 
Martien Verbruggen              | My friend has a baby. I'm writing
Interactive Media Division      | down all the noises the baby makes so
Commercial Dynamics Pty. Ltd.   | later I can ask him what he meant -
NSW, Australia                  | Steven Wright


From JamesL at Lugoj.Com  Fri Mar  9 00:41:35 2001
From: JamesL at Lugoj.Com (James Logajan)
Date: Thu, 08 Mar 2001 21:41:35 -0800
Subject: "!=" is bad form. Re: sorry....never mind
References: <DZkp6.18811$pF4.3015652@news1.telusplanet.net> <fdlp6.16982$go.2157953@news0.telusplanet.net> <3AA6E308.5B63ECC6@cosc.canterbury.ac.nz> <3AA72253.FB021815@Lugoj.Com> <slrn9aehgr.33d.tim@vegeta.ath.cx> <984047188.296438@newsmaster-04.atnet.at> <slrn9agdkv.9ms.tim@vegeta.ath.cx>
Message-ID: <3AA86D0F.E36B0BDC@Lugoj.Com>

Tim Hammerquist wrote:
> I realize I seem to be in the minority in this group when I say I like
> C/C++.  NB: I like them as _languages_, but would prefer NOT to do
> everyday programming in them, as they're just a pain in the ass most of
> the time.

Ironically, I like C and have been coding in C on and off for it seems just
about forever (well, 18 some years I think). And I find programming in C a
mostly pleasurable experience. Yet as I stated earlier, for Python "<>"
feels "righter" to me than "!=". I suspect it is a triviality for both
programmers to understand, and the language to support, both forms. While
deprecating "<>" in Python would not mean the end of civilization as we know
it, it would be a damn shame anyway.


(a > b) or (a == b)  --->  (a >= b)
(a < b) or (a == b)  --->  (a <= b)
(a < b) or (a >  b)  --->  (a <> b)


From aleaxit at yahoo.com  Sun Mar 18 05:22:17 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sun, 18 Mar 2001 11:22:17 +0100
Subject: Cameron Laird Quote on www.unixinsider.com -- Probably old news to y'all, but new to me.
References: <bJOr6.9481$TW.38047@tor-nn1.netcom.ca> <pkZs6.8451$992.49916@news-server.bigpond.net.au>
Message-ID: <99225o0205i@news2.newsguy.com>

"Neil Hodgson" <neilh at scintilla.org> wrote in message
news:pkZs6.8451$992.49916 at news-server.bigpond.net.au...
    [snip]
>    C compilation time is also close to zero for much code now. For

...but not for large programs, unless a LOT of care has been
taken in dependency management when splitting them up
into very carefully separated components.  In my experience,
such painstaking care is hardly ever exercized with total success.

And then, of course, you're hosed anyway when you need to
change a iota in a low-level headerfile and 'recompile the
world' (advanced dependency-checkers that purport to NOT
do so, by basing dependency on finer grain than files, are not
reliable today in my experience -- they end up making things
worse, not better -- but I'll admit I haven't tried IBM's
VisualAge C++ in that regard).


>    Modern C++ that uses the standard C++ library and so requires a lot of
> template expansion may slow down compilation to half a second but that is
> still quite competitive with Python.

That depends on the compiler and build-system; we've had
several recent reports on this group of (gcc?) users very happy
with Boost Python _except_ for the little detail that compilation
of a very simple component can take several minutes and over
100 MB of memory (I don't see that problem with Microsoft VC++,
but it has other issues, such as a notoriously unreliable dependency
manager -- I've lost track of the times one gives up and hits
'rebuild all' because the incremental only-at-need builds are
not working...).


>    This is one benefit of the 'scripting languages' that is rapidly
> disappearing as machines get faster and C++ compilers are optimised.

I just wish it was this way -- since my mostly-Python apps will just
about always have SOME lower-level for-speed component (and I
far prefer to code those in C++ rather than in C).  Unfortunately,
while the trend is indeed as you mention, the progress along that
street is slower than the growth of program size and program use
of templates & such...!-)


Alex





From alf at leo.logilab.fr  Thu Mar 15 04:05:13 2001
From: alf at leo.logilab.fr (Alexandre Fayolle)
Date: Thu, 15 Mar 2001 09:05:13 +0000 (UTC)
Subject: Regular Expressions and Python 1.5.2
Message-ID: <98q0k8$2b77$1@norfair.nerim.net>

Is this a known bug in the module re of python 1.5.2, or am I missing 
something obvious?

>>> import re
>>> t = "m&n"
>>> re.sub('[^/_ -]','*',t)
'***'
>>> re.sub('[^_ -/]','*',t)
'*&*'

TIA

Alexandre Fayolle
-- 
http://www.logilab.com 
Narval is the first software agent available as free software (GPL).
LOGILAB, Paris (France).


From loewis at informatik.hu-berlin.de  Wed Mar  7 08:34:14 2001
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 07 Mar 2001 14:34:14 +0100
Subject: How to Xml escaping of characters above 7 bit ascii.
References: <N6pp6.711$tR1.9978@news1.oke.nextra.no>
Message-ID: <j4vgpln1nt.fsf@informatik.hu-berlin.de>

"Syver Enstad" <syver at NOSPAMcyberwatcher.com> writes:

> Is there a method in Python that takes a unicode string and escapes it
> according to the xml file format? Or will I have to do it myself?

You could write a codec yourself; several people have already written
code to do so (see PyXML for examples). Standard Python does not
provide such a codec, so you have to write it yourself.

Regards,
Martin



From m.linnemann at addcom.de  Wed Mar 28 08:54:07 2001
From: m.linnemann at addcom.de (Michael Linnemann)
Date: Wed, 28 Mar 2001 15:54:07 +0200
Subject: Python scripts as Drop targets in Windows Explorer
References: <99qogo$o6g$1@troll.powertech.no> <99qqq3$qt7$0@216.39.170.247> <99r043$qqr$1@troll.powertech.no>
Message-ID: <vdqs99.p16.ln@haegar.fqdn.th-h.de>

Am Tue, 27 Mar 2001 23:16:44 +0200 schrieb Syver Enstad:

>That's what I did, but the default on Win2k is to save as unicode. I guess
>you're using NT or Win9x?

>begin 666 python filedrop.reg
>M4D5'141)5#0-"@T*6TA+15E?0TQ!4U-%4U]23T]47%!Y=&AO;BY&:6QE7'-H
>M96QL97A<1')O<$AA;F1L97)=#0I /2)[.#9#.#8W,C M-#)!,"TQ,#8Y+4$R
>513 at M,#@P,#)",S S,#E$?2(-"@T*

Yeah, looks cool. What is it, perl? Czech? EBCDIC?

Come on, tune your W2k a little and make it talk plain :-)

Regards
Michael


From mcherm at destiny.com  Sat Mar  3 01:29:05 2001
From: mcherm at destiny.com (Michael Chermside)
Date: Sat, 03 Mar 2001 00:29:05 -0600
Subject: A counter-proposal to __future__ in PEP 236
Message-ID: <3AA08F31.55CB6ADB@destiny.com>

On the topic of 
    from __future__ import foo
versus
    directive foo

I'm hearing two different messages here. On the one hand, I hear Martin saying
that the from __future__ syntax is ugly and misleading. On the other hand
I hear Tim saying that this is an incompatible language change, and requires
a different mechanism than what #pragma would offer. (Yes, I know other
people spoke too... I'm using Martin and Tim as representatives.)

So let me chime in.

On the first point, I have to agree with Martin. "from __future__ import foo"
is the sort of horrible ugliness which should never be seen outside of C++.
It is *NOT* an actual import statement, and it does *NOT* scream "black-magic"
-- just looks like another of those special "built-in" things; you know...
the functions __cmp__ and __import__ and the modules __builtin__ and 
__future__. But that's NOT what's happening here... we're doing something
VERY drastic (changing the language grammer) and the use of a special keyword,
like directive, may help signal that.

On the other hand, let's consider the second argument: don't get this mixed
up with pragmas. I realize that #pragma has been used for many different
purposes throughout it's colorful and inglorious history, but in MY mind
(you're free to disagree) the REAL purpose is "hints" like optimize-for-space
or inline-this-function, or I-won't-be-using-exec-so-don't-bother-including-
the-parser-when-you-compile. The thing that distinguishes these is that
they do NOT affect the behavior of the language... a "valid" program will
produce the same results whether the pragma is obeyed or ignored. In fact,
that's where the idea came from originally of putting it in comments...
because it didn't really help define the program, just provided pragmatic
hints for the compiler/interpreter.

Well, if THAT'S what we mean, then Martin's directive syntax will never do.
First and foremost, he requires the line to come at the beginning of the
file... which unreasonably restricts what you could do with it. If you wanted
to have favor_time and favor_space as options, you'd want to be able to turn
them on and off with finer granularity [debatable point]. In fact, the only
things that the directive keyword would really be useful for (in it's proposed 
form) are ones which affect an ENTIRE file... such as changes to the language
grammer (including keywords), or to the source syntax (eg latin-1). Including
pragmas is a separate issue.

So I'm coming out strongly in favor (+0.8) of the directive syntax. But in
doing so, I'd better point out the one place where I think the "from
__future__ import" syntax is better. Through the creation of an (empty)
"__future__" module, it becomes possible to write code that compiles and and
runs fine in pre-2.1 versions of python, which would choke up on encountering
the "directive" statement. But why? After all, if it USES the new feature
(nested scopes in this case, but it could be ANY new feature), then it
obviously isn't going to run in the pre-2.1 version. And if it doesn't use
the feature, then just don't put anything in! All post 2.0 versions of 
python could choose to fail when reading input with a directive statement
that they don't recognize (or maybe someone has a better behavior to suggest?).


How's-THAT-for-a-long-winded-and-strongly-opinionated-delurk-ly y'rs

-- Michael Chermside
   (who's too opinionated to be a good lurker anyhow)



From new_name at mit.edu  Thu Mar  1 19:33:02 2001
From: new_name at mit.edu (Alex)
Date: 01 Mar 2001 19:33:02 -0500
Subject: compile() source code from a string
References: <3a9ee600.391990@News.CIS.DFN.DE>
Message-ID: <etdd7c1c8n5.fsf@pickled-herring.mit.edu>

>>> print compile.__doc__
compile(source, filename, mode) -> code object

Compile the source string (a Python module, statement or expression)
into a code object that can be executed by the exec statement or eval().
The filename will be used for run-time error messages.
The mode must be 'exec' to compile a module, 'single' to compile a
single (interactive) statement, or 'eval' to compile an expression.
>>> 

So, you want the 'exec' option.

Alex.



From paulp at ActiveState.com  Fri Mar 23 17:27:42 2001
From: paulp at ActiveState.com (Paul Prescod)
Date: Fri, 23 Mar 2001 14:27:42 -0800
Subject: static methods
References: <Pine.LNX.4.21.0103231640070.5553-100000@clarkevans.com>
Message-ID: <3ABBCDDE.546FEB5A@ActiveState.com>

"Clark C. Evans" wrote:
> 
> I was wondering if there is any rationale as to why
> Python classes do not have static methods?

Probably because they are simply not neccessary. 

def _static():
	XXX.foo=bar
	return XXX.foo

class Class:
	staticmethod = _static
	...

I might like a nicer syntax but I also like the fact that Python is
syntactically small...

-- 
Take a recipe. Leave a recipe.  
Python Cookbook!  http://www.activestate.com/pythoncookbook



From grante at visi.com  Tue Mar 13 18:21:21 2001
From: grante at visi.com (Grant Edwards)
Date: Tue, 13 Mar 2001 23:21:21 GMT
Subject: This math scares me
References: <mailman.984523936.12653.python-list@python.org>
Message-ID: <RXxr6.1244$Tg.169940@ruti.visi.com>

In article <mailman.984523936.12653.python-list at python.org>, Steven D. Majewski wrote:

>> I don't know if you can credit Gates with doing anything other
>> than following the pack.  Back when I first started doing
>> software (CP/M 1.7), all language packages for "personal"
>> computers came with both BCD and binary FP math libraries so
>> you could do exact monetary calculations if you wanted.
>
>Exactly what pack do you believe Gates was following ? 
>
>You do know (don't you?) that Bill's BASIC predates not just
>MS-DOS (which microsoft bought from another company) but CP/M
>and just about any other "PC" software (and most of the hardware)
>since the first version was written to run on the Altair computer --
>which was the first microprocessor based personal computer. 

You're right.  I had forgotten about the MITS Altair version
and was thinking of the Apple II and IBM-PC version of MS-BASIC
(I never ran into MS products until the IBM-PC came out).

-- 
Grant Edwards                   grante             Yow!  I represent a
                                  at               sardine!!
                               visi.com            


From ben.hutchings at roundpoint.com  Fri Mar  9 03:47:38 2001
From: ben.hutchings at roundpoint.com (Ben Hutchings)
Date: 09 Mar 2001 00:47:38 -0800
Subject: "Wrap" a python-cgi
References: <3aa7c5f9$0$21482@SSP1NO25.highway.telekom.at> <mailman.984078311.6628.python-list@python.org>
Message-ID: <u8zmf8h1x.fsf@roundpoint.com>

Kalle Svensson <kalle at gnupung.net> writes:
<snip>
> if __name__ == "__main__":
>     print "Content-Type: text/html\n"
>     try:
>         main()
>     except:
>         import traceback, sys
> 	print "<h1>Error!</h1>\n<pre>"
> 	traceback.print_exc(file=sys.stdout)
> 	print "</pre>"

Eek!  There's a cross-site scripting attack in the making!

Instead of printing the traceback directly to standard output, print
it to a string-file, HTML-escape the string and then print that.

-- 
Any opinions expressed are my own and not necessarily those of Roundpoint.


From not.this at seebelow.org  Sat Mar 17 00:45:51 2001
From: not.this at seebelow.org (Grant Griffin)
Date: Fri, 16 Mar 2001 23:45:51 -0600
Subject: Thoughts about extensions to the Python language
References: <mailman.984608060.6280.python-list@python.org>
Message-ID: <3AB2FA0F.E61862B0@seebelow.org>

Carlos Ribeiro wrote:
> 
> (flames off please ...)

If you want flames, you came to the wrong place.  (If you don't mind me
saying so. <wink>)

...
> What attracted me to Python was the simplicity and cleanliness of the
> language. Code written in Python is as clean as code can be.

Oh, except list comprehensions.  (Yup, I use 'em--but I always feel
guilty afterwards. <wink>)

> I have written
> systems using several other languages, including C, C++ and Perl, and none
> of these languages can even come close of Python in terms of readability.

Still, one can write FORTRAN in any language.  (But it takes so much
work to write FORTRAN in Python that lazy folks don't bother.)

...

> 6) Most important of all, it's clear that Guido had used a 'common sense'
> approach to design the original language.

Are you sure?  It appears to me that the strength of his sense is its
very _un_common-ness <wink>.  Anyway, I guess we just have to trust him
(at least until he leads us wrong--then we'll take him to meet Jimmy
Hoffa <wink>.)

As to the PEP process, it seems pretty good to me, except that it
sometimes gives the impression (perhaps mistaken) of "design by
committee".  In other words, new things seem to happen because somebody
thought of a good idea, and Guido agreed.  Now that's not all bad (after
all, it _was_ a good idea), but ideally, there would be some larger
vision for the future of the language.  (Guido has probably spelled that
out somewhere, but I guess I missed it.  Sling me a link, /F. <wink>)

Now, the usual manifestation of "growth" in software is new features
(often ones that merely clutter the forest with more trees), but Python
has painted itself into a corner in terms of its major feature being its
lack of features.  We love it for that, but that works against
"growth".  (Contrariwise, Perl has painted itself into the corner of
already having so many kitchen sinks that it's hard to find yet another
one that needs throwing in.)

So where does language growth come from?  Modules, Benjamin.  Modules. 
And documentation, Benamin.  Documentation.  I guess I'd like to see
more stuff added to the standard distribution.  For example, I think if
Numeric were in the standard distribution (or maybe in an alternate
standard distribution), that might make it easier for me to sell snake
oil to my friends.  (When I tell 'em they've got to download and install
six packages to run my software, they roll their eyes--or nod off.)

Oh, and the fact that Python doesn't have any single, compelling
data-plotting system (rather than several not-very-compelling ones)
makes it hard to sell as an alternative to Matlab.  Numeric users find
themselves pretty blind without that.

And from another angle, if Python ever shed its dynamic skin and came in
a static/compilable version (let's call the new one...oh, lemme
see..."P3K"), that would be a *big* step towards growth: scripting
languages are great, but there are very good reasons most of us still
use statically typed languages such as C/C++ for most of our work--and
will do so into the indefinite future.

embedded-real-time-ly y'rs

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation            http://www.iowegian.com


From lluang at northwestern.edu  Tue Mar 13 23:18:28 2001
From: lluang at northwestern.edu (Louis Luangkesorn)
Date: Tue, 13 Mar 2001 22:18:28 -0600
Subject: cursor.execute fails under mxODBC
References: <3AABF77F.7B77D51@northwestern.edu> <3AAE59AE.630C148C@lemburg.com>
Message-ID: <3AAEF113.E928A434@northwestern.edu>

[and I'm adding it back into the newsgroup comp.lang.python: Louis]
"M.-A. Lemburg" wrote:
[CCed to the Python newsgroup, since I believe this is of general
 interest; hope you don't mind.]

> Thank you for forwarding the messages. I have had numerous questions
> about this problem and will look into it later this week (just came
> back from the Python conference and am still fighting the jet lag).
>
> One thing I've noted about the problem reports is that all seem
> to use ActiveState Python 2.0. There may be an issue here, since
> the mxODBC (alas the egenix-commercial) package was compiled against
> the BeOpen Python 2.0 version which is available from python.org.
>
> Could you perhaps check whether you experience the same problems
> using BeOpen Python 2.0 ?
>
> Thanks,
> Marc.

Marc

No, that did not work. Here it is under the BeOpen-Python-2.0  After the
cursor.execute(...) the Python shell hangs.

Louis
---begin output---
Python 2.0 (#8, Oct 16 2000, 17:27:58) [MSC 32 bit (Intel)] on win32
Type "copyright", "credits" or "license" for more information.
IDLE 0.6 -- press F1 for help
>>> import mx.ODBC.Windows
>>> channel = mx.ODBC.Windows.Connect('sqltutor')

>>> cursor = channel.cursor()
>>> cursor.execute('select FirstName, LastName, Address, City, State FROM
EmployeeAddressTable')

---end output---
--
K Louis Luangkesorn
lluang at northwestern.edu  http://pubweb.nwu.edu/~kll560  PGP:0xF3E2D362

Whatsoever things are true, ... honest, ... just, ... pure, ... lovely, ... of
good report; if there be any virtue, and if there be any praise, think on
these things.
     - motto - Northwestern University (Phil 4:8)




From norman at earth.ox.ac.uk  Fri Mar 30 09:46:51 2001
From: norman at earth.ox.ac.uk (Norman Charnley)
Date: Fri, 30 Mar 2001 15:46:51 +0100
Subject: BLT extensions with BeOpen Python 2.0
References: <Pine.GSO.4.21.0103291300160.11408-100000@babbage>
Message-ID: <3AC49C5B.23F5F95@earth.ox.ac.uk>

Martyn Quick wrote:
> 
> I've been trying to install the BLT extensions for Tk into my BeOpen
> distribution of Python 2.0.  However, it seems that the directory
> structure that the BLT distribution expects (see www.tcltk.com/blt) is
> different to that provided by the python distribution.

I had a similar problem (trying to see BLT from Pmw under Win98,
BeOpen2.0 Python), and a solution provided by Gordon Lightfoot fixed it
for me:

install the BLT 2.4 directory (with the pkgIndex.tcl file) into the
\Python20\tcl\tcl8.3 tree (so that the path to the pkgIndex.tcl looks
like
    \Python20\tcl\tcl8.3\blt2.4

Good Luck,

Norman

 =================================================
 Dr. Norman Charnley
 Department of Earth Sciences
 University of Oxford
 Oxford OX1 3PR, UK.
 
 Telephone:
 +44 1865 272053 (Office)
 +44 1865 272009/283741 (SEM/Electron Microprobe Labs)
 +44 1865 282131 (Microsims Ion Probe Lab.)
 
 Fax: +44 1865 272072
==================================================


From slhath at home.com  Mon Mar 12 21:26:54 2001
From: slhath at home.com (Scott Hathaway)
Date: Tue, 13 Mar 2001 02:26:54 GMT
Subject: sql2xml available
Message-ID: <Ozfr6.162210$bb.1564026@news1.rdc1.tx.home.com>

Hello Everyone,

I have created a simple utility called sql2xml that takes an SQL call and
returns the resultset as xml.  In the support file (db.py), there are
functions to pull out the field names from the SQL, or you can pass your own
if you want to use friendlier names than those in your database.  This
utility is in the public domain.  I hope you find it useful.  (To use the
db.py file, you must have mxODBC, but changing to another database module
should be very easy since the function just needs the resultset in the
standard python db api format).  The utility is just a work in progress and
your comments/suggestions are needed and welcome.

http://members.home.net/slhath/downloads.html

Also, Dr. David Mertz has a complimentary utility called sql2dtd which takes
an SQL call and provides an XML DTD from it.  The two tools will work
together in the near future (they are very close to doing so right now).
You can get David's tool at:

  http://gnosis.cx/download/sql2dtd.py

Thanks,
Scott Hathaway




From aahz at panix.com  Tue Mar 20 10:32:51 2001
From: aahz at panix.com (Aahz Maruch)
Date: 20 Mar 2001 07:32:51 -0800
Subject: relational/boolean operator overloading -> PEP 207
References: <98tjq6$qre$1@news.flashnet.it> <994h71$ohq$1@news.flashnet.it> <996r1u$67f$1@panix6.panix.com> <9977ap$l05$1@news.flashnet.it>
Message-ID: <997t73$p7i$1@panix3.panix.com>

In article <9977ap$l05$1 at news.flashnet.it>,
Renzo Tomaselli <renzo.tomaselli at tecnotp.it> wrote:
>
>Nevertheless, even when forgetting about my topic for a while, I still
>cannot understand how PEP 207 will handle 'and/or' operators applied to
>"generic" objects which the new relational overloadings could return
>anytime. E.g. how an expression such as "obj1 and obj2" will be evaluated by
>Python 2.1.

This appears to be the appropriate section from the PEP:

    5 In the current proposal, when A<B returns an array of
      elementwise comparisons, this outcome is considered non-Boolean,
      and its interpretation as Boolean by the shortcut operators
      raises an exception.  David Ascher's proposal tries to deal
      with this; I don't think this is worth the additional complexity
      in the code generator.  Instead of A<B<C, you can write
      (A<B)&(B<C).

If you want a better answer, you might try posting to
python-dev at python.org
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

There's a difference between a person who gets shit zie doesn't deserve
and a person who gets more shit than zie deserves.  --Aahz


From digitig at cix.co.uk  Wed Mar 14 19:42:00 2001
From: digitig at cix.co.uk (Tim Rowe)
Date: Thu, 15 Mar 2001 00:42 +0000 (GMT Standard Time)
Subject: Einstein's Riddle
References: <98glrd$1si1a$1@ID-75892.news.dfncis.de>
Message-ID: <memo.20010315004232.1527D@digitig.compulink.co.uk>

In article <98glrd$1si1a$1 at ID-75892.news.dfncis.de>, 
no-spam-popov at bigserver.com (Boris Popov) wrote:

<snip>

> house         1 2 3 4 5
> color            yellow blue red green white
> nationality   norweigian dane brit german swede
> drink            water tea milk coffee beer
> smoke        dunhill blend pall mall prince bluemaster
> pet              cats horse birds fish dogs

Nope. Because each person keeps a certain pet (singular), but the person 
who smokes Pall Mall rears birds (plural). So they can't be his pets, 
presumably he does it professionally (poultry farmer?) 

And you also assumed the first house is the one on the left.

Formal notations are the only way to go :-)

BTW, did Einstein mean that only 2% of the population could be /bothered/ 
to solve it?


From loriendNOSPAM at bigpond.com  Fri Mar 16 21:29:50 2001
From: loriendNOSPAM at bigpond.com (Lorien Dunn)
Date: Sat, 17 Mar 2001 13:29:50 +1100
Subject: Meta-class inheritance problem (Python2.0 bug?)
Message-ID: <6Kzs6.35062$v5.73895@newsfeeds.bigpond.com>

Hullo all,
        I've been using boost::python for a little while now- long enough to have 
seemingly successfully wrapped my extension, and I've run into a bug in 
python 2.0. The real code that the following pseudo code represents throws 
a TypeError. 

class MyDerivedClass(MyPurePythonClass, MyBoostClass): 
        def __init__(self):
                MyPurePythonClass.__init__(self)
                MyBoostClass.__init__(self)

The problem occurs in MyPurePythonClass- Python says "unbound method must 
be called with class instance 1st argument". This makes sense: I'm passing 
a meta-class instead of a class. David Abrahams (the main author of 
boost::python) agrees that this is a Python problem, and that it should 
affect Zope as well.

He also said that it should be an easy modification, but he doesn't know 
what ramifications it would have throughout the rest of the python 
interpreter.

Advice anyone?

Thanks,
Lorien Dunn
                


From tim_one at email.msn.com  Sun Mar 18 16:04:06 2001
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 18 Mar 2001 16:04:06 -0500
Subject: Why "from __future__" stinks; a counter-offer
In-Reply-To: <AD70ED184AFED311ADFD009027AC917F83617A@dmc1nt.zoho.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCMEICJGAA.tim_one@email.msn.com>

[Jason Orendorff]
> Okay, nested scopes are a good idea.  Having the feature
> "disabled by default" in 2.1 is a good idea.  But this
> "from __future__" stuff isn't so hot.
>
> First, it's ugly.

Good -- then it will stick out <0.5 wink>.  Double double-underscore names
are intended to "stick out", btw, and future_statements are intended to be
temporary.  In that sense, the uglier they are, the more incentive to get rid
of them ASAP.

> Second, owing mostly to its ugliness, it will be
> *really* embarrassing to try and explain this to my
> (non-Python-bitten) friends.

If they don't use Python, I don't know why you'd bother to mention this to
them regardless of spelling.

> Third, it's extremely silly to have a module named
> "__future__" that describes only the present and the past.

At the moment, __future__.py contains only one entry, which refers to 2.2
final.  Sounds like the future to me.

> Fourth, Python 2.0 and earlier report an
>     ImportError: No module named __future__
> when they encounter a future_statement.  Surely a
> compiler error, even a SyntaxError, would be more
> appropriate and less confusing.

I agree that would be better.  Over time, though, it becomes less
interesting, as people move to more recent releases (which will raise
exceptions at compile-time for future_statements they can't satisfy).

> Lastly and most importantly, it's not too late to do
> something saner.

Unclear, although in the absence of a competing PEP it's certain to remain as
it is, and the last 2.1 beta release is scheduled for this coming Friday (as
noted in PEP 226).

> I agree it's desirable to have the code itself advertise that it
> uses a certain nascent feature.

Since syntax and/or semantics change in the presence of a future_statement,
it's *mandatory* that modules desiring the new behavior explicitly say so --
"desirable" has nothing to do with it.

> So, what if we do something like this:
>
>   1. Pick a name.  Suggestions follow:
>          require
>          using
>          pragma
>          this_code_uses
>          feature
> ...

Did you read Martin's messages on this?  He picked "directive", and submitted
a patch to implement it (or some part of it).  But he didn't write a PEP, so
you'll have to search c.l.py and SourceForge for the relevant bits.  He had
in mind a more general facility, and in particular wants to specify
source-file character encodings someday; I actively want not to conflate
future_statements with "general directives" (although "directive future xxx"
might be OK -- alas, without a PEP, it's a non-starter).

BTW, from the names you suggested, I have to guess that you think a
future_statement is for an "optional feature".  It is not:  it's for an
incompatible language change that will become *mandatory* in a future release
(and there's no going back then).  future_statements are a temporary aid for
use during a transition period.  "require", "using", "this_code_uses" and
"feature" all send a wrong message in that respect.  "pragma" is
inappropriate because future_statements are about syntax and semantics, not
pragmatics.

patiently-awaiting-the-end-of-the-world-ly y'rs  - tim




From kalle at gnupung.net  Tue Mar 27 08:28:44 2001
From: kalle at gnupung.net (Kalle Svensson)
Date: Tue, 27 Mar 2001 15:28:44 +0200
Subject: Netscape and Python
In-Reply-To: <13EE655665F4D311AB4D0008C789498A01EE2CA5@zei02exm02.cork.cig.mot.com>; from corour01@motorola.com on Tue, Mar 27, 2001 at 02:15:31PM +0100
References: <13EE655665F4D311AB4D0008C789498A01EE2CA5@zei02exm02.cork.cig.mot.com>
Message-ID: <20010327152844.A17383@father>

Sez O'Rourke Clodagh-corour01:
> Hi all, 
> 
> Does anyone know how to launch Netscape from Python?

Use the webbrowser module in 2.0, otherwise look at os.system, os.exec*
os.spawn*.

Peace,
  Kalle
-- 
Email: kalle at gnupung.net     | You can tune a filesystem, but you
Web: http://www.gnupung.net/ | can't tune a fish. -- man tunefs(8)
PGP fingerprint: 0C56 B171 8159 327F 1824 F5DE 74D7 80D7 BF3B B1DD
 [ Not signed due to lossage.  Blame Microsoft Outlook Express. ]



From sr63085 at aol.com  Sat Mar  3 23:14:46 2001
From: sr63085 at aol.com (Sr63085)
Date: 04 Mar 2001 04:14:46 GMT
Subject: Clearing Screen
Message-ID: <20010303231446.29909.00000348@ng-fn1.aol.com>

Is there a way to clear all text from the screen in a Python Script
window?(similar to clreol() with conio.h in C/C++)
Thanks in advance.


From aleaxit at yahoo.com  Fri Mar  9 08:57:34 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 9 Mar 2001 14:57:34 +0100
Subject: UserDict and items()
References: <98656v$cud$1@news.rrz.Uni-Koeln.DE> <slrn9aed8i.cg1.scarblac@pino.selwerd.nl>
Message-ID: <98anga01trm@news2.newsguy.com>

"Remco Gerlich" <scarblac at pino.selwerd.nl> wrote in message
news:slrn9aed8i.cg1.scarblac at pino.selwerd.nl...
    [snip]
> The cuteness introduced in later Python versions would allow that to be:
>
>    def __getitem__(self, key):
>       return self.data.setdefault(key, DatabaseCall(key))
>
> Ie, look up everything in the database that's not in the dictionary yet,
put
> it in, and return what's in the dict.

Unfortunately, this approach is not lazy -- DatabaseCall is here
being called also for keys which ARE already among those of self.data.
A more explicit expression is needed to allow the laziness.


Alex





From bsb at winnegan.de  Wed Mar 21 15:08:18 2001
From: bsb at winnegan.de (Siggy Brentrup)
Date: 21 Mar 2001 21:08:18 +0100
Subject: Substitute for join()
In-Reply-To: <Mo7u6.58$Hp.2896@typhoon.aracnet.com>
References: <Mo7u6.58$Hp.2896@typhoon.aracnet.com>
Message-ID: <878zlylwb1.fsf@winnegan.de>

"Daniel Klein" <DanielK at jBASE.com> writes:

> I've got to make a script (written in version 2.0) to work on a machine
> running 1.5.2. The only incompatible line is one that uses the
> "".join(mylist) method. To 'get out of the woods' quickly, I substituted it
> with
> 
> s = ''
> for element in mylist: s = s + element
> 
> Is this the most efficient way to code this?

Nope, use

import string
string.join(mylist,'')

HIH
  Siggy
-- 
Siggy Brentrup - bsb at winnegan.de - http://www.winnegan.de/
****** ceterum censeo javascriptum esse restrictam *******




From sholden at holdenweb.com  Mon Mar  5 11:35:49 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Mon, 5 Mar 2001 11:35:49 -0500
Subject: Can SAX parser return <!-- COMMENTS -->  ???
References: <MWxn6.15038$x27.906254@news1.rdc2.on.home.com> <j4d7c0jh03.fsf@informatik.hu-berlin.de> <wzwva5shuy.fsf@sunshine.cs.uu.nl> <j48zmko6nl.fsf@informatik.hu-berlin.de>
Message-ID: <rjPo6.21305$1D5.840563@e420r-atl1.usenetserver.com>

"Martin von Loewis" <loewis at informatik.hu-berlin.de> wrote in message
news:j48zmko6nl.fsf at informatik.hu-berlin.de...
> Piet van Oostrum <piet at cs.uu.nl> writes:
>
> > MvL> It is a matter of ongoing debate whether applications should even
deal
> > MvL> with comments; if an application needs XML comments for proper
> > MvL> operation, something is wrong with that application.
> >
> > What about a prettyprinter? Doesn't that count as an application?
>
> Sure. But why does it have to preserve comments?
>
I would say "because that particular application is using XML as a tool for
human communication" -- obviously no XML application in its right mind would
require input to be formatted. The comments, being primarily for humans,
should therefore be preserved.

begs-the-question-of-how,-of-course-ly y'rs  - steve




From grey at despair.rpglink.com  Thu Mar 22 08:12:11 2001
From: grey at despair.rpglink.com (Steve Lamb)
Date: Thu, 22 Mar 2001 13:12:11 GMT
Subject: Please rate these Python books
References: <20010322.1445209@mis.configured.host> <_Meu6.8736$8O6.1818817@typhoon.austin.rr.com>
Message-ID: <slrn9bjuha.4qv.grey@teleute.dmiyu.org>

On Thu, 22 Mar 2001 03:59:54 GMT, alan runyan <runyaga at thisbox.com> wrote:
>how are you approaching python?  what is your level of experience w/ python?
>what kinda books do you like? reference, explanatory, or 'by example' ?  My
>favorite is Essential Reference, David Beazley did a incredible job.

    No disrespect towards Beazley, he's a large reason why I work in Python as
much as I can now but his book semmed to be pretty much a dump of the
postscript of the website to a printer.  Not that that is a bad thing as it is
the /only/ Python book I own and one I recommend to anyone who has passing
familiarity with other scripting languages (Perl comes to mind).  You're
correct in stating it is sparse on the examples but it is a rare case where
I've really needed the examples.  Python being as interactive as it is I tend
to learn by firing up Python in a separate window and then playing with the
syntax until it works out in that session before plopping it into my script.

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


From jafo at tummy.com  Mon Mar 19 21:13:07 2001
From: jafo at tummy.com (Sean Reifschneider)
Date: Mon, 19 Mar 2001 19:13:07 -0700
Subject: Serial Port on Unix
In-Reply-To: <3AB637A6.FDEAE8FF@nipltd.com>; from chrisw@nipltd.com on Mon, Mar 19, 2001 at 04:45:26PM +0000
References: <3AB637A6.FDEAE8FF@nipltd.com>
Message-ID: <20010319191307.A1093@tummy.com>

On Mon, Mar 19, 2001 at 04:45:26PM +0000, Chris Withers wrote:
>Anyway, I had a hunt on python.org without much luck. A few obscure postings and

Check on ftp.tummy.com:/pub/tummy including cid and the power* thing --
they both do serial port interaction.

Sean
-- 
 If the code and the comments disagree, then both are probably wrong. 
                 -- Norm Schryer
Sean Reifschneider, Inimitably Superfluous <jafo at tummy.com>
tummy.com - Linux Consulting since 1995. Qmail, KRUD, Firewalls, Python



From amk at mira.erols.com  Sun Mar 11 22:38:03 2001
From: amk at mira.erols.com (A.M. Kuchling)
Date: 12 Mar 2001 03:38:03 GMT
Subject: Going ZODB
References: <mailman.984324988.25505.python-list@python.org> <m3wv9w3x6h.fsf@gnu.org>
Message-ID: <slrn9aoi0i.rt.amk@mira.erols.com>

On 11 Mar 2001 12:47:18 -0700, John Wiegley <johnw at gnu.org> wrote:
>I tried to use ZODB as the persistence engine for my MOO clone
>(http://doop.sourceforge.net).  I found it's restrictions too onerous,
>most especially the restriction that I not override __getattr__ or
>__setattr__.  Lastly, I found that debugging problems was hideously

Current versions of the ZODB *do* support __getattr__ and __setattr__,
so you may want to take another look at the ZODB.  

--amk


From alex at server01.shindich.com  Mon Mar 26 01:33:14 2001
From: alex at server01.shindich.com (Alex Shindich)
Date: Sun, 25 Mar 2001 22:33:14 -0800 (PST)
Subject: Yet Another PEP:  Interface Adapter Mechanism __adapt__
In-Reply-To: <Pine.LNX.4.21.0103252208070.10200-100000@clarkevans.com>
Message-ID: <Pine.LNX.4.21.0103252141090.9359-100000@server01.shindich.com>

On Sun, 25 Mar 2001, Clark C. Evans wrote:

> On Mon, 26 Mar 2001, Alex Shindich wrote:
> > It sounds like your PEP suggests adding Strategy pattern
> > behavior to all Python objects.
> 
> I'm suggesting a bi-directional adapter pattern.  IMHO,
> It has nothing to do with the strategy pattern.   Did you
> read the proposal?
I have to confess... I only read the beginning... OK! Shoot me in the head
now!
>Evidently my explanation failed, hunh?
Let's see... The abstract section of the PEP says: "This proposal allows 
a standard and extensible method to ask two types of questions:  Is a
particular object compliant with a given protocol?" The following
describes the strategy pattern: "The Strategy pattern provides a means of
parameterizing a component to accept an algorithm.  This keeps the
component simpler, and lets clients pick and choose an algorithm
specifically tailored to their needs." See any similarities?
> 
> > Another way of looking at your PEP is to say that you are 
> > asking for Python Java-like interfaces.
> 
> No, actually there are other PEPs that do that.
> 
That is true, there is PEP-245. But in some sense, an ability to determine
"is-a" relationship is no different from asking "do you support
protocol Blah?" 

Other comments:
1. You are making an assumption that each class will provide adaptors to
some set of protocols, thus entangling the class and available set of
protocol conversions. Why can't protocol conversion live outside of the
class?
2. Is adaption symmetric? If class A provides an adaptor for class B,
mustn't class B provide an adaptor for class A?
 
3. Versioning is not mentioned anywhere in your PEP. Once a class defines
an adaptor for a protocol, the protocol better not change. Otherwise every
adaptor needs to change to match the new definition of the
protocol. 
 
That's all I have got so far... I'll send you more once I read more of the
PEP. ;))

-- 
Alex Shindich
mailto:alex at shindich.com
Visit http://www.shindich.com/




From alexander.stirzel at daimlerchrysler.com  Tue Mar 13 06:36:50 2001
From: alexander.stirzel at daimlerchrysler.com (Alexander Stirzel)
Date: Tue, 13 Mar 2001 12:36:50 +0100
Subject: What's wrong with me/Python ?
Message-ID: <3AAE0652.D68F253E@daimlerchrysler.com>

Hi!

I just wrote a small script for killing some hanging netscape processes,
the source is listed below.
Now if I run this script I get this error message:

8<----------- error message -----------------------------
Traceback (most recent call last):
  File "/usr/sbin/netscape-killer.py", line 11, in ?
    splitted = string.split(line)
NameError: There is no variable named 'string'
8<----------- error message -----------------------------

Now here's the script:

8<------------ script -----------------------------------
#!/usr/local/bin/python
import os
import time

while 1:
  time.sleep(600)
  f = os.popen("ps aux")
  lines = f.read()
  f.close()
  for line in lines:
    splitted = string.split(line)
    if (string.find('netscape',splitted.pop()))-1:
      cputime = splitted.pop()
      cputime = string.replace(cputime, ":", " ")
      cputime = string.split(cputime)
      cputime.reverse
      cputime = cputime.pop()
      if (string.atoi(cputime)>15):
        splitted.reverse
        pid = splitted.pop()
        pid = splitted.pop()
        os.kill(pid,9)
8<------------ script -----------------------------------

What did I do wrong? And why does Python want a "variable named
'string'" here?

Thanks in advance

-- 
Alexander Stirzel 
E-Mail: alexander.stirzel at informatik.uni-ulm.de


From grante at visi.com  Thu Mar 15 13:04:24 2001
From: grante at visi.com (Grant Edwards)
Date: Thu, 15 Mar 2001 18:04:24 GMT
Subject: This math scares me
References: <3aad4062.1290021@News.CIS.DFN.DE> <20010312170356.B3513@better.net> <mailman.984480856.18097.python-list@python.org> <984486395.407379@newsmaster-04.atnet.at> <slrn9asf1d.ja3.qrczak@qrnik.zagroda> <984518199.127140@newsmaster-04.atnet.at> <jwbnews-09D75F.13313313032001@corp.supernews.com> <98ohuc$du9$1@news.udel.edu> <7bQr6.644$lz1.17686@ruti.visi.com> <98oku1$f6a$1@news.udel.edu> <erRr6.659$lz1.20662@ruti.visi.com> <98qv55$f5d$1@news.udel.edu>
Message-ID: <Iu7s6.897$lz1.80703@ruti.visi.com>

In article <98qv55$f5d$1 at news.udel.edu>, Terry Reedy wrote:

>So, .1% gets represented as the binary equivalent of, perhaps,
>.0010000000000000001.
>So the interest on 2,000,000,000 pennies is calculated as
>2000000.0000000002 instead of
>2000000 pennies.  When rounded off, they are the same.  To me, this is, as
>I claimed. 'well enough' (for many purposes, at least), even though not
>'exact'.  

If that's good enough for you, then go ahead and use FP.

But, the people who actually write financial SW for a living
seem to think it's not good enough.  You would have to talk to
one of them to find out the exact situations where FP proved
insufficient.  But, I don't think anybody would have gone to
the trouble of implimenting BCD math just for the fun of it.
OTOH, I guess I know people who would do stuff like that just
for fun.

I do remember that BCD math was a big selling point of many
compilers.

>Now suppose a financial instrument carries a yearly interest rate of .077
>(7.7%) with montly payments.  Then the monthly multiplier is
>.0006416666666666666666666666666666666666...
>Whoops, the BCD representation has to be approximate also, just like the
>binary one.

>Even if a monthly multiplier like .00643 (corresponding to a
>yearly 7.716%) is exactly represented in BCD, there may still
>be fractional amounts that need to be rounded to the nearest
>penny by one rule or another.  BCD does not eliminate the
>problems of finite precision arithmetic.

Probably not.  But it apparently makes the problems easier in
the financial world.

-- 
Grant Edwards                   grante             Yow!  The FALAFEL SANDWICH
                                  at               lands on my HEAD and I
                               visi.com            become a VEGETARIAN...


From grzegorz at ugrad.cs.ualberta.ca  Fri Mar  9 01:03:11 2001
From: grzegorz at ugrad.cs.ualberta.ca (Grzegorz Dostatni)
Date: Thu, 8 Mar 2001 23:03:11 -0700
Subject: python, swig and typemaps..
Message-ID: <Pine.GSO.4.21.0103082257230.9953-100000@csu038.cs.ualberta.ca>

Hello... 
I've got a problem with swig and typemaps.  What I'm trying to do is to
have a c subroutine working on some data.  I'd love to pass the data into
c just as a char* (that's what it is).  The problem is it is full of
zeroes (so c null-terminated string will not work very well...)  The best
way would be to have a it so that from python I can call:
example("Hello")
and swig will call the corresponding c function as 
void example(char*, int size).  If that's not possible it's ok to pass in
the size of the string. ie python:
example( "Hello", len("Hello"))
when I'm allocating the char *, how do I get the other parameter (ie
size?)...

Thanx a lot..
Grzegorz Dostatni


########################################################################
The world is full of people who have never, since childhood, met an
open doorway with an open mind.
                -- E.B. White

Chemistry is physics without thought.
Mathematics is physics without purpose.

"If you want to know what God thinks about money, just look at the people
He gives it to." 
-- Old Irish Saying 




From pearu at cens.ioc.ee  Sat Mar 10 14:33:05 2001
From: pearu at cens.ioc.ee (Pearu Peterson)
Date: Sat, 10 Mar 2001 21:33:05 +0200
Subject: import headaches/gripes
In-Reply-To: <mailman.984250586.21661.python-list@python.org>
References: <3aaa3f94.15940553@News.CIS.DFN.DE> <mailman.984250586.21661.python-list@python.org>
Message-ID: <Pine.LNX.4.21.0103102113210.7964-100000@kev.ioc.ee>

On Sat, 10 Mar 2001, Paul Prescod wrote:

> I think there are two reasons:
> 
>  1. Even checking file dates has a performance cost. It involves going
> to the hard disk after all. Code like this isn't rare in Python

I think this is not true in Linux where recently read files stay in
memory buffer so that later access to the files does not involve
(physical) actions on the hard disk. Don't know if Windows has such a
feature..

Regards,
	Pearu




From DanielK at jBASE.com  Wed Mar 21 14:17:19 2001
From: DanielK at jBASE.com (Daniel Klein)
Date: Wed, 21 Mar 2001 11:17:19 -0800
Subject: learning python...
References: <3AB8ECB5.2197E940@noaa.gov>
Message-ID: <m77u6.56$Hp.3284@typhoon.aracnet.com>

I don't think you could go wrong with 'Learning Python' as it patiently
walks you thru all of the basics of the language, and then some. The book is
dated so it does not cover some of the more recent additions to the language
like list comprehensions and new string methods (like join()), etc.
Personally, from the standpoint of an experienced programmer, I found 'The
Quick Python Book' provided an accelerated introduction along with some
additional advanced concepts. From there, I would recommend "Programming
Python 2nd Ed." as well as the indispensible "Python Essential Reference"
(altho I think a new edition is in the works so you might want to wait for
that).

I've also got "Core Python Programming" and if I had to review it, I would
say that it tries to be all things to everyone but never quite hits the
mark. The examples always seem to fall short of what you are really looking
for. The author seems to explain the fundamental stuff very well but just as
it starts to get interesting, the section/chapter ends. I think instead of
using extra thick paper and large fonts to give the visual appearance of an
impressive body of work, the author should have provided a bit more detail
in some of the more advanced areas like network programming.

I hope some of this help.

Dan

"Benjamin.Altman" <benjamin.altman at noaa.gov> wrote in message
news:3AB8ECB5.2197E940 at noaa.gov...
> Hello.  From the perspective of someone who already has programming
> experience, would anyone know if O'reillys "Learning Python" book would
> be better than say Prentice Hall's "Core Python Programming"?  Or would
> it be better to go straight into something like "Programming Python"?
>
> Thanks,
> Ben
>




From paul.moore at uk.origin-it.com  Tue Mar  6 09:47:57 2001
From: paul.moore at uk.origin-it.com (Paul Moore)
Date: Tue, 6 Mar 2001 15:47:57 +0100 
Subject: urllib.urlopen and proxies
References: <vse4atgcgcfafth7c5tq3po7srd3l32f3h@4ax.com> <008301c0a57d$1331cde0$a100000a@local> <mailman.983804894.17415.python-list@python.org>
Message-ID: <BfakOm9OYyn2V0Hg9h1SSS=a7t7Y@4ax.com>

On Mon, 5 Mar 2001 16:07:42 +0100 , Siggy Brentrup <bsb at winnegan.de>
wrote:

>"Chris Gonnerman" <chris.gonnerman at usa.net> writes:
>
>> It *appears* (from reading the source) that username and password
>> information for
>> the proxy server must be passed as part of the URL:
>> 
>>     http://username:password at my.host.com/index.html
>> 
>
>I suggest using urllib2, changing the proxy setup described in the
>module's docstring to
>
>  proxy_support = urllib2.ProxyHandler({"http" :
>                                        "http://user:pw at proxy:port"})
>
>Untested too, since my proxy doesn't require authorization.

I had a look at urllib2. The module docstring even includes an
example!

----------
import urllib2
 
# set up authentication info
authinfo = urllib2.HTTPBasicAuthHandler()
authinfo.add_password('realm', 'host', 'username', 'password')
 
proxy_support = urllib2.ProxyHandler({"http" :
"http://ahad-haam:3128"})
 
# build a new opener that adds authentication and caching FTP handlers
opener = urllib2.build_opener(proxy_support, authinfo,
urllib2.CacheFTPHandler)
 
# install it
urllib2.install_opener(opener)
 
f = urllib2.urlopen('http://www.python.org/')
----------

The only problem is that I don't know what the 'realm' and 'host'
parameters to authinfo.add_password should be. I need to supply a
username and password for everything...

Paul.



From cfelling at iae.nl  Thu Mar  8 15:38:43 2001
From: cfelling at iae.nl (Carel Fellinger)
Date: 8 Mar 2001 21:38:43 +0100
Subject: polymorphjsm &c (was Re: I come to praise .join, not to bury it...)
References: <mailman.983485923.1136.python-list@python.org> <97p7iq$kk2$1@nntp6.u.washington.edu> <97rka002k9@news1.newsguy.com> <9811a8$k5k$1@nntp6.u.washington.edu> <982cja0v5c@news1.newsguy.com> <983u1l$6mk$1@animus.fel.iae.nl> <9851gf014vq@news1.newsguy.com>
Message-ID: <988qkj$7qt$1@animus.fel.iae.nl>

Alex Martelli <aleaxit at yahoo.com> wrote:
...some code corrections sniped

So much for testing then:) I expected some errors, got them, so I was
pleased.  And forget to really look at the errors:(

> but neither do you get what you want, because a.split(b) works
> irregularly...:

>     -- if a is a re object, or if it's module-object os.path,
>         then string b is split using splitter-object a
>     -- if a is a string object (including a Unicode string),
>         then string a is split using splitter-object b (which
>         must also be a string)

Ops, overlooked this one completely.  Should have known better though
having been bit by it in the past.

> I.e., the semantics are backwards in the two cases, alas.

Yes, and one is wrong according to your arguments on joiner.join:)

Though I agree with what you hinted at in an other post, that much of
the gripes over it have to do with the fact that join is Intransitive
and Transitive as well.  Would we choose a different word for it the
reading of joiner.join(to_be_joined) would be clear without us needing
to spell it out like here.  At least, that's what I tent to think now.
[[ besides in dutch this ambiguity gets resolved nicely, as mostly the
   intransitive form has an extra 'zich' (self) attached ]]

-- 
groetjes, carel


From tim.one at home.com  Wed Mar 21 20:20:59 2001
From: tim.one at home.com (Tim Peters)
Date: Wed, 21 Mar 2001 20:20:59 -0500
Subject: New PEP: The directive statement
In-Reply-To: <20010321165943.A25383@glacier.fnational.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCGEGJJHAA.tim.one@home.com>

[piet at cs.uu.nl]
> It would be much less of a joke when __future__ would be replaced by
> __features__, I think.

[Neil Schemenauer]
> __features__ is a terrible name.  Its pretty obvious that you
> (and a lot of other people on this list) don't understand what
> __future__ does.

When I first read the Pentium architecture manual, its description of how the
Pentium handled floating-point exceptions was so off the wall I simply
couldn't believe they meant what they said, and went off writing code after
mentally substituting a *reasonable* (but related) meaning of my own.

Heh.  Turns out it meant exactly what it said!  Cost me plenty to recover
from that, too.

I'm not sure we're not suffering a similar confusion wrt __future__:  there
are no exact analogues in any other language I know of, so I wouldn't be
surprised if people think this *is* for "optional features", and that
"optional" is what they'll *read* no matter what the PEP says.  I may like
Martin's "transitional" better for this reason -- provided that it wakes
people up enough to think about what they're reading.  Alas,

    from __transitional__ import nested_scopes

is not only ugly, it's not even funny, so would be worse than "from
__future__ ...".  It works better with "directive" because it keeps it
uniformly bland.

if-you-can't-laugh-at-your-tools-you're-taking-them-too-seriously-ly
    y'rs  - tim




From tim.one at home.com  Sat Mar  3 14:55:09 2001
From: tim.one at home.com (Tim Peters)
Date: Sat, 3 Mar 2001 14:55:09 -0500
Subject: [Python-Dev] Re: Bug fix releases
In-Reply-To: <15009.17989.88203.844343@cj42289-a.reston1.va.home.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCEELHJDAA.tim.one@home.com>

[Fred L. Drake, Jr.]
> ...
>   Managing the bugfix releases would also be an excellent task for
> someone who's expecting to use the bugfix releases more than the
> feature releases -- the mentality has to be right for the task.  I
> know I'm much more of a "features" person, and would have a hard time
> not crossing the line if it were up to me what went into a bugfix
> release.

Note there was never a bugfix release for 1.5.2, despite that 1.5.2 had some
serious bugs, and that 1.5.2 was current for an unprecedentedly long time.
Guido put out a call for volunteers to produce a 1.5.2 bugfix release, but
nobody responded.  Past is prelude ...

everyone-is-generous-with-everyone-else's-time-ly y'rs  - tim




From skip at mojam.com  Mon Mar  5 20:45:05 2001
From: skip at mojam.com (Skip Montanaro)
Date: Mon, 5 Mar 2001 19:45:05 -0600 (CST)
Subject: gah! I hate the new string syntax
In-Reply-To: <XFMail.20010302085406.shaleh@valinux.com>
References: <97ohh2$polt0$1@ID-77331.news.dfncis.de>
	<XFMail.20010302085406.shaleh@valinux.com>
Message-ID: <15012.16673.237937.603550@beluga.mojam.com>

    Sean> Turning a string literal (";") suddenly into a string object and
    Sean> then using its methods is not an easy read.

In Python, everything is an object.  Thus, a string literal is a string
object - no "turning into" required.  Just because it wasn't created by
instantiating a class doesn't make it any less an object.

The presence of join methods for strings and why that's where they landed
has been hashed and rehashed several times.  Please check out FAQ 4.96 for
details.

    Sean> ";" . join() # in perl, this means concat ";" with the result of
    Sean> join()

Nobody ever claimed that Python was trying to be Perl... ;-) Just because
'";".join' is a fragment of valid expressions in both languages doesn't mean
they have to mean the same thing.

(In reality, I actually do my best to make my Perl code look as much like
Python to my Python-infected mind as I can.)

-- 
Skip Montanaro (skip at mojam.com)
Support Mojam & Musi-Cal: http://www.musi-cal.com/sponsor.shtml
(847)971-7098



From paulp at ActiveState.com  Tue Mar 27 14:37:23 2001
From: paulp at ActiveState.com (Paul Prescod)
Date: Tue, 27 Mar 2001 11:37:23 -0800
Subject: PEP 240 scares me
References: <Pine.LNX.4.21.0103222338530.29360-100000@kev.ioc.ee>
		<99e2h61aao@news2.newsguy.com> <15040.53203.949420.930912@beluga.mojam.com>
Message-ID: <3AC0EBF3.F8FF1F68@ActiveState.com>

Skip Montanaro wrote:
> 
>...
> Okay, dumb question time.  Suppose the compiler sees the string "1.2" in the
> input stream.  How does it translate that into a rational number (6/5 or
> 12/10 or 24/20 or ...)?  Does it simply move the decimal point over
> sufficiently and create the corresponding rational number that contains the
> appropriate power-of-ten denominator?  

My understanding is that it moves the decimal point over and then it
factors out common factors from the top and bottom.

> What about rationals that can't be
> represented correctly using decimal notation (e.g. how does one get from
> 0.3333 to 1/3, or is it simply approximated by 3333/10000?)

0.3333 has no relation to 1/3. If you want 1/3 you would just type 1/3.
The idea is that division would yield rationals also!

-- 
Take a recipe. Leave a recipe.  
Python Cookbook!  http://www.activestate.com/pythoncookbook



From tim.hochberg at ieee.org  Tue Mar  6 15:19:23 2001
From: tim.hochberg at ieee.org (Tim Hochberg)
Date: Tue, 06 Mar 2001 20:19:23 GMT
Subject: Long Shifty Bits
References: <kmap6.15$8Q.28605@news.pacbell.net>
Message-ID: <fDbp6.2219$Xt3.521106@news1.rdc1.az.home.com>

Have you tried it and found a problem?

My understanding is that this stuff should "just work"with long integers.
Not that I've tried it, since this isn't my problem domain. Trying your
particular problem one gets:

>>> x = 1L << 200
>>> hex(x)
'0x100000000000000000000000000000000000000000000000000L'

Which appears, at least to casual inspection,  to be the correct answer.

-tim

"Phlip" <phlip_cpp at my-deja.com> wrote in message
news:kmap6.15$8Q.28605 at news.pacbell.net...
> Not hyp:
>
> I want this expression to make sense:
>
>         x = 1L << 200
>
> If someone tells me "it will roll over at the 32nd bit", or mayby even "it
> will roll over when the CPU damn well feels like it", that's okay. I'l
just
> construct a Bit Vector of some ilk and drop my bits into it. But the above
> notation is closest to our problem space, so it would be nice if it worked
> as expected.
>
> --
>   Phlip                          phlip_cpp at my-deja.com
> ============== http://phlip.webjump.com ==============
>   --  "In my experience, the customer doesn't know what he wants
>        until you don't give it to him." --David Brady  --




From aleaxit at yahoo.com  Thu Mar 29 04:29:37 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 29 Mar 2001 11:29:37 +0200
Subject: Is inheritance broken?
References: <mailman.985808317.25055.python-list@python.org>
Message-ID: <99uva501v6o@news2.newsguy.com>

"Jonathan Claggett" <jcc.ugm at ix.netcom.com> wrote in message
news:mailman.985808317.25055.python-list at python.org...
> >    Jonathan> Isn't this wrong?
> >
> >Nope.  It's a feature of multiple inheritance, not a bug.  The search
> >through base classes is left-to-right, depth first, thus
> >parallelogram.angle is found before rectangle.angle.
>
> Thank you, I think I understand now. I think the problem is that the depth
> first searching works fine so long as you don't have a common ancestor in
> your class tree. If you do have a common ancestor, it will be searched
ahead
> of some of its descendant classes.

Good summary.

> I wonder if there is a (quick) way to do
> a depth first search that accounts for common ancestors...

Not all that quick, but, what about...:


# a toy example of the potential problem-case:
class ancestor:
    def a(self): return 'ancestor.a'
    def b(self): return 'ancestor.b'
class descen_a(ancestor):
    def a(self): return 'descen_a.a'
class descen_b(ancestor):
    def b(self): return 'descen_b.b'
class desc_both(descen_a, descen_b):
    pass

# let's show the problem in action:
db = desc_both()
print db.a()
print db.b()

# a function to diagnose this possible problem:
def study(obj, attrib_name):
    attrib = getattr(obj.__class__, attrib_name)
    at_clas = attrib.im_class
    print 'Found:', attrib
    for base in obj.__class__.__bases__:
        found = getattr(base, attrib_name, None)
        if found is not None and found != attrib:
            print '  also:', found,
            fo_clas = found.im_class
            if issubclass(fo_clas, at_clas):
                print "(potential problem, override ignored)"
            else:
                print "(no problem, unrelated)"

# let's show the function at work:
print
study(db, 'a')
study(db, 'b')



D:\>python a.py
descen_a.a
ancestor.b

Found: <unbound method descen_a.a>
  also: <unbound method ancestor.a> (no problem, unrelated)
Found: <unbound method ancestor.b>
  also: <unbound method descen_b.b> (potential problem, override ignored)


Alex





From tim.hochberg at ieee.org  Thu Mar  1 12:32:49 2001
From: tim.hochberg at ieee.org (Tim Hochberg)
Date: Thu, 01 Mar 2001 17:32:49 GMT
Subject: Combinations of n Lists
References: <NI8n6.2979$TW.15640@tor-nn1.netcom.ca> <89bn6.1121$DY.186465@news1.rdc1.az.home.com> <97lqcm030au@news2.newsguy.com>
Message-ID: <5Jvn6.615$ls4.163844@news1.rdc1.az.home.com>

"Alex Martelli" <aleaxit at yahoo.com> wrote in message
news:97lqcm030au at news2.newsguy.com...
> "Tim Hochberg" <tim.hochberg at ieee.org> wrote in message
> news:89bn6.1121$DY.186465 at news1.rdc1.az.home.com...
>     [snip]
> > > def combinations(list1,list2):
> > >     return [ (i,j) for i in list1 for j in list2 ]
> > >
> > > print combinations( [1,2,3], ['a','b','c'] )
> > >
> > > [(1, 'a'), (1, 'b'), (1, 'c'), (2, 'a'), (2, 'b'), (2, 'c'), (3, 'a'),
> (3,
>     [snip]
> > zip([1,2,3], [4,5,6], [7,8,9], .... )
>
> Nope -- this only gives [(1,4,7),(2,5,8),(3,6,9)], *not* all of the
> other 24 requested combinations (1,4,8), (1,4,9), ..., (3,6,8).

Foiled again. I didn't read the original posting very closely I guess.
Sigh....

-tim




From hzhu at mars.localdomain  Wed Mar 21 04:03:28 2001
From: hzhu at mars.localdomain (Huaiyu Zhu)
Date: Wed, 21 Mar 2001 09:03:28 GMT
Subject: Yet Another PEP:  Query Protocol Interface or __query__
References: <mailman.985144046.3492.python-list@python.org>
Message-ID: <slrn9bgr5h.rv.hzhu@mars.localdomain>

On Wed, 21 Mar 2001 14:06:07 +1100, Delaney, Timothy <tdelaney at avaya.com> wrote:
>So far as I can see, this PEP does not solve one basic problem - interfaces
>which the class writer doesn't know about.
>
>For example, I have an interface, mapped to "org.mydomain.something". Then I
>have a slightly different interface mapped to "org.mydomain.somethingelse".
>
>I have a class which implements org.mydomain.something. However, whilst it
>can perfectly implement org.mydomain.somethingelse, it would be rejected as
>it doesn't actually know about it.

This is not difficult to solve.  See 

http://www.geocities.com/huaiyu_zhu/python/Interface/typecheck.txt

Just change the file name to typecheck.py to test if it does what you want.

The idea is that instead of a class stating what interfaces it belong to,
and interfaces actively states what it covers.  This allows inclusion of
things into an interface based on various criteria:
- instantance of class
- having a given method
- is of a specific type
- is a particular object
- etc

This also allows both interetance base on classes and reversed inheritance
based on interfaces themselves.

>I may well just try implementing this some time ... most of it shouldn't be
>too hard. Unfortunately, I've got a nasty deadline coming up ... :(

That implementation has been sitting on my page for many months now :-)
I think I mentioned it here once.  Maybe I should just post it here,
as most people don't usually follow links.

-- 
Huaiyu Zhu   <hzhu at users.sourceforge.net>


From dietmar at schwertberger.de  Wed Mar  7 16:55:56 2001
From: dietmar at schwertberger.de (Dietmar Schwertberger)
Date: Wed, 7 Mar 2001 22:55:56 +0100
Subject: National Intruments GPIB
References: <3AA46593.753BAE1@newfocus.com>
Message-ID: <ant0721560b02%2U@schwertberger.freenet.de>

Hi Ken!

In article <3AA46593.753BAE1 at newfocus.com>, Ken Koller
<URL:mailto:kkoller at newfocus.com> wrote:
> Has anyone written a python interface to some of the National
> Instruments libraries? I'm in the process of using Python to connect to
> some instrumentation and I'm writing the C code to interface to their
> GPIB API. Was hoping someone else had done it :)

Instead of using C you could just call the DLL using windll/calldll.


Example(s):
(not necessarily functional)

= NI_GPIB =================================================================
[snip]

from GPIB import *

import windll, calldll
gpib = windll.module("gpib-32")

def ibdev(BdIndx=0, address=(0,0), more=None):
    '''ibdev(BdIndx=0, address=(0, 0), more=(11,1,0)):
    Open and initialize a device descriptor (handle) for
    use with subsequent calls for e.g. ibwrt and ibrd.
        BdIndx Index of the access interface for the device
        address  The (primary, secondary) GPIB address of the device
        more     (The I/O timeout value, EOI mode of the device, EOS character and modes)
    Returns the device descriptor or raises GPIBError.
    '''
    if more==None:
        more=(11,1,0)
    handle = gpib.ibdev( BdIndx, address[0], address[1], more[0], more[1], more[2])
    if handle > 0:
        return handle
    _raise( GPIBIOError, "ibdev(BdIndx=%d, address=%s, more=%s): Could not get handle."%(BdIndx, str(address), str(more)) )

get_handle=ibdev

def ibwrt(handle, s):
    '''ibwrt(handle, s):
    Write data to a device from a user buffer.
        handle: Interface or device descriptor
        s:      bytes to write
    Returns None or raises GPIBIOError (or GPIBTimeoutError).
    '''
    wrbuffer = windll.cstring(s,len(s))
    ibsta = gpib.ibwrt(handle,wrbuffer.address(),len(wrbuffer))
    if ibsta <> _iocompletedstatus:
        _raise(GPIBIOError, "ibwrt(handle=%d, s='%s'): Could not send string."%(handle, s), ibsta )

write = ibwrt


[snip]

boardID = 0
deviceaddress = (9,0)

# low level:
handle = ibdev(boardID, l)
ibwrt(handle, "*IDN?")

# higher level
bus = NI_GPIBInterface(boardID)
device = GPIBDevice(bus, address)
device.write("*IDN")

# highest level
multimeter1 = HP34970A(bus, address)
voltage = multimeter1.measure(range="10VAC", channel=5)

=========================================================================


I'm still working on the highest level classes for multimeters, sources,
source-monitor-units, relais scanners, ...


You may also try a search at google/deja for a posting by Sam Schulenburg:
# Forum: comp.lang.python
# Thread: Python and National Instruments
# Subject: Re: Python and National Instruments
# Date: 03/15/2000
# Author: Sam Schulenburg <samschul at pacbell.net>



Regards,

Dietmar



From kens at sightreader.com  Wed Mar 14 15:41:45 2001
From: kens at sightreader.com (Ken Seehof)
Date: Wed, 14 Mar 2001 14:41:45 -0600
Subject: Fw: reading binary data
Message-ID: <002f01c0acc7$301202d0$04090a0a@upcast.com>

----- Original Message -----
From: "Ken Seehof" <kens at sightreader.com>
To: "Francois Forest" <francois_forest at yahoo.fr>
Sent: Wednesday, March 14, 2001 2:37 PM
Subject: Re: reading binary data


> Take a look at the "struct" module.  It has all you need to mess around
with
> binary data.  I have an interactive tool (called filecracker) for reverse
> engineering binary files using the struct module, but I haven't found time
> to release it yet.  If you want, I can email it to you (if enough people
> respond to this, I'll be motivated to post it in parnassus).
>
> BTW, there's a gotcha, that everyone learns the hard way early on.  Make
> sure you use the 'b' flag when opening binary files.
>
> f = open('ni.wav', 'rb')
>
> If you don't, the file will be truncated at the first ascii end-of-file
> marker that happens to be in the binary data.
>
> ----- Original Message -----
> From: "Francois Forest" <francois_forest at yahoo.fr>
> Newsgroups: comp.lang.python
> To: <python-list at python.org>
> Sent: Wednesday, March 14, 2001 3:46 PM
> Subject: reading binary data
>
>
> > I would like to read and write xxx.wav files,
> > this assume to alternatively read characters,  integers 16 bits and
> > integers 32 bits.
> > I found that I can read a binary file and get the result in a string,
> > Is there a way to convert a string slice into an int16, int32, or
> > floatting point ?
> > (what is usually done in C through a cast).
> >
> > More generally, how can I browse through standard libraries ?
> >
> > --
> > http://mail.python.org/mailman/listinfo/python-list
> >
>




From bryan at eevolved.com  Fri Mar 23 03:13:20 2001
From: bryan at eevolved.com (Bryan Mongeau)
Date: Fri, 23 Mar 2001 08:13:20 GMT
Subject: Good GUI for Python
References: <mailman.985314033.5120.python-list@python.org>
Message-ID: <AADu6.436656$f36.14223632@news20.bellglobal.com>

For some reason, people tend to overlook FXPy, the python bindings for the 
FOX GUI toolkit:

fxpy.sourceforge.net

There is also pyQT available from thekompany.com but the legality of the 
windows licensing remains a thornbush.

Regards,
-- 
<=====================================>
Bryan Mongeau
Lead Developer, Director
eEvolved Real-Time Technologies Inc.
Website: http://www.eevolved.com
Public key: http://eevolved.com/bcm.pk
<=====================================>

"Seek ye first the good things of the mind, and the rest will either be 
given you, or its absence will not be missed." -- Francis Bacon



From kevin_cazabon at hotmail.nospamplease!.com  Sun Mar 11 23:28:31 2001
From: kevin_cazabon at hotmail.nospamplease!.com (Kevin Cazabon)
Date: Sun, 11 Mar 2001 21:28:31 -0700
Subject: Who am I: can a class instance determine its own name?
References: <mailman.984034150.9635.python-list@python.org> <DYV4H3fqAHA.187@typhoon.i3s.net> <mailman.984324974.25480.python-list@python.org>
Message-ID: <vZJg10qqAHA.210@typhoon.i3s.net>

The reason I say it's not elegant is that there is the possibility (read
"probability") that the assigned name and the actual name can differ.

We're assuming that the programmer assigns exactly the same name to both,
which doesn't seem like a hard task, but it introduces the possibility of
problems, and probably hard ones to debug too.

If there were an automatic method, it would be MUCH more reliable.

Just MHO.

Kevin.



"Paul Prescod" <paulp at ActiveState.com> wrote in message
news:mailman.984324974.25480.python-list at python.org...
| Kevin Cazabon wrote:
| >
| > ...
| > You could also
| > make it an initialization parameter.
| >
| > FooBar = Foo(name="FooBar")
| >
| > Just trying to help... not elegantly though.
|
| I disagree Kevin. This is not only the canonical way to give an instance
| a name, it is also the elegant way. It gives you a lot more flexibility
| because the name it has at runtime is unrelated to the variable name you
| happened to have given it in your code. If you need to change one or the
| other for any reason, you can do so safely.
|
|  Paul Prescod
|




From aahz at panix.com  Sun Mar 11 14:21:15 2001
From: aahz at panix.com (Aahz Maruch)
Date: 11 Mar 2001 11:21:15 -0800
Subject: any interest in type-scoped static constants?
References: <3AAB9578.4C8754E4@fxtech.com> <98g85d$cih$1@panix6.panix.com> <3AABCA0E.CB3E2F4A@fxtech.com>
Message-ID: <98gj7b$83u$1@panix2.panix.com>

In article <3AABCA0E.CB3E2F4A at fxtech.com>,
Paul Miller  <paul at fxtech.com> wrote:
>Aahz:
>>
>> Python itself, the way I'd do this is to create a module Point that
>> contains the Point class, and following the class definition create the
>> ZERO object.  Then you refer to it as Point.ZERO and create new points
>> with Point.Point().
>
>You are right that I could do this. However, I'm talking about extension
>types. If I am embedding Python in an app, I may be providing several
>extension types, that provide access to my internal C++ classes.
>However, I do not want to make a module for each type.

Again talking in terms of what I'd do in Python, I'd handle this by
writing a package with internal modules.  I don't know if it's possible
to build virtual packages in extensions.
-- 
                      --- Aahz (Copyright 2001 by aahz at pobox.com)

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

"Perhaps God rewards martyrs, but life seldom does..." --Ulrika O'Brien


From clarence at netlojix.com  Tue Mar  6 10:53:37 2001
From: clarence at netlojix.com (Clarence Gardner)
Date: Tue, 6 Mar 2001 07:53:37 -0800
Subject: I come to praise .join, not to bury it...
References: <mailman.983890030.6303.python-list@python.org>
Message-ID: <983894109.2077137603@news.silcom.com>

On Tue, 06 Mar 2001, Mike C. Fletcher wrote:
>Here's a professional aesthetic's take:
>
>	mystring.join( )
>
>This is a perfectly normal and good-looking construct.  It seems pythonic
>and beautiful, object orientation is part of Python, and this works nicely.
>
>	"'".join( )
>	";".join( )
>	".".join( )
>	"!".join( )
>	",".join( )
>
>All look like executable line noise.  That is, the literal string's syntax
>makes the use of dotted attributes on such a string look jarring and
>off-putting.
<snip>

>That said, I don't think we'll get this particular wart out of the language,
>so we'll have to deal with the flaw in the way artists have through the
>ages, enhance it and accent it so people start to think it's a beauty mark.

comma.join()
period.join()
bang.join()
etc.




From simone at oreilly.com  Fri Mar  9 12:28:53 2001
From: simone at oreilly.com (Simone Paddock)
Date: Fri, 09 Mar 2001 17:28:53 GMT
Subject: Updated edition of Programming Python out
Message-ID: <pj4iat8kevh2fp52gkua87ig16h76oh3m0@4ax.com>

The updated, 1256 pages strong tome is finally here:
"Programming Python, 2nd Edition"
http://www.oreilly.com/catalog/python2/

A sample chapter is here: 
http://www.oreilly.com/catalog/python2/chapter/ch15.html

Enjoy.




From cg at schlund.de  Sat Mar 10 13:38:00 2001
From: cg at schlund.de (Carsten Gaebler)
Date: Sat, 10 Mar 2001 19:38:00 +0100
Subject: How to efficently build a nested dictionary
References: <3AAA2042.E69A53FC@gmx.de> <98dl0n$30p$1@animus.fel.iae.nl>
Message-ID: <3AAA7487.38AF182@schlund.de>

Carel Fellinger wrote:
> 
> or use newer Python goodies like:
> 
>    m = traffic.setdefault(dir,{}).setdefault(year,{}).setdefault(month,{})
>    m[day] = m[day] + bytes
> 

That's it. Thanks!

cg.


From JamesL at Lugoj.Com  Sat Mar 10 16:44:52 2001
From: JamesL at Lugoj.Com (James Logajan)
Date: Sat, 10 Mar 2001 13:44:52 -0800
Subject: "!=" is bad form. Re: sorry....never mind
References: <mailman.984252135.24141.python-list@python.org>
Message-ID: <3AAAA054.7C536BB0@Lugoj.Com>

Tim Peters wrote:
> print>>-got-in-for-"consistency-with-c"-too-ly y'rs  - tim

Oh dear; an ominous trend. When can we expect { } notation? I need to know
because it takes some time to pull my old arguments out of the closet to
support:
if ....:
{
   blah blah
}

over

if ....: {
   yuck yuck
}


From nospam at packeteer.com  Tue Mar  6 13:25:04 2001
From: nospam at packeteer.com (Bob Purvy)
Date: Tue, 06 Mar 2001 10:25:04 -0800
Subject: setjmp.h
Message-ID: <3AA52B80.507E23A7@packeteer.com>

hi,

I'm trying to port Python 2.0 to pSOS (yes, that's just our little cross
to bear here!).  I hit a roadblock with pypcre.c:  it seems to require
long jumps, and there is no way to configure that out.  I'm still
checking whether it's possible for us to enable them in our software,
but we definitely don't have them right now.

Is there any way around this requirement?  Maybe an earlier version of
Python?






From jarober at mail.com  Wed Mar 21 08:06:10 2001
From: jarober at mail.com (James A. Robertson)
Date: Wed, 21 Mar 2001 13:06:10 GMT
Subject: Collection interfaces
References: <3a90fa97.0@news1.mweb.co.za>   <MCjm6.474652$U46.14163288@news1.sttls1.wa.home.com> <MPG.150394195f7f04a19896a7@news.earthlink.net> <3A9A86B1.9A8C69A2@ix.netcom.com> <MPG.1504d14563dce6499896af@news.earthlink.net> <97fjla128ir@news2.newsguy.com> <MPG.1505a76831c34b749896b3@news.earthlink.net> <97h5s40vhc@news2.newsguy.com> <MPG.1506007591cf818b9896bc@news.earthlink.net> <slrn99opop.gr.amk@mira.erols.com> <MPG.1506265ebf6d934d9896be@news.earthlink.net> <97ibmc0k47@news1.newsguy.com> <MPG.1506cf3293b094b49896c1@news.earthlink.net> <97m0c7080c@news2.newsguy.com> <MPG.15085517debf4b649896ca@news.earthlink.net> <zb%n6.31033$o%6.586026@news1.rdc1.bc.home.com> <MPG.150b383cea226d409896d0@news.earthlink.net> <kNko6.6986$eq3.707121@news.uswest.net> <MPG.150c46db9cec283e9896d4@news.earthlink.net> <3AA2A898.450B659E@mail.com> <gRCt6.12224$Im6.1308560@newsread1.prod.itd.earthlink.net> <3AB755F8.67E84126@mail.com> <%a%t6.17245$227.1601569@newsread2.prod.itd.earthlink.net>
Message-ID: <3AB8A730.4BE13D22@mail.com>

topmind wrote:
> 

> > http://www.cincom.com/smalltalk
> >
> 
> I only see a front page, not a parsing example.
> 

Follow the tutorial link - it's right there on that page.  


> Besides, what are you comparing it to so say
> it is better than a procedural/relational approach?

It's vastly easier than similar code done in C or basic, for example

> 
> > >
> > > > [snip]
> > >
> > > -tmind-
> > > oop.ismad.com
> >
> > --
> > James A. Robertson
> > Product Manager (Smalltalk), Cincom
> > jarober at mail.com
> > <Talk Small and Carry a Big Class Library>
> >

-- 
James A. Robertson
Product Manager (Smalltalk), Cincom
jarober at mail.com
<Talk Small and Carry a Big Class Library>


From grey at despair.rpglink.com  Thu Mar 22 08:24:06 2001
From: grey at despair.rpglink.com (Steve Lamb)
Date: Thu, 22 Mar 2001 13:24:06 GMT
Subject: Good Editor?
References: <3ab9a32b$1_1@news2.one.net> <3AB9FAC9.97EA420E@student.liu.se>
Message-ID: <slrn9bjv7m.4qv.grey@teleute.dmiyu.org>

On Thu, 22 Mar 2001 14:14:49 +0100, Rikard Bosnjakovic
<rikbo716 at student.liu.se> wrote:
>I prefer Emacs and python-mode, it's really neat. But I know a lot of
>guys who won't touch Emacs even with a long stick :)

<ObEMACSJoke>
That's because Emacs won't fit on a long stick of RAM.  Everyone knows it
takes 3 long sticks of RAM before Emacs will run.
</ObEMACSJoke>

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


From aahz at panix.com  Wed Mar 14 00:10:37 2001
From: aahz at panix.com (Aahz Maruch)
Date: 13 Mar 2001 21:10:37 -0800
Subject: Help doing simple threading
References: <3AAEF93E.F817D852@theworld.com>
Message-ID: <98mugd$fmm$1@panix3.panix.com>

In article <3AAEF93E.F817D852 at theworld.com>,
David Lees  <deblNonOspammy at theworld.com> wrote:
>
>I am trying to learn how to use threads.  I want multiple threads to be
>able to communicate.  I thought that threads started by a common process
>would share globals in that module.  I want thread to be able to store
>results in a common array.  However the following returns all zeros, so
>clearly I am not passing anything back to the top level list 'z'.  Can
>someone tell me how I should be passing things back from threads?

Um.  No offense, but I think you'd find it more productive if you learn
a bit more about Python before trying to use threads.  For example,
"global" has no effect at the module level.
-- 
                      --- Aahz <*>  (Copyright 2001 by aahz at pobox.com)

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

Python:
    Programming the way
    Guido indented it


From andreas at andreas-jung.com  Mon Mar 19 20:46:56 2001
From: andreas at andreas-jung.com (Andreas Jung)
Date: Mon, 19 Mar 2001 20:46:56 -0500
Subject: Paring down diskspace required by a Python-2.0 installation
References: <6dbdbtsg6beh5p73cvlubeb192oocohfbb@4ax.com>
Message-ID: <996ct7$g3d$1@nntp9.atl.mindspring.net>

Usually you only need the python interpreter and all the stuff under
lib/pythonXX
You can also remove the tests directory in the lib/python dir.

Andreas
"Sheila King" <sheila at spamcop.net> schrieb im Newsbeitrag
news:6dbdbtsg6beh5p73cvlubeb192oocohfbb at 4ax.com...
> directories. Like Grammar/ and Modules/ which have only .c and .h and .o
files
> in them. These files are not used for running Python, only for creating
the
> executable, right? So can I safely delete them?
>
> Is there anything else I can remove?
>





From mcfletch at home.com  Wed Mar  7 14:43:39 2001
From: mcfletch at home.com (Mike C. Fletcher)
Date: Wed, 7 Mar 2001 14:43:39 -0500
Subject: I come to praise .join, not to bury it...
In-Reply-To: <984tcd010qp@news1.newsguy.com>
Message-ID: <000101c0a73e$e8ec5ff0$a9807018@cr706570a>

Okay, I've built a few modules now (this Python thing is fun, why didn't I
find out about it earlier ;) ), but I'm not sure I've really got the hang of
it...

	http://members.home.net/mcfletch/programming/

Still haven't bought the books though.  Hmm, only one I have is Win32
Programming.  Was disappointing that it dealt mostly with Python, I was
hoping more for a tome containing distilled knowledge of Win32 arcana.

Enjoy yourself,
Mike


-----Original Message-----
From: Alex Martelli [mailto:aleaxit at yahoo.com]
Sent: Wednesday, March 07, 2001 4:01 AM
To: python-list at python.org
Subject: Re: I come to praise .join, not to bury it...


"Mike C. Fletcher" <mcfletch at home.com> wrote in message
news:mailman.983928669.28324.python-list at python.org...
...
> I'll have to give that Python thing a try.  Can you recommend any good
books
> for someone who's been stuck in a backwater language for the last 6 years?

My favourite books for Python beginners are:
...




From mertz at gnosis.cx  Sat Mar 17 20:18:27 2001
From: mertz at gnosis.cx (Dr. David Mertz)
Date: Sat, 17 Mar 2001 20:18:27 -0500
Subject: Looking for web indexer in python (fwd)
Message-ID: <jzAt6kKkXEFZ092yn@bellatlantic.net>                         

Roy Smith <roy at panix.com> wrote:
|The software development group I work in has a small internal web site (a
|few hundred documents) which we don't currently have any way of searching.
|I'd like to put an automated indexer/search system in place.  If possible,
|I'd like to use python, since I think it would be a nifty way to introduce
|the group to the language.

Well...  I'd love it if my recent indexer turns out to prove useful to
you:

  http://gnosis.cx/download/indexer.py

Public domain, so you can do what you like with it.  It does not have a
CGI interface yet, but it does the indexing/searching part pretty
nicely.  Best of luck.

Yours, David...

P.S. I *did* actually submit this to Parnassus a week or two ago... it
seems to have gotten lost there.  Dunno why.  I guess I'll try doing
their forms again (although I have had a similar submission experience
in the past).



From hernan.foffani at iname.com  Tue Mar 20 05:38:04 2001
From: hernan.foffani at iname.com (Hernan M. Foffani)
Date: 20 Mar 2001 10:38:04 GMT
Subject: Writing PEPs (was Re: Why "from __future__" stinks; a counter-offer)
References: <mailman.984949523.1896.python-list@python.org> <4Aft6.2171$pP2.372173@paloalto-snr1.gtei.net> <mailman.985029625.6313.python-list@python.org> <995pl4$jta$1@news.mathworks.com> <995rod$en6$1@panix3.panix.com>
Message-ID: <20010320053804.682$tV@newsreader.com>

aahz at panix.com (Aahz Maruch) wrote:
> ...[snip]...
> Why?  BDFL checks off on *all* PEPs.  Thus the worst case scenario (e.g.
> "print >>" ;-) is that a proposal has enthusiastic proponents *and* is
> good enough to satisfy Guido.
> ...[snip]...

Speaking as someone who doesn't write PEPs and neither wants to, I truly
believe that the PEP mechanism along with the PSF (Python Software
Fundation) helps the Python Language to stand out in the software jungle
making it unique compared to open source and commercial products.

BTW, I didn't see many discussion here about the PSF. And it's THE Python's
news of the year! hmmm, maybe because there's nothing to whin about it :-)

regards,
-Hernan

-- 
-------------------- http://NewsReader.Com/ --------------------
                       Usenet for the Web


From korpo at 01019freenet.de  Thu Mar 22 17:35:12 2001
From: korpo at 01019freenet.de (Oliver Korpilla)
Date: Thu, 22 Mar 2001 23:35:12 +0100
Subject: Python and emacs
References: <mailman.985286728.17308.python-list@python.org>
Message-ID: <3ABA7E1F.869BD5CB@01019freenet.de>

Don Dwiggins wrote:

> A semi-idle question: has anyone looked into embedding Python as an
> extension language in Emacs, alongside Elisp?  I'm toying with the idea of
> doing some Emacs enhancements/modifications, and given the choice, I'd much
> rather do them in Python.  (One such enhancement would be to add
> Pythonwin-style "intellisense" popups to python-mode.)
>
>
>
>
>

Well, good luck. Emacs is written in LISP, so doing anything with Emacs
requires you to know LISP thoroughly. LISP is functional, so you want to
interface between Emacs functional buildup to Python. Sounds tough too
me.. and if you do learn LISP well enough to install a Python
interpreter within Emacs, why then not use LISP AND when you've done
writing that LISP interface to Python, will you wrap all LISP functions
to work within Python, or will you write a new Python-comprehensing
editor replacing LISP emacs ?
Sounds funny.

Ciao, Korpo



From debl at NoNoSpammytheworld.com  Sat Mar 31 20:58:36 2001
From: debl at NoNoSpammytheworld.com (David Lees)
Date: Sun, 01 Apr 2001 01:58:36 GMT
Subject: PythonLabs Team Moves to ActiveState
References: <mailman.986086501.5576.python-list@python.org>
Message-ID: <3AC68B5A.E2A41D0E@NoNoSpammytheworld.com>

Could someone explain exactly what this means?  I read over PEP 228 and
am
wondering if this is what you are referring to below with respect to the
'Perl numerical model'?

I assume Python will still keep long integers, integers and complex.
All that will change is coersion to float on the division of integers?

Would it be possible to clarify this change?

Thanks,

David Lees


   "...Also, after some talks with Larry, I've decided to adopt the Perl 
   numerical model for Python. The Perl model is simple, predictable and
   easy to explain. This is a small enough change that it does not need
   a PEP, and Moshe will implement it before 2.1rc1 is out."


From brendhanhorne at bellsouth.net  Sat Mar 17 19:22:15 2001
From: brendhanhorne at bellsouth.net (Brendhan Horne)
Date: Sat, 17 Mar 2001 19:22:15 -0500
Subject: Interactive startup file
Message-ID: <3dTs6.2144$ge.55105@news3.mia>

In the tutorial it  says in section 2.2.3
When you use Python interactively, it is frequently handy to have some
standard commands executed every time the interpreter is started. You can do
this by setting an environment variable named $PYTHONSTARTUP to the name of
a file containing your start-up commands. This is similar to the .profile
feature of the Unix shells.

This file is only read in interactive sessions, not when Python reads
commands from a script, and not when /dev/tty is given as the explicit
source of commands (which otherwise behaves like an interactive session). It
is executed in the same namespace where interactive commands are executed,
so that objects that it defines or imports can be used without qualification
in the interactive session. You can also change the prompts sys.ps1 and
sys.ps2 in this file.

If you want to read an additional start-up file from the current directory,
you can program this in the global start-up file, e.g. "if
os.path.isfile('.pythonrc.py'): execfile('.pythonrc.py')". If you want to
use the startup file in a script, you must do this explicitly in the script:


import os
filename = os.environ.get('PYTHONSTARTUP')
if filename and os.path.isfile(filename):
    execfile(filename)




Can some one please explain this to me in novice english and for a guy who
is using windows98 with his python.

--
Thanks,
Brendhan




From guido at digicool.com  Thu Mar  1 15:32:37 2001
From: guido at digicool.com (Guido van Rossum)
Date: Thu, 01 Mar 2001 15:32:37 -0500
Subject: [Python-Dev] Re: d = {}; d[0:1] = 1; d[0:1] = 2; print d[:]
In-Reply-To: Your message of "Thu, 01 Mar 2001 11:37:53 EST."
             <15006.31457.377477.65547@w221.z064000254.bwi-md.dsl.cnc.net> 
References: <4.3.1.2.20010228223714.00d29430@exchange.starmedia.net>  
            <15006.31457.377477.65547@w221.z064000254.bwi-md.dsl.cnc.net> 
Message-ID: <200103012032.PAA18322@cj20424-a.reston1.va.home.com>

> >>>>> "RT" == Robin Thomas <robin.thomas at starmedia.net> writes:
> 
>   RT> Using Python 2.0 on Win32. Am I the only person to be depressed
>   RT> by the following behavior now that __getitem__ does the work of
>   RT> __getslice__?

Jeremy:
> You may the only person to have tried it :-).
> 
>   RT> Python 2.0 (#8, Oct 16 2000, 17:27:58) [MSC 32 bit (Intel)] on win32
>   >>> d = {}
>   >>> d[0:1] = 1
>   >>> d
>   {slice(0, 1, None): 1}
> 
> I think this should raise a TypeError (as you suggested later).

Me too, but it's such an unusual corner-case that I can't worry about
it too much.  The problem has to do with being backwards compatible --
we couldn't add the 3rd argument to the slice API that we wanted.

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



From tuttledon at hotmail.com  Sat Mar 10 01:25:03 2001
From: tuttledon at hotmail.com (Don Tuttle)
Date: Sat, 10 Mar 2001 06:25:03 GMT
Subject: Python In A Nutshell
References: <nG5q6.54910$__6.11376601@typhoon.southeast.rr.com> <Ufcq6.140786$47.2078615@news.bc.tac.net> <Blfq6.60716$__6.11886688@typhoon.southeast.rr.com> <98c4ok$fhd$1@panix2.panix.com>
Message-ID: <3Njq6.78213$Z8.15223029@typhoon.southeast.rr.com>

> [Don Tuttle]
> >Planning stages?  We've all seen his posts (a.k.a. treatises) here.  I'd
say
> >he should have the final draft to you in 3 weeks tops. ;-)
>
>[Aahz ]
> ITYM "first draft".  It will take another three-to-six months for Alex
> to edit it down.  ;-)

Good point Aahz. And of course another six months for O'reilly to cut it
down to a mere 1,200 pages!  All in all, it MIGHT be out just in time for
Christmas 2002  ;-)

Don





From andymac at bullseye.apana.org.au  Thu Mar  8 04:14:22 2001
From: andymac at bullseye.apana.org.au (Andrew MacIntyre)
Date: Thu, 8 Mar 2001 20:14:22 +1100 (EDT)
Subject: I come to praise .join, not to bury it...
In-Reply-To: <9850ge013ns@news1.newsguy.com>
Message-ID: <Pine.OS2.4.32.0103081958120.6600-100000@central>

On Wed, 7 Mar 2001, Alex Martelli wrote:

> Suppose I'm building a line of a CSV file, a popular
> textual format where fields are separated (aka joined)
> with commas -- the acronym stands for 'comma-separated
> values'.  So, what I want to do is
>     'comma-separate these values'
> and, since 'separate' equates to join, this equates to:
>     'comma-join these values'
> so, I code:
>     comma.join(these_values)
> *WHAT'S BACKWARDS ABOUT IT*?!

Well thats fine from your point of view, but to me you've just argued
against your point of basing programming around natural language
(although I may have misconstrued your argument).

>From my POV, a method is something that logically "sends a message" to its
object.  In this context, having the joiner be the object is backwards,
because it makes more sense to me to think in terms of:
"send message <join members with separator parameter and return string> to
a sequence"
rather than:
"send message <use self as a separator when joining members of a sequence
parameter and return string> to a string".

But then as a long time C & FORTRAN programmer, I have difficulty with
knowing when an object design is a good one...

--
Andrew I MacIntyre                     "These thoughts are mine alone..."
E-mail: andrew.macintyre at aba.gov.au    (work) | Snail: PO Box 370
        andymac at bullseye.apana.org.au  (play) |        Belconnen  ACT  2616
        andymac at pcug.org.au           (play2) |        Australia




From tim.one at home.com  Sun Mar  4 23:29:09 2001
From: tim.one at home.com (Tim Peters)
Date: Sun, 4 Mar 2001 23:29:09 -0500
Subject: suggestions for more docs (was Re: Bug fix releases)
In-Reply-To: <3AA2EDBC.E933E78E@seebelow.org>
Message-ID: <LNBBLJKPBEHFEDALKOLCOEOIJDAA.tim.one@home.com>

[Grant Griffin]
> ...
> From the preceeding, I'm gonna deduce that the primary excitement the
> PythonLabs team gets is implementing new language features.

I wouldn't know -- I haven't had time to implement a new feature since I
started this job <0.5 wink>.

> But what about writing more documentation?--*that's* pretty exciting!
> In fact, as one who publishes software myself, I can attest to the
> fact that it's even _more_ exciting than implementing new
> features. <wink>
>
> [followed by excellent suggestions for doc projects]

A point may be getting missed here.  PythonLabs isn't especially looking for
new projects to soak up time; to the contrary, we're already overloaded and
are looking to shed Python work.  So new projects require volunteers, be it
additional docs or additional bugfix releases.  That isn't a reflection on
the value of new projects, it's just the reality of how much sheer grunt work
there is in keeping an every-platform-on-earth language + extensive library
in good working order.

As is, we owe DC a major chunk of our time and are behind on our commitments
to them.  So the amount of work we can devote to core Python is actually
going to decrease significantly and soon.

I don't know whether that's good news or bad news for Python's users.  It
could be good news, if 5x the number of volunteers pop up as are needed just
to fill the gap; or it could be bad news, if it falls in the other direction.

great-ideas-are-a-nickel-a-hundred-but-a-competent-masochist-who-
    never-sleeps-is-rare-as-size-11-fairy-earlobes<wink>-ly y'rs  - tim




From qrczak at knm.org.pl  Mon Mar 19 04:01:22 2001
From: qrczak at knm.org.pl (Marcin 'Qrczak' Kowalczyk)
Date: 19 Mar 2001 09:01:22 GMT
Subject: Who's minister of propaganda this week?
References: <B98041329020E007.7DA84CAD9D0562AB.817BBC572F480869@lp.airnews.net> <98nu1m$12e@dispatch.concentric.net> <660ACCA1E4816F7E.C359CDF4168EF3FB.3B6B05538D619770@lp.airnews.net> <3AAFA13D.76E32C3F@divalsim.it> <98oa4o01i4v@news2.newsguy.com> <mailman.984619039.7856.python-list@python.org> <98q9h60122v@news2.newsguy.com> <mailman.984937946.5318.python-list@python.org> <9937en0bge@news2.newsguy.com> <slrn9bagbm.toq.qrczak@qrnik.zagroda> <994ab801h5o@news2.newsguy.com>
Message-ID: <slrn9bbin2.11r.qrczak@qrnik.zagroda>

Mon, 19 Mar 2001 07:53:13 +0100, Alex Martelli <aleaxit at yahoo.com> pisze:

> But if the container has a list of typed references, they have to
> be typed to SOMETHING (which can be either a leaf of any
> kind or another nested container).

In Haskell's and Clean's syntax:
    type Container a = [Something a]
    data Something a = Leaf a | Nested (Container a)

In OCaml's syntax:
    type 'a container = 'a something list
    and  'a something = Leaf of 'a | Nested of 'a container

> The classic Design Pattern uses just such a 'certain interface'
> as I just described.

Design patterns designed for C++ and Java are not applicable to
languages which use algebraic types and don't use subtyping. C++
and Java don't have a convenient way to express "either this or that
or these" (wrapping each case in a class is not convenient enough).
OTOH Haskell and OCaml don't work by implicit coercion of elements
to a common supertype.

Note that in these languages it's meaningless to mix all types of
objects together in a container. If they are unrelated, you can't
apply any operation to an unspecified member of them.

Tell what do you really want to archieve. The whole thing looks like
a poor fit to the style of these languages. You don't want to have a
net of containers of everything. For example if they are registered
to let somebody do something with them, you would rather register
"do something with them" actions instead of objects themselves.
Objects are rarely mutable so even this looks very suspicious.

> So what (single) type do you make your container's references?

"Something a", where a is the type of objects in leafs.

-- 
 __("<  Marcin Kowalczyk * qrczak at knm.org.pl http://qrczak.ids.net.pl/
 \__/
  ^^                      SYGNATURA ZAST?PCZA
QRCZAK


From carmstro at dynup.net  Sun Mar  4 08:32:37 2001
From: carmstro at dynup.net (Chris Armstrong)
Date: Sun, 04 Mar 2001 18:32:37 +0500
Subject: screen + sound access
References: <3AA2BBBC.760DE6CB@yahoo.fr>
Message-ID: <ta5k4gm41pguc9@corp.supernews.com>

Check out PyGame (http://pygame.seul.org) - it's a really nice
interface to SDL, which is a cross-platform media library.

For documentation and books, check out www.python.org/doc/

In article <3AA2BBBC.760DE6CB at yahoo.fr>, "Francois Forest"
<francois_forest at yahoo.fr> wrote:

> I plan to teach programming to teen agers and would need drawing
> capabilities as well as paying sounds. Would Python be a proper
> candidate ? Is there any book or tutorial about Python and drawing or
> Python and accessing a device ?
> 


-- 
Chris Armstrong   http://twistedmatrix.com/~carmstro     carmstro at dynup.net
There is a 90% chance that this message was written when the author's been
awake longer than he should have. Please disregard any senseless drivel.


From xeno at bigger.aa.net  Sat Mar 10 22:12:58 2001
From: xeno at bigger.aa.net (Xeno Campanoli)
Date: 10 Mar 2001 19:12:58 -0800
Subject: LWP session problems
References: <3aa9c338$1_3@huge.aa.net> <98eo6h$k3j$1@clematis.singnet.com.sg>
Message-ID: <3aaaed3a$1_3@huge.aa.net>

Ng Pheng Siong <ngps at madcap.dyndns.org> wrote:

Okay, I'll take a look at it.  Thanks.


From matso at dtek.chalmers.se  Sun Mar 25 11:37:28 2001
From: matso at dtek.chalmers.se (Mats Olsson)
Date: 25 Mar 2001 16:37:28 GMT
Subject: Collection interfaces
References: <3a90fa97.0@news1.mweb.co.za> <MPG.15085517debf4b649896ca@news.earthlink.net> <zb%n6.31033$o%6.586026@news1.rdc1.bc.home.com> <o1jv6.9635$P4.789083@newsread1.prod.itd.earthlink.net>
Message-ID: <99l6s8$gpu$1@nyheter.chalmers.se>

In article <o1jv6.9635$P4.789083 at newsread1.prod.itd.earthlink.net>,
topmind <topmind at technologist.com> wrote:
>
>The last thing I need is yet more foriegn software on my PC making it crash
>more and confusing other DLL's.

    Yea, use a Java applet instead. Guaranted not to fiddle with your
DLL's :-)

    /Mats


From aleaxit at yahoo.com  Fri Mar  9 10:41:39 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 9 Mar 2001 16:41:39 +0100
Subject: Can anyone recomend a good intoduction to C...
References: <mailman.983894288.19965.python-list@python.org> <3AA5EF59.507F1D91@olsen.ch> <983958878.976344@newsmaster-04.atnet.at> <77up6.1289$y6.235541@ruti.visi.com> <985vqs01ipr@drn.newsguy.com> <ewwp6.1466$y6.241005@ruti.visi.com> <slrn9aeblb.cg1.scarblac@pino.selwerd.nl> <djagat4kk3cs9hg0c3k39cudk6goblhkg6@4ax.com>
Message-ID: <98atjg025pn@news2.newsguy.com>

"Courageous" <jkraska1 at san.rr.com> wrote in message
news:djagat4kk3cs9hg0c3k39cudk6goblhkg6 at 4ax.com...
>
> >All this bloat added to C++ was not necessary for making an OO language.
>
> Echo that. What's "object oriented" about templates?
>
> A: NOTHING.

True.  They're just the best programming concept to hit
statically-typed languages since sliced bread, and happen
to work with superb sinergy with OO and overloading, but
that doesn't make them 'object-oriented' -- just GREAT.

There's a lot of parallelism between generic programming
(via templates in a statically-typed language) and the
way one designs and codes in Python (or other dynamically
typed languages with polymorphism, I guess), except of
course that in the the static case everything really
interesting happens at compile-time (which has pluses
_and_ minuses).


Alex





From root at rainerdeyke.com  Sun Mar 11 15:50:11 2001
From: root at rainerdeyke.com (Rainer Deyke)
Date: Sun, 11 Mar 2001 20:50:11 GMT
Subject: Static typing (was Re: Java guy interested in Python)
References: <mailman.984334635.27345.python-list@python.org>
Message-ID: <7yRq6.396886$ge4.138738128@news2.rdc2.tx.home.com>

"Tim Peters" <tim.one at home.com> wrote in message
news:mailman.984334635.27345.python-list at python.org...
> [Rainer Deyke]
> > When I type 'int i;' in C++, what I really mean is "'i' behaves like an
> > integer variable.".
>
> Since that's all "int i;" *can* mean in C++, seems tautological <wink>.

My mental model of what I'm trying to do is in terms of behavior, not sets
of values.  C++ merely reflects that.

> > I'm not saying anything about the values 'i' can take,
> > except that the values must support integer semantics.
>
> You're also explicitly agreeing to let the implementation decide which
subset
> of all the integers you're allowed to use, and are insisting on signed
ints,
> both of which say quite a bit about the values 'i' can take.  That this is
> inadequate in practice is why C99 goes on to define a few dozen related
> typdefs (int_least32_t, int_fast32_t, int32_t, etc).  It's still often
(not
> always) just approximating the set of values you really have in mind.

'int' is basically short for 'int_fast16_t'.

> > The limited range of type 'int' is part of its defined semantics,
> > as is its fast performance.
>
> I expect you'll search in vain for a promise in the std that ints are
"fast"
> <wink>.

The intention, if not the specification, of 'int' is "the fastest
representation of an integer on the target machine that can respresent the
range from -(2**15) to 2**15, inclusive".

> > If I want to restrict 'i' to a subset of its possible values, I use an
> > assertion.
>
> While if you were programming in Pascal instead, you could express
subranges
> directly in *its* type system; if you were programming in SLIM, you could
> express concepts such as that a variable is restricted to even or prime
> integer values; etc.  You use an assertion in C++ or *any other language*
> because its type system isn't rich enough to capture all you want to say,
not
> because type systems are inherently incapable of capturing more than how
big
> machine registers are <0.5 wink>.

Actually I could express all of these things in the type system of C++
through the use of template classes.  That would be a hack.  There are
assertion which cannot be built into the type system because they are
independent of type.  Given that, it makes no sense to try to model sets of
possible values through types.  In C++ (like in Python), types are intended
to express behavior.


--
Rainer Deyke (root at rainerdeyke.com)
Shareware computer games           -           http://rainerdeyke.com
"In ihren Reihen zu stehen heisst unter Feinden zu kaempfen" - Abigor




From aleaxit at yahoo.com  Thu Mar 29 04:07:14 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 29 Mar 2001 11:07:14 +0200
Subject: Is inheritance broken?
References: <mailman.985808317.25055.python-list@python.org> <3AC25FB7.7FD8759D@accessforall.nl>
Message-ID: <99uu0501u4h@news2.newsguy.com>

"Ype Kingma" <ykingma at accessforall.nl> wrote in message
news:3AC25FB7.7FD8759D at accessforall.nl...
    [snip]
> This code follows a rule: never instantiate a class that
> is inherited from. (In other languages your 'flat' classes
> might be called 'abstract' in order to indicate that they cannot
> be instantiated.)
> IIRC this is a design pattern in class hierarchies.

I agree.  So does C++ guru Scott Meyers, who recommends
the same rule for C++ design -- whenever you find yourself
with two instantiable classes inheriting from each other,
refactor to extract their commonality as an abstract class.

[It's also consistent with the concept of _taxonomy_ --
'leaves' of a taxonomy tree (concrete-classes in its
equivalent object model) should be a partition, not just
a cover; of course, taxonomies have their own issues,
related to those of single-inheritance constraints in OO].

Inheriting-from-concrete-classes is "pragmatically OK" in
small-enough subsystems, but mercilessly refactoring to
abstract the reusable abstractions is an excellent principle
for 'programming in the large'.  Adopting in-the-large ideas
for moderately-sized subsystems is not a must, but it may
serve you well if and when a subsystem unexpectedly grows.


Alex





From mcfletch at home.com  Tue Mar  6 09:46:53 2001
From: mcfletch at home.com (Mike C. Fletcher)
Date: Tue, 6 Mar 2001 09:46:53 -0500
Subject: I come to praise .join, not to bury it...
In-Reply-To: <kuofvfjb77.fsf@lasipalatsi.fi>
Message-ID: <000001c0a64c$497df250$a9807018@cr706570a>

Here's a professional aesthetic's take:

	mystring.join( )

This is a perfectly normal and good-looking construct.  It seems pythonic
and beautiful, object orientation is part of Python, and this works nicely.

	"'".join( )
	";".join( )
	".".join( )
	"!".join( )
	",".join( )

All look like executable line noise.  That is, the literal string's syntax
makes the use of dotted attributes on such a string look jarring and
off-putting.  You almost expect some weird function to jump out of this
syntax.  It looks very Ruby/Perl-esque.

Note particularly the amount of space between the " character and the .
character.  Typographically, " characters serve to set the enclosing data
off from the outer context, and that's the opposite of what you're trying to
do with dot notation.  Dot notation normally looks natural because almost
every character that gets used right before the . comes fairly close to the
.  (even characters such as r ).  Also note: in the common cases of
single-character joins, there's no significant visual weight difference
between what's inside and outside the quotes).  Consider what happens when
you alter the proportions of space and the size of the connector...

	"."__join( )
	"."-->join( )
	"."..join( )
	" "__join( )

You should find those slightly easier on the eyes (except possibly the " "
one) because the entire "." entity can be understood as a single graphic
with a connector to "join".

That said, I don't think we'll get this particular wart out of the language,
so we'll have to deal with the flaw in the way artists have through the
ages, enhance it and accent it so people start to think it's a beauty mark.
Mike




From immanuel at yours.com  Tue Mar 20 09:34:25 2001
From: immanuel at yours.com (Immanuel)
Date: Tue, 20 Mar 2001 06:34:25 -0800
Subject: pdb with threads & PyEval_CallObject
Message-ID: <ZkLt6.1$Fk4.2000@news.pacbell.net>

I have a Linux multi-threaded application which both extend & embed Python.
The threads are created in C++, and the entry point of the application is
started via
Python; one of these threads, however, have a PyEval_CallObject in it. Since
I need
to acquire the interpreter lock before calling PyEval_CallObject, this means
that pdb
is unable to break at the callback function. Does anyone know how to get
around this?
The alternative to debugging this is, of course, good old prints.

Thanks.
  Immanuel

Code snippet:

=======================
"main" python thread:

def cbfunc(arg):
    # callback func
    ...

app.start()         # threads created
app.installCallback(cbfunc)
...

=======================
"embedded" thread:
tstate = PyThreadState_New(interp)
Py_AcquireThread(tstate)
...
PyEval_CallObject(cbfunc)
....
PyThreadState_Clear(tstate)
Py_ReleaseThread(tstate)
PyThreadState_Delete(tstate)









From dougfort at downright.com  Thu Mar 29 07:18:47 2001
From: dougfort at downright.com (Doug Fort)
Date: Thu, 29 Mar 2001 07:18:47 -0500
Subject: asyncore
References: <3AC316CB.597E85F3@utbm.fr>
Message-ID: <3AC32827.67B138F1@downright.com>

Matias GUIJARRO wrote:

> - The server must be always listenning for new connections, so I call
> the
> asyncore.loop() method, but it seems to be "blocking" : the code after
> isn't
> executed. How can I make an "asyncronous" listenner ?

If your sockets are blocking, I suspect that you didn't use
asyncore.create_socket() to create them.  If so, try that.

You might want to  take a look at asynchat.py as well.  It doesn't seem to
be in the Python Library documentation, but it uses asyncore and handles a
lot of the dirty work.

You might also get some use out of asynchttp which uses asyncore and
asynchat.  http://asynchttp.sourceforge.net

--

Doug Fort (dougfort at downright.com)
Senior Meat Manager
Downright Software LLC
http://www.dougfort.net





From bbhaydon at bigpond.com  Sat Mar 24 01:00:50 2001
From: bbhaydon at bigpond.com (Brett Haydon)
Date: Sat, 24 Mar 2001 17:00:50 +1100
Subject: Please rate these Python books
References: <20010322.1445209@mis.configured.host> <_Meu6.8736$8O6.1818817@typhoon.austin.rr.com> <slrn9bjuha.4qv.grey@teleute.dmiyu.org> <blzu6.8974$G72.207211@e420r-atl1.usenetserver.com>
Message-ID: <vwWu6.2238$45.10474@newsfeeds.bigpond.com>

Snap!
> Prime collection is currently "Learning Python" (despite its age), "Win32
> Programming in Python" and "Python Essential Reference". The rest have
their
> place, but it's in the bookshelf, not the briefcase.

though haven't really used Python Essential Ref much..
Looking foward to the new OReilly Python book, though I fear fitting it
in the briefcase may be out of the question.

Brett




From piet at cs.uu.nl  Thu Mar 22 15:43:29 2001
From: piet at cs.uu.nl (Piet van Oostrum)
Date: 22 Mar 2001 21:43:29 +0100
Subject: New PEP: The directive statement
References: <200103201042.LAA20248@pandora.informatik.hu-berlin.de> <mailman.985107140.4586.python-list@python.org> <9988ie02k7u@drn.newsguy.com> <ulmpzjrpb.fsf@cs.uu.nl> <mailman.985222709.15240.python-list@python.org>
Message-ID: <wzr8zpwn4e.fsf@sunshine.cs.uu.nl>

>>>>> Neil Schemenauer <nas at arctrix.com> (NS) writes:

NS> On Wed, Mar 21, 2001 at 12:18:24PM +0100, piet at cs.uu.nl wrote:
>> It would be much less of a joke when __future__ would be replaced by
>> __features__, I think.

NS> __features__ is a terrible name.  Its pretty obvious that you
NS> (and a lot of other people on this list) don't understand what
NS> __future__ does.

NS> A small hint: pretend you are using 2.2 and want to use nested
NS> scopes.  Will you be using "from __future__ import
NS> nested_scopes"?  What if you are using 2.2 and don't want nested
NS> scopes?

If the suggestion is that after 2.2. you would remove the "from __future__
import ..." then there will be problems. E.g. consider the integer
division. For the transitional releases you would have to add a
future_statement. But if you remove it after the feature has become
standard your new modules that use the new division semantics become
ambiguous. Suppose you have to use Python 1.5.2. or 2.0 because that's all
your ISP provides. And you have this nifty module that someone wrote. How
do you know what the semantics of division is in that module? It would be
much better if the module indicated that it uses the new semantics, for the
ages to come.
-- 
Piet van Oostrum <piet at cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: P.van.Oostrum at hccnet.nl


From sandipan at vsnl.com  Sat Mar 10 15:08:36 2001
From: sandipan at vsnl.com (Sandipan Gangopadhyay)
Date: Sun, 11 Mar 2001 01:38:36 +0530
Subject: CGISSL
References: <mailman.984248593.17249.python-list@python.org> <3AAA7E1B.B411E6AE@stroeder.com>
Message-ID: <004f01c0a99d$e4811bc0$020000c3@node02>

cgissl.py is part of your web2ldap. Thanks for the great software. The
reason that I asked is this:
1. I have an Apache1.3.9 with Mod_SSL
2. It has a certificate .key and .crt installed. The .crt is signed by a
Root CA, called (say) XXX.
3. On the other hand, I have an MSIE5.5 browser that has a Digital ID
installed.
4. This digital ID was generated with Xenroll and signed by the same Root CA
XXX.

5. When a cgi program on the server is accessed by the client, the following
method of cgissl is returning 1 (server side SSL only). I want it to be 2
(client side as well). I want to find out how...

sec_sslacceptedciphers = [
    'IDEA-CBC-SHA',
    'RC4-MD5',
    'EXP-RC4-MD5',
    'RC4-SHA',
    'IDEA-CBC-MD5',
    'DES-CBC3-SHA',
    'DH-DSS-DES-CBC3-SHA',
    'DH-RSA-DES-CBC3-SHA',
    'EDH-DSS-DES-CBC3-SHA',
    'EDH-RSA-DES-CBC3-SHA',
    'ADH-RC4-MD5',
    'ADH-DES-CBC3-SHA',
    'FZA-RC4-SHA',
    'RC2-CBC-MD5',
    'DES-CBC3-MD5'
  ]
intLevel = cgissl.SecLevel(sec_sslacceptedciphers, '/.*', '/.*')

I am only interested in ldap_clientdn2binddn = ['CN'] for now.

My OpenSSL is 0.9.4

What I am basically asking is how do I ensure that Apache and IE negotiate
the client cert as well as the server cert in the HTTPS handshake ?

Is the problem because IE is not using the Digital ID at all ?
Is the problem because Apache cant find the Root CA certificate to match the
signer/issuer of the Digital ID ?
Is the regexp failing ?
How do I find out ?

Thanks for your fast response,

Sandipan

----- Original Message -----
From: "Michael Str?der" <michael at stroeder.com>
To: "Sandipan Gangopadhyay" <sandipan at vsnl.com>
Sent: Sunday, March 11, 2001 12:48 AM
Subject: Re: CGISSL


> Sandipan Gangopadhyay wrote:
> >
> > Is the cgissl module documentation available online ?
>
> Which module do you mean?
>
> Ciao, Michael.
>




From ws-news at gmx.at  Tue Mar  6 15:44:30 2001
From: ws-news at gmx.at (Werner Schiendl)
Date: Tue, 6 Mar 2001 21:44:30 +0100
Subject: Can anyone recomend a good intoduction to C...
References: <mailman.983894288.19965.python-list@python.org> <983896344.701456@newsmaster-04.atnet.at> <m3lmqibw0a.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <983911749.217622@newsmaster-04.atnet.at>

Michael Hudson <mwh21 at cam.ac.uk> wrote in message
news:m3lmqibw0a.fsf at atrus.jesus.cam.ac.uk...
> "Werner Schiendl" <ws-news at gmx.at> writes:
>
> > Hi Laura,
> >
> > The only reason I can imagine at this point to go with plain C is that
there
> > are a lot more platforms that support C than do support C++. This is
mainly
> > a topic in embedded development and realtime platforms.
> [...]
> > Did I miss an important point?
>
> Sprinkle IMHO's throughout what follows.
>
> Yes, you're missing the fact that C++ sucks large rocks through very
> small straws.  There's some merit in learning C for when low-levelness
> is wanted, eg. when developing Python.  If you're writing an

I do not see the point, why not use C++ for low-levelness.
Years ago I had a lot of discussions why use C instead of Assembler.
The arguments were mainly for performance.
And still C made the race, guess why?
It is easier, more powerful, more safe (the compiler can catch more bugs
then the assembler)

> application, then you might think that C++ would be a good language to
> use.  You'd be wrong (FWIW, you'd be wrong to use C here, too).  Use a
> sane programming language like Python or Common Lisp (and if you use
> Python, then you can write portions in C - but leave the application
> logic out of it!).

If you have such an language AND your customer accepts that you use it you
are lucky.
Go with python or whatever you like.

Maybe I'm a bit influenced from my industrial programming background.
An RTOS or embedded ?C usually doe not come with python or lisp or such.

But you always have C, and increasingly often C++.
And if this is the choice, I would take C++.

That's all I wanted say with my original post.
I would do an python extension for sure in C++ rather than in C.
The much more powerful libs rectify that on their own.

>
> One of my more recent additions to my sig collection is:
>
>   C++ is a siren song.  It *looks* like a HLL in which you ought to
>   be able to write an application, but it really isn't.
>                                        -- Alain Picard, comp.lang.lisp
>
> I hope everyone noticed the:
>
> Sprinkle IMHO's throughout what follows.
>
> at the top of this article.  My feelings for C++ oscillate between
> mild dislike and outright searing hatred - and I have programmed in
> it, I've formed my opinion on experience, and I think it sucks.  You
> may have a different opinion, but this is mine.
>

Of course your opinion is welcome!
I hope mine is as well...

> Cheers,
> M.
>
> --
>   There's a difference between random people with stripy jumpers, and
>   a respected scientist with a reputation.
>                                             -- Steve Kitson, ucam.chat

regards
werner




From ws-news at gmx.at  Tue Mar  6 12:06:20 2001
From: ws-news at gmx.at (Werner Schiendl)
Date: Tue, 6 Mar 2001 18:06:20 +0100
Subject: Can Python replace Visual Basic? Should it?
References: <5DZo6.11126$hn5.1578410@news1.rdc1.mb.home.com> <98266k0p5h@news1.newsguy.com> <3q8p6.12338$hn5.1798569@news1.rdc1.mb.home.com>
Message-ID: <983898657.328387@newsmaster-04.atnet.at>

I never tried to develop a GUI with Python, so i cannot comment on how easy
or difficult that would be.

But I think Visual Basic is really great for such simple toy programs that
do not do much but are necessary to get the customer satisfied. And they (of
course) fit perfectly the native look and feel of windows and support all
kind of fancy things like OLE Drag & Drop. That does not mean that i think
this is necessary to make the tool good, but there are people asking for
such stuff and than its getting hard if your environment forces you to do
that by hand.

In general I agree that one should know how tha mechanics of the program
works. But I think it is not necessary to know how Visual Basic registers
Active X Controls in detail. It works and I know how to invoke it, what's
the problem? I do not know in detail how the .EXE file is made up or how my
harddisk handles CRCs. And still the programs work fine...

For a small, one or two window GUI with a fixed layout Visual Basic will
stay my favorite tool for the next time I guess.
But, of course, everyone may use the tools he likes best.

regards
werner

Brad Bollenbach <bbollenbach at homenospam.com> wrote in message
news:3q8p6.12338$hn5.1798569 at news1.rdc1.mb.home.com...
>
> "Alex Martelli" <aleaxit at yahoo.com> wrote in message
> news:98266k0p5h at news1.newsguy.com...
> > "Brad Bollenbach" <bbollenbach at homenospam.com> wrote in message
> > news:5DZo6.11126$hn5.1578410 at news1.rdc1.mb.home.com...
> > > Again, I'm much more interested in the real world experience of those
> who
> > > used to use the big visual development tools in Windows like VB,
Delphi
> or
> > > Powerbuilder rather than theoretical answers.
> >
> > Why do you focus on 'used to use'?  One may perfectly well keep using
> > some VB (&c) for tasks it's particularly well-suited to, while using
> > Python wherever feasible (and, indeed, using both in the same project
>
> Certainly. And for all intents and purposes "used to use" isn't necessary,
> but the question happened to be directed (as its subject suggests) at
those
> who _have_ replaced VB with Python. :)
>
> But realistically, I just wanted to emphasize my interest in hearing
answers
> from those who have or currently are using VB or one of the other Win32
> visual development tools. Personally, I'm a former VB programmer now
working
> in Powerbuilder and am unhappy with both, again based on my dislike of
IDE's
> that obscure the programmer's view of what's going on.
>
> I could explain _why_ I don't like the 40,000 foot views of IDE's, but
> that's another thread.
>
> Brad
>
>




From claird at starbase.neosoft.com  Wed Mar 14 13:40:02 2001
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 14 Mar 2001 12:40:02 -0600
Subject: Who's minister of propaganda this week?
References: <mailman.984591979.3013.python-list@python.org>
Message-ID: <888BC9CCDC831605.DFA63C543972E752.F70781F6998E0A03@lp.airnews.net>

In article <mailman.984591979.3013.python-list at python.org>,
Steven D. Majewski <sdm7g at virginia.edu> wrote:
			.
			.
			.
>An object is a piece of data with procedures attached to it.
>A closure is a procedure with a piece of data attached to it. 
>
>Obviously, they are complete opposites! 
>
>Well -- maybe "mirror images" is a better term.
Mathematicians talk of "duals".  In signal processing,
for example, one considers a voice recording--or its
"frequency graph" (periodogram--I'm simplifying, of
course).  An important instinct for all developers is
to think, when they see code (data), how to regard it
as data (code).
>
>If you read the HOPL articles on Lisp and Smalltalk, it's
>clear that both groups were in communication with each other
>and were aware of this -- they called it "orthigonality" --
>both a different syntax, and a different way of looking at the 
>same thing. 
Worth repeating.
>
>
>-- Steve Majewski
>
>


-- 

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


From sholden at holdenweb.com  Thu Mar  8 22:28:59 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 8 Mar 2001 22:28:59 -0500
Subject: gah! I hate the new string syntax
References: <mailman.983615170.23313.python-list@python.org> <tN5o6.2746$eH3.14913@NewsReader> <Koco6.7518$1D5.244602@e420r-atl1.usenetserver.com> <mailman.983997252.5628.python-list@python.org> <m2vgpkyfjo.fsf@mycroft.actrix.gen.nz> <mailman.984063199.18779.python-list@python.org>
Message-ID: <s9Yp6.42541$ML1.1889131@e420r-atl2.usenetserver.com>

"D-Man" <dsh8290 at rit.edu> wrote in message
news:mailman.984063199.18779.python-list at python.org...
> On Thu, Mar 08, 2001 at 06:51:07PM +1300, Paul Foley wrote:
> | On Wed, 7 Mar 2001 15:33:30 -0500, D-Man  wrote:
> |
> | > What about one who is both English and Scots, yet never set foot on
> | > that island ;-)?
> |
> | AFAIK, nobody outside the US would admit the existence of such a
> | thing.  The American habit of calling themselves "Irish" or "Italian"
> | or whatever when they're not is *very* strange.  I've never heard
> | anything like it anywhere else.
>
> Heh.  Maybe Americans are very odd.  Wouldn't surprise me at all.  ;-)
> I was born in America and have lived in America all my life.  I do
> have English, Scots, Irish and Welsh ancestry though.  One could argue
> that I am not really American since none of my ancestors came from
> America.  "Real" Americans would be the so-called "Indians", or more
> appropriately "Native Americans".  (Also Aztecs, Inca, etc., and their
> descendants such as the Quiche)
>
> -D
>
QED. You are neither English nor Scottish. But since you're American, and
therefore very odd, this probably won't bother you ;-)

regards
 Steve




From lance.hill at removespamfilter.olsinc.net  Fri Mar 23 16:07:15 2001
From: lance.hill at removespamfilter.olsinc.net (lance.hill at removespamfilter.olsinc.net)
Date: Fri, 23 Mar 2001 21:07:15 GMT
Subject: XML/XSLT for newbies
Message-ID: <3abbbb0f.69122775@nnrp.wiznet.ca>

All the documentation I have been able to find on how to do xsl
transformation assumes you have ASP,VB,JAVA or such. Does anyone know
of learning resources for xml/xsl in python?

In particular, I would like to be able to do things such as display a
graphic image depending on the text value of an element.

All help greatly appreciated.

L. Hill


From paulp at ActiveState.com  Sun Mar 11 14:19:24 2001
From: paulp at ActiveState.com (Paul Prescod)
Date: Sun, 11 Mar 2001 11:19:24 -0800
Subject: Questions about no more nested imports
References: <98gi4l$n9n$0@216.39.170.247>
Message-ID: <3AABCFBC.D164CDB3@ActiveState.com>

Dave LeBlanc wrote:
> 
> I'm a bit confused by this.
> 
> It appears that if this is enforced, one can no longer dynamically
> load extensions/whatever dynamically and all imports must be done at
> the top of a module prior to anticipated use - whether used or not. If
> this is true, then load times for python apps could go up drastically.

Nobody has proposed that nested imports be disallowed. The only form
that is disallowed is a nested form of "from X import *".

-- 
Python:
    Programming the way
    Guido
    indented it.
       - (originated with Skip Montanaro?)



From jschmitt at vmlabs.com  Fri Mar 23 00:55:19 2001
From: jschmitt at vmlabs.com (John Schmitt)
Date: Thu, 22 Mar 2001 21:55:19 -0800
Subject: Good Editor? (What's the big deal with syntax colouring?)
Message-ID: <008F0A63472BD311AF9800104BCD102561CD45@minirex.vmlabs.com>

I do typos all the time.  But they very rarely take time to debug because
the compiler will tell me about those instantly.  I spend much more time
with bugs related to poor design or poor code or broken hardware.  

Anyway, the syntax colouring that I've seen also differentiates between
comments, keywords, expressions, literal numerals, literal strings,
pre-processor directives, etc.  Do any of them require, I don't know, more
attention than others?  To me they're equally significant.  Does syntax
colouring move your attention to something important?  Other than typos, I
mean - I can kind of see the value in emphasizing typos.  I also find it
distracting if the editor changes colours on me if I'm adding or deleting
quotes or if the cursor jumps around because I'm adding or deleting
brackets.  My attention span is too short already.

John

-----Original Message-----
From: Steve Holden [mailto:sholden at holdenweb.com]
Sent: Thursday, March 22, 2001 8:24 PM
To: python-list at python.org
Subject: Re: Good Editor? (What's the big deal with syntax colouring?)


"Timothy Grant" <tjg at exceptionalminds.com> wrote in message
news:mailman.985295730.14145.python-list at python.org...
> On Thu, Mar 22, 2001 at 12:10:59PM -0800, John Schmitt wrote:
> > Why do people want syntax colouring?  I can't stand it.  One of my minor
> > beefs with CodeWright is that there isn't a one button way to tell it to
> > change _all_ of my colours all at once.  Am I missing something by
disabling
> > syntax colouring?  It always reminds me of the artwork my 6-year-old
nephew
> > brings home from school to show his mommy.
>
> Hey, I'm colour blind, and I still need syntax highlighting
> though I can't see all of the colour differenctials.
>
> It's pretty simple. Forget to close a ("|')--or open one for
> that matter, and you automatically know somethings wrong
> because the colours are wrong, misspell(sp?) a keyword and you
> know it because it doesn't hi-light in the correct colour.
>
> It is a huge benefit to getting syntax correct before you save
> your file.
>
> Now if only else didn't highlight but else: did<wink>
>
I seem to remember that having to write your source on coding forms, have it
punched up on tape or cards and waiting six hours until the computer center
had deposited the printout in your pigenhole made me similarly careful about
my coding style.

that's-when-i-discovered-semantics-ly y'rs  - steve



-- 
http://mail.python.org/mailman/listinfo/python-list



From chrisw at nipltd.com  Mon Mar 12 07:38:28 2001
From: chrisw at nipltd.com (Chris Withers)
Date: Mon, 12 Mar 2001 12:38:28 +0000
Subject: I have no idea
References: <g13r6.201$qt5.10920@news3.mia>
Message-ID: <3AACC344.3BC212F8@nipltd.com>

Brendhan Horne wrote:
> 
> In basic you use to be able to do this:
> 10 print "Hello World"
> 20 goto 10
> Run
> 
> If you did that it would run up the screen in an endless loop until you hit
> the ESC key. If you added   ; it would cover the whole screen. How do you
> write the equivalent in python?

while 1:
   print "Hello World"

or 

while 1:
   print "Hello World",

Replace ESC with Ctrl-C :-)

cheers,

Chris



From neelk at alum.mit.edu  Tue Mar 13 08:26:50 2001
From: neelk at alum.mit.edu (Neelakantan Krishnaswami)
Date: 13 Mar 2001 13:26:50 GMT
Subject: This math scares me
References: <3aad4062.1290021@News.CIS.DFN.DE> <Hwbr6.1139$Tg.127571@ruti.visi.com> <3aad4485.2348499@News.CIS.DFN.DE> <jwbnews-7CC03F.19050312032001@corp.supernews.com> <yu993dcimjt8.fsf@europa.research.att.com>
Message-ID: <slrn9asbvc.dfq.neelk@alum.mit.edu>

On Tue, 13 Mar 2001 03:25:39 GMT, Andrew Koenig <ark at research.att.com> wrote:
>
>> Many of us feel that the present truthful display is correct, and the 
>> former (before Python 2) display of 10.55 was wrong because it was a lie 
>> (some don't feel quite *that* strongly about it but still prefer the new 
>> form).
>
> I'm curious -- Does Python now follow either the IEEE or the more
> restrictive Scheme rules for floating-point conversion?

Python uses whatever the underlying (usually C) platform provides. It
used to do rounding on the output, and now it doesn't. If you wanted
to implement the algorithms in William Clinger's _Reading Floating
Point Numbers Accurately_ and Byrger and Dybvig's _Printing Floating
Point Numbers Quickly and Accurately_ and submit it as a patch I don't
think anyone would object. :)


Neel


From embed at geocities.com  Wed Mar 14 13:24:10 2001
From: embed at geocities.com (Warren Postma)
Date: Wed, 14 Mar 2001 13:24:10 -0500
Subject: Cameron Laird Quote on www.unixinsider.com -- Probably old news to y'all, but new to me.
Message-ID: <bJOr6.9481$TW.38047@tor-nn1.netcom.ca>

I found this quote on www.unixinsider.com's main page when I went there
today:

"I think we are seeing an end to the emphasis on statically typed languages
like C++ and Java. As this decade progresses, I expect to see an ever
increasing use of dynamically typed languages, such as Python, Ruby, and
even Smalltalk. These languages are often referred to as 'scripting
languages;' I think this is a gross injustice. It is these languages, and
languages of their kind, that will be mainstream industrial languages in the
coming years. They are much easier to refactor and have virtually zero
compile time."
--Robert C. Martin, talking to Cameron Laird in the Interviews forum

---

I apologize for trotting it out again, if I am repeating something earlier
discussed, but I couldn't find any direct reference to it (my nntp server
sucks, only goes back 7 days).

All I'll add is: Well said, Cameron!

Warren Postma





From claird at starbase.neosoft.com  Mon Mar 19 09:50:44 2001
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 19 Mar 2001 08:50:44 -0600
Subject: How to find out about free space on a given drive?
References: <3AB5C19E.FA40DDA9@darwin.in-berlin.de> <mailman.984991890.14766.python-list@python.org>
Message-ID: <CDE3F34783E0C6CB.0DCEA469EF336A5B.984705F2492598CF@lp.airnews.net>

In article <mailman.984991890.14766.python-list at python.org>,
Steve Purcell  <stephen_purcell at yahoo.com> wrote:
>Dinu Gherman wrote:
>> is there a simple way (or any way at all) to find out for 
>> any given hard disk how much free space is left on that
>> device? I looked into the os module, but either not hard
>> enough or there is no such function. Of course, the ideal
>> solution would be platform-independant, too... :)
>
>os.statvfs() seems to be what you want for UNIX. I don't think you'll find
>a cross-platform solution.
			.
			.
			.
When it matters enough to me, maybe I'll write up (partial)
os.statvfs() implementations for Win*, or even MacOS, perhaps
starting from <URL: http://mini.net/cgi-bin/wikit/526.html >
(see "Disk free capacity") and submit the sources for inclu-
sion in the distribution library.  It surprises me a teensy
bit when I read, "I don't think you'll find a cross-platform
solution", without the "yet" which belongs at the end.
-- 

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


From kevin at GamesandTheory.com  Mon Mar  5 03:23:29 2001
From: kevin at GamesandTheory.com (Kevin Douglas)
Date: Mon, 5 Mar 2001 03:23:29 -0500
Subject: Need a free lancer python programmer - to do bots (flowing water and open systems)
In-Reply-To: <4XFo6.19441$1D5.730654@e420r-atl1.usenetserver.com>
Message-ID: <DKEEJHGMIIOMNIJFOCAIIEBMDHAA.kevin@GamesandTheory.com>

hey the list pans out some spanish bullion for me AND some
support - excellent!

maybe you guys won't mind when i promote my forthcoming
album here either... trying to raise money for technology
through non-technology specific markets, to bring more
concentration into the arena - drawing lines in the sand
from one castle to the next..

and yes!  they're channels for water conduits - free flow
is best!

the water of the world is the world's largest open system!

it's all free to go anywhere it wants - yet it all sticks
together!

;)


-kev  <-- creative designer/technologist/MBA-in-progress


> -----Original Message-----
> From: python-list-admin at python.org
> [mailto:python-list-admin at python.org]On Behalf Of Steve Holden
> Sent: Monday, March 05, 2001 12:55 AM
> To: python-list at python.org
> Subject: Re: Need a free lancer python programmer - to do bots
> 
> 
> "Kevin Douglas" <kcocaine at GamesandTheory.com> wrote in message
> news:mailman.983753706.16462.python-list at python.org...
> > crap!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
> >
> > shoot me..
> >
> >
> > -kev
> >
> Bang. Good luck.
> 
> Steve
> 
> 
> -- 
> http://mail.python.org/mailman/listinfo/python-list
> 
> 



From LangaK at discoveryhealth.co.za  Fri Mar  2 04:16:40 2001
From: LangaK at discoveryhealth.co.za (Langa Kentane)
Date: Fri, 2 Mar 2001 11:16:40 +0200 
Subject: playing with IP's
Message-ID: <A69BC9CA4B39D411BF2200105A45B45B0B14ECE4@dhexchange.discoveryhealth.co.za>

Greetings.
Here is what I wanna do:
Given a network number and mask ie:
192.168.1.0/32
I want to create a list with all the valid hosts on that range.
Problem is I don't know how to play with decimal.

Where would I start.. any sample code?
Thanks

_________________________________________________________
Langa Kentane		|	Tel:	[011] 290 3218
Security Administrator	|	Cell:	082 606 1515
[CNA MCSE CCSA CCNA]	|	www.discoveryhealth.co.za
_________________________________________________________



From othello at javanet.com  Sat Mar 24 01:48:57 2001
From: othello at javanet.com (Raymond Hettinger)
Date: Sat, 24 Mar 2001 01:48:57 -0500
Subject: PEP 245: Python interfaces
References: <mailman.985377931.22542.python-list@python.org>
Message-ID: <3ABC4359.F24C5974@javanet.com>

Michel Pelletier wrote:

>
> >    d) What is also missing from Python is
> >       an "auto-delegation".  An analysis
> >       of how this fits in or doesn't fit
> >       in would be helpful.
>
> What is that?  I've neve heard of auto-delegation.
>

class AutoDelegate:
    'Mixin class autodelegates to objects listed in "delegationVars"'
    def __getattr__( self, key ):
        if not self.__dict__.has_key('delegationVars'): delegationVars=[]
        for str in self.delegationVars: # if deleVars not defined, skip
            try:
                return getattr( self.__dict__[str], key )
            except KeyError:
                pass
            except AttributeError:
                pass
        raise AttributeError, key

class Movie( AutoDelegate ):        # installs the AutoDelegate mixin
    delegationVars = ['priceCode']  # list of references to delegates
    def __init__( self, title ):
        self.priceCode = RegularPrice()
        self.title = title
    def changeCode( self, aPriceObj ):
        self.priceCode = aPriceObj
    ##def getPrice( self, qty ):    # Delegated method no longer needed
    ##    self.priceCode.getPrice(qty)

class RegularPrice:
    def getPrice( self, qty ):
        return 10*qty

class DiscountedPrice:
    def getPrice( self, qty ):
        return 10 + 9 * (qty-1)

a = Movie('Rambo')
b = Movie('Aliens')
b.changeCode( DiscountedPrice() )
print a.getPrice(10), b.getPrice(12) # these calls automatically forwarded



From lac at cd.chalmers.se  Thu Mar  8 11:03:57 2001
From: lac at cd.chalmers.se (Laura Creighton)
Date: Thu, 8 Mar 2001 17:03:57 +0100 (MET)
Subject: I got 2.1b to dump core.
Message-ID: <200103081603.RAA17793@boris.cd.chalmers.se>

Here is a traceback.  Should I send this some place other than here?  
Laura
--------------

#0  0x400dc223 in malloc () from /lib/libc.so.6
#1  0x400dc0ce in malloc () from /lib/libc.so.6
#2  0x80927aa in PyString_FromStringAndSize (
    str=0x81c0696 "184684445458205368372152666662828241785498762663290423991282
9087750391537648943213531422278166751524252378540140409965127726779702015558498
1700322047559471699792855379646431352368268044945323567473669"..., size=309) at
 Objects/stringobject.c:60
#3  0x80931a6 in string_slice (a=0x81c0680, i=2, j=311) at Objects/stringobject
.c:517
#4  0x807fab8 in PySequence_GetSlice (s=0x81c0680, i1=2, i2=311) at Objects/abs
tract.c:1040
#5  0x805eaa5 in apply_slice (u=0x81c0680, v=0x80e5180, w=0x81563e0) at Python/
ceval.c:3169
#6  0x805b77c in eval_code2 (co=0x819c910, globals=0x819d3c4, locals=0x0, args=
0x80edf54, argcount=2, kws=0x80edf5c, kwcount=0, defs=0x0, defcount=0, closure=
0x0)
    at Python/ceval.c:1203
#7  0x805e414 in fast_function (func=0x8187d44, pp_stack=0xbffff5b4, n=2, na=2,
 nk=0) at Python/ceval.c:2913
#8  0x805cc61 in eval_code2 (co=0x8120420, globals=0x8130c14, locals=0x0, args=
0x80fd5c8, argcount=3, kws=0x0, kwcount=0, defs=0x0, defcount=0, closure=0x0) a
t Python/ceval.c:1913
#9  0x805e316 in call_eval_code2 (func=0x81b29ec, arg=0x80fd5bc, kw=0x0) at Pyt
hon/ceval.c:2857
#10 0x805df22 in call_object (func=0x81b29ec, arg=0x80fd5bc, kw=0x0) at Python/
ceval.c:2701
#11 0x805e1df in call_method (func=0x81b29ec, arg=0x81b97f4, kw=0x0) at Python/
ceval.c:2814
#12 0x805df05 in call_object (func=0x81768d4, arg=0x81b97f4, kw=0x0) at Python/
ceval.c:2699
#13 0x805debe in PyEval_CallObjectWithKeywords (func=0x81768d4, arg=0x81b97f4, 
kw=0x0) at Python/ceval.c:2673
#14 0x808166e in PyInstance_New (klass=0x81b294c, arg=0x81b97f4, kw=0x0) at Obj
ects/classobject.c:488
#15 0x805df4a in call_object (func=0x81b294c, arg=0x81b97f4, kw=0x0) at Python/
ceval.c:2705
#16 0x805e6d7 in do_call (func=0x81b294c, pp_stack=0xbffff814, na=2, nk=0) at P
ython/ceval.c:3014
#17 0x805cc7f in eval_code2 (co=0x80f2020, globals=0x80e46cc, locals=0x0, args=
0x81ae320, argcount=6, kws=0x0, kwcount=0, defs=0x0, defcount=0, closure=0x0) a
t Python/ceval.c:1916
#18 0x805e316 in call_eval_code2 (func=0x81b2b34, arg=0x81ae314, kw=0x0) at Pyt
hon/ceval.c:2857
#19 0x805df22 in call_object (func=0x81b2b34, arg=0x81ae314, kw=0x0) at Python/
ceval.c:2701
#20 0x805e1df in call_method (func=0x81b2b34, arg=0x81205dc, kw=0x0) at Python/
ceval.c:2814
#21 0x805df05 in call_object (func=0x80ec354, arg=0x81205dc, kw=0x0) at Python/
ceval.c:2699
#22 0x805debe in PyEval_CallObjectWithKeywords (func=0x80ec354, arg=0x81205dc, 
kw=0x0) at Python/ceval.c:2673
#23 0x808166e in PyInstance_New (klass=0x81b2a94, arg=0x81205dc, kw=0x0) at Obj
ects/classobject.c:488
#24 0x805df4a in call_object (func=0x81b2a94, arg=0x81205dc, kw=0x0) at Python/
ceval.c:2705
#25 0x805e6d7 in do_call (func=0x81b2a94, pp_stack=0xbffffa74, na=5, nk=0) at P
ython/ceval.c:3014
#26 0x805cc7f in eval_code2 (co=0x8121110, globals=0x80e6fe4, locals=0x80e6fe4,
 args=0x0, argcount=0, kws=0x0, kwcount=0, defs=0x0, defcount=0, closure=0x0) a
t Python/ceval.c:1916
#27 0x805a349 in PyEval_EvalCode (co=0x8121110, globals=0x80e6fe4, locals=0x80e
6fe4) at Python/ceval.c:336
#28 0x8071467 in run_node (n=0x80f0d20, filename=0xbffffdb6 "./ifc.pys", global
s=0x80e6fe4, locals=0x80e6fe4, flags=0x0) at Python/pythonrun.c:957
#29 0x8071422 in run_err_node (n=0x80f0d20, filename=0xbffffdb6 "./ifc.pys", gl
obals=0x80e6fe4, locals=0x80e6fe4, flags=0x0) at Python/pythonrun.c:943
#30 0x80713f5 in PyRun_FileEx (fp=0x80e2e58, filename=0xbffffdb6 "./ifc.pys", s
tart=257, globals=0x80e6fe4, locals=0x80e6fe4, closeit=1) at Python/pythonrun.c
:934
#31 0x8070a9b in PyRun_SimpleFileEx (fp=0x80e2e58, filename=0xbffffdb6 "./ifc.p
ys", closeit=1) at Python/pythonrun.c:635
#32 0x807063b in PyRun_AnyFileEx (fp=0x80e2e58, filename=0xbffffdb6 "./ifc.pys"
, closeit=1) at Python/pythonrun.c:465
#33 0x8057443 in Py_Main (argc=2, argv=0xbffffca4) at Modules/main.c:312
#34 0x8056f18 in main (argc=2, argv=0xbffffca4) at Modules/python.c:10




From bogus at ddress.com  Tue Mar 13 10:43:29 2001
From: bogus at ddress.com (J. A. Durieux (reply from http://www.biep.org/))
Date: Tue, 13 Mar 2001 16:43:29 +0100
Subject: Einstein's Riddle
References: <oNPq6.396644$ge4.138479566@news2.rdc2.tx.home.com> <98glrd$1si1a$1@ID-75892.news.dfncis.de> <3AABE712.499CCF4C@home.com> <98gp5b$1u0oi$1@ID-75892.news.dfncis.de>
Message-ID: <98lf4g$2dqj2$1@ID-63952.news.dfncis.de>

"Boris Popov" <no-spam-popov at bigserver.com> wrote in message
news:98gp5b$1u0oi$1 at ID-75892.news.dfncis.de...
> P.S. What's emu ?

A European currency unit, now replaced by the Euro..

--
Biep
http://www.biep.org/
(email to me can be sent from there)






From martin.skott at propylon.com  Thu Mar  8 08:39:58 2001
From: martin.skott at propylon.com (Martin Skott)
Date: 08 Mar 2001 13:39:58 +0000
Subject: virtual hosting in Zope
References: <987lbo$119t$1@infocom.km.ua>
Message-ID: <8zmg2xch.fsf@propylon.com>

"Anton A. Zotov" <futurb at nh.km.ua> writes:

> Urgent: I'd appreciate a person who would tell me how to implement virtual
> hosting under Zope/Medusa environment

What do you actually mean with virtual hosting?
Is your Zope server placed behind a webserver (eg. Apache) or is it
standalone Zope?

I think nomatter your setup you should take a look at the SiteAccess
product which you can find on www.zope.org. I have heard rumors that
it should be included in the latest Zope, but haven't checked myself.

-- 
Martin Skott
Software Developer
Propylon - Enabling Universal Mobility
http://www.propylon.com
Tel: +353 096 37245
Mobile: +353 087 9680370


From paul.moore at uk.origin-it.com  Thu Mar  8 05:46:23 2001
From: paul.moore at uk.origin-it.com (Paul Moore)
Date: Thu, 8 Mar 2001 11:46:23 +0100 
Subject: urllib.urlopen and proxies
References: <vse4atgcgcfafth7c5tq3po7srd3l32f3h@4ax.com> <008301c0a57d$1331cde0$a100000a@local> <mailman.983804894.17415.python-list@python.org> <BfakOm9OYyn2V0Hg9h1SSS=a7t7Y@4ax.com> <mailman.983901970.11969.python-list@python.org> <vBKmOkB47iH8JfPSAvl1kJShzae+@4ax.com> <mailman.983969111.11625.python-list@python.org>
Message-ID: <XWKnOkoTSX9D5gaI8iSioP21b6Iv@4ax.com>

On Wed, 7 Mar 2001 13:44:11 +0100 , Siggy Brentrup <bsb at winnegan.de> wrote:

>Paul Moore <paul.moore at uk.origin-it.com> writes:
>Well, obviously I should have tested before posting :-( There's one
>omission in my code as well as three typos in urllib2.py (sample and
>patch attached)

Hmm. I was using Python 2.0. I switched to Python 2.1, and applied your patch
(did you send it to SourceForge as a bug report? If not, would you like me
to?) Running your script (with my proxy info) I still get an error...

C:\Data>python21 proxy_auth.py
Traceback (most recent call last):
  File "proxy_auth.py", line 18, in ?
    f = urllib2.urlopen('http://www.python.org/')
  File "c:\applications\python21\lib\urllib2.py", line 135, in urlopen
    return _opener.open(url, data)
  File "c:\applications\python21\lib\urllib2.py", line 318, in open
    '_open', req)
  File "c:\applications\python21\lib\urllib2.py", line 297, in _call_chain
    result = func(*args)
  File "c:\applications\python21\lib\urllib2.py", line 822, in http_open
    return self.do_open(httplib.HTTP, req)
  File "c:\applications\python21\lib\urllib2.py", line 816, in do_open
    return self.parent.error('http', req, fp, code, msg, hdrs)
  File "c:\applications\python21\lib\urllib2.py", line 344, in error
    return self._call_chain(*args)
  File "c:\applications\python21\lib\urllib2.py", line 297, in _call_chain
    result = func(*args)
  File "c:\applications\python21\lib\urllib2.py", line 425, in
http_error_default
    raise HTTPError(req.get_full_url(), code, msg, hdrs, fp)
urllib2.HTTPError: HTTP Error 407: Proxy Authentication Required

I dumped out a few of the objects, and it looks like the request is sending a
Proxy-Authorization header to the proxy, with the base64-encoded user ID and
password in it. Is there any chance my proxy doesn't work that way, and
requires a different form of authorisation? I know nothing much about proxies,
but I seem to recall seeing "Squid" in some of the proxy messages - is Squid a
type of proxy????

If this is some sort of bizarre peculiarity of my proxy setup, then feel free
to drop the issue - I can live without getting this to work, and I don't want
to take up too much of your time with this.

Thanks for all the help you've provided so far - I feel that I at least
understand the problem much better now...!

Paul.



From ws-news at gmx.at  Tue Mar 13 16:11:53 2001
From: ws-news at gmx.at (Werner Schiendl)
Date: Tue, 13 Mar 2001 22:11:53 +0100
Subject: This math scares me
References: <3aad4062.1290021@News.CIS.DFN.DE> <20010312170356.B3513@better.net> <mailman.984480856.18097.python-list@python.org> <984486395.407379@newsmaster-04.atnet.at> <slrn9asf1d.ja3.qrczak@qrnik.zagroda>
Message-ID: <984518199.127140@newsmaster-04.atnet.at>

Marcin 'Qrczak' Kowalczyk <qrczak at knm.org.pl> wrote in message
news:slrn9asf1d.ja3.qrczak at qrnik.zagroda...
> Tue, 13 Mar 2001 13:21:50 +0100, Werner Schiendl <ws-news at gmx.at> pisze:
>
> > The most useful approach is to round the numbers to 15 digits for
display,
> > because this is the precision double values can really correctly
represent.
> > The drawback is, that the rounded value does not discover all the
> > information actually stored in the variable.
>
> The most useful approach is to display as much as is needed to read
> the value back, but no more. I.e. repr(float("0.2")) == "0.2", and
> there is no other float value whose repr is "0.2".
>

15 to 16 digits are the precision double values will provide. So rounding to
15 digits will normally display the value you have entered. There is however
no guarantee that your value is displayed exactly. Even if is looks nice in
decimal writing. I did not try for 0.2 (nor any other, because that does not
matter, I guess) but there are numerous values that cannot be converted to
double exactly.

If such a number is converted to its representation and rounded once more,
it may not yield the same value again if it is converted back afterwards.
The use of __repr__, as to my knowledge, is to provide the state of the
object in a way that will evaluate to the same object value.

> This doesn't solve the problem from the beginning of this thread,
> because float("0.1") + float("0.2") needs not be equal to float("0.3").

True, but even 0.1 and 0.2 need not necessarily be exact values.

> But at least all short numbers entered at the prompt don't have those
> unnecessary 00000s or 99999s, and conversion to a string and back
> preserves the value.
>

Non fractional numbers that do not exceed the range of the mantissa will be
stored as exact values.
(At least I think this is so...)

regards
werner





From stephen_purcell at yahoo.com  Mon Mar 26 08:27:21 2001
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Mon, 26 Mar 2001 15:27:21 +0200
Subject: Help how do I read `word' variables from a binary file?
In-Reply-To: <3abf3b0b.1487830@news.is.co.za>; from nomad***@***freemail.absa.co.za on Mon, Mar 26, 2001 at 12:57:31PM +0000
References: <3abf3b0b.1487830@news.is.co.za>
Message-ID: <20010326152721.A13900@freedom.puma-ag.com>

nomad wrote:
> I can open the file without problems, and I can read from the file
> without a hitch, _BUT_ the trouble comes in that some of the data
> records in the file have fields that would be a double word or even a
> normal word in C++, so in effect the data may be 0x80002310, but
> python reads the data directly from the file, and as such is reading
> hex bytes as (in this example) `10 23 00 80' etc.  I'm sure that there
> must be a simple way to read in data like this, but I can't figure it
> out.


Look at the 'struct' module. For example, to read a network-byte-ordered word
(i.e. 2 bytes) from a stream you can use the following code:

    def readWord(self, stream):
        b2 = stream.read(2)     
        return struct.unpack('!H',b2)[0]

-Steve


-- 
Steve Purcell, Pythangelist
Get testing at http://pyunit.sourceforge.net/
Any opinions expressed herein are my own and not necessarily those of Yahoo



From chrisw at nipltd.com  Mon Mar 26 12:16:21 2001
From: chrisw at nipltd.com (Chris Withers)
Date: Mon, 26 Mar 2001 18:16:21 +0100
Subject: Namespaces question
Message-ID: <3ABF7965.D8445EFB@nipltd.com>

Hi,

I've got a method as follows:

def create(self):

    for name in ['x','y','z']:
        value = session.get(name,None)
        if value is None:
            errors.append('No %s supplied.' % name)
        else:
            locals()[name]=value

    if errors:
        return errors

    print x

...now I know it's not right, 'cos I get a NameError on x at the print
statement.

What should I be using instead of locals()?

cheers,

Chris



From root at rainerdeyke.com  Sat Mar  3 19:41:29 2001
From: root at rainerdeyke.com (Rainer Deyke)
Date: Sun, 04 Mar 2001 00:41:29 GMT
Subject: Missing the functional mind set
References: <mlb2at8ptrr7sh5dgq8qlk2201kquh6tfl@4ax.com> <uDao6.335972$ge4.119457077@news2.rdc2.tx.home.com> <t6f2at870pk82q96bnd5a2od4n8ns5kf83@4ax.com>
Message-ID: <Zago6.339134$ge4.120232013@news2.rdc2.tx.home.com>

"Daniel Klein" <danielk at aracnet.com> wrote in message
news:t6f2at870pk82q96bnd5a2od4n8ns5kf83 at 4ax.com...
> On Sat, 03 Mar 2001 18:22:18 GMT, "Rainer Deyke" <root at rainerdeyke.com>
wrote:
>
> >But if you are indexing with integers from 0 to n, why are you using a
> >dictionary in the first place?
>
> I'm not sure I understand your question. I need the dictionary to
associate a
> numbered key with a value. Rather than force the user to enter elements
'in
> order' I am letting them specify what the order is on-the-fly, ie enter
element
> 3 first, then 1, then 2, then 0. The method will the place the values in
the
> proper order for further processing.

Assuming the elements are numbered from '0' to 'n - 1' *and* 'n' is know
ahead of time, you could just create the list as '[None] * n' and use it
instead of the dictionary.  If 'n' is not known ahead of time, this may be
more trouble than it's worth.


--
Rainer Deyke (root at rainerdeyke.com)
Shareware computer games           -           http://rainerdeyke.com
"In ihren Reihen zu stehen heisst unter Feinden zu kaempfen" - Abigor




From grante at visi.com  Thu Mar 15 12:51:39 2001
From: grante at visi.com (Grant Edwards)
Date: Thu, 15 Mar 2001 17:51:39 GMT
Subject: Who's minister of propaganda this week?
References: <B98041329020E007.7DA84CAD9D0562AB.817BBC572F480869@lp.airnews.net> <98nu1m$12e@dispatch.concentric.net> <660ACCA1E4816F7E.C359CDF4168EF3FB.3B6B05538D619770@lp.airnews.net> <3AAFA13D.76E32C3F@divalsim.it> <slrn9b0ijs.ccb.kzaragoza@kzaragoza.ne.mediaone.net>
Message-ID: <Li7s6.895$lz1.79367@ruti.visi.com>

In article <slrn9b0ijs.ccb.kzaragoza at kzaragoza.ne.mediaone.net>, Kris J. Zaragoza wrote:

>Just a minor nit:  The languanges that are being discussed are not
>actually typeless.  They are typed, but dynamically.  Every object in
>Python, for example, has some type associated with it, and as a
>result, the semantics that follow from that type.  You cannot thus put
>a string in the middle of a mathematical expression.  Relatively few
>of the "scripting" languages in popular use today are weakly typed or
>typeless (Perl and TCL come to mind.)

I would consider TCL to be degenerately typed: everything is a
string.

I tried to write a program in TCL once many years ago.  After
the bleeding from the ears abated, I switched to Scheme.  ;)

-- 
Grant Edwards                   grante             Yow!  If Robert Di Niro
                                  at               assassinates Walter Slezak,
                               visi.com            will Jodie Foster marry
                                                   Bonzo??


From cce at clarkevans.com  Mon Mar 26 06:23:32 2001
From: cce at clarkevans.com (Clark C. Evans)
Date: Mon, 26 Mar 2001 06:23:32 -0500 (EST)
Subject: portable string localization and resource bundles...
Message-ID: <Pine.LNX.4.21.0103260619180.21288-100000@clarkevans.com>

I was wondering if there is a solution to the 
common problem of string localization 
(menu and message box titles, for instance).

Thank you,

Clark




From embed at geocities.com  Fri Mar  2 15:34:02 2001
From: embed at geocities.com (Warren Postma)
Date: Fri, 02 Mar 2001 20:34:02 GMT
Subject: A counter-counter-proposal for PEP 236: #pragma( nested_scopes)
References: <mailman.983475064.30859.python-list@python.org>
Message-ID: <_sTn6.15738$UZ4.4336745@news4.rdc1.on.home.com>

Sallied for Warren:
> > Has anyone considered borrowing a C notation:
> >
> > #pragma( nested_scopes )
> >
> > Benefits:
> >     - It's a comment therefore it doesn't break anything else.

Thusly spake Peter:
> But that's dead-wrong behavior for future_statements.  If your module
needs
> nested_scopes semantics, it will do unpredictable wrong things at runtime
if
> it doesn't get them.  That's why PEP 236 is emphatically *not* a "pragma"
> proposal.

Hmm.  I wonder how many people would be happier with this syntax

from __twilightzone__ import nested_scopes

<grin>

Warren





From scarblac at pino.selwerd.nl  Tue Mar 27 09:29:11 2001
From: scarblac at pino.selwerd.nl (Remco Gerlich)
Date: 27 Mar 2001 14:29:11 GMT
Subject: who is my caller?
References: <mailman.985700854.31956.python-list@python.org>
Message-ID: <slrn9c18o0.ojm.scarblac@pino.selwerd.nl>

Clark C. Evans <cce at clarkevans.com> wrote in comp.lang.python:
> In many mud languages, you can query the
> call-stack to find out who called you.
> Is there a stack of function objects
> that can be obtained somehow?  Is this
> non-portable black magic?  
> 
> I'm asking beacuse this technique is
> often useful for implementing security
> (ok, not very good security)

if (this_player(1) != this_object())
   return !notify_fail("Back off, cheater.\n")
   
Ahh, the memories...

Anyways, yes, it's more like black magic in Python. There is a call stack
you can inspect, but only if you just caught an exception.

import sys

try:
   raise "whee"
except "whee":
   traceback = sys.exc_info()[2]
   # extract whatever info you need from the traceback
finally:
   del traceback # to avoid cyclic references

See the 'traceback' module to see how you can print out a traceback, etc.
You can also play around with the object in the interpreter to see what kind
of attributes it has, and its attributes, and...

But, what kind of security are you thinking of? You're not going to mix
untrusted code with server code in some sort of mud? The mud's code (like
LPC on a lpmud) has to run in some sort of sandbox, otherwise all the
server's internals will be open.

Python has 'rexec' to run things inside a sandbox, but I don't think it's
quite industrial strength enough to build a mud on. Too many good coders
with too much time trying to cheat :). There has been discussion before,
I seem to remember a module, but I can't think of it at the moment, search
around for "mud", "rexec", "python", that sort of thing...

-- 
Remco Gerlich


From sholden at holdenweb.com  Thu Mar 22 22:34:18 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 22 Mar 2001 22:34:18 -0500
Subject: Why Python/Jython?
References: <99cl9e$b2b$1@Merlin.et.bocholt.fh-gelsenkirchen.de> <mailman.985260089.28444.python-list@python.org> <1901D50DC30D30E4.16551C1AB0FB4363.C0D72E208759F8A7@lp.airnews.net> <99dkg301894@news1.newsguy.com> <3ABAB2F7.127D3872@webamused.com> <slrn9bl0ka.9ck.mmealman@loki.thebisgroup.com>
Message-ID: <Eyzu6.9018$G72.213538@e420r-atl1.usenetserver.com>

"Mark" <mmealman at loki.thebisgroup.com> wrote in message
news:slrn9bl0ka.9ck.mmealman at loki.thebisgroup.com...
[ ... ]

> But I'm still real fuzzy on how their paths/jars/imports fully work and
it's
> a pain in the ass for me to install any app that wants to run off the JDK
or
> JRE, mostly because getting the stupid CLASSPATH right makes Quantum
physics
> look easy.
>
> Java = good intentions, bad implementations.
>
Write once, debug everywhere. I'd like to avoid this for Python if it
doesn't get in the way of the world domination plan so successfully scripted
by the PS




From costas at meezon.com  Thu Mar  1 21:56:45 2001
From: costas at meezon.com (Costas Menico)
Date: Fri, 02 Mar 2001 02:56:45 GMT
Subject: define a new func on the fly?
References: <3A9EE927.8744154E@troikanetworks.com> <o6Cn6.846$45.1115@www.newsranger.com>
Message-ID: <3a9f0bc2.10059041@News.CIS.DFN.DE>

Gregory Jorgensen<gregj at pobox.com> wrote:

>eval() evaluates an expression. exec() executes a statement. Try this:
>
>>>> exec("def f():\n\treturn 'hello'\t")

When I try the above I get a syntax error... I am using PythonWin. Is
this a bug?

Costas


>>>> f()
>'hello'
>
>>>> eval("f()")
>'hello'



From dougfort at downright.com  Thu Mar  8 14:44:53 2001
From: dougfort at downright.com (Doug Fort)
Date: Thu, 08 Mar 2001 14:44:53 -0500
Subject: Documentation/Examples about the htmllib?
References: <3aa7c372$0$22464@SSP1NO25.highway.telekom.at>
Message-ID: <3AA7E135.EBDA80F1@downright.com>

httplib seems to want to  prettyprint only.

To get actual tags, we use sgmllib.  I've attached the module we use  to
extract <form> tag stuff.   Getting <area>  tags shoulld be similar.
--
Doug Fort <dougfort at downright.com>
Senior  Meat Manager
Downright Software LLC
http://www.dougfort.net


Hermann Himmelbauer wrote:

> Hi,
> I want to parse a htmlpage with python, so I thought using the htmllib
> would be good for that task.
>
> In my html-page I have this tag:
> <area href="/html/page.html?key" .... >
>
> What I want to do is extract this link parameter "key".
>
> It would be perfect if the htmllib would extract me all the <area ...> tags
> into a list so that I could simply find the rigth tag and extract the key.
>
> I did manage to get data between a tag like <title> data </title> but could
> not get the data "in" the tag itself. Of course I could do this with a
> regular expression but I thought using this module would give me better
> results, what do you think?
>
> Does anyone have a clue?
>
>                 Best Regards,
>                 Hermann
>
> --
>  ,_,
> (O,O)     "There is more to life than increasing its speed."
> (   )     -- Gandhi
> -"-"--------------------------------------------------------------

--
Doug Fort (dougfort at downright.com)
Senior Meat Manager
Downright Software LLC
http://www.dougfort.net


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20010308/ed37edd9/attachment.html>
-------------- next part --------------
#!/usr/bin/env python
"""
FormFieldParser

This object parses HTML text and builds a dictionary of
dictionaries of form fields

$Id: formfieldparser.py,v 1.1 2001/01/26 15:18:30 dougfort Exp $
"""
__author__="Downright Software LLC"
__version__="$Revision: 1.1 $"[11:-2]

import sgmllib
import string
import cStringIO
import urllib
import re

import webnudge.util.misc
import webnudge.util.document

class FormFieldParserException:
    def __init__(self, message):
        self._message = message
    def __str__(self):
        return self._message

###########################################################
class FormFieldParser(sgmllib.SGMLParser):
###########################################################
    """
    FormFieldParser class. Parse a page from a website,
    creating a dictionary of dictionairies of form
    fields
    """

    #----------------------------------------------------------
    def __init__(self):
    #----------------------------------------------------------
        """
        Constructor
        """
        sgmllib.SGMLParser.__init__(self)

        self._formcount = 0
        self._formdict = {}

    #----------------------------------------------------------
    def parse(self, text):
    #----------------------------------------------------------
        """
        parse some text, without trashing javascript
        """
        self.feed(text)
        self.close()
        return self._formdict
        
    #----------------------------------------------------------
    def start_form(self,attributes):
    #----------------------------------------------------------
        """
        start a form
        """
        self._formdict[self._formcount] = {}
    #----------------------------------------------------------
    def end_form(self):
    #----------------------------------------------------------
        """
        end a form
        """
        self._formcount += 1
        
    #----------------------------------------------------------
    def _storeformfield(self,attributes,multivalue=0):
    #----------------------------------------------------------
        """
        Capture name and value attributes of a form field
        """
        tagname = None
        tagvalue = ""
        selected = 0
        for key, value in attributes:
            if key == "name":
                tagname = value
                continue
            if key == "value":
                tagvalue = value
                continue
            if key == "selected":
                selected = 1
                continue
        if multivalue and not selected:
            return
            
        if tagname:
            self._formdict[self._formcount][tagname] = tagvalue
        
    #----------------------------------------------------------
    def do_input(self,attributes):
    #----------------------------------------------------------
        """
        Capture <input> element
        """
        self._storeformfield(attributes)
        
    #----------------------------------------------------------
    def do_option(self,attributes):
    #----------------------------------------------------------
        """
        Capture <option> element
        """
        self._storeformfield(attributes, multivalue=1)
        
    #----------------------------------------------------------
    def do_select(self,attributes):
    #----------------------------------------------------------
        """
        Capture <select> element
        """
        self._storeformfield(attributes, multivalue=1)
        
    #----------------------------------------------------------
    def do_textarea(self,attributes):
    #----------------------------------------------------------
        """
        Capture <textarea> element
        """
        self._storeformfield(attributes)
        
#----------------------------------------------------------
if __name__ == "__main__":
#----------------------------------------------------------
    """
    Code for commandline testing
    """
    import sys
    if len(sys.argv) != 2:
        print "Usage:  filteringparser.py <url>"
        sys.exit(-1)

    import webnudge.util.rawhtmlpage
    page = webnudge.util.rawhtmlpage.RawHTMLPage()
    page.load(sys.argv[1])
    if not page:
        print "*** Error *** %s" % (page._message)
        sys.exit(-1)

    result = FormFieldParser().parse(page._data)
    
    sys.stdout.write(repr(result))



From chris.gonnerman at usa.net  Thu Mar  8 21:46:44 2001
From: chris.gonnerman at usa.net (Chris Gonnerman)
Date: Thu, 8 Mar 2001 20:46:44 -0600
Subject: kbhit()  and  getch() question
References: <taecoqaa2mot62@corp.supernews.com>
Message-ID: <009301c0a843$5a8dede0$a100000a@local>

----- Original Message -----
From: "Jeremy Reed" <jp_reed at yahoo.com>
Subject: kbhit() and getch() question

> In a previous question, I asked how to get non-blocking input from the
user.
> The answer proposed is as follows:
>
> import msvcrt
>
> ch = 0
> while ch != 'x':
>     if msvcrt.kbhit():
>         ch = msvcrt.getch()
>         print ch
>
> I have run into a few problems.  First of all, the kbhit() doesn't seem to
> work.  I am unable to find a way for the function to return true.
[brutal hack]

This code as given works for me on Win98SE, Python 2.0.  What is your
platform
and Python version?





From chris.gonnerman at usa.net  Thu Mar 15 21:54:24 2001
From: chris.gonnerman at usa.net (Chris Gonnerman)
Date: Thu, 15 Mar 2001 20:54:24 -0600
Subject: site.py on Windows
References: <%tKr6.2570$pX.26174@news1.onlynews.com><98oq3r$1e0$1@troll.powertech.no> <u3dcfagwh.fsf_-_@advancedmp.net>
Message-ID: <013501c0adc4$6a94b620$a100000a@local>

Frankly, I just hack site.py like so:

# site.py, Python 2.0, about line 259
addsitedir("c:/python/lib/site-packages")

It may not be the right way to do it, but it works for me.  Don't
forget to substitute your Python install directory.

----- Original Message -----
From: "Don Dwiggins" <dwig at advancedmp.net>
To: <python-list at python.org>
Sent: Wednesday, March 14, 2001 8:46 PM
Subject: site.py on Windows


> I'm trying to use the site-packages facility, as documented for site.py,
in
> my ActiveState distribution (2.03, I think).  After some investigation, I
> found that the site.py code essentially punts if the path separator
> character isn't '/'.  After seeing that, I also noticed in the fine print
of
> the documentation: "Note that for some non-Unix systems, sys.prefix and
> sys.exec_prefix are empty, and the path manipulations are skipped; however
> the import of sitecustomize is still attempted."  Doh!
>
> OK, has anyone fixed site.py to do the right stuff under Windows?  (I'd
> rather not resort to writing a sitecustomize.)  If not, I'm willing to
have
> a go at it, and submit what I come up with.  It doesn't look particularly
> difficult, but if there are subtle issues I'm not seeing, please let me
> know.
>
> --
> Don Dwiggins                    "Solvitur Ambulando"
> Advanced MP Technology
> dwig at advancedmp.net
>
>
> --
> http://mail.python.org/mailman/listinfo/python-list
>




From bellman at lysator.liu.se  Sat Mar 17 20:58:48 2001
From: bellman at lysator.liu.se (Thomas Bellman)
Date: 18 Mar 2001 01:58:48 GMT
Subject: [Python-Dev] Revive the types sig?
References: <jeremy@alum.mit.edu> <15020.9404.557943.164934@w221.z064000254.bwi-md.dsl.cnc.net> <200103120711.AAA09711@localhost.localdomain> <mailman.984439040.24419.python-list@python.org> <s44tatolisnipdfm4r4p1aqt2e09e458hk@4ax.com> <mailman.984581571.10100.python-list@python.org> <1pWvOlsMJ9Zxk0d4s9BhFmMCSusy@4ax.com> <n3Nr6.263$4d3.55227@newsb.telia.net>
Message-ID: <9914oo$ij6$1@newsy.ifm.liu.se>

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

> Paul Moore wrote:

>>     def f(n):
>>         assert <n supports the "integer" interface>

> footnote: the correct way to spell this is:

>         assert isinstance(n, type(1)) # or types.IntType

But that doesn't mean the same thing!  There is a difference
between following a certain interface specification, and
inhereting from a common base class.


-- 
Thomas Bellman,   Lysator Computer Club,   Link?ping University,  Sweden
"Adde parvum parvo magnus acervus erit"       ! bellman @ lysator.liu.se
          (From The Mythical Man-Month)       ! Make Love -- Nicht Wahr!


From qrczak at knm.org.pl  Mon Mar 12 17:44:17 2001
From: qrczak at knm.org.pl (Marcin 'Qrczak' Kowalczyk)
Date: 12 Mar 2001 22:44:17 GMT
Subject: file - codecs - unicode ???
References: <3aace9da$0$382$446d705c@news.skynet.be>
Message-ID: <slrn9aqka1.3vk.qrczak@qrnik.zagroda>

Mon, 12 Mar 2001 16:29:42 +0100, S?bastien Libert <sebastien.libert at comexis.com> pisze:

> >>> line
> '\377\3760\000.\0000\0000\000\011\000i\000n\000f\000o\000\011\000L\000o\000g
> \000_\000S\000t\000a\000n\000d\000a\000r\000d\000\011\000n\000g\000L\000o\00
> 0g\000\015\000\012'
> 
> What can i do with this kind of thing ?????

It's encoded in UTF-16, so:
    unicode(line, 'UTF-16').encode('ASCII')
except that the '\012' at the end is bogus. It shouldn't be there.

You may want to use a different encoding than ASCII, because ASCII
is only able to encode Latin letters without accents. If the string
contains other character, you will get an exception.

-- 
 __("<  Marcin Kowalczyk * qrczak at knm.org.pl http://qrczak.ids.net.pl/
 \__/
  ^^                      SYGNATURA ZAST?PCZA
QRCZAK


From cce at clarkevans.com  Tue Mar  6 14:07:02 2001
From: cce at clarkevans.com (Clark C. Evans)
Date: Tue, 6 Mar 2001 14:07:02 -0500 (EST)
Subject: Can Python replace Visual Basic? Should it?...and generalities
 :)
In-Reply-To: <CM9p6.12769$hn5.1803901@news1.rdc1.mb.home.com>
Message-ID: <Pine.LNX.4.21.0103061336560.21280-100000@clarkevans.com>

On Tue, 6 Mar 2001, Brad Bollenbach wrote:
> I was really hoping someone would come along and say something like:
> 
>     "Yes, our company used to use VB for our main product, a universal
> scheduler program with a clickable ledger. We saw the benefits of Python and
> rewrote the whole thing in it, and our company have saved more time and
> money than ever before, and our customers are thanking us for it."
> 
> Unfortunately, maybe that time won't come.

Over the last 3-4 months I've written a pretty sophisticated
"front-end" program using Visual Basic; it is essentially an
editor for a specific type of XML file.  (I've used Visual Basic
since Version 1.0 and have written VBX's, but not ATL OCX's yet.)   
A few years ago '97, I switched over to Java, and got sick to 
my stomach (Java is far worse than Visual Basic) after a year
of giving it a fair shake.  So, I switched back...

I did spend some time (about a week) investigating Python.  Here 
are my impressions:

  1.  I very much like Python language, and I will learn much more.
      In particular, if you interface is "dynamic", where user interface
      widgets are being created/destroyed/moved all of the time then
      Python could be a big winner!
  2.  It was not "obvious" how to put together a Windows installation
      program that is python friendly.  We need to have a seperate
      user-level python/wxwindows module and then a way for the
      install program to check to see if that module is installed,
      downloading/fetching-from-cd if the module isn't.
  3.  It looks significantly slower, and the particular application
      that I have does a ton of re-calculating and re-drawing and
      has to work on out-dated Win95 boxes on office desks.
      I wrote a small sample, and as I feared, it showed significant
      "lag" on a Pentium 60.  The visual basic program is 'snappy.
  4.  The re-draw speeds got somewhat faster when I re-wrote my 
      small sample program with wxWindows.  wxWindows looks cool.
  5.  I did not see a sutiable replacement for the visual basic
      data control.  Writing all of the widget to data binding
      code is *very* tedious... Python just isn't there yet in
      this respect.
  6.  On the plus side, Python has some wonderful XML tools
      (4suite.org), and it's string processing kicks the shit
      out of visual basic (I still use VSAWK for string processing).
  
If Pythoners seriously want to have people in VB land convert
over to Python, then a few "immediate" things need to be done:

  A Python "OCX" control needs to be written.  This control
  would simply encapsulate Python for us Visual Basic idiots.
  This way we could 'pop in the control in our VB project and
  start using Python for parts of the program.  
  
  At first, this control could be non-visual.  The OCX would
  have to expose through IDispatch the objects managed within
  the embedded python system. Then, it  would be cool if it 
  became a wxWindows frame, so that small client areas could 
  be re-written in Python, etc.  For extra bonus points,
  the visual control would allow OCX components to be dropped.

  Eventually the VB programmer could 'drop the VB shell after 
  their entire program is re-written.  In the corporate world, 
  it will be damn near impossible to get funding for re-writing
  a program in Python.  However, it is very possible to have
  this conversion be INCREMENTAL, one sub-routine at a time!
  In short, a complete re-write is just too high of risk for
  most corporate progammers to do...  

By the way, the component does have a server-side part.  I've
decided to write this in Python since cross-platform compatibility
would be a big bonus.  I'll give a report on my findings here
in about two to three months.

Kind Regards,

Clark Evans

  




From cce at clarkevans.com  Wed Mar 21 21:08:40 2001
From: cce at clarkevans.com (Clark C. Evans)
Date: Wed, 21 Mar 2001 21:08:40 -0500 (EST)
Subject: Yet Another PEP:  Query Protocol Interface or __query__
In-Reply-To: <B43D149A9AB2D411971300B0D03D7E8B0FBFBE@natasha.auslabs.avaya.com>
Message-ID: <Pine.LNX.4.21.0103212101370.11584-100000@clarkevans.com>

On Thu, 22 Mar 2001, Delaney, Timothy wrote:
> I think the major trap would be going with a system which did not have
> sufficient expressiveness. One of the problems with passing a base class
> object (or class instance) to __adapt__() is that there is no standard way
> to specify requirements other than "interface has these methods".

The goal of this PEP (if it gets accepted as a PEP...) is to 
provide a simple mechanism for "adapt", _without_ having to 
dive into the many many issues that the type special interest
group is exploring.  If done right, the mechanism can be expanded 
later to handle interfaces or signature specifications, instead
of a class.  I want to keep this simple...

> In interface descriptor class is much more general, and much more easily
> extended. For example, the first iteration could simply be used to add
> method requirements. Then we could add method parameter and return type
> signatures. Then pre- and post- conditions (such as parameter #3 <= 5). And
> so on.

Right.  And this PEP proposal would not preclude this at
all, and infact, this type of stuff could be plugged in
as additional "work" that the adapt method can perform.

> I think the adapt stuff would work as follows...

You should read up on Michel Pelletier's work PEP 245
"Python Interfaces" which is at:

  http://python.sourceforge.net/peps/pep-0245.html

Clark




From guido at digicool.com  Thu Mar  1 16:01:52 2001
From: guido at digicool.com (Guido van Rossum)
Date: Thu, 01 Mar 2001 16:01:52 -0500
Subject: d = {}; d[0:1] = 1; d[0:1] = 2; print d[:]
In-Reply-To: Your message of "Thu, 01 Mar 2001 15:54:55 EST."
             <15006.46879.607192.367739@w221.z064000254.bwi-md.dsl.cnc.net> 
References: <15006.46879.607192.367739@w221.z064000254.bwi-md.dsl.cnc.net> 
Message-ID: <200103012101.QAA18516@cj20424-a.reston1.va.home.com>

(Adding python-dev, keeping python-list)

> Quoth Robin Thomas <robin.thomas at starmedia.net>:
> | Using Python 2.0 on Win32. Am I the only person to be depressed by the 
> | following behavior now that __getitem__ does the work of __getslice__?
> |
> | Python 2.0 (#8, Oct 16 2000, 17:27:58) [MSC 32 bit (Intel)] on win32
> |  >>> d = {}
> |  >>> d[0:1] = 1
> |  >>> d
> | {slice(0, 1, None): 1}
> |
> | And then, for more depression:
> |
> |  >>> d[0:1] = 2
> |  >>> d
> | {slice(0, 1, None): 1, slice(0, 1, None): 2}
> |
> | And then, for extra extra chagrin:
> |
> |  >>> print d[0:1]
> | Traceback (innermost last):
> |    File "<pyshell#11>", line 1, in ?
> |      d[0:1]
> | KeyError: slice(0, 1, None)
> 
> If it helps, you ruined my day.

Mine too. :-)

> | So, questions:
> |
> | 1) Is this behavior considered a bug by the BDFL or the community at large?

I can't speak for the community, but it smells like a bug to me.

> | If so, has a fix been conceived? Am I re-opening a long-resolved issue?

No, and no.

> | 2) If we're still open to proposed solutions, which of the following do you 
> | like:
> |
> |     a) make slices hash and cmp as their 3-tuple (start,stop,step),
> |        so that if I accidentally set a slice object as a key,
> |        I can at least re-set it or get it or del it :)

Good idea.  The SF patch manager is always open.

> |     b) have dict.__setitem__ expressly reject objects of SliceType
> |        as keys, raising your favorite in (TypeError, ValueError)

This is *also* a good idea.

> From: Donn Cave <donn at oz.net>
> 
> I think we might be able to do better.  I hacked in a quick fix
> in ceval.c that looks to me like it has the desired effect without
> closing the door to intentional slice keys (however unlikely.)
[...]
> *** Python/ceval.c.dist Thu Feb  1 14:48:12 2001
> --- Python/ceval.c      Wed Feb 28 21:52:55 2001
> ***************
> *** 3168,3173 ****
> --- 3168,3178 ----
>         /* u[v:w] = x */
>   {
>         int ilow = 0, ihigh = INT_MAX;
> +       if (u->ob_type->tp_as_mapping) {
> +               PyErr_SetString(PyExc_TypeError,
> +                       "dict object doesn't support slice assignment");
> +               return -1;
> +       }
>         if (!_PyEval_SliceIndex(v, &ilow))
>                 return -1;
>         if (!_PyEval_SliceIndex(w, &ihigh))

Alas, this isn't right.  It defeats the purpose completely: the whole
point was that you should be able to write a sequence class that
supports extended slices.  This uses __getitem__ and __setitem__, but
class instances have a nonzero tp_as_mapping pointer too!

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



From nelson at crynwr.com  Mon Mar 26 23:25:27 2001
From: nelson at crynwr.com (Russell Nelson)
Date: Mon, 26 Mar 2001 23:25:27 -0500
Subject: Linux GPS program
Message-ID: <3AC01637.4EF3A6F2@crynwr.com>

Hi.  I have written a GPS program in Python for Linux.  I'm sure that
someone 
could port it to run under Windows, but I have no interest in doing so. 
I haven't
done a real release of the software yet, since it's still under heavy
development.
But you're welcome to download a copy from handhelds.org's CVS.  The
location is "apps/pygps".  For info on handhelds.org's CVS:
http://www.handhelds.org/sources.html

Right now, it does moving maps (UTM bitmaps), track recording, and
playback.
Will select the "best" of multiple maps.

-- 
-russ nelson will be speaking at http://www.osdn.com/conferences/brie/
Crynwr sells support for free software  | PGPok | Watch out!  He's got
an
521 Pleasant Valley Rd. | +1 315 268 1925 voice | opinion, and he's not
Potsdam, NY 13676-3213  | +1 315 268 9201 FAX   | afraid to share it!


From sholden at holdenweb.com  Thu Mar 15 03:37:27 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 15 Mar 2001 03:37:27 -0500
Subject: Better string.translate?
References: <3aaf758f$1_2@news.zet.net> <mailman.984590438.506.python-list@python.org> <3ab0749f$1_3@news.zet.net>
Message-ID: <Be%r6.83512$ML1.3899965@e420r-atl2.usenetserver.com>

"Stowasser Harald" <stowasser.h at idowa.de> wrote in message
news:3ab0749f$1_3 at news.zet.net...
> "Steve Purcell" <stephen_purcell at yahoo.com> schrieb :
> > Use the 'sub' function in the 're' module:
>
> Thank you, but i don't want do delete the characters. I need to
> translate them into Spaces.
>
> Anybody knows how?
>
I'm not sure I understand ... your original code was intended to translate
punctuation marks into spaces? In my newsreader I suspect that a load of
spaces in the second argument to maketrans() were trimmed as part of a
line-wrap, so the code wasn't too obvious.

Perhaps you could explain again what it is you want to do.

regards
 Steve





From aleaxit at yahoo.com  Thu Mar 15 04:18:09 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 15 Mar 2001 10:18:09 +0100
Subject: WebBrowser as a panacea (was Re: wxPython UI refresh)
References: <98jku0$ifi$1@newstest.laserlink.net> <slrn9avc13.p8n.zeitlin@seth.lpthe.jussieu.fr> <98p123$ls$1@newstest.laserlink.net> <tTYr6.39$yBO.17760287@news.randori.com>
Message-ID: <98q6rn1uoa@news2.newsguy.com>

"Robin Dunn" <robin at stop.spam.alldunn.com> wrote in message
news:tTYr6.39$yBO.17760287 at news.randori.com...
    [snip]
> There is a ray of hope however.  The WebBrowser ActiveX object (internet
> explorer) can view excel workbooks just fine.  In the wxPython demo for
> ActiveXWrapper_IE type the path of a .xls file into the combo box and it

This can't be stressed strongly enough: WebBrowser (and to a
lesser extents other bits and pieces of IE) are _extremely_
useful and handy -- a very full-fledged container, able to
house just about anything you can throw at it, let you script
it, catch events, etc, etc; at the same time, it's rather
undemanding -- will let just about any halfway-ripe container
house it, without making many demands on said container.  As
an "impedance matcher" between minimalist containers and very
demanding controls, it can easily be a life-saver.


Alex





From nas at arctrix.com  Sun Mar  4 02:02:51 2001
From: nas at arctrix.com (Neil Schemenauer)
Date: Sat, 3 Mar 2001 23:02:51 -0800
Subject: GC and extension types
In-Reply-To: <jKio6.8061$3F2.2558420@typhoon.ne.mediaone.net>; from abrahams@mediaone.net on Sun, Mar 04, 2001 at 03:35:43AM +0000
References: <jKio6.8061$3F2.2558420@typhoon.ne.mediaone.net>
Message-ID: <20010303230250.A14497@glacier.fnational.com>

On Sun, Mar 04, 2001 at 03:35:43AM +0000, David Abrahams wrote:
> Is there documentation someplace on what I must do to make my extension
> types play nicely with GC? I am worried that if I store a pointer to a
> PyObject* in some arbitrary memory (and increment its reference count, of
> course!), the collector might decide it was unreachable anyway and collect
> it.

Relax, it won't.  The GC only breaks reference cycles it can
find.  It actually doesn't even free memory itself.

> From looking at the source code, I have an inkling that it might try to
> account for all the references in a cycle, so that objects will only be
> collected when all known references can be accounted for by the (reachable)
> cycles in which it participates. I guess this would create the opposite but
> more benign problem of being too conservative.

Yes, it errors on the conservative side.  If you extension type
can create reference cycles you should tell the GC how to follow
the references it keeps.  Its easy to do but not well documented
right now.  This may help:

    http://python.ca/nas/python/gc_api.txt

> Anyway, I'm hoping there is authoritative text somewhere that describes
> what's really going on... and that the text isn't just 'C' code!

The C code is authoritative.  Anything else is probably
inaccurate.  :-)

  Neil



From aleaxit at yahoo.com  Wed Mar  7 04:54:15 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Wed, 7 Mar 2001 10:54:15 +0100
Subject: I come to praise .join, not to bury it...
References: <mailman.983485923.1136.python-list@python.org> <97p7iq$kk2$1@nntp6.u.washington.edu> <97rka002k9@news1.newsguy.com> <3AA2F4B3.A7496E76@cosc.canterbury.ac.nz> <97vmoh061p@news2.newsguy.com> <3AA46134.D385208B@cosc.canterbury.ac.nz> <98292h0s0g@news1.newsguy.com> <3AA58BD0.3324C709@cosc.canterbury.ac.nz>
Message-ID: <9850ge013ns@news1.newsguy.com>

"Greg Ewing" <greg at cosc.canterbury.ac.nz> wrote in message
news:3AA58BD0.3324C709 at cosc.canterbury.ac.nz...
> Alex Martelli wrote:
> >
> > No!  string_join (in stringobject.c) currently just _checks_
> > PyString_Check on the items -- if any item fails that, it
> > delegates to PyUnicode_Join (if PyUnicode_Check OK's that).
>
> And that's somehow not a typeswitch? If it's a regular
> string do this... if it's a Unicode string do that...
> smells like a typeswitch to me...

It's (needlessly) coded as a typeswitch in the Python 2
sources for string_join, yes.  The correct coding to
reflect the design would be to use the existing standard
interface to ask each item to _behave like_ a string,
rather than resorting to the "IS of identity" -- Korzibski's
intimations against it are perfectly valid in the context
of polymorphic programming, no less than in epistemology;
in fact, just about any test for 'IS object X of exact type
Y' you can find in the Python sources, unless they are
shortcut-like 'accelerators' for a special case _before_
the more-general request 'please o mr X try to behave like
an Y, are you able to?', can be seen as implementation
defects, in my personal opinion.

But we're arguing about the DESIGN, not the _current
implementation_ thereof, with its limitations and bugs.


> You seemed to be arguing that dispatching on the joiner
> somehow resulted in an efficiency gain by going straight
> to the right code for the kind of strings being joined,
> and I was pointing out that that's not the case. But
> maybe I misunderstood what you were saying.

The joiner-dispatch DESIGN *affords* efficiency gains;
the implementation, at one stage, may not take advantage
of the enabling-effects of a good design, but that's no
reason to make the design worse, contorting it to refect
the current limitations of an implementation!


> > You gain a clean approach, which the implementation could
> > (and should) exploit to access the items in the sequence
> > through the _appropriate_ standard interface
>
> It's only clean according to one measure of cleanness.
> It could be considered cleaner to treat all the arguments
> uniformly, and access them *all* through a standard
> interface.

And *WHAT* 'standard interface' would apply on the joiner
object, pray?  The only response I see that makes sense is
"a joiner interface" -- something that the joiner object
exposes, which has a method to be called to do the join.
Which, surprise surprise, is EXACTLY what we have now --
the method in question being called 'join' (why not?!).

Note that pairwise-joining of items is NOT sufficient,
because of efficiency considerations -- O(N) versus
O(N squared).  (In other terms: general joining of a
sequence may be _functionally_ constructed by iterating
a join-two-items primitive, BUT that is unacceptable as
it leads to O(N*N) performance).


> > There are NO technical costs associated with these
> > advantages.
>
> I disagree. I've mentioned one already -- it bloats the
> string object, and makes the Python core larger and
> less modular.

The string object and the string module are both in
the Python core, so moving functionality between them
can in no way be considered 'bloating'.  Somebody
who's _subsetting_ the Python core is (by definition)
operating outside of Python's problem-space, and
his or her problems are not technical Python problems.


> > Only technical advantages and costs
> > can be fairly put on the balance
>
> I disagree with that, too. I think it's quite possible
> and appropriate for a small technical advantage to be
> outweighed by a larger aesthetic one. For example, there
> are technical advantages to braces instead of indentation:
> the scanner is easier to implement, it's easier to write
> programs which generate code, and there's no chance of
> space-tab confusion. On the indentation side, there's the
> advantage that indentation and block structure always
> agree. If you're counting technical advantages, that's
> one in favour of indentation and three against. But I
> doubt Guido would switch on that basis.

'Counting' technical advantages is as idiotic as 'counting'
coins as a measure of wealth -- if I have one 1-dollar
coin, and you have three 5-cent ones, such an idiot counter
would conclude that you must be richer than me by a
three to one margin.  *Obviously*, the balancing of
technical considerations must take account of their
importance and relevance; a 'count' is inappropriate.

For the brace/indentation case, I would count only the
space/tab issue as being of any substantial worth -- the
miniscule extra workload on the scanner, and the need
for the output-stage of source-generators to keep track
of nesting (advisable on other grounds anyway) being
absolutely-unessential trifles.  As the space/tab issue
can easily be ameliorated via tabnanny & friends, I judge
the blocking==indenting advantage strongly dominant (and
I'm happy Guido's technical evaluation is the same:-).


> > Note that I've seen NO arguments AT ALL for having
> > .join be a method on SEQUENCE object ... despite this
> > being often mentioned on an irrational-purely-aesthetical
> > basis.
>
> I don't think the basis is irrational at all. It
> seems to me that there are two quite distinct possible
> reasons for deciding to make something a method of a
> particular object.
>
> One is to get polymorphism on that object. The other

So far, so good.

> is because the object makes sense as the direct object
> (in the English grammar sense) of the operation, so
> that you can naturally read x.verb(y) as "verb x
> with y" or "verb x using y".

And here, we disagree VERY deeply.  Natural language is
NO sensible basis on which to approach software design!
Please re-read Wittgenstein's works (the Philosophical
Investigations will suffice) for many deep, convincing
arguments on the issue -- W's youthful hopes to clean up
natural language of 'inappropriate superstructure' and
make it a usably solid basis for anything rigorous (as
vigorously, even brilliantly, pursued in the Tractatus)
being inevitably dashed by the deep nature of language.

Yeah, yeah, I know -- the world is full of assassinated
trees, tragically chomped into pulp for the purpose of
making paper on which such absurdities as basing software
design on natural-language foibles are _advised_.  But
then, in the literature for any subject you can find many
widespread examples of both obvious and subtle absurdities.

When you do O-O design in the context of a language that
does NOT force you to express any function as a method
of something, x.verb(y) MEANS that object x has some say
(which it may of course choose to delegate) on how verb
is implemented, while object y does not necessarily have
any such implementation-freedom aspect regarding verb.
(This assumes a single-dispatch setting -- multimethods,
such as Dylan's, may change the picture drastically).

If the language forces you to express everything as a
method, then the implications are correspondingly weaker --
but, still there to some extent, since you MIGHT have
chosen to introduce a third-party object to act (to all
intent and purposes) as a 'collection of functions', to
be able to write functions.verb(x,y) and avoiding the
(supposedly undesired) dissimetry between x and y's roles.


> When these two considerations pull in the same
> direction, all is well. And with most of the string
> methods they do, but in the case of join, they don't.
> There are good reasons for not making it a method of
> the sequence, but when it is made a method of the
> joiner, the resulting expression sounds backwards[1].

Suppose I'm building a line of a CSV file, a popular
textual format where fields are separated (aka joined)
with commas -- the acronym stands for 'comma-separated
values'.  So, what I want to do is
    'comma-separate these values'
and, since 'separate' equates to join, this equates to:
    'comma-join these values'
so, I code:
    comma.join(these_values)
*WHAT'S BACKWARDS ABOUT IT*?!  (Please note that the
CSV acronym IS an English one -- maybe _your_ English
native-speaker preference would have been for 'VSWC',
'Values Separated With Commas', but English native
speakers from _North_ of the Equator coined this term,
so, maybe it's a Coriolis-effect issue...?-).


> So we are faced with two alternatives, neither
> of which are entirely good. In such cases, my
> preference is not to choose either alternative --
> i.e. make join a function, not a method.

A definite technical loss on the altar of endlessly
debatable aesthetic grounds?  ***Include me out***!


Alex





From phd at phd.russ.ru  Mon Mar 26 11:07:00 2001
From: phd at phd.russ.ru (Oleg Broytmann)
Date: 26 Mar 2001 10:07:00 -0600
Subject: Dr. Dobb's Python-URL! - weekly Python news and links (Mar 26)
Message-ID: <FC718F7B3AF20F62.09ECC4E1E82F69E7.7E08A6AB55BAC1B9@lp.airnews.net>

    Guido van Rossum announces Python 2.1b2
        http://groups.yahoo.com/group/python-announce-list/message/1104
    
    On the heels of the announcement Sean Reifschneider announces Python
    2.1b2 SRPM
        http://groups.yahoo.com/group/python-announce-list/message/1105
    
    Fred Drake updates documentation for the second beta release of
    Python 2.1
        http://python.sourceforge.net/devel-docs/
    
    Andrew Kuchling reminds there are Python HOWTO Documents
        http://py-howto.sourceforge.net/
    
    It was a week of PEPs. Barry Warsaw updates PEP 1, PEP Purpose and
    Guidelines
        http://groups.yahoo.com/group/python-list/message/128991
    
    Andrew Kuchling writes PEP 241: Metadata for Python Software Packages
        http://groups.yahoo.com/group/python-announce-list/message/1099
    
    Clark Evans writes a new PEP for Query Protocol Interface; the
    discussion is deep
        http://groups.google.com/groups?lr=&group=comp.lang.python&safe=off&ic=1&th=41a1861d6f82a50d&seekd=911967025
        http://groups.google.com/groups?lr=&group=comp.lang.python&safe=off&ic=1&th=67d8c4d642a4b0b&seekd=911967026
    He applies for a PEP number
        http://groups.yahoo.com/group/python-list/message/129126
    Michel Pelletier reminds about PEP 245: Python interfaces
        http://groups.yahoo.com/group/python-list/message/129327
    Clark Evans comments
        http://groups.yahoo.com/group/python-list/message/129331
    
    Paul Dubois writes PEP 242: Numeric Kinds
        http://groups.yahoo.com/group/python-list/message/128862
    
    Sean Reifschneider makes PEP243: Module Repository Upload Mechanism
        http://groups.yahoo.com/group/python-list/message/129570
    and Distutils upload patch
        http://groups.yahoo.com/group/python-list/message/129671
    
    Moshe Zadka announces new list for discussion of Python's Numeric Model
        http://groups.yahoo.com/group/python-list/message/128911
    
    John Shipman reveals new Tkinter and PIL documentation
        http://groups.yahoo.com/group/python-list/message/129319
    84 pages in Postsvript or PDF format
    
    Object inspectors just like Smalltalk! Visually inspect objects as
    you code.  Written using tkinter; interfaces nicely into IDLE
        http://www.smalltalksource.org/anth/python/inspector.zip
    
    Thinking Outside the Box
        http://washingtonpost.com/wp-dyn/articles/A23937-2001Mar18.html
    The government should give you copyright. But in exchange, you owe
    me a buck every time your software crashes. (Not related to Python,
    just a good reading)
    
    TuxBot Programming with Python
        http://www.oreillynet.com/pub/a/python/2001/03/21/pythonnews.html
    
    Bruce Edge shows how to preserve history between interactive sessions
    (using readline, of course)
        http://groups.yahoo.com/group/python-list/message/128961
    
    Doug Fort publishes another cookie module
        http://groups.yahoo.com/group/python-list/message/129121
    
    Building Zope and Python on Mac OS X
        http://www.zope.org/Members/jshell/ZopeOnOSX
    
    Zope:
    Zope 2.3.1 beta 3 released
        http://lists.zope.org/pipermail/zope-announce/2001-March/000393.html
    
    An Interview with Paul Everitt and Ken Manheimer of Digital Creations
        http://oss4lib.org/readings/interview-everitt-manheimer-2001-03.php
    
    A Zope Spreadsheet Generator: Perl, Python, And DTML Working Together
        http://www.byte.com/column/BYT20010314S0002
    
    Chris Withers finally releases Squishdot 1.0.0
        http://lists.zope.org/pipermail/zope/2001-March/086183.html
    
    Hot threads of the week:
    New PEP: The directive statement
        http://groups.google.com/groups?lr=&group=comp.lang.python&safe=off&ic=1&th=b137a4ad27f7a6f9&seekd=911921785
    Static methods
        http://groups.google.com/groups?lr=&group=comp.lang.python.*&safe=off&ic=1&th=162da83d63248a97&seekd=910508206


========================================================================

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

      Python.org's Python Language Website is the traditional
      center of Pythonia
          http://www.python.org
      Notice especially the master FAQ
          http://www.python.org/doc/FAQ.html

      PythonWare complements the digest you're reading with the
      daily python url
           http://www.pythonware.com/daily  

      comp.lang.python.announce announces new Python software.  Be
      sure to scan this newly-revitalized newsgroup at least weekly.
	  http://groups.google.com/groups?oi=djq&as_ugroup=comp.lang.python.announce

      Michael Hudson continues Andrew Kuchling's marvelous tradition
      of summarizing action on the python-dev mailing list once every
      other week.
	  http://starship.python.net/crew/mwh/summaries/
	  http://www.amk.ca/python/dev

      The Vaults of Parnassus ambitiously collect Python resources
          http://www.vex.net/~x/parnassus/   

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

      The Python Software Foundation has replaced the Python Consortium
      as an independent nexus of activity
	  http://www.python.org/psf/

      Cetus does much of the same
          http://www.cetus-links.de/oo_python.html

      Python FAQTS
          http://python.faqts.com/

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

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

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

      Archive probing trick of the trade:
	  http://groups.google.com/groups?oi=djq&as_ugroup=comp.lang.python&num=100


Previous - (U)se the (R)esource, (L)uke! - messages are listed here:
  http://purl.org/thecliff/python/url.html
or
  http://groups.google.com/groups?oi=djq&as_q=+Python-URL!&as_ugroup=comp.lang.python


Suggestions/corrections for next week's posting are always welcome.
[http://www.egroups.com/list/python-url-leads/ is hibernating.  Just
e-mail us ideas directly.]

To receive a new issue of this posting in e-mail each Monday morning,
ask <claird at neosoft.com> to subscribe.  Mention "Python-URL!".


-- The Python-URL! Team--

Dr. Dobb's Journal (http://www.ddj.com) is pleased to participate in and
sponsor the "Python-URL!" project.
-- 

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


From emile at fenx.com  Sun Mar 18 18:15:07 2001
From: emile at fenx.com (Emile van Sebille)
Date: Sun, 18 Mar 2001 15:15:07 -0800
Subject: Help with new scope problem
References: <993bar$7h5$1@news.chatlink.com>
Message-ID: <993fnr$49h2h$1@ID-11957.news.dfncis.de>

Luis,

Your code doesn't use globals: all the variables are local to the function.
Also note that you check for <4, >4, >5 but never for anything =4.  I'd say
this is part of it.

Are you invoking this as:

box, gifname, newgifname = getarguments()

in your main code?

HTH,

Emile van Sebille
emile at fenx.com


"Luis Cortes" <lcortes at aodinc.com> wrote in message
news:993bar$7h5$1 at news.chatlink.com...
> Hello,
>
> I have a piece of code that I think should work fine, but I believe that
> I've hit a scope problem (Python 2.1b1).  Does anyone out there have a
hint
> as to how to fix it??
>
> code:
>
> def getarguments():
>
>         box = ( 160, 120 )
>         newgifname = 'new.gif'
>         gifname = ''
>
>         if (len(sys.argv) < 4 ):
>                 print 'usage: trans.py width height image.xxx
newimage.gif'
>                 sys.exit(0)
>
>         if (len(sys.argv) > 4):
>                 # program width, height, image.xxx, newimage.gif
>                 box = ( int(sys.argv[1]), int(sys.argv[2]) )
>                 gifname = sys.argv[3]
>
>         if (len(sys.argv) > 5 ):
>                 newgifname = sys.argv[4]
>
>         print box, gifname, newgifname
>         sys.exit(0)
>         return ( box, gifname, newgifname )
>
> THE PROBLEM:  when I have more than 3 arguments, the global variables do
not
> change to the correct variables, but instead print their default values.
> Help!
>
>
>




From danielk at aracnet.com  Thu Mar 29 10:06:31 2001
From: danielk at aracnet.com (Daniel Klein)
Date: Thu, 29 Mar 2001 07:06:31 -0800
Subject: Converting an integer base 10 to a binary number
References: <3AC1E898.D5929C44@fh-hof.de>
Message-ID: <d7j6ct0r6mu73c42jkfvucq87kl07khbe1@4ax.com>

On Wed, 28 Mar 2001 15:35:20 +0200, Peter St?hr <peter.stoehr at fh-hof.de> wrote:

>Hi out there,
>
>I know the int(x[,radix]) function that converts a string or a number
>(for a give radix) to an integer. Is there a built-in function that
>provides a conversion the other way round?
>Something like
>    bin(32) = "10000"

Methinks you will have to code this yourself. Hows about something like:

def toBinary(dec):
	bin = ''
	for x in range(8):
		bin = str(dec % 2) + bin
		dec = int(dec/2)
	return bin

>>> toBinary(32)
'00100000'

Of course this only works for 0 <= dec <= 255

Daniel Klein
Portland OR USA


From mcfletch at home.com  Tue Mar 20 14:24:47 2001
From: mcfletch at home.com (Mike C. Fletcher)
Date: Tue, 20 Mar 2001 14:24:47 -0500
Subject: Insertin **keywords into a class
In-Reply-To: <Pine.GHP.4.21.0103201151090.4991-100000@don.et.byu.edu>
Message-ID: <000001c0b173$6d7c9350$a9807018@cr706570a>

self.kw = keywords[kw] sets the attribute named 'kw'

setattr( self, kw, keywords[kw] ) would work, as would

self.__dict__.update( keywords ), though it would be somewhat scary for
security-paranoid folks ;) .

Enjoy,
Mike

-----Original Message-----
From: C. Porter Bassett [mailto:porter at et.byu.edu]
Sent: Tuesday, March 20, 2001 1:53 PM
To: python-list at python.org
Cc: python-list at python.org
Subject: Insertin **keywords into a class


cCan you please tell me why the following code snippet doesn't update the
value of radius?


class myClass:
   def __init__(self, *arguments, **keywords):
      self.radius = 0.0
      for kw in keywords.keys():
         print kw, ":", keywords[kw]
         self.kw = keywords[kw]
      print "self.radius =", self.radius


b = myClass(radius = 1.0)


--------------------------------------------------------------------------
"Pretend like this is a really witty saying." - Anonymous
--------------------------------------------------------------------------


--
http://mail.python.org/mailman/listinfo/python-list




From echuck at mindspring.com  Thu Mar 22 10:52:39 2001
From: echuck at mindspring.com (Chuck Esterbrook)
Date: Thu, 22 Mar 2001 10:52:39 -0500
Subject: [ANNOUNCE] Article: Using Mix-ins with Python
Message-ID: <5.0.2.1.0.20010322105220.0219eb00@mail.mindspring.com>

I am re-announcing this because at the time of the original announcement, 
it was not known that the article would be online. This announcement 
includes the new URL:

There is a Python article in the April 2001 issue of the Linux Journal, 
"Using Mix-ins with Python". It covers the general topic of mix-in 
programming with some comments on various languages, and then dives into 
Python specifics including examples.

http://noframes.linuxjournal.com/lj-issues/issue84/4540.html


I used these concepts when creating the MiddleKit component of Webware 
(http://webware.sourceforge.net/), which includes a mix-in function in 
Webware.MiscUtils.MixIn.


-Chuck
--
Get web dev at http://webware.sourceforge.net/ 




From topmind at technologist.com  Thu Mar 22 01:57:10 2001
From: topmind at technologist.com (topmind)
Date: Thu, 22 Mar 2001 06:57:10 GMT
Subject: Collection interfaces
References: <3a90fa97.0@news1.mweb.co.za>   <MPG.150394195f7f04a19896a7@news.earthlink.net> <3A9A86B1.9A8C69A2@ix.netcom.com> <MPG.1504d14563dce6499896af@news.earthlink.net> <MPG.1505a76831c34b749896b3@news.earthlink.net> <97h5s40vhc@news2.newsguy.com> <MPG.1506007591cf818b9896bc@news.earthlink.net> <slrn99opop.gr.amk@mira.erols.com> <MPG.1506265ebf6d934d9896be@news.earthlink.net> <97ibmc0k47@news1.newsguy.com> <MPG.1506cf3293b094b49896c1@news.earthlink.net> <97m0c7080c@news2.newsguy.com> <MPG.15085517debf4b649896ca@news.earthlink.net> <zb%n6.31033$o%6.586026@news1.rdc1.bc.home.com> <MPG.150b383cea226d409896d0@news.earthlink.net> <kNko6.6986$eq3.707121@news.uswest.net> <MPG.150c46db9cec283e9896d4@news.earthlink.net> <3AA2A898.450B659E@mail.com> <gRCt6.12224$Im6.1308560@newsread1.prod.itd.earthlink.net> <3AB755F8.67E84126@mail.com> <%a%t6.17245$227.1601569@newsread2.prod.itd.earthlink.net> <3AB8A730.4BE13D22@mail.com> <B6DE17D2.D6B0%t*b*r*u*m*l*e*y@f*u*s*e.n*e*t>
Message-ID: <anhu6.18497$Im6.1938203@newsread1.prod.itd.earthlink.net>

"Troy Brumley" <t*b*r*u*m*l*e*y at f*u*s*e.n*e*t> wrote in message
news:B6DE17D2.D6B0%t*b*r*u*m*l*e*y at f*u*s*e.n*e*t...
> in article 3AB8A730.4BE13D22 at mail.com, James A. Robertson at
> jarober at mail.com wrote on 3/21/01 8:06 AM:
>
> > topmind wrote:
> >>
> >
> >>> http://www.cincom.com/smalltalk
> >>>
> >>
> >> I only see a front page, not a parsing example.
> >>
> >
> > Follow the tutorial link - it's right there on that page.
> >
> >
> >> Besides, what are you comparing it to so say
> >> it is better than a procedural/relational approach?
> >
> > It's vastly easier than similar code done in C or basic, for example
>
> User testimonial here ... last week I had to rip thorugh a pile of IIS
logs
> to get some statistics for some people, and I used the original (non-gui)
> tutorial code as a starting point. I was once again impressed with how
much
> easier it was to do several common data processing tasks in Smalltalk than
> it was in Assembly language or C or Pascal.
>

Give me specific requirements and I bet I can match it on code size.

C, Pascal, and Assembly are hardly what I would use for code-size
contests.

Of course, there is more to comparing than code size, but those often
get very subjective or fuzzy.

> I'm a long time procedural type, learning to really think in and use
> Smalltalk has been an interesting experience.
>
> I've also done a lot of database "fixing" and Smalltalk has usually beat
out
> raw SQL for anything but the most trivial tasks.
>

Fixing what? SQL is not good at string parsing.

> All that said, I think it's the environment (interactive) that improves
the
> experience as much as the language itself. If I had a reasonable
> interpretive BASIC environment available, I could probably do many of the
> things I did in Smalltalk almost as easily ... certainly more easily than
> with C.
>

I find interpretive environments potentially more compact code-wise than
static ones.

>
> >
> >>
> >>>>
> >>>>> [snip]
> >>>>
> >>>> -tmind-
> >>>> oop.ismad.com
> >>>
> >>> --
> >>> James A. Robertson
> >>> Product Manager (Smalltalk), Cincom
> >>> jarober at mail.com
> >>> <Talk Small and Carry a Big Class Library>
> >>>
>




From paulp at ActiveState.com  Fri Mar 23 17:33:42 2001
From: paulp at ActiveState.com (Paul Prescod)
Date: Fri, 23 Mar 2001 14:33:42 -0800
Subject: Good GUI for Python
References: <002a01c0b33f$9595ade0$340210ac@winson>
Message-ID: <3ABBCF46.F3671763@ActiveState.com>

> ?i???q wrote:
> 
> Hi, All
> 
> I surveyed 3 GUI framework for Python including wxPython, PythonWin's
> included win32ui and Tkinter. My development environment is Mark
> Hammond's PythonWin downloaded from ActiveStates.

In general, the open source Python IDEs do not deal well with code
written in a GUI framework OTHER than the one the IDE was written in. So
Python MFC code will not work well in IDLE and Tk code does not work
well in PythonWin.

More advanced IDEs do not have this problem because they run the Python
programs in a separate process. ActiveState's Komodo is an example of an
IDE that can support any GUI (or even code embedded in Zope or Apache
etc.).
-- 
Take a recipe. Leave a recipe.  
Python Cookbook!  http://www.activestate.com/pythoncookbook



From schaffer at optonline.net  Fri Mar 23 17:06:56 2001
From: schaffer at optonline.net (Les Schaffer)
Date: Fri, 23 Mar 2001 22:06:56 GMT
Subject: VPython/Tk app works on single processor but not dual
References: <995gi1$9ce$1@bob.news.rcn.net> <acjmbtsg82rm7biao0e3cmnchff8et6qqr@4ax.com> <99fqm2$sp5$1@bob.news.rcn.net>
Message-ID: <l1inbt85rgeaqp6nrujlrdq5upajn06grf@4ax.com>

On 23 Mar 2001 15:38:42 GMT, fcy at enteract.com (Fred Yankowski) wrote:

>I really want to build VPython from source myself, so that I can
>instrument and tweak the C code to check out some theories I've got
>about why it's hanging.  If you've got a VC++ project file, I'd like
>to have that.  Is the VPython build process for Win32 documented
>anywhere?

yup. just download the sources from 

http://cil.andrew.cmu.edu/projects/visual/download/VisualSource-2001-02-27.zip

and unzip into your Python folder. i had to change the link line to
/MD from /MT, other than that, it works out of the box.

your residue.py script still runs fine on my machine with the latest
version of the source code and the visual python scripts
(Visual-2001-03-08).

>The "The Book of Numbers" book by Conway
>and Guy looks to be wonderful source of material that could be adapted
>into such programs.

thanks. was looking it over the other day at the store. will grab it
and give a closer look.

>P.S.  I'm using a 3DFx VooDoo-3 3000 16MB AGP video card.

do other GL programs run okay on your system? for example, do you have
the VTK stuff? does that run too?

>P.P.S.  I get hundreds of email messages every day.  I have seen that at
>least one of the mail servers involved will refuse incoming messages
>that appear to come from invalid domain names.

i send thousands of email messages and never had one bounce like this.



From aleaxit at yahoo.com  Mon Mar  5 04:37:21 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Mon, 5 Mar 2001 10:37:21 +0100
Subject: I come to praise .join, not to bury it...
References: <mailman.983485923.1136.python-list@python.org> <97p7iq$kk2$1@nntp6.u.washington.edu> <97rka002k9@news1.newsguy.com> <3AA2F4B3.A7496E76@cosc.canterbury.ac.nz>
Message-ID: <97vmoh061p@news2.newsguy.com>

"Greg Ewing" <greg at cosc.canterbury.ac.nz> wrote in message
news:3AA2F4B3.A7496E76 at cosc.canterbury.ac.nz...
> Alex Martelli wrote:
> >
> > I have not heard ANY _technical_ arguments opposed to
> > this last time the discussion went around -- nothing but
> > vague aesthetic, "it should be the other way", "I find it
> > ugly" kind of complaints.
>
> While your arguments are _technically_ sound, we are yet
> to be convinced that the benefits outweigh the drawbacks
> of being forced to use what many find to be an obscure and
> hard-to-follow syntax for all string joining.

Note that there is no 'forcing': once the right polymorphic
architecture has been chosen, one can still choose HIDE the
fact that polymorphism is happening behind a function call --
and, indeed, the string module supplies ready-made versions
of such calls (for backwards compatibility reasons, no doubt,
but, as a side effect, you can use it 'aesthetically'...!):

def replace(s, old, new, maxsplit=-1):
    return s.replace(old, new, maxsplit)

and so on, and so forth.  Personally, I prefer not to hide
such things (explicit better than implicit), but, you refined
aesthetes are being positively _pampered_ here -- a whole
module's worth of such syntax-sugar-providing functions...!


When you choose the right underlying architecture, the one
that best matches the actual needs, you MAY later decide to
'paper it over' with syntax-sugary superstructures if your
aesthetics so dictate -- since I am an engineer, I'm sure you
can guess what my attitude might be to such aesthetics, but
that's a different flamewar ("form follows function" crusaders,
this IS a call to arms!-).

The key point is that the actual *function* (the one which
I strongly opine 'form' should follow -- while acknowledging
the widespread human foible for 'dressing it up' in trompe
l'oeil 'prettiness':-) for joining IS: polymorphism on the
joiner object, access through standard sequence interface
for the sequence whose items are to be joined.

As long as that '_technically_ sound' architecture is the
underlying one (as, fortunately!, it is in Python 2), the
amount of actual damage you can do by covering it in your
chosen stucco and/or wallpaper is reasonably limited.

IF the underlying architecture did NOT reflect actual needs
(e.g., if the string.join function had to typeswitch on the
joiner-object to distinguish Unicode from single-byte --
and *every other* string module function had to similarly
typeswitch, rather than each being able to just delegate
to string-object's modules!), _then_ we'd be in deep doodoo.
But, fortunately, Python continued its tradition of pretty
good design indeed, introducing string methods in Python 2.


> The vast majority of string-joining operations in Python
> code that I've seen and written haven't had any need for
> polymorphism at all.

I guess you feel no need for Unicode -- what's the rest of
the world after all, we KNOW that all things that matter
only happen in countries where good ol' ASCII is quite
adequate for all of our string needs, right?


Alex





From effbot at hotmail.com  Thu Mar  8 12:58:52 2001
From: effbot at hotmail.com (Fredrik Lundh)
Date: Thu, 08 Mar 2001 17:58:52 -0000
Subject: news from python9!!!
Message-ID: <F113BpdFfQJE1G9IVir00015e95@hotmail.com>

>Shouldn't this read 'bugfix release of ActivePython 2.0' ?

yes.

(and the subject should probably read "news from active-
state's marketing department" ;-)

in other news, there were lots of great talks, pythoneers
were seen having a great time all over long beach, and
PythonWare showed the world's first refactoring browser
for Python.

you can find some other conference reports via:

   http://www.pythonware.com/daily

Cheers /F

_________________________________________________________________
Get your FREE download of MSN Explorer at http://explorer.msn.com




From stepheb at comm.mot.com  Wed Mar  7 12:56:40 2001
From: stepheb at comm.mot.com (Stephen Boulet)
Date: Wed, 07 Mar 2001 11:56:40 -0600
Subject: Extracting cell contents from html tables
Message-ID: <3AA67658.B39580CB@comm.mot.com>

Is there a module for extracting cell contents from html tables
(preferably into an array of strings)?

-- Stephen



From grey at despair.rpglink.com  Wed Mar 28 11:55:41 2001
From: grey at despair.rpglink.com (Steve Lamb)
Date: Wed, 28 Mar 2001 16:55:41 -0000
Subject: learning python...
References: <3AB8F4E1.184DD9C@noaa.gov> <a2uh99.o71.ln@pingu> <3AC21453.3A6F8524@noaa.gov>
Message-ID: <slrn9c45sd.qav.grey@teleute.dmiyu.org>

On Wed, 28 Mar 2001 11:41:55 -0500, Benjamin.Altman <benjamin.altman at noaa.gov>
wrote:
>What about the aspect of it covering a version less than 2.0 when
>Programming Python would be more up to date?

    Actually, for <2.0 learning for someone who has programming experience I'd
say go with the Beazley Book, Python Essential Reference.  It is a concise,
well written reference book on the language.  It doesn't try to teach
programming, it is a reference to the Python language with enough attentioned
paid to how to program in Python for the experienced programmer to pick it up
quite quickly.

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


From Klaus.Bruns at teleatlas.com  Tue Mar 27 09:40:13 2001
From: Klaus.Bruns at teleatlas.com (Klaus Bruns)
Date: Tue, 27 Mar 2001 16:40:13 +0200
Subject: Combining Lists
References: <99q887$hl9$1@ctb-nnrp2.saix.net>
Message-ID: <3AC0A64C.B9DB6353@teleatlas.com>

Mix wrote:

> How does one take two or more lists and combined them?
> For example, suppose one has the following 2 lists :
>  list1=['1','2']
>   and
>  list2=['3','9,'15']
>
> Now how  does one get the following list
>  list3=['1','2','3','9','15']
> from a combing the two lists (i.e list1 and list2)?
>
> Thanks.
> Mix

Try this:

list3 = list1 + list2

--
Klaus



From fgeiger at datec.at  Fri Mar  9 01:56:14 2001
From: fgeiger at datec.at (Franz GEIGER)
Date: Fri, 9 Mar 2001 07:56:14 +0100
Subject: "Wrap" a python-cgi
References: <3aa7c5f9$0$21482@SSP1NO25.highway.telekom.at>
Message-ID: <989utg$8b6$1@newsreaderm1.core.theplanet.net>

Have a look at Webware (currently 0.5). There's a lot more you can rely on
to make your "CGI life" easier...

Regards
Franz

"Hermann Himmelbauer" <dusty at strike.wu-wien.ac.at> wrote in message
news:3aa7c5f9$0$21482 at SSP1NO25.highway.telekom.at...
> Hi,
> I am currently using python for several cgi programs. One problem is that
> if my cgi-script has a bug it exits with an exception and the user gets an
> ugly error message done by the webserver.
>
> I would like to put some really simple "bugfree" (if something like this
> can exist :-) "wrapper" around my cgi-scripts, let it trap Exceptions, log
> the error and let it output a nice error page to the user.
>
> My first approach would perhaps be to wrap the whole script into a "try:
> except" clause but at first I don't know if nested try/except clauses are
> allowed and moreover the whole script looks somehow ugly as every line is
> indented.
>
> Perhaps you know a better solution?
>
>                 Best Regards,
>                 Hermann
>
> --
>  ,_,
> (O,O)     "There is more to life than increasing its speed."
> (   )     -- Gandhi
> -"-"--------------------------------------------------------------




From cce at clarkevans.com  Wed Mar 28 22:34:58 2001
From: cce at clarkevans.com (Clark C. Evans)
Date: Wed, 28 Mar 2001 22:34:58 -0500 (EST)
Subject: FYI: PEP 246: Object Adaptation
In-Reply-To: <Pine.LNX.4.21.0103201837570.21343-100000@clarkevans.com>
References: <Pine.LNX.4.21.0103201837570.21343-100000@clarkevans.com>
Message-ID: <Pine.LNX.4.21.0103282232500.14408-100000@clarkevans.com>

http://python.sourceforge.net/peps/pep-0246.txt

Thus far I have one comment:

  * Make AdaptForceFailError more consistent with Python exception naming

Your comments would be appreciated.

Thank you!

Clark





From mjbjr at best.com  Tue Mar 20 03:05:02 2001
From: mjbjr at best.com (Martin Brown)
Date: Tue, 20 Mar 2001 08:05:02 GMT
Subject: Python2.0 rpm's don't have IDLE?
Message-ID: <OaEt6.2328$Up.121023@sea-read.news.verio.net>

I've downloaded and installed the 3 Python2.0 rpm's retrieved:

   BeOpen-Python-2.0-1.i386.rpm
   Tkinter-2.0-8.0-3.i386.rpm
   expat-1.1-1.i386.rpm

I've checked the installation directories, but there doesn't seem to be a
copy of IDLE.  The web site says IDLE 0.6 is included in Python2.0.  Also,
as a separate package, apparently, only version 0.5 is available.

Suggestions on how to get version 0.6 of IDLE into my Python2.0
installation greatly appreciated.

--
                            - Martin J. Brown, Jr. -                           
                            - mjbjr at beaudesign.com -                           
                                                                               
    PGP Public Key ID: 0xCED9BD8A  Key Server: http://www.keyserver.net/en/


From fredrik at pythonware.com  Tue Mar 13 05:09:03 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Tue, 13 Mar 2001 10:09:03 GMT
Subject: file - codecs - unicode ???
References: <3aace9da$0$382$446d705c@news.skynet.be> <slrn9aqka1.3vk.qrczak@qrnik.zagroda> <3aadf0bf$0$384$5a62924d@news.skynet.be>
Message-ID: <3lmr6.6$4d3.1784@newsb.telia.net>

S?bastien Libert wrote:

> >>> f=open("c:\\mylog.log")

better make that:

    f=open("c:\\mylog.log", "rb")

> >>> fline=f.readline()
> >>> fline=unicode(fline,'UTF-16').encode('utf-8')

Cheers /F




From stephen_purcell at yahoo.com  Tue Mar 27 04:28:05 2001
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Tue, 27 Mar 2001 11:28:05 +0200
Subject: Pickling functions
In-Reply-To: <sac0d8c7.078@dohprimo.ccf.health.nsw.gov.au>; from TCHUR@doh.health.nsw.gov.au on Tue, Mar 27, 2001 at 06:15:29PM +1000
References: <sac0d8c7.078@dohprimo.ccf.health.nsw.gov.au>
Message-ID: <20010327112805.A21122@freedom.puma-ag.com>

Tim CHURCHES wrote:
> How does one go about pickling and unpickling a function object? All my
> attempts so far result in a reference to the function object being
> pickled but not the function itself.


No promises, but it looks like you could pickle a list of

 [func.func_code, func.func_globals, func.__name__, func.func_defaults]

and then re-create the original function by unpickling that list and
passing it to 'new.function'.

(see http://www.python.org/doc/lib/module-new.html)

You should probably avoid pickling func.func_globals, though.

-Steve

-- 
Steve Purcell, Pythangelist
Get testing at http://pyunit.sourceforge.net/
Any opinions expressed herein are my own and not necessarily those of Yahoo



From mhuening at zedat.fu-berlin.de  Tue Mar 13 06:44:08 2001
From: mhuening at zedat.fu-berlin.de (Matthias Huening)
Date: Tue, 13 Mar 2001 12:44:08 +0100
Subject: What's wrong with me/Python ?
References: <3AAE0652.D68F253E@daimlerchrysler.com>
Message-ID: <98l16c$29cvf$1@fu-berlin.de>

>What did I do wrong? And why does Python want a "variable
>named 'string'" here?

Because you didn't import the string-module...

Matthias




From sholden at holdenweb.com  Tue Mar 27 08:53:45 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Tue, 27 Mar 2001 08:53:45 -0500
Subject: file locking
References: <5zCv6.583$p5.2667@news1.rivrw1.nsw.optushome.com.au> <mailman.985596999.26319.python-list@python.org> <99n47u02fk@news2.newsguy.com>
Message-ID: <n%0w6.10132$2q.437602@e420r-atl1.usenetserver.com>

"Alex Martelli" <aleaxit at yahoo.com> wrote in message
news:99n47u02fk at news2.newsguy.com...
> "Steve Purcell" <stephen_purcell at yahoo.com> wrote in message
> news:mailman.985596999.26319.python-list at python.org...
> > deadmeat wrote:
> > > How do I lock a file on opening it?  the bultin open() only allows me
to
> set
> > > the read/write mode, not any deny modes.
> >
> > On Unix, look at the 'flock' function in the module 'fcntl'.
> >
> > You can also try the 'posixfile' module -- posixfile objects support
> locking.
>
> On Win32, look at module win32file; its CreateFile function has
> a shareMode parameter.  Getting a Python fileobject from the
> HANDLE returned by CreateFile is a tad tedious -- I _think_ you
> have to use win32file._open_osfhandle to get an (integer) C
> level handle from the opaque HANDLE, then os.fdopen to get from
> said int-handle to a Python fileobject.
>
Perhaps another place where the platform-specific libraries need a common
shim layer over them to give a portable API. Does the Mac have any quirks in
this area?

regards
 Steve





From nas at arctrix.com  Sat Mar 17 22:30:54 2001
From: nas at arctrix.com (Neil Schemenauer)
Date: Sat, 17 Mar 2001 19:30:54 -0800
Subject: reading stdinput
In-Reply-To: <Pine.LNX.4.10.10103181104580.12584-100000@localhost.localdomain>; from admin@mulga.com.au on Sun, Mar 18, 2001 at 11:11:00AM +0930
References: <Pine.LNX.4.10.10103181104580.12584-100000@localhost.localdomain>
Message-ID: <20010317193054.A12376@glacier.fnational.com>

On Sun, Mar 18, 2001 at 11:11:00AM +0930, Michael Hall wrote:
> How do I read standard input into a Python script? I'm trying to do the
> same as 'read' in a shell script to create a variable. I've played about
> with various combinations of 'sys.stdin.read()' and such but no go.

You haven't defined your problem very well.  sys.stdin.read()
will indeed read from stdin.  Perhaps you want
sys.stdin.readline() or raw_input()?  For example:

    x = raw_input("Your name: ")
    print "Hello", x

Cheers,

  Neil



From qrczak at knm.org.pl  Tue Mar  6 14:14:50 2001
From: qrczak at knm.org.pl (Marcin 'Qrczak' Kowalczyk)
Date: 6 Mar 2001 19:14:50 GMT
Subject: Long Shifty Bits
References: <kmap6.15$8Q.28605@news.pacbell.net>
Message-ID: <slrn9aadpa.uri.qrczak@qrnik.zagroda>

Tue, 6 Mar 2001 10:50:21 -0800, Phlip <phlip_cpp at my-deja.com> pisze:

> I want this expression to make sense:
> 
>         x = 1L << 200

It does make sense. At least in Python 2.0 I get
1606938044258990275541962092341162602522202993782792835301376L
and Haskell gives the same answer so they are probably both right :-)

-- 
 __("<  Marcin Kowalczyk * qrczak at knm.org.pl http://qrczak.ids.net.pl/
 \__/
  ^^                      SYGNATURA ZAST?PCZA
QRCZAK


From larskl at klassekampen.no  Tue Mar 27 10:44:35 2001
From: larskl at klassekampen.no (Lars Klęboe)
Date: Tue, 27 Mar 2001 17:44:35 +0200
Subject: Changing every other instance of <B> in a file
References: <mailman.985705112.10486.python-list@python.org>
Message-ID: <sqc1ct8km0a7p84pdj7k450k9nqe3r1v05@4ax.com>

On Tue, 27 Mar 2001 09:49:39 -0500, Jonathan Soons <jsoons at juilliard.edu>
wrote:
>>>Here's one way (although I'm interested to see how re might do this):
>>
>>[--- interresting code ---]
>>
>>This is another (fancier ;)) implementation of the solution I've
>>been working on, but I still refuse to believe that the script has
>>to be this elaborate/big. *twisting my tired gray matter*
>>
>>I guess re would do the trick (at least if combined with a simple
>>loop, I'll look into it when I'm really awake.
>>
>This seems to fail for tags that do not have a closing tag, which occurs
>frequently in non-automated HTML.

The quark files are checked before they reach my script, and so far
not a single error has occured. Besides the minimal risk, my post check
of the HTML will produce a warning - and even if this check fails,
an error for every umphteenth thousand file is acceptable for this task.

-- 
Lars Kl?boe



From sav at ulmen.mv.ru  Sun Mar 11 16:47:49 2001
From: sav at ulmen.mv.ru (Alexander Semenov)
Date: Mon, 12 Mar 2001 00:47:49 +0300
Subject: any interest in type-scoped static constants?
References: <3AAB9578.4C8754E4@fxtech.com>
Message-ID: <98gs2i$2vtp$1@news1.simtel.ru>

"Paul Miller" <paul at fxtech.com> wrote in message
news:3AAB9578.4C8754E4 at fxtech.com...
> If I write an extension type that mirrors the functionality of a C++
> class, and the C++ has static constants within its scope, I can do this
> in C++:
>
> Point p = Point::ZERO;
>
> Where ZERO is defined as:
>
> class Point
> {
> static Point ZERO = Point(0, 0);
> };
>
> If I create an extension type for Point, I can't do this:
>
> p = Point.ZERO

In Python:

class Point:
    def __init(self, x, y): pass

Point.ZERO = Point(0,0)

Then you can write p = Point.ZERO.
It's even simpler and clearer then C++

WBR,
Alexander Semenov.





From wolfson at midway.uchicago.edu  Mon Mar 26 20:00:24 2001
From: wolfson at midway.uchicago.edu (Ben Wolfson)
Date: Tue, 27 Mar 2001 01:00:24 GMT
Subject: fixing map(None, [1,2,3]) ?
References: <mailman.985648235.13881.python-list@python.org>
Message-ID: <ICRv6.122$E4.5826@uchinews>

In article <mailman.985648235.13881.python-list at python.org>,
Sean 'Shaleh' Perry <shaleh at valinux.com> wrote:
>
>On 26-Mar-2001 Mike Coleman wrote:
>> Shouldn't "map(None, [1,2,3])" return "[(1,), (2,), (3,)]"?
>> 
>> Is there any hope that this could be fixed?
>> 
>
>map runs the function on each element of the list it is passed and returns an
>element of a new list.  Why should it magically return a list of one-tuples?


>>> def t(*a): return a
...
>>> map(None, range(3), range(4,7))
[(0,4), (1,5), (2,6)]
>>> map(t, range(3), range(4,7)) #same as None
[(0,4), (1,5), (2,6)]

>>> map(None, range(3))
[0, 1, 2]
>>> map(t, range(3))             #differnt
[(0,), (1,), (2,)]





-- 
Barnabas T. Rumjuggler
Fried Something with Noodle:  Very good for ugly, sad, lonely, sick
people with the blood and the grasses all over the floor...
 -- Joe Frank, _A Call in the Night_


From bos at hack.org  Tue Mar 27 01:13:34 2001
From: bos at hack.org (Rikard Bosnjakovic)
Date: Tue, 27 Mar 2001 08:13:34 +0200
Subject: math.choice attribute error
References: <Pine.LNX.4.21.0103270124290.3010-100000@moroni.pp.asu.edu>
Message-ID: <3AC02F8E.A04D502D@hack.org>

iddwb wrote:

> from the ref man math.choice says it takes a sequence object and returns a
> random choice. perfect for my needs. the follow happens:
[...]
> what have I missed?

The method choice lives in the module random, which the math-module
imports (IIRC). Try to import random and run choice instead, it should
work.

   >>> import random
   >>> random.choice((1,42,4711))
   42


-- 
Rikard Bosnjakovic - http://a214.ryd.student.liu.se/cv/ - ICQ: 1158217

Anyone sending unwanted advertising e-mail to my address will be
charged $250 for network traffic and computing time. By extracting my
address from this message or its header, you agree to these terms.


From mwh21 at cam.ac.uk  Thu Mar 29 17:33:07 2001
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 29 Mar 2001 23:33:07 +0100
Subject: Best Python documentation ->HTML?
References: <pg0qbt0e598g1ecnfj5lj0lpa1q1pqvagq@4ax.com>
Message-ID: <m37l18cijg.fsf@atrus.jesus.cam.ac.uk>

Brian Glendenning <bglenden at hotmail.com> writes:

> I am a Python newbie and am confused about what the best way to turn
> Python documentation into HTML is (i.e., a "javadoc" equivalent). If
> it could also make good printed documentation (e.g., page numbers) as
> well as HTML, all the better.

If you know latex, writing howtos is pretty easy, and produces nice
printed and HTML output (the latter only with the luck and following
wind you need to get latex2html going, but you only have to do that
once).

Cheers,
M.

-- 
  If you give someone Fortran, he has Fortran.
  If you give someone Lisp, he has any language he pleases.
    -- Guy L. Steele Jr, quoted by David Rush in comp.lang.scheme.scsh


From JamesL at Lugoj.Com  Tue Mar  6 23:38:12 2001
From: JamesL at Lugoj.Com (James Logajan)
Date: Tue, 06 Mar 2001 20:38:12 -0800
Subject: Open Source Organic Chemistry software project
References: <taaseuid59k047@corp.supernews.com> <QYep6.29320$1D5.1247980@e420r-atl1.usenetserver.com>
Message-ID: <3AA5BB34.722ED2C4@Lugoj.Com>

Steve Holden wrote:
> 
> Don't forhet to solicit the chemictry community, much of which won't read
> c.l.py
> 
> not-done-organic-chemistry-in-thirty-five-years-ly y'rs  - steve

Same for the nanotechnology community, some of whom haunt sci.nanotech.
Always useful to have more tools to work with!

> 
> "mapMonkey" <jziniti at speakeasy.org> wrote in message
> news:taaseuid59k047 at corp.supernews.com...
> > Pythonistas,
> >
> > I am encouraging my current employer to GPL at least some of
> > our code and I would like to see what kind of interest I can
> > generate from the Python community ( everyone's favorite
> > language ;).  The amount of interest is very likely to swing the
> > vote ...
...
> >
> > John Ziniti


From gReGlAnDrUm at earthlink.net  Fri Mar  2 10:28:04 2001
From: gReGlAnDrUm at earthlink.net (Greg Landrum)
Date: Fri, 2 Mar 2001 07:28:04 -0800
Subject: Has anyone been able to convince Bruce Eckel to write "Thinking in Python"
References: <mailman.983480583.14914.python-list@python.org>
Message-ID: <97oe4g$g5q$1@nntp9.atl.mindspring.net>

"Chris Richard Adams" <chrisa at ASPATECH.COM.BR> wrote in message
news:mailman.983480583.14914.python-list at python.org...
If anyone has worked with Java - they would know how Bruce's bible
'Thinking in Java' has helped thousands of developers.  Could we be as
fortunate with him considering a Python book?? Can only pray and beg.


http://www.bruceeckel.com/Python/ThinkingInPython.html

-greg




From crawford at goingware.com  Mon Mar  5 03:20:59 2001
From: crawford at goingware.com (Michael D. Crawford)
Date: Mon, 05 Mar 2001 04:50:59 -0330
Subject: Test Your Web Apps - article at LinuxQuality
Message-ID: <3AA34C6B.EC12B25B@goingware.com>

Tonight I posted:

Use Validators and Load Generators to Test Your Web Applications
http://linuxquality.sunsite.dk/articles/webapptesting/

The article promotes the idea that one should check the HTML
generated by a web application with a validator to ensure that
it conforms to W3C standards, and that you should use a load
generator to stress-test your server during development to
detect performance problems early.

Links to some load generators and validators are given, with
tips on using them.

I further suggest that if you generate XHTML documents rather
than HTML, you can take advantage of the many available
XML software packages to validate and process your documents.

Comments, criticisms, corrections and suggested additions are
appreciated.

The Linux Quality Database at http://linuxquality.sunsite.dk/
has the dual purpose of promoting higher quality among Free
and Open Source software by publishing articles such as this,
and also the development of an easy-to-use but powerful bug
database to enable widespread public quality assurance of the
linux kernel.  

You're welcome to contribute articles on the general topics
of quality assurance and better programming practice.  I'm also
looking for knowledgeable people who can help me design the schema
for the bug database.

-- 
Michael D. Crawford
GoingWare Inc. - Expert Software Development and Consulting
http://www.goingware.com
crawford at goingware.com

  Tilting at Windmills for a Better Tomorrow.


From tjg at exceptionalminds.com  Fri Mar  9 23:03:19 2001
From: tjg at exceptionalminds.com (Timothy Grant)
Date: Fri, 9 Mar 2001 20:03:19 -0800
Subject: gah! I hate the new string syntax
In-Reply-To: <98c4rn$flq$1@panix2.panix.com>; from aahz@panix.com on Fri, Mar 09, 2001 at 06:51:35PM -0800
References: <mailman.983615170.23313.python-list@python.org> <98asrq$645$1@panix6.panix.com> <G5eq6.47941$ML1.2142034@e420r-atl2.usenetserver.com> <jwbnews-8E2915.18132009032001@corp.supernews.com> <98c4rn$flq$1@panix2.panix.com>
Message-ID: <20010309200319.A2402@trufflehunter.avalongroup.net>

> But that's a US ship.  English Ships Are Different.
> -- 
>                       --- Aahz (Copyright 2001 by aahz at pobox.com)
As a dual citizen Britain/US, I can vouch for Danish ships as
that is what I arrived in this country on.

BTW: weren't all y'all just next door to one of the most famous
British ships of all time?
-- 
Stand Fast,
    tjg.

Timothy Grant                         tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.         <><       (503) 246-3630
>>>>>>>>>>>>>Linux, because rebooting is *NOT* normal<<<<<<<<<
>>>>This machine was last rebooted:  52 days  8:15 hours ago<<



From jurgen.defurne at philips.com  Wed Mar 14 03:30:43 2001
From: jurgen.defurne at philips.com (jurgen.defurne at philips.com)
Date: Wed, 14 Mar 2001 09:30:43 +0100
Subject: This math scares me
Message-ID: <0056900016584737000002L072*@MHS>

If you search the internet for 'Gary Kildall' you will find that he already
had a version of CP/M before the Altair, that he developed together
with Intel.

Jurgen




sdm7g at virginia.edu@SMTP at python.org on 14/03/2001 00:05:03
Sent by:	python-list-admin at python.org
To:	grante at visi.com@SMTP
cc:	python-list at python.org@SMTP 
Subject:	Re: This math scares me
Classification:	



On Tue, 13 Mar 2001, Grant Edwards wrote:

> In article <jwbnews-09D75F.13313313032001 at corp.supernews.com>, John W. Baxter wrote:
> >In article <984518199.127140 at newsmaster-04.atnet.at>,
> > "Werner Schiendl" <ws-news at gmx.at> wrote:
> >
> >Bill Gates got this right in the early days, delivering two
> >Basic interpreters: binary floating point for speed, and
> >decimal floating point for money and such.
>
> I don't know if you can credit Gates with doing anything other
> than following the pack.  Back when I first started doing
> software (CP/M 1.7), all language packages for "personal"
> computers came with both BCD and binary FP math libraries so
> you could do exact monetary calculations if you wanted.

Exactly what pack do you believe Gates was following ?

You do know (don't you?) that Bill's BASIC predates not just
MS-DOS (which microsoft bought from another company) but CP/M
and just about any other "PC" software (and most of the hardware)
since the first version was written to run on the Altair computer --
which was the first microprocessor based personal computer.
( I think you probably had to boost the memory up to 4K to get
  Basic to run. )

> Of course we'd never imagined we'd ever see hardware FP and BCD
> libraries were only marginally slower than binary ones. Now
> that FP is usually in hardware, BCD FP (or fixed point) will be
> pretty slow by comparison.

-- Steve Majewski



--
http://mail.python.org/mailman/listinfo/python-list





From moshez at zadka.site.co.il  Fri Mar 16 08:18:43 2001
From: moshez at zadka.site.co.il (Moshe Zadka)
Date: Fri, 16 Mar 2001 15:18:43 +0200
Subject: [Very Long (11K)] Numeric PEPs, first public posts
Message-ID: <E14du7v-0004Xn-00@darjeeling>

After the brouhaha at IPC9, it was decided that while PEP-0228 should stay
as a possible goal, there should be more concrete PEPs suggesting specific
changes in Python numerical model, with implementation suggestions and
migration paths fleshed out. So, there are four new PEPs now, all proposing
changes to Python's numeric model. There are some connections between them,
but each is supposed to be accepted or rejected according to its own merits.

To facilitate discussion, I'm including copies of the PEPs concerned
(for reference purposes, these are PEPs 0237-0240, and the latest public
version is always in the Python CVS under non-dist/peps/ . A reasonably
up to date version is linked from http://python.sourceforge.net)

Please direct all future discussion to python-numerics at lists.sourceforge.net
This list has been especially set-up to discuss those subjects.

PEP: 237
Title: Unifying Long Integers and Integers
Version: $Revision: 1.2 $
Author: pep at zadka.site.co.il (Moshe Zadka)
Status: Draft
Type: Standards Track
Created: 11-Mar-2001
Python-Version: 2.2
Post-History:


Abstract

    Python has both integers (machine word size integral) types, and
    long integers (unbounded integral) types.  When integers
    operations overflow the machine registers, they raise an error.
    This PEP proposes to do away with the distinction, and unify the
    types from the perspective of both the Python interpreter and the
    C API.


Rationale

    Having the machine word size exposed to the language hinders
    portability.  For examples Python source files and .pyc's are not
    portable because of this.  Many programs find a need to deal with
    larger numbers after the fact, and changing the algorithms later
    is not only bothersome, but hinders performance in the normal
    case.


Literals

    A trailing 'L' at the end of an integer literal will stop having
    any meaning, and will be eventually phased out.  This will be done
    using warnings when encountering such literals.  The warning will
    be off by default in Python 2.2, on for 12 months, which will
    probably mean Python 2.3 and 2.4, and then will no longer be
    supported.


Builtin Functions

    The function long() will call the function int(), issuing a
    warning.  The warning will be off in 2.2, and on for two revisions
    before removing the function.  A FAQ will be added to explain that
    a solutions for old modules are:

         long=int

    at the top of the module, or:

         import __builtin__
         __builtin__.long=int

    In site.py.


C API

    All PyLong_As* will call PyInt_As*.  If PyInt_As* does not exist,
    it will be added.  Similarly for PyLong_From*.  A similar path of
    warnings as for the Python builtins will be followed.


Overflows

    When an arithmetic operation on two numbers whose internal
    representation is as machine-level integers returns something
    whose internal representation is a bignum, a warning which is
    turned off by default will be issued.  This is only a debugging
    aid, and has no guaranteed semantics.


Implementation

    The PyInt type's slot for a C long will be turned into a 

        union {
            long i;
            struct {
                unsigned long length;
                digit digits[1];
            } bignum;
        };

    Only the n-1 lower bits of the long have any meaning; the top bit
    is always set.  This distinguishes the union.  All PyInt functions
    will check this bit before deciding which types of operations to
    use.


Jython Issues

    Jython will have a PyInt interface which is implemented by both
    from PyFixNum and PyBigNum.


Open Issues

    What to do about sys.maxint?

    What to do about PyInt_AS_LONG failures?

    What do do about %u, %o, %x formatting operators?

    How to warn about << not cutting integers?

    Should the overflow warning be on a portable maximum size?

    Will unification of types and classes help with a more straightforward
    implementations?


Copyright

    This document has been placed in the public domain.


PEP: 238
Title: Non-integer Division
Version: $Revision: 1.1 $
Author: pep at zadka.site.co.il (Moshe Zadka)
Status: Draft
Type: Standards Track
Created: 11-Mar-2001
Python-Version: 2.2
Post-History:


Abstract

    Dividing integers currently returns the floor of the quantities.
    This behavior is known as integer division, and is similar to what
    C and FORTRAN do.  This has the useful property that all
    operations on integers return integers, but it does tend to put a
    hump in the learning curve when new programmers are surprised that

        1/2 == 0

    This proposal shows a way to change this while keeping backward
    compatibility issues in mind.


Rationale

    The behavior of integer division is a major stumbling block found
    in user testing of Python.  This manages to trip up new
    programmers regularly and even causes the experienced programmer
    to make the occasional mistake.  The workarounds, like explicitly
    coercing one of the operands to float or use a non-integer
    literal, are very non-intuitive and lower the readability of the
    program.


// Operator

    A `//' operator which will be introduced, which will call the
    nb_intdivide or __intdiv__ slots.  This operator will be
    implemented in all the Python numeric types, and will have the
    semantics of

        a // b == floor(a/b)

    Except that the type of a//b will be the type a and b will be
    coerced into.  Specifically, if a and b are of the same type, a//b
    will be of that type too.


Changing the Semantics of the / Operator

    The nb_divide slot on integers (and long integers, if these are a
    separate type, but see PEP 237[1]) will issue a warning when given
    integers a and b such that

        a % b != 0

    The warning will be off by default in the 2.2 release, and on by
    default for in the next Python release, and will stay in effect
    for 24 months.  The next Python release after 24 months, it will
    implement

        (a/b) * b = a (more or less)

    The type of a/b will be either a float or a rational, depending on
    other PEPs[2, 3].


__future__

    A special opcode, FUTURE_DIV will be added that does the
    equivalent of:

        if type(a) in (types.IntType, types.LongType):
           if type(b) in (types.IntType, types.LongType):
               if a % b != 0:
                    return float(a)/b
        return a/b

    (or rational(a)/b, depending on whether 0.5 is rational or float).

    If "from __future__ import non_integer_division" is present in the
    module, until the IntType nb_divide is changed, the "/" operator
    is compiled to FUTURE_DIV.


Open Issues

    Should the // operator be renamed to "div"?


References

    [1] PEP 237, Unifying Long Integers and Integers, Zadka,
        http://python.sourceforge.net/peps/pep-0237.html

    [2] PEP 239, Adding a Rational Type to Python, Zadka,
        http://python.sourceforge.net/peps/pep-0239.html

    [3] PEP 240, Adding a Rational Literal to Python, Zadka,
        http://python.sourceforge.net/peps/pep-0240.html


Copyright

    This document has been placed in the public domain.


PEP: 239
Title: Adding a Rational Type to Python
Version: $Revision: 1.1 $
Author: pep at zadka.site.co.il (Moshe Zadka)
Status: Draft
Type: Standards Track
Created: 11-Mar-2001
Python-Version: 2.2
Post-History:


Abstract

    Python has no numeric type with the semantics of an unboundedly
    precise rational number.  This proposal explains the semantics of
    such a type, and suggests builtin functions and literals to
    support such a type.  This PEP suggests no literals for rational
    numbers; that is left for another PEP[1].


Rationale

    While sometimes slower and more memory intensive (in general,
    unboundedly so) rational arithmetic captures more closely the
    mathematical ideal of numbers, and tends to have behavior which is
    less surprising to newbies.  Though many Python implementations of
    rational numbers have been written, none of these exist in the
    core, or are documented in any way.  This has made them much less
    accessible to people who are less Python-savvy.


RationalType

    There will be a new numeric type added called RationalType.  Its
    unary operators will do the obvious thing.  Binary operators will
    coerce integers and long integers to rationals, and rationals to
    floats and complexes.

    The following attributes will be supported: .numerator and
    .denominator.  The language definition will not define these other
    then that:

        r.denominator * r == r.numerator

    In particular, no guarantees are made regarding the GCD or the
    sign of the denominator, even though in the proposed
    implementation, the GCD is always 1 and the denominator is always
    positive.

    The method r.trim(max_denominator) will return the closest
    rational s to r such that abs(s.denominator) <= max_denominator.


The rational() Builtin

    This function will have the signature rational(n, d=1).  n and d
    must both be integers, long integers or rationals.  A guarantee is
    made that

        rational(n, d) * d == n


References

    [1] PEP 240, Adding a Rational Literal to Python, Zadka,
        http://python.sourceforge.net/peps/pep-0240.html


Copyright

    This document has been placed in the public domain.


PEP: 240
Title: Adding a Rational Literal to Python
Version: $Revision: 1.1 $
Author: pep at zadka.site.co.il (Moshe Zadka)
Status: Draft
Type: Standards Track
Created: 11-Mar-2001
Python-Version: 2.2
Post-History:


Abstract

    A different PEP[1] suggests adding a builtin rational type to
    Python.  This PEP suggests changing the ddd.ddd float literal to a
    rational in Python, and modifying non-integer division to return
    it.


Rationale

    Rational numbers are useful, and are much harder to use without
    literals.  Making the "obvious" non-integer type one with more
    predictable semantics will surprise new programmers less then
    using floating point numbers.


Proposal

    Literals conforming to the regular expression '\d*.\d*' will be
    rational numbers.


Backwards Compatibility

    The only backwards compatible issue is the type of literals
    mentioned above.  The following migration is suggested:

    1. "from __future__ import rational_literals" will cause all such
       literals to be treated as rational numbers.

    2. Python 2.2 will have a warning, turned off by default, about
       such literals in the absence of a __future__ statement.  The
       warning message will contain information about the __future__
       statement, and indicate that to get floating point literals,
       they should be suffixed with "e0".

    3. Python 2.3 will have the warning turned on by default.  This
       warning will stay in place for 24 months, at which time the
       literals will be rationals and the warning will be removed.


References

    [1] PEP 239, Adding a Rational Type to Python, Zadka,
        http://python.sourceforge.net/peps/pep-0239.html


Copyright

    This document has been placed in the public domain.



From not.this at seebelow.org  Tue Mar 20 13:46:38 2001
From: not.this at seebelow.org (Grant Griffin)
Date: 20 Mar 2001 10:46:38 -0800
Subject: New PEP: The directive statement
References: <200103201042.LAA20248@pandora.informatik.hu-berlin.de> <mailman.985107140.4586.python-list@python.org>
Message-ID: <9988ie02k7u@drn.newsguy.com>

In article <mailman.985107140.4586.python-list at python.org>, Timothy says...
>
>I am not one of the gods of Python,

me neither

>but I'm not Python illiterate either.

me neither

>When I read PEP 236. I cringed.

me too

>I understand the value,

me too

>but the solution seemed far from elegant,

me too

>and in Python, most things seem elegant

me too

>(except for lambda<wink>.

me too

>When I read the below suggestion it made far more sense to me,

me too

>and seemed to be a more elegant solution to the problem.

me too

By "elegant", perhaps you mean that it reads more like what it really is.

>On Tue, Mar 20, 2001 at 11:42:34AM +0100, Martin von Loewis wrote:
>> Please have a look at the PEP below. It is intended to replace the
>> __future__ import suggested in PEP 236.

No kidding, Gang, when I first ran across posts about the __future__ directive,
I thought it was another one of Tim's beloved witicisms.  (That's a polite way
of saying I thought it was a joke. <wink>)  It took me awhile to figure out it
was serious, because it _reads_ like a joke!: "from __future__ import..."

The "directive" thing reads a little drier--but maybe that's the way a
programming language _should_ read.  Also, using "import" for something that
really isn't an import raises our instinctive alarms: programming experience
shows that one should always use a different gizmo to do a different job. 
Specifically, "directive" is less likely to confuse those who run across it and
don't understand it because they will ignore it or look it up, rather than
trying to figure it out.

if-it's-hard-to-explain-it's-a-bad-idea
   --- or --  
from-__future__-import-directive-ly y'rs,

=g2
p.s.  It's nice to run into yet another Grant here. <wink>

_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation            http://www.iowegian.com



From fredrik at pythonware.com  Thu Mar 29 08:23:57 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Thu, 29 Mar 2001 13:23:57 GMT
Subject: CGI XML-RPC (xmlrpccgi.py) demo?
References: <mailman.985869402.3326.python-list@python.org>
Message-ID: <NHGw6.4158$4N4.847384@newsc.telia.net>

Eugene Leitl wrote:
> I'm trying to wrap the coils of my puny little mind around XML-RPC,
> specifically Kalle Svensson's CGI XML-RPC server:
>
> http://mail.python.org/pipermail/python-announce-list/2001-March/000698.html
>
> (interesting enough, according to Google there seems to be also Jeff's
>
> http://starship.python.net/crew/jjkunce/python/xmlrpccgi.tgz
>
> advertised as "A XML-RPC "server" that is accessed via a web server
> through CGI calls (as opposed to a long-running XML-RPC server that
> monitors a dedicated port).", which seems 1) entirely unrelated to 2) much
> smaller than Kalle's code 3) designed for Winders).
>
> To reiterate, I'm trying to be able to trap URI's looking like
>
> http://blah.net:666/cgi/boo.cgi?hiss=aargh#boom
>
> or parameters POSTed via HTML like
>
> <FORM TARGET="result">
> <INPUT TYPE=button VALUE="Start search now!" onClick="jsCopyFormData();return (0);">
> </FORM>
> <FORM name=submitform METHOD=post TARGET="result" ACTION="">
> <INPUT TYPE=hidden NAME="smi"     VALUE="">
> <INPUT TYPE=hidden NAME="jme"     VALUE="">
> <INPUT TYPE=hidden NAME="stype"   VALUE="">
> <INPUT TYPE=hidden NAME="sstype"  VALUE="">
> <INPUT TYPE=hidden NAME="checkin" VALUE="$cwkey">
> <INPUT TYPE=hidden NAME="usr" VALUE="$user">
> </FORM>

sorry, it looks like I misread your original question -- I thought
you were looking for a way to pass arguments (i.e. do RPC calls)
to a server process via HTTP and CGI, not that you had to use
a certain syntax.

under XML-RPC, you send XML-RPC requests, not FORMs.  see
http://www.xmlrpc.com for more info.

Cheers /F




From aleaxit at yahoo.com  Sat Mar  3 17:20:57 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sat, 3 Mar 2001 23:20:57 +0100
Subject: A counter-proposal to __future__ in PEP 236
References: <j47l29lpo8.fsf@informatik.hu-berlin.de> <97ll2u02qmp@news2.newsguy.com> <3A9FBE5C.95615241@sewardweb.com>
Message-ID: <97rqq209c1@news1.newsguy.com>

"Lloyd Sommerer" <lsommerer at sewardweb.com> wrote in message
news:3A9FBE5C.95615241 at sewardweb.com...
> Speaking as someone who is very new to python (3 weeks) and has only
> written two programs, using a colon in this manner would have been
> confusing.  Right now, my understanding is that the colon exists (apart
> from slicing) only to allow statements to follow immediately after a
> looping or control structure (on the same line).

Actually, colons are used, for readability, in various situations where
it's quite rare for the suite to follow on the same line -- who ever
does that with 'class Name:', for example?


Alex





From emile at fenx.com  Sat Mar 24 12:12:46 2001
From: emile at fenx.com (Emile van Sebille)
Date: Sat, 24 Mar 2001 09:12:46 -0800
Subject: difference between script and prompt ! looking for the Problem
References: <3ABCCA03.90871133@noos.fr> <99iht4$1c7hg$1@ID-11957.news.dfncis.de> <3ABCD57F.DC395F88@noos.fr>
Message-ID: <99ikon$1dbnf$1@ID-11957.news.dfncis.de>

Well, you're printing the individual attributes.  (3.0, 4.5) is tuple
representation.  You can print the tuple by using:

print (c.r, c.i)


HTH
--

Emile van Sebille
emile at fenx.com

---------
"William Famy" <william.famy at noos.fr> wrote in message
news:3ABCD57F.DC395F88 at noos.fr...
> Thanks.
>
> but if i print in the script i have 3.0 4.5 insted of (3.0, 4.5)
>
>
>
> Emile van Sebille wrote:
> >
> > You need to print from within a script.  Interactive mode assumes it.
> >
> > ie:
> >
> > print c.r, c.i
> >
> > HTH
> > --
> >
> > Emile van Sebille
> > emile at fenx.com
>
> --
>        \\\|///
>                            \\  - -  //
>                             (  @ @  )
> +-------------------------oOOo-(_)-oOOo-------------------------+
> |  Famy William  36 avenue des barattes  74000 Annecy  France   |
> |  email:william.famy at noos.fr william.famy at mail-enseignant.com  |
> +---------------------------------Oooo--------------------------+
>                            oooO   (   )
>                           (   )    ) /
>                            \ (    (_/
>                             \_)




From dwig at advancedmp.net  Thu Mar 22 20:01:07 2001
From: dwig at advancedmp.net (Don Dwiggins)
Date: 22 Mar 2001 17:01:07 -0800
Subject: Python and emacs
In-Reply-To: "Sean 'Shaleh' Perry"'s message of "Thu, 22 Mar 2001 10:56:15 -0800 (PST)"
References: <XFMail.20010322105615.shaleh@valinux.com>
Message-ID: <u7l1hjo30.fsf@advancedmp.net>

Sean 'Shaleh' Perry writes:
> On 22-Mar-2001 Don Dwiggins wrote:
>> A semi-idle question: has anyone looked into embedding Python as an
>> extension language in Emacs, alongside Elisp?  I'm toying with the idea of
>> doing some Emacs enhancements/modifications, and given the choice, I'd much
>> rather do them in Python.  (One such enhancement would be to add
>> Pythonwin-style "intellisense" popups to python-mode.)
>> 

> An editor that spoke python would not be a bad thing.  However, trying to get
> emacs to speak python strikes me as an amazingly good way to waste whatever
> brain cells you have left.

I'm not sure we're on the same page.  If by "trying to get emacs to speak
python" you mean abandoning Elisp and rewriting the entire .el library in
python, I'd agree wholeheartedly.  

What I had in mind was the ability to "load" a .py file (.epy?) _in addition
to_ .el files.  Then you could (re)write extensions in either language.
Yes, of course it would likely be an ambitious project, which is why I
called it "semi-idle" -- I just thought someone might have been
motivated/foolish enough (delete whichever is inappropriate) to tackle it.

-- 
Don Dwiggins                    "Solvitur Ambulando"
Advanced MP Technology
dwig at advancedmp.net




From jwbnews at scandaroon.com  Sun Mar  4 14:12:26 2001
From: jwbnews at scandaroon.com (John W. Baxter)
Date: Sun, 04 Mar 2001 11:12:26 -0800
Subject: I come to praise .join, not to bury it...
References: <mailman.983485923.1136.python-list@python.org> <97p7iq$kk2$1@nntp6.u.washington.edu> <97rka002k9@news1.newsguy.com>
Message-ID: <jwbnews-665CC3.11122604032001@corp.supernews.com>

In article <97rka002k9 at news1.newsguy.com>,
 "Alex Martelli" <aleaxit at yahoo.com> wrote:

> [Excellent exposition omitted.]

> So, since we want general polymorphism on the joiner
> object, but are quite content with polymorphism through
> the standard sequence interface on the sequence object,
> it is _just right_ that .join be a method on the joiner
> object (e.g., a string) and that it take the sequence of
> string to be joined as its argument.

OK, thanks.  I'm convinced.  For the first time.  (Which won't 
necessarily keep me from writing non-polymorphic things with the string 
module, but it might make me *try* to remember.)

> 
> I have not heard ANY _technical_ arguments opposed to
> this last time the discussion went around -- nothing but
> vague aesthetic, "it should be the other way", "I find it
> ugly" kind of complaints.  Unless and until further notice,
> then, I class these together with the complaints of people
> who dislike 0-based indexing, on similar vague bases -- as
> 0-based indexing _works_ better than 1-based, then, for
> me, it is a superior choice.

Which doesn't mean I have to *like* zero-based indexing...only that I 
have to use it.  But then, *all* of the looping that was available in 
the IBM 704 via the TXI, TXL, TXH family of instructions had fence-post 
problems*, so I suppose I shouldn't be surprised.

  --John (* if you build a straight fence with 10 posts 6 feet apart, 
how long is the fence?)

-- 
John W. Baxter   Port Ludlow, WA USA  jwbnews at scandaroon.com


From aleaxit at yahoo.com  Thu Mar 22 03:05:06 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 22 Mar 2001 09:05:06 +0100
Subject: Puzzling input problem
References: <3AB9449C.A1D14738@monitor.net> <mailman.985233449.28064.python-list@python.org> <3AB97E58.7B01D485@monitor.net>
Message-ID: <99cbni0149a@news2.newsguy.com>

"Stephen R. Figgins" <fig at monitor.net> wrote in message
news:3AB97E58.7B01D485 at monitor.net...
> > Actually, raw_input isn't the culprit, exactly.  The print
> > statement's format handler doesn't know that raw_input() has
> > modified the screen contents, so it still behaves like the cursor is
> > just after the "prompt string: " on the screen, inserting a space
> > as the ref guide says it will.
>
> Hmmm... very interesting.  The first print doesn't put the space in
> right away, but waits for the next call to the print statement to add
> the space.  That is not what I would have expected.  Interesting!

Makes sense if you think of it -- if the next character to
be output is a newline, you probably don't want a trailing
space on the previous line.  To control this, the print
statement uses an attribute named 'softspace' on the file
(or file-like) object it's printing to (normally, the file
object referenced by sys.stdout, unless you're using a
'print>>bah' statement): when it's about to emit a space
that is 'soft' (that is, is present just because print
space-separates fields), rather than emitting it at once,
it sets the softspace attribute to one; when it's about to
emit to that file again, it checks the softspace attribute,
and, if 1, first emits a space, then resets softspace to 0.

This gives you a chance to retroactively 'cancel' the
soft space that would normally be emitted, by manually
resetting the softspace attribute to 0, although it's
not a frequent need -- still, it does work:

>>> print 'one',; sys.stdout.softspace=0; print 'two'
onetwo
>>>


Alex





From ben.hutchings at roundpoint.com  Tue Mar 20 15:21:37 2001
From: ben.hutchings at roundpoint.com (Ben Hutchings)
Date: 20 Mar 2001 12:21:37 -0800
Subject: How to increase PythonWin v2.0: COM browser limits?
References: <994psh$so5$1@news1.xs4all.nl>
Message-ID: <uy9u06vji.fsf@roundpoint.com>

"Twan van der Schoot" <twanvds at xs4all.nl> writes:
<snip>
> To reproduce the problem: On a heavily used NT 4.0 system with a lot of
> registered COM stuff, start Pythonwin version 2.0 go to the menu entry: COM
> Browser to start the COM browser and expand the "Registered Categories"
> branch in the tree view.

This happens on my computer running Windows 2000 as well.

<snip>
> Pythonwin shows in the interactive window a backtrace on an exception.  The
> culprit raising this exception was the response:
> 
> File "D:\Python20\win32com\client\util.py", line 53, in __GetIndex
>     result = self._oleobj_.Next(1)
> pywintypes.com_error: (-2147024882, 'Not enough storage is available to
> complete this operation.', None, None)
> win32ui: Exception in OnNotify() handler
> 
> Something suggests that the number of registered objects is too large for
> the COM browser.

No, that's almost certainly *not* it.  For a start 'Registered
Categories' are entirely different from registered classes, and there
are likely to be very few of them on your computer.

> As I've -of course- the sources available, I was wonder if
> there is some parameter which I can increase to solve the problem.  Or am I
> running into limitations of the treeview control of MS Windows?

No, not that either.  'Out of memory' error messages are very often
the result of a programmer incorrectly assuming that some operation
can only fail due to lack of memory.

In fact, by adding a print statement to the loop that's iterating over
the categories to add them to the tree, I discovered that it yields
more or less the same results as OLEView (a similar program available
from Microsoft).  The exception is only raised after these results
have been returned.

The wrapper for COM enumerators, in true Pythonic style, does not
check the size of the collection in advance, but attempts to read
items until this yields an error.  I think that the category
enumerator implementation is reporting the no-more-items error in the
wrong way (returning E_OUTOFMEMORY rather than S_FALSE), so the
wrapper does not translate it into an IndexError exception which
would cause the loop to terminate successfully.  OLEView probably
just doesn't check for the error but displays what it already got.

-- 
Any opinions expressed are my own and not necessarily those of Roundpoint.


From jeremy at alum.mit.edu  Fri Mar 30 14:39:56 2001
From: jeremy at alum.mit.edu (Jeremy Hylton)
Date: Fri, 30 Mar 2001 14:39:56 -0500 (EST)
Subject: Should I close after popen??
In-Reply-To: <oqelvfm25h.fsf@lin2.sram.qc.ca>
References: <m38zm6r8ih.fsf@master.athome>
	<fhOs6.510$Fb.777424@typhoon2.ba-dsg.net>
	<mailman.985919149.7252.python-list@python.org>
	<uofujnng0.fsf@ctwd0143.fitlinxx.com>
	<oqelvfm25h.fsf@lin2.sram.qc.ca>
Message-ID: <15044.57612.465798.404399@w221.z064000254.bwi-md.dsl.cnc.net>

>>>>> "FP" == =?iso-8859-1?q?Fran=E7ois?= Pinard <pinard at iro.umontreal.ca> writes:

  FP> [David Bolen]
  >> pinard at iro.umontreal.ca (=?iso-8859-1?q?Fran=E7ois?= Pinard)
  >> writes:

  >> > I would completely agree with the above for someone who wants
  >> > to write Jython, or write Python that could be quickly turned
  >> > into Jython.  But if you stick with Python, you can safely rely
  >> > on the fact that a file is automagically closed as soon as you
  >> > do not refer to it anymore.

  >> I'd probably change Python in that sentence to CPython, since
  >> it's an implementation issue and not a language one (the language
  >> makes no such guarantees).

  FP> As it was pointed on this list quite a while ago, Python has no
  FP> language definition per itself.  It is rather defined by the
  FP> documentation of its main implementation. 

Yes and no.  I believe the language reference is intended to define
the language, not the implementation.  It may not be perfect in this
regard, but the language reference is not just "the documentation of
its main implementation."
 
  FP>                                            It is true that the
  FP> documentation of (C)Python does not commit it on the particular
  FP> issue above, and this is why I checked with this list before
  FP> starting to firmly rely on this feature, in real code.

I believe there is an FAQ entry about finalization.  The official
position is that immediate finalization via reference counting is not
part of the language definition.  It's just a property of a particular
implementation.  I'd recommend calling close explicitly; I believe
that is what the FAQ says.

However, I don't expect CPython to change any time soon.  (But who
knows what changes are lurking unforseen :-).

Jeremy



From twanvds at xs4all.nl  Mon Mar 19 21:32:24 2001
From: twanvds at xs4all.nl (Twan van der Schoot)
Date: Tue, 20 Mar 2001 03:32:24 +0100
Subject: How to increase PythonWin v2.0: COM browser limits?
References: <994psh$so5$1@news1.xs4all.nl> <995sgi$7u0$04$1@news.t-online.com>
Message-ID: <996flh$e9m$1@news1.xs4all.nl>

thanks,


But I think it is strange that nobody else seems to have the problem, wonder
if anybody uses the COM facilities in Pythonwin.

gr

Twan


"Eduard Hiti" <KatMouse at gmx.de> wrote in message
news:995sgi$7u0$04$1 at news.t-online.com...
> I just tried the COM browser here on my machine, and it gives me the same
> error. I'm suspecting an error in Pythonwin...
>
> Eduard
>
> "Twan van der Schoot" <twanvds at xs4all.nl> wrote in message
> news:994psh$so5$1 at news1.xs4all.nl...
> | I'm trying to use the COM Browser in Pythonwin version 2.0.  But I run
> into
> | limitations of some kind.  As a Python newby I dutifully scanned FAQs
and
> | newsgroups, but alas!
> |
> | To reproduce the problem: On a heavily used NT 4.0 system with a lot of
> | registered COM stuff, start Pythonwin version 2.0 go to the menu entry:
> COM
> | Browser to start the COM browser and expand the "Registered Categories"
> | branch in the tree view.
> |
> | Pythonwin shows in the interactive window a backtrace on an exception.
> The
> | culprit raising this exception was the response:
> |
> | File "D:\Python20\win32com\client\util.py", line 53, in __GetIndex
> |     result = self._oleobj_.Next(1)
> | pywintypes.com_error: (-2147024882, 'Not enough storage is available to
> | complete this operation.', None, None)
> | win32ui: Exception in OnNotify() handler
> |
> | Something suggests that the number of registered objects is too large
for
> | the COM browser.  As I've -of course- the sources available, I was
wonder
> if
> | there is some parameter which I can increase to solve the problem.  Or
am
> I
> | running into limitations of the treeview control of MS Windows?
> |
> | TIA
> |
> | Twan van der Schoot
> | twanvds at NOSPAM.xs4all.nl
> |
> |
>
>




From gkiffney at my-deja.com  Wed Mar 21 10:18:52 2001
From: gkiffney at my-deja.com (gkiffney)
Date: Wed, 21 Mar 2001 10:18:52 -0500
Subject: Accessing Oracle from Python remotely ?
References: <Xns906AB9C901807PaCmAnRDLM@161.48.76.20>
Message-ID: <eE3u6.776$9%.203510@newsr1.maine.rr.com>

There's http://www.zope.org/Products/DCOracle which should get you started.
There was an article by Uche Ogbuji in Linuxworld that helped me get going
with this, but it's apparently fallen off the Internet map into the bit
bucket.  Using the dcoracle module (which is a bit of a pain to get going,
look at the Zope docs carefully, and if you can find the prebuilt dll's, use
those), you can do things like this:

[uche at malatesta uche]$ python
Python 1.5.1 (#1, Sep 3 1998, 22:51:17) [GCC 2.7.2.3] on linux-i386
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import DCOracle
>>> #connect to DB. scott/tiger are the test database username/password
>>> dbc=DCOracle.Connect("scott/tiger at sforza") #sforza is the local Oracle
server
>>> c = dbc.cursor()
>>> c.execute("SELECT * FROM emp WHERE emp.deptno = 10")
>>> rows = c.fetchall()
>>> print rows
[(7782, 'CLARK', 'MANAGER', 7839, dbiDate('1981-06-09 00:00:00'), 2450.0,
None, 10),
(7839, 'KING', 'PRESIDENT', None, dbiDate('1981-11-17 00:00:00'), 5000.0,
None, 10),
(7934, 'MILLER', 'CLERK', 7782, dbiDate('1982-01-23 00:00:00'), 1300.0,
None, 10)]



"Fred Pacquier" <fredp at mygale.org.nospam> wrote in message
news:Xns906AB9C901807PaCmAnRDLM at 161.48.76.20...
> Hi,
>
> I have an Oracle (7) database running on a Windows NT 4 server. Is it
> possible (easy, difficult ?) to read the data in those tables from a
python
> script running on a remote Unix (Solaris) box ?
>
> Sorry if the question is silly -- I do know a bit of python but I'm
> inexperienced in databases and their connectivity. Any pointers welcome
> though.
>
> Thanks,
> fp
>
> --
> YAFAP : http://www.multimania.com/fredp/




From lluang at northwestern.edu  Mon Mar 12 09:49:48 2001
From: lluang at northwestern.edu (Louis Luangkesorn)
Date: Mon, 12 Mar 2001 08:49:48 -0600
Subject: Python In A Nutshell - Numpy essential
References: <3AAE7E0D@MailAndNews.com> <slrn9aolph.ccf.jimhill@meldrick.swcp.com>
Message-ID: <3AACE20C.AC8D2EA2@northwestern.edu>

I have to second that Numeric Python belongs in the ESSENTIAL column.  For me,
the fact that Python can replace Matlab was the reason I started using Python.
And when my university discontinued its campus-wide license, and every
installation of Matlab must be paid for, that it is an open-source replacement
for Matlab makes it uniquely valuable (Octave was too slow).

Louis

Jim Hill wrote:

> >>===== Original Message From Hamish Lawson <hamish_lawson at yahoo.co.uk> =====
> >>
> >>For the third-party modules I'd nominate the following, grouped by
> >>desirability of inclusion:
> >>
> >>    ESSENTIAL
> >>
> >>    NICE
> >>       Numerical Python
>
> I'd put this one in the ESSENTIAL column.
>
> Jim
> --

--
K Louis Luangkesorn
lluang at northwestern.edu  http://pubweb.nwu.edu/~kll560  PGP:0xF3E2D362
Whatsoever things are true, ... honest, ... just, ... pure, ... lovely, ... of
good report; if there be any virtue, and if there be any praise, think on these
things.- motto - Northwestern University




From smnordby at yahoo.com  Mon Mar 12 07:04:17 2001
From: smnordby at yahoo.com (Steve)
Date: Mon, 12 Mar 2001 04:04:17 -0800
Subject: Einstein's Riddle
References: <mailman.984369618.12493.python-list@python.org>
Message-ID: <3AACBB41.5854666A@yahoo.com>

Making assumptions based on vague statements is not a feature of
"advanced" IQ tests.  Only one IQ test, the Stanford-Binet form L-M, has
been shown to reliably discern among IQ's higher than about 150, and if
this question were on that test, the subject would probably be awarded
extra points if he or she pointed out that this puzzle is unsolvable
without making an assumption. 

-SteveN-
Glossary of Gifted Education:
http://members.aol.com/svennord/ed/GiftedGlossary.htm

Tim Peters wrote:
> 
> [Gregory Jorgensen]
> > The 2% that can solve it must be psychic or deranged. The riddle
> > as posed is not solvable.
> > ...
> > But only four types of pet are mentioned: dogs, birds, cats, horse. We
> > have no evidence that any of the five owns fish.
> >
> > More telling: how many people can't quickly determine that this puzzle is
> > unsolvable?
> 
> It's a std feature of "advanced" IQ tests that the problem stmts are
> intentionally somewhat vague:  part of what they're testing is whether you
> can come up with the most reasonable assumptions necessary to make the
> problem *interesting*.  Since the problem statement here asked "Who owns the
> fish?", and the problem admits of a unique solution if you assume *someone*
> owns a fish, that's a more interesting assumption to make than to complain
> that the question is ill-defined; the ability to fill in gaps reasonably is
> as much a kind of intelligence as the ability to follow chains of reasoning.
> 
> You could cover both cheeks by answering "if anyone owns a fish, it must be
> the German".  Then again, you may prefer to complain that "on the left of"
> doesn't necessarily imply adjacent to; or that the problem statement never
> says that a person who "lives in" a house is also its "owner"; or etc etc.
> 
> reading-a-question-as-trivial-is-best-reserved-for-saints-ly y'rs  - tim


From paulp at ActiveState.com  Fri Mar 23 20:40:50 2001
From: paulp at ActiveState.com (Paul Prescod)
Date: Fri, 23 Mar 2001 17:40:50 -0800
Subject: newbe how to create 2 __init__ in a class ?
References: <3ABBEB7D.6E75E5B8@noos.fr>
Message-ID: <3ABBFB21.761A50F5@ActiveState.com>

(year,month,day,price1,price2,price3) 
or*                                     
(string) "2000/12/28,12.3,56.3,45.3"

> Is it possible te create a class with 2 __init__
> 
> exemple i whant to create a classe with
> (year,month,day,price1,price2,price3)
> or*                                     (string) "2000/12/28,12.3,56.3,45.3"
> 
> in C++ I tell the type but how to do it in python.

StringTypes = types.UnicodeType, types.StringType

class Foo:
    def __init__(self, string_or_year, 
                   month=None, day=None, price1=None, 
                   price2=None, price3=None):
        if month:
		assert day and price1 and price2 and ...
		year = string_or_year
	else:
		assert string_or_year in StringTypes
		year, month, day, price1, price2, price3 = \
			mysplit(sring_or_year)

-- 
Take a recipe. Leave a recipe.  
Python Cookbook!  http://www.activestate.com/pythoncookbook



From jgbrawley at earthlink.net  Fri Mar 16 17:21:18 2001
From: jgbrawley at earthlink.net (John Brawley)
Date: Fri, 16 Mar 2001 22:21:18 GMT
Subject: Type mismatchings and the rank newbie
Message-ID: <ylws6.2639$227.274111@newsread2.prod.itd.earthlink.net>

Rank newbie here, trying to do something specific.
I need to use a set of numbers (say, 1 to 25), and create 25 lists from the
numbers (1 thru 25), which lists have a character "P" and a number (one of
the 25).

The lists I want look like :
P1=[x, y, z]
P2=[x, y, z]
P3=[x, y, z]
(etcetera down to P25)

The user inputs the number that creates the lists, so I can't just write 25
lists (the number could be anywhere from 13 to 500 or more).

I need to write a loop that takes each number in sequence and sticks it next
to the ascii "P" so that when the loop is finished, I end up with a set of
lists each of whose names is different by one number.

The loop is not a problem; I just can't figure how to stick a different
number form the user input total, into the second place of the list name
(  Pn=[x, y, z] where 'n' is the number stuck in there by the loop) with
every iteration of the loop.

The docs are confusing to me (I'm not a programmer and don't want to be, but
there is no other way for me to do what I need to to see what this program
will display for me).

Can anyone suggest a way out of this hole?

Thanks!

--
JB
jgbrawley at earthlink.net
Web: http://www.jcn1.com/jbrawley
          http://home.earthlink.net/~jgbrawley




From msoulier at nortelnetworks.com  Tue Mar 20 14:38:54 2001
From: msoulier at nortelnetworks.com (Michael P. Soulier)
Date: 20 Mar 2001 19:38:54 GMT
Subject: re.match
References: <997skf$qoc$1@bmerhc5e.ca.nortel.com> <TrMt6.2490$4N4.318602@newsc.telia.net> <mailman.985114106.26896.python-list@python.org>
Message-ID: <998bke$88i$1@bmerhc5e.ca.nortel.com>

On Tue, 20 Mar 2001 10:53:16 -0800, Ken Seehof <kens at sightreader.com> wrote:
>re.match tests whether or not a string -matches- a regular expression.
>re.search -searches- a string for the first occurrence of a regular
>expression.

    And hence the non-intuitive aspect of it. If you include an anchor in your
regexp, ie. ^ or $, then the definition of a match is such that the pattern
matches at the beginning and/or end of the line respectively. If you do not
include anchors, then by the definition of the regexp syntax, the entire line
should be searched. 
    Now, it appears that re.match has special capability with respect to
multi-line strings, but again, this is non-standard regexp syntax, and hence
non-intuitive. 

    No biggy, but surprising compared to every other implementation of regexps
that I've seen. 

    Cheers,

    Mike

-- 
Michael P. Soulier, TD12, SKY  Tel: 613-765-4699 (ESN: 39-54699)
Optical Networks, Nortel Networks, SDE Pegasus
"With sufficient thrust, pigs fly just fine. However, this is not necessarily a
good idea. It is hard to be sure where they are going to land, and it could be
dangerous sitting under them as they fly overhead." -- RFC 1925
Nortel Linux User's Group Ottawa: (internal) http://nlug.ca.nortel.com


From jafo at tummy.com  Sat Mar 24 18:17:35 2001
From: jafo at tummy.com (Sean Reifschneider)
Date: Sat, 24 Mar 2001 16:17:35 -0700
Subject: RFC: PEP243: Module Repository Upload Mechanism
Message-ID: <20010324161735.A19818@tummy.com>

Included below is the version of PEP243 after it's initial round of review.
I welcome any feedback.

Thanks,
Sean

============================================================================
PEP: 243
Title: Module Repository Upload Mechanism
Version: $Revision$
Author: jafo-pep at tummy.com (Sean Reifschneider)
Status: Draft
Type: Standards Track
Created: 18-Mar-2001
Python-Version: 2.1
Post-History: 
Discussions-To: distutils-sig at python.org


Abstract

    For a module repository system (such as Perl's CPAN) to be
    successful, it must be as easy as possible for module authors to
    submit their work.  An obvious place for this submit to happen is
    in the Distutils tools after the distribution archive has been
    successfully created.  For example, after a module author has
    tested their software (verifying the results of "setup.py sdist"),
    they might type "setup.py sdist --submit".  This would flag
    Distutils to submit the source distribution to the archive server
    for inclusion and distribution to the mirrors.

    This PEP only deals with the mechanism for submitting the software
    distributions to the archive, and does not deal with the actual
    archive/catalog server.


Upload Process

    The upload will include the Distutils "PKG-INFO" meta-data
    information (as specified in PEP-241 [1]), the actual software
    distribution, and other optional information.  This information
    will be uploaded as a multi-part form encoded the same as a
    regular HTML file upload request.  This form is posted using
    ENCTYPE="multipart/form-data" encoding [RFC1867].

    The upload will be made to the host "modules.python.org" on port
    80/tcp (POST http://modules.python.org:80/swalowpost.cgi).  The form
    will consist of the following fields:

        distribution -- The file containing the module software (for
        example, a .tar.gz or .zip file).

        distmd5sum -- The MD5 hash of the uploaded distribution,
        encoded in ASCII representing the hexadecimal representation
        of the digest ("for byte in digest: s = s + ('%02x' %
        ord(byte))").

        pkginfo (optional) -- The file containing the distribution
        meta-data (as specified in PEP-241 [1]).  Note that if this is not
        included, the distribution file is expected to be in .tar format
        (gzipped and bzipped compreesed are allowed) or .zip format, with a
        "PKG-INFO" file in the top-level directory it extracts
        ("package-1.00/PKG-INFO").

        infomd5sum (required if pkginfo field is present) -- The MD5 hash
        of the uploaded meta-data, encoded in ASCII representing the
        hexadecimal representation of the digest ("for byte in digest:
        s = s + ('%02x' % ord(byte))").

        platform (optional) -- A string representing the target
        platform for this distribution.  This is only for binary
        distributions.  It is encoded as
        "<os_name>-<os_version>-<platform architecture>-<python
        version>".

        signature (optional) -- A OpenPGP-compatible signature [RFC2440]
        of the uploaded distribution as signed by the author.  This may be
        used by the cataloging system to automate acceptance of uploads.

        protocol_version -- A string indicating the protocol version that
        the client supports.  This document describes protocol version "1".


Return Data

    The status of the upload will be reported using HTTP non-standard
    ("X-*)" headers.  The "X-Swalow-Status" header may have the following
    values:

        SUCCESS -- Indicates that the upload has succeeded.

        FAILURE -- The upload is, for some reason, unable to be
        processed.

        TRYAGAIN -- The server is unable to accept the upload at this
        time, but the client should try again at a later time.
        Potential causes of this are resource shortages on the server,
        administrative down-time, etc...

    Optionally, there may be a "X-Swalow-Reason" header which includes a
    human-readable string which provides more detailed information about
    the "X-Swalow-Status".

    If there is no "X-Swalow-Status" header, or it does not contain one of
    the three strings above, it should be treated as a temporary failure.

    Example:

        >>> f = urllib.urlopen('http://modules.python.org:80/swalowpost.cgi')
        >>> s = f.headers['x-swalow-status']
        >>> s = s + ': ' + f.headers.get('x-swalow-reason', '<None>')
        >>> print s
        FAILURE: Required field "distribution" missing.


Sample Form

    The upload client must submit the page in the same form as
    Netscape Navigator version 4.76 for Linux produces when presented
    with the following form:

        <H1>Upload file</H1>
        <FORM NAME="fileupload" METHOD="POST" ACTION="swalowpost.cgi"
              ENCTYPE="multipart/form-data">
        <INPUT TYPE="file" NAME="distribution"><BR>
        <INPUT TYPE="text" NAME="distmd5sum"><BR>
        <INPUT TYPE="file" NAME="pkginfo"><BR>
        <INPUT TYPE="text" NAME="infomd5sum"><BR>
        <INPUT TYPE="text" NAME="platform"><BR>
        <INPUT TYPE="text" NAME="signature"><BR>
        <INPUT TYPE="hidden" NAME="protocol_version" VALUE="1"><BR>
        <INPUT TYPE="SUBMIT" VALUE="Upload">
        </FORM>


Platforms

    The following are valid os names:

        aix beos debian dos freebsd hpux mac macos mandrake netbsd
        openbsd qnx redhat solaris suse windows yellowdog

    The above include a number of different types of distributions of
    Linux.  Because of versioning issues these must be split out, and
    it is expected that when it makes sense for one system to use
    distributions made on other similar systems, the download client
    will make the distinction.

    Version is the official version string specified by the vendor for
    the particular release.  For example, "2000" and "nt" (Windows),
    "9.04" (HP-UX), "7.0" (RedHat, Mandrake).

    The following are valid architectures:

        alpha hppa ix86 powerpc sparc ultrasparc


Status

    I currently have a proof-of-concept client and server implemented.
    I plan to have the Distutils patches ready for the 2.1 release.
    Combined with Andrew's PEP-241 [1] for specifying distribution
    meta-data, I hope to have a platform which will allow us to gather
    real-world data for finalizing the catalog system for the 2.2
    release.


References

    [1] Metadata for Python Software Package, Kuchling,
        http://python.sourceforge.net/peps/pep-0241.html

    [RFC1867] Form-based File Upload in HTML
        http://www.faqs.org/rfcs/rfc1867.html

    [RFC2440] OpenPGP Message Format
        http://www.faqs.org/rfcs/rfc2440.html


Copyright

    This document has been placed in the public domain.



Local Variables:
mode: indented-text
indent-tabs-mode: nil
End:
-- 
 A smart terminal is not a smart*ass* terminal, but rather a terminal
 you can educate.  -- Rob Pike
Sean Reifschneider, Inimitably Superfluous <jafo at tummy.com>
tummy.com - Linux Consulting since 1995. Qmail, KRUD, Firewalls, Python



From roy at panix.com  Sat Mar 31 23:25:37 2001
From: roy at panix.com (Roy Smith)
Date: Sat, 31 Mar 2001 23:25:37 -0500
Subject: Conditional operator in Python?
References: <mailman.986094301.20973.python-list@python.org>
Message-ID: <roy-8A40AC.23253731032001@news.panix.com>

"Tim Peters" <tim.one at home.com> wrote:
> Eric, there's no feature from *any* language that isn't regularly suggested
> for Python.

I have never once seen somebody suggest that we have either FORTRAN-style 
computed gotos or counted hollerith constants :-)


From chrisa at ASPATECH.COM.BR  Thu Mar  1 16:06:13 2001
From: chrisa at ASPATECH.COM.BR (Chris Richard Adams)
Date: Thu, 1 Mar 2001 18:06:13 -0300
Subject: Has anyone been able to convince Bruce Eckel to write "Thinking in Python"
Message-ID: <BD674CDC0A68C642B60038986FBB56DF015C47@mailsrv.ASPATECH.COM.BR>

If anyone has worked with Java - they would know how Bruce's bible
'Thinking in Java' has helped thousands of developers.  Could we be as
fortunate with him considering a Python book?? Can only pray and beg.  
 
 



From msoulier at nortelnetworks.com  Tue Mar 20 13:22:05 2001
From: msoulier at nortelnetworks.com (Michael P. Soulier)
Date: 20 Mar 2001 18:22:05 GMT
Subject: re.match
References: <997skf$qoc$1@bmerhc5e.ca.nortel.com> <TrMt6.2490$4N4.318602@newsc.telia.net>
Message-ID: <99874d$4tp$1@bmerhc5e.ca.nortel.com>

On Tue, 20 Mar 2001 17:29:23 GMT, Fredrik Lundh <fredrik at pythonware.com>
wrote:
>Michael P. Soulier wrote:
>>     Hey people. Could someone explain the purpose of the match function in
>>     the re module? If it's only difference from search is to match at the
>>     beginning of the line, why does it exist? That's what the ^ in the
>>     regexp is for.
>
>no, it's not.  read the documentation again.

    That answer was so warm and informative I'd think this was
comp.lang.perl.misc. 

---quote---
Note that match may differ from search using a regular expression beginning
with "^": "^" matches only at the start of the string, or in MULTILINE mode
also immediately following a newline. The ``match'' operation succeeds only if
the pattern matches at the start of the string regardless of mode, or at the
starting position given by the optional pos argument regardless of whether a
newline precedes it. 
---end quote---

    The difference is apparently multiline mode. My point was that it doesn't
seem to offer enough of a difference to justify going against standard regexp
syntax and ignore the ^ character. It is misleading, IMHO, and Python seems to
be all about being easy to learn. Thus being misleading is not beneficial. 

    Again, thanks for the extremely helpful response. If someone would
actually care to have a discussion with me on this so that I might understand
the reasons why this was done, please do so. As always, one asks questions not
to simply annoy, but to learn. There was a time when Usenet was a good place
to do that. 

    Cheers,

    Mike

-- 
Michael P. Soulier, TD12, SKY  Tel: 613-765-4699 (ESN: 39-54699)
Optical Networks, Nortel Networks, SDE Pegasus
"With sufficient thrust, pigs fly just fine. However, this is not necessarily a
good idea. It is hard to be sure where they are going to land, and it could be
dangerous sitting under them as they fly overhead." -- RFC 1925
Nortel Linux User's Group Ottawa: (internal) http://nlug.ca.nortel.com


From kcocaine at GamesandTheory.com  Sun Mar  4 19:54:08 2001
From: kcocaine at GamesandTheory.com (Kevin Douglas)
Date: Sun, 4 Mar 2001 19:54:08 -0500 (EST)
Subject: Need a free lancer python programmer - to do bots
In-Reply-To: <Pine.LNX.4.21.0103041948210.19952-100000@critical.GamesandTheory.com>
Message-ID: <Pine.LNX.4.21.0103041953481.19952-100000@critical.GamesandTheory.com>

crap!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

shoot me..


-kev






From renzo.tomaselli at tecnotp.it  Fri Mar  9 11:53:58 2001
From: renzo.tomaselli at tecnotp.it (Renzo Tomaselli)
Date: Fri, 9 Mar 2001 17:53:58 +0100
Subject: embedding python: where to put imported modules
Message-ID: <98b1qi$265$1@news.flashnet.it>

Hi all,
    as many others I want to embed Python in my components which take a
"physical" form of a dll/shlib. Since I want to embed a restricted version
which preimports a well defined module set, an interesting issue arises:
where to put such modules so that no further external files are needed and
such components will be fully self-consistent.
A possible solution would be to zip them as a dll resource for the win32
version, or a static data block for a shlib. Then modules would be
unzipped/loaded at first usage or at startup.
Did anybody manage a similar issue for embedding Python ? Any comment ?
Thanks,
                                             Renzo Tomaselli
 -------------------------------------------------------------------------
TecnoTP s.n.c. Special Information System Design
Maso Pelauchi I38050 Ronchi Valsugana,  Trento TN  ITALY
Tel. +39 0461 773164      Fax. +39 0461 771514
e-mail: renzo.tomaselli at tecnotp.it
-------------------------------------------------------------------------





From loewis at informatik.hu-berlin.de  Sun Mar 11 15:12:35 2001
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 11 Mar 2001 21:12:35 +0100
Subject: supporting asyncronous messages
References: <mailman.983793067.17928.python-list@python.org> <j466hloqof.fsf@informatik.hu-berlin.de> <mailman.984312437.15426.python-list@python.org>
Message-ID: <j4k85wm5e4.fsf@informatik.hu-berlin.de>

Joshua Spoerri <josh at pobox.com> writes:

> can python overload the logical operators, if not, are there any
> plans to enable it to? 

No. Even if you could, you would not get your three-state logic.

> do you see any gotchas in this approach?

Many. If you want a special meaning, why not write methods? Instead of

   a and b

you could write

   a.my_special_and(b)

If you want that to operate on many arguments, how about

   my_special_and(a,b,c,d)

Provided you have the right syntax, how are you going to implement it?
It is easy to block, it is not that easy to unblock.

Regards,
Martin



From aleaxit at yahoo.com  Mon Mar  5 10:09:37 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Mon, 5 Mar 2001 16:09:37 +0100
Subject: Name of the Script
References: <20010304.185354.2114937732.8277@localhost.localdomain> <NJCo6.6720$qv3.2834377@nnrp5-w.sbc.net> <FpmjOjvwe5bm0Bg09LDTMPhh5vji@4ax.com>
Message-ID: <980a7j1rg0@news2.newsguy.com>

"Paul Moore" <paul.moore at uk.origin-it.com> wrote in message
news:FpmjOjvwe5bm0Bg09LDTMPhh5vji at 4ax.com...
    [snip]
> --- a.py ---
> import sys
> print sys.argv[0]
> ---
>
> python a.py
>
> just prints "a.py". Is there a way of getting the *full* script name.

import sys, os.path

print os.path.abspath(sys.argv[0])


Alex





From robin.thomas at starmedia.net  Sun Mar 25 12:58:53 2001
From: robin.thomas at starmedia.net (Robin Thomas)
Date: Sun, 25 Mar 2001 12:58:53 -0500
Subject: Comparing efficiency in sending mail
In-Reply-To: <dm8sbt84f8jrvcu085a8pt8hvhdfvvlv9k@4ax.com>
References: <3A7FA5BA.82FB7593@esec.com.au>
 <mailman.981444758.4937.python-list@python.org>
Message-ID: <4.3.1.2.20010325124756.01bc4cb0@exchange.starmedia.net>

At 05:08 PM 3/25/01 +0000, Sheila King wrote:

>Anyhow, that last part is really off-topic here, as it is an issue with how to
>use sendmail/qmail. I just wanted to point out, that using the smtplib module
>is easier and gives me more control over the messages I send, and therefore
>I'm curious as to why I would want to use popen to send to sendmail instead.

If you use smtplib, your program is the MTA. Your program must do the 
actual message transfer by itself: checking/munging the message to be sent, 
blocking on connect to destination MTA, handling errors, retrying if 
desired, bouncing message to a mailbox, etc.

If you use sendmail, sendmail is the MTA. Your program, if it gives 
sendmail a minimally complete message, will complete the popen() function 
quickly, and sendmail will then worry about the actual sending of the 
message. Your program can do other things. As a "full-service" MTA, 
sendmail can retry sending, bounce the message to a mailbox, queue the 
message for sending later in a big batch, etc. Of course, "full-service" 
also means that sendmail does things to the message or with the message 
that you didn't ask it to do.

The choice you make is informed by your application's needs.



--
Robin Thomas
Engineering
StarMedia Network, Inc.
robin.thomas at starmedia.net




From kbk at shore.net  Sat Mar 10 14:56:20 2001
From: kbk at shore.net (Kurt B. Kaiser)
Date: Sat, 10 Mar 2001 19:56:20 GMT
Subject: Can anyone recomend a good intoduction to C...
References: <mailman.983894288.19965.python-list@python.org> <ewwp6.1466$y6.241005@ruti.visi.com> <slrn9aeblb.cg1.scarblac@pino.selwerd.nl> <djagat4kk3cs9hg0c3k39cudk6goblhkg6@4ax.com> <989mbk$24nuv$1@fido.engr.sgi.com> <slrn9ah66m.k09.qrczak@qrnik.zagroda> <98auhn0270j@news2.newsguy.com> <slrn9aiaq7.rmu.qrczak@qrnik.zagroda> <98dpoe12ckn@news1.newsguy.com>
Message-ID: <3AAA85C1.81FB90AA@shore.net>

Alex Martelli wrote:

> Your opinion is widespread, and self-perpetuating, but it has no
> current technical basis for such tasks as Python extensions on Linux.
> 
My understanding is that the C++ compiler for Linux is still
considered to be unstable and that that is a significant technical
factor for sticking with C?


From jkraska1 at san.rr.com  Sun Mar 18 01:47:31 2001
From: jkraska1 at san.rr.com (Courageous)
Date: Sun, 18 Mar 2001 06:47:31 GMT
Subject: Pep 238 Why new // operator (Non-integer Division)
References: <98tc2s$qir$1@news2.ottawa.cyberus.ca> <mailman.984822021.13503.python-list@python.org> <vag7bt09r54tv05nohtssd0ge891v3fa02@4ax.com> <3AB42890.AD9E9FF0@wwc.com>
Message-ID: <lgm8btgpa5ntkaoehqngtj059ctgri191e@4ax.com>

On Sun, 18 Mar 2001 03:16:03 GMT, Steve Williams <stevewilliams at wwc.com> wrote:
>Courageous wrote:
>> This isn't true; the moment you encounter code written by someone else,
>> you DO have to use them. This simple principle is the magic of Python
>> minimalism, and the primary driving reason why new Python programmers
>> can become so effective so fast (IMO).

>           For simplicity's sake I do all my computations in the base-pi number
>system. In this system pi=10.

While I perceive a shadowy point here, the fact of the matter is that the more
features you add to a language, the more a new programmer must know in
order to read code produced by more experienced programmers. I'm a big
fan of Python's minimalism for this case, and believe it's one of its strong
selling points. The relative paucity of forms in Python is a strength, not a
weakness. If Python departs from this aspect of its origin, it will no longer
really be Python.

This isn't to say that I'm against the growth of the language. I merely would
like to point out that all additions are at odds with Python's principal of
minimalism. If some addition can already be achieved through a simple
means in Python, perhaps it shouldn't be added.

Python is not Perl.

C//



From adaminthedomaindevtty.net  Thu Mar 22 01:05:36 2001
From: adaminthedomaindevtty.net (Adam Logghe)
Date: Wed, 21 Mar 2001 22:05:36 -0800
Subject: XLink/XPointer processing in Python
References: <mailman.985163671.1382.python-list@python.org>
Message-ID: <3ab8eddb_1@news.nwlink.com>

Try www.fourthought.com has produced 4suite (python and c).

They have produced about as full featured XML suite as I've seen on any
platform. XLINK is implemented though I haven't used it yet.

It's the good stuff.

Adam
adam in the domain devtty.net

"Lars von Wedel" <vonWedel at lfpt.rwth-aachen.de> wrote in message
news:mailman.985163671.1382.python-list at python.org...
> Hi,
>
> I want to build an XML processing app which shall be
> able to process "simple" links according to the XLink
> spec (using SAX).
>
> I haven't found any information on how these are handled
> in the processing application, so I assume, I must handle
> the xlink:href attribute content on 'my' own? Are there
> any reusable pieces of software to do so?
>
> Lars




From syver at NOSPAMcyberwatcher.com  Fri Mar 30 16:24:12 2001
From: syver at NOSPAMcyberwatcher.com (Syver Enstad)
Date: Fri, 30 Mar 2001 23:24:12 +0200
Subject: Use site-packages on all platforms
References: <mailman.985955621.21096.python-list@python.org> <EB3x6.114662$Ch.15365680@typhoon.we.rr.com>
Message-ID: <9a2tlv$o8k$1@troll.powertech.no>

"Pete Shinners" <shredwheat at mediaone.net> wrote in message
> i heartily agree with this proposal. let's keep user installed
> packages/modules/scripts out of the executable directory. it's
> unclean.
I agree too :-)

I recently installed 4dom and it threw directories all over the place, very
messy stuff.





From cfelling at iae.nl  Sat Mar 17 16:03:40 2001
From: cfelling at iae.nl (Carel Fellinger)
Date: 17 Mar 2001 22:03:40 +0100
Subject: I come to praise .join, not to bury it...
References: <mailman.983485923.1136.python-list@python.org> <97p7iq$kk2$1@nntp6.u.washington.edu> <97rka002k9@news1.newsguy.com> <3AAF7E47.294617E1@sage.att.com> <98qeco02hn7@news1.newsguy.com> <98rmod$c8p$1@animus.fel.iae.nl> <98tacr0q5s@news1.newsguy.com> <98u86d$6p9$1@animus.fel.iae.nl> <98v88s01ri1@news2.newsguy.com>
Message-ID: <990jfc$51m$1@animus.fel.iae.nl>

Alex Martelli <aleaxit at yahoo.com> wrote:
> "Carel Fellinger" <cfelling at iae.nl> wrote in message
> news:98u86d$6p9$1 at animus.fel.iae.nl...
...
>> Here we agree, and clearly Rococo, though having a beauty of its own,
>> has nothing to do with the aesthetic principles that guided the
>> Bauhaus people, so aesthetics in it self as the sole guiding light has
>> it pitfals, to say the least:)  But we have discussed this before,
>> aesthettics as a design guide only works for the `Enlightend Ones'.

> So you're saying Brunelleschi was not 'Enlightend' enough for you,
> for example, because he ornamented aplenty?  This keeps sounding
> to me as high-falutin' cover-up for "what _I_ like is good".

I don't know Brunelleschi, so I can't comment on his `Enlightenment',
but surely the plenty of curls in the letters of the Rococo period did
no good to the *readability*, and *that* is what is relevant it the
context of our current discussion.




-- 
groetjes, carel


From jepler at inetnebr.com  Mon Mar 26 08:58:39 2001
From: jepler at inetnebr.com (Jeff Epler)
Date: Mon, 26 Mar 2001 13:58:39 GMT
Subject: close(), exceptions and problems
References: <slrn9bm5u5.klj.kuntsi@myntti.helsinki.fi> <99i3j7$lqr$1@news.cybercity.dk> <slrn9bpiej.a63.qrczak@qrnik.zagroda> <99l8f4$1fob$1@news.cybercity.dk>
Message-ID: <slrn9buioe.3ta.jepler@potty.housenet>

On Sun, 25 Mar 2001 17:04:36 +0000 (UTC), Erwin S. Andreasen
 <erwin at andreasen.com> wrote:
>But this only happens when SA_RESTART is not set (OK, I thought it was
>SA_NORESTART to *disable* restarting but it's SA_RESTART to enable
>restarting).
>
>Read the "Interrupted Primitives" node for more information.
>
>Maybe the threading library on the platform the original poster uses does not
>set the SA_RESTART flag (if its threads are userspace, switched by receiving
>SIUGUSR1) ? But that's hard to believe.

Well, *either*
	1) The system returns -EINTR on close() and the file is not closed.
	   This should cause fclose() to return EOF, and Python to raise
	   an IOError.
	2) The system restarts the close() syscall internally in fclose().
	   When the fclose() call returns, it must have succeeded.
I don't see how anything else could happen.

However, I did just notice one potential problem when the system raises
IOError on close.  Regardless of the returned value from f->f_close(f->f_fp),
f->f_fp is set to NULL.  Thus, if close() returns -EINTR, the underlying fd
is not closed, but the file is marked as closed, and in effect the file is
"leaked".

Could this be the behavior the original poster was referring to?

Untested pseudo-patch below.

Jeff

static PyObject *
file_close(PyFileObject *f, PyObject *args)
{
        int sts = 0;
        if (!PyArg_NoArgs(args))
                return NULL;
        if (f->f_fp != NULL) {
                if (f->f_close != NULL) {
                        Py_BEGIN_ALLOW_THREADS
                        errno = 0;
                        sts = (*f->f_close)(f->f_fp);
                        Py_END_ALLOW_THREADS
                }
-               f->f_fp = NULL; 
+		if (sts != EOF) {
+                       f->f_fp = NULL;
+               }
        }
        if (sts == EOF)
                return PyErr_SetFromErrno(PyExc_IOError);
        if (sts != 0)
                return PyInt_FromLong((long)sts);
        Py_INCREF(Py_None);
        return Py_None;
}



From sholden at holdenweb.com  Thu Mar 22 23:15:15 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 22 Mar 2001 23:15:15 -0500
Subject: Python and emacs
References: <XFMail.20010322105615.shaleh@valinux.com> <mailman.985309350.25506.python-list@python.org>
Message-ID: <F8Au6.13015$_46.493149@e420r-atl2.usenetserver.com>

"Don Dwiggins" <dwig at advancedmp.net> wrote in message
news:mailman.985309350.25506.python-list at python.org...
> Sean 'Shaleh' Perry writes:
> > On 22-Mar-2001 Don Dwiggins wrote:
> >> A semi-idle question: has anyone looked into embedding Python as an
> >> extension language in Emacs, alongside Elisp?  I'm toying with the idea
of
> >> doing some Emacs enhancements/modifications, and given the choice, I'd
much
> >> rather do them in Python.  (One such enhancement would be to add
> >> Pythonwin-style "intellisense" popups to python-mode.)
> >>
>
> > An editor that spoke python would not be a bad thing.  However, trying
to get
> > emacs to speak python strikes me as an amazingly good way to waste
whatever
> > brain cells you have left.
>
> I'm not sure we're on the same page.  If by "trying to get emacs to speak
> python" you mean abandoning Elisp and rewriting the entire .el library in
> python, I'd agree wholeheartedly.
>
> What I had in mind was the ability to "load" a .py file (.epy?) _in
addition
> to_ .el files.  Then you could (re)write extensions in either language.
> Yes, of course it would likely be an ambitious project, which is why I
> called it "semi-idle" -- I just thought someone might have been
> motivated/foolish enough (delete whichever is inappropriate) to tackle it.
>
>From (fading...) memory, I seem to remember Tim Peters was fairly convincing
about why this would end up being pretty much a waste of time.

you-need-six-megabytes-of-python-to-interface-to-eighteen-of-elisp-ly
'rs  - steve





From loewis at informatik.hu-berlin.de  Wed Mar  7 04:58:58 2001
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 07 Mar 2001 10:58:58 +0100
Subject: Python 2.1b1 setup.py bug for HPUX11
References: <3AA41658.AAA154F1@email.sps.mot.com>
Message-ID: <j41ys9oq71.fsf@informatik.hu-berlin.de>

Norman Shelley <Norman_Shelley-RRDN60 at email.sps.mot.com> writes:

> In order to build python 2.1b1 for
> % uname -a
> HP-UX wshelley B.11.00 A 9000/785 2014123772 two-user license
> 
> I had to modify the setup.py script as follows. BTW, I did not request
> curses or termios in Modules/Setup either.

Obviously not - if you had, setup.py would not attempt to build it

> % diff setup.py setup.py.orig
> 385c385
> <             ###exts.append( Extension('termios', ['termios.c']) )
> ---
> >             exts.append( Extension('termios', ['termios.c']) )
> 418,419c418,419
> <             ###exts.append( Extension('_curses', ['_cursesmodule.c'],
> <             ###                       libraries = curses_libs) )
> ---
> >             exts.append( Extension('_curses', ['_cursesmodule.c'],
> >                                    libraries = curses_libs) )

What is the problem with building the termios and _curses modules? Is
it generally impossible to do so on HP/UX? Are some libraries or
header files necessary that are not generally available, but require
additional packages being installed? Are there additional #defines or
other magic needed to compile these modules?

It would be desirable to find a patch to setup.py that

a) automatically detects whether it can build these modules on HP/UX,
b) automatically skips them when building them is not possible, and
c) has no impact on other systems.

Your patch won't fulfil requirement c), as it drops termios
altogether. If you can come up with a patch that solves the problem
automatically, please submit it to sourceforge.net/projects/python.

Thanks,
Martin


From mcfletch at home.com  Fri Mar 23 13:49:26 2001
From: mcfletch at home.com (Mike C. Fletcher)
Date: Fri, 23 Mar 2001 13:49:26 -0500
Subject: Slightly Off-Topic: Python featured in Severance: Blade of Darkness
In-Reply-To: <GBKu6.3193$4N4.474429@newsc.telia.net>
Message-ID: <000001c0b3c9$fc864430$a9807018@cr706570a>

Ooh!  There's a job that sounds like heaven, writing Python to script
virtual worlds :) .  Wonder who gets paid to do it?  You out there lucky
people?  Speak up, we want to know more!  Write an article or something on
Gamasutra.

*poof*

-----Original Message-----
From: Fredrik Lundh [mailto:fredrik at pythonware.com]
Sent: Friday, March 23, 2001 11:12 AM
To: python-list at python.org
Subject: Re: Slightly Off-Topic: Python featured in Severance: Blade of
Darkness
...
it contains about 370,000 lines of Python source code...  game
logic, sound f/x, animations, translations, you name it.
...




From lluang at northwestern.edu  Sat Mar 10 11:01:40 2001
From: lluang at northwestern.edu (Louis Luangkesorn)
Date: Sat, 10 Mar 2001 10:01:40 -0600
Subject: Python In A Nutshell - suggestions
References: <nG5q6.54910$__6.11376601@typhoon.southeast.rr.com> <Ufcq6.140786$47.2078615@news.bc.tac.net>
Message-ID: <3AAA4FE3.B471B689@northwestern.edu>

Laura,

Well, since your asking for suggestions . . .

One thing I would like the Nutshell book to include are sections on some of the commonly used modules that may not be included in the base
distribution.  I say this because for the language itself and standard modules, there is already printed documentation in the form of the Python
Essential Reference by David M. Beazley and presumably the The Standard Python Library by Fredrik Lundh that is coming out.  But in no place I've seen
is there documentation on the third party modules.

Of course this begs the question of "what are commonly used modules?"  There are a couple of lists out there.  One is the list of modules that is on the
ActiveState page, http://www.activeState.com/Products/ActivePython/Extensions.html,
These include:  tkinter, wxPython, expat, Numeric Python, and ODBC (they have mxODCB, but using DBI 2.0 compliant modules in general would make for a
useful section) Another useful list are the modules that are mentioned in Python Programing for Windows32 by Hammond and Robinson in Chapter 3.  They
only have space for a paragraph each as a teaser, but it would be good to have a few pages in a Nutshell book.  (and if you really want to know, the
modules I use are Numeric Python and ODBC, and I'm going to be choosing between tkinter and wxPython soon)

And while I'm at it, one useful book to have would be on wxWindows/wxPython.  I see references to it in several places, and a few intros on the web, but
nothing that looks like good documentation (besides help files), even on the internet.  (and this book would be useful to C, C++ as well as Python
programmers.)

Thank you. (as I look at my shelf with several binders of printed documentation)

Louis
LLewin at oreilly.com wrote:

> Hi,
> Yep--Python Nutshell is "forthcoming"...we're just in the planning stages at this point--very early on in the process.  Any suggestions are welcome!
> Laura
> LLewin at oreilly.com

--
K Louis Luangkesorn
lluang at northwestern.edu  http://pubweb.nwu.edu/~kll560  PGP:0xF3E2D362
Whatsoever things are true, ... honest, ... just, ... pure, ... lovely, ... of good report; if there be any virtue, and if there be any praise, think on
these things.- motto - Northwestern University




From chris.gonnerman at usa.net  Thu Mar  1 19:45:23 2001
From: chris.gonnerman at usa.net (Chris Gonnerman)
Date: Thu, 1 Mar 2001 18:45:23 -0600
Subject: Lower bounds of lists
References: <2mjt9tcfa0o55ap99p452trjj3u7h3qst6@4ax.com>
Message-ID: <00cb01c0a2b2$d1063fe0$a100000a@local>

----- Original Message ----- 
From: "Michael Prager" <Mike.Prager at noaa.gov>
Subject: Lower bounds of lists
> As a newcomer to Python, I'm curious about the choice to number
> list elements starting at zero, rather than one.  My impression
> is that there is no user option to change that in a specific
> program.  Correct?

Correct.

> It seems an odd choice, as humans count from one, and Python in
> other respects seems quite logical and well thought out.

Humans count from 1.  Hackers count from 0.

> Starting lists at zero also seems to have had ripple effects,
> such as 
> 
> range(1,100) giving (1, 2, 3, ... , 99)

This gives me headaches sometimes also; JavaScript does similar things.
 
> which leads to such odd conscructions as 
> 
> for i in range(start ,stop+1):
> #do something

Sometimes.  If the values of i are not directly important (just the
number of iterations), I just use

    for i in range(iterations):

> Though I don't consider quirks in computer software lovable, I'm
> convinced that Python's virtues far outweigh any inconvenience
> caused by this.  I am curious, though, how this choice came
> about.  Can anyone refer me to source material on that, or
> explain the logic behind it?

C does it that way.  So, in general, does assembler, and the internal
representation of unsigned integer numbers runs from 0 to (2^bits)-1, 
where bits is the number of bits of data storable.

Counting from 0 is counter-intuitive to humans, but so is much of
computer science.  Ultimately, it's because Guido wanted it that way.





From laird at gunsmoke.ecn.purdue.edu  Thu Mar 15 00:42:18 2001
From: laird at gunsmoke.ecn.purdue.edu (Kyler Laird)
Date: Thu, 15 Mar 2001 05:42:18 +0000 (UTC)
Subject: PIREP: LAF-HYS-FMN-LGB-DRO-HYS-LAF (Python 9)
Message-ID: <98pknq$m7h$1@mozo.cc.purdue.edu>

Some notes from our family trip to the Python 9 conference
in Long Beach:
	http://www.ecn.purdue.edu/~laird/flying/flights/2001/Long_Beach-A

--kyler


From fredrik at pythonware.com  Wed Mar 14 11:27:35 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 14 Mar 2001 16:27:35 GMT
Subject: Extending Python: rewriting a single method in C
References: <g0wv9vxms0.fsf@scumbag.ecs.soton.ac.uk> <slrn9apn8f.382.jepler@potty.housenet> <g0ofv6xbj4.fsf@scumbag.ecs.soton.ac.uk> <98l1i901k4u@news1.newsguy.com> <g0lmq9xxz0.fsf@scumbag.ecs.soton.ac.uk> <gYur6.119$4d3.23079@newsb.telia.net> <g0hf0wy7cw.fsf@scumbag.ecs.soton.ac.uk>
Message-ID: <XZMr6.261$4d3.54975@newsb.telia.net>

Jacek Generowicz wrote:
> No, I'm quite sure that I didn't . . . for some reason I was buried
> in http://www.python.org/doc/current/ext
>
> (Oops! Blush)
>
> > http://www.python.org/doc/current/api/
>
> This is much more like it.

maybe it's time to merge these two documents into one?

(or at least make them refer/link to each other)

Cheers /F




From spamdropbox at myrealbox.com  Sun Mar  4 09:41:28 2001
From: spamdropbox at myrealbox.com (Walter Hofmann)
Date: Sun, 4 Mar 2001 15:41:28 +0100
Subject: Clearing Screen
References: <20010303231446.29909.00000348@ng-fn1.aol.com>
Message-ID: <slrn9a4l0o.1u1.spamdropbox@frodo.uni-erlangen.de>

On 04 Mar 2001 04:14:46 GMT, Sr63085 <sr63085 at aol.com> wrote:
>
>Is there a way to clear all text from the screen in a Python Script
>window?(similar to clreol() with conio.h in C/C++)
>Thanks in advance.

import os
clear = os.popen ("clear").read ()

print "Hallo"
print clear


Walter


From wware at world.std.com  Mon Mar  5 21:35:08 2001
From: wware at world.std.com (Will Ware)
Date: Tue, 6 Mar 2001 02:35:08 GMT
Subject: MySQL-python build griefs
References: <mailman.983838187.21896.python-list@python.org>
Message-ID: <G9r8IK.n58@world.std.com>

Mark Wilson (m.wilson at bibliocraft.com) wrote:
> this may well have _nothing_ to do with your problem however, I experienced
> similar grief with distutils on a similar box.

I finally did find a fix. There's a point in sysconfig.py around line
195 or so where it says something like:

    try: v = string.atoi(v)
    except ValueError: pass
    done[n] = v

Then later it depends upon the fact that done[n] is a string, when it
tries to concatenate it with other strings. I replaced the first line
above with "try: vvvv = string.atoi(v)" so that v would remain a string
and not be converted to a number. I think I could have equivalently
just removed the "try" and "except" lines entirely.
-- 
-----------------------------------+---------------------
 22nd century: Esperanto, geodesic | Will Ware
 domes, hovercrafts, metric system | wware at world.std.com


From thomas.heller at ion-tof.com  Wed Mar  7 05:34:10 2001
From: thomas.heller at ion-tof.com (Thomas Heller)
Date: Wed, 7 Mar 2001 11:34:10 +0100
Subject: wxPython+py2exe+win98 ImportError: DLL load failed: A device attachedto the system is not functioning
References: <mailman.983957596.4438.python-list@python.org>
Message-ID: <9852r3$9mlb$1@ID-59885.news.dfncis.de>

"Clark C. Evans" <cce at clarkevans.com> wrote in message news:mailman.983957596.4438.python-list at python.org...
> Hello.  I've recently discovered py2exe and wanted to give
> it a test run (I'm looking for a good software distribution
> system for Python on Windoz).
>
> I was able to make an executable of a command-line python
> program, and it workes just great on my Win2K development
> box, and my Win98 install.  Further, I was able to create a
> wxPython program (the first example from the wxpython.org
> tutorial) and it works great on Win2K.  However, the
> executable created gives the following error when I run it
> on a fresh copy of Win98:
>
>   Traceback (most recent call last):
>     File "<string>", line 1, in ?
>     File "imputil.pyc", line 88, in _import_hook
>     File "<string>", line 20, in _import_top_module
>     File "imputil.pyc", line 197, in import_top
>     File "imputil.pyc", line 252, in _import_one
>     File "<string>", line 95, in _process_result
>     File "wxPython\__init__.pyc", line 20, in ?
>     File "imputil.pyc", line 78, in _import_hook
>     File "imputil.pyc", line 323, in _do_import
>     File "imputil.pyc", line 248, in _import_one
>     File "<string>", line 130, in get_code
>   ImportError: DLL load failed: A device attached to
>                the system is not functioning.
>
> I was wondering if other people have the same problem
> or if I'm doing something wrong here.  Thanks!  Clark Evans
>
Yes, this came up before. The problem is that MSVCRT.DLL and
MSVCIRT.DLL are system dlls, which must live in the system
directory. py2exe shouldn't copy them at all into the dist\
directory.

You have two options:
- Delete these two files from the dist directory
and hope that they are present on the target system
in the correct location.

- Create an installation program (with WISE or InnoSetup
for example) for your exe, and make sure these files are
installed into the SYSTEM directory with proper version checking.

Thomas





From tjreedy at udel.edu  Wed Mar  7 19:19:07 2001
From: tjreedy at udel.edu (Terry Reedy)
Date: Wed, 7 Mar 2001 19:19:07 -0500
Subject: I come to praise .join, not to bury it...
References: <kuofvfjb77.fsf@lasipalatsi.fi> <000001c0a64c$497df250$a9807018@cr706570a> <mailman.983997159.4842.python-list@python.org> <3AA6B157.2B62BFA5@wwc.com>
Message-ID: <986jqe$gre$1@news.udel.edu>

"Steve Williams" <stevewilliams at wwc.com> wrote in message
news:3AA6B157.2B62BFA5 at wwc.com...
> > Continuing on into the darkness:
>
> ten = +.join(1,2,3,4)
>
> and
>
> sixfactorial = *.join(range(1,7))

We already have reduce and operator.x for these.

join:string::reduce:number -- almost

I believe that ''.join(str_list) == reduce(operator.add, str_list).

TJ Reedy





From barry at digicool.com  Wed Mar 28 15:04:16 2001
From: barry at digicool.com (Barry A. Warsaw)
Date: Wed, 28 Mar 2001 15:04:16 -0500
Subject: PEP 1, PEP Purpose and Guidelines
References: <mailman.985196613.7994.python-list@python.org>
	<99ifgu$jei$1@panix3.panix.com>
	<200103241546.KAA19937@panix3.panix.com>
	<mailman.985679316.32485.python-list@python.org>
	<99q9t6$5un$1@panix6.panix.com>
Message-ID: <15042.17344.327023.537293@anthem.wooz.org>

>>>>> "AM" == Aahz Maruch <aahz at panix.com> writes:

    AM> Not easily; I'm reading from the newsgroup, not the mailing
    AM> list.

Ah, and your newsreader doesn't support both followups and replies.  I
love XEmacs. :)

    >> What about Requires:?  Or Dependencies: where each entry on the
    >> line would be a PEP number (which we could hyperlink fairly
    >> easily).

    AM> Those would certainly be fine; I'm not wedded to any
    AM> particular naming scheme.  And, yes, I was assuming that the
    AM> field could contain multiple entries.

Okay, for lack of anything better (or any other suggestions), let's go
with Requires:.  I'll update PEP 1.

-Barry



From tim.hochberg at ieee.org  Tue Mar  6 09:45:56 2001
From: tim.hochberg at ieee.org (Tim Hochberg)
Date: Tue, 06 Mar 2001 14:45:56 GMT
Subject: National Intruments GPIB
References: <3AA46593.753BAE1@newfocus.com>
Message-ID: <EK6p6.2072$Xt3.397306@news1.rdc1.az.home.com>

"Ken Koller" <kkoller at newfocus.com> wrote in message
news:3AA46593.753BAE1 at newfocus.com...
> Has anyone written a python interface to some of the National
> Instruments libraries? I'm in the process of using Python to connect to
> some instrumentation and I'm writing the C code to interface to their
> GPIB API. Was hoping someone else had done it :)

I have a simple interface to NI44.2 GPIB that I wrote. I'll send you a copy
via private email to avoid cluttering C.L.P. And if anyone else needs a
copy, let me know. You're welcome to use it as a starting point.

-tim




From quinn at seniti.ugcs.caltech.edu  Wed Mar 14 18:41:50 2001
From: quinn at seniti.ugcs.caltech.edu (Quinn Dunkan)
Date: 14 Mar 2001 23:41:50 GMT
Subject: This math scares me
References: <3aad4062.1290021@News.CIS.DFN.DE> <Hwbr6.1139$Tg.127571@ruti.visi.com> <3aad4485.2348499@News.CIS.DFN.DE> <DZbr6.17917$Qb7.3117518@newsb.telia.net> <zrcr6.1144$Tg.129254@ruti.visi.com> <3AAE46B5.963FC3CE@wwc.com>
Message-ID: <slrn9b00du.edn.quinn@seniti.ugcs.caltech.edu>

On Tue, 13 Mar 2001 16:11:13 GMT, Steve Williams <stevewilliams at wwc.com> wrote:
>Languages supporting serious commercial numeric processing have the dreaded
>static typing.  Viz:
>
>    Picture 99v99 in COBOL
>
>or
>
>    Decimal(4,2) in DDL
>
>That way you 'do the right thing' in your *intermediate* calculations without
>having to sit the exam for floating point or remembering to 'adjust the scale
>and precision' for fixed point.

Except that doesn't really have anything to do with "static typing".  Some
languages may have built-in fixed precision types, but you should be able to
write your own in any language.

class Fixed(UserInt):
    def __init__(self, n, accuracy):
        self.accuracy = accuracy
        self.data = long(n) * accuracy * 10
    def __repr__(self):
        return '%d.%d' % divmod(self.data, self.accuracy * 10)
    __str__ = __repr__

How slow this winds up making your program might have a bit more relevance to
"static typing" :)

>And using print formatting to solve the problem is [wait for it] missing the
>point.

Ok, so what's the syntax of the rotten-tomato mime type again?


From garry at sage.att.com  Mon Mar 26 11:48:49 2001
From: garry at sage.att.com (Garry Hodgson)
Date: Mon, 26 Mar 2001 16:48:49 GMT
Subject: warrior geeks
Message-ID: <3ABF72F1.98CB78F@sage.att.com>

a funny thing happened at IPC9.  a bunch of were sitting in the
lounge (bar) talking, and one person mentioned that he was going
to study kendo while in japan this year.  i asked if he had previous
martial arts experience, and he said yes, he'd studied judo.
it then turned out that in this group of maybe 6-7 people, we had
this judo/kendo student, another who's done karate and now kickboxing,
another who did karate, myself (okinawan karate, aikijutsu and others),
and a pa qua practitioner.  at the next table sat another tae kwon
do/aikido
student.

it seemed like an odd coincidence, but it made me wonder: is there an 
unusually high percentage of MA practitioners among the python
community?
or the open source community?  or software in general?

enquiring minds want to know...

-- 
Garry Hodgson                   Once in a while
Senior Hacker                   you can get shown the light
Software Innovation Services    in the strangest of places
AT&T Labs                       if you look at it right


From tjg at exceptionalminds.com  Fri Mar  2 22:59:03 2001
From: tjg at exceptionalminds.com (Timothy Grant)
Date: Fri, 2 Mar 2001 19:59:03 -0800
Subject: python irc channel?
In-Reply-To: <20010301203006.A1619@tummy.com>; from jafo@tummy.com on Thu, Mar 01, 2001 at 08:30:06PM -0700
References: <97n15e$9g3$1@nntp1.ba.best.com> <20010301203006.A1619@tummy.com>
Message-ID: <20010302195903.C11607@trufflehunter.avalongroup.net>

On Thu, Mar 01, 2001 at 08:30:06PM -0700, Sean Reifschneider wrote:
> On Fri, Mar 02, 2001 at 02:39:42AM +0000, Martin Brown wrote:
> >So, is there a preferred irc channel where python people hangout?
> 
> I believe that the EFNet #python is gatewayed to the IRC server at
> irc.openprojects.net #python.  I haven't been on EFNet in quite a
> while.  It was more a social channel that occasionally broke out
> into code, but usually was talking about computers and cars and
> motorcycles.  ;-)

Motorcycles? One of my favourite subjects. Where do I sign up?

-- 
Stand Fast,
    tjg.

Timothy Grant                         tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.         <><       (503) 246-3630
>>>>>>>>>>>>>Linux, because rebooting is *NOT* normal<<<<<<<<<
>>>>This machine was last rebooted:  45 days  8:10 hours ago<<



From Jaeho.Lee at brooks.com  Thu Mar 22 11:19:13 2001
From: Jaeho.Lee at brooks.com (Lee, Jaeho)
Date: Thu, 22 Mar 2001 11:19:13 -0500
Subject: Good Editor?
Message-ID: <857F15D7E3D8D3118D290008C7CF058602C97583@mail-naeast1.brooks.com>

Brian Langenberger wrote
> I've spent countless hours editing Python code in Emacs, really.
> Between that and the interactive interpreter, I simply don't
> use anything else.

Same to me. I can not think any programming without Emacs. On Unix and
Windows both.
If you are using python only, there can be better option. 
But for some like me to use C++, python, SQL, Java ..., Emacs is the best
option for all languages, IMO.

Personally, I used vi for 6 yrs. I was most-advanced-user of vi in my
company.
After changing to Emacs for 2 months, I could not go back. 

/Jaeho



From hamish_lawson at yahoo.co.uk  Fri Mar  9 04:52:09 2001
From: hamish_lawson at yahoo.co.uk (=?iso-8859-1?q?Hamish=20Lawson?=)
Date: Fri, 9 Mar 2001 09:52:09 +0000 (GMT)
Subject: Static typing (was Re: Java guy interested in Python)
Message-ID: <20010309095209.24339.qmail@web206.mail.yahoo.com>

Michael Hudson wrote:

> If you type-annotate this, you might write (using entirely made up
> syntax):
> 
> def write16(file : <file>, int : <int>):
>     file.write(struct.pack("h",int))
> 
> but then what about passing in a StringIO object, or some user 
> wrapper of a file object?  What makes a "<file>"?  At the moment, 
> it's just a loose collection of methods, some of which are optional 
> in some circumstances.

Wouldn't there be a hierarchy of interface types for file-like objects,
with some interfaces specifying more methods than others? You'd then
choose the interface type that specified the collection of methods that
you require of your passed object.

Not-that-I'd-necessarily-want-this-all-of-the-time-ly yours,
Hamish Lawson


____________________________________________________________
Do You Yahoo!?
Get your free @yahoo.co.uk address at http://mail.yahoo.co.uk
or your free @yahoo.ie address at http://mail.yahoo.ie



From cepticismo at yahoo.com  Sun Mar  4 00:43:46 2001
From: cepticismo at yahoo.com (Filipe)
Date: Sat, 3 Mar 2001 21:43:46 -0800 (PST)
Subject: Any documentation about Python daemons?
In-Reply-To: <e0gq79.8og.ln@127.0.0.1>
Message-ID: <20010304054346.63089.qmail@web11708.mail.yahoo.com>

Take a look at the following:

Socket Programming HOWTO
http://www.python.org/doc/howto/sockets/sockets.html

Filipe



--- gradha at iname.com wrote:
> Hello.
> 
> I would like to write now a program which acts as a server on the
> same
> machine for other clients in a Linux environment, something in the
> spirit
> of chronyd, if you know the program.
> 
> However, I am new both to Python and process' intercommunication.
> Could
> you suggest me ways of doing this and/or point me to documentation
> explaining such communication methods? I first thought about nice
> pipes,
> but then I realized that the server program should be running always,
> not
> being allowed to fork, and clients could mantain multiple
> communication
> channels at the same time (ej: daemon + 2 clients). How could I do
> this?
> 
> -- 
> 
>  Grzegorz Adam Hankiewicz   gradha at iname.com -
> http://gradha.infierno.org
>  Other web pages:  http://glub.ehu.es/  - 
> http://welcome.to/gogosoftware/
> -- 
> http://mail.python.org/mailman/listinfo/python-list


__________________________________________________
Do You Yahoo!?
Get email at your own domain with Yahoo! Mail. 
http://personal.mail.yahoo.com/



From runyaga at thisbox.com  Wed Mar 21 22:59:54 2001
From: runyaga at thisbox.com (alan runyan)
Date: Thu, 22 Mar 2001 03:59:54 GMT
Subject: Please rate these Python books
References: <20010322.1445209@mis.configured.host>
Message-ID: <_Meu6.8736$8O6.1818817@typhoon.austin.rr.com>

how are you approaching python?  what is your level of experience w/ python?
what kinda books do you like? reference, explanatory, or 'by example' ?  My
favorite is Essential Reference, David Beazley did a incredible job.

1/Programming Python 2nd edition @ O'reilly

Very big, has quite a bit of information (4-5 chapters) on Tkinter (which I
dont use at all), socket programming (which I do use), internet applications
(HTMLgen, ZOPE, etc).  The book is massive and has small type face, packed
full of information.  Examples tend to be a bit longer than Core Python.  If
you want tons of information and especially plan on using Tkinter, this is
definetly the best book out of the three.  I feel this book is very good and
tends to get a bit more detailed than Core Python, but I only use ~40% of
the information.

2/ Core Python Programming / Chun / Prentice Hall

I recommended a friend of mine this book, and I read quite a bit of it.
Very well laid out, nice type face, very good (brief) examples.  I would
recommend this book.  My friend read Core Python and Programming Ruby at the
same time -- comparing the two languages.  At first he didnt like it, but
after he finished it he said it was a *very* good intro and has a very idea
of python now.  *NOTE* he finished the book.  I wouldnt imagine finishing
Programming Python.  He also concluded Ruby didnt offer anything else that
Python didnt have (that he would use) and likes the elegance.

3/ Python Essential Reference / David Beazley / New Riders Publishing

My favorite python book, great reference book if you already know python, or
like Nutshell books.  type face is very small ;(  but its compact and full
of all sorts of goodies.  not as many examples as a newbie would like.  at
work, all the people who use python, snatch this from each others desk.  if
you like books and already familiar w/ python -- this is a must.

cheers,
~runyaga





From digitig at cix.co.uk  Tue Mar  6 13:17:00 2001
From: digitig at cix.co.uk (Tim Rowe)
Date: Tue, 6 Mar 2001 18:17 +0000 (GMT Standard Time)
Subject: Proposal: allow '?' and '!' in identifiers
References: <96sp4s$ni7$1@news.mathworks.com>
Message-ID: <memo.20010306181748.34635A@digitig.compulink.co.uk>

In article <96sp4s$ni7$1 at news.mathworks.com>, jmarshal at mathworks.com 
(Joshua Marshall) wrote:

> Introducing some strange behavior with '!' may very well not be worth
> it.  However, while adding '?' as a legal identifier character does
> not make the Python more expressive, it can make Python programs more
> self-documenting.  The convention of ending predicate function names
> with '?' is useful, in my opinion.

Just end it with _p and everyone will think you're a converted Lisp 
hacker!



From sholden at holdenweb.com  Thu Mar 22 00:44:41 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 22 Mar 2001 00:44:41 -0500
Subject: re.match
References: <mailman.985169080.17724.python-list@python.org> <BN3u6.894$sk3.291185@newsb.telia.net>
Message-ID: <Wmgu6.4572$WC2.157632@e420r-atl1.usenetserver.com>

"Fredrik Lundh" <fredrik at pythonware.com> wrote in message
news:BN3u6.894$sk3.291185 at newsb.telia.net...
> Oleg Broytmann wrote:
> > But I, personnaly, still do not know Dylan outside of TW :)
>
> ...just like many people still don't know that Ken Thompson did
> regular expression stuff before he started hacking on Unix ;-)
>
> (and others did regular expression stuff before him, of course)
>
> Cheers /F
>
Name dropper! I thought Unix was Ken's attempt to extend the reach of
regular expressions to the user base?

regards
 Steve





From aahz at panix.com  Thu Mar  8 21:05:17 2001
From: aahz at panix.com (Aahz Maruch)
Date: 8 Mar 2001 18:05:17 -0800
Subject: from Tkinter import *
References: <mailman.983920449.5230.python-list@python.org> <slrn9ab0g7.s6.matt@happy-hour.mondoinfo.com>
Message-ID: <989dot$t6t$1@panix3.panix.com>

In article <slrn9ab0g7.s6.matt at happy-hour.mondoinfo.com>,
Matthew Dixon Cowles <matt at mondoinfo.com> wrote:
>On Tue, 6 Mar 2001 15:16:31 -0800, Timothy Grant
><tjg at exceptionalminds.com> wrote:
>>
>> All of Fredrik's examples in his guide on Tkinter use
>>
>>from Tkinter import *
>>
>>Now, I'm aware that the effbot is never wrong about anything<wink>,
>>but is it safe to assume that I should not be using the above idiom,
>>but the import Tkinter idiom instead?
>
>Rest assured, the effbot is right as usual.

Actually, I disagree with the effbot.  I used to think it was okay to
special-case "from Tkinter import *", but I've come to the conclusion
that "import *" is _always_ a Bad Idea in any sort of production code.
Sure it takes more work to write

from Tkinter import Tk, Frame, Button, Entry, Canvas, Text

but I think the payoff in clarity is worth it.
-- 
                      --- Aahz (Copyright 2001 by aahz at pobox.com)

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

"..., and some of you may regard all women as evil traps that exist only
to tease, torture, and suck out your very soul."  --DrMax


From syver at NOSPAMcyberwatcher.com  Tue Mar 13 19:53:51 2001
From: syver at NOSPAMcyberwatcher.com (Syver Enstad)
Date: Wed, 14 Mar 2001 01:53:51 +0100
Subject: win32all buld 137 win32gui fails
References: <98l8qk$oeo$1@troll.powertech.no> <DYqr6.83020$__6.16207072@typhoon.southeast.rr.com> <984516152.813137@clam-ext>
Message-ID: <98mfj7$6nt$1@troll.powertech.no>

"Mark Hadfield" <m.hadfield at niwa.cri.nz> wrote in message > As a work-around
you could try making a copy of PyWinTypes20.dll and
> renaming it PyWinTypes21.dll.

That's what I did, and it seems to be working ok in the script I used it.
But I am not entirely pleased with it.




From scarblac at pino.selwerd.nl  Wed Mar 21 18:16:02 2001
From: scarblac at pino.selwerd.nl (Remco Gerlich)
Date: 21 Mar 2001 23:16:02 GMT
Subject: I'm sure there's a good reason!
References: <memo.20010321220225.57471D@digitig.compulink.co.uk>
Message-ID: <slrn9bidcd.ejm.scarblac@pino.selwerd.nl>

Tim Rowe <digitig at cix.co.uk> wrote in comp.lang.python:
> Can any of the assembled experts explain to me why:
> 	print r"\hello"
> is fine, but
> 	print r"hello\"
> isn't? In other words, why does the backslash escape quotes in a raw 
> string? At one level I'm going to guess (without reading the spec) that 
> it's because "that's what the spec says", but why? I thought the idea of 
> raw strings was that they /didn't/ handle escapes!

They do handle escapes, but leave the backslash inside the string. Without
that it wouldn't be possible to have " inside a raw "" string. Now, you
can't end a raw string with an odd numbers of backslashes... Just some weird
consequence...

r"hello" "\\"

works.

-- 
Remco Gerlich


From thomas at cintra.no  Thu Mar  8 06:19:51 2001
From: thomas at cintra.no (Thomas Weholt)
Date: Thu, 8 Mar 2001 12:19:51 +0100
Subject: Added namespaces; now nothing works!
References: <3UJp6.10068$t21.271693@news3.oke.nextra.no>
Message-ID: <rVJp6.10070$t21.271602@news3.oke.nextra.no>

Sorry, wrong group. Was suppose to be comp.text.xml.

sorry again.

Thomas




From syver at NOSPAMcyberwatcher.com  Wed Mar 14 17:09:42 2001
From: syver at NOSPAMcyberwatcher.com (Syver Enstad)
Date: Wed, 14 Mar 2001 23:09:42 +0100
Subject: Cameron Laird Quote on www.unixinsider.com -- Probably old news to y'all, but new to me.
References: <bJOr6.9481$TW.38047@tor-nn1.netcom.ca>
Message-ID: <98oqbe$1k9$1@troll.powertech.no>

"Warren Postma" <embed at geocities.com> wrote
> All I'll add is: Well said, Cameron!
I thought that was an Uncle Bob quote Warren. I have read him state that
opinion several times before.




From alex at shindich.com  Wed Mar 14 02:57:03 2001
From: alex at shindich.com (Alex Shindich)
Date: Wed, 14 Mar 2001 07:57:03 +0000 (UTC)
Subject: Perl test
References: <98mtve$enl$1@panix3.panix.com>
Message-ID: <3AAF2426.93D10554@shindich.com>

>  [Previous]   [Next]   [Reply]   [Index]   [Home]   [Find]
>
>    * Newsgroups: comp.lang.python
>    * From: aahz at panix.com (Aahz Maruch)
>    * Subject: Perl test
>    * Date: 13 Mar 2001 21:01:34 -0800
>    * Organization: The Cat & Dragon
>
>   ------------------------------------------------------------------------
>
> http://www.bbspot.com/News/2001/03/perl_test.html
> --
>                       --- Aahz <*>  (Copyright 2001 by aahz at pobox.com)
>
> Androgynous poly kinky vanilla queer het Pythonista   http://www.rahul.net/aahz/
> Hugs and backrubs -- I break Rule 6
>
> Python:
>     Programming the way
>     Guido indented it
>
>   ------------------------------------------------------------------------
>  [Previous]   [Next]   [Reply]   [Index]   [Home]   [Find]
> Mailgate.ORG is maintained online by Dario Centofanti

I loved it! ;)))))))))))))))))))))))))))))

--
Alex Shindich
mailto:alex at shindich.com
Visit http://www.shindich.com/


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20010314/db26cbcf/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: alex.vcf
Type: text/x-vcard
Size: 194 bytes
Desc: Card for Alex Shindich
URL: <http://mail.python.org/pipermail/python-list/attachments/20010314/db26cbcf/attachment.vcf>

From danielk at aracnet.com  Thu Mar 22 00:00:56 2001
From: danielk at aracnet.com (Daniel Klein)
Date: Wed, 21 Mar 2001 21:00:56 -0800
Subject: learning python...
References: <3AB8ECB5.2197E940@noaa.gov> <m77u6.56$Hp.3284@typhoon.aracnet.com> <mailman.985205007.3092.python-list@python.org>
Message-ID: <mf1jbts73s80185cl0cmj60tj9gib0aufm@4ax.com>

On Wed, 21 Mar 2001 12:05:03 -0800, Timothy Grant <tjg at exceptionalminds.com>
wrote:

>> additional advanced concepts. From there, I would recommend "Programming
>> Python 2nd Ed." as well as the indispensible "Python Essential Reference"
>> (altho I think a new edition is in the works so you might want to wait for
>> that).
>
>I'm am searching for a review of PP2nd. Is it a significantly
>different book from PP1st?

PP2nd is BIG book with a lot of information.  Personnally I think it's too new
for anyone to have read (and absorbed) the entire book enough to review it
fairly. However, I have already used several examples from the book in my own
code and I am pleased so far.

>Also. I'd love to see a new version of Beazley's book. I know
>that Ivan isn't reading this list at the moment, but I found
>the colophon of PER to be a fascinating and enjoyable bit of
>reading. Hope he gets to do something in the update too.

I understand that one will be published in June.

Dan



From sholden at holdenweb.com  Thu Mar 22 22:23:48 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 22 Mar 2001 22:23:48 -0500
Subject: Good Editor?
References: <3ab9a32b$1_1@news2.one.net> <3AB9FAC9.97EA420E@student.liu.se> <slrn9bjv7m.4qv.grey@teleute.dmiyu.org> <99d780$l2g$1@laurel.tc.umn.edu> <3ABA20F7.AB71D3B4@student.liu.se>
Message-ID: <Rozu6.8994$G72.208991@e420r-atl1.usenetserver.com>

"Rikard Bosnjakovic" <rikbo716 at student.liu.se> wrote in message
news:3ABA20F7.AB71D3B4 at student.liu.se...
> Brian Langenberger wrote:
>
> > (E)ventually (M)alloc()s (A)ll (C)omputer (S)torage
>
> Escape Meta Alt Control Shift
> Eighteen Megs And Constantly Swapping
>
>
> > I've spent countless hours editing Python code in Emacs, really.
> > Between that and the interactive interpreter, I simply don't
> > use anything else.
>
> The best part is C-c ! and C-c C-c in the python-mode. IMNSHO ofcourse.
>
(E)very (M)inor (A)ttention (C)osts (S)anity?

regards
 Steve





From doughellmann at bigfoot.com  Sat Mar 24 12:10:34 2001
From: doughellmann at bigfoot.com (Doug Hellmann)
Date: 24 Mar 2001 12:10:34 -0500
Subject: [ANN] HappyDoc 1.4 releasedh
Message-ID: <20010324171010.LHZO28032.femail2.sdc1.sfba.home.com@localhost>

Announcing the latest version of HappyDoc, a Python documentation
extraction tool.

  HappyDoc is a tool for extracting documentation from Python source
  code.  It differs from other such applications by the fact that it
  uses the parse tree for a module to derive the information used in
  its output, rather that importing the module directly.  This allows
  the user to generate documentation for modules which need special
  context to be imported.

  Version 1.4 -- 

      - Handle Packages separately from Modules.  This behavior is
        optional and consists mostly of creating intermediate level
        TOC pages and not showing subdirectories of a Package on the
        Package's parent TOC page.

      - Changed default title to include instructions about how to
        change the title

      - Show exceptions in plugins when there is a problem loading a
        module

      - Added a command line argument to support ignoring specific
        subdirectories.

      - Fixed problem with ignoring directories by correcting name
        comparison logic.

      - Change &quot;link&quot;:URL style references which were HTML
        quoted back to StructuredText so that hyperlinks can be embedded
        within docstrings. (reported by "Mihalopoulos Evangelos"
        <mihalop at yahoo.com>)

      - Changed base class arrangement for docset so the basic interface
        is documented as part of the docset base class.

      - Moved some formatter functions into the docset, since they match
        that API better and more properly belong there.

      - Fixed bug 131751, problem with string dereference in
parseinfo.py

      - Set the DOCTYPE in hdformatter_htmlfile.py.  (Thanks to
        Shannon -jj Behrens <jjinux at yahoo.com> for the patch.)

      - Fixed problem with references to extra files when using output
        prefixes.

      - Changed test harness to use the HappyDoc class directly instead
        of calling out to a separate program.

      - Output a oneliner for referenced pre-formatted documents.

      - Modified parseinfo.py to improve speed

      - Expanded information extracted about exceptions thrown by
        functions

      - Simplified the regressino test to make it faster by removing
        Zope docs from the default regression test suite

Download

  Download the latest version of HappyDoc from the home page on
  SourceForge:

     http://sourceforge.net/projects/happydoc

Doc-string Format

  How does an author write documentation so that it will be marked up
  and look fancy?  This is a perennial question for Python, and seems
  to have introduced a roadblock into the development of more robust
  and useful documentation tools.  By separating the formatter classes
  from the docset classes, HappyDoc allows a user to create their own
  formatter to interpret comments in any way they see fit.

  The default for the HTMLTableFormatter (the default formatter for
  HappyDoc) is to treat __doc__ strings as StructuredText.  Don't like
  StructuredText?  Write your own formatter that does something
  different and drop it into place.

Documentation not in Doc-strings

  It is not always desirable to put all documentation in __doc__
  strings.  Sometime, notably when working with Zope, special meaning
  is attached to the presence of __doc__ strings.  For this reason,
  and to support existing code which might not have __doc__ strings,
  HappyDoc will find and extract documentation in Python comments.

  Comment documentation can contain all of the same formatting as
  __doc__ strings.  The preceding comment marker will be stripped off
  and the lines will be assembled and treated as a block of
  StructuredText.

  To use this feature, it is important to place the comments
  **before** the named object which they describe.  In this example:

      #
      # Class documentation goes here
      #
      class ClassWithNoDocStrings:
         "Using __doc__ strings overrides comment documentation."

         def method1(self, params):
             "This method uses a __doc__ string."
             pass

         #
         # Method2 does not use a __doc__ string.
         #
         def method2(self):
             pass

  The output would include the __doc__ strings for the class and for
  method1.  It would also make it appear that method2 had a __doc__
  string with the contents "Method2 does not use a __doc__ string."






From will at nospam.misconception.org.uk  Mon Mar 19 19:59:56 2001
From: will at nospam.misconception.org.uk (Will Newton)
Date: Tue, 20 Mar 2001 00:59:56 +0000
Subject: Why "from __future__" stinks; a counter-offer
References: <mailman.985029808.7088.python-list@python.org> <memo.20010319213204.8771C@digitig.compulink.co.uk> <9965t7$s57$1@panix6.panix.com>
Message-ID: <9969ta$2is$1@newsg3.svr.pol.co.uk>

Aahz Maruch wrote:

> So far, no other person has actually come up with a scheme that addreses
> the problem at hand, as opposed to simply whining, suggesting minor and
> worthless variations, or putting forth even *more* complicated solutions.
> One of the nice things about this solution is that if it is decided to be
> an ugly wart, it can be easily dropped for Python 2.2.

How often is this "from __future__" expected to be used?

Just experimental code written for bleeding edge versions of Python or a 
significant portion of code (that will eventually become legacy code)?



From aleaxit at yahoo.com  Thu Mar 22 14:35:33 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 22 Mar 2001 20:35:33 +0100
Subject: Why Python/Jython?
References: <99cl9e$b2b$1@Merlin.et.bocholt.fh-gelsenkirchen.de> <mailman.985260089.28444.python-list@python.org> <1901D50DC30D30E4.16551C1AB0FB4363.C0D72E208759F8A7@lp.airnews.net>
Message-ID: <99dkg301894@news1.newsguy.com>

"Cameron Laird" <claird at starbase.neosoft.com> wrote in message
news:1901D50DC30D30E4.16551C1AB0FB4363.C0D72E208759F8A7 at lp.airnews.net...
    [snip]
>     meaningful.  People do NOT start learning
>     Java lightly; it often takes hours just to
>     install and configure a minimal JDK, let

*blink* -- I'm hardly a Java-ite, myself, but
"install and configure a minimal JDK" I would
estimate as typically a 10-minute job (on, say,
NT -- is it an order of magnitude harder on,
say, Linux?!).  Maybe I misunderstand what
you mean...?

I'll underwrite the rest of your post (haven't
had occasion to use Jython 'in anger', yet, but
from just playing with it it does seem a great
way to exploit the huge libraries available in
and for Java), but this point seems strange.


Alex





From owen at astrono.junkwashington.emu  Fri Mar  2 17:41:30 2001
From: owen at astrono.junkwashington.emu (Russell E. Owen)
Date: Fri, 02 Mar 2001 14:41:30 -0800
Subject: gah! I hate the new string syntax
References: <mailman.983485923.1136.python-list@python.org>
Message-ID: <97p7iq$kk2$1@nntp6.u.washington.edu>

In article <mailman.983485923.1136.python-list at python.org>,
 "Sean 'Shaleh' Perry" <shaleh at valinux.com> wrote:

>return ";".join(["%s=%s" % (k, params[k]) for k in params.keys()])
>
>every day python seems to be moving closer to the line noise aspect of
>coding.
>

Mostly I think string methods are a great idea. One thing that annoyed 
me about python early on was the plethora of global functions that made 
more sense to me (a smalltalk lover, I admit) as methods. So overall I'm 
very pleased to see the new string methods.

However, I do agree that join is not intuitive. I believe the problem 
(at least for the way I look at it) is that join should be a list 
method, not a string method. I.e.:

  joined_string = ['a', 'b'].join(', ')

makes a lot of sense to me. The current join string method does not.

-- Russell


From robin at alldunn.com  Fri Mar 16 15:02:38 2001
From: robin at alldunn.com (Robin Dunn)
Date: Fri, 16 Mar 2001 12:02:38 -0800
Subject: socket.ssl for win32
Message-ID: <01ff01c0ae54$0d6ff1e0$0100a8c0@Rogue>

Does anybody have a pre-built _socket.pyd (and a _socket_d.pyd too if
possible) for Python 2.0 that includes the OpenSSL support?

--
Robin Dunn
Software Craftsman
robin at AllDunn.com       Java give you jitters?
http://wxPython.org      Relax with wxPython!







From stephen_purcell at yahoo.com  Fri Mar  9 11:38:03 2001
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Fri, 9 Mar 2001 17:38:03 +0100
Subject: Static typing (was Re: Java guy interested in Python)
In-Reply-To: <20010309112734.G4546@harmony.cs.rit.edu>; from dsh8290@rit.edu on Fri, Mar 09, 2001 at 11:27:34AM -0500
References: <20010309095209.24339.qmail@web206.mail.yahoo.com> <20010309112734.G4546@harmony.cs.rit.edu>
Message-ID: <20010309173802.A1914@freedom.puma-ag.com>

D-Man wrote:
> I do like the ERR side of compile-time type checking however.  Perhaps
> someone should create a Py-lint that will check python source for this
> sort of thing.

http://www.dotfunk.com/projects/pylint

> Then again maybe it's already called PyUnit or doctest.

You could indeed consider those an alternative, since a thorough test suite
would find all the same errors. <wink>

-Steve

-- 
Steve Purcell, Pythangelist
Get testing at http://pyunit.sourceforge.net/
Any opinions expressed herein are my own and not necessarily those of Yahoo



From steffen.ries at sympatico.ca  Sat Mar 24 09:38:59 2001
From: steffen.ries at sympatico.ca (Steffen Ries)
Date: Sat, 24 Mar 2001 14:38:59 GMT
Subject: newbe   how to read in a string?
References: <3ABB4E70.54885C9F@noos.fr> <mailman.985355670.6600.python-list@python.org> <99g1ac02fug@news2.newsguy.com>
Message-ID: <m366gz44g1.fsf@gondolin.beleriand>

"Alex Martelli" <aleaxit at yahoo.com> writes:

> "Chris Gonnerman" <chris.gonnerman at usa.net> wrote in message
> news:mailman.985355670.6600.python-list at python.org...
>     [snip]
> >     import string
> >     seq = string.split(a, ",")
> >     for i in range(1, len(seq)): # this skips the 0 index,
> >         seq[i] = float(seq)      # which needs to stay a string.
> >
> > In Python 2.0 the code might be shorter, using string methods and list
> > comprehension, but I personally prefer the "old way" myself.
> 
> The 'old way' would be
>     seq = map(float, string.split(a, ','))
> and it still works fine of course, though I'd rather spell it
>     seq = map(float, ','.split(a))

You're (ab)using ",".join(seq) too much, it seems :-)

What you wanted to say is: "seq = map(float, a.split(','))" (not
counting the conversion problem of the first element...)

/steffen
-- 
steffen.ries at sympatico.ca	<> Gravity is a myth -- the Earth sucks!


From pearu at cens.ioc.ee  Sat Mar 17 02:38:17 2001
From: pearu at cens.ioc.ee (Pearu Peterson)
Date: Sat, 17 Mar 2001 09:38:17 +0200
Subject: Thoughts about extensions to the Python language
In-Reply-To: <3AB2FA0F.E61862B0@seebelow.org>
References: <mailman.984608060.6280.python-list@python.org> <3AB2FA0F.E61862B0@seebelow.org>
Message-ID: <Pine.LNX.4.21.0103170930420.26659-100000@kev.ioc.ee>

On Fri, 16 Mar 2001, Grant Griffin wrote:

> > I have written
> > systems using several other languages, including C, C++ and Perl, and none
> > of these languages can even come close of Python in terms of readability.
> 
> Still, one can write FORTRAN in any language.  (But it takes so much
> work to write FORTRAN in Python that lazy folks don't bother.)

What do you mean in the parenthesis? Wrapping Fortran programs to Python
is now even easier than wrapping C programs. Take a look at f2py or
pyfort. By `easier' I mean that e.g. f2py can scan your Fortran sources
and automatically generate extension modules for Python. I don't know any
such tool for C, do you?

Pearu



From costas at meezon.com  Thu Mar  1 19:24:02 2001
From: costas at meezon.com (Costas Menico)
Date: Fri, 02 Mar 2001 00:24:02 GMT
Subject: compile() source code from a string
Message-ID: <3a9ee600.391990@News.CIS.DFN.DE>

I would like to use the compile() function to compile source code from
a string variable.

The commands I use is as follows:

code='for i in range(3): print i'
comp=compile(code,'','exec')
exec(comp)

This works well.

However I would like to execute the following example from string
variable:

if x == 1:
  print "1"
  print "2"
  print "3"
elif x = 2:
  print 'A'
  print 'B'
  print 'C'  

print 1+2


How can I put this into a string variable that the compile() function
would accept? I tried putting chr(13)+chr(10) between each line but it
gives a syntax error. Also putting a \n or \r does not help either.

The only option that I see is to write it to a file and use
compile_file() but that would be time consuming.

Does anyone have any ideas how to do this?

Thanks

Costas





From bencr at bigpond.com  Wed Mar 21 07:17:36 2001
From: bencr at bigpond.com (Ben)
Date: Wed, 21 Mar 2001 23:17:36 +1100
Subject: Pick Systems D3 Database
Message-ID: <rG%t6.555$45.2416@newsfeeds.bigpond.com>

Hi,

I have not been able to find any direct information about this so I thought
I would ask here in the slim chance someone may have had to do something
similar ... what would be the best way to push/pull information from a D3
(PickSystems) database running on AIX using Python? Is ODBC my only option
or are there others? If you have done something similar how succesful was
your solution?

Thanks in advance




From aahz at panix.com  Tue Mar 20 20:47:35 2001
From: aahz at panix.com (Aahz Maruch)
Date: 20 Mar 2001 17:47:35 -0800
Subject: Yet Another PEP:  Query Protocol Interface or __query__
References: <Pine.LNX.4.21.0103201837570.21343-100000@clarkevans.com>
Message-ID: <99917n$2jf$1@panix6.panix.com>

[posted & e-mailed]

In article <Pine.LNX.4.21.0103201837570.21343-100000 at clarkevans.com>,
Clark C. Evans <cce at clarkevans.com> wrote:
>
>    A new built-in method, __query__ is proposed.  This method has
>    a single argument, an interface protocol identifier, and either
>    returns an object supporting the given protocol, or throws an
>    unknown exception.
>
>    An interface protocol identifier is a lower-case string having
>    a reverse DNS host name string:   TLD '.' DOMAIN ['.' NAME ]*
>    Where TLD is a top level domain such as 'com' or 'uk.co'.  DOMAIN
>    is a registered name in the given TLD.  And one or more optional
>    NAME separated by a period.  NAME is a sequence of one or more
>    alphabetic characters including the dash '-' character.  See the
>    relevant ITEF specifications for specific details.

It's not at all clear to me why this domain thing is being suggested.
Is there supposed to be some kind of lookup mechanism associated with
it?  Why is this a string instead of a sequence object (tuple or list)?
Most importantly, how does one find out what interfaces an object *does*
support?

>Example Usage
>
>    >>> class EggsOnly:
>            def eggs(self,str): print "eggs!" + str
>            def __query__(self,protocol):
>                if protocol == "org.python.example.eggs": return self
>                raise "unknown: " + protocol
>
>    >>> class EggsAndHam:
>            def ham(self,str): print "ham!"+str
>            def __query__(self,protocol):
>                if protocol == "org.python.example.ham": return self
>                if protocol == "org.python.example.eggs":
>                    return EggsOnly()
>                raise "unknown" + protocol

This seems to be implying that it is appropriate to return a brand-new
object with no association with the original object.  Is that correct?

Does it make sense to have something like:

class HamAndEggs:
    def __query__(self, protocol):
        if protocol.find("org.python.example.ham") == 0:
            return self
h = HamAndEggs()
e = h.__query__("org.python.example.ham.eggs")
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

"I won't accept a model of the universe in which free will, omniscient
gods, and atheism are simultaneously true."  -- M


From f8dy at yahoo.com  Fri Mar  2 11:26:48 2001
From: f8dy at yahoo.com (Mark Pilgrim)
Date: Fri, 2 Mar 2001 11:26:48 -0500
Subject: gah! I hate the new string syntax
References: <mailman.983485923.1136.python-list@python.org> <B7In6.16281$Qb7.2658902@newsb.telia.net>
Message-ID: <97ohh2$polt0$1@ID-77331.news.dfncis.de>

"Fredrik Lundh" <fredrik at pythonware.com> wrote in message
news:B7In6.16281$Qb7.2658902 at newsb.telia.net...
> Sean 'Shaleh' Perry wrote:
> > return ";".join(["%s=%s" % (k, params[k]) for k in params.keys()])
> >
> > every day python seems to be moving closer to the line noise aspect of
> > coding.
>
> you're using list comprehensions and the wrong dictionary method,
> and then you're complaining about string methods? ;-)
>
> ("items()" returns a key/value tuple that you can pass directly to
> the formatting operator.  and nobody will complain if you stick to
> string.join for stuff like this...)

I recognize this code as an example from my book "Dive Into Python",
specifically http://diveintopython.org/odbchelper_join.html, and would like
to (partially) defend it in that context.

1. I used Python 1.5.x very little; I really only ramped up when 1.6/2.0 was
released.  I *like* list comprehensions.  I prefer them over map/lambda
because of scope issues, and I prefer them over for loops because they force
me to think Pythonically, and they emphasize Python's strengths (less code,
richer data structures).  To my eyes, list comprehensions fulfill Python's
promise of being readable pseudo-code, not readable line noise.  YMMV.
2. I considered using .items() and rejected it for this example, because I
didn't want to get into explaining multi-variable assignment in chapter 1.
(It is discussed later in chapter 3:
http://diveintopython.org/fileinfo_multiassign.html, including an example of
iterating through a dictionary using items().)  I still believe this was the
right choice; you can't cover everything at once, even when you're talking
to smart people.  However, I will add a footnote after the discussion of
this example, noting that there is an even *more* elegant way of
accomplishing the same thing using multi-variable assignment, to be covered
in chapter 3.
3. I *like* string methods, even .join().  I am aware that this puts me in a
very small minority on comp.lang.python.  But I would maintain that
";".join(...) only seems odd to you because you see it as the result of a
strained migration from string.join(..., ";"), instead of seeing it with
fresh eyes.  However, I may add a footnote explaining the old style
string.join, if only because new programmers will eventually come across old
code that uses it.

If current trends continue, *far more people* will learn Python 2.0 than
ever learned Python 1.x.  They will all take list comprehensions and string
methods for granted, and stare at you blankly when you reminisce about the
good old days of Python 1.5.2.

-M feeling-old-and-crotchety-already-ly
You're smart; why haven't you learned Python yet?
http://diveintopython.org/





From joconnor at cybermesa.com  Thu Mar  1 18:31:47 2001
From: joconnor at cybermesa.com (joconnor at cybermesa.com)
Date: Thu, 01 Mar 2001 23:31:47 GMT
Subject: Newbie: Casting Int to string....
References: <3a9ed412$1@news1>
Message-ID: <3a9edb8a.95346080@news.uswest.net>

On Thu, 1 Mar 2001 18:01:09 -0500, "Digital Hyakugei"
<jos at trapeze.removespam.com> wrote:

>Ok, pretty simple, no? I can not find out how to do the following...
>
>i've got a variable:
>
>aNumber = 1
>
>Now i have a function
>
>nntp.stat( aVal )
>
>where aVal has to be a string.
>
>I cant do
>
>nntp.stat( aNumber )
>
>cuz it expects a string, not an int.
>
>How, oh how, do i cast an int to a string?
>I'm sure its something that is forehead slapping obvious, but i just cant
>seem to find it...
>


Yes, it is :)

You can  do either....

	nntp.stat (str(aNumber))

or teh shorter
	nntp.stat (`aNumber`)

Notice that the single quotes are backwards

Take care,
Jay O'Connor


From Eugene.Leitl at lrz.uni-muenchen.de  Thu Mar 29 07:32:39 2001
From: Eugene.Leitl at lrz.uni-muenchen.de (Eugene Leitl)
Date: Thu, 29 Mar 2001 14:32:39 +0200 (MET DST)
Subject: CGI XML-RPC (xmlrpccgi.py) demo?
In-Reply-To: <xUEw6.4147$4N4.842877@newsc.telia.net>
Message-ID: <Pine.GSO.4.03.10103291336490.23334-100000@sun1.lrz-muenchen.de>

Caution, confused braindump follows.

I'm trying to wrap the coils of my puny little mind around XML-RPC,
specifically Kalle Svensson's CGI XML-RPC server:

	http://mail.python.org/pipermail/python-announce-list/2001-March/000698.html

(interesting enough, according to Google there seems to be also Jeff's

	http://starship.python.net/crew/jjkunce/python/xmlrpccgi.tgz

advertised as "A XML-RPC "server" that is accessed via a web server
through CGI calls (as opposed to a long-running XML-RPC server that
monitors a dedicated port).", which seems 1) entirely unrelated to 2) much
smaller than Kalle's code 3) designed for Winders).

To reiterate, I'm trying to be able to trap URI's looking like

	http://blah.net:666/cgi/boo.cgi?hiss=aargh#boom

or parameters POSTed via HTML like

<FORM TARGET="result">
<INPUT TYPE=button VALUE="Start search now!" onClick="jsCopyFormData();return (0);">
</FORM>
<FORM name=submitform METHOD=post TARGET="result" ACTION="">
<INPUT TYPE=hidden NAME="smi"     VALUE="">
<INPUT TYPE=hidden NAME="jme"     VALUE="">
<INPUT TYPE=hidden NAME="stype"   VALUE="">
<INPUT TYPE=hidden NAME="sstype"  VALUE="">
<INPUT TYPE=hidden NAME="checkin" VALUE="$cwkey">
<INPUT TYPE=hidden NAME="usr" VALUE="$user">
</FORM> 

to some piece of code called boo.cgi, using above arguments as parameters.
Lacking that, I don't mind going the usual route with CGI

import cgi
results=cgi.FieldStorage()
etc.

To make things more difficult, I also might have to process parameters
passed via GET, which current (ick, feh) Perl bit does thusly:

sub get_params {
        if ($ENV{'REQUEST_METHOD'} eq 'GET') {
                @pairs = split(/&/, $ENV{'QUERY_STRING'});
        } elsif ($ENV{'REQUEST_METHOD'} eq 'POST') {
                read(STDIN, $buf, $ENV{'CONTENT_LENGTH'});
                @pairs = split(/&/, $buf);
        } else {
#               &error('request');
        }
        foreach $pair (@pairs) {
                local($name, $value) = split(/=/, $pair);
                $name =~ tr/+/ /;
                $name =~ s/%([a-fA-F0-9][a-fA-F0-9])/pack("C",hex($1))/eg;
                $value =~ tr/+/ /;
                $value =~ s/%([a-fA-F0-9][a-fA-F0-9])/pack("C",hex($1))/eg;
                $value =~ s/<!--(.|\n)-->//g;
                $form{$name} = $value;
#               last;
        }
        return(%form);
}
1;                 

Anyway, as above xmlrpccgi demo doesn't seem to work without input, and I
can't generate input without knowing what it does I'd appreciate any
pointers to simple demo code which I could put into cgi-bin and test
drive a bit.

I guess I could also use a regexp looking for a cgi-bin/boo.cgi pattern in
the code below instead of a specific reference, and then parse the
self.path string by hand, but that looks unnecessary ugly. I might have to
go that route, though, if I can't figure any better way to process
arguments. Anyway, comments welcome.

import StringIO
import SimpleHTTPServer
 
BaseClass = SimpleHTTPServer.SimpleHTTPRequestHandler
 
CounterTemplate = """ <H1>Server
Statistics</H1>
 
This <A HREF=./>server</A> has been accessed
<b>%d</b> times.  """
 
count = 0
 
class MyRequestHandler(BaseClass):
 
    def do_GET(self):
        global count
        count = count + 1
        BaseClass.do_GET(self)
 
    def send_head(self):
        if self.path == "/counter.html":
            return self.send_counter()
        else:
            return BaseClass.send_head(self)
 
    def send_counter(self):
        self.send_response(200)
        self.send_header("Content-type", "text/html")
        self.end_headers()
        text = CounterTemplate % count
        return StringIO.StringIO(text)
 
def test():
    SimpleHTTPServer.test(MyRequestHandler)
 
test()  




From dnew at san.rr.com  Thu Mar 15 13:11:12 2001
From: dnew at san.rr.com (Darren New)
Date: Thu, 15 Mar 2001 18:11:12 GMT
Subject: Who's minister of propaganda this week?
References: <B98041329020E007.7DA84CAD9D0562AB.817BBC572F480869@lp.airnews.net> <98nu1m$12e@dispatch.concentric.net> <660ACCA1E4816F7E.C359CDF4168EF3FB.3B6B05538D619770@lp.airnews.net> <3AAFA13D.76E32C3F@divalsim.it> <slrn9b0ijs.ccb.kzaragoza@kzaragoza.ne.mediaone.net> <Li7s6.895$lz1.79367@ruti.visi.com>
Message-ID: <3AB105C0.A106233@san.rr.com>

Grant Edwards wrote:
> I would consider TCL to be degenerately typed: everything is a
> string.

Right.

FORTH, however, is untyped. You can pass a string to something expecting an
integer and you'll be surprised. (At least, last time I used Forth, way back
when.)
 
> I tried to write a program in TCL once many years ago.  After
> the bleeding from the ears abated, I switched to Scheme.  ;)

Tcl has gotten better!

-- 
Darren New / Senior MTS & Free Radical / Invisible Worlds Inc.
San Diego, CA, USA (PST).  Cryptokeys on demand.


From MarkH at ActiveState.com  Wed Mar 21 19:26:47 2001
From: MarkH at ActiveState.com (Mark Hammond)
Date: Thu, 22 Mar 2001 00:26:47 GMT
Subject: ASCII encoding error: ordinal not in range(128)
References: <mailman.985144766.14809.python-list@python.org>
Message-ID: <3AB9480F.4020104@ActiveState.com>

Tim Peters wrote:

 >>> UnicodeError: ASCII encoding error: ordinal not in range(128)
...
 > Anything is better than what it says now!    even-a-ham-sandwich-ly
 > y'rs  - tim

I'm a little confused.

Are you proposing:

UnicodeError: ASCII encoding error: ordinal not in ham sandwich

or

UnicodeError: ASCII encoding error: ham sandwich not in range(128)

??

I prefer the former as that is something you would never find in my 
samdwiches.  However, trying to predict exactly what Tim will stick in 
his mouth is a fools folly <wink>

mmmmm-ham-ly,

Mark.






From park997 at aol.comnotospam  Sun Mar 11 10:46:06 2001
From: park997 at aol.comnotospam (Park997)
Date: 11 Mar 2001 15:46:06 GMT
Subject: DOS python?
Message-ID: <20010311104606.07469.00000559@ng-cj1.aol.com>

Hi,

I would appreciate any pointers on obtaining a version of
python that operates on an HP LX200 (DOS 5, 80186 CPU; 
essentially 8 bit IBM XT equivalent). I found a python16 that
runs on Windows, but not on the LX200.

Thanks,

Wendell Cropper


From vecernik at aon.at  Fri Mar 30 11:14:48 2001
From: vecernik at aon.at (Oliver Vecernik)
Date: Fri, 30 Mar 2001 18:14:48 +0200
Subject: micro navigation
Message-ID: <3AC4B0F8.E923CA11@aon.at>

Hi,

I'd like to add a micro navigation to my homepage. Actually there should
be links to the title of the index.html pages. For example:

Home / Products / Hardware /

It's not a problem to generate a dict with the path as key and the title
as value. But what is the easiest way to build the path? Is there a
module to step up the path (html/products/hardware/index.html)?

Oliver


From park997 at aol.comnotospam  Wed Mar 14 09:47:57 2001
From: park997 at aol.comnotospam (Park997)
Date: 14 Mar 2001 14:47:57 GMT
Subject: main in mac?
References: <mailman.984569899.20412.python-list@python.org>
Message-ID: <20010314094757.02225.00001967@ng-cg1.aol.com>

>> >I don't know what the run all button is (not on mac), but the whole reason
>> >for this idiom is so it wont 'do anything' when imported. If a file named
>> >x.py is imported, then it's __name__ is x. If it is being run as a program,
>> >then it's __name__ is __main__. 
>> >
>> >
>> >David
>> 
>> I knew that was the expected behavior, and the
>> disadvantage of replacing if __name__ is __main__. 
>> with if 1: is that it does run when imported. Turning the run as main
>option on
>> or off produces the expected behavior on the mac.
>
>It's not that you're running on a Mac, it's that you're using an IDE. If
>you double click the saved file in the Finder the __name__ would be set
>to '__main__' as you expect -- just like if a Windows or UN*X user runs
>the file directly.
>
>Doug

This is true for .pyc files (created after import in
interactive window), but .py files only invoke the
IDE if "run as main" is not saved. I have also learned
that the pull down menu "scripts" in the ide will give
run as main behavior.

Wendell


From aleaxit at yahoo.com  Tue Mar  6 03:05:23 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Tue, 6 Mar 2001 09:05:23 +0100
Subject: Static class methods
References: <rdzo6.154$gc4.80445@news.pacbell.net> <aqa7at0b6cbqjaq23mli0flmumlp8n6s9s@4ax.com> <7SZo6.7063$qv3.3125634@nnrp5-w.sbc.net>
Message-ID: <9825od0omq@news1.newsguy.com>

"Alex Shindich" <alex at shindich.com> wrote in message
news:7SZo6.7063$qv3.3125634 at nnrp5-w.sbc.net...
> There are indeed many ways to simulate the behavior. But shouldn't such a
> strong OOP language like Python have support for static methods?

I don't think it should, not necessarily.  OOP does not have to
mean class-centered; Python does offer classes, but does not
focus on them the way some other OOP languages (such as Eiffel)
do -- rather, Python offers different kinds of objects for
different purposes.

Thus, there is no necessary reason why functions that are
not associated with a specific instance should _have_ to be
connected to a class-object rather than to a module-object
(or other kinds of objects yet -- remember that packages
as well as modules exist, for example).

I suspect that the strong desire to associate such functions
with class-objects may come from previous experience with
languages which gave one no real alternative regarding the
organizing-objects choice for subsystems, besides classes.

Personally, I don't feel any particular constraint from not
being able to assign functions as class-object attributes,
any more than I do for the similar inability regarding, say,
dictionaries, or lists, or tuples.  Sure, some more flexibility
would not be a problem (unless it imposed costs elsewhere), but
nor would it make my Python use any more 'object-oriented' --
at most, it would move some of my functions from attributes of
objects of one kind to attributes of objects of another kind.


Alex





From kelvin.chu at uvm.edu  Wed Mar  7 09:35:46 2001
From: kelvin.chu at uvm.edu (kelvin.chu at uvm.edu)
Date: 7 Mar 2001 14:35:46 GMT
Subject: Python script for deMSWordification?
Message-ID: <985h02$1pbm$1@swen.emba.uvm.edu>

Dear fellow Pythoneers;

Numerous colleagues and administration types send memos, colloqium
announcements and other things around in Micro$oft word format.  Has anyone
written a filter in Python that can convert these things into normal text?

If not, is there a basic set of rules for decoding .doc format files
anywhere that I can look at?

Cheers,

-k

-- 

  kelvin.chu at uvm.edu				     (802) 656-0064
  http://www.uvm.edu/~kchu/			FAX: (802) 656-0817



From ngps at madcap.dyndns.org  Sat Mar 10 21:33:53 2001
From: ngps at madcap.dyndns.org (Ng Pheng Siong)
Date: 11 Mar 2001 02:33:53 GMT
Subject: LWP session problems
References: <3aa9c338$1_3@huge.aa.net>
Message-ID: <98eo6h$k3j$1@clematis.singnet.com.sg>

According to Xeno Campanoli  <xeno at bigger.aa.net>:
> So, there must be a good (probably superior) web robot that does
> everything without problems or obscurity or breakage in Python...Is there
> a class in Python I can use effectively to program multiple requests in
> an SSL session?  If so, I'll try to use it soon.  I keep hearing that
> Python and Python People do it better..so?


"""M2Crypto.SSL.Session client demo: This program requests a URL from 
a HTTPS server, saves the negotiated SSL session id, parses the HTML 
returned by the server, then requests each HREF in a separate thread 
using the saved SSL session id.

Copyright (c) 1999-2000 Ng Pheng Siong. All rights reserved."""

RCS_id='$Id: sess.py,v 1.2 2000/09/11 14:52:29 ngps Exp ngps $'

from M2Crypto import Err, Rand, SSL, X509, threading
m2_threading = threading; del threading

import formatter, getopt, htmllib, sys
from threading import Thread
from socket import gethostname


def handler(sslctx, host, port, href, recurs=0, sslsess=None):

    s = SSL.Connection(sslctx)
    if sslsess:
        s.set_session(sslsess)
        s.connect((host, port))
    else:
        s.connect((host, port))
        sslsess = s.get_session()
    #print sslsess.as_text()

    if recurs:
        p = htmllib.HTMLParser(formatter.NullFormatter())

    f = s.makefile("rw")
    f.write(href)
    f.flush()

    while 1:
        data = f.read()
        if not data:
            break
        if recurs:
            p.feed(data)

    if recurs:
        p.close()

    f.close()

    if recurs:
        for a in p.anchorlist:
            req = 'GET %s HTTP/1.0\r\n\r\n' % a
            thr = Thread(target=handler, 
                        args=(sslctx, host, port, req, recurs-1, sslsess))
            print "Thread =", thr.getName()
            thr.start()
    

if __name__ == '__main__':

    m2_threading.init()
    Rand.load_file('../randpool.dat', -1) 

    host = '127.0.0.1'
    port = 443
    req = '/'

    optlist, optarg = getopt.getopt(sys.argv[1:], 'h:p:r:')
    for opt in optlist:
        if '-h' in opt:
            host = opt[1]
        elif '-p' in opt:
            port = int(opt[1])
        elif '-r' in opt:
            req = opt[1]
    
    ctx = SSL.Context('sslv3')
    ctx.load_cert('client.pem')
    ctx.load_verify_info('ca.pem')
    ctx.load_client_ca('ca.pem')
    ctx.set_verify(SSL.verify_none, 10)
    
    req = 'GET %s HTTP/1.0\r\n\r\n' % req

    start = Thread(target=handler, args=(ctx, host, port, req, 1))
    print "Thread =", start.getName()
    start.start()
    start.join()
    
    m2_threading.cleanup()
    Rand.save_file('../randpool.dat')



This is a demo program with M2Crypto, found here:

    http://www.post1.com/home/ngps/m2


Cheers.

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



From pj at sgi.com  Wed Mar  7 18:09:05 2001
From: pj at sgi.com (Paul Jackson)
Date: 7 Mar 2001 23:09:05 GMT
Subject: I come to praise .join, not to bury it...
References: <kuofvfjb77.fsf@lasipalatsi.fi> <000001c0a64c$497df250$a9807018@cr706570a> <mailman.983997159.4842.python-list@python.org>
Message-ID: <986f2h$1legl$1@fido.engr.sgi.com>

On Tue, Mar 06, 2001 at 09:46:53AM -0500, Mike C. Fletcher wrote:
| Here's a professional aesthetic's take:
| 
|       mystring.join( )
| 
| This is a perfectly normal and good-looking construct.  It seems pythonic
| and beautiful, object orientation is part of Python, and this works nicely.
| 
|       "'".join( )
|       ";".join( )
|       ".".join( )
|       "!".join( )
|       ",".join( )
| 
| All look like executable line noise.  That is, the literal string's syntax
| makes the use of dotted attributes on such a string look jarring and
| off-putting.


Hmmmm ... the following looks nicer to the part of my
brain that agreed with the above sighting of noise:

    str(",").join(x)

as in:

    >>> x=['a', 'b', 'c']
    >>> str(",").join(x)
    'a,b,c'

But my hunch is that the above will quickly look like a silly
example of excess elaboration to a frequent reader of Python code.

Time to get used to simply:

    ",".join(x)

-- 
                          I won't rest till it's the best ...
                          Manager, Linux System Software
                          Paul Jackson <pj at sgi.com> 1.650.933.1373


From aleaxit at yahoo.com  Fri Mar 16 10:39:53 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 16 Mar 2001 16:39:53 +0100
Subject: windows script host and python
References: <%tKr6.2570$pX.26174@news1.onlynews.com> <98oq3r$1e0$1@troll.powertech.no> <8Aps6.104$pl.1515@newsfeed.slurp.net>
Message-ID: <98tc4d0rmk@news1.newsguy.com>

"bowman" <bowman at montana.com> wrote in message
news:8Aps6.104$pl.1515 at newsfeed.slurp.net...
>
> Syver Enstad <syver at NOSPAMcyberwatcher.com> wrote in message
> news:98oq3r$1e0$1 at troll.powertech.no...
> >
> > Just curious but what do plan to acheive by using python with wsh that
you
> > couldn't do using normal .py files?
>
> even more curious: can Python hook up and use WMI from a normal .py?

Sure, why not?  As MS say, in their "Microsoft Windows Management
Instrumentation Scripting" paper,
"""
The WMI SDK includes a scripting API that you use to create scripts
and Visual Basic-based applications to interact with the CIM Object
Manager. It supports the following languages:
    [snip]
Additionally, you can also use any scripting language implementation
that supports Microsoft's Active Scripting technologies with this API
(such as a Perl scripting engine).
"""
and Python (with win32all) DOES support ActiveScripting, of course.

Here, for example, is the "list all running processes" WMI scripting
example translated into Python:

from win32com.client import GetObject

processes = GetObject("winmgmts:{impersonationLevel=impersonate}").\
    InstancesOf("Win32_process")

for process in processes:
    print process.Name


Alex






From bsass at freenet.edmonton.ab.ca  Thu Mar 22 17:45:23 2001
From: bsass at freenet.edmonton.ab.ca (Bruce Sass)
Date: Thu, 22 Mar 2001 15:45:23 -0700 (MST)
Subject: New PEP: The directive statement
In-Reply-To: <j4n1aeywze.fsf@informatik.hu-berlin.de>
Message-ID: <Pine.LNX.4.33.0103221408241.23215-100000@bms>

On 22 Mar 2001, Martin von Loewis wrote:
> Bruce Sass <bsass at freenet.edmonton.ab.ca> writes:
>
> > Why was (something along the lines of)...
> >
> > use:
> >     __future__.nested_scopes
> [...]
> > ...rejected?
>
> It was not rejected - it simply was not proposed by anybody. I'd feel
> that a block structure for "settings" is overkill; it's not that you'd
> activate dozens of features that way.

I'm generalizing a little more than either the __future__ or directive
proposals.  I see wanting to `test beta features out on the general
population' as a subclass of the `is that feature/module/etc. available'
problem faced when writing a program.

For example, I'd rather do...

use:
    new_math.//
    result = new_math_based_function()
except NewMathError:
    result = old_math_based_function()

...and have the compiler use the appropriate source depending on what
the interpreter can handle, rather than do a check of the
"sys.version"[1] or try to catch a run-time exception.

I'd also rather do...

use:
    __future__.nested_scopes

...or even...

use: nested_scopes

...than...

from __future__ import nested_scopes

...or...

directive nested_scopes

...but for purely aesthetic reasons.

The way I see it there is no reason for "use: something.whatever" to
have a specific interpretation.  e.g., "use: nested_scopes" could be a
compile time switch, "use: new_math.//" acts like a preprocessor.

"use:" looks like an intuitive way to handle both (and any number of
different) types of extras to the language or its implementation, and
even be a general wy of dealing with coding for multiple versions of
the language.


> In any case, if you really want to see such syntax, draft a few lines
> of how exactly that should work, and I put them into my text (or you
> can write your own PEP).

I don't know enough about Python's internals to author a PEP[2], but
would be open to working with someone who did (so that the few lines
produced could be implemented without rewriting all of Python, etc.)


- Bruce


[1] What I'm talking about is the difference between each program
needing to know what versions of Python support which features, and
telling the compiler what to do if it (or the interpreter) can not
handle a certain feature (or a specific module, or package, is
unavailable).

[2] If I did, the "motivation" would be: "from __future__ import" is
confusing, "directive" looks messy, neither addresses things from the
programmer's perspective...




From moshez at zadka.site.co.il  Fri Mar 23 12:14:18 2001
From: moshez at zadka.site.co.il (Moshe Zadka)
Date: Fri, 23 Mar 2001 19:14:18 +0200
Subject: sharing data between modules
In-Reply-To: <20010323.073801.76065818.26070@mead.troikanetworks.com>
References: <20010323.073801.76065818.26070@mead.troikanetworks.com>
Message-ID: <E14gV8k-0000tB-00@darjeeling>

On Fri, 23 Mar 2001, "Bruce Edge" <bedge at troikanetworks.com> wrote:

> Currently I register the classes with each other using class static data, 
> so they all have a reference to each other. Ideally, I'd like to be able
> to have a few globals that every module can see.  

Every module can see every other module. If you don't have circular
dependancies, life is even easier.

-- 
"I'll be ex-DPL soon anyway so I'm        |LUKE: Is Perl better than Python?
looking for someplace else to grab power."|YODA: No...no... no. Quicker,
   -- Wichert Akkerman (on debian-private)|      easier, more seductive.
For public key, finger moshez at debian.org  |http://www.{python,debian,gnu}.org



From paul at fxtech.com  Fri Mar 23 09:52:58 2001
From: paul at fxtech.com (Paul T. Miller)
Date: Fri, 23 Mar 2001 08:52:58 -0600
Subject: Slightly Off-Topic: Python featured in Severance: Blade of Darkness
References: <NMHu6.144$NR.1628@news3.oke.nextra.no>
Message-ID: <3ABB634A.997748B@fxtech.com>

> Wouldn't Python fit just perfectly for this purpose?

Python was also used to script the RTS game, Kingdom Under Fire.


From db3l at fitlinxx.com  Thu Mar  8 22:12:23 2001
From: db3l at fitlinxx.com (David Bolen)
Date: 08 Mar 2001 22:12:23 -0500
Subject: Name of the Script
References: <20010304.185354.2114937732.8277@localhost.localdomain> <NJCo6.6720$qv3.2834377@nnrp5-w.sbc.net> <FpmjOjvwe5bm0Bg09LDTMPhh5vji@4ax.com> <Pine.LNX.4.21.0103051715150.18263-100000@kev.ioc.ee> <cbKjOqPRwNza00jOm7A1HBDkRYj7@4ax.com> <upuft9z0s.fsf@ctwd0143.fitlinxx.com> <uFCnOk=ag9iJ0BBH3CPw9xrMiTRB@4ax.com>
Message-ID: <ug0gnk548.fsf@ctwd0143.fitlinxx.com>

Paul Moore <paul.moore at uk.origin-it.com> writes:

> Agreed, in principle. However, in practice, I do have occasion to need
> this sort of thing. As a specific example, suppose I had a module
> which provided some form of logging service
> 
> def log(msg):
>     logfile.write(timestamp + ": ")
>     logfile.writeline(msg)
> 
> ... you know the sort of thing. Now, I want my log file to reside in
> the same directory as the calling script (not the same directory as
> the logging module), but with an extension of .log instead of .py.
> 
> The obvious, and most user-friendly, way of doing this is using the
> script name to generate a logfile name. This means looking at
> sys.argv[0] from within the module.

By "user-friendly" do you mean the user of the script or the
developer?  If the developer then I would think a clean maintainable
interface for the logging subsystem with a minimum of implicit
dependencies would be more user-friendly.  If you mean the end user of
the script, I have to admit that I rare place log files where my
scripts are - scripts are executable code and logs are data, and I
like to keep those things separate.

In terms of implementation, I'd probably (and I do) encapsulate my
logging functionality in a class (since I also like to re-route
sys.stdout there), and simply include the location and/or log file I
want to use when initializing the instance of the class.  After all,
why not have a generic log class that can write anywhere, and let the
intelligence of the user (code wise) of that log class instruct it
where to go on an as needed basis?  Better re-use, and more flexible
(for those cases where you _don't_ want the log file to go wherever
that first top level script happened to be run from).

> A less friendly way is to require the main script to call
> logger.init(os.path.abspath(sys.argv[0])) before starting processing.

Essentially this (although in my case during the instance creation).
I'm not sure why it's less friendly though.

> (Which in turn requires sys and os imports which the main script may
> otherwise not use).

True, but I wouldn't avoid it for that reason - or rather, needing
standard library modules shouldn't be a negative.  Even if the main
script doesn't use sys or os, if it imports anything else that does
you're still going to load them at some point, so you won't have a
performance hit.  And it's hard to get through all but tiny scripts
without needing sys or os.  (And sys is really a builtin anyway)

> This is a real-life case - I had a (VB Script, sorry!) program which
> did exactly this. And the initialisation step was a definit pain...

Can you clarify the pain part further?  I see this as normal
initialization that just makes sense and clarity.

This isn't to discourage your desire for another solution, but only to
point out that making the search for such a solution might also be an
indication that the approach could be re-evaluated.

> Maybe having sys.argv[0] always be the full script pathname would be
> something which could be added to Python? The interpreter must know
> where the script is, so it would be easy to do from the startup
> code...

Well, today I'm pretty sure the interpreter just passes on argv as
already initialized by the C RTL without having to do any extra work,
so it would have to encode the equivalent operation as your use of
abspath() above.  Given that you don't want the overhead of that in
your script, why impose it on everyone without an option?

There's also an interesting (or perhaps just debatable) question of
information content in the path.  If the interpreter were to decide to
munge the path to the file as specified on the command line - is what
do you do in the case of ambiguitities or where an absolute path might
lead to information loss.  For example, having the path as specified
by the user would include (under Unix) any symbolic directory links
traversed whereas an absolute path would lose that.  Under Windows, a
UNC path might (I'd have to check this) translate back to a mapped
drive if available, if the underlying Windows API call was made.
That's a loss of information content unnecessarily, since the script
itself can make an absolute path if it really wants one.

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


From aleaxit at yahoo.com  Thu Mar 15 06:41:21 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 15 Mar 2001 12:41:21 +0100
Subject: Cameron Laird Quote on www.unixinsider.com -- Probably old news to  y'all, but new to me.
References: <bJOr6.9481$TW.38047@tor-nn1.netcom.ca> <3AB00593.D1C4BBBF@fxtech.com>
Message-ID: <98qacl12c1e@news1.newsguy.com>

"Paul T. Miller" <paul at fxtech.com> wrote in message
news:3AB00593.D1C4BBBF at fxtech.com...
> > I apologize for trotting it out again, if I am repeating something
earlier
> > discussed, but I couldn't find any direct reference to it (my nntp
server
> > sucks, only goes back 7 days).
> >
> > All I'll add is: Well said, Cameron!
>
> I love Python. I especially love it as an embedded scripting language.
> But it is not going to replace graphics applications where speed is
> paramount. Many of these articles fail to take that into account. But
> then again, many problem domains do not have the interaction constraints
> that 3D or compositing applications have.

*Fast* components (coded in C++) for about (tops!) 20-30% of the
actual codebase, 'scripting' (higher-level, easier-to-use languages)
for MOST of it -- and, I _am_ working specifically in the 3D field...


Alex





From othello at javanet.com  Mon Mar  5 09:49:00 2001
From: othello at javanet.com (Raymond Hettinger)
Date: Mon, 05 Mar 2001 09:49:00 -0500
Subject: Little problem with the "+" operator
References: <983802310.155838463@news.t-online.de>
Message-ID: <3AA3A75C.966D3B01@javanet.com>

Zamurai wrote:

> I want to make the following:
>
> variableOne = 500
> myVariable = variableOne + "cool"
>
> But I get the following error: "unsupported operand types for +"
>
> Does someone know how to solve the problem?

Try:
myVariable = str(variableOne) + "cool"





From grante at visi.com  Sat Mar 17 11:26:30 2001
From: grante at visi.com (Grant Edwards)
Date: Sat, 17 Mar 2001 16:26:30 GMT
Subject: More accurate summations [was Re: This math scares me]
References: <mailman.984726142.2994.python-list@python.org> <98umht$7e0$1@agate.berkeley.edu>
Message-ID: <slrn9b7b6k.10s.grante@tuxtop.visi.com>

On 17 Mar 2001 03:43:57 GMT, Edward Jason Riedy <ejr at cs.berkeley.edu> wrote:
>And <jurgen.defurne at philips.com> writes:
> - 
> - Try summing all financial operations of one day together, or
> - better, all operations involving percentages and then adding them
> - together : cumulative errors.
>
>More concretely, look at the Vancouver Stock Exchange index's
>problems of the late 70s-early 80s.  They had chosen to
>truncate floating point values rather than round them, causing
>a downward drift of around 20 points a month[1].
>
>You could say that's because of poor handling of floating-point
>arithmetic, and you'd be right.  However, just about every
>inaccuracy in floating-point financial calculations can be
>attributed to poor handling of fp arithmetic.

The problem is that proper handling of fp arithmetic is not
easy even though it appears trivial to the uneducated. It
requires quite a bit of knowlege, thought, and work. After one
undergrad numerical analysis class, I know that I don't know
enough to do it right.

>That doesn't mean floating point is a poor choice for all
>financial programs.  Indeed, the HP financial calculators _all_
>use floating point, although decimal.

That was (I think) the original topic: BCD floating point vs.
binary floating point.

-- 
Grant Edwards                   grante             Yow!  Now KEN is having
                                  at               a MENTAL CRISIS beacuse
                               visi.com            his "R.V." PAYMENTS are
                                                   OVER-DUE!!


From dsh8290 at rit.edu  Fri Mar 23 17:22:02 2001
From: dsh8290 at rit.edu (D-Man)
Date: Fri, 23 Mar 2001 17:22:02 -0500
Subject: Jython is cool?
In-Reply-To: <20010322103954.A18539@sync.nyct.net>; from mbac@nyct.net on Thu, Mar 22, 2001 at 10:39:54AM -0500
References: <20010322103954.A18539@sync.nyct.net>
Message-ID: <20010323172202.D12082@harmony.cs.rit.edu>

On Thu, Mar 22, 2001 at 10:39:54AM -0500, Michael Bacarella wrote:
| 
| 	When I first read about Jython, what immediately popped
| into my head was that you get a Python interpreter written in Java
| that you can say, embedd in a web page and pass a parameter which says
| which python script to execute.
| 

| 	I find myself frustrated by Java which is why the thought of
| doing web applets in Python was so exciting. Unfortunately, it seems
| that in downloading Jython, a JDK is still required.

The jdk requirement isn't a problem.  Any of the web browsers you
would target with an applet has a jdk (actually, jre -- just the
runtime and libraries, no compiler) already in it.  Use the jythonc
script to build a jar file from your Python sources and embed it in a
web page just like you would a java applet.

-D




From gregj at pobox.com  Wed Mar 14 23:52:47 2001
From: gregj at pobox.com (Greg Jorgensen)
Date: Thu, 15 Mar 2001 04:52:47 GMT
Subject: tiny extension crashes
References: <Pine.SOL.4.30.0103141957240.10526-100000@mimosa.csv.warwick.ac.uk>
Message-ID: <zUXr6.571195$U46.16976599@news1.sttls1.wa.home.com>

You have an uninitialized pointer, intp, in your iround() function. Do this:

double intp;
...
if (fabs(modf(number, &intp)) >= 0.5) {
        intp >= 0 ? (intp)++ : (intp)--;
...
return Py_BuildValue("l", (long)intp);

But you don't need to do so much work with the floating-point functions to
round a number. Just add +/- 0.5 and truncate. Converting a float or double
to an int or long will truncate the fractional portion.

double number;
long intp;

if (PyArg_ParseTuple(args, "d", &number)) {
    intp = (number > 0.0) ? (number + 0.5) : (number - 0.5);
    return Py_BuildValue("l", intp);
}
else
    return NULL;


Since you posted your source code, let me suggest an alternative
implementation of array_max:

static long array_max(PyArrayObject *array, int absolute)
{
    long *a = array->data;
    long stride = array->strides[0];
    long max = 0;

    if (absolute) {
        for (i = 0; i < array->dimensions[0]; i++) {
            if (labs(*a) > max)
                max = labs(*a);
            a += stride;
        }
    }
    else {
        for (i = 0; i < array->dimensions[0]; i++) {
            if (*a > max)
                max = *a;
            a += stride;
        }
    }
    return max;
}

The test on absolute only happens once, not at every loop iteration. (I
would make this two functions, one that works on the actual values and one
that works on the absolute values.) The array->data[] index calculation is
changed to pointer addition. And I used the labs() function rather than the
unnecessary floating-point fabs() version. And I changed max to a long, and
made the function return a long.

--
Greg Jorgensen
PDXperts
Portland, Oregon, USA
gregj at pobox.com




From danielk at aracnet.com  Sat Mar  3 13:53:31 2001
From: danielk at aracnet.com (Daniel Klein)
Date: Sat, 03 Mar 2001 10:53:31 -0800
Subject: Missing the functional mind set
References: <mlb2at8ptrr7sh5dgq8qlk2201kquh6tfl@4ax.com> <uDao6.335972$ge4.119457077@news2.rdc2.tx.home.com>
Message-ID: <t6f2at870pk82q96bnd5a2od4n8ns5kf83@4ax.com>

On Sat, 03 Mar 2001 18:22:18 GMT, "Rainer Deyke" <root at rainerdeyke.com> wrote:

>But if you are indexing with integers from 0 to n, why are you using a
>dictionary in the first place?

I'm not sure I understand your question. I need the dictionary to associate a
numbered key with a value. Rather than force the user to enter elements 'in
order' I am letting them specify what the order is on-the-fly, ie enter element
3 first, then 1, then 2, then 0. The method will the place the values in the
proper order for further processing.

Is there a better approach?

Daniel Klein



From donn at oz.net  Fri Mar  2 02:08:09 2001
From: donn at oz.net (Donn Cave)
Date: 2 Mar 2001 07:08:09 GMT
Subject: d = {}; d[0:1] = 1; d[0:1] = 2; print d[0:1]
References: <mailman.983419142.24027.python-list@python.org> <mailman.983429941.26766.python-list@python.org>
Message-ID: <97ngsp$i4h$0@216.39.151.169>

Quoth Robin Thomas <robin.thomas at starmedia.net>:
...
| PySequence_SetSlice, strangely, has a big chunk of code expressly written 
| to support slice assignment to mapping objects. From Objects/abstract.c 
| below. If we remove that code, we get the same effect.
...
|          /* the evil begins */
|          } else if ((mp = s->ob_type->tp_as_mapping) && mp->mp_ass_subscript) {
|                  int res;
|                  PyObject *slice = sliceobj_from_intint(i1, i2);
|                  if (!slice)
|                          return -1;
|                  res = mp->mp_ass_subscript(s, slice, o);
|                  Py_DECREF(slice);
|                  return res;
|          }
...
| Anybody have an explanation why this behavior was desired?

So far, I haven't seen any.  The class instance that is supposed
to benefit from all this already has tp_as_sequence, so it doesn't
reach that branch.

I think you ought to submit that as a patch, i.e., delete that
whole block.

	Donn Cave, donn at oz.net


From timr at probo.com  Fri Mar  2 02:33:59 2001
From: timr at probo.com (Tim Roberts)
Date: Thu, 01 Mar 2001 23:33:59 -0800
Subject: gah! I hate the new string syntax
References: <mailman.983485923.1136.python-list@python.org>
Message-ID: <71ju9to3jlp6l6u1t2peihlc7gegku3a99@4ax.com>

"Sean 'Shaleh' Perry" <shaleh at valinux.com> wrote:

>return ";".join(["%s=%s" % (k, params[k]) for k in params.keys()])
>
>every day python seems to be moving closer to the line noise aspect of
>coding.

But join is really the only one of the string methods that comes out so
awkwardly.  The rest of them are quite sensible.  And, at least for now,
you do still have the traditional string.join alternative.

Wouldn't your code be slightly quicker if you used items() instead of
looking up the individual keys:

return ";".join(["%s=%s" % it for it in params.items()])
--
- Tim Roberts, timr at probo.com
  Providenza & Boekelheide, Inc.


From tim.one at home.com  Wed Mar 21 17:41:12 2001
From: tim.one at home.com (Tim Peters)
Date: Wed, 21 Mar 2001 17:41:12 -0500
Subject: I'm sure there's a good reason!
In-Reply-To: <memo.20010321220225.57471D@digitig.compulink.co.uk>
Message-ID: <LNBBLJKPBEHFEDALKOLCAEFMJHAA.tim.one@home.com>

[Tim Rowe]
> Can any of the assembled experts explain to me why:
> 	print r"\hello"
> is fine, but
> 	print r"hello\"
> isn't? In other words, why does the backslash escape quotes in a raw
> string? At one level I'm going to guess (without reading the spec) that
> it's because "that's what the spec says", but why? I thought the idea of
> raw strings was that they /didn't/ handle escapes!

The idea of raw strings was to make it easier to create input for systems
that want to do their own backslash escape processing (you can think of the
"r" as meaning "regexp" instead ...).  There's no reason to believe such
systems don't need string-quote characters, but much reason to believe
they'll consider an odd number of trailing backslashes to be an error.  So
r-strings were designed to let you pass escaped quote characters easily, at
the cost of forbidding an odd number of trailing backslashes.

See the FAQ for more info; there are many ways to get an odd number of
trailing backslashes on a string.

if-one-string-gimmick-were-enough-for-all-purposes-there-wouldn't-
    be-two<wink>-ly y'rs  - tim




From aleaxit at yahoo.com  Thu Mar 29 07:45:16 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 29 Mar 2001 14:45:16 +0200
Subject: Is it possible to sort a dictionary?
References: <GEtv6.59583$Lm2.6378825@news0.telusplanet.net> <3ABE771B.7B9E1631@student.liu.se> <99o33k$aq4$0@216.39.170.247> <99o51301qgb@news1.newsguy.com> <mailman.985660594.13367.python-list@python.org> <99q70i$1le$1@animus.fel.iae.nl>
Message-ID: <99vap202a1b@news2.newsguy.com>

"Carel Fellinger" <cfelling at iae.nl> wrote in message
news:99q70i$1le$1 at animus.fel.iae.nl...
    [snip]
> I just wonder why the items() method is slooow; values() and keys()
> are both much faster, going over the whole dict indexing from keys()
> is even faster.  Is tuple making/unpacking so expensive?  Or is a dict

I can't reproduce this observation... here's a datapoint:

D:\>python b.py
sortWithItems: 0.0438600314118
List with 2969 items:
  ('\012\012 905', 905)
  ('\012\012 906', 906)
  ('\012\012 907', 907)
  ...
  ('}\012 697', 697)
  ('}\012 698', 698)
  ('}\012 699', 699)
sortWithKeys: 0.0460969072614
List with 2969 items:
  ('\012\012 905', 905)
  ('\012\012 906', 906)
  ('\012\012 907', 907)
  ...
  ('}\012 697', 697)
  ('}\012 698', 698)
  ('}\012 699', 699)

i.e., items() seems a bit _faster_ than keys() then
going over the dict -- with the script b.py being:

import time,sys,pprint

def sortWithItems(adict):
    result = adict.items()
    result.sort()
    return result

def sortWithKeys(adict):
    keys = adict.keys()
    keys.sort()
    return [ (key, adict[key]) for key in keys ]

def timeit(afunc, *someargs):
    start = time.clock()
    result = afunc(*someargs)
    stend = time.clock()
    print "%s: %s" % (afunc.__name__, stend-start)
    return result

def plist(alist):
    print "List with %d items:"%len(alist)
    if len(alist)<=7:
        for item in alist: print ' ',item
    else:
        for item in alist[:3]: print ' ',item
        print '  ...'
        for item in alist[-3:]: print ' ',item

if __name__=='__main__':
    adict = {}
    sometext=open(sys.argv[0]).read()
    for i in range(1,len(sometext)-1):
        adict[sometext[i-1:i+2]+str(i)] = i
        adict[sometext[i:i+3]+str(i)] = i
        adict[sometext[i+1:i+4]+str(i)] = i

    result = timeit(sortWithItems, adict)
    plist(result)

    result = timeit(sortWithKeys, adict)
    plist(result)


Alex





From hamish_lawson at yahoo.co.uk  Fri Mar  9 12:53:17 2001
From: hamish_lawson at yahoo.co.uk (=?iso-8859-1?q?Hamish=20Lawson?=)
Date: Fri, 9 Mar 2001 17:53:17 +0000 (GMT)
Subject: Static typing (was Re: Java guy interested in Python)
Message-ID: <20010309175317.66881.qmail@web219.mail.yahoo.com>

Steve Purcell:

    You could indeed consider those an alternative, since a thorough
    test suite would find all the same errors.

Yes, I've tried to convince my static-typing-loving colleagues of the
claim of dynamic typing and test suites: a test suite can catch errors
that static typing can't; the static-typing furniture clutters up the
logic of your code; and the less flexible nature of statically-typed
code often makes the code longer (and hence takes a longer time to
write
and understand) than the dynamically-typed equivalent (even when a test
suite is lumped in with the latter).

Michael Hudson:

    I suspect that you'd basically end up listing the methods required,
    eg:

    def func(file : <has 'write', 'read', 'seek'>, ...): pass

    I'm not sure that makes a "type system".  Hmm.  More thinking
    required.

Not sure that makes a type system? Isn't that what the modern
understanding of a type is - a list of operations that are expected to
be supported, as with Java's idea of interfaces being distinct from
classes? I actually thought your example showed an interesting idea -
anonymous, defined-on-the-fly interfaces. (That is, if static typing is
wanted.)


Hamish Lawson


____________________________________________________________
Do You Yahoo!?
Get your free @yahoo.co.uk address at http://mail.yahoo.co.uk
or your free @yahoo.ie address at http://mail.yahoo.ie



From erno-news at erno.iki.fi  Sun Mar  4 13:45:58 2001
From: erno-news at erno.iki.fi (Erno Kuusela)
Date: 04 Mar 2001 20:45:58 +0200
Subject: Concurrency and shelve
References: <3AA26D5E.2FD28B40@stroeder.com>
Message-ID: <kuk865l6e1.fsf@lasipalatsi.fi>

In article <3AA26D5E.2FD28B40 at stroeder.com>, Michael Str?der
<michael at stroeder.com> writes:

| HI!
| Can anybody summarize issues with shelve (based on anydbm) and
| concurrent access? I experienced with bsddb that changes to the DB
| within one instance are only visible after calling close() to a
| second instance of the same database.

it depends on the dbm implementation that anydbm happens to
pick up. afaik only gdbm does locking, the others break.
if you look in the shelve module, it is easy to see how to
make it use gdbm if you want. the classes there are not
documented however so it may not be safe to rely on them being there.

  -- erno


From kalle at gnupung.net  Mon Mar  5 15:36:18 2001
From: kalle at gnupung.net (Kalle Svensson)
Date: Mon, 5 Mar 2001 21:36:18 +0100
Subject: Coroutines, and memory leaks
In-Reply-To: <8oh089.7it.ln@127.0.0.1>; from dg@pearl.tao.co.uk on Mon, Mar 05, 2001 at 05:18:00PM +0000
References: <8oh089.7it.ln@127.0.0.1>
Message-ID: <20010305213618.A846@apone.network.loc>

Sez David Given:
> 2. I have this massive memory leak. How can I find it?
> 
> The more I use the program, the more memory it uses. I've seen it use
> >100MB before I killed it. I guess that I've got cyclic memory structures
> in the message cache, so it's never freeing any of the mail I look at. Are
> there any suggestions for how I can find these and kill them?

Search on parnassus for Cyclops.  It's a cycle detecting script by Tim
Peters.

Peace,
  Kalle
-- 
Email: kalle at gnupung.net     | You can tune a filesystem, but you
Web: http://www.gnupung.net/ | can't tune a fish. -- man tunefs(8)
PGP fingerprint: 0C56 B171 8159 327F 1824 F5DE 74D7 80D7 BF3B B1DD
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 240 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/20010305/52241912/attachment.sig>

From sholden at holdenweb.com  Tue Mar 13 18:52:01 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Tue, 13 Mar 2001 18:52:01 -0500
Subject: Einstein's Riddle
References: <oNPq6.396644$ge4.138479566@news2.rdc2.tx.home.com> <98glrd$1si1a$1@ID-75892.news.dfncis.de> <3AABE712.499CCF4C@home.com> <98gp5b$1u0oi$1@ID-75892.news.dfncis.de> <98lf4g$2dqj2$1@ID-63952.news.dfncis.de>
Message-ID: <isyr6.68132$1D5.3483144@e420r-atl1.usenetserver.com>

"J. A. Durieux (reply from http://www.biep.org/)" <bogus at ddress.com> wrote
in message news:98lf4g$2dqj2$1 at ID-63952.news.dfncis.de...
> "Boris Popov" <no-spam-popov at bigserver.com> wrote in message
> news:98gp5b$1u0oi$1 at ID-75892.news.dfncis.de...
> > P.S. What's emu ?
>
> A European currency unit, now replaced by the Euro..
>
Bzzzt. EMU is the European Monetary Union, the group of countries which have
collectively agreed to bind their exchange rates to the ECU, which is the
European Currency Unit, commonly referred to as the Euro.

still-european-even-though-living-in-the-usa-ly y'rs  - steve





From gunturu_neelima at rediffmail.com  Tue Mar  6 01:59:20 2001
From: gunturu_neelima at rediffmail.com (neelima  g)
Date: 6 Mar 2001 06:59:20 -0000
Subject: (no subject)
Message-ID: <20010306065920.9744.qmail@mailweb23.rediffmail.com>

sir,

     thank you very much for giving reply.you asked me to give more details.it is text database only.
i have to create a database with fields

userid,from,subject,body,attachment information.

using gdbm module in python.hope you will be giving reply in soon.
                                     thanking you,
                                         neelima g.










_____________________________________________________
Chat with your friends as soon as they come online. Get Rediff Bol at
http://bol.rediff.com







From bobhicks at adelphia.net  Mon Mar 12 08:18:21 2001
From: bobhicks at adelphia.net (Robert L Hicks)
Date: Mon, 12 Mar 2001 13:18:21 GMT
Subject: ZODB and OSX
References: <mailman.984324988.25505.python-list@python.org> <m3wv9w3x6h.fsf@gnu.org> <slrn9aoi0i.rt.amk@mira.erols.com> <m3bsr7h9rt.fsf@gnu.org>
Message-ID: <B6D236BD.2310%bobhicks@adelphia.net>

The current beta says that it will compile on OSX. ZODB would be a great
thing to have as well!

- Bob



From mda at idatar.com  Sat Mar 17 13:38:56 2001
From: mda at idatar.com (David Allen)
Date: Sat, 17 Mar 2001 18:38:56 GMT
Subject: can a class intantiate itself to None?
References: <mailman.984770427.21248.python-list@python.org>
Message-ID: <4bOs6.503$Fb.772856@typhoon2.ba-dsg.net>

In article <mailman.984770427.21248.python-list at python.org>, "Timothy Grant"
<tjg at exceptionalminds.com> wrote:

> I'd like to create a class that can instantiate itself to None when
> instantiation fails...
> 
> class foo:
> 	def __init__(self, bar):
> 		if type(bar) == StringType:
> 				self.data = 'This is a String'
> 				
> 
> x = foo('a')
> 
> y = foo(1)  # In this instance I would like y to be None
> 
> So, can it be done, or do I need to raise an exception, catch it and then
> set y to None?

Raise an exception and set y to None.

You have the option of setting all class data to 
None if something doesn't happen that you want to 
happen, but there isn't a way that you can actually
return None from a constructor as far as I know of.
Besides, that would be confusing as hell, and that's
partially what exceptions are for, so use 'em.

-- 
David Allen
http://opop.nols.com/
----------------------------------------
Be wary of strong drink. It can make you shoot at tax collectors and 
miss 
- Lazarus Long, "Time Enough for Love"


From no-spam-popov at bigserver.com  Mon Mar 12 00:29:25 2001
From: no-spam-popov at bigserver.com (Boris Popov)
Date: Sun, 11 Mar 2001 21:29:25 -0800
Subject: Einstein's Riddle
References: <oNPq6.396644$ge4.138479566@news2.rdc2.tx.home.com> <98glrd$1si1a$1@ID-75892.news.dfncis.de> <98h62k$knn$1@abbenay.CS.Berkeley.EDU>
Message-ID: <98hmsv$216kg$1@ID-75892.news.dfncis.de>

"Brian Harvey" <bh at abbenay.cs.berkeley.edu> wrote in message
news:98h62k$knn$1 at abbenay.CS.Berkeley.EDU...
> "Boris Popov" <no-spam-popov at bigserver.com> writes:
> >house         1 2 3 4 5
> >color            yellow blue red green white
> >nationality   norweigian dane brit german swede
> >drink            water tea milk coffee beer
> >smoke        dunhill blend pall mall prince bluemaster
> >pet              cats horse birds fish dogs
> >
> >P.S. Did I have to write software to solve it ? :)
>
> I can't resist being (apparently) first to post a computer program.
> But it doesn't really belong in any of these newsgroups, because it's
> written in Logo.

Khm... I thought that software is impossible to write for this particular
problem. The thing is that there's no fifth pet element (fish) explicitly
defined in the puzzle. And it will definetely blow up if you run it with no
"fish"... :)

-Boris




From tim.one at home.com  Sat Mar 10 15:33:41 2001
From: tim.one at home.com (Tim Peters)
Date: Sat, 10 Mar 2001 15:33:41 -0500
Subject: Can anyone recomend a good intoduction to C...
In-Reply-To: <1eq0a7k.1yanuaitkzmn4N%francois.granger@free.fr>
Message-ID: <LNBBLJKPBEHFEDALKOLCMELOJEAA.tim.one@home.com>

[Werner Schiendl]
> What's TMTOWTDI, please?

[Fran?ois Granger]
> There is More Than One Way To Do It
>
> which is a Perl motto as far as I know.

Right!  My automobile license plate reads

    TOOWTDI

Honest.  I will know that my work on Python is finally complete when the
average American looks at my license plate and nods in silent agreement.  I'm
not there yet.  So far, approximately 100 out of 100 look at it, ponder, then
start making noises like some demented 170-pound bird with a broken beak.

perhaps-i'd-have-better-luck-if-i-moved-to-california-ly y'rs  - tim




From Patrick.Bussi at space.alcatel.fr  Tue Mar 27 10:32:20 2001
From: Patrick.Bussi at space.alcatel.fr (Patrick.Bussi at space.alcatel.fr)
Date: Tue, 27 Mar 2001 17:32:20 +0200
Subject: Newbie question : string processing with RE
Message-ID: <C1256A1C.00555479.00@vzmta01.netfr.alcatel.fr>

Hello,

I have difficulty with the search method of the re module, when applied to an
object retrieved from urllib.In the following program, f3 is read once and the
search is effective. But when I try to read it again, it acts as if the index of
a file had to be "rewinded" (e.g. as would do a "seek" on a "true" file). I
thought that f3.read was to be considered as a string. Could someone help me
understand my mistake and give me a nice solution (means not storing data into a
file).

Thank you in advance.


 #! /usr/bin/env python
 # file test 22
 import re, socket, urllib
 encoded_data = ... # bla-bla-bla
 f3 = urllib.urlopen ("www.such-a-nice-webpage.com/index.html", encoded_data)
 p=re.compile("window.open")
 q=re.compile("document.location =" )
 print p,'\n',q
 t1=p.search(f3.read())
 t2=q.search(f3.read())
 print t1.group()
 print t2.group()
 # ---- end of file test22 ----


 $ test22
 <re.RegexObject instance at 8135458>
 <re.RegexObject instance at 81354e0>
 window.open
 Traceback (innermost last):
   File "./test22", line 12, in ?
     print t2.group()
 AttributeError: 'None' object has no attribute 'group'


---
Patrick Bussi
patrick.bussi at space.alcatel.fr


Any opinions expressed are my own and not necessarily those of my Company.





From db3l at fitlinxx.com  Mon Mar 26 21:55:01 2001
From: db3l at fitlinxx.com (David Bolen)
Date: 26 Mar 2001 21:55:01 -0500
Subject: WinNT problem with Python and Cygwin Bash
References: <mailman.985632095.6515.python-list@python.org>
Message-ID: <uwv9b53ay.fsf@ctwd0143.fitlinxx.com>

Don Dwiggins <dwig at advancedmp.net> writes:

> I've just installed Happydoc (nice!) on my WinNT box, and am having a bit of
> a problem running the shell script generated by the installer under Cygwin
> using Bash.  It looks as though Bash is passing Python a Unix-style pathname
> (/e/Python20/Scripts/happydoc), but Python expects a Windows-style pathname,
> so complains it can't find the file.  I can work around it by applying some
> "Bash grease", but I'd like to know if anyone else has run across this, and
> if there's a simple fix.

In general, Python (or any program using the Win32 API) can handle
forward slashes or backslashes in paths without any problem.

However, am I correct in guessing that "/e" isn't actually a top level
directory on your current drive, but rather a Cygwin mount point for
your E: drive?  If that's the case, then you'll definitely have a
problem because there's no way for Python (unless you've built a
Cygwin version of it) to handle that mount translation, as such
translation is built into the Cygwin DLL.

Alternatively, if /e really is a directory off of your Cygwin root,
but your Cygwin root drive is not the same drive you are "on" when
running the script, again, Windows won't like it since it will
interpret the leading "/" as being the root directory of the _current_
drive.

In either case, you can use the cygpath utility to convert any Cygwin
path into a Windows path (or vice versa, or other conversions), and it
will handle replacing any mount references with appropriate Windows
native paths and adding drive references as necessary.

So if you modify the created shell script slightly to take whatever
pathname it would normally send into Python and run it through
"cygpath -w" first, you should be ok.  Or, just create a small
front-end shell script to run instead of Python which can do the
expansion and then call Python, and tell the Happydoc created script
to run that instead of Python.

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


From thomas at cintra.no  Fri Mar 23 07:59:24 2001
From: thomas at cintra.no (Thomas Weholt)
Date: Fri, 23 Mar 2001 13:59:24 +0100
Subject: Slightly Off-Topic: Python featured in Severance: Blade of Darkness
Message-ID: <NMHu6.144$NR.1628@news3.oke.nextra.no>

Hi,

A friend of mine just informed me that while starting up his copy of the
game Blade of Darkness he saw a reference to Python 1.5.2 during the loading
process. Anybody know anything about that? What has Python been used for?
Writing mods/extensions in Python would be cool in any game. :->

Wouldn't Python fit just perfectly for this purpose?

Best regards,
Thomas Weholt




From chris.gonnerman at usa.net  Fri Mar 23 00:29:30 2001
From: chris.gonnerman at usa.net (Chris Gonnerman)
Date: Thu, 22 Mar 2001 23:29:30 -0600
Subject: IPC module
References: <984438966.381139@basic.maxnet.ru>
Message-ID: <00f001c0b35a$40c8bda0$a100000a@local>

Hey, it might help if you post the ERROR MESSAGES.

If you are using the standard make command, do this:

    make >makelog 2>&1

and then attach or cut-and-paste the makelog file to your email/posting.

----- Original Message -----
From: "Alexei Zalivine" <zalivin at krona.obninsk.ru>
Subject: IPC module


> Hello,
> Please help me to compile shm module for linux. I'm not a C programmer,
and
> I can't understand errors from compiler. You can get the source from
> http://sirac.inrialpes.fr/~marangoz/python/shm/
>
> Are there other IPC modules for Python?
>
> Thank you,
> Alexei.
>
>
> --
> http://mail.python.org/mailman/listinfo/python-list
>





From aleaxit at yahoo.com  Fri Mar 16 16:56:30 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 16 Mar 2001 22:56:30 +0100
Subject: Python style questions
References: <98tb91$9t9$1@saltmine.radix.net> <98tdu1$cfk$1@panix2.panix.com> <1eqdo5u.1xxmzfi2kjk0qN%francois.granger@free.fr>
Message-ID: <98u24p0a98@news2.newsguy.com>

"Fran?ois Granger" <francois.granger at free.fr> wrote in message
news:1eqdo5u.1xxmzfi2kjk0qN%francois.granger at free.fr...
> Aahz Maruch <aahz at panix.com> wrote:
>
> > In article <98tb91$9t9$1 at saltmine.radix.net>,
> > Cary O'Brien <cobrien at Radix.Net> wrote:
>
> > >3. I really miss not having a "switch" or "case" statement.  Sniff.
>
> > Use a dictionary.
>
> Some code snippet as exemple ?
> At this point, I can't imagine a good way, and I don't remember seeing
> something on this.

There are several ways you can use a dictionary to dispatch pieces
of code; any callable (function or otherwise) can be the value that
corresponds to a given key.  Somewhat less obvious, and thus less
advisable, is to use compiled fragments (and exec to run them) as
said values.

To give one toy example, suppose there is some current code you
have that is:

def toy1(fileob, operation, *data):
    if operation=='leggi': return fileob.read(*data)
    elif operation=='scrivi': return fileob.write(*data)
    elif operation=='chiudi': return fileob.close(*data)
    else: raise ValueError, operation

and you're pining to be able to code this as a switch on the
operation rather an an if/elif tree.  OK, easy:

def toy2(fileob, operation, *data):
    switch = {'leggi': fileob.read, 'scrivi': fileob.write,
        'chiudi': fileob.close }
    return switch[operation](*data)

(it raises a KeyError rather than ValueError if it fails, but
if that's important you can catch this and raise whatever),
or, a syntax-sugar alternative:

def toy3(fileob, operation, *data):
    return {'leggi': fileob.read,
        'scrivi': fileob.write,
        'chiudi': fileob.close }[operation](*data)

although I far prefer the toy2 version.


If you give us a specific, similarly-small example that
you dream to be able to code with a switch (and are
presumably now coding with if/elif), we can help you
translate this into some form of dispatching...


Alex





From oliphant at ee.byu.edu  Mon Mar 19 00:11:22 2001
From: oliphant at ee.byu.edu (Travis Oliphant)
Date: Sun, 18 Mar 2001 22:11:22 -0700
Subject: Python In A Nutshell - Numpy essential
In-Reply-To: <3AAE5DA3.B736C2D0@northwestern.edu>
References: <3AAE7E0D@MailAndNews.com> <slrn9aolph.ccf.jimhill@meldrick.swcp.com>
 <3AACE20C.AC8D2EA2@northwestern.edu> <Ikkr6.68807$ML1.3285992@e420r-atl2.usenetserver.com>
 <3AAE5DA3.B736C2D0@northwestern.edu>
Message-ID: <Pine.HPX.4.33.0103182208511.10898-100000@sesame.ee.byu.edu>

I used to be one who really wanted Numeric in the Python core.  At this
point it is not critical to me.  Maybe someday, the multidimensional array
aspect of Numeric will be distributed with Python.  For now, though what
is needed instead is an easy to install package (which includes Numeric)
and provides a competitive numerical enviroment (something like SciPy
which is coming this summer...)

I'm happy to keep Numeric out of the core as it gives the math-heads more
control over it's optimization.

-Travis O.





From Mike_B at T-Online.de  Tue Mar 27 21:22:14 2001
From: Mike_B at T-Online.de (Michael Bauer)
Date: Wed, 28 Mar 2001 04:22:14 +0200
Subject: daemon app in win32
References: <99nk49$a2q$02$1@news.t-online.com> <mailman.985619314.10299.python-list@python.org> <99nsn3$q39$06$1@news.t-online.com> <99o2ge$9on$0@216.39.170.247> <uu24ebw63.fsf@roundpoint.com>
Message-ID: <99rhsp$i4$06$1@news.t-online.com>

Ben Hutchings wrote:

> whisper at oz.net (Dave LeBlanc) writes:
> 
>> On Mon, 26 Mar 2001 19:02:27 +0200, Michael Bauer <Mike_B at T-Online.de>
>> wrote:
>> 
>> >Hmm, seems like this is only available under NT. Is there a more general
>> >way to accomplish this task independently of the underlying
>> >Windows-version?
>> >
>> >But maybe i should just get a life and write different wrappers for
>> >Win95, Win98, WinME, Win200, WinNT... :-(
> 
> You should only need two.  The Apache web server code for Windows
> might be instructive.

Okay i will give it a close look. 
 
> Console applications can register to receive such notifications in the
> same way they receive control-C notifications under *some* versions of
> Windows.

Do you have any more details about this? Because i experienced that my app 
does actually *not* receive the CTRL-Cs or similar i am pressing when i try 
to cancel it "by hand"... (btw, this is win98)
If this would work, i would be happy, i really dont need more than an 
exception i could catch ;-) 

I already spent some time reading the documentation of Marc Hammonds win32, 
but it seems to me that it is clearly aimed at experienced windows 
programmers.

sigh, and i thougt i could do it really OS-independent :-(

thanks, Mike


From alphatrooper at skynet.be  Fri Mar 30 06:14:10 2001
From: alphatrooper at skynet.be (alphatrooper)
Date: Fri, 30 Mar 2001 13:14:10 +0200
Subject: Fetch object in imaplib
Message-ID: <3ac47845$0$30795$456d72a3@news.skynet.be>

hello ,
I'm a bit stuck with the fecth object from the iamplib .
in the lib ref they mention :

       Fetch (parts of) messages. message_parts should be a string of
message part names enclosed within parentheses, eg: ""(UID
       BODY[TEXT])"". Returned data are tuples of message part envelope
and data.

And in the examples they use :
M.fetch(num, '(RFC822)')

Now my question is : what are all the possibilities of "message part
names", I looked in the rfc's and on the net but can't find anything
about it.
What can I put in place of : (UID BODY[TEXT])
Or if you could send me a pointer where to find that info.
Thanks
Dimitri Haentjens


From max at alcyone.com  Thu Mar 29 11:02:12 2001
From: max at alcyone.com (Erik Max Francis)
Date: Thu, 29 Mar 2001 08:02:12 -0800
Subject: Quoted identifiers in Python?
References: <3ac3387f$0$12248$cc9e4d1f@news.dial.pipex.com>
Message-ID: <3AC35C84.FBED128@alcyone.com>

Eddie and Babs wrote:

> NOW... would this be a good thing for Python to have in the future? It
> would
> make element access consistent with subscripting:-
> 
>         x = obj."Ivor the engine"
> 
>             ...would be the same as
> 
>         x = obj["Ivor the engine"]
> 
> ..no?

I would hope they would not be the same, since one invokes __getattr__
while the other invokes __getitem__.  They're different things, they
shouldn't be forced to be the same thing.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ Traveler, there are no paths.  Paths are made by walking.
\__/ Antonio Machado
    REALpolitik / http://www.realpolitik.com/
 Get your own customized newsfeed online in realtime ... for free!


From scarblac at pino.selwerd.nl  Mon Mar 12 17:56:47 2001
From: scarblac at pino.selwerd.nl (Remco Gerlich)
Date: 12 Mar 2001 22:56:47 GMT
Subject: This math scares me
References: <3aad4062.1290021@News.CIS.DFN.DE>
Message-ID: <slrn9aqkth.aki.scarblac@pino.selwerd.nl>

costas at springmail.com <costas at springmail.com> wrote in comp.lang.python:
> Ok, I can see maybe division having problems. But why does addition of
> the two numbers below:
> 
> 5.01+5.54
> 
> give me this?
> 
> 10.550000000000001
> 
> I hope its some setting that can take care of this. I am using Python
> under Windows 98.

It's worse than that. Try:

>>> 5.01
5.0099999999999998

The number is stored in binary. It's not possible to represent everything in
a finite number of binary digits (just like you can't represent 1/3 in a
finite number of decimal digits - you don't protest either when you see
0.3333333333333333, do you? It's just as wrong).

This is not a Python problem, this is due to the way computers store
floating point numbers.

If you can avoid them, don't use floats. If you're working with money, count
cents instead (and use longs if you have numbers over a few hundred million
cents).

-- 
Remco Gerlich


From aleaxit at yahoo.com  Fri Mar  9 10:05:08 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 9 Mar 2001 16:05:08 +0100
Subject: COM support (was Re: Can Python replace Visual Basic? Should it?)
References: <5DZo6.11126$hn5.1578410@news1.rdc1.mb.home.com> <mailman.984010931.26257.python-list@python.org> <iSBp6.14005$hn5.2035325@news1.rdc1.mb.home.com> <mailman.984094514.29514.python-list@python.org> <984133550.452523@newsmaster-04.atnet.at> <98ai2j01o8j@news2.newsguy.com> <984145785.263773@newsmaster-04.atnet.at> <fU5q6.45262$1D5.2046670@e420r-atl1.usenetserver.com>
Message-ID: <98arf0022o4@news2.newsguy.com>

"Steve Holden" <sholden at holdenweb.com> wrote in message
news:fU5q6.45262$1D5.2046670 at e420r-atl1.usenetserver.com...
    [snip -- sorry but I didn't see the post Steve is
     responding to, so I'm taking advantage to reply...]
> > interface classes and things like that and this is one point Visual
Basic
> > handles nicely.

...until you smack into something really hard-to-handle, and then you're
on a rudderless raft in the middle of the ocean.  Which doesn't happen
with C++... or Python.

> > ActiveX control containment is also one thing I would have
> > no idea on how to start.

Instantiate a WebBrowser (an instance of the Internet Explorer
core functionality, minus toolbars, menus, etc), and that will
easily supply all the active-document (and active-control)
containment functionality you might ever possibly desire, _and
then some_.  A tiny amount of HTML and a modest amount of COM
let you do WONDERS with WebBrowser and friends.


>  COM is much more  fun in Python than in VB, until you stub your toe
against
> somethig that really does require an understanding of the underlying
> implementation.

...which happens more rarely in VB, but DOES happen (been there,
done that, got the T-shirt), and when it does you ARE in a truly
desperate situation (which is not the case when you have sources
to your tools -- the situation can then be serious, but never
truly hopeless).  As I mentioned, this is the main thing I'm
currently doing to earn the daily caviar & champagne -- playing
guru on all COM issues (and a few others, such as C++, VB, &c)
in support of a few hundred application developers -- so, of course,
I get _mostly_ exposed to the very difficult problem cases, which
may distort my perception but gives me ample proof of existence
of such cases:-).


> > P.S.: Of course one should (or must, to bring it to the point)
understand
> > the technology in use. This applies heavily to COM, and I know quite a
lot
> > about that stuff. This is one of the reasons I think that well designed
> > interfaces are more important for remoting than the actual programming
> > language. And you can use IDL defined Interfaces with VB, otherwise it
> > would be a headache some time...

I much prefer IDL-described interfaces, too, no matter how they're
going to be implemented and consumed; I consider it a defect of .NET
that it includes no official interface-description language, but
rather appears to assume each programming language will be able to
fully and transparently map any existing or being-designed interface
into its own terms -- a seriously-wrong architecture decision IMHO.


Alex





From mjackson at wc.eso.mc.xerox.com  Thu Mar  1 14:50:31 2001
From: mjackson at wc.eso.mc.xerox.com (Mark Jackson)
Date: 1 Mar 2001 19:50:31 GMT
Subject: [Q] Question from Python tutorial
References: <hxxn6.6969$dL4.91744@vixen.cso.uiuc.edu>
Message-ID: <97m967$lq8$1@news.wrc.xerox.com>

"Young-Jin Lee" <ylee12 at uiuc.edu> writes:
> "Jay O'Connor" <joconnor at cybermesa.com> wrote in message
> news:3A9EA472.F15AF847 at cybermesa.com...
> >
> >
> > Young-Jin Lee wrote:

> > > x.counter = 1
> > > while x.counter < 10:
> > >     x.counter = x.counter * 2
> > > print x.counter
> > >       ^
> > > SyntaxError: invalid syntax

> > You probably cut and pasted the code from the tutorial.  When I cut and

> I typed them manually. I dind't copy and paste.

In the interpreter you need an empty line to signal the end of an
indented code block.  (Code read from a file is scanned completely
before execution, so this hint isn't necessary there).  I think you'll
find that

while x.counter < 10:
    x.counter = x.counter * 2

print x.counter

works just fine. . .

-- 
Mark Jackson - http://www.alumni.caltech.edu/~mjackson
    They are the sort of people who think that no observation is
    so intuitive that it can't be improved by regression analysis.
				- Louis Menand




From brian.j.geddes at intel.com  Fri Mar  2 17:46:05 2001
From: brian.j.geddes at intel.com (Brian Geddes)
Date: Fri, 2 Mar 2001 14:46:05 -0800
Subject: Can I capture console output from an os.system() call?
Message-ID: <97p7nt$a2s@news.or.intel.com>

All -

When I call a foreign program using an os.system("program.exe") call, is
possible to get at the console output of "program.exe" from my python
script?

Just in case I'm not being entirely clear: (I'm pretty brain-fried at the
moment!)

If "program.exe" spits out a message "Hello World!" to the screen when run
from the DOS command line, and I call 'os.system("program.exe")' as a line
of my Python script, how do I get at the "Hello World!" message?

Thanks,
-Brian




From lee at joramo.com  Tue Mar 13 09:09:52 2001
From: lee at joramo.com (Lee Joramo)
Date: Tue, 13 Mar 2001 07:09:52 -0700
Subject: Converting a text data file from positional to tab delimited.
Message-ID: <B6D37840.2124%lee@joramo.com>

I am looking for suggestions to speed up the process of converting a large
text data file from 'positional' layout to tab delimited. The data file is
over 200MB in size containing over 40,000 lines which have over 600 fields.

I suspect that the 'for' loop that splits each line into tab delimited,
could be optimized. Perhaps it could be replaced with a regex or other
technique.

Thanks for any suggestions.

Lee Joramo

==================

#'layout' list elements [field name, start position, end position]
#For brievty, I have only included 10 fields.
#The contents of 'layout' are extracted from a text file that
#describes the layout of the datafile.
layout = [
    ['STUDY', 0, 7]
    ['MDLNO', 8, 12]
    ['DASH', 13, 16]
    ['INCENT', 17, 17]
    ['CODE1', 18, 18]
    ['CODE2', 19, 19]
    ['COVLET', 20, 20]
    ['VERSN', 21, 21]
    ['MAILNO', 22, 23]
    ['MLDYY', 24, 27]]

inFile = open('rawdata.dat', 'r')
outFile = open('delimted.dat', 'w')
while 1:
    lines = inFile.readlines(1000)
    if not lines: break
    for line in lines:
        delimitedLine = ""
        delimit = ""
        for field in layout:
            #
            #can this loop be improved??
            #
            fieldValue = line[field[1]:field[2]]
            delimitedLine = delimitedLine + delimit + fieldValue
            delimit = "\t"
        outFile.write(delimitedLine+"\n")
inFile.close()
outFile.close()
del lines



From ralf.claus at teleatlas.com  Thu Mar 29 05:47:38 2001
From: ralf.claus at teleatlas.com (Ralf Claus)
Date: Thu, 29 Mar 2001 12:47:38 +0200
Subject: how to write an error message in a file....
References: <3AC2FFC2.A60F886E@teleatlas.com>
Message-ID: <3AC312CA.7CC7ACD2@teleatlas.com>

I am really beginner :-)

sys.stderr = open('file','a')

operates perfectly !

ciao

Ralf Claus wrote:
> 
> ....befor the programm quit by an error ?
> 
> I would like to store the different errors in a file!?
> Is there a possibility ?
> 
> cu
> 
> Ralf


From ransen_spam_me_not at nemo.it  Thu Mar  1 10:03:58 2001
From: ransen_spam_me_not at nemo.it (Owen F. Ransen)
Date: Thu, 01 Mar 2001 15:03:58 GMT
Subject: Py_INCREF / DECREF not needed in simple C functions?
Message-ID: <3a9e5301.1251797@news.newsguy.com>

Trundling along here and reading the docs I've come
to the conclusion that if I pass simple values to 
my C extensions to Python I do not need to worry
about Py_INCREF and Py_DECREF.

Am I right? For example if I only pass, say, 3
double values to a C extension function, the values
of which I extract using PyArgs_ParseTuple, then
I do not need to worry about reference counts?

Concrete example:

static PyObject* Py_ReallyWildFunc (PyObject *self, PyObject *args)
{
    double x,y,z ;
    if (!PyArg_ParseTuple(args, "ddd",&x,&y,&z)) {
        return (NULL) ;
    }

    ...do some really wild stuff with x,y,z...

    return (Py_None) ;
}

(I know that in the future I may do more complex things,
 but at the moment I'm concentraing on this simple stuff)
--
Owen F. Ransen
http://www.ransen.com/
Home of Gliftic & Repligator Image Generators


From rol9999 at attglobal.net  Thu Mar 29 15:20:14 2001
From: rol9999 at attglobal.net (Roland Schlenker)
Date: Thu, 29 Mar 2001 15:20:14 -0500
Subject: Looking for Python equivalent to a Perl'ism
References: <3AC375A7.A4BA87F1@spammraqia.com>
Message-ID: <3AC398FE.D04351B3@attglobal.net>

David Lees wrote:
> 
> I am trying to figure out how to do the Python equivalent of the
> following perl code, which finds all the groups tagged by 'xx' and
> replaces them with text looked up in a dictionary:
> ----------
> my %dict;
> my $string="this is xxfooxx this is junk xxmumblexx and more stuff";
> %dict=(foo=>"yechfoo",mumble=>"yukkomumble");
> 
> $string =~ s/xx(\S+)xx/$dict{$1}/g;
> 
> print $string;
> ---------------------
> 
> My Python equivalent without the backreference dictionary lookup is:
> 
> dict ={'xxx':'foo',
>        'yyy':'barf',
>        }
> p=re.compile(r'zz(\w+)zz')
> s='this is and so is zzxxxzz abc that an zzyyyzz abc zzxxxzz or else'
> print re.sub(p,'junk',s)
> 
How about a non re solution.

import string
s = string.split(s, "zz")
for r in range(len(s)):
    try:
        s[r] = dict[s[r]]
    except KeyError:
        pass
s = string.join(s, "")

Roland Schlenker


From sbrunning at trisystems.co.uk  Wed Mar  7 07:36:52 2001
From: sbrunning at trisystems.co.uk (Simon Brunning)
Date: Wed, 7 Mar 2001 12:36:52 -0000
Subject: Using a jar with jython
References: <3AA6115E.7B2ACDEE@ifi.it>
Message-ID: <3aa62b59_1@nnrp1.news.uk.psi.net>

accattatevillo <^NOSPAM^ifi at ifi.it> wrote in message
<3AA6115E.7B2ACDEE at ifi.it>...
>I'm trying to convert a java application to jython.
>
>I use the IBM java toolbox JT400.jar to access data on a AS/400 bat
I'm
>not able to do the same thing with jython.
>
>With java I added jt400.jar to the CLASSPATH but in jython???
>Anyone could help me?

Works for me:

Jython 2.0 on java1.1.8 (JIT: symcjit)
Type "copyright", "credits" or "license" for more information.
>>> import com.ibm.as400.access
>>> my400 = com.ibm.as400.access.AS400()
>>> my400.getVersion()
4

What are the problems that you are having? Error messages?

BTW, I'm cross-posting this the Jython on AS/400 mailing list. If
you are using Jython to access an AS/400, you might want to join -
see <http://groups.yahoo.com/group/JythonAS400>.

Cheers,
Simon Brunning
sbrunning at bigfoot.com
On two occasions I have been asked (by members of Parliament!),
'Pray, Mr. Babbage, if you put into the machine wrong figures, will
the right answers come out?' I am not able rightly to apprehend the
kind of confusion of ideas that could provoke such a question. -
Charles Babbage





From olczyk at interaccess.com  Tue Mar  6 03:09:47 2001
From: olczyk at interaccess.com (Thaddeus L. Olczyk)
Date: Tue, 06 Mar 2001 08:09:47 GMT
Subject: Can anyone recomend a good intoduction to C...
References: <3aa366c6_2@nnrp1.news.uk.psi.net> <8gNo6.440$y6.128357@ruti.visi.com>
Message-ID: <3aa79a8a.466828531@nntp.interaccess.com>

On Mon, 05 Mar 2001 14:19:48 GMT, grante at visi.com (Grant Edwards)
wrote:

>In article <3aa366c6_2 at nnrp1.news.uk.psi.net>, Simon Brunning wrote:
>>... for a Python programmer?
>>
>>I'm going to have to bite the bullet and learn a bit of C. I'd at
>>least like to learn the *right* way to do it.
>
>The "right" way to do it is with a PDP-11, a Lear-Siegler ADM3a
>and the v7 source code.  ;)  Some would argue in favor of an
>ASR-33, but they tend to annoy the downstairs neighbors.
Recently the original  UNIX was released( isn't the v7 source a bit
old? I would think you get something a bit earlier).
If you surf a bit you can find the source ( finding the source is
really all you need the page has links to everything else ), also the
source to a C compiler and PDP11 emulators for free download.
In fact go to eitherr Ken Thompsons web page or Dennis Ritchies web
page, and you will find it there ).


From bedge at troikanetworks.com  Fri Mar  2 13:34:42 2001
From: bedge at troikanetworks.com (Bruce Edge)
Date: Fri, 02 Mar 2001 10:34:42 -0800
Subject: create new statement?
Message-ID: <3A9FE7C2.1F5525D2@troikanetworks.com>

Can you add a new statement to python?
AFAICT the only type of commands that don't require () or . after them
are statements.

I want to add my own command to the interp, but not have to use object
calling syntax
eg:
    create obj opt
not
    create(obj, opt)

How do I define new statements, or, a way of calling functions using the
former syntax?

Thanks, Bruce.


From sbrunning at trisystems.co.uk  Mon Mar  5 05:14:35 2001
From: sbrunning at trisystems.co.uk (Simon Brunning)
Date: Mon, 5 Mar 2001 10:14:35 -0000
Subject: Can anyone recomend a good intoduction to C...
Message-ID: <3aa366c6_2@nnrp1.news.uk.psi.net>

... for a Python programmer?

I'm going to have to bite the bullet and learn a bit of C. I'd at
least like to learn the *right* way to do it.

--
Cheers,
Simon Brunning
sbrunning at bigfoot.com





From paulp at ActiveState.com  Sat Mar 17 21:05:49 2001
From: paulp at ActiveState.com (Paul Prescod)
Date: Sat, 17 Mar 2001 18:05:49 -0800
Subject: Who's minister of propaganda this week?
References: <B98041329020E007.7DA84CAD9D0562AB.817BBC572F480869@lp.airnews.net> <98nu1m$12e@dispatch.concentric.net> <660ACCA1E4816F7E.C359CDF4168EF3FB.3B6B05538D619770@lp.airnews.net> <3AAFA13D.76E32C3F@divalsim.it> <98oa4o01i4v@news2.newsguy.com> <3AB0176D.8070903@yahoo.com> <3AB0C70A.C6A55DF5@ActiveState.com> <3AB40BE3.3080001@yahoo.com>
Message-ID: <3AB417FD.1905AC3D@ActiveState.com>

Michael Chermside wrote:
> 
> ...
> 
> Yes, I can see that. But sometimes it's awefully nice to have the
> compiler do it for you automatically. 

I absolutely agree that compiler checks of code can catch thing in
statically typed languages earlier and this is a good thing. 

> It certainly doesn't hurt.

I think it does hurt because you end up writing more code and that code
is prone (like any code!) to the subtle semantic errors that are really
tough to find. So you are trading trivial, relatively easy-to-find
errors for subtle, buried in the code errors. It's like driving twice
the distance to avoid a particularly busy intersection. You feel safer
because you can look at the news and say: "Whew, I missed a car crash at
that intersection today" but you have no idea how much you increased
your risk of an accident just by driving farther.

> (But see Alex's response, where he argues that perhaps it DOES hurt,
> for other reasons.)

I think it takes a lot of work with both kinds of languages to recognize
that you lose something significant in exchange for the earlier error
messages. And I think that slowly over the next few years, programmers
are going to to come to that realization.

It is highly akin to the GC situation. A few years back, garbage
collection of any sort was considered unaffordable for real apps (even
in the face of evidence to the contrary) merely because most programmers
did not have real-world experience with it. Now, nobody designs a new
language without GC. Of course it wasn't all just education. Moore's law
helped too. The situation is the same with dynamic typing. Moore's law
and education will bring programmers over.

-- 
Take a recipe. Leave a recipe.  
Python Cookbook!  http://www.activestate.com/pythoncookbook



From aahz at panix.com  Fri Mar  9 12:29:49 2001
From: aahz at panix.com (Aahz Maruch)
Date: 9 Mar 2001 09:29:49 -0800
Subject: Python RMI?
References: <ihvhatcm5v9rn943etearsp15k4q9425ia@4ax.com>
Message-ID: <98b3ud$m9u$1@panix6.panix.com>

In article <ihvhatcm5v9rn943etearsp15k4q9425ia at 4ax.com>,
Daniel Klein  <danielk at aracnet.com> wrote:
>
>Does Python have an equivalent or substitute for Java's RMI?

Yes.








I'm sorry, did you actually want a useful answer?  ;-)  Anyway, go to
the Vaults of Parnassus (http://www.vex.net/parnassus/) and select the
link for "Networking", then select the link for "Remote and
Distributed".
-- 
                      --- Aahz (Copyright 2001 by aahz at pobox.com)

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

"..., and some of you may regard all women as evil traps that exist only
to tease, torture, and suck out your very soul."  --DrMax


From timr at probo.com  Sat Mar 17 23:57:12 2001
From: timr at probo.com (Tim Roberts)
Date: Sat, 17 Mar 2001 20:57:12 -0800
Subject: windows script host and python
References: <%tKr6.2570$pX.26174@news1.onlynews.com> <98oq3r$1e0$1@troll.powertech.no> <8Aps6.104$pl.1515@newsfeed.slurp.net> <98tc4d0rmk@news1.newsguy.com>
Message-ID: <lvf8bt4p6e6naq0vimgnkhs175eoqbajfr@4ax.com>

"Alex Martelli" <aleaxit at yahoo.com> wrote:
>
>Here, for example, is the "list all running processes" WMI scripting
>example translated into Python:
>
>from win32com.client import GetObject
>
>processes = GetObject("winmgmts:{impersonationLevel=impersonate}").\
>    InstancesOf("Win32_process")
>
>for process in processes:
>    print process.Name

Cool!  Is this supposed to be NT-only?  It doesn't fly on my Win98 box.
--
- Tim Roberts, timr at probo.com
  Providenza & Boekelheide, Inc.


From ejr at cs.berkeley.edu  Tue Mar 13 17:16:57 2001
From: ejr at cs.berkeley.edu (Edward Jason Riedy)
Date: 13 Mar 2001 22:16:57 GMT
Subject: This math scares me
References: <3aad4062.1290021@News.CIS.DFN.DE> <slrn9aqkth.aki.scarblac@pino.selwerd.nl> <3aad6ffe.3692710@News.CIS.DFN.DE>
Message-ID: <98m68p$gog$2@agate.berkeley.edu>

And Costas Menico <costas at meezon.com> writes:
 - Fortunately I discovered I can use "print 5.01 + 5.54" and it does the
 - job. It has the extra intelligence built into it to autoformat

It's called cosmetic rounding, and it's evil.  It tricks you
into believing you have a number you don't.  See 
	W. Kahan, ``Marketing versus Mathematics.''
	http://www.cs.berkeley.edu/~wkahan/MktgMath.pdf
starting on page 13.

Well, ok, there's one circumstance when it's less than purely
evil.  Some laws and regulatory rules mandate certain types of
rounding.  But then it's done with some forethought (and malice?)
and not just silently hoisted upon an unsuspecting programmer.

Jason
-- 


From aahz at panix.com  Fri Mar 16 19:11:32 2001
From: aahz at panix.com (Aahz Maruch)
Date: 16 Mar 2001 16:11:32 -0800
Subject: HTMLparsing abnormal html pages
References: <98pvp1$15t$1@news.netmar.com> <B6D81330.4FDA%f8dy@diveintopython.org>
Message-ID: <98ua3k$9vi$1@panix2.panix.com>

In article <B6D81330.4FDA%f8dy at diveintopython.org>,
Mark Pilgrim  <f8dy at diveintopython.org> wrote:
>in article 98pvp1$15t$1 at news.netmar.com, asle at spam.com at asle at spam.com
>wrote on 3/15/01 3:50 AM:
>>
>> Considering the small program below. Running it will show that the
>> HTMLparser
>> is truncating urls in the HTML page.
>> [...]
>> import htmllib
>> [...]
>> One solution is of course to preprosess the whole HTML page and
>> replacing invalid URL's whith valid URL's (using regex??), however I have
>
>Don't use htmllib, use sgmllib.  It does exactly what you want: uses regular
>expressions to pull out the tags and attributes of potentially messy HTML,
>then calls methods on itself based on the tags.  You can subclass it and
>provide methods for each tag.

That doesn't work for truly malformed HTML.  For example, sgmllib will
still produce incorrect results for something like this:

<a href="asdfasdfasdf>malformed link</a>
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

Three sins: BJ, B&J, B&J


From loewis at informatik.hu-berlin.de  Tue Mar 20 15:29:31 2001
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 20 Mar 2001 21:29:31 +0100
Subject: re.match
References: <997skf$qoc$1@bmerhc5e.ca.nortel.com> <TrMt6.2490$4N4.318602@newsc.telia.net> <99874d$4tp$1@bmerhc5e.ca.nortel.com>
Message-ID: <j4r8zsi3pw.fsf@informatik.hu-berlin.de>

msoulier at nortelnetworks.com (Michael P. Soulier) writes:

> >no, it's not.  read the documentation again.
> 
>     That answer was so warm and informative I'd think this was
> comp.lang.perl.misc. 

You have to take two things into account:

a) you were critizing the author of the regex code (in its sre
   incarnation), and
b) this is regular expressions, which seem be unfriendly by nature
   (which perhaps had an influence on comp.lang.perl.misc as well :-)
c) Fredrik spent quite some time getting all the "start of line"
   and "start of string" cases right. It was difficult to write,
   so it should be difficult to understand :-)

Well, that's three things :-)

>     The difference is apparently multiline mode. My point was that
> it doesn't seem to offer enough of a difference to justify going
> against standard regexp syntax and ignore the ^ character. It is
> misleading, IMHO, and Python seems to be all about being easy to
> learn. Thus being misleading is not beneficial.

I find match extremely useful, and I never ever use search. The reason
is that I use regular expressions for tokenization of languages with a
formal syntax. There, I don't care whether the point where I look for
the next token happens to be the start of a line - but I absolutely
must make sure that any match I get is at the current position (or
else there is a syntax error).

Please have a look at the undocumented sre.Scanner class for an
example of using match in a loop.

> As always, one asks questions not to simply annoy, but to
> learn. There was a time when Usenet was a good place to do that.

Well, maybe you should phrase your question in a way that encourages
discussion. You've stated 

# If it's only difference from search is to match at the beginning of
# the line, why does it exist? That's what the ^ in the regexp is for.

quite factually, implying that there can't be any possible other use
or difference. The manual would have told you that there is a
difference, so if you had read the manual, you could have shown in the
article that you know the difference, and ask the question that you
were really interested in "Why would anybody ever use match?" :-)

Regards,
Martin



From gorleyd at unbc.ca  Thu Mar 29 15:05:34 2001
From: gorleyd at unbc.ca (Doug Gorley)
Date: Thu, 29 Mar 2001 12:05:34 -0800
Subject: Printing mixtures in integers and strings
In-Reply-To: <3ac38cf6$0$12246$ed9e5944@reading.news.pipex.net>
References: <3ac38cf6$0$12246$ed9e5944@reading.news.pipex.net>
Message-ID: <Pine.LNX.4.21.0103291204310.6448-100000@ugrad.unbc.ca>

On Wed, 28 Mar 2001, Neil Benn wrote:

> Hello,
> 
>             I'm trying to use something like this:-
> 
> def move(row): #Moves the dispenser to the row on the plate
>     print "Moved to Row - " + row
> 

Try defining your function as:

def move( row ) :
   print "Moved to Row -", row

Use the comma instead of the + sign.  This should solve your problem.

-- 

 -------------------
 | Douglas Gorley  |
 | gorleyd at unbc.ca |
 -------------------



From donn at u.washington.edu  Wed Mar 14 14:32:35 2001
From: donn at u.washington.edu (Donn Cave)
Date: 14 Mar 2001 19:32:35 GMT
Subject: Python, threads, and signals (oh my!)
References: <98jqeh$6ut$1@newshost.mot.com>
Message-ID: <98oh0j$tik$1@nntp6.u.washington.edu>

Quoth jlowe at mentos.urbana.css.mot.com (Jason Lowe):
[
  I'm quoting this lengthy post intact partly just because it was
  interesting and well researched and could be missed in the heavy
  traffic here.
]
| I've been playing around with Python and threads, and I've noticed some
| odd and often unstable behavior.  In particular, on my Solaris 8 box I
| can get Python 1.5.2, 1.6, or 2.0 to core dump every time with the
| following sequence.  I've also seen this happen on Solaris 6.
|
|
| 1. Enter the following code into the interactive interpreter:
| --
| import threading
|
| def loopingfunc():
|   while 1: pass
|
| threading.Thread(target=loopingfunc).start()
| --
|
| 2. Send a SIGINT signal (usually Ctrl-C, your terminal settings may
|    vary).  "Keyboard Interrupt" is displayed and so far everything looks 
|    fine.
|
| 3. Now simply press the <Enter> key to enter a blank line in the
|    interpreter.  For my Solaris 8 box with the GNU readline 2.2 module
|    present, this always ends up in a core dump.  It may take a while,
|    since at this point the readline signal handler is being re-entered
|    recursively until the stack overflows.
|
|
| Looking more into this, it appears that on Solaris, more than one
| thread is processing the signal handler installed by the readline
| module (according to truss output of the process).  Unfortunately, the
| readline signal handlers don't support being re-entered, as they have
| global data that's not protected.  
|
| Now granted, signals and threads are a dangerous business.  However, it
| would be nice if the user sending SIGINT to a script wouldn't cause
| instability.  Looking at the Python module sources, I noticed that the
| signal module is somewhat "thread aware" -- it allows signal
| handlers to be installed only by the main thread.  However, I found it a
| little odd that the thread support, both in the core interpreter and in
| the thread module, had no support code for signals -- even in the
| specific thread cases like POSIX (pthreads).
|
| According to various pthread documentation, it appears the "right" way
| to handle the often volatile signal/thread mix is to mask all signals
| except in one thread (usually the main thread) so that only one thread
| will receive the signal.  Unfortunately, I couldn't find any Python
| module that would allow one to change the signal mask of a thread
| (pthread_sigmask) or even a process (sigmask, sigprocmask, etc.).  I'm
| assuming the lack of this interface has to do with portability across
| platforms.
|
| I was able to solve the problem by modifying Python/thread_pthread.h's
| PyThread_start_new_thread() to block all signals before creating the new 
| thread and then restoring the signal mask after the new thread was off
| and running.  Therefore, all threads created by Python except the
| initial thread will have all signals masked with this change.  Masking
| all signals in new Python threads makes sense to me, given that the
| signal module doesn't like other threads installing handlers anyway.
|
| As a side note, I think part of the problem with sending SIGINT at the
| interactive prompt while threading is aggravated by the Python readline
| module's signal handler.  It setjmp()'s and longjmp()'s to do its dirty
| work, and many thread implementations require the longjmp() be performed 
| by the thread that did the setjmp().  If signals aren't masked in all
| threads except the one doing the readline() call, then this isn't
| guaranteed.
|
| So what's everyone else's take on this?  Has anyone else experienced
| problems in Python when using threads and receiving signals (like
| SIGINT)?  Should the thread code be at least a little more "signal
| aware", so that new Python threads have all signals masked on platforms
| that support this?
|
| Thanks in advance.
|
| Jason Lowe
| --
| Jason Lowe                                    Urbana Design Center
| Motorola Personal Communications Sector       1800 South Oak Street
| jlowe at urbana.css.mot.com                      Champaign, IL  61820-6947
|                                               (217) 384-8513

I ran some experiments on other UNIX platforms, and Solaris is not
alone if signals aren't handled well when there are multiple concurrent
threads.  In general, signals were trapped but Python handler functions
weren't executed.

My take is that it's complicated.  Does it make sense to split out
readline's special problems as an issue that could be addressed
separately?  In my experiments, without readline, there were still
these issues, and even without readline, signal handling at the
interpreter prompt differed from an interpreted disk file.

I can't think of any reason why it would hurt anything to block
signals by default per non-main thread.  I don't think it will
work for many platforms, but if it works for anyone's I guess
that's better than nothing.  (And the one I would be most concerned
about, BeOS, doesn't seem to need it, as signals seem to work more
or less as intended in multithreaded programs there.)

I think the ability to block signals in general should be added,
in the form of the POSIX 1003.1 sigprocmask function.  It would
be easy enough to do, if anyone cared enough.  Is that how you
masked off the signal from the new thread - do new pthreads copy
their signal handling from the parent at creation?

	Donn Cave, donn at u.washington.edu


From ykingma at accessforall.nl  Wed Mar 28 05:33:09 2001
From: ykingma at accessforall.nl (Ype Kingma)
Date: Wed, 28 Mar 2001 11:33:09 +0100
Subject: Jython and JPython Questions...How similar to Python, Tkinter?
References: <3ABFF14A.30D0096D@earthlink.net>
Message-ID: <3AC1BDDB.11D8E367@accessforall.nl>

Ron,

you wrote wrote:
> 
> Does Jython support Tkinter? OK, maybe that question as worded is
> nonsense, let me re-phrase it: If I code a program in Python and put a
> GUI front end on it with Tkinter, how hard would it be to do the same
> program in Jython? Or is there a better solution for Jython GUI's?
>

From: http://www.jython.org/docs/differences.html 
Under extension modules:

 However, Finn Bock has a JNI implementation called jTkinter which supports the
 full _tkinter API. Very cool stuff! 

and this is the link: http://jTkinter.sourceforge.net/

>     By the way, as I am new to Python and learning it, but I am
> interested in Jython for embedding (relatively simple) applets into
> webpages, how different is Jython from Python? How steep will the
> learning curve be for Jython, after learning a reasonable amount of
> Python?

>From a language point of view the implementations are almost as equal as
possible.
To make an applet in Jython have a look at:

http://www.jython.org/applets/index.html


> 
>     How hard is it to translate Python programs into Jython?

Check the differences page above. It mostly depends on how much use
you make of extension modules that are not available 'on the other side'.
Beware that from jython it is very easy to use java features:

from java.lang import System
System.out.writeln('Hello')

Not very useful (python print is better), but you can also do that with swing
for example.

>     How good is Jython? Can I write Jython programs, and then turn them

Just subscribe to the jython-users list.
Or browse the archives a bit and then realize that most of
the Python language itself is _not_ discussed. Jython works.

http://lists.sourceforge.net/lists/listinfo/jython-users

> into Java applets that easily embed into web pages? (I have a small but
> working knowledge of embedding Java applets in web pages, despite my
> overwhelming newbie status in all things coding related ;-)))

That plus some basic Python knowledge should give you a head start.

>     Please forgive my newbieness and all replies will be deeply
> appreciated!!!
> 
> Ron Stephens

My pleasure. Newbieness is easily lost in the Python environment.
Ype

-- 
email at xs4all.nl


From jwbnews at scandaroon.com  Thu Mar 15 03:13:38 2001
From: jwbnews at scandaroon.com (John W. Baxter)
Date: Thu, 15 Mar 2001 00:13:38 -0800
Subject: [Q] Bug in Python?
References: <t6Zr6.92712$__6.18256422@typhoon.southeast.rr.com>
Message-ID: <jwbnews-91FA17.00133815032001@corp.supernews.com>

In article <t6Zr6.92712$__6.18256422 at typhoon.southeast.rr.com>,
 "Daehyok Shin" <DSHIN at nc.rr.com> wrote:

> Recently, I met a strange behavior of Python 6.1.
> When I tried the following module,
> I got an error message in "print var":
>     UnboundLocalError: var
> 
> But, when I changed "var=None" to "pass",
> the error message was gone and I got an expected result:--printing None and
> None.
> What happens here? Is this a bug in Python or I missed something?
> 
> ############# test.py ###############
> var = None
> 
> def a():
>      print var #UnboundLocalError
>      if var != None:
>           var = None  # >> pass
> 
> def b():
>      print var
>      a()
> 
> b()

> 

You're getting the unbound local error because, with the var = None 
form, the 
   print var
is indeed trying to print an unbound local variable.  var is a local 
because it is assigned to in a() but that hasn't happened by the time 
print needs to use it.  When you substitute the pass, then var becomes a 
global variable.  If you want to use the (module) global var in a() with 
the assignment to var, declare var global:

    global var

  --JOhn

-- 
John W. Baxter   Port Ludlow, WA USA  jwbnews at scandaroon.com


From gregj at pobox.com  Tue Mar 13 04:35:42 2001
From: gregj at pobox.com (Gregory Jorgensen)
Date: Tue, 13 Mar 2001 09:35:42 GMT
Subject: Subject with SMTPLIB?
References: <3AADE48A.5D8AA43@daimlerchrysler.com>
Message-ID: <ORlr6.2821$54.2794@www.newsranger.com>

You need to add a subject header to your message. Email message format is
simple: headers, blank line, message body.

import smtplib

msg = "subject: hello, world!\n\nmessage body"

s = smtplib.SMTP(host)
s.sendmail(fromaddr, toaddrs, msg)

You'll find a good SMTP quick reference on Dan Bernstein's site (he's the author
of qmail). See http://cr.yp.to/immhf.html.


In article <3AADE48A.5D8AA43 at daimlerchrysler.com>, root says...
>
>Hello!
>
>I need to write some mail for my fax-log evaluating script. I'm
>currently using
>the smtplib to simply send some mails. But how can I get (without
>studying RFCs
>and papers on SMTP for years)  a SUBJECT for my mails? It must be
>this easy, but the library reference in the python documentation package
>tells
>me nothing about it.  I was disappointed that the documentation on the
>smtplib
>module didn't tell me that.

Greg Jorgensen
Deschooling Society
Portland, Oregon, USA
gregj at pobox.com


From bwalker at earth1.net  Sat Mar 31 10:30:38 2001
From: bwalker at earth1.net (Bill Walker)
Date: Sat, 31 Mar 2001 10:30:38 -0500
Subject: Starship Compiler Question
Message-ID: <tcbtng94tbju68@corp.supernews.com>

I downloaded the Starship Compiler for Python, then I compiled a test
routine using Tkinter to plot a simple graph. Worked like a charm. Using the
Standalone method, it created a directory with all the dlls and other
support files necessary, plus the test.exe file that runs stand alone.

The question: I tried compiling both a test.py and a test.pyw file and both
generate an .exe that produces the annoying DOS window under the Tk window.
How can I get rid of it?






From dougfort at downright.com  Thu Mar 29 07:24:13 2001
From: dougfort at downright.com (Doug Fort)
Date: Thu, 29 Mar 2001 07:24:13 -0500
Subject: htmllib samples
References: <3AC2F48E.4EF77C71@aon.at>
Message-ID: <3AC3296D.E24DBA02@downright.com>

Oliver Vecernik wrote:

> Hi,
>
> has anyone got some samples or point me to where to find them on how to
> implement this module?
>
> Oliver

I've attached a simple parser we use.  It may help.

Note that Frederick  Lundh modestly failed to mention his upcoming book on
'The Python Library'.  There's an ebook version of his 'eff-bot' Guide,
which is the  first place I look for code  examples.

--
Doug Fort (dougfort at downright.com)
Senior Meat Manager
Downright Software LLC
http://www.dougfort.net


-------------- next part --------------
#!/usr/bin/env python
"""
FormFieldParser

This object parses HTML text and builds a dictionary of
dictionaries of form fields

$Id: formfieldparser.py,v 1.1 2001/01/26 15:18:30 dougfort Exp $
"""
__author__="Downright Software LLC"
__version__="$Revision: 1.1 $"[11:-2]

import sgmllib
import string
import cStringIO
import urllib
import re

import webnudge.util.misc
import webnudge.util.document

class FormFieldParserException:
    def __init__(self, message):
        self._message = message
    def __str__(self):
        return self._message

###########################################################
class FormFieldParser(sgmllib.SGMLParser):
###########################################################
    """
    FormFieldParser class. Parse a page from a website,
    creating a dictionary of dictionairies of form
    fields
    """

    #----------------------------------------------------------
    def __init__(self):
    #----------------------------------------------------------
        """
        Constructor
        """
        sgmllib.SGMLParser.__init__(self)

        self._formcount = 0
        self._formdict = {}

    #----------------------------------------------------------
    def parse(self, text):
    #----------------------------------------------------------
        """
        parse some text, without trashing javascript
        """
        self.feed(text)
        self.close()
        return self._formdict
        
    #----------------------------------------------------------
    def start_form(self,attributes):
    #----------------------------------------------------------
        """
        start a form
        """
        self._formdict[self._formcount] = {}
    #----------------------------------------------------------
    def end_form(self):
    #----------------------------------------------------------
        """
        end a form
        """
        self._formcount += 1
        
    #----------------------------------------------------------
    def _storeformfield(self,attributes,multivalue=0):
    #----------------------------------------------------------
        """
        Capture name and value attributes of a form field
        """
        tagname = None
        tagvalue = ""
        selected = 0
        for key, value in attributes:
            if key == "name":
                tagname = value
                continue
            if key == "value":
                tagvalue = value
                continue
            if key == "selected":
                selected = 1
                continue
        if multivalue and not selected:
            return
            
        if tagname:
            self._formdict[self._formcount][tagname] = tagvalue
        
    #----------------------------------------------------------
    def do_input(self,attributes):
    #----------------------------------------------------------
        """
        Capture <input> element
        """
        self._storeformfield(attributes)
        
    #----------------------------------------------------------
    def do_option(self,attributes):
    #----------------------------------------------------------
        """
        Capture <option> element
        """
        self._storeformfield(attributes, multivalue=1)
        
    #----------------------------------------------------------
    def do_select(self,attributes):
    #----------------------------------------------------------
        """
        Capture <select> element
        """
        self._storeformfield(attributes, multivalue=1)
        
    #----------------------------------------------------------
    def do_textarea(self,attributes):
    #----------------------------------------------------------
        """
        Capture <textarea> element
        """
        self._storeformfield(attributes)
        
#----------------------------------------------------------
if __name__ == "__main__":
#----------------------------------------------------------
    """
    Code for commandline testing
    """
    import sys
    if len(sys.argv) != 2:
        print "Usage:  filteringparser.py <url>"
        sys.exit(-1)

    import webnudge.util.rawhtmlpage
    page = webnudge.util.rawhtmlpage.RawHTMLPage()
    page.load(sys.argv[1])
    if not page:
        print "*** Error *** %s" % (page._message)
        sys.exit(-1)

    result = FormFieldParser().parse(page._data)
    
    sys.stdout.write(repr(result))



From amitp at Xenon.Stanford.EDU  Sun Mar  4 01:45:40 2001
From: amitp at Xenon.Stanford.EDU (Amit Patel)
Date: 4 Mar 2001 06:45:40 GMT
Subject: gah! I hate the new string syntax
References: <mailman.983485923.1136.python-list@python.org> <97p7iq$kk2$1@nntp6.u.washington.edu>
Message-ID: <97soak$qh9$1@nntp.Stanford.EDU>

 Russell E. Owen <owen at astrono.junkwashington.emu> wrote:
| In article <mailman.983485923.1136.python-list at python.org>,
|  "Sean 'Shaleh' Perry" <shaleh at valinux.com> wrote:
| 
| >return ";".join(["%s=%s" % (k, params[k]) for k in params.keys()])
| >
| >every day python seems to be moving closer to the line noise aspect of
| >coding.
| >
| 
| Mostly I think string methods are a great idea. One thing that annoyed 
| me about python early on was the plethora of global functions that made 
| more sense to me (a smalltalk lover, I admit) as methods. So overall I'm 
| very pleased to see the new string methods.
| 
| However, I do agree that join is not intuitive. I believe the problem 
| (at least for the way I look at it) is that join should be a list 
| method, not a string method. I.e.:
| 
|   joined_string = ['a', 'b'].join(', ')
| 
| makes a lot of sense to me. The current join string method does not.

One thing that I *liked* about Python was that it had a split(a,b)
function instead of a method:  a.split(b) or b.split(a).

When I see split(a,b), I think, "Oh, this isn't an inherent part of a
string-- *I* could write it, even though I didn't write the string
object."  

When I see b.split(a), I think, "This is a basic part of a string, and
I *cannot* write it."


If I have a split-like function I'd like to write (perhaps
re.split, or some custom split function), then what I'll see is:

          function style              method style

          string.split(a,b)           b.split(a)
          re.split(a,b)               re.split(a,b)
          mysplit(a,b)                mysplit(a,b)

Note that with the function style, they're consistent -- MY function
is on equal footing with string.split.  But with the new method style,
the split method on strings is elevated to a privileged position that
I cannot do for mysplit.  So now I have inconsistent syntax.

So I will continue to use string.split in my code.  I'll only use the
methods for things that are conceptually inherently tied to strings.
*Maybe* replace and index.  But certainly not capitalize or center.
They feel like a user definable function that are in the string module
for convenience.  At least in my mind, capitalize is NOT part of the
basic functionality of a string class.



      - Amit
-- 
--
Amit J Patel, Computer Science Department, Stanford University
http://www-cs-students.stanford.edu/~amitp/


From nils at hhk.de  Wed Mar 21 07:01:47 2001
From: nils at hhk.de (Nils Hensel)
Date: Wed, 21 Mar 2001 13:01:47 +0100
Subject: Checking the type
References: <99a493$bhl$1@sun.rhrk.uni-kl.de>
Message-ID: <99a539$buco$1@ID-59026.news.dfncis.de>

Look for the types module:

from types import *
if type(a) is ListType:
  # do list stuff

HTH,
Nils

"Markus Reitz" <Markus_Reitz at yahoo.com> schrieb im Newsbeitrag
news:99a493$bhl$1 at sun.rhrk.uni-kl.de...
> Hi,
>
> is there a possibilty to check if a variable refers to an object of type
x?
> Does a typeof-command exist in Python:
>
>   if a typeof list:
>      #Commands only useable with list
>
> I read the Python Tutorial, but have not found a hint to this topic.
>
> Thanks
>
> Markus
>
>




From aleaxit at yahoo.com  Wed Mar  7 03:14:23 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Wed, 7 Mar 2001 09:14:23 +0100
Subject: Python, ASP, ADO, and Image string/binary problems
References: <ta0amgebq0sn0b@corp.supernews.com> <tabjei1p4a9481@corp.supernews.com>
Message-ID: <984ql60tdk@news1.newsguy.com>

"Shonn Gilson" <shonn at shonn.com> wrote in message
news:tabjei1p4a9481 at corp.supernews.com...
    [snip]
> that '\377' is octal not decimal.). The problem seems to be that when
Python
> passes strings through COM that they get translated to Unicode. This
happens

Yes, this IS what COM (actually, Automation) specifies -- *all* strings
are Unicode (note that Java specifies exactly the same thing, but COM
was there first:-).

Bravo for diagnosing that this was the problem in your case, though!


> for things like Response.BinaryWrite along with things like recordsets
> (objRS = Server.CreateObject("ADODB.Recordset")).  The result is that the
> beginning of my jpg file looks like
> '\377\000\330\000\377\000\340\000\000\000\020\000J\000F\000I\000F' instead
> of '\377\330\377\340\000\020JFIF'.
>
> Does anyone know how to handle this?

If you don't want to use *STRINGS* (normally-readable sequences of
characters), then don't.  Python offers other datatypes, which will
help you map (for example) *sequences of BYTES* (a byte is not the
same thing as a character -- even though decades of programming with
single-byte-characters strings have obscured this for many:).


Specifically, I think that what you want is the built-in *BUFFER*
type, which Python/COM will map into (in low-level COM terms) a
SAFEARRAY of VT_UI8 (a sequence-of-bytes, uninterpreted).


Try just changing this line:

> > Response.BinaryWrite(bstrOut.getvalue())

to:

> > Response.BinaryWrite(buffer(bstrOut.getvalue()))

and see if that solves it...


Alex





From phd at phd.russ.ru  Mon Mar  5 08:05:15 2001
From: phd at phd.russ.ru (Oleg Broytmann)
Date: 5 Mar 2001 07:05:15 -0600
Subject: Dr. Dobb's Python-URL! - weekly Python news and links (Mar  5)
Message-ID: <FFE41EE542E94336.B7A7E90E01745DB0.38760D529048CCC3@lp.airnews.net>

    Guido van Rossum releases Python 1.6.1 to fix GPL incompatibility
        http://mail.python.org/pipermail/python-announce-list/2001-February/000682.html
    
    On the Python2 side: Python 2.1b1 released
        http://groups.yahoo.com/group/python-list/message/126320
    New in this release: Nested Scopes are now optional, new tool pydoc, full
    support for MacOS X, Cygwin, and RISCOS.
    
    Tim Peters makes PEP 236: "Back to the __future__" and thus starts hot
    discussions, proposals, counter-proposals and counter-counter-proposals
        http://groups.yahoo.com/group/python-list/messagesearch?query=PEP%20236%3A%20%20Back%20to%20the%20__future__
        http://groups.yahoo.com/group/python-list/messagesearch?query=counter-proposal%20for%20PEP%20236
    
    Michael Hudson publishes python-dev summary for 2001-02-15 - 2001-03-01
        http://groups.yahoo.com/group/python-list/message/126217
    
    Cameron Laird and Kathryn Soraiz: REGULAR EXPRESSIONS - Pulling different
    threads
        http://www.sunworld.com/unixinsideronline/swol-02-2001/swol-0216-regex.html
    
    A Python role playing game! Play your favorite Python community member! While
    said member watches! Travel through time! Save the planet! Martijn Faassen
    proposes Python RPG (joke)
        http://groups.yahoo.com/group/python-list/message/125822
    Read March instead of January, of course
    
    Oleg Broytmann takes a SpamCop perl script and rewrites it to Python;
    fighting spam has never been easier!
        http://groups.yahoo.com/group/python-list/message/125520
        http://spamcop.net/fom-serve/cache/166.html
    
    GNU Solfege is an eartraining program for X written in Python, using
    the GTK+ and GNOME libraries
        http://www.solfege.org/
    
    Charming Python - Updating your Python reading list
    David Mertz: Reviewing eight of the best known Python books
        http://www-106.ibm.com/developerworks/library/l-cp12.html?dwzone=linux
    
    David also reveals a code for full-text indexing and searching
        http://groups.yahoo.com/group/python-list/message/126371
    
    Peter Norvig: Extreme Rapid Development
        http://www.sdmagazine.com/articles/2001/0103/0103e/0103e.htm
    
    Tom Schwaller: Scheduling with Python and Webware
        http://www.python.de/docs/QuickStart.html
    
    A.M. Kuchling: Quotation Exchange Language, or QEL, is an XML
    language for exchanging collections of quotations
        http://www.amk.ca/qel/
        http://xml.coverpages.org/ni2001-03-03-a.html
    
    Python Helps Disney Write a New Script
        http://python.oreilly.com/news/disney_0201.html
    
    Chris Gonnerman posts C source to a shadow module which contains
    the functions lckpwdf() and ulckpwdf()
        http://groups.yahoo.com/group/python-list/message/125999
    
    Python must matter if it can be used with Ada, no?  pyAda
    enables programmers to write Python extensions in Ada 95 or
    extend Ada 95 programs with a Python interpreter
        http://groups.yahoo.com/group/python-list/message/125989
        http://pyada.sourceforge.net/
    
    Federico Di Gregorio, the same man who brought to us PoPy
    PostgreSQL driver, writes new Postgres driver and Zope
    Database Adapter
        http://initd.org/
    
    PythonJournal is alive! It becomes interactive!
        http://groups.yahoo.com/group/python-list/message/126159
    Not all people are happy, though
        http://groups.yahoo.com/group/python-list/message/126180
    
    W3C: Python modules for RDF
    Tim Berners-Lee: By the way... Python is cool!
        http://www.w3.org/2000/10/swap/
    
    Kirby Urner: Oregon curriculum network.  Hitting two targets:
    OO + group theory (pedagogy). Python as intro language
        http://groups.yahoo.com/group/python-list/message/126344
    
    Barry Warsaw announces Mailman 2.0.2 - important privacy patch
        http://groups.yahoo.com/group/python-list/message/126373
    
    Beware the southern pythons! (joke)
        http://groups.yahoo.com/group/python-list/message/125616
    
    Alex Martelli: I come to praise .join, not to bury it
        http://groups.yahoo.com/group/python-list/message/126436
    
    
    Zope:
    Paul Tchistopolskii makes Zope Prouct API easy with Zope Beans
        http://lists.zope.org/pipermail/zope/2001-March/084640.html
        http://www.pault.com/zope_beans.html
    
    
    Hot threads of the week:
    
    CPAN functionality for python
        http://groups.yahoo.com/group/python-list/messagesearch?query=CPAN%20functionality%20for%20python
    Combinations of n Lists
        http://groups.google.com/groups?lr=&group=comp.lang.python&safe=off&ic=1&th=275c44135cba024f&seekd=920137360
    Collection interfaces
        http://groups.google.com/groups?num=25&lr=&group=comp.lang.python&safe=off&th=34d9ba9d0f261c1a&ic=1&seekd=920732844


========================================================================

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

      Python.org's Python Language Website is the traditional
      center of Pythonia
          http://www.python.org
      Notice especially the master FAQ
          http://www.python.org/doc/FAQ.html

      PythonWare complements the digest you're reading with the
      daily python url
           http://www.pythonware.com/daily  

      comp.lang.python.announce announces new Python software.  Be
      sure to scan this newly-revitalized newsgroup at least weekly.
	  http://groups.google.com/groups?oi=djq&as_ugroup=comp.lang.python.announce

      Michael Hudson continues Andrew Kuchling's marvelous tradition
      of summarizing action on the python-dev mailing list twice a
      month.
	  http://starship.python.net/crew/mwh/summaries/
	  http://www.amk.ca/python/dev

      The Vaults of Parnassus ambitiously collect Python resources
          http://www.vex.net/~x/parnassus/   

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

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

      Cetus does much of the same
          http://www.cetus-links.de/oo_python.html

      Python FAQTS
          http://python.faqts.com/

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

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

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

      Archive probing trick of the trade:
	  http://groups.google.com/groups?oi=djq&as_ugroup=comp.lang.python&num=100


Previous - (U)se the (R)esource, (L)uke! - messages are listed here:
  http://purl.org/thecliff/python/url.html
or
  http://groups.google.com/groups?oi=djq&as_q=+Python-URL!&as_ugroup=comp.lang.python


Suggestions/corrections for next week's posting are always welcome.
[http://www.egroups.com/list/python-url-leads/ is hibernating.  Just
e-mail us ideas directly.]

To receive a new issue of this posting in e-mail each Monday morning,
ask <claird at neosoft.com> to subscribe.  Mention "Python-URL!".


-- The Python-URL! Team--

Dr. Dobb's Journal (http://www.ddj.com) is pleased to participate in and
sponsor the "Python-URL!" project.
-- 

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


From chris.gonnerman at usa.net  Mon Mar  5 09:03:35 2001
From: chris.gonnerman at usa.net (Chris Gonnerman)
Date: Mon, 5 Mar 2001 08:03:35 -0600
Subject: urllib.urlopen and proxies
References: <vse4atgcgcfafth7c5tq3po7srd3l32f3h@4ax.com>
Message-ID: <008301c0a57d$1331cde0$a100000a@local>

It *appears* (from reading the source) that username and password
information for
the proxy server must be passed as part of the URL:

    http://username:password at my.host.com/index.html

for instance.  I haven't tested this yet.  The source for urllib is a bit
dense
IMHO but the answers are in there.

----- Original Message -----
From: "Paul Moore" <gustav at morpheus.demon.co.uk>
Newsgroups: comp.lang.python
To: <python-list at python.org>
Sent: Sunday, March 04, 2001 7:44 AM
Subject: urllib.urlopen and proxies


> I've just looked at urllib for the first time. I was pleased to find
> that it automatically picks up my proxy configuration from the system
> (this is ActiveState's Python 2.0 on Windows 2000). However, I have a
> problem - my proxy requires a user name and password, and there is no
> place that I can see to enter them. So all I get back from urlopen()
> is the polite "you need authorisation" message that our proxy gives
> :-(
>
> Can anyone tell me how to pass a proxy userid and password to
> urllib.urlopen()?
>
> Thanks,
> Paul Moore
> --
> http://mail.python.org/mailman/listinfo/python-list
>




From euripides2002 at yahoo.co.uk  Fri Mar 23 05:38:13 2001
From: euripides2002 at yahoo.co.uk (SpagMan)
Date: Fri, 23 Mar 2001 10:38:13 +0000
Subject: ANN: SPAG - for your amusement
Message-ID: <3ABB2794.31259FC@yahoo.co.uk>

- with apologies - a small offering you might find fun

It does demonstrate the power of OO, recursion, Python
along with pygtk and the gnomecanvas.
- also that you can write bad code in Python but
it is still manageable :-/

http://uk.geocities.com/euripides2002/spag.html

- it actually started just as an excercise to learn
pygtk and gnome....but I find it so useful I am using it!!!!

Bon Voyage !!




From phd at phd.pp.ru  Tue Mar 27 10:20:13 2001
From: phd at phd.pp.ru (Oleg Broytmann)
Date: Tue, 27 Mar 2001 19:20:13 +0400 (MSD)
Subject: Regular expressions and non-standard character set
In-Reply-To: <ka8hf0fp86k.fsf@cheerful.cs.hut.fi>
Message-ID: <Pine.LNX.4.33.0103271919001.609-100000@phd.fep.ru>

On 27 Mar 2001, Petri Mikael Kuittinen wrote:
> I tried to find information about setting locale. The Linux man pages
> nor the standard Python were not helpful on this matter.

   Just look into docs for python module locale.

> > >>> import re
> > >>> re.findall(r"\b...\b", "spam, egg, bacon, and ???")
> > ['egg', 'and']
> > >>> re.findall(r"(?L)\b...\b", "spam, egg, bacon, and ???")
> > ['egg', 'and', '???']
>
> I tried the above. It didn't work under Python 2.0 under Windows 2000,

   This is known bug. Will be fixed in Python 2.1 and, I hope, in Python
2.0.1.

Oleg.
----
     Oleg Broytmann            http://phd.pp.ru/            phd at phd.pp.ru
           Programmers don't die, they just GOSUB without RETURN.




From no-spam-popov at bigserver.com  Sun Mar 11 16:02:10 2001
From: no-spam-popov at bigserver.com (Boris Popov)
Date: Sun, 11 Mar 2001 13:02:10 -0800
Subject: Einstein's Riddle
References: <oNPq6.396644$ge4.138479566@news2.rdc2.tx.home.com> <98glrd$1si1a$1@ID-75892.news.dfncis.de> <3AABE712.499CCF4C@home.com>
Message-ID: <98gp5b$1u0oi$1@ID-75892.news.dfncis.de>

"Bruce Elrick" <belrick at home.com> wrote in message
news:3AABE712.499CCF4C at home.com...
> I swear there is no evidence that any of them own a fish.  The German
> owns an emu.
>
> Bruce

I am missing something then. I thought my table satisfies all given
statements and it has fish right under the german...

-Boris
P.S. What's emu ?

>
> Boris Popov wrote:
> >
> > "Steve Graham" <js.graham at home.com> wrote in message
> > news:oNPq6.396644$ge4.138479566 at news2.rdc2.tx.home.com...
> > > Who would be interested in using his/her brain (and his computer) to
solve
> > > the following?
> > >
> > >
> > > Steve Graham
> > >
> > > ===
> > >
> > > Einstein's Riddle
> > >
> > > Albert Einstein wrote this riddle this century [ed. 20th century].  He
> > said
> > > 98% of the world could not solve it.
> > >
> > > There are 5 houses in 5 different colors.  In each house lives a
person
> > with
> > > a different nationality.  The 5 owners drink a certain type of
beverage,
> > > smoke a certain brand of cigar, and keep a certain pet.  No owners
> > > have the same pet, smoke the same brand of cigar or drink the same
> > beverage.
> > >
> > > The question is: "Who owns the fish?"
> > >
> > > Hints:
> > >
> > > The Brit lives in the red house.
> > >
> > > The Sweed keeps dogs as pets.
> > >
> > > The Dane drinks tea.
> > >
> > > The green house is on the left of the white house.
> > >
> > > The green house's owner drinks coffee.
> > >
> > > The person who smokes Pall Mall rears birds.
> > >
> > > The owner of the yellow house smokes Dunhill.
> > >
> > > The man living in the center house drinks milk.
> > >
> > > The Norwegian lives in the first house.
> > >
> > > The man who smokes Blends lives next to the one who keeps cats.
> > >
> > > The man who keeps the horse lives next to the man who smokes Dunhill.
> > >
> > > The owner who smokes Bluemasters drinks beer.
> > >
> > > The German smokes Prince.
> > >
> > > The Norwegian lives next to the blue house.
> > >
> > > The man who smokes Blends has a neighbor who drinks water.
> > >
> >
> > I don't beleive he said that only 2% of the world could solve it. It's
not
> > hard, you just need to pay attention to the details. The German owns the
> > fish. It was supposed to be a table that I used to layout all the
details,
> > but I prefer to post plain-text:
> >
> > house         1 2 3 4 5
> > color            yellow blue red green white
> > nationality   norweigian dane brit german swede
> > drink            water tea milk coffee beer
> > smoke        dunhill blend pall mall prince bluemaster
> > pet              cats horse birds fish dogs
> >
> > -Boris
> > P.S. Did I have to write software to solve it ? :)




From sulfugor at rocketmail.cum  Mon Mar 12 01:56:25 2001
From: sulfugor at rocketmail.cum (sulfugor)
Date: Mon, 12 Mar 2001 06:56:25 GMT
Subject: Einstein's Riddle
References: <oNPq6.396644$ge4.138479566@news2.rdc2.tx.home.com> <UdSq6.2184$54.2166@www.newsranger.com>
Message-ID: <984380185.768920@news.aei.ca>

Gregory Jorgensen wrote in message ...
>The 2% that can solve it must be psychic or deranged. The riddle as posed
is not
>solvable.
>
>We are told of five houses anf five people. The hints mention five colors
(red,
>green, white, yellow, blue), five drinks (tea, coffee, milk, beer, water),
five
>smokes (Pall Mall, Dunhill, Blends, Bluemaster, Prince), and five
nationalities
>(Brit, Swede, Dane, Norwegian, German). But only four types of pet are
>mentioned: dogs, birds, cats, horse. We have no evidence that any of the
five
>owns fish.


But the fish is mentionned in the question before the clues are given. We
are able to find who owns each of the 4 pets mentionned in the hint list. is
it so unreasonable to assume then that the last pet is a fish, since we were
asked about it ? The question IMplied that one of the 5 people owned a fish.

sulfugor

ps = by the way, how long is this supposed to take ( solving the riddle) ?



>
>More telling: how many people can't quickly determine that this puzzle is
>unsolvable?
>
>
>In article <oNPq6.396644$ge4.138479566 at news2.rdc2.tx.home.com>, Steve
Graham
>says...
>>
>>Who would be interested in using his/her brain (and his computer) to solve
>>the following?
>>
>>
>>Steve Graham
>>
>>===
>>
>>Einstein's Riddle
>>
>>Albert Einstein wrote this riddle this century [ed. 20th century].  He
said
>>98% of the world could not solve it.
>>
>>There are 5 houses in 5 different colors.  In each house lives a person
with
>>a different nationality.  The 5 owners drink a certain type of beverage,
>>smoke a certain brand of cigar, and keep a certain pet.  No owners
>>have the same pet, smoke the same brand of cigar or drink the same
beverage.
>>
>>The question is: "Who owns the fish?"
>>
>>Hints:
>>
>>The Brit lives in the red house.
>>
>>The Sweed keeps dogs as pets.
>>
>>The Dane drinks tea.
>>
>>The green house is on the left of the white house.
>>
>>The green house's owner drinks coffee.
>>
>>The person who smokes Pall Mall rears birds.
>>
>>The owner of the yellow house smokes Dunhill.
>>
>>The man living in the center house drinks milk.
>>
>>The Norwegian lives in the first house.
>>
>>The man who smokes Blends lives next to the one who keeps cats.
>>
>>The man who keeps the horse lives next to the man who smokes Dunhill.
>>
>>The owner who smokes Bluemasters drinks beer.
>>
>>The German smokes Prince.
>>
>>The Norwegian lives next to the blue house.
>>
>>The man who smokes Blends has a neighbor who drinks water.
>
>Greg Jorgensen
>Deschooling Society
>Portland, Oregon, USA
>gregj at pobox.com




From paulp at ActiveState.com  Sat Mar 10 12:06:45 2001
From: paulp at ActiveState.com (Paul Prescod)
Date: Sat, 10 Mar 2001 09:06:45 -0800
Subject: import headaches/gripes
References: <3aaa3f94.15940553@News.CIS.DFN.DE>
Message-ID: <3AAA5F25.B900AA9F@ActiveState.com>

Costas Menico wrote:
> 
>...
> 
> Why can't the import statement be smart enough to check if a file was
> changed to import it anyway? All it has to do is compare .py to the
> last import date/time.

I think there are two reasons:

 1. Even checking file dates has a performance cost. It involves going
to the hard disk after all. Code like this isn't rare in Python

def foo():
   import bar
   ....

 2. What would happen if you built a big, long-running application and
then changed a file while it was running. Would you really want one
import statement in the file to load the new version? Then you would
have objects floating around from the old version alongside objects from
the new version. When it crashed due to the inconsistency, it would be
very confusing.

Pythonwin is one big long-running Python app. In that particular
situation you might prefer it to act differently but Python doesn't
really have a feature that allows what you want...and even if it did,
you'd have to be careful because you could get into the same situation
where you had old objects hanging around with the old behavior. That
happens even with a reload today.

> if module_exists(utils):
>         reload(utils)
> else:
>         import utils

You can make it simpler than that:

import utils
reload(utils)

The first operation will continue silently and quickly if the module is
loaded. Then it will be reloaded. The only "cost" is that the very first
time you would load the module twice in a row...

-- 
Python:
    Programming the way
    Guido
    indented it.



From tjreedy at udel.edu  Wed Mar 14 15:46:02 2001
From: tjreedy at udel.edu (Terry Reedy)
Date: Wed, 14 Mar 2001 15:46:02 -0500
Subject: This math scares me
References: <3aad4062.1290021@News.CIS.DFN.DE> <20010312170356.B3513@better.net> <mailman.984480856.18097.python-list@python.org> <984486395.407379@newsmaster-04.atnet.at> <slrn9asf1d.ja3.qrczak@qrnik.zagroda> <984518199.127140@newsmaster-04.atnet.at> <jwbnews-09D75F.13313313032001@corp.supernews.com> <98ohuc$du9$1@news.udel.edu> <7bQr6.644$lz1.17686@ruti.visi.com>
Message-ID: <98oku1$f6a$1@news.udel.edu>

"Grant Edwards" <grante at visi.com> wrote in message
news:7bQr6.644$lz1.17686 at ruti.visi.com...
> In article <98ohuc$du9$1 at news.udel.edu>, Terry Reedy wrote:
>
> >> Bill Gates got this right in the early days, delivering two Basic
> >> interpreters:  binary floating point for speed, and decimal floating
> >> point for money and such
> >
> >If one counts pennies instead of dollars, then 32 bit ints (+- $20
million)
> >will serve for many purposes.   Exact, faster, and more widely available
> >than BCD
>
> In fiancial calculations you also need to represent numbers
> that aren't amounts of money.  Interest rates for example.

So?  Floats (double precision) work well enough for this, I believe.

Note that binary fraction stock prices are exact with floats and not with
BCD.




From nas at python.ca  Tue Mar 27 13:02:57 2001
From: nas at python.ca (Neil Schemenauer)
Date: Tue, 27 Mar 2001 10:02:57 -0800
Subject: Eat this one
In-Reply-To: <3AC0B349.EE0AA9B2@hack.org>; from bos@hack.org on Tue, Mar 27, 2001 at 05:35:37PM +0200
References: <3AC0B349.EE0AA9B2@hack.org>
Message-ID: <20010327100257.A18517@glacier.fnational.com>

Rikard Bosnjakovic wrote:
> I feel ashamed to show the code, but I didn't find any other solution :)

That's okay.  The code you have is pretty bad though (think about
the complexity of the algorithm).  Here is a solution which would
be much faster for large values of txt:

    import string

    _notrans = string.maketrans("", "") # don't translate any chars

    def remove_chars(txt, chars):
        return string.translate(txt, _notrans, chars)

    txt = cgi.escape(open("somefile").read())
    txt = remove_chars(txt, "\n<>&")

Cheers,

  Neil



From parz at home.com  Fri Mar  2 00:41:35 2001
From: parz at home.com (Parzival Herzog)
Date: Fri, 02 Mar 2001 05:41:35 GMT
Subject: "import X" and "from X import y" produce different results, Why?
References: <ifmn6.6635$hn5.857862@news1.rdc1.mb.home.com> <97l3e8027ho@news2.newsguy.com>
Message-ID: <joGn6.7530$hn5.943423@news1.rdc1.mb.home.com>

Thanks guys!, I guess I got caught by the word "IDENTICAL" in the Essential Reference,
and I didn't see the implication of the words "creates a reference".

"from x import y" is anything BUT identical to plain import!

humbly-acknowledging-the-higher-pythonic-wisdom-of-those-who-know-about-effbot.org-ly
    - Parzival

"Alex Martelli" <aleaxit at yahoo.com> wrote in message news:97l3e8027ho at news2.newsguy.com...

> Right.  I.e., it's identical, except that it isn't:-).





From dbrueck at edgix.com  Mon Mar 26 09:19:56 2001
From: dbrueck at edgix.com (Dave Brueck)
Date: Mon, 26 Mar 2001 07:19:56 -0700
Subject: python 2.0 rpm for redhat 7.0?
Message-ID: <055701c0b5ff$e3de2b90$450514ac@PRODUT2KDAVE>

> The BeOpen rpm for py2.0 complains of missing libraries. I decided to
> make py2.0 from source (normally a trivial exercise). It complains that
> gcc appears to be broken. I install the many update rpms for gcc from
> redhat. It's STILL broken....

Hi Bob,

RH 7.0's glibc is broken. After you install the glibc updates off redhat.com, building 2.0 becomes a trivial exercise once more.

-Dave

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20010326/f6c2bf22/attachment.html>

From jos at trapeze.removespam.com  Fri Mar  2 09:33:42 2001
From: jos at trapeze.removespam.com (Digital Hyakugei)
Date: Fri, 2 Mar 2001 09:33:42 -0500
Subject: WDDX in Python. 
Message-ID: <3a9faea1$1@news1>

Hello again. I've scoured the python.org site, used Google and have only had
hints that there _might_ be a WDDX parser/encoder for Python... If anyone
has a more concrete example, source code, link, anything, i'd really
appreciate it!
Thanks!
jos





From michael at stroeder.com  Fri Mar 23 09:42:44 2001
From: michael at stroeder.com (Michael =?iso-8859-1?Q?Str=F6der?=)
Date: Fri, 23 Mar 2001 15:42:44 +0100
Subject: Python Module for Server-side Includes (SSI)
Message-ID: <3ABB60E4.7F376C1E@stroeder.com>

HI!

Does anybody know of a Python module for processing server-side
includes (SSI) as described on
http://httpd.apache.org/docs/howto/ssi.html?

If I'm going to write my own SSI processor which module to use?
htmllib or sgmllib?

Ciao, Michael.


From db3l at fitlinxx.com  Thu Mar  8 22:26:15 2001
From: db3l at fitlinxx.com (David Bolen)
Date: 08 Mar 2001 22:26:15 -0500
Subject: grabbing return codes from os.system() call
References: <%wFp6.9481$dL4.126404@vixen.cso.uiuc.edu> <x8Gp6.899$54.943@www.newsranger.com> <984083439.472914@clam-ext>
Message-ID: <ud7brk4h4.fsf@ctwd0143.fitlinxx.com>

"Mark Hadfield" <m.hadfield at niwa.cri.nz> writes:

> "Gregory Jorgensen" <gregj at pobox.com> wrote in message
> news:x8Gp6.899$54.943 at www.newsranger.com...
> > os.system returns the exit status. ... on Windows the
> > exit status is always 0.
> 
> Not so.
> 
> On my Win 2000 box with Python 2.0, successful commands return 0 and
> unsuccessful commands (where the command is not found) return 1.
> 
> Of course there may be more to it, but that's what I found in 20 s of
> experimentation...

The problem is that in order to properly return an exit code of the
child process, the shell in the middle has to pass that information on
up.  This is because os.system() will actually start up a child shell
(e.g., something like "/bin/sh" under Unix or typically "command.com"
or "cmd.exe" under Windows), so the actual command being executed is
one layer removed from the Python script.

This is much more reliable under Unix than in my experience under
various command interpreters under various Windows versions, although
NT (I don't yet use 2K consistently) has been reasonably solid.

I've had decent luck with Windows NT and 2K command interpreters
passing up result codes (and the interpreter will generally return 1
if it can't find the command to execute), but recall problems in old
times with Windows 95 command.com, so if that's what the original
poster had, it may account for the consistent 0 return.

It can also be impacted if the command you are going to run is 16-bit,
since that involves other machinery on each Windows version that often
loses the result code along the way.  All in all a pain, and best
tested in the environment in which its needed.

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


From kkoller at newfocus.com  Mon Mar  5 23:20:35 2001
From: kkoller at newfocus.com (Ken Koller)
Date: Mon, 05 Mar 2001 20:20:35 -0800
Subject: National Intruments GPIB
Message-ID: <3AA46593.753BAE1@newfocus.com>

Has anyone written a python interface to some of the National
Instruments libraries? I'm in the process of using Python to connect to
some instrumentation and I'm writing the C code to interface to their
GPIB API. Was hoping someone else had done it :)

Thanks,

Ken



From lcortes at aodinc.com  Sun Mar 18 17:02:56 2001
From: lcortes at aodinc.com (Luis Cortes)
Date: Sun, 18 Mar 2001 15:02:56 -0700
Subject: Help with new scope problem
Message-ID: <993bar$7h5$1@news.chatlink.com>

Hello,

I have a piece of code that I think should work fine, but I believe that
I've hit a scope problem (Python 2.1b1).  Does anyone out there have a hint
as to how to fix it??

code:

def getarguments():

        box = ( 160, 120 )
        newgifname = 'new.gif'
        gifname = ''

        if (len(sys.argv) < 4 ):
                print 'usage: trans.py width height image.xxx newimage.gif'
                sys.exit(0)

        if (len(sys.argv) > 4):
                # program width, height, image.xxx, newimage.gif
                box = ( int(sys.argv[1]), int(sys.argv[2]) )
                gifname = sys.argv[3]

        if (len(sys.argv) > 5 ):
                newgifname = sys.argv[4]

        print box, gifname, newgifname
        sys.exit(0)
        return ( box, gifname, newgifname )

THE PROBLEM:  when I have more than 3 arguments, the global variables do not
change to the correct variables, but instead print their default values.
Help!





From skip at pobox.com  Tue Mar 13 19:02:30 2001
From: skip at pobox.com (Skip Montanaro)
Date: Tue, 13 Mar 2001 18:02:30 -0600
Subject: How do I get Win32 to locate DLLs elsewhere (not really a Python thing)?
Message-ID: <3AAEB516.3040205@pobox.com>

I'm trying for the first time to use Python (2.0) on Win98 with external 
DLLs (GTK/GDK/...).  I installed all the various bits and pieces for 
GIMP/GTK and PyGTK, but three DLLs that come with GIMP/GTK are not where 
Windows is searching for them, so I get import errors.  (GIMP runs 
fine.) Using a tool called the dependency walker (something from MS) I 
am able to extend its module search directories to include C:\Program 
Files\Common Files\GNU so it will resolve those DLLS.

How do I perform that trick so that Microsoft's equivalent of ld.so can 
find them at run-time?  I assume there is some key in the registry I can 
set, but I haven't stumbled on anything obvious so far.  I did extend 
the registry key HKEY_LOCAL_MACHINE\Software\Python\PythonCore\2.0 
\PythonPath with the above directory, but that had no discernable 
effect.  I also checked to see what GIMP does.  It has a Path value 
associated with gimp.exe that refers to the DLL-containing directory, so 
I tried adding a string named Path with C:\Program Files\Common 
Files\GNU as its value to python.exe.  That didn't work either.  (Is 
this the equivalent of a Unix environment variable?) Is this another one 
of those "you must reboot for the changes to take effect" things?

Thanks,

unix-weenie-ly y'rs,

Skip Montanaro
skip at pobox.com




From ppvisser at hotmail.com  Mon Mar  5 09:42:36 2001
From: ppvisser at hotmail.com (patrick)
Date: Mon, 5 Mar 2001 15:42:36 +0100
Subject: help
Message-ID: <3aa3a6cd$0$16795@reader2>

I need a dutch tutorial for python 2.1
Please send it to ppvisser at hotmail.com
or just give me an download link or a good site




From fredrik at pythonware.com  Wed Mar 21 10:29:37 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 21 Mar 2001 15:29:37 GMT
Subject: re.match
References: <mailman.985169080.17724.python-list@python.org>
Message-ID: <BN3u6.894$sk3.291185@newsb.telia.net>

Oleg Broytmann wrote:
> But I, personnaly, still do not know Dylan outside of TW :)

...just like many people still don't know that Ken Thompson did
regular expression stuff before he started hacking on Unix ;-)

(and others did regular expression stuff before him, of course)

Cheers /F




From fredrik at pythonware.com  Sat Mar 24 06:32:05 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Sat, 24 Mar 2001 12:32:05 +0100
Subject: [Python-Dev] Anyone using weakrefs?
References: <15035.46175.599654.851399@localhost.localdomain>
Message-ID: <018e01c0b456$6710b830$e46940d5@hagrid>

Fred wrote:
>   Is anyone out there playing with the weak references support yet?
> I'd *really* appreciate receiving a short snippet of non-contrived
> code that makes use of weak references to use in the documentation.

am I the only one who gets a bit nervous when the guy who
designed and implemented a feature has a hard time finding a
non-contrived use case? ;-)

Cheers /F




From cce at clarkevans.com  Wed Mar 21 22:14:54 2001
From: cce at clarkevans.com (Clark C. Evans)
Date: Wed, 21 Mar 2001 22:14:54 -0500 (EST)
Subject: Yet Another PEP:  Interface Adapter Mechanism __adapt__
Message-ID: <Pine.LNX.4.21.0103212211320.11584-100000@clarkevans.com>

On Wed, 21 Mar 2001, Rainer Deyke wrote:
> This 'check_base' stuff is a mess. 

No joke.  Sorry for being so brain dead.  The following
should much better illustrate the concept... (had two bugs 
that are now fixed).

...

Here is the "merged" version where check_only is an
optional argument to adapt.  Imagine the operator,

  a isa b -> adapt(a,b,check_only=true)

--------------------------------------------------------
adapter/__init__.py
--------------------------------------------------------

import types
def adapt(obj,ident,check_only=0):

    # check to see if the current object is ok as is

    if obj is ident: return obj
    if type(obj) is ident: return obj
    if type(obj) is type(ident): return obj
    if type(ident) is types.ClassType:
        if type(obj) is types.InstanceType:
            if isinstance(obj,ident): return obj
        if type(obj) is types.ClassType:
            if issubclass(obj,ident): return obj

       # if we get a new interface type, code should
       # be added in this spot to check if obj is an 
       # instance of ident, where ident is an types.Interface
       # *THIS IS ANOTHER PEP*

    # the obj may have the answer, so ask it about the ident
    

    adapt = getattr(obj, '__adapt__',None)
    if adapt:
        retval = adapt(ident,check_only)
        if retval: return retval

    # the ident may have the answer, so ask it about the obj

       # some code here for a class, type, or interface,
       # or whatever to check to see if it can adapt the
       # class.  Unfortunately, calling __adapt__ won't
       # work exactly here...  we could add another
       # argument, "forward"... *THIS IS ANOTHER PEP*
 
    # a registry may exist, ask it about adapters.

       # there may be a registry-based  solution 
       # adapters that can be added to the system without
       # modifying either A or B.  *THIS IS ANOTHER PEP*

--------------------------------------------------------------
adapter/example.py
--------------------------------------------------------------

class KnightsWhoSayNi: pass

class EggsOnly:  # an unrelated class/interface
    def eggs(self,str): print "eggs!" + str

class HamOnly:  # used as an interface, no inhertance
    def ham(self,str): pass
    def _bugger(self): pass  # irritating a private member

class SpamOnly: # a base class, inheritance used
    def spam(self,str): print "spam!" + str

class EggsSpamAndHam (SpamOnly):
    def ham(self,str): print "ham!" + str
    def __adapt__(self,ident,check_only):
        if ident is HamOnly:
            return self       # implements HamOnly implicitly, no _bugger
        if ident is EggsOnly and not check_only:
            return EggsOnly() # Knows how to create the eggs!

------------------------------------------------------------------
testadapter.py
------------------------------------------------------------------
import adapter.example
from adapter import adapt
def testadapter():
    x = adapter.example.EggsSpamAndHam()
    adapt(x,adapter.example.SpamOnly).spam("Ni!")
    adapt(x,adapter.example.EggsOnly).eggs("Ni!")
    adapt(x,adapter.example.HamOnly).ham("Ni!")
    adapt(x,adapter.example.EggsSpamAndHam).ham("Ni!")
    #
    # imagine operator syntax for check_only:
    #
    #   if x isa adapter.example.SpamOnly: print "SpamOnly"
    #
    if adapt(x,adapter.example.SpamOnly,1): print "SpamOnly"
    if adapt(x,adapter.example.EggsOnly,1): print "EggsOnly"
    if adapt(x,adapter.example.HamOnly,1): print "HamOnly"
    if adapt(x,adapter.example.EggsSpamAndHam,1): print "EggsAndSpam"
    if adapt(x,adapter.example.KnightsWhoSayNi,1): print "NightsWhoSayNi"
    adapt(x,adapter.example.KnightsWhoSayNi,1).spam("Ni!")


--------------------------------------------------------------------
example run
--------------------------------------------------------------------

>>> import testadapter
>>> testadapter.testadapter()
spam!Ni!
eggs!Ni!
ham!Ni!
ham!Ni!
SpamOnly
HamOnly
EggsAndSpam
Traceback (innermost last):
  File "<interactive input>", line 1, in ?
  File "c:\work\testadapter.py", line 14, in testadapter
    adapt(x,adapter.example.KnightsWhoSayNi,1).spam("Ni!")
AttributeError: 'None' object has no attribute 'spam'





From cfelling at iae.nl  Fri Mar  9 17:44:00 2001
From: cfelling at iae.nl (Carel Fellinger)
Date: 9 Mar 2001 23:44:00 +0100
Subject: Dictionary attributes and functions
References: <Pine.LNX.4.21.0103092014190.19640-100000@moroni.pp.asu.edu>
Message-ID: <98bmbg$61b$1@animus.fel.iae.nl>

iddwb <iddwb at imap1.asu.edu> wrote:
...
> attributes that can be reference regarding dictionaries... specifically if
> key(a) is already in dictionsary x I'd like to know prior to
> adding.  rather than change the value of key(a) I'd like to add the value
> of key(`a) to the value of key(a) so I end up with

> {keya: [123]}
> on insert of keya againg have something like
> {keya: [123, 234]}

in python 2.0 (or was this already in 1.6?) and newer:

   d = dict.setdefault(key, [])
   d.append(value)

in older versions still around:

   if dict.has_key(key):
	dict[key].append(value)
   else:
      dict[key] = [value]
-- 
groetjes, carel


From othello at javanet.com  Mon Mar  5 11:20:01 2001
From: othello at javanet.com (Raymond Hettinger)
Date: Mon, 05 Mar 2001 11:20:01 -0500
Subject: Q: Operator precedence
References: <Pine.LNX.4.21.0103051051180.12138-100000@kev.ioc.ee>
Message-ID: <3AA3BCB1.13CC1479@javanet.com>

Pearu Peterson wrote:

> Section 5.12 in Python reference manual summarizes operator precedence.
>
> I would like to have the following function:
>
> def getprec(s):
>      return <the precedence number of a keyword/operator `s'>

Here is an approach to rank ordering of precedence at run-itme:

def higherOrEqual( op1, op2 ):
    return eval('51 %s 83 %s 97 == ( 51 %s 83 ) %s 97' % (op1,op2,op1,op2))

order = ['Lower', 'HigherOrEqual']

binops = ['+', '-', '&', '^', '/']

for x in binops:
    for y in binops:
        print x, order[higherOrEqual(x,y)], y


Raymond




From gahanr at gofree.indigo.ie  Wed Mar  7 14:42:17 2001
From: gahanr at gofree.indigo.ie (Robert Gahan)
Date: Wed, 7 Mar 2001 19:42:17 -0000
Subject: Remote script execution with Python ...
References: <7Ybp6.7729$r4.7336@news.indigo.ie> <ulmqil8qu.fsf@roundpoint.com>
Message-ID: <Lpwp6.8059$r4.7712@news.indigo.ie>

Thanks Ben.

os.popen2 is not available to windows programmers but
win32pipe came to the rescue !!. win32pipe I think is available
from the following location
http://www.activestate.com/Products/ActivePython/win32all.html

Got something like the following to work ....

import win32pipe

stdin, stdout = win32pipe.popen2("rsh <hostname> -l <username>
<scriptname>", "t")
stdin.write("<whatever you want to write to stdin !!>")
print stdout.readlines() # This will print the result.

Regards,
Robert.

Ben Hutchings <ben.hutchings at roundpoint.com> wrote in message
news:ulmqil8qu.fsf at roundpoint.com...
> "Robert Gahan" <gahanr at gofree.indigo.ie> writes:
>
> > Hi there,
> >
> > I have installed python on a Windows PC and I am looking to
> > remotely execute a script on a UNIX box from a python program.
> > I need to be able to pipe a "y" to stdin also. Is there a windows
> > equivalent of the unix "rexec" that could help me ?. What is the
> > most appropriate way of doing this ?.
>
> You might be able to do this with something like:
>
>     script_in, script_out = os.popen2("rsh otherhost -c myscript")
>     print >>script_in 'y'
>     result = script_out.readlines()
>
> Windows 2000 includes an rsh client but I'm not sure about other
> versions of Windows.  You might want to use ssh instead for greater
> security.
>
> --
> Any opinions expressed are my own and not necessarily those of Roundpoint.




From jurgen.defurne at philips.com  Mon Mar 19 07:02:33 2001
From: jurgen.defurne at philips.com (jurgen.defurne at philips.com)
Date: Mon, 19 Mar 2001 13:02:33 +0100
Subject: More accurate summations [was Re: This math scares me]
Message-ID: <0056900016700784000002L042*@MHS>

grante at visi.com@SMTP at python.org on 17/03/2001 17:42:57
Sent by:	python-list-admin at python.org
To:	python-list at python.org@SMTP
cc:	 
Subject:	Re: More accurate summations [was Re: This math scares me]
Classification:	


On 17 Mar 2001 03:43:57 GMT, Edward Jason Riedy <ejr at cs.berkeley.edu> wrote:
>And <jurgen.defurne at philips.com> writes:
> -
> - Try summing all financial operations of one day together, or
> - better, all operations involving percentages and then adding them
> - together : cumulative errors.
>
>More concretely, look at the Vancouver Stock Exchange index's
>problems of the late 70s-early 80s.  They had chosen to
>truncate floating point values rather than round them, causing
>a downward drift of around 20 points a month[1].
>
>You could say that's because of poor handling of floating-point
>arithmetic, and you'd be right.  However, just about every
>inaccuracy in floating-point financial calculations can be
>attributed to poor handling of fp arithmetic.

The problem is that proper handling of fp arithmetic is not
easy even though it appears trivial to the uneducated. It
requires quite a bit of knowlege, thought, and work. After one
undergrad numerical analysis class, I know that I don't know
enough to do it right.

>That doesn't mean floating point is a poor choice for all
>financial programs.  Indeed, the HP financial calculators _all_
>use floating point, although decimal.

That was (I think) the original topic: BCD floating point vs.
binary floating point.

--
Grant Edwards                   grante             Yow!  Now KEN is having
                                  at               a MENTAL CRISIS beacuse
                               visi.com            his "R.V." PAYMENTS are
                                                   OVER-DUE!!
--
http://mail.python.org/mailman/listinfo/python-list

And that is why abovementioned financial operations ARE all carried out using BCD, mostly supported
by hardware, or optimized libraries.

BTW, I have never heard about a library which supports the x87 extensions for using 18 decimal BCD
math. This is based on the internal 80-bit FP format of the x87 numeric processors.

Jurgen



From fredrik at pythonware.com  Wed Mar 14 11:27:35 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 14 Mar 2001 16:27:35 GMT
Subject: How can I get the logged in users username on Win32?
References: <3aaf7ab4.3152221@news.is.co.za>
Message-ID: <XZMr6.262$4d3.54933@newsb.telia.net>

"nomad" wrote:
> After searching through the Python Docs, I found that on Un*x I could
> use getpass.getuser(), but there is no equivalent for the Win32
> platform.
>
> In other words, how can I get the username in Windows?

in case you really meant to post this again, here
are the responses you got two weeks ago:

fredrik:
> > In other words, how can I get the username in Windows?
>
> getpass.getuser()
>
> (it works as long as USERNAME is set, which it usually is)

mark:
> And win32api.GetUserName() should work even if it isnt :)

Cheers /F




From akuchlin at mems-exchange.org  Sat Mar 10 15:15:49 2001
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: Sat, 10 Mar 2001 15:15:49 -0500
Subject: IPC9 notes
Message-ID: <E14bpmH-00019X-00@ute.cnri.reston.va.us>

Continuing my usual tradition, my diary entries for
IPC9 at now available at http://www.amk.ca/python/writing/ipc9.html .

--amk



From renzo.tomaselli at tecnotp.it  Mon Mar 19 03:49:37 2001
From: renzo.tomaselli at tecnotp.it (Renzo Tomaselli)
Date: Mon, 19 Mar 2001 09:49:37 +0100
Subject: relational/boolean operator overloading
References: <98tjq6$qre$1@news.flashnet.it> <98tk4b$sbv$1@panix3.panix.com>
Message-ID: <994h71$ohq$1@news.flashnet.it>

Aahz Maruch wrote in message <98tk4b$sbv$1 at panix3.panix.com>...
>In article <98tjq6$qre$1 at news.flashnet.it>,
>Renzo Tomaselli <renzo.tomaselli at tecnotp.it> wrote:
>>
>Not sure what advice to give you with current Python, but Python 2.1
>introduces rich comparisons, which may give you what you want.  See
>PEP 207, http://python.sourceforge.net/peps/pep-0207.html

Pep 207 looks great for my posted topic. However its solves half of the
problem. The missing part is about overloading 'and'/'or' operators, also
because this PEP allows for comparisons returning any object. How to chain
such results with booleans, beside a trivial truth test ?
Say we have: "(sa > 10) and (sb < 100)". I would expect PEP overloading to
carry out relational filtering for the two sets and to return two result
sets, while some "and" overloading would connect them to return a final
result set.
Thanks,
                Renzo Tomaselli





From fredrik at pythonware.com  Wed Mar 14 13:29:23 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 14 Mar 2001 18:29:23 GMT
Subject: Dynamic construction of a print format string
References: <3AAFB663.57FCCE7@troikanetworks.com>
Message-ID: <7MOr6.281$4d3.61498@newsb.telia.net>

Bruce Edge wrote:
> I'm using:
>
>   fmtstr = "%s%s%s" % ( "    %", namelen, "s - %s" )
>
> to get:
>
>   "    %19s - %s"

how about:

    format = "    %%%ds - %%s" % namelen
    text = format % (arg1, arg2)

or

    text = "    %*s - %s" % (namelen, arg1, arg2)

Cheers /F




From costas at meezon.com  Tue Mar 20 23:48:13 2001
From: costas at meezon.com (Costas Menico)
Date: Wed, 21 Mar 2001 04:48:13 GMT
Subject: Saving history in interactive Python
References: <mailman.985113411.25785.python-list@python.org>
Message-ID: <3ab831a7.10008641@News.CIS.DFN.DE>

Timothy Grant <tjg at exceptionalminds.com> wrote:

>Is there a way to save the history list in the Python
>interpreter.

You can use IDLE or Pythonwin (Windows) to do that.

It is such a pain using the command line version of Python. If you
make a syntax error you have to retype the whole line. It would be
nice if there was something similar to DOSKEY where you use the up
arrow and view/edit prior lines. 

Costas



From KatMouse at gmx.de  Sun Mar 18 05:58:04 2001
From: KatMouse at gmx.de (Eduard Hiti)
Date: Sun, 18 Mar 2001 11:58:04 +0100
Subject: windows script host and python
References: <%tKr6.2570$pX.26174@news1.onlynews.com> <98oq3r$1e0$1@troll.powertech.no> <8Aps6.104$pl.1515@newsfeed.slurp.net> <98tc4d0rmk@news1.newsguy.com> <lvf8bt4p6e6naq0vimgnkhs175eoqbajfr@4ax.com>
Message-ID: <9924ea$4js$07$1@news.t-online.com>

I think WMI support can be added to Win98 with a download from microsoft.
Search for the WMI SDK on www.microsoft.com/downloads

"Tim Roberts" <timr at probo.com> wrote in message
news:lvf8bt4p6e6naq0vimgnkhs175eoqbajfr at 4ax.com...
| "Alex Martelli" <aleaxit at yahoo.com> wrote:
| >
| >Here, for example, is the "list all running processes" WMI scripting
| >example translated into Python:
| >
| >from win32com.client import GetObject
| >
| >processes = GetObject("winmgmts:{impersonationLevel=impersonate}").\
| >    InstancesOf("Win32_process")
| >
| >for process in processes:
| >    print process.Name
|
| Cool!  Is this supposed to be NT-only?  It doesn't fly on my Win98 box.
| --
| - Tim Roberts, timr at probo.com
|   Providenza & Boekelheide, Inc.




From pj at sgi.com  Fri Mar 16 14:09:06 2001
From: pj at sgi.com (Paul Jackson)
Date: 16 Mar 2001 19:09:06 GMT
Subject: How do I avoid the Tkinter 30 msec polling loop.
Message-ID: <98toci$ktjh$1@fido.engr.sgi.com>

I have a little 'biff' (BSD-speak for email watchdog) like
program, written in Python, using Tkinter, that shows one of
two text messages in a little window, depending on whether my
email box has new mail in it.  The app polls the mailbox every
minute, with an os.stat() call.

Unfortunately, the app consumes cpu almost continuously,
even when I wish it would sleep for the minute.

The same busy loop can be see even on the smallest "Hello,
world" example, such as:

    import Tkinter
    root = Tkinter.Tk()
    label = Tkinter.Label(root, text = "Hello, world.")
    label.pack()
    root.mainloop()

A system call trace of the above, on Irix 6.5.11, using

    Python 2.1a1 (#23, Feb  2 2001, 17:25:50) [C] on irix6,

shows the following 2 select() system calls, every 30 msec,
non-stop:

 hello.py( 3693): select(4, [3], [<empty>], [<empty>], {sec=0, usec=0})
 hello.py( 3693): select(0, 0, 0, 0, {sec=0, usec=20000})

This is on a system with a 10 msec TICK, and a timeout on
the select of 20 msec, which ends up taking 30 msec each
loop (timeouts are rounded up to the next TICK).

===

Anyhow, is there anyway to stop or radically slow this loop?

The little window responds to no key or mouse input.  It just
changes the text message when Label.configure (text="you have
mail") or some such is executed after the once a minute thread
wakes up and notices a change in the mail box.

                          I won't rest till it's the best ...
                          Manager, Linux System Software
                          Paul Jackson <pj at sgi.com> 1.650.933.1373
-- 
                          I won't rest till it's the best ...
                          Manager, Linux System Software
                          Paul Jackson <pj at sgi.com> 1.650.933.1373


From jacobus at geo.co.za  Tue Mar 27 07:51:08 2001
From: jacobus at geo.co.za (Jacobus van der Merwe)
Date: Tue, 27 Mar 2001 14:51:08 +0200
Subject: Reading e-mail body (newbie)
Message-ID: <3AC08CBB.52061537@geo.co.za>

Hi,

I have a script to which e-mail to a certain address is relayed.

I need to write the body of the e-mail to a file.

I use

rfc822.Message(sys.stdin)

to read the e-mail, but I can't figure out how to access the body.
It seems that rfc822 only have methods to access the headers. ????

Any help will be appreciated.

Jacobus





From dyoo at hkn.eecs.berkeley.edu  Mon Mar 12 06:36:15 2001
From: dyoo at hkn.eecs.berkeley.edu (Daniel Yoo)
Date: Mon, 12 Mar 2001 03:36:15 -0800 (PST)
Subject: true newbie
Message-ID: <Pine.LNX.4.21.0103120331530.13241-100000@hkn.eecs.berkeley.edu>

> can anyone help me find a tutorial for python thats for people just
> starting out in programming. The one one that comes with it is ok but
> its still too difficult.

Also, you might want to subscribe with tutor at python.org: we're a group of
tutors and tutees that help each other learn Python:

    http://mail.python.org/mailman/listinfo/tutor

... join us... join us...

*grin*  Good luck to you.




From loewis at informatik.hu-berlin.de  Sat Mar 24 06:47:26 2001
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 24 Mar 2001 12:47:26 +0100
Subject: How 2 get rid of console
References: <tborq16qctlm42@corp.supernews.com>
Message-ID: <j4r8znmlrl.fsf@informatik.hu-berlin.de>

"Bill Walker" <bwalker at earth1.net> writes:

> I'm using Tkinter to build a GUI application. How do I get rid of the
> distracting dos console under the Tk window?

Rename your main program script into .pyw. It will be run by
pythonw.exe, which won't open a console window.

Regards,
Martin


From rambo at linuxsecurity.co.kr  Fri Mar 16 12:42:56 2001
From: rambo at linuxsecurity.co.kr (rambo)
Date: Fri, 16 Mar 2001 09:42:56 -0800
Subject: How can install Tkinter on linux.
Message-ID: <98rnem$465$2@news2-2.kornet.net>

Hello,

How can I install Tkinter on linux.
I download Tkinter-2.0-8.0.tar.gz, and extract it ( tar xvzf
Tkinter-2.0-8.0.tar.gz  )

    Tkinter-2.0-8.0/
    Tkinter-2.0-8.0/src/
    Tkinter-2.0-8.0/src/_tkinter.c
    Tkinter-2.0-8.0/src/tkappinit.c
    Tkinter-2.0-8.0/setup.cfg
    Tkinter-2.0-8.0/setup.py

 to use Tkinter on linux, What should I do?
and how to execute setup.py on linux prompt?

thank you for your time.
    hava a nice day.







From greg at cosc.canterbury.ac.nz  Thu Mar  1 21:50:21 2001
From: greg at cosc.canterbury.ac.nz (Greg Ewing)
Date: Fri, 02 Mar 2001 15:50:21 +1300
Subject: PEP 227 (was Re: Nested scopes resolution -- you can breathe again!)
References: <LNBBLJKPBEHFEDALKOLCMEEPJBAA.tim.one@home.com>
		<200102230259.VAA19238@cj20424-a.reston1.va.home.com>
		<mailman.983384008.21434.python-list@python.org>
		<97lqsr0311d@news2.newsguy.com> <mailman.983466367.2968.python-list@python.org>
Message-ID: <3A9F0A6D.49278A8E@cosc.canterbury.ac.nz>

Jeremy Hylton wrote:
> 
>             global x in spam, z in eggs

Hee, hee! I like it! Now we would have a way for
people to explicitly declare all variables if they
wanted, if we let it work on the local scope as
well:

  def spam():
    global x in spam, y in spam
    ...

or, to make that a little shorter,

    global (x, y) in spam

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


From ws-news at gmx.at  Tue Mar  6 12:29:38 2001
From: ws-news at gmx.at (Werner Schiendl)
Date: Tue, 6 Mar 2001 18:29:38 +0100
Subject: Pickling arbitrary character strings on Windows
References: <3AA51D2D.3251921E@lbl.gov>
Message-ID: <983900055.132778@newsmaster-04.atnet.at>

Is there a particular reason why you open the files in ASCII mode (text
mode)??
This will convert linefeeds to CRLF and use Ctrl-Z as a file delimiter (on
DOS and compatibles)

To work around that, use 'wb' instead of 'w' and 'rb' instead of 'r' as the
file mode.

This will do the work

hth
werner

Nicholas K. Sauter <nksauter at lbl.gov> wrote in message
news:3AA51D2D.3251921E at lbl.gov...
> It would be good to be able to pickle strings containing arbitrary ASCII
> characters, for example to encode data at the C level.  But on Windows
> it is impossible to do a binary unpickle of a string that has an
> embedded ASCII 26 (octal 032).  This example code tells the whole story:
>
> from pickle import dump,load
>
> def myfunc(i):
>     pro = "I_am_a_" +chr(i)+ "string"
>     f = open('try.bin','w'); dump(pro,f,1); f.close()
>     g = open('try.bin','r'); str = g.readline()# i=10 is a linefeed
> (that's OK)
>     print "read string",i,len(str),"characters"
>
> if __name__=="__main__":
>     for i in xrange(256):
>       myfunc(i)
>
> Some obvious workarounds are to use formatted output (dump(pro,f,0)) or
> to use Unix instead of Windows.  But is there any way to implement a
> practical binary string pickle on Windows?
>
> Nick Sauter, LBNL
>




From pj at sgi.com  Fri Mar 16 00:10:25 2001
From: pj at sgi.com (Paul Jackson)
Date: 16 Mar 2001 05:10:25 GMT
Subject: How do I avoid the Tkinter 30 msec polling loop.
Message-ID: <98s781$7a8o$1@fido.engr.sgi.com>

 [Sorry if this is a repeat - I don't see my post from yesterday. -pj]

I have a little 'biff' (BSD-speak for email watchdog) like
program, written in Python, using Tkinter, that shows one of
two text messages in a little window, depending on whether my
email box has new mail in it.  The app polls the mailbox every
minute, with an os.stat() call.

Unfortunately, the app consumes cpu almost continuously,
even when I wish it would sleep for the minute.

The same busy loop can be see even on the smallest "Hello,
world" example, such as:

    import Tkinter
    root = Tkinter.Tk()
    label = Tkinter.Label(root, text = "Hello, world.")
    label.pack()
    root.mainloop()

A system call trace of the above, on Irix 6.5.11, using

    Python 2.1a1 (#23, Feb  2 2001, 17:25:50) [C] on irix6,

shows the following 2 select() system calls, every 30 msec,
non-stop:

 hello.py( 3693): select(4, [3], [<empty>], [<empty>], {sec=0, usec=0})
 hello.py( 3693): select(0, 0, 0, 0, {sec=0, usec=20000})

This is on a system with a 10 msec TICK, and a timeout on
the select of 20 msec, which ends up taking 30 msec each
loop (timeouts are rounded up to the next TICK).

===

Anyhow, is there anyway to stop or radically slow this loop?

The little window responds to no key or mouse input.  It just
changes the text message when Label.configure (text="you have
mail") or some such is executed after the once a minute thread
wakes up and notices a change in the mail box.
-- 
                          I won't rest till it's the best ...
                          Manager, Linux System Software
                          Paul Jackson <pj at sgi.com> 1.650.933.1373


From noen at none.com  Wed Mar 28 16:30:00 2001
From: noen at none.com (Wong Da Qiang)
Date: Thu, 29 Mar 2001 05:30:00 +0800
Subject: Recommendations for Python Books
Message-ID: <3ac2582c$1@news.starhub.net.sg>

Just a newbie to programming and python, need some recommendations on books
to buy




From moshez at zadka.site.co.il  Sat Mar 31 16:03:15 2001
From: moshez at zadka.site.co.il (Moshe Zadka)
Date: Sat, 31 Mar 2001 23:03:15 +0200
Subject: Know of Substantial Apps Written in Python?
In-Reply-To: <Pine.SOL.4.30.0103312050260.15965-100000@mimosa.csv.warwick.ac.uk>
References: <Pine.SOL.4.30.0103312050260.15965-100000@mimosa.csv.warwick.ac.uk>, <3ac4e4a8_1@news4.newsfeeds.com> <KE6x6.415$SR6.235983@typhoon1.ba-dsg.net> <3ac4ff12$1_2@news4.newsfeeds.com> <mailman.986048021.30049.python-list@python.org>
Message-ID: <E14jSWh-0003L4-00@darjeeling>

On Sat, 31 Mar 2001, "John J. Lee" <phrxy at csv.warwick.ac.uk> wrote:
 
> This reminds me: why aren't the Redhat (or any other linux / unix vendor)
> boot scripts in Python, too?  Whenever I've made small changes to them,
> I've wished they were!  I have seen it stated that 'obviously' boot
> scripts can't be in Python, but nobody has explained to me why this is.
> How much bigger is a minimal Python interpreter than bash?

Much bigger considering that "minimal" is worthless -- without Python's
batteries, Python wouldn't be the same. What I wonder is why nobody
writes a system in which the boot levels are arranged so *most* functionality
can be used in *most* boot scripts.

> I'm doubtful that there are any advantages to shell, even for short
> scripts.

If the scripts are well written, they can run under shells much smaller
then bash, like ash for example.
Python would never fit on a floppy.
-- 
"I'll be ex-DPL soon anyway so I'm        |LUKE: Is Perl better than Python?
looking for someplace else to grab power."|YODA: No...no... no. Quicker,
   -- Wichert Akkerman (on debian-private)|      easier, more seductive.
For public key, finger moshez at debian.org  |http://www.{python,debian,gnu}.org



From geek+ at cmu.edu  Tue Mar  6 13:43:54 2001
From: geek+ at cmu.edu (Brian E Gallew)
Date: Tue, 6 Mar 2001 13:43:54 -0500
Subject: Ingres module?
Message-ID: <200103061843.NAA13897@smtp2.andrew.cmu.edu>

I noticed (from searching the web site) that Jim Fulton had at one
time talked about an Ingres module.  Has anyone actually seen such a
beast?  Even parts of one?  I'm not actually looking forward to
writing this myself, though I guess I will if I have to.

-- 
=====================================================================
| JAVA must have been developed in the wilds of West Virginia.      |
| After all, why else would it support only single inheritance??    |
=====================================================================
| Finger geek at cmu.edu for my public key.                            |
=====================================================================



From dwig at advancedmp.net  Wed Mar  7 19:00:08 2001
From: dwig at advancedmp.net (Don Dwiggins)
Date: 07 Mar 2001 16:00:08 -0800
Subject: Static typing (was Re: Java guy interested in Python)
In-Reply-To: qrczak@knm.org.pl's message of "5 Mar 2001 16:49:57 GMT"
References: <003801c0a459$a29c97f0$63da1542@wisdom>
	<mailman.983688785.707.python-list@python.org>
	<slrn9a7gtl.ca6.qrczak@qrnik.zagroda>
Message-ID: <ud7btyvsn.fsf_-_@advancedmp.net>

Marcin Kowalczyk writes:
> I use static typing to protect myself from errors, and to define
> interfaces, and to dispatch polymorphic code (on types rather than
> values), and to think about a piece of code before I write it. So it
> definitely is a purpose.

...

> Neither static typing nor dynamic typing is better suited than the
> other for all purposes. Both approaches have very strong good reasons.
>> From this it follows that static vs. dynamic typing flamewars will
> never end :-)

Well said.

So, are there any efforts in the Python community to provide something in
this area?  I'm thinking of things like type inference, and of something
like "permissive typing" (compiler/interpreter to programmer: "you don't have
to tell me the type of a variable or argument, but if you do, I'll use that
information to help you wherever I can").

-- 
Don Dwiggins                    "Solvitur Ambulando"
Advanced MP Technology
dwig at advancedmp.net




From scarblac at pino.selwerd.nl  Wed Mar 28 06:44:39 2001
From: scarblac at pino.selwerd.nl (Remco Gerlich)
Date: 28 Mar 2001 11:44:39 GMT
Subject: Getting all the *files* from a directory -- A better way??
References: <mailman.985732718.3725.python-list@python.org> <99r6v7$4hef2$1@fido.engr.sgi.com>
Message-ID: <slrn9c3jf1.qb3.scarblac@pino.selwerd.nl>

Paul Jackson <pj at sgi.com> wrote in comp.lang.python:
 > Close - better to use os.path.join (more portable across
> operating environments with different path component
> separator characters, such as '\' in DOS/Windows).
> 
> Try:
> 
>     dir="a/b/c"
>     files = filter(
> 	os.path.isfile,
> 	[os.path.join(dir, f) for f in os.listdir(dir)]
>     )

Or even

dir="a/b/c"
files = [os.path.join(dir, f) for f in os.listdir(dir)
                              if os.path.isfile(f)]

If you do the map() with a list comprehension, why not the filter() as well.

I don't know if this is the Pythonic way to do it, but it comes pretty close
to the Haskellic way :).

-- 
Remco Gerlich


From fredrik at pythonware.com  Wed Mar 21 10:29:38 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 21 Mar 2001 15:29:38 GMT
Subject: New PEP: The directive statement
References: <mailman.985117312.5806.python-list@python.org><j4u24oi4e8.fsf@informatik.hu-berlin.de><yHPt6.2553$4N4.334967@newsc.telia.net><998hv2$5ub$1@panix3.panix.com><oOQt6.2576$4N4.342378@newsc.telia.net> <uhf0njqq5.fsf@cs.uu.nl>
Message-ID: <CN3u6.895$sk3.290814@newsb.telia.net>

Piet van Oostrum wrote:
> FL> import declares things, in all existing Python versions (hint: it's
> FL> an assignment statement)
>
> On the other hand, the effect of import is the same as the calling of the
> magical __import__ function, of which the definition says:
>
> This function is invoked by the import statement (not declaration).
> And it is done at runtime, not compiletime.

Yup, but import has compile-time behaviour too, like any
other assignment operator.

(and vice versa, the "from __future__" statement has run-
time behaviour as well...)

But this doesn't matter much; the from __future__ statement is
a hack designed to deal with a another hack (all the code jeremy
had to add to dumb down his new compiler).  All last-minute
attempts to replace that with something that appears to be
more future proof (but really isn't) falls under the good old X
windows adage:

    "the only thing worse than generalizing from one example is
    generalizing from no example at all"

Cheers /F




From olive at conectiva.com.br  Thu Mar 15 15:52:02 2001
From: olive at conectiva.com.br (Fabio Olive Leite)
Date: Thu, 15 Mar 2001 17:52:02 -0300
Subject: This math scares me
In-Reply-To: <HR4s6.877$lz1.73512@ruti.visi.com>; from grante@visi.com on Thu, Mar 15, 2001 at 03:04:07PM +0000
References: <984486395.407379@newsmaster-04.atnet.at> <slrn9asf1d.ja3.qrczak@qrnik.zagroda> <984518199.127140@newsmaster-04.atnet.at> <jwbnews-09D75F.13313313032001@corp.supernews.com> <98ohuc$du9$1@news.udel.edu> <7bQr6.644$lz1.17686@ruti.visi.com> <98oku1$f6a$1@news.udel.edu> <erRr6.659$lz1.20662@ruti.visi.com> <mailman.984664520.26463.python-list@python.org> <HR4s6.877$lz1.73512@ruti.visi.com>
Message-ID: <20010315175202.L1757@conectiva.com.br>

On Thu, Mar 15, 2001 at 03:04:07PM +0000, Grant Edwards wrote:
) 
) One step further is base-60, which is still used for many
) things.  Time and angular measurement for example is partially
) a base-60 system.  I believe that one or more of the ancient
) Aztecs/Incans/Mayans used a base-60 number system.

Indeed, they used it. And some other people too, as is seen even today in
the most common languages of the word. French, for example, has no "real
words" for seventy, eighty and ninety, and uses "sixty ten", "four twenty"
and "four twenty ten".

Base 12 was used a lot for commerce, as it has more integer fractions than
10. :)

-- 
( F?bio Oliv? Leite -*- http://www.conectiva.com.br/~olive )



From gradha at iname.com  Thu Mar  1 16:18:11 2001
From: gradha at iname.com (gradha at iname.com)
Date: Thu, 1 Mar 2001 22:18:11 +0100
Subject: Looking for better Tkinter's listbox widget
References: <kr5379.9hl.ln@127.0.0.1> <3A9BEF9D.2050507@mobileway.com> <mailman.983346794.30823.python-list@python.org>
Message-ID: <jaem79.2pa.ln@127.0.0.1>

Martin Franklin <martin.franklin at westerngeco.com> wrote:
> Try MCScrolledListBox it is a Pmw/PmwContribD by Doug Hellmann
>  http://members.home.net/doughellmann/PmwContribD/index.html


Really nice package. Thanks!

-- 

 Grzegorz Adam Hankiewicz   gradha at iname.com - http://gradha.infierno.org
 Other web pages:  http://glub.ehu.es/  -  http://welcome.to/gogosoftware/


From n8gray at caltech.edu.is.my.e-mail.address  Thu Mar  8 01:00:33 2001
From: n8gray at caltech.edu.is.my.e-mail.address (Nathaniel Gray)
Date: Wed, 7 Mar 2001 22:00:33 -0800
Subject: Long Shifty Bits
References: <kmap6.15$8Q.28605@news.pacbell.net> <m3itlmbu3g.fsf@atrus.jesus.cam.ac.uk> <MLap6.18$8Q.33391@news.pacbell.net>
Message-ID: <9877h8$m2i@gap.cco.caltech.edu>

Phlip wrote:

> Michael Hudson wrote:
> 
>> Works for me.  Have you tried it?
> 
>>>> x = 1L << 200
>>>> x
> 1606938044258990275541962092341162602522202993782792835301376L
> 
> Ouch. Just trying it did not occur to me. The expression's deep inside a
> blob of logic, and I just assumed when I hit a bug up around 32 elements
> that I had misrepresented <<'s prowess.
> 

Just make sure your variable is *definitely* a long and not just an int!

[n8gray at golux n8gray]$ python
Python 1.5.2 (#1, Dec 21 2000, 15:29:08)  [GCC egcs-2.91.66 19990314/Linux 
(egcs- on linux-i386
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> 1 << 200
0

-n8

-- 
_.~'^`~._.~'^`~._.~'^`~._.~'^`~._.~'^`~._
             Nathaniel Gray
   California Institute of Technology
     Computation and Neural Systems
     n8gray <at> caltech <dot> edu
_.~'^`~._.~'^`~._.~'^`~._.~'^`~._.~'^`~._



From SBrunning at trisystems.co.uk  Wed Mar 28 09:17:12 2001
From: SBrunning at trisystems.co.uk (Simon Brunning)
Date: Wed, 28 Mar 2001 15:17:12 +0100
Subject: __init__.py files in PMW
Message-ID: <31575A892FF6D1118F5800600846864D78BA43@intrepid>

> From:	Laura Creighton [SMTP:lac at cd.chalmers.se]
> PMW comes in several directories, and each one has as file named
> __init__.py which contains this line:
> # File to allow this directory to be treated as a python 1.5 package.
> 
> I don't understand.  Who needs this file and for what?
 
See <http://www.python.org/doc/essays/packages.html>.

Cheers,
Simon Brunning
TriSystems Ltd.
sbrunning at trisystems.co.uk





-----------------------------------------------------------------------
The information in this email is confidential and may be legally privileged.
It is intended solely for the addressee. Access to this email by anyone else
is unauthorised. If you are not the intended recipient, any disclosure,
copying, distribution, or any action taken or omitted to be taken in
reliance on it, is prohibited and may be unlawful. TriSystems Ltd. cannot
accept liability for statements made which are clearly the senders own.



From n8gray at caltech.edu.is.my.e-mail.address  Fri Mar 23 01:38:00 2001
From: n8gray at caltech.edu.is.my.e-mail.address (Nathaniel Gray)
Date: Thu, 22 Mar 2001 22:38:00 -0800
Subject: Good Editor? (What's the big deal with syntax colouring?)
References: <008F0A63472BD311AF9800104BCD102561CD3E@minirex.vmlabs.com> <mailman.985295730.14145.python-list@python.org>
Message-ID: <99erbn$3e@gap.cco.caltech.edu>

Timothy Grant wrote:
> Now if only else didn't highlight but else: did<wink>

Ah, but that's what auto-indent is for!  When NEdit doesn't indent the 
first line of a block I know I forgot the colon at the end of the last line.

One of these days I'll send in my ultra-improved auto-indent Python macros 
for NEdit.  I'm still working on the auto-dedent macros... <wink>

-n8

-- 
_.~'^`~._.~'^`~._.~'^`~._.~'^`~._.~'^`~._
             Nathaniel Gray
   California Institute of Technology
     Computation and Neural Systems
     n8gray <at> caltech <dot> edu
_.~'^`~._.~'^`~._.~'^`~._.~'^`~._.~'^`~._



From tbryan at python.net  Sun Mar 11 00:54:33 2001
From: tbryan at python.net (Thomas A. Bryan)
Date: Sun, 11 Mar 2001 00:54:33 -0500
Subject: Open Source
References: <9806i4$2e97$1@news3.infoave.net>
Message-ID: <3AAB1319.46B1B9AE@python.net>

Alton Luder III wrote:
> 
> Is there any place on the web that offers open source python programs. It
> might be helpful for me to learn if I see other people's program's source
> code.
> Thanks in advance,
> Alton Luder III

You can browse the Sourceforge software map by programming language.  It 
looks like setting the language to Python maxes out the query at over 
300 results. 

http://sourceforge.net/softwaremap/trove_list.php?form_cat=178

---Tom


From aahz at panix.com  Sun Mar 18 10:39:53 2001
From: aahz at panix.com (Aahz Maruch)
Date: 18 Mar 2001 07:39:53 -0800
Subject: How do I avoid the Tkinter 30 msec polling loop.
References: <98s781$7a8o$1@fido.engr.sgi.com> <3ab3d191.6695740@24.0.228.33>
Message-ID: <992ks9$c97$1@panix2.panix.com>

In article <3ab3d191.6695740 at 24.0.228.33>,
Mike Clarkson <support @ internetdiscovery.com> wrote:
>
>The idea would be to establish a Tcl Event procedure that would check
>for Python signals every (say) 100 msec., and handle Tcl events all of
>the rest of the time.
>
>Does anyone know how to do this in Tcl - it shouldn't be more than a
>few lines I suspect. Look at Tcl_CreateEventSource and
>Tcl_SetMaxBlockTime in tcl/generic/tclNotify.c

That sounds like a good idea.  After doing some *very* basic Tkinter
work as part of my threads tutorial, I would suggest that such a feature
include the ability to set the length of the delay; 100ms is pretty
large-grained (though fine as a default).
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

Three sins: BJ, B&J, B&J


From duncan at rcp.co.uk  Mon Mar 12 04:41:05 2001
From: duncan at rcp.co.uk (Duncan Booth)
Date: Mon, 12 Mar 2001 09:41:05 +0000 (UTC)
Subject: question , Does pytyon excist on a pda, like psion?
References: <xDJq6.28348$KE1.1127955@zwoll1.home.nl> <98frub$1mhg0$1@fu-berlin.de>
Message-ID: <Xns9062621467D5Dduncanrcpcouk@127.0.0.1>

"Matthias Huening" <mhuening at zedat.fu-berlin.de> wrote in <98frub$1mhg0$1
@fu-berlin.de>:

>
>> Is Python already ported to the psion or other pda, s?
>>
>
>
>http://www.cs.uu.nl/~otfried/Python/
>http://dales.rmplc.co.uk:9080/Duncan/PyPsion.htm
>(The second was unavailable when I tried it five minutes ago)

Try http://dales.rmplc.co.uk/Duncan/PyPsion.htm
and if you know of anywhere publicising the 9080 port number, please can you 
ask them to remove it.


From erno-news at erno.iki.fi  Tue Mar 27 21:47:31 2001
From: erno-news at erno.iki.fi (Erno Kuusela)
Date: 28 Mar 2001 05:47:31 +0300
Subject: Threading legality/morality
References: <fsDv6.1247$NR.100201@news3.oke.nextra.no> <99nnbv$phs$1@panix6.panix.com> <h0%v6.1842$NR.144463@news3.oke.nextra.no> <99qa3v$6e0$1@panix6.panix.com> <99qoje$o6q$1@troll.powertech.no>
Message-ID: <ku7l1awqws.fsf@lasipalatsi.fi>

In article <99qoje$o6q$1 at troll.powertech.no>, "Syver Enstad"
<syver at NOSPAMcyberwatcher.com> writes:

| Think so yes, how can it be fixed? I am pretty well versed in python but a
| pointer could be nice so I didn't use quite so much time on it.

you can override the FancyURLopener.get_user_passwd method.

  -- erno


From cce at clarkevans.com  Tue Mar 27 07:49:46 2001
From: cce at clarkevans.com (Clark C. Evans)
Date: Tue, 27 Mar 2001 07:49:46 -0500 (EST)
Subject: Yet Another PEP:  Query Protocol Interface or __query__
In-Reply-To: <Pine.LNX.4.21.0103201837570.21343-100000@clarkevans.com>
References: <Pine.LNX.4.21.0103201837570.21343-100000@clarkevans.com>
Message-ID: <Pine.LNX.4.21.0103270747160.17716-100000@clarkevans.com>

Below is an updated version of the proposal I've been
a champion for.  It incorporates a good amount of feedback
that I've received.  This proposal is starting to be come
more solid and the implementation is now hosted at 
http://sourceforge.net/projects/adapt.  Changes include:

    *   Checking component "can_wrap" and "isa" have been
        removed as suggested by Alex Martelli
    *   Replaced __adapt__ with __conform__ and then __prodapt__ 
        with __adapt__ as suggested by Magnus Lie Hetland
    *   A _check private function was added to demonstrate how
        we work in conjunction with Paul Prescod's checking 
        proposal and Michel Pelletier's interface proposal.
    *   Borrowed more descriptive sections from Alex Martelli
        for the abstract. 
    *   Updated the title to reflect the removal of 
        checking from the goal and to reflect the bi-directional
        lookup mechanism as proposed; is this a good name?
    *   Added a section about how this proposal relates to the
        automagical proxy construction techniques posted by 
        Carlos Ribeiro.

Thank you once again for your kind feedback. 

Best,

Clark


PEP: XXX
Title: Protocol Conformance and Object Adaptation
Version: $Revision$
Author: Clark Evans
Python-Version: 2.2
Status: Draft
Type: Standards Track
Created: 21-Mar-2001
Updated: 23-Mar-2001

Abstract

    This proposal puts forth an extensible mechanism for the 
    adaptation of an object to a context where a specific type, 
    class, interface, or other protocol is expected.

    This proposal provides a built-in "adapt" function that, for any
    object X and protocol Y, can be used to ask the Python environment
    for a version of X complaint with Y.  Behind scenes the mechanism
    asks the object X: "Are you now, or do you know how to wrap 
    yourself to provide, a supporter of protocol Y?".  And, if this
    request fails, the function then asks the protocol Y: "Does object
    X support you, or do you know how to wrap it to obtain such a 
    supporter?"  This duality is important, because protocols can be
    developed after objects are, OR vice-versa, and this PEP lets
    either case be supported non-invasively with regard to the 
    pre-existing component[s].

    This proposal does not limit what a protocol is, what 
    compliance to the protocol means, nor what a wrapper 
    constitutes.  This mechanism leverages existing
    protocol categories such as the type system and class 
    hierarchy and can be expanded to support future protocol
    categories such as the pending interface proposal [1] 
    and signature based type-checking system [2]

Motivation

    Currently there is no standardized mechanism in Python for
    asking if an object supports a particular protocol. Typically,
    existence of particular methods, particularly those that are
    built-in such as __getitem__, is used as an indicator of
    support for a particular protocol.  This technique works for
    protocols blessed by GvR, such as the new enumerator proposal
    identified by a new built-in __iter__.  However, this technique
    does not admit an infallible way to identify interfaces lacking
    a unique, built-in signature method.

    More so, there is no standardized way to obtain an adapter
    for an object.  Typically, with objects passed to a context
    expecting a particular protocol, either the object knows about
    the context and provides its own wrapper or the context knows
    about the object and wraps it appropriately.  The difficulty 
    with these approaches is that such adaptations are one-offs,
    are not centralized in a single place of the users code, and
    are not executed with a common technique, etc.  This lack of
    standardization increases code duplication with the same
    adapter occurring in more than one place or it encourages
    classes to be re-written instead of adapted.  In either case,
    maintainability suffers.

    It would be very nice to have a standard function that can
    be called upon to verify an objects compliance with a 
    particular protocol and provide for a wrapper if one is
    readily available -- all without having to hunt through 
    a library's documentation for the appropriate incantation.

Requirements

    When considering an objects compliance with a protocol, 
    there are several cases to be examined:

     a) When the protocol is a type or class, and the object
        has exactly that type or is a member of the class.
        In this case compliance is automatic.

     b) When the object knows about the protocol and either
        considers itself compliant or knows how to wrap itself
        appropriately.

     c) When the protocol knows about the object and either
        the object already complies or can be wrapped accordingly.

     d) When the protocol is a class, and the object is a
        member of a subclass.  This is distinct from the first
        case (a) above, since inheritance does not necessarily
        imply substitutability and must be handled carefully.

     e) When the context knows about the object and the
        protocol and knows how to adapt the object so that
        the required protocol is satisfied.  This could use an
        adapter registry or similar method.

    For this proposal's requirements, the first case should be
    come for free and the next three cases should be relatively
    relatively easy to accomplish.  This proposal does not address 
    the last case, however it provides a base mechanism upon which 
    such an approach could be developed.  Further, with only minor
    implementation changes, this proposal should be able to 
    incorporate a new interface type or type checking system.

    The fourth case above is subtle. A lack of substitutability can
    occur when a method restricts an argument's domain or raises an 
    exception which a base class does not or extends the co-domain to
    include return values which the base class may never produce.  
    While compliance based on class inheritance should be automatic, 
    this proposal should allow an object to signal that it is not 
    compliant with a base class protocol.

Specification

    This proposal introduces a new built-in function, adapt, which
    is the basis for supporting these requirements. 

    The adapt function has three parameters: 1) the object to be
    adapted ("obj"), 2) the protocol requested of the object
    ("protocol"), and 3) an optional object to return if the object
    could not be adapted ("alternate").

    A successful result of the adapt function returns either
    the object passed ("obj") if the object is already compliant 
    with the protocol, or a secondary object ("wrapper"), which
    provides a view of the object compliant with the protocol.  
    The definition of wrapper is explicitly vague and a wrapper
    is allowed to be a full object with its own state if necessary.
    A failure to adapt the object to the protocol will raise a
    TypeError unless the alternate parameter is used, in this
    case the alternate argument is returned.

    To enable the first case listed in the requirements, 
    the adapt function first checks to see if the objects 
    type or the objects class are identical to the protocol.
    If so, then the adapt function returns the object directly
    without further ado.

    To enable the second case, when the object knows about the
    protocol, the object must have an __conform__ method.  This 
    optional method takes two arguments, the object being conformed
    ("self") and the protocol requested ("protocol").  The object  
    may return itself through this method to indicate compliance.
    Alternatively, the object also has the option of returning a
    wrapper object compliant with the protocol.  Finally, if the 
    object cannot determine its compliance, it should either return
    None or raise a TypeError to enable the remaining mechanisms.

    To enable the third case, when the protocol knows about the object,
    the protocol must have a __adapt__ method.  This optional method
    takes two arguments, the protocol requested ("self") and the
    object being adapted ("obj"). If the protocol finds the object
    to be compliant, it can return obj directly.  Alternatively, the 
    method may return a wrapper compliant with the protocol.  Finally,
    compliance cannot be determined, this method should either 
    return None or raise a TypeError so other mechanisms can be tried.

    The fourth case, when the object's class is a sub-class of the 
    protocol, is handled by the built-in adapt function.  Under normal
    circumstances, if isinstance(object,protcol) then adapt returns
    the object directly.  However, if the object is not substitutable,
    either the __conform__ or __adapt__ methods above may raise an
    adaptForceFailException to prevent this default behavior.

    Please note two important things.  First, this proposal does not
    preclude the addition of other protocols.  Second, this proposal
    does not preclude other possible cases where adapter pattern may
    hold, such as the context knowing the object and the protocol 
    (the last case in the requirements).  In fact, this proposal opens
    the gate for these other mechanisms to be added.

Reference Implementation and Test Cases

    -----------------------------------------------------------------
    adapt.py
    -----------------------------------------------------------------
    import types

    adaptRaiseTypeException = "(raise a type exception on failure)"
    adaptForceFailException = "(forced failure of adapt)"

    # look to see if the object passes other protocols
    def _check(obj,protocol,default):
            return default

    def adapt(obj, protocol, alternate = adaptRaiseTypeException):

        # first check to see if object has the exact protocol
        if type(obj) is types.InstanceType and \
           obj.__class__ is protocol: return obj
        if type(obj) is protocol: return obj

        # next check other protocols for exact conformance
        # before calling __conform__ or __adapt__
        if _check(obj,protocol,0):
            return obj

        # procedure to execute on success
        def succeed(obj,retval,protocol,alternate):
            if _check(retval,protocol,1):
                return retval
            else:
                return fail(obj,alternate)

        # procedure to execute on failure
        def fail(obj,protocol,alternate):
            if alternate is adaptRaiseTypeException:
                raise TypeError("%s cannot be adapted to %s" \
                                 % (obj,protocol))
            return alternate

        # try to use the object's adapting mechanism
        conform = getattr(obj, '__conform__',None)
        if conform:
            try:
                retval = conform(protocol)
                if retval:
                    return succeed(obj,retval,protocol,alternate)
            except adaptForceFailException:
                return fail(obj,protocol,alternate)
            except TypeError: pass

        # try to use the protocol's adapting mechanism
        adapt = getattr(protocol, '__adapt__',None)
        if adapt:
            try:
                retval = adapt(obj)
                if retval:
                    return succeed(obj,retval,protocol,alternate)
            except adaptForceFailException:
                return fail(obj,protocol,alternate)
            except TypeError: pass

        # check to see if the object is an instance
        try:
            if isinstance(obj,protocol):
                return obj
        except TypeError: pass

        # no-adaptation-possible case
        return fail(obj,protocol,alternate)

    -----------------------------------------------------------------
    test.py
    -----------------------------------------------------------------
    import types
    from adapt import adaptForceFailException
    from adapt import adapt

    class KnightsWhoSayNi: pass

    class Eggs:  # an unrelated class/interface
        def eggs(self): print "eggs!"
        word = "Nee-womm"

    class Ham:  # used as an interface, no inhertance
        def ham(self): pass
        word = "Ping"

    class Spam: # a base class, inheritance used
        def spam(self): print "spam!"

    class EggsSpamAndHam (Spam,KnightsWhoSayNi):
        def ham(self): print "ham!"
        def __conform__(self,protocol):
            if protocol is Ham:
                # implements Ham's ham, but does not have a word
                return self
            if protocol is KnightsWhoSayNi:
                # we are no longer the Knights who say Ni!
                raise adaptForceFailException
            if protocol is Eggs:
                # Knows how to create the eggs!
                return Eggs()

    class SacredWord:
        class HasSecredWord:
            def __call__(self, obj):
                if getattr(obj,'word',None): return obj
        __adapt__= HasSecredWord()

    class Bing (Ham):
        def __conform__(self,protcol):
            raise adaptForceFailException

    def test():
        x = EggsSpamAndHam()
        adapt(x,Spam).spam()
        adapt(x,Eggs).eggs()
        adapt(x,Ham).ham()
        adapt(x,EggsSpamAndHam).ham()
        print adapt(Eggs(),SacredWord).word
        print adapt(Ham(),SacredWord).word
        pass
        if adapt(x,KnightsWhoSayNi,None): raise "IckyIcky"
        if not adapt(x,Spam,None): raise "Spam"
        if not adapt(x,Eggs,None): raise "Eggs"
        if not adapt(x,Ham,None): raise "Ham"
        if not adapt(x,EggsSpamAndHam,None): raise "EggsAndSpam"
        if     adapt(x,KnightsWhoSayNi,None): raise "NightsWhoSayNi"
        if     adapt(x,SacredWord,None): raise "SacredWord"
        try:
            adapt(x,SacredWord)
        except TypeError: pass
        else: raise "SacredWord"
        try:
            adapt(x,KnightsWhoSayNi)
        except TypeError: print "Ekky-ekky-ekky-ekky-z'Bang, " \
                                + "zoom-Boing, z'nourrrwringmm"
        else: raise "NightsWhoSayNi"
        pass
        b = Bing()
        if not adapt(b,Bing,None): raise "Not a Bing"
        if adapt(b,Ham,None): raise "Not a Ham!"
        if adapt(1,types.FloatType,None): raise "Not a float!"
        if adapt(b,types.FloatType,None): raise "Not a float!"
        if adapt(1,Ham,None):             raise "Not a Ham!"
        if not adapt(1,types.IntType,None): raise "Is an Int!"

    -----------------------------------------------------------------
    Expected Output
    -----------------------------------------------------------------
    >>> import test
    >>> test.test()
    spam!
    eggs!
    ham!
    ham!
    Nee-womm
    Ping
    Ekky-ekky-ekky-ekky-z'Bang, zoom-Boing, z'nourrrwringmm
    >>>

Relationship To Paul Prescod and Tim Hochberg's Type Assertion method

    Paul and Tim had proposed an type checking mechanism, where the
    Interface is passed an object to verify.  The example syntax Paul
    put forth recently [2] was:

        interface Interface
            def __check__(self,obj)

    For discussion purposes, here would be a protocol with __check__:

        class Interface:
            class Checker:
                def __call__(self, obj): pass  #check the object
            __check__= Checker()

    The built-in adapt function could be augmented to use this 
    checking mechanism updating the _check method as follows:

    # look to see if the object passes other protocols
    def _check(obj,protocol,default):
        check = getattr(protocol, '__check__',None)
        if check:
            try:
               if check(obj): return 1
            except TypeError: pass
            return 0
        else:
            return default

    In short, the work put forth by Paul and company is great, and
    I don't see any problems why these two proposals couldn't work
    together in harmony, if not be completely complementary.

Relationship to Python Interfaces [1] by Michel Pelletier

    The relationship to this proposal to Michel's proposal could
    also be complementary.  Following is how the _check method 
    would be updated for this mechanism:

    # look to see if the object passes other protocols
    def _check(obj,protocol,default):
        if type(protocol) is types.InterfaceType:
            return implements(obj,protcol)
        return default

Relationship to Carlos Ribeiro's proxy technique [7] and [8]

    Carlos presented a technique where this method could return 
    a proxy instead of self or a wrapper.  The advantage of this
    approach is that the internal details of the object are
    protected.  This is very neat.  No changes are necessary to 
    this proposal to support this usage as a standardized mechanism 
    to obtain named proxies.

Relationship To Microsoft's Query Interface

    Although this proposal may sounds similar to Microsoft's
    QueryInterface, it differs by a number of aspects.  First,
    it is bi-directional allowing the interface to be queried 
    as well giving more dynamic abilities (more pythonic).  Second,
    there is not a special "IUnknown" interface which can be used
    for object identity, although this could be proposed as one
    of those "special" blessed interface protocol identifiers.
    Third, with QueryInterface, once an object supports a particular
    interface it must always there after support this interface;
    this proposal makes no such guarantee, although this may be
    added at a later time. Fourth, implementations of Microsoft's
    QueryInterface must support a kind of equivalence relation.
    By reflexive they mean the querying an interface for itself
    must always succeed.  By symmetrical they mean that if one
    can successfully query an interface IA for a second interface
    IB, then one must also be able to successfully query the
    interface IB for IA.  And finally, by transitive they mean if
    one can successfully query IA for IB and one can successfully
    query IB for IC, then one must be able to successfully query
    IA for IC.  Ability to support this type of equivalence relation
    should be encouraged, but may not be possible.  Further research
    on this topic (by someone familiar with Microsoft COM) would be
    helpful in further determining how compatible this proposal is.

Question and Answer

    Q:  What benefit does this provide?

        The typical python programmer is an integrator, someone
        who is connecting components from various vendors.  Often
        times the interfaces between these components require 
        an intermediate adapter.  Usually the burden falls upon
        the programmer to study the interface exposed by one
        component and required by another, determine if they
         are directly compatible, or develop an adapter.  Sometimes
        a vendor may even include the appropriate adapter, but
        then searching for the adapter and figuring out how to 
        deploy the adapter takes time.
    
        This technique enables vendors to work with each other
        directly by implementing __conform__ or __adapt__ as
        necessary.  This frees the integrator from making their
        own adapters.  In essence, this allows the components
        to have a simple dialogue among themselves.  The integrator
        simply connects one component to another, and if the types
        don't automatically match an adapting mechanism is built-in.

        For example, consider SAX1 and SAX2 interfaces, there is 
        an adapter required to switch between them.  Normally the
        programmer must be aware of this; however, with this 
        adaptation framework this is no longer the case.

    Q:  Why does this have to be built-in, can't it be standalone?

        Yes, it does work standalone.  However, if it is built-in,
        it has a greater chance of usage.  The value of this proposal
        is primarily in standardization.  Furthermore:

        0.  The mechanism is by its very nature a singleton.
        1.  If used frequently, it will be much faster as a built-in
        2.  It is extensible and unassuming.
        3.  A whole-program optimizing compiler could optimize it out
            in particular cases (ok, this one is far fetched)

    Q:  Why the verbs __conform__ and __adapt__?

        conform, verb intransitive
            1. To correspond in form or character; be similar. 
            2. To act or be in accord or agreement; comply.  
            3. To act in accordance with current customs or modes.

        adapt, verb transitive
            1. To make suitable to or fit for a specific use or
               situation.

        Source:  The American Heritage Dictionary of the English
                 Language, Third Edition 

Backwards Compatibility

    There should be no problem with backwards compatibility unless
    someone had used __conform__ or __adapt__, but this seems 
    unlikely.  Indeed this proposal, save an built-in adapt() 
    function, could be tested without changes to the interpreter.

Credits

    This proposal was created in large part by the feedback
    of the talented individuals on both the main mailing list
    and also the type signature list.  Specific contributors
    include (sorry if I missed someone).

    This proposal is based largely off the suggestions from
    Alex Martelli and Paul Prescod with significant feedback
    from Robin Thomas and borrowing ideas from Marcin 'Qrczak'
    Kowalczyk and Carlos Ribeiro.  Other contributors (via comments)
    include:

        Michel Pelletier, Jeremy Hylton, Aahz Maruch, 
        Fredrik Lundh, Rainer Deyke, Timothy Delaney, 
        and Huaiyu Zhu

Copyright

    This document has been placed in the public domain.


References and Footnotes

    [1] http://python.sourceforge.net/peps/pep-0245.html
    [2] http://mail.python.org/pipermail/types-sig/2001-March/001223.html
    [3] http://www.zope.org/Members/michel/types-sig/TreasureTrove
    [4] http://mail.python.org/pipermail/types-sig/2001-March/001105.html
    [5] http://mail.python.org/pipermail/types-sig/2001-March/001206.html
    [6] http://mail.python.org/pipermail/types-sig/2001-March/001223.html
    [7]
http://mail.python.org/pipermail/python-list/2001-March/035136.html
    [8]
http://mail.python.org/pipermail/python-list/2001-March/035197.html









From db3l at fitlinxx.com  Wed Mar  7 14:16:48 2001
From: db3l at fitlinxx.com (David Bolen)
Date: 07 Mar 2001 14:16:48 -0500
Subject: Overriding UserList methods
References: <b566atck0ruu4tae8n1srrkntbmme26gor@4ax.com> <2oc6atgkn0ror77rj3b6j8d8d2s0gaa5k7@4ax.com>
Message-ID: <un1ax9yov.fsf@ctwd0143.fitlinxx.com>

Daniel Klein <danielk at aracnet.com> writes:

> I found the answer in the Beazley book. It's
> 
> def append(self, object):
>         self.data.append(object)
> 
> using the 'data' instance variable automatically set up in the UserList
> class.

While this is a particular solution that works with the implementation
exposed by UserList, it's not a general answer to subclassing and
overriding in general.  For that, the right approach is to have your
subclass method call it's superclass when or if appropriate in any
overridden method.

> But I'll try the other suggestions anyway just to get a better
> understanding.

Think of this as exactly the same thing you are already doing in your
__init__ method (although you may just have copied that from other
examples).

When you subclass from UserList.UserList, you will inherit all of its
methods, such that when code tries to call a method in your instance
that doesn't exist, Python will walk the list of superclasses (only
UserList in this case) and try to call the method there.

But when you override a method - by defining it in your own class -
then Python calls that method on your own instance.  This is true in
your original example with your local definition of __init__.  So in
order to execute your superclass' __init__ code you have to call it
yourself.  In other languages this may happen automatically, whereas
in Python you have to be explicit about it, and in particular you
explicitly reference the superclass name (UserList.UserList in your
example) when making the call, so it's very clear that you're calling
your superclass.

But __init__ isn't special in this regard - this very same approach is
used for any method you may override.  And since you control when the
superclass function is called, you can perform pre or post processing
in your overridden method as appropriate, or even ignore completely
the original superclass method.

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


From wayne.izatt at _nospam_myself.com  Tue Mar  6 02:25:27 2001
From: wayne.izatt at _nospam_myself.com (Wayne Izatt)
Date: Tue, 06 Mar 2001 07:25:27 GMT
Subject: Trouble running a Grayson Example
References: <3AA3E573.6DBE6D2E@theworld.com>
Message-ID: <Hh0p6.183446$df5.5139642@news1.crdva1.bc.home.com>

We don't all have the book. If you want help with code, include it.

cheers

"David Lees" <deblNonOspammy at theworld.com> wrote in message
news:3AA3E573.6DBE6D2E at theworld.com...
> I am trying to run example 8.9 (Example_8_9.py) from Graysons book,
> under Python 2.0 and get the following errors.  Can someone tell me what
> is going on?  Perhaps this has to do with changes between 1.5.2 and 2.0
> Python?
>
> Thanks in advance.
>
> david lees
>
>
> Traceback (most recent call last):  File
> "d:\Python20\Pythonwin\pywin\framework\scriptutils.py", line 301, in
> RunScript
>     exec codeObject in __main__.__dict__
>   File "E:\Grayson1\Chapter08\Example_8_9.py", line 79, in ?
>    ddnotebook = DDNotebook()
>   File "E:\Grayson1\Chapter08\AppShell.py", line 55, in __init__
>   self.__createInterface()
>   File "E:\Grayson1\Chapter08\AppShell.py", line 282, in
> __createInterface
>    self.createInterface()
>   File "E:\Grayson1\Chapter08\Example_8_9.py", line 75, in
> createInterface
>     self.createNotebook()
>   File "E:\Grayson1\Chapter08\Example_8_9.py", line 26, in
> createNotebook  Pmw.NoteBookR,(self.interior(),),)
>   File "E:\Downloads\pmw\Pmw\Pmw_0_8_4\lib\PmwLoader.py", line 174, in
> __getattr__
> AttributeError: NoteBookR




From chrisw at nipltd.com  Tue Mar 27 06:17:02 2001
From: chrisw at nipltd.com (Chris Withers)
Date: Tue, 27 Mar 2001 12:17:02 +0100
Subject: Namespaces question
References: <mailman.985627029.26291.python-list@python.org> <kubsqnwkvc.fsf@lasipalatsi.fi>
Message-ID: <3AC076AE.59442A69@nipltd.com>

Erno Kuusela wrote:
> 
> you aren't supposed to alter the dictionary returned by locals.
> try using exec if you really want to do this, although it would

Thanks, that's wha I was looking for...

> probably be better to just put the values in a regular dictionary.

...agreed, but theres' a whole load of legacy code that follows it that I don't
really want to mess with...

cheers,

Chris



From tim.one at home.com  Fri Mar  2 12:41:42 2001
From: tim.one at home.com (Tim Peters)
Date: Fri, 2 Mar 2001 12:41:42 -0500
Subject: A counter-counter-proposal for PEP 236: #pragma( nested_scopes)
In-Reply-To: <j4g0gwjh36.fsf@informatik.hu-berlin.de>
Message-ID: <LNBBLJKPBEHFEDALKOLCIEGPJDAA.tim.one@home.com>

[Tim]
> Hmm.  Maybe that's why I so dislike Martin's attempt to hijack
> PEP 236:  his current formulation of an alternative leaves only the raw
> names, without the __future__ or __possible_world__ qualifiers.

[Martin von Loewis]
> If *that* is the complaint, then the patch can be easily changed
> into reading
>
> directive future nested_scopes
>
> (future would be the directive name, and nested_scopes the atom
> argument).

That would be much better to me.  You need to write a PEP for this, though!
As is, "the patch" is a moving target and it's not at all clear how you
intend for this gimmick to evolve.  For example, in other msgs I read today,
you're saying that "directive" *will* be a keyword, someday -- I had not
guessed that (to the contrary, the emphasis before today was on that
"directive" was *not* being made a reserved word).  It needs a spec.

> In that case, I'd prefer /F's interpretation of the state
> of things, i.e.
>
> directive experimental nested_scopes
>
> since this does not talk about the future, but about the present (just
> as the future import does not talk about the future, but about
> directives that are presently needed).

nested_scopes talks about the future because, barring an unexpected disaster,
nested_scopes WILL BE the rule in 2.2, but ARE NOT in 2.1. "will be but are
not" == "the future", wrt any release (incl. 2.1) in which nested_scopes are
not yet the rule.  "will be", "not yet", ... it's impossible to tell the
nested_scopes story without invoking the future!  "experimental" misses the
point:  this isn't an option, this isn't going away, it's not up for a vote,
it's a preview of Python's future.

"experimental" would make more sense for a "hmm, Guido can't make up his mind
whether Python wants this or not, so we're enabling it as an experiment --
let us know what you think!" kind of gimmick.  The only good example of this
in Python's history I can recall is the now-defunct "access" statement (which
few people knew existed, because Guido was so wary of it that it was never
documented -- but that meant it didn't get a good trial, either).

"experimental" is a "use at your own risk" thing; "future" is a "*don't* use
at your own risk, because in a future release you'll be forced to live with
it" thing.  Character encodings are neither of those, they're a "use or not,
it's up to you, the language will support whatever decision you make, today
and forever more" thing.  Those distinctions are crucial for users:  they're
the differences among facing the reality of personal mortality, trying Perl
for a week, and deciding whether to hit the snooze button when the alarm goes
off <wink>.

without-a-pep-it-ain't-hep-ly y'rs  - tim




From chris at voodooland.net  Sat Mar 10 04:10:34 2001
From: chris at voodooland.net (Chris Watson)
Date: Sat, 10 Mar 2001 03:10:34 -0600 (CST)
Subject: Urgent Question about Python licensing
In-Reply-To: <98cpfe$5oj$0@216.39.170.247>
Message-ID: <Pine.BSF.4.21.0103100307160.26511-100000@open-systems.net>

> 2. Is it currently so, or will Python end up with, some license that
> requires that python scripts be covered by the GPL?

That would be when Python starts shipping under something like a GPL+. And
the day that happens I for one will drop Python like a bad habbit. I would
prefer a BSD or BSD style license. But im sure im out-voted on that.

--
=============================================================================
-Chris Watson         (316) 326-3862 | FreeBSD Consultant, FreeBSD Geek 
Work:              scanner at jurai.net | Open Systems Inc., Wellington, Kansas
Home:  scanner at deceptively.shady.org | http://open-systems.net
=============================================================================
WINDOWS: "Where do you want to go today?"
LINUX: "Where do you want to go tomorrow?"
BSD: "Are you guys coming or what?"
=============================================================================
GNU = Gnu's Non-portable & Unstable
irc.openprojects.net #FreeBSD -Join the revolution!
ICQ: 20016186




From whisper at oz.net  Mon Mar 26 13:44:50 2001
From: whisper at oz.net (Dave LeBlanc)
Date: 26 Mar 2001 18:44:50 GMT
Subject: CfP: Babel01 Workshop on Multi-Language Infrastructure and Interoperability
References: <3abf4ae1@msrnews.research.microsoft.com>
Message-ID: <99o2n2$9on$1@216.39.170.247>

Gotta wonder why Microsoft is asking for papers to be submitted in a
format it doesn't support.

I bet this ends up all C-sharpish.

Dave LeBlanc

On Mon, 26 Mar 2001 14:58:07 +0100, "Andrew Kennedy"
<akenn at microsoft.com> wrote:

>			       BABEL'01
>			   CALL FOR PAPERS
>
>		   First Workshop on Multi-Language
>		 Infrastructure and Interoperability
>
>	   http://research.microsoft.com/~nick/babel01.htm
>
>			 Satellite to PLI'01
>
>		  Firenze, Italy, 8th September 2001
>
>
>AIMS AND SCOPE
>
>Recent years have seen a resurgence of interest in multi-language
>tools and intermediate languages, and in interoperability between
>programs and components written in different programming
>languages. Shared infrastructure such as code generators, analysis
>tools and garbage collectors can greatly ease the task of producing a
>high-quality implementation of a new programming language, whilst
>being able to interoperate easily with code written in existing
>languages is essential for such an implementation to be useful in
>practice.  This workshop aims to bring together researchers and
>developers working on multi-language integration. Contributions are
>welcome on topics including, but not limited to:
>
>
>* Compilation of high level languages to common executable formats
>  such as Sun's Java Virtual Machine, the DRA's ANDF or Microsoft's
>  .NET Common Language Runtime.
>
>* Defining and using bindings for component middleware such as OMG's
>  CORBA or Microsoft's COM.
>
>* Language constructs to support interoperability between different
>  languages, particularly from different paradigms (e.g. OO/functional).
>
>* Multi-language backends for compilation and/or analysis
>  (e.g. MLRISC, FLINT, C--, TAL, SUIF).
>
>* Type theoretic and semantic foundations for multi-language work.
>
>* Multi-language development environments and tools (e.g. debuggers,
>  profilers).
>
>Submissions may address any programming paradigm.  Experience papers
>which describe the use of multiple languages in application
>development are also welcome. Authors unsure of the appropriateness of
>a potential submission should email the PC chair (nick at microsoft.com).
>
>
>PROGRAMME COMMITTEE
>
>Nick Benton (chair)          Microsoft Research
>Fergus Henderson             University of Melbourne
>Andrew Kennedy (organiser)   Microsoft Research
>Greg Morrisett               Cornell University
>Martin Odersky               Ecole Polytechnique Federale de Lausanne
>John Reppy                   Bell Labs
>Andrew Tolmach               Portland State University
>David Wakeling               University of Exeter
>
>
>INVITED SPEAKERS
>
>Zhong Shao                   Yale University
>
>
>IMPORTANT DATES
>
>Deadline for submission      1st June 2001
>Notification of acceptance   9th July 2001
>Final version due            10th August 2001
>
>
>SUBMISSION DETAILS
>
>Papers should be at most 14 pages and should be submitted in
>Ghostscript-compatible PostScript or PDF format and be printable on
>both US letter and A4 paper.  Authors are strongly encouraged to use
>ENTCS style files (see http://math.tulane.edu/~entcs/).  Papers should
>be submitted electronically from the workshop website given above.
>Submissions should not overlap significantly with work which has
>already been published or submitted to any other conference or
>journal.  An informal proceedings will be published as a technical
>report and distributed at the workshop.  It is intended that a final
>proceedings will be published in a volume of ENTCS.
>
>
>



From kriol at boopsie.fnal.gov  Thu Mar  8 17:18:12 2001
From: kriol at boopsie.fnal.gov (Oleg Krivosheev)
Date: 08 Mar 2001 16:18:12 -0600
Subject: How to call C function in shared library?
References: <vi81ys8gke1.fsf@wally.fnal.gov> <mailman.984079871.14235.python-list@python.org>
Message-ID: <vi8g0gn3nx7.fsf@boopsie.fnal.gov>

Kalle Svensson <kalle at gnupung.net> writes:

> Sez Oleg Krivosheev:
> > i already have shared library
> > with simple function which takes double
> > and returns double. Is it simple way 
> > to call it directly from Python?
> 
> You could take a look at SWIG (http://www.swig.org/), or make your own
> wrapper.  Something like (extremely untested):
> 
> /************  BEGIN **************/
> #include "Python.h"
> #include "myheader.h"
> 
> static PyObject *
> wrap_myfunction(PyObject *self, PyObject *args)
> {
>     double arg, ret;
>     if (!PyArg_ParseTuple(args, "d", &arg))
>         return NULL;
>     ret = myfunction(arg);
>     return Py_BuildValue("d", ret);
> }
> 
> /* Module initialisation stuff goes here...     */
> /* Check http://www.python.org/doc/current/ext/ */
> /* and http://www.python.org/doc/current/api/   */
> /* for more information...                      */
> /************  END **************/
> 
> > Or some python encapsulation for dlopen/dlsym/dlclose ?
> 
> There is the dl module:
> http://www.python.org/doc/current/lib/module-dl.html

umm...

it doesn't support double it seems...

thanks anyway

OK


From jurgen.defurne at philips.com  Fri Mar 16 02:32:56 2001
From: jurgen.defurne at philips.com (jurgen.defurne at philips.com)
Date: Fri, 16 Mar 2001 08:32:56 +0100
Subject: Announcement: Python Programming: Taming the Snake.
Message-ID: <0056900016652174000002L042*@MHS>

You can tame a snake, but you can't tame Monty.

Jurgen


From phrxy at csv.warwick.ac.uk  Thu Mar 29 13:18:32 2001
From: phrxy at csv.warwick.ac.uk (John J. Lee)
Date: Thu, 29 Mar 2001 19:18:32 +0100
Subject: NumPY: Reading/Writing Arrays To/From Files
In-Reply-To: <3AC1F37D.E8ADA48@seebelow.org>
References: <3aae3bd6$0$42882$272ea4a1@news.execpc.com> <3AC1F37D.E8ADA48@seebelow.org>
Message-ID: <Pine.SOL.4.30.0103291907490.13267-100000@primrose.csv.warwick.ac.uk>

On Wed, 28 Mar 2001, Grant Griffin wrote:

> Ben Held wrote:
> >
> > I would like to be able to read/write NumPY array objects to/from ascii
> > files.  I could not find support for this in the NumPY code.  Any ideas?
> >
> > Ben
>
> I forget exactly, but I think there's something out there called
> "arrayio".  Look for Travis Oliphant's page: I think you can find it
> there.
[...]

ArrayIO I think is for binary data, not ASCII.  As far as I can tell,
Travis' pages haven't resurfaced since they disappeared a few months ago.
There is a copy at

http://starship.python.net/~da/Travis/numpyio/

but this may be ancient for all I know.  You can do the same thing with
the Python standard library too, IIRC.

TableIO reads ASCII tables.

http://php.iupui.edu/~mmiller3/python/

There is a bug somewhere around line 40 that you'll get if you use Numeric
arrays where it wants lists.  Just delete the line that checks for array
type, and put in a try / except if you feel like it.

This is faster than a python version of course, but the GNU C library
implementation of strtok() has trouble with input like .3 or 3. (one or
the other).


John



From db3l at fitlinxx.com  Mon Mar  5 16:26:26 2001
From: db3l at fitlinxx.com (David Bolen)
Date: 05 Mar 2001 16:26:26 -0500
Subject: str behaviour
References: <mailman.983520482.8490.python-list@python.org>
Message-ID: <u66hnaovx.fsf@ctwd0143.fitlinxx.com>

tanzer at swing.co.at (Christian Tanzer) writes:

> Just off the top of my head.
> 
> Would it work to have a third function, say `display' (*), besides `str'
> and `repr'? The container types would "pass display down" regardless
> of whether str() or repr() is applied at the top level.
> 
> For classes, the designer could choose whatever meaning applied best
> for `__display__' -- whether it be an alias of `str' or of `repr' or
> something entirely different.

But then you'd defeat the case where repr() was really desired such
that the string could be reprocessed back into its original form.  In
such a case, the class doesn't necessarily have enough information to
know that this is destined for "display" rather than processing.

For example, I like the fact that I can print a nested object and end
up with a string that I can then eval() back into it's original form
at some later time.  Any necessary quoting and what not is handled.
Neither solution is perfect, but at least passing down repr() ensures
that invariant.  (Mostly, as more complex types may not always be
eval()able back into the same object they started as, but it's still
better than defaulting to str() which guarantees that you can't do
that).

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


From gerhard.nospam at bigfoot.de  Sat Mar 10 17:24:14 2001
From: gerhard.nospam at bigfoot.de (Gerhard =?iso-8859-1?Q?H=E4ring?=)
Date: Sat, 10 Mar 2001 23:24:14 +0100
Subject: [ Cry for help ] Please bugtest my full-text indexer
References: <Pe6p6.8119$t21.206104@news3.oke.nextra.no>
Message-ID: <slrn9alb87.ag8.gerhard.nospam@lilith.hqd-internal>

On Tue, 6 Mar 2001 15:11:57 +0100, Thomas Weholt <thomas at cintra.no> wrote:
>Hi,
>
>Just finished my own full-text indexer which is easy to use and seem to be
>quite fast. BUT, when tested against a lot of data, my archive of RFCs, the
>thing croaked after inserting approx. 500.000 keys ( size of data stored was
>~40 MB ), giving me an error-message from the bsddb-module, (0,'error').

I just ran this under my system and it worked quite fine. After some more
than 3.000.000 keys, I aborted the program.

I used Python 2.0 on SuSE 7.1, the bsd module used has version db-3.1.17. So
maybe it's just that you have a buggy version of BSDdb?

Gerhard
-- 
mail:   gerhard <at> bigfoot <dot> de
web:    http://highqualdev.com


From jcc.ugm at ix.netcom.com  Thu Mar 29 11:21:53 2001
From: jcc.ugm at ix.netcom.com (Jonathan Claggett)
Date: Thu, 29 Mar 2001 11:21:53 -0500
Subject: Is inheritance broken?
In-Reply-To: <99uu0501u4h@news2.newsguy.com>
Message-ID: <NDBBLAKACLGJJBDNGGCMOEKECCAA.jcc.ugm@ix.netcom.com>

Thank you all very much for your insights. I am getting the impression that
there as many ways to use inheritance as there are martial art fighting
styles (OOP Fu :-). I just checked* and it appears Scott Meyers has a book
on effective programming which I now intend to purchase. Obviously, there is
more to inheritance than is covered by my borrowed copy of "Learning
Python". Thanks again,

Jonathan

*
http://www.amazon.com/exec/obidos/ASIN/0201924889/qid=985881640/sr=1-1/ref=s
c_b_2/103-7754247-2945461




From shredwheat at mediaone.net  Sat Mar 31 21:08:27 2001
From: shredwheat at mediaone.net (Pete Shinners)
Date: Sun, 01 Apr 2001 02:08:27 GMT
Subject: PythonLabs Team Moves to ActiveState
References: <mailman.986086501.5576.python-list@python.org>
Message-ID: <v4wx6.121194$Ch.16134973@typhoon.we.rr.com>

"Guido van Rossum" <guido at python.org> wrote
> To all Python users and developers:
>
>    Less than half a year ago, I moved with my team to Digital Crations...

i was reading through the announcement thinking 'hmmm, another
exciting new owner for python...?'  then i got towards the end of
the announcement and ?!?!?

then i thought, hmm, must be april fools day.
(please, please, please)

i've got a knot in my stomach. i can't tell if this is a terrific
april fools joke or just about the worst news i've heard in a long
time.






From tim.one at home.com  Sun Mar 11 01:12:23 2001
From: tim.one at home.com (Tim Peters)
Date: Sun, 11 Mar 2001 01:12:23 -0500
Subject: grabbing return codes from os.system() call
In-Reply-To: <pjEq6.2038$54.2165@www.newsranger.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCKENCJEAA.tim.one@home.com>

[Gregory Jorgensen]
> Actually the current docs at python.org say that Windows 95 and 98
> always return 0. Apparently Windows NT and 2000 do return something.
> I don't know about Win ME. The Beazley book "Python Essential Reference"
> (which I used) says "On Windows, the exit code is always 0" in the
> description of os.system.

On any Windows box,

    Python's os.system() calls Microsoft's system().
    Microsoft's system() constructs a command line, and
        effective executes
            %COMSPEC% /c the_command_line
        in a new shell.
    The return value of the shell (%COMSPEC%) is returned
        by Microsoft's system(), which is in turn returned by
        Python's os.system().

My Win98SE box is typical of all Win9X systems:

C:\Code>echo %COMSPEC%
C:\WINDOWS\COMMAND.COM

C:\Code>

While it doesn't appear to be documented anywhere, command.com returns 0 no
matter what happens.

NT/Win2K generally use cmd.exe instead, and that returns the same exit code
as the command line it runs.

There is no way to work around the Win9X behavior except via installing a
different DOS shell and setting %COMSPEC% to point to that.




From clee at gnwy100.wuh.wustl.edu  Mon Mar  5 23:23:08 2001
From: clee at gnwy100.wuh.wustl.edu (Christopher Lee)
Date: 05 Mar 2001 22:23:08 -0600
Subject: PIL and Windows file formats?
References: <3AA45311.150DB844@alcyone.com> <GdZo6.24467$1D5.1007863@e420r-atl1.usenetserver.com>
Message-ID: <7u8zmjil03.fsf@gnwy100.wuh.wustl.edu>

Is the PIL directory part of your python path?  The png image plugin is
only loaded automatically if the PIL directory is on your pythonpath.

[Are you perhaps using PIL as a package (as in import PIL.Image )?]

You can load the plugin yourself.  In a message on the image-sig list, /F
suggested: 

    from PIL import Image, PngImagePlugin # etc.

-chris


From aleaxit at yahoo.com  Sat Mar  3 15:29:55 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sat, 3 Mar 2001 21:29:55 +0100
Subject: I come to praise .join, not to bury it...
References: <mailman.983485923.1136.python-list@python.org> <97p7iq$kk2$1@nntp6.u.washington.edu>
Message-ID: <97rka002k9@news1.newsguy.com>

"Russell E. Owen" <owen at astrono.junkwashington.emu> wrote in message
news:97p7iq$kk2$1 at nntp6.u.washington.edu...
    [snip]
> However, I do agree that join is not intuitive. I believe the problem
> (at least for the way I look at it) is that join should be a list
> method, not a string method. I.e.:
>
>   joined_string = ['a', 'b'].join(', ')
>
> makes a lot of sense to me. The current join string method does not.

This seems to be a widespread opinion, and I've already tried
to explain why my take on it differs, but that was a few months
ago, and apparently the current crop of discussants hasn't read
those discussions, so, let's give it one more spin.

Python does not have multi-methods: it's a single-dispatch
language.  Thus, quite apart from minor issues of syntax sugar,
there is ONE important difference between:

    paak.join(klop)                # form 1
    klop.join(paak)                # form 2
    amodule.join(paak, klop)  # form 3

Form 1 states: this .join call CAN be directly polymorphic on
(==directly dispatch on) paak, but NOT on klop -- if it's at
all polymorphic on klop, it must be by accessing it through
some standard interface ('interfaces' are not formalized in
Python, or Smalltalk, but, informally, they do exist and are
widely used -- "this argument must be a file-like object
and implement method write", "this must be a sequence",
etc).

Form 2 states the reverse: this .join call CAN be directly
polymorphic on (==directly dispatch on) klop, but NOT on
paak -- if it's at all polymorphic on paak, it must be by
accessing it through some standard interface.

Form 3 states that this call (on a module function) is not
directly polymorphic on _either_ klop or paak, though it
may access either or both through standard interfaces and
thus attain a (constrained) degree of polymorphism on
either or both of them.

It would be different in a multi-dispatch language, of
course, but Python isn't one (nor is Smalltalk).

So, when one states "join should be a method of klop, not
of paak" ('form 2 is better than form 1'), one is really
stating something like...: "this .join may more easily need
to be generally-polymorphic on klop, while, if it needs any
polymorphism on paak, it will more reasonably be able to
obtain it indirectly by accessing paak through some standard
interface".
If one expresses a preference for form 3 over either 1 or 2,
then one is stating that general polymorphism is not very
necessary on either argument, that standard interfaces
will do the job satisfactorily enough if any polymorphism at
all is needed.

I hope we do agree so far -- it doesn't seem to me that,
so far, there is really anything contentious or debatable
in what I've written.  If one considers syntax sugar in se
and per se more important than polymorphic semantics,
of course, then one _might_ think my assertions so far
are contentious and debatable -- but I wouldn't be very
interested in debating them: for me, the importance of
substance always dominates that of mere form... I'd
rather read a great novel in a poor, cheap binding, than
trash hand-bound in the best-quality leather.  Those who
disagree are welcome to their opinions (and particularly
their expensively-bound trash literature!-) but we'll never
manage to convince each other of anything anyway.


So, back to the specific discussion: Python 2 implicitly
asserts "when a string ``joiner'' is used to join the items
in a sequence of strings, it's more important than the
method be polymorphic on the joiner-object; polymorphism
on the sequence of strings can be obtained indirectly by
accessing it through a standard interface".

You guys who disagree are implicitly asserting that full
fledged polymorphism over the sequence of strings being
joined is more important -- than either no polymorphism
at all is needed on the joiner object, or that what is needed
can be obtained by accessing it through a standard interface.

So, let's compared these implicit underlying assertions --
what kind of polymorphism is needed and where.


Say that I'm implementing a sequence-like object.  Would
having .join as a part of the set of methods I must write
be an _advantage_ to me?  As things stand now, to "be a
sequence" I must just implement __getitem__ in the
appropriate manner -- accepting an index from 0 to N-1,
and raising IndexError for indices outside the range -- and
maybe implement __length__ to return N.  With this small
amount of effort, I gain the ability to be used as sequence
in any existing context -- from the for statement onwards.

For example, if my object gets passed to a file-like object's
writelines method, then, voila, my string items, which I
return in sequence from __getitem__, get 'written' to the
file-like object one after another, in order -- I don't have
to do anything peculiar to allow that, nor am I given any
chance to interfere with the procedure in case of need --
that's what it means to get polymorphism by being accessed
through a standard interface, rather than direct, generalized
polymorphism.  The standard interface makes you work much
less (you implement it once, and it comes in handy for many
kinds of client code), but it gives you no say on what is to
happen in _specific_ situations -- you're not even informed
for what exact purpose / in what exact contect your standard
interface's methods are being called.

I'm pretty happy not having to worry about "how are my
items to be joined [written one after another] when I am
passed to writelines", etc -- I implement the sequence
standard interface, and I get 'for free' the ability of being
polymorphically used as a sequence "everywhere".  Cool!

At least in all use cases I've ever found, my object is
quite happy NOT being told 'your items are to be written
one after another to this file-like object, do whatever is
appropriate' -- rather, it's asked for its items one after
the other, and the asker does whatever is appropriate --
the sequence object's task is JUST to be a sequence, to
present the items appropriately and let it be known when
there are no more.

Is the case 'your items are to be made into a single
string by being joined by this joiner-object' -- the join
method -- all that different from the writelines method
case?  I don't think so -- why would I _want_ to do
something different and peculiar when what is to be
done with my items (one after another in sequence)
is to be concatenated into a string rather than written
into a file?  In either case I'll just present my items
in sequence, one after the other, and whatever object
is asking for them will do whatever operation it needs
to do.  So, it's quite appropriate for the sequence to be
passed as an argument to .writelines, .join, etc; all the
polymorphism needed on the sequence object is well
handled by accessing it through the standard sequence
interface.


What about the joiner object?  Could IT be accessed
through a standard interface, rather than provide the
general polymorphism via dispatch?
Well, WHAT standard interface would that be?  What
methods would it provide, and for what purposes?
I can't think of any appropriate interface except one
with a .join method -- which would rather beg the
question, since a .join method is exactly what we're
trying to design...!  We could _name_ it differently
in a standard interface, but, what for?  We may as
well accept the name 'join'.

So, IF we need any polymorphism at all on the part
of the joiner-object, it had better be general dispatch
polymorphism -- either that, or no polymorphism at
all.  Well, IS any polymorphism at all useful on the
joiner object...?


As it happens, it is.  As for other string methods, for
example, the fact that .join is a method of string
objects (and, polymorphically, Unicode strings) saves
the type-switching that a module-level function would
have to do, each and every time, to distinguish whether
single-byte character strings or Unicode ones are
involved.  But, there are other opportunities -- consider,
for example:

class MixedJoiner:
    def __init__(self, joiner1, joiner2):
        self.joiner1 = joiner1
        self.joiner2 = joiner2
    def join(self, sequence):
        return joiner2.join((
            joiner1.join(sequence[:-1]),
            sequence[-1]))

Now, if we get sequences of strings and want to join
them in such wise that the LAST joining uses a different
joiner than the others, we just have to instantiate the
appropriate MixedJoiner instance and use it:

>>> comma_and = MixedJoiner(', ',' and ')
>>> print comma_and.join(('wine','women','song'))
wine, women and song
>>> print comma_and.join(('spam','butter','bacon','eggs'))
spam, butter, bacon and eggs

Now, any piece of client code written to use a joiner
object polymorphically through its join method will
work just as well with our MixedJoiner instance:

def print_joined(joiner, *sequence):
    print joiner.join(sequence)

we can now call
>>> print_joined(comma_and, 'wine', 'women', 'song')
to get the output of:
wine, women and song
just as we call it to get the printing to use space-joining:
>>> print_joined(' ','wine', 'women', 'song')
the latter emitting, of course:
wine women song


So, since we want general polymorphism on the joiner
object, but are quite content with polymorphism through
the standard sequence interface on the sequence object,
it is _just right_ that .join be a method on the joiner
object (e.g., a string) and that it take the sequence of
string to be joined as its argument.

I have not heard ANY _technical_ arguments opposed to
this last time the discussion went around -- nothing but
vague aesthetic, "it should be the other way", "I find it
ugly" kind of complaints.  Unless and until further notice,
then, I class these together with the complaints of people
who dislike 0-based indexing, on similar vague bases -- as
0-based indexing _works_ better than 1-based, then, for
me, it is a superior choice.  Aesthetics is in the eye of the
beholder, and the beholder _can_ be retrained and grow
a 'technically better and more appropriate' view of the
world -- one that will let him or her 'see' as 'beautiful'
those approaches which offer maximal effectiveness and
simplicity.  Technical issues such as where we want general
polymorphism versus standardized-interface access, just
like ones about arithmetic, tend to be more persistent, as
well as applying more generally, and impersonally, while
aesthetics are more subjective, individual, and fickle.


Alex





From tim.one at home.com  Thu Mar  1 18:47:58 2001
From: tim.one at home.com (Tim Peters)
Date: Thu, 1 Mar 2001 18:47:58 -0500
Subject: PEP 227 (was Re: Nested scopes resolution -- you can breathe again!)
In-Reply-To: <KNSz0UAhfpn6EwZj@jessikat.fsnet.co.uk>
Message-ID: <LNBBLJKPBEHFEDALKOLCGEDDJDAA.tim.one@home.com>

[Jeremy Hylton]
> Or: global NAME in FUNCTION
>
> def eggs():
>     z = 2
>     def spam():
>         def inner():
>             global x in spam, z in eggs
>             x += 1
>             z += x
>         x = 23
>         inner()

[Robin Becker]
> I like that a lot better than having it default possibly wrongly, but
> how would this be doable/spelled in the lambda case?

Wouldn't.  Python lambdas are restricted to a single expression, and all
(re)binding constructs in Python are (non-expression) statements.  So there's
no possibility to do a binding inside a Python lambda anyway (note that the
point of the above was to provide a means for changing names in outer scopes;
but lambdas can't even bind names in the local scope).




From aahz at panix.com  Mon Mar 26 10:31:11 2001
From: aahz at panix.com (Aahz Maruch)
Date: 26 Mar 2001 07:31:11 -0800
Subject: Threading legality/morality
References: <fsDv6.1247$NR.100201@news3.oke.nextra.no>
Message-ID: <99nnbv$phs$1@panix6.panix.com>

In article <fsDv6.1247$NR.100201 at news3.oke.nextra.no>,
Syver Enstad <syver at NOSPAMcyberwatcher.com> wrote:
>
>How is calling print from more than one thread looked upon?

Depends what you're trying to do.  If it's debugging output and if you
concatenate strings before trying to do a print (don't use multiple
prints in succession), you're safe.  Anything other than that is
probably risky -- but unlikely to do actual damage (you'll just get
messed up output).
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

"I won't accept a model of the universe in which free will, omniscient
gods, and atheism are simultaneously true."  -- M


From digitig at cix.co.uk  Fri Mar  9 15:34:00 2001
From: digitig at cix.co.uk (Tim Rowe)
Date: Fri, 9 Mar 2001 20:34 +0000 (GMT Standard Time)
Subject: XML Module problems
References: <j4y9uhnb1m.fsf@informatik.hu-berlin.de>
Message-ID: <memo.20010309203450.37329B@digitig.compulink.co.uk>

In article <j4y9uhnb1m.fsf at informatik.hu-berlin.de>, 
loewis at informatik.hu-berlin.de (Martin von Loewis) wrote:

> digitig at cix.co.uk (Tim Rowe) writes:
> 
> > That's what I have with my (ActiveState) 2.0 installation. 
> > xml.parsers.expat is very like pyexpat but with some subtle 
> > differences 
> 
> ??? xml.parsers.expat should be in the file xml/parsers/expat.py,
> which should read
> 
> from pyexpat import *
> 
> So I'd call it *highly* unlikely that they are different.

<snip>

So it does! I only noticed that Pyxie (the DOS version on Sean McGrath's 
book, not the rpm) broke because:

a. It couldn't find pyexpat (though evidently xml/parsers/expat.py 
manages) and

b. When I changed pyxie to point to xml.parsers.expat, it found that the 
attribute list was a dictionary, not a list.

The version on the Pyxie site worked, though, and it's rather too 
important for me at the moment for me to re-install the broken version to 
investigate further!


From fredrik at pythonware.com  Tue Mar 13 05:29:02 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Tue, 13 Mar 2001 10:29:02 GMT
Subject: module importing and variable syntax
References: <3AADF3DC.1030003@dynworks.com>
Message-ID: <ODmr6.10$4d3.2359@newsb.telia.net>

Jeff Davis wrote:
> What I am trying to do is have an arbitrary string with an arbitrary
> type (type with respect to my program, not python) access the function
> associated with the string in the module associated with the type.

look for __import__ and getattr under built-in functions
in the library reference.

Cheers /F




From dkuhlman at rexx.com  Fri Mar 30 12:18:58 2001
From: dkuhlman at rexx.com (Dave Kuhlman)
Date: Fri, 30 Mar 2001 17:18:58 -0000
Subject: Best Python documentation ->HTML?
References: <pg0qbt0e598g1ecnfj5lj0lpa1q1pqvagq@4ax.com>
Message-ID: <985972738.615308@rexx.com>

Go to Vaults of Parnassus (http://www.vex.net/parnassus/), then
search for "gendoc".

gendoc worked pretty well for me.

It looks like Pythondoc is a newer replacement for gendoc.

  - Dave



Brian Glendenning <bglenden at hotmail.com> wrote:
> 
> I am a Python newbie and am confused about what the best way to turn
> Python documentation into HTML is (i.e., a "javadoc" equivalent). If
> it could also make good printed documentation (e.g., page numbers) as
> well as HTML, all the better.
> 
> Pointers and suggestions welcome! -
> 
> Cheers,
> Brian Glendenning, bglenden at nrao.edu

-- 
Dave Kuhlman
dkuhlman at rexx.com


From carribeiro at yahoo.com  Fri Mar 23 17:27:34 2001
From: carribeiro at yahoo.com (Carlos Ribeiro)
Date: Fri, 23 Mar 2001 19:27:34 -0300
Subject: New PEP: The directive statement
In-Reply-To: <j4k85gkiye.fsf@informatik.hu-berlin.de>
References: <mailman.985085006.24236.python-list@python.org>
 <slrn9bk7fk.fnm.scarblac@pino.selwerd.nl>
Message-ID: <5.0.2.1.0.20010323183750.009f2510@mail.inet.com.br>

I have some questions that I collected around several posts that aren't 
fully explored. First the questions, the some comments of mine:

1) Topic: People seems to be divided over the value of "introspection".

1.1) Is introspection a desirable feature of any "directive-like" solution?
1.2) Do we have time to implement introspection in the "directive" patch?
1.3) If so, isn't sufficient to implement a global dictionary called 
__directive__ to store it?

2) Topic: Right now, directives operate on a per-module basis.

2.1) What's going to happen in a multi module project with different 
directive settings for each module?
2.2) What happens if one function inside a module calls a function on a 
module where the directive setting is different?

3) Topic: In the future directive may become a keyword, and it will be 
possible to place directives anywhere inside the source code.

3.1) Will we be able to set/reset/test/save directives (features?)?
3.2) What about directive scoping? Some directives may be valid at module 
scope, other for a class, others for a function, others anywhere.
3.3) If we have different scopes for some directives, how are we going to 
handle introspection? A global dictionary is not enough anymore.

4) Topic: Directives have both runtime and compile-time behavior.

4.1) Do we need directives for things that may be implemented using *only* 
runtime behavior?

5) Topic: CPython x The world

5.1) Is it possible to patch directives into all Python implementations?
5.2) Are we going to see directives that apply only to some selected 
platforms or implementations?


My comments:

a) Directives should only be used to control compile time behavior. This 
solve a lot of the open issues. One's code runtime behavior should not 
*ever* depend upon directive settings. Otherwise, we're going to have 
*lots* of scope related problems. The only allowable exception would be the 
use for documentation purposes. Anything else may be implemented using 
"flags", be it through parameters, global variables, etc.

b) Directives *may* be useful in different scopes - module, class or 
function. Restriction (a) helps to minimize any scoping problems.

c) Please keep the proposal *simple*.


Man, this is turning into a nightmare...


Carlos Ribeiro
cribeiro at mail.inet.com.br
carribeiro at yahoo.com


_________________________________________________________
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com




From lluang at northwestern.edu  Wed Mar 14 15:19:15 2001
From: lluang at northwestern.edu (Louis Luangkesorn)
Date: Wed, 14 Mar 2001 14:19:15 -0600
Subject: New windows installer for mxODBC works (was: cursor.execute fails under 
 mxODBC)
References: <3AABF77F.7B77D51@northwestern.edu> <3AAE59AE.630C148C@lemburg.com> <3AAEF113.E928A434@northwestern.edu> <3AAFBC7C.2A1CD536@lemburg.com>
Message-ID: <3AAFD243.3EDAF2A3@northwestern.edu>

[It looks like you were trying to send this to the newsgroup so I'll forward my
response there]

"M.-A. Lemburg" wrote:

>
> I've uploaded a new version of the Windows installer to:
>
> http://www.lemburg.com/files/python/egenix-mx-commercial-2.0.0.win32-py2.0.exe
>
> There seems to have been some build error during the release
> session which was introduced after the final tests. Please check
> whether you still get the indicated problems with the new
> build.
>
> One thing this teaches is that distutils should probably clear
> the build directory before rebuilding a package... I'll add that
> to my batch file.
>
> Thank you for looking into this and sorry for the trouble
> this has caused you. I hope the new build works for you as well
> (it does for me on Windows 9x -- don't have a W2k to test on).
>
> --
> Marc
> --
> Marc-Andre Lemburg
> ______________________________________________________________________
> Company & Consulting:                           http://www.egenix.com/
> Python Pages:                           http://www.lemburg.com/python/

The new installer works!  Thanks for fixing it.

Louis
--
K Louis Luangkesorn
lluang at northwestern.edu  http://pubweb.nwu.edu/~kll560  PGP:0xF3E2D362
Whatsoever things are true, ... honest, ... just, ... pure, ... lovely, ... of good
report; if there be any virtue, and if there be any praise, think on these things.-
motto - Northwestern University




From rwklee at home.com  Sat Mar 24 23:39:15 2001
From: rwklee at home.com (Rick Lee)
Date: Sun, 25 Mar 2001 04:39:15 GMT
Subject: static methods
References: <mailman.985474834.2256.python-list@python.org>
Message-ID: <3ABD7673.9D447A32@home.com>

Thanks, and thanks to Alex.  This is very useful, also very telling that it
is pretty obscure.  For me, static, or class-wide methods are very useful.
With this, you only have to pass the class reference to the caller.
Otherwise, you have to pass both the class reference and the module
reference to the caller.

- Rick Lee

"Clark C. Evans" wrote:

> On 24 Mar 2001, Marcin 'Qrczak' Kowalczyk wrote:
> | Fri, 23 Mar 2001 14:27:42 -0800, Paul Prescod <paulp at ActiveState.com>:
> | | Probably because they are simply not neccessary.
> | |
> | | def _static():
> | |     XXX.foo=bar
> | |     return XXX.foo
> | |
> | | class Class:
> | |     staticmethod = _static
> | |     ...
> |
> | It doesn't work. Try to call it.
> |
> | The fact that you are not the first person who thinks it works
> | suggests that the current Python's solution is not ideal :-)
>
> >>> class Class:
>         class StaticFudge:
>             def __call__(self):
>                 print "I am a static"
>         staticmethod = StaticFudge()
> >>> Class.staticmethod()
> I am static
>
>     It took much looking and many wrong answers before
>     Alex Martelli pointed me to this solution.
>

<snip>



From niemeyer at conectiva.com  Thu Mar 15 08:36:58 2001
From: niemeyer at conectiva.com (Gustavo Niemeyer)
Date: Thu, 15 Mar 2001 10:36:58 -0300
Subject: Linuxconf modules in python
Message-ID: <20010315103658.F28764@tux.distro.conectiva>

Hello people!!

I'd like to announce the release of Pythonmod, a Linuxconf module
written in C++ that embeds and extends python, allowing one to
build new modules for Linuxconf with just a few lines of Python.
Most of the functionalities available in the C++ interface have
been wrapped and are available at the python layer.

As usual in Linuxconf, when you build a python module you don't
have to care about every interface available. The same python
code will make the module available in all interfaces (web, X,
console, and java).

I must write some documentation about it, but the best way to
understand how it works is probably to take a look at the example
included in the modules/pythonmod/examples directory of Linuxconf's
sources (pythonmod is already included in standard Linuxconf).

To make a module be loaded by Linuxconf, just drop the python file
inside /usr/lib/linuxconf/pythonmod. If you are using the treemenu,
remove /var/state/linuxconf/treemenu.cache to reset the treemenu
cache (I must find a better way to do this).

Please, contact me if you have any doubts or would like more
information about it (#linuxconf at irc.openprojects.net is also
a good place).

Best regards!!

-- 
Gustavo Niemeyer

[ 2AAC 7928 0FBF 0299 5EB5  60E2 2253 B29A 6664 3A0C ]



From bbhaydon at bigpond.com  Sat Mar 24 21:50:40 2001
From: bbhaydon at bigpond.com (Brett Haydon)
Date: Sun, 25 Mar 2001 12:50:40 +1000
Subject: Please rate these Python books
References: <mailman.985443811.30792.python-list@python.org>
Message-ID: <%Pcv6.2770$45.12741@newsfeeds.bigpond.com>

Trumped by a Joker...
"Jonathan Soons" <jsoons at juilliard.edu> wrote in message
news:mailman.985443811.30792.python-list at python.org...
>
> "Snap!" is not understood outside former Commonwealth countries.
> It's use causes culture shock.
>
...Sorry now I'm getting images of Monty Pythons "and now for something
different" phrase book sketch.

Is Monty Python universal??

Brett

--Warning--.
This email and thread will combust in 5 iterations to prevent serious
digression.




From syver at NOSPAMcyberwatcher.com  Fri Mar 30 15:23:18 2001
From: syver at NOSPAMcyberwatcher.com (Syver Enstad)
Date: Fri, 30 Mar 2001 22:23:18 +0200
Subject: Python scripts as Drop targets in Windows Explorer
References: <99qogo$o6g$1@troll.powertech.no> <99qqq3$qt7$0@216.39.170.247> <99r043$qqr$1@troll.powertech.no> <vdqs99.p16.ln@haegar.fqdn.th-h.de> <8kAw6.780$R6.22057@news1.oke.nextra.no> <gi7v99.hp.ln@haegar.fqdn.th-h.de>
Message-ID: <9a2q3p$maq$1@troll.powertech.no>

"Michael Linnemann" <m.linnemann at addcom.de> wrote and wrote
news:gi7v99.hp.ln at haegar.fqdn.th-h.de...
> Am Thu, 29 Mar 2001 08:10:08 +0200 schrieb Syver Enstad:

> It is not really difficult, especially for me in particular. I don ot
> care about registries because I have none :-)

Aaah, so it's not difficult for you, you just enjoy being difficult. I see.
Nice to know anyway.





From leo at iems.nwu.edu  Fri Mar  2 15:14:24 2001
From: leo at iems.nwu.edu (Leonardo B Lopes)
Date: Fri, 02 Mar 2001 14:14:24 -0600
Subject: docs for mysqldb
References: <3A9D63BC.C3EB9246@iems.nwu.edu> <gcxxuDAA4sn6Ewsc@gol.com>
Message-ID: <3A9FFF20.CB9D0D3B@iems.nwu.edu>

Yes, my dist. also came with docs, but cursor.execute() and
cursor.executemany() are not documented there. The user is referred to
the Python DBI 2.0 docs, which say that the convention for passing
parameters is open to the implementer. Searching the web for examples
show some orcale example where parameters are referred to as ":1", ":2". 

The examples in the dist have a regular pythonlike "%s" placeholder, but
the docs talk about defining types of parameters beforehand to speed
things up, and I can't see how that applies if we're just doing the
equivalent of a sprintf operation.

Leo.

Ian Parker wrote:
> 
> In article <3A9D63BC.C3EB9246 at iems.nwu.edu>, Leonardo B Lopes
> <leo at iems.nwu.edu> writes
> >Sorry for the crossposting, but this actually interests both lists...
> >
> >Does anyone use the python module mysqldb? Does anyone have any decent
> >docs for it? In particular, how do you do cursor.execute() with
> >different parameters?
> 
> The MySQLdb-0.2.2.tar.gz kit that I downloaded includes both
> documentation and examples.  IIRC correctly, I got it from
> http://home.t-online.de/home/err666/
> 
> Checking again, it looks like there are later version available.
> 
> Regards
> 
> Ian
> --
> Ian Parker

-- 
=======================================================================
Leonardo B. Lopes                                      leo at iems.nwu.edu 
Ph.D. Student                                             (847)491-8470
IEMS - Northwestern University             http://www.iems.nwu.edu/~leo


From qrczak at knm.org.pl  Mon Mar 19 14:56:33 2001
From: qrczak at knm.org.pl (Marcin 'Qrczak' Kowalczyk)
Date: 19 Mar 2001 19:56:33 GMT
Subject: [Very Long (11K)] Numeric PEPs, first public posts
References: <mailman.984748941.31248.python-list@python.org> <Q5Ns6.432845$ge4.152574719@news2.rdc2.tx.home.com>
Message-ID: <slrn9bcp3h.ns2.qrczak@qrnik.zagroda>

Sat, 17 Mar 2001 17:25:04 GMT, Rainer Deyke <root at rainerdeyke.com> pisze:

> A simpler way to solve this problem would be to standardize Python
> ints as 32 bit.

I would definitely not do it (especially as 64-bit platforms will be
more popular as the time goes). Relying on the size of a type called
int is evil. Call a type/class int32 if you want to guarantee that
it's exactly 32 bits.

-- 
 __("<  Marcin Kowalczyk * qrczak at knm.org.pl http://qrczak.ids.net.pl/
 \__/
  ^^                      SYGNATURA ZAST?PCZA
QRCZAK


From othello at javanet.com  Mon Mar 12 20:11:11 2001
From: othello at javanet.com (Raymond Hettinger)
Date: Mon, 12 Mar 2001 20:11:11 -0500
Subject: Help with Regular Expressions
References: <mailman.984440295.31538.python-list@python.org>
Message-ID: <3AAD73AE.86CBA2AE@javanet.com>

Tim Peters wrote:

> [Raymond Hettinger]
> > Is there an idiom for how to use regular expressions for lexing?
> >
> > My attempt below is unsatisfactory because it has to filter the
> > entire match group dictionary to find-out which token caused
> > the match. This approach isn't scalable because every token
> > match will require a loop over all possible token types.
> >
> > I've fiddled with this one for hours and can't seem to find a
> > direct way get a group dictionary that contains only matches.
>
> That's because there isn't a direct way; best you can do now is seek to order
> your alternatives most-likely first (which is a good idea anyway, given the
> way the engine works).
>
> If you peek inside sre.py (2.0 or later), you'll find an undocumented class
> Scanner that uses the undocumented .lastindex attribute of match objects.
> Someday I hope this will be the basis for solving exactly the problem you're
> facing.  There's also an undocumented .lastgroup attribute:
>
> Python 2.1b1 (#11, Mar  2 2001, 11:23:29) [MSC 32 bit (Intel)] on win32
> Type "copyright", "credits" or "license" for more information.
> IDLE 0.6 -- press F1 for help
> >>> import re
> >>> pat = re.compile(r"(?P<a>aa)|(?P<b>bb)")
> >>> m = pat.search("baab")
> >>> m.lastindex  # numeral of group that matched
> 1
> >>> m.lastgroup  # name of group that matched
> 'a'
> >>> m = pat.search("ababba")
> >>> m.lastindex
> 2
> >>> m.lastgroup
> 'b'
> >>>
>
> They're not documented yet because we're not yet sure whether we want to make
> them permanent parts of the language.  So feel free to play, but don't count
> on them staying around forever.  If you like them, drop a note to the effbot
> saying so.
>
> for-more-docs-read-the-source-code-ly y'rs  - tim

Thanks Tim,

I changed the last function line to:
        return (m.lastgroup, m.group(m.lastgroup))
and it worked perfectly.

I'll try out the undocumented variables and send back a recommendation.
Right now, I like them but would prefer them as functions.  Possibly better
is to have a variation of  m.groupdict() or flag that causes it to return a
dictionary
of matches, instead of None for non-matches.

always-amazed-at-how-much-the-bots-know
Raymond




From NoSpam at NoSpam.com  Thu Mar  8 12:53:17 2001
From: NoSpam at NoSpam.com (Tom)
Date: Thu, 08 Mar 2001 17:53:17 GMT
Subject: Memory leaks in VC6 2 line Python 2.1b1 apps?
References: <3aa88f52.5419185@news.newsguy.com>
Message-ID: <hGPp6.53719$UZ4.13963947@news4.rdc1.on.home.com>

"Owen F. Ransen" <ransen_spam_me_not at nemo.it> wrote in message
news:3aa88f52.5419185 at news.newsguy.com...
> Python 2.1b1
>
> I tried to create a simple VC MFC app using the
> wizard as follows:
>
> 1) Multidoc shared MFC-DLL application.
> 2) Run time library debug multithreaded DLL
> 3) Added #include <python.h> to the main file
> 4) Added
>         Py_Initialize()
>         Py_Finalize()
>    at the end of the InitInstance()
> 5) When I run I get memory leaks visible at
>    exit.
>
> Any clues anyone?

For comparison sake.
I use VC6 and Python 2.0.  I create an EXE that calls the same same
functions that you mention, and Python itself doesn't leak (I get various
wierd leaks from other sources, but not Python so far as I can tell).

Tom.

>
> (I can't yet report this via Source Forge because my account
>  problems have not be sorted out yet.)
>
>
> --
> Owen F. Ransen
> http://www.ransen.com/
> Home of Gliftic & Repligator Image Generators




From akuchlin at mems-exchange.org  Fri Mar 23 18:07:23 2001
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 23 Mar 2001 18:07:23 -0500
Subject: Good Editor?
References: <mailman.985291768.27561.python-list@python.org> <JStu6.1735$SB2.192020@ruti.visi.com>
Message-ID: <3dk85gythw.fsf@ute.cnri.reston.va.us>

Everyone's aware that there's an Editor HOWTO, right?
(http://py-howto.sourceforge.net/editor/) And that it could always use
more information on how to configure various editors for writing Python code?

--amk



From sburr at home.com  Tue Mar 20 11:08:49 2001
From: sburr at home.com (Steven Burr)
Date: Tue, 20 Mar 2001 16:08:49 GMT
Subject: Functionalism, aesthetics Was:(RE: I come to praise .join, not to bury it...)
References: <mailman.985067371.2582.python-list@python.org> <997fp901i44@news2.newsguy.com>
Message-ID: <sburr-2003010908490001@cx937488-a.chnd1.az.home.com>

In article <997fp901i44 at news2.newsguy.com>, "Alex Martelli"
<aleaxit at yahoo.com> wrote:

<snip (to state the obvious)>

>                            I might even suggest that, in
> some fields, a productive strategy might be to proceed to
> determine by traditional means all that the users _say_
> they want, then do exactly the opposite (but that would
> be pure provocation on my part, so I won't:-).

I'd like to know where you got a copy of our IT department's mission
statement. :)


From parkw at better.net  Mon Mar 12 17:03:56 2001
From: parkw at better.net (William Park)
Date: Mon, 12 Mar 2001 17:03:56 -0500
Subject: This math scares me
In-Reply-To: <3aad4062.1290021@News.CIS.DFN.DE>; from costas@springmail.com on Mon, Mar 12, 2001 at 09:34:58PM +0000
References: <3aad4062.1290021@News.CIS.DFN.DE>
Message-ID: <20010312170356.B3513@better.net>

On Mon, Mar 12, 2001 at 09:34:58PM +0000, costas at springmail.com wrote:
> Ok, I can see maybe division having problems. But why does addition of
> the two numbers below:
> 
> 5.01+5.54
> 
> give me this?
> 
> 10.550000000000001

It's called 'Round-off error'.

---William Park, Open Geometry Consulting, Linux/Python, 8 CPUs.



From paulp at ActiveState.com  Thu Mar 15 08:43:38 2001
From: paulp at ActiveState.com (Paul Prescod)
Date: Thu, 15 Mar 2001 05:43:38 -0800
Subject: Who's minister of propaganda this week?
References: <B98041329020E007.7DA84CAD9D0562AB.817BBC572F480869@lp.airnews.net> <98nu1m$12e@dispatch.concentric.net> <660ACCA1E4816F7E.C359CDF4168EF3FB.3B6B05538D619770@lp.airnews.net> <3AAFA13D.76E32C3F@divalsim.it> <98oa4o01i4v@news2.newsguy.com> <3AB0176D.8070903@yahoo.com>
Message-ID: <3AB0C70A.C6A55DF5@ActiveState.com>

Michael Chermside wrote:
> 
>...
> I'm really not sure I see it this way. If the method  foo(x) is known to
> take a FancyDateObject there are three kinds of errors we could make. 
> One is that foo() is written badly so it doesn't do what it's supposed 
> to. The unit tests of foo() need to guard against this.
> Another possible error is that FancyDateObject isn't written properly.
> The unit tests of FancyDateObject need to guard against this. And 
> the third type of error is that that somewhere where we CALL foo(), 
> we might pass it a DateObject instead... or even a String... which 
> will cause it to perform wrong. To guard against this in a
> dynamically typed language, we have to write unit tests for *every
> single place* that we call foo(). 

Your analysis is incomplete in two ways:

1. I don't think it makes sense to speak of writing a unit test for "one
thing calling another". You test the first thing, you test the second
thing and if the second thing depends upon the first then you've
implicitly tested the connection between them.

2. In object oriented programming languages, we can't just test one
function against one input object class and be confident that the code
is correct. There are an infinite number of potential subclasses of
FancyDateObject. Therefore we should NOT depend on testing every single
one of them with foo. We just need to make sure that foo is correct,
that each FancyDateObject is correct and that the communication protocol
between them is well-understood and tested by both.

3. Statically typed languages are absolutely not immune to type errors
for a variety of reasons:

 * Null pointer exceptions are type errors. You've passed the null
object when the code expected some other object. 

 * Almost every statically typed language also has a set of features for
working around the type system (dynamic casts). 

 * Even when the compiler thinks the "type" of your object is correct,
there are a variety of type-related errors you can make such as passing
a negative number where only positive ones are allowed, passing an
gopher: URL where only an http: URL is allowed, passing an absolute
pathname where only a relative pathname is allowed and so forth. If you
try to catch all of these in the static type checking system you wlil
spend more of your life doing conversions than actually coding. *PLUS*
you will need to do the runtime conversion-compatibility checks
YOURSELF!

4. Yes, if you unit test every unit in your program, and do good
coverage testing you WILL catch ever occurrence of some code calling
foo. That's what good testing is about. And you need it, not to catch
the basic type errors but to catch the subtle semantic errors. Catching
the basic type errors comes for free.

-- 
Take a recipe. Leave a recipe.  
Python Cookbook!  http://www.activestate.com/pythoncookbook



From python at commandprompt.com  Mon Mar  5 16:51:38 2001
From: python at commandprompt.com (python at commandprompt.com)
Date: Mon, 5 Mar 2001 13:51:38 -0800
Subject: OT: TUX issue 6 released
Message-ID: <20010305135138.A28011@crazypenguins.commandprompt.com>

Hello,

We posted this OT last time and a lot of people loved it. We just released a
new version so we thought we let people know.

http://www.linuxports.com/TUX

Nothing like a post-apocolyptic, nuclear mutated TUX to brighten your day.

:P



From ljohnson at resgen.com  Tue Mar 20 09:46:52 2001
From: ljohnson at resgen.com (Lyle Johnson)
Date: Tue, 20 Mar 2001 08:46:52 -0600
Subject: The most important modules
References: <Pine.GSO.4.21.0103191732210.22234-100000@laplace>
Message-ID: <tber6micgl2hd3@corp.supernews.com>

> I've seen a few comments along the lines of "When you are a newbie it will
> take a while to realize which the important modules are."  As someone who
> is in the middle of wading through Chapter 3 of the Library Reference, it
> would be very helpful to know which ones are particularly worth
> noting.  (Presumably  sys, (c)Pickle and shelve, from the ones I've so far
> in the chapter - but what other treats are out there for me?)

For writing scripts that deal with files (whether for system administration
or other tasks), you should get familiar with the "fileinput", "glob" and
"os" modules. All of Python's support for regular expressions comes in
through the "re" module, so that will be one you'll want for some
text-processing needs. And I've been meaning to teach myself how to use the
source-level Python debugger module, "pdb", but haven't gotten around to
that yet;

A lot of the other modules get into more application-specific needs, whether
it's for network support ("ftplib", "httplib", ...), XML processing
("xml.*") or others. But it's all good stuff. And of course the standard
Python library doesn't even scratch the surface of what's available; you'll
also want to check some of the other sources for contributed software, like
the Vaults of Parnassus:

    http://www.vex.net/parnassus/

Hope this helps,

Lyle





From phrxy at csv.warwick.ac.uk  Thu Mar 29 12:35:10 2001
From: phrxy at csv.warwick.ac.uk (John J. Lee)
Date: Thu, 29 Mar 2001 18:35:10 +0100
Subject: BLT extensions with BeOpen Python 2.0
In-Reply-To: <Pine.GSO.4.21.0103291300160.11408-100000@babbage>
References: <Pine.GSO.4.21.0103291300160.11408-100000@babbage>
Message-ID: <Pine.SOL.4.30.0103291832080.13267-100000@primrose.csv.warwick.ac.uk>

On Thu, 29 Mar 2001, Martyn Quick wrote:

> I've been trying to install the BLT extensions for Tk into my BeOpen
> distribution of Python 2.0.  However, it seems that the directory
> structure that the BLT distribution expects (see www.tcltk.com/blt) is
> different to that provided by the python distribution.
>
> Has anyone succeeded in installing BLT with the distribution, or do I need
> to switch distributions?  (and to which one?)

I'm pretty sure I had BLT working with BeOpen 2.0 at some point, at least
on linux, and perhaps windows too.  I couldn't tell you exactly where I
put it, though.  Make sure you have the right BLT version for the tcl that
comes with python 2.0 .


John




From syver at NOSPAMcyberwatcher.com  Thu Mar 15 15:46:03 2001
From: syver at NOSPAMcyberwatcher.com (Syver Enstad)
Date: Thu, 15 Mar 2001 21:46:03 +0100
Subject: Python and ADO
Message-ID: <98r9qn$tbi$1@troll.powertech.no>

Has anyone considered making a DB API compliant wrapper around MS ADO? Or
isn't this considered worthwhile?

As ADO is the database api that I use, I could have a look at making a
wrapper for it if there was any interest in it and if making such a thing is
not too involved.




From sholden at holdenweb.com  Thu Mar 22 22:25:53 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 22 Mar 2001 22:25:53 -0500
Subject: Good Editor?
References: <3ab9a32b$1_1@news2.one.net> <99cv2t$f20g$1@ID-11957.news.dfncis.de>
Message-ID: <sqzu6.12865$_46.475246@e420r-atl2.usenetserver.com>

"Emile van Sebille" <emile at fenx.com> wrote in message
news:99cv2t$f20g$1 at ID-11957.news.dfncis.de...
>
> For windows, I like textpad http://www.textpad.com/ beause it uses
> workspaces, has a hex display mode, runs code in a separate process,
> supports other languages (programming and spoken), and generally behaves
> well.
>
> Emile van Sebille
> emile at fenx.com
>
Does it show Unicode string constants correctly in hex?

regards
 Steve





From tim.one at home.com  Mon Mar 12 01:25:02 2001
From: tim.one at home.com (Tim Peters)
Date: Mon, 12 Mar 2001 01:25:02 -0500
Subject: popen problem
In-Reply-To: <98hgtt$5627@NWNEWS.PCT.EDU>
Message-ID: <LNBBLJKPBEHFEDALKOLCKEPPJEAA.tim.one@home.com>

[Marco Catunda]
> What's wrong with this code?
>
> ##########################
> import threading
> import os
>
> class TestPopen( threading.Thread ):
>
>    def run( self ):
>       while 1:
>          f = os.popen( "ls -l /" )
>          lines = f.read()
>          f.close()
>
> if __name__ == "__main__":
>    for i in range(50):
>       t = TestPopen()
>       t.start()
> ###########################
> [and it eventually dies w/ IOErrors in the f.close() calls]

You should report this as a bug on SourceForge, and make it very clear then
which version of Python you're using and under which operating system.  Best
guess is that your OS is simply running out of resources, but no way to tell
for sure based on what you've said here.




From prairiesasquatch at crosswinds.net  Sun Mar 18 23:31:06 2001
From: prairiesasquatch at crosswinds.net (Kevin Russell)
Date: Sun, 18 Mar 2001 20:31:06 -0800
Subject: text file to array and back
References: <990fgo$cdm$04$1@news.t-online.com>
Message-ID: <3AB58B8A.E1EAFA44@crosswinds.net>

Andreas Kremer wrote:

> Hi,
>
> i am trying to write a python class to make access to a collection of
> scientific analysis programs easier for me. In this case I need to read a
> file which consists of a time series, collect it into an array (numpy
> preferred) and after some operations write it back to disk as a text file.
> My solutions are attached. These are my questions:
> 1. as I am new to python i suspect my joining and splitting not to be very
> efficient, especially the joining.

The reason you're getting those square brackets and commas is that
you're converting the entire list for a line to a string *first*, and
only then worrying about joining it, which is too late, since you're
now dealing with just a single string.

You could use:
    dataline = string.join(map(str, line))

or:
    for line in array:
        for element in line:
            file.write("%s " % element)
        file.write("\n")

(though that's probably less efficient I/O-wise.)

or even, if you're willing to mess around with the definition of
sys.stdout:

    realstdout = sys.stdout
    sys.stdout = open(...)
    for line in array:
        for element in line:
            print element,
        print
    sys.stdout = realstdout

> 2. Is there a way while reading the file
> to store the data immediately into the array?

If you're asking "Is there a magical procedure that will already do
exactly what I want?", there probably is, but I don't know it.

If all you're asking is "Can I just read one line at a time instead
of reading the whole file into memory first?", then definitely.
In Python 2.1, just use xreadlines() instead of readlines().
In older versions of Python, you can get the same effect using
the fileinput module.


> 3. I am concerned about a memory leak. what happens to self.array if I read
> the file again? It obviously doesnt append the data to the old, but are the
> old data deleted from the memory?

Yes.  As soon as Python figures out that nobody is referring to the
old object any more, it will be garbage-collected.


>
> Thanks in advance. Andreas Kremer.
>
>     def __createarray(self):
>         "creates an array of the outputfile"
>         # open file, transfer content into list self.array
>         file = open(self.outputfile, 'r')
>         for line in file.readlines():
>             dataline = map(float, string.split(line))
>             self.array.append(dataline)
>         file.close()
>         # convert list self.array into an array (Numpy)
>         self.array = array(self.array, Float64)
>
>     def __writearray(self):
>         "writes array to outputfile"
>         # open file, format array and write it to disk
>         file = open(self.outputfile, 'w')
>         for line in self.array:
>             dataline = string.join(str(line),'')
>             dataline = dataline+'\n'
>             dataline = string.replace(dataline,'[','')
>             dataline = string.replace(dataline,']','')
>             dataline = string.replace(dataline,',','')
>             file.write(dataline)
>         file.close()



From aahz at panix.com  Mon Mar 19 13:24:53 2001
From: aahz at panix.com (Aahz Maruch)
Date: 19 Mar 2001 10:24:53 -0800
Subject: Subsetting a dictionary
References: <993dkm$ttr$1@news6.svr.pol.co.uk> <993qq3$4bajg$1@ID-11957.news.dfncis.de> <995hj6$dk6$1@newsg3.svr.pol.co.uk>
Message-ID: <995itl$2d2$1@panix2.panix.com>

In article <995hj6$dk6$1 at newsg3.svr.pol.co.uk>,
Will Newton  <will at nospam.misconception.org.uk> wrote:
>Emile van Sebille wrote:
>>
>> Rather than copying and then deleting from the copy, try building a new
>> dictionary while adding only those that qualify.
>
>On average this makes the code ~10x slower. I haven't done an in depth 
>analysis of how the routine is called, but my "backwards" way of doing 
>things seems much faster for my task.

While it's possible you're right, my gut instinct is that Emile's way
will be faster unless your subset almost always includes most of the
dictionary.
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

There's a difference between a person who gets shit zie doesn't deserve
and a person who gets more shit than zie deserves.  --Aahz


From bedge at troikanetworks.com  Fri Mar 23 12:52:49 2001
From: bedge at troikanetworks.com (Bruce Edge)
Date: Fri, 23 Mar 2001 09:52:49 -0800
Subject: sharing data between modules
References: <20010323.073801.76065818.26070@mead.troikanetworks.com> <99fvbq$mmg$1@panix3.panix.com>
Message-ID: <20010323.095248.1034949299.26070@mead.troikanetworks.com>

In article <99fvbq$mmg$1 at panix3.panix.com>, "Aahz Maruch" <aahz at panix.com>
wrote:

> In article <20010323.073801.76065818.26070 at mead.troikanetworks.com>,
> Bruce Edge <bedge at troikanetworks.com> wrote:
>>
>>As a C++ guy I tend to put each class in a different file. In python
>>this gives me a different module for each class. My app uses this group
>>of classes, ie: modules, what's the best way to share data between them?
>>
>>Currently I register the classes with each other using class static
>>data,  so they all have a reference to each other. Ideally, I'd like to
>>be able to have a few globals that every module can see.  
> 
> If you're sharing instance data, references are the best way to go.  If
> it's truly global data, one of the advantages of Python modules is that
> they *are* namespaces.  So either create a globals.py if variables are
> shared across multiple classes or set variables in your class modules if
> it's specific to a single class.

global.py! Argh, it so simple, why didn't I think of that. 
Probably because my sum total experience with python is almost a month
:-) 

Thank you! That makes my life much easier (and my code smaller).

-Bruce.


From tuttledon at hotmail.com  Thu Mar  8 15:13:14 2001
From: tuttledon at hotmail.com (Don Tuttle)
Date: Thu, 08 Mar 2001 20:13:14 GMT
Subject: news from python9!!!
References: <mailman.984074414.26983.python-list@python.org>
Message-ID: <uJRp6.46512$__6.10501931@typhoon.southeast.rr.com>

"Fredrik Lundh"
> (and the subject should probably read "news from active-
> state's marketing department" ;-)

Hey, you gotta run with what you can find, seeing how all you conventioneers
where having way too much fun to drop into the newsgroup and give us an
update! <wink>

Don

(Sorry about the award.  Maybe next year.)




From robin at jessikat.fsnet.co.uk  Thu Mar  8 13:17:23 2001
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Thu, 8 Mar 2001 18:17:23 +0000
Subject: gah! I hate the new string syntax
References: <mailman.983615170.23313.python-list@python.org>
 <tN5o6.2746$eH3.14913@NewsReader>
 <Koco6.7518$1D5.244602@e420r-atl1.usenetserver.com>
 <mailman.983997252.5628.python-list@python.org>
 <m2vgpkyfjo.fsf@mycroft.actrix.gen.nz>
 <mailman.984063199.18779.python-list@python.org>
Message-ID: <g44pZKAzy8p6EweT@jessikat.demon.co.uk>

In article <mailman.984063199.18779.python-list at python.org>, D-Man
<dsh8290 at rit.edu> writes
>On Thu, Mar 08, 2001 at 06:51:07PM +1300, Paul Foley wrote:
>| On Wed, 7 Mar 2001 15:33:30 -0500, D-Man  wrote:
>| 
>| > What about one who is both English and Scots, yet never set foot on
>| > that island ;-)?
>| 
>| AFAIK, nobody outside the US would admit the existence of such a
>| thing.  The American habit of calling themselves "Irish" or "Italian"
>| or whatever when they're not is *very* strange.  I've never heard
>| anything like it anywhere else.
>
>Heh.  Maybe Americans are very odd.  Wouldn't surprise me at all.  ;-)
>I was born in America and have lived in America all my life.  I do
>have English, Scots, Irish and Welsh ancestry though.  One could argue
>that I am not really American since none of my ancestors came from
>America.  "Real" Americans would be the so-called "Indians", or more
>appropriately "Native Americans".  (Also Aztecs, Inca, etc., and their
>descendants such as the Quiche)
>
>-D
wow you're lucky; when I was a lad there was Professor Hoyle who said we
probably didn't even evolve on Earth. Latest observations seem to
support his contention that the vast majority of biological like
molecules are in outer space. If true we're all immigrants :)
-- 
Robin Becker


From fluxent at yahoo.com  Tue Mar 20 11:53:15 2001
From: fluxent at yahoo.com (fluxent at yahoo.com)
Date: Tue, 20 Mar 2001 16:53:15 -0000
Subject: IIS forgetting PythonPath?
Message-ID: <9981tr+3n1j@eGroups.com>

I'm calling Python code from with ASP pages on Win2K. I have 
ActiveState PyWin installed. Everything's been working fine.

Then my machine had a power glitch this morning. Now, whenever I call 
pages that do an import, I get an error message about not finding the 
module being asked for. Pages calling some vanilla Python work fine. 
The registry still shows my subkey of PythonPath fine, and I can 
perform the import from within the PythonWin interactive window fine. 
And if I copy the module up to the Python root directory, then the 
ASP page is delivered.

So it appears to me that IIS is ignoring the PythonPath, even though 
it seems to be OK. Is there another place in the registry that has a 
copy of PythonPath that could be buggered?

If it matters, there's an old 1.5x version of Python also on this 
machine within c:\python-old\. I'm pretty sure it's not involved, 
that IIS has always been running the ActiveState v2, but not sure how 
to prove that...





From sholden at holdenweb.com  Mon Mar  5 20:57:15 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Mon, 5 Mar 2001 20:57:15 -0500
Subject: Little problem with the "+" operator
References: <983802310.155838463@news.t-online.de> <mailman.983804468.16428.python-list@python.org> <980i6q$crs$1@news.mathworks.com> <mt1yscm7yu.fsf@astron.berkeley.edu> <3aa3d3a7.420988249@news.uswest.net>
Message-ID: <MxXo6.24186$1D5.970545@e420r-atl1.usenetserver.com>

"Jay O'Connor" <joconnor at cybermesa.com> wrote in message
news:3aa3d3a7.420988249 at news.uswest.net...
> On 05 Mar 2001 09:38:49 -0800, Johann Hibschman
> <johann at physics.berkeley.edu> wrote:
>
> >Joshua Marshall writes:
> >
> >>> This is infinitely more sane than:
> >
> >>> % perl -e 'print 500 + "cool"'
> >>> 500
> >
> >> Comparing Python to Perl here doesn't seem very fair.  They're very
> >> different languages.  In general, I'm surprised by the urgency with
> >> which people often jump to compare them.
> >
> >Well, Perl's the only language that I know of that lets you add 500 to
> >"cool" without an error.  Well, I suppose tcl might.  But it's still a
> >fairly safe inference that the original poster was coming from Perl.
>
>
> Actually, JavaScript will allow you to add numbers to strings, with
> weird results
>
> '100' + 200 = '100200'
> 100 + '200' = 300
>
> I *think* that's how it works, I just know it's not very safe
>
Kind of "operator underloading"?

too-bad-a-typist-to-program-in-java-ly y'rs  -steve





From ws-news at gmx.at  Wed Mar  7 04:49:59 2001
From: ws-news at gmx.at (Werner Schiendl)
Date: Wed, 7 Mar 2001 10:49:59 +0100
Subject: Can anyone recomend a good intoduction to C...
References: <mailman.983894288.19965.python-list@python.org> <3AA5EF59.507F1D91@olsen.ch>
Message-ID: <983958878.976344@newsmaster-04.atnet.at>

> I don't think C deserves all the superlatives you apply to it, but I agree
that
> it's instructive to study C++ as an example of a language that gets it
wrong.
> Take a good language, C; add some good ideas; and end up with a
> baroque mess. Fascinating.

To be completely honest, I can not see the obvius reason why C++ should be
such a bad programming language.
It is widely available (on almost any platform, I would guess).

It supports for a real large number of different design approaches.
I do not know of any one more flexible language at all, combined with the
performance of the executables and the quality of syntax check while
compiling.

It maybe somewhat overwhelming when you try to master all the features in
one shot.
It is of course nonsense to use any existing feature in a little toy proggy.
Compare it to using regular expression to split up a comma separated list in
python, its the same I think.
But I love the features to be in the box so that I can take them out right
when the need arises.

regards
werner




From James_Althoff at i2.com  Thu Mar 22 13:21:02 2001
From: James_Althoff at i2.com (James_Althoff at i2.com)
Date: Thu, 22 Mar 2001 10:21:02 -0800
Subject: New PEP: The directive statement
Message-ID: <OF4DDCB0A8.53D4E538-ON88256A17.0064866D@i2.com>

ok, how about:

from  __readyOrNot__  import nested_scopes

or perhaps,

from  __likeItOrNot__  import nested_scopes

???

jim



                                                                                                                   
                    "Tim Peters"                                                                                   
                    <tim.one at home.com>        To:     <python-list at python.org>                                     
                    Sent by:                  cc:                                                                  
                    python-list-admin@        Subject:     RE: New PEP: The directive statement                    
                    python.org                                                                                     
                                                                                                                   
                                                                                                                   
                    03/21/01 05:20 PM                                                                              
                                                                                                                   
                                                                                                                   




[piet at cs.uu.nl]
> It would be much less of a joke when __future__ would be replaced by
> __features__, I think.

[Neil Schemenauer]
> __features__ is a terrible name.  Its pretty obvious that you
> (and a lot of other people on this list) don't understand what
> __future__ does.

When I first read the Pentium architecture manual, its description of how
the
Pentium handled floating-point exceptions was so off the wall I simply
couldn't believe they meant what they said, and went off writing code after
mentally substituting a *reasonable* (but related) meaning of my own.

Heh.  Turns out it meant exactly what it said!  Cost me plenty to recover
from that, too.

I'm not sure we're not suffering a similar confusion wrt __future__:  there
are no exact analogues in any other language I know of, so I wouldn't be
surprised if people think this *is* for "optional features", and that
"optional" is what they'll *read* no matter what the PEP says.  I may like
Martin's "transitional" better for this reason -- provided that it wakes
people up enough to think about what they're reading.  Alas,

    from __transitional__ import nested_scopes

is not only ugly, it's not even funny, so would be worse than "from
__future__ ...".  It works better with "directive" because it keeps it
uniformly bland.

if-you-can't-laugh-at-your-tools-you're-taking-them-too-seriously-ly
    y'rs  - tim


--
http://mail.python.org/mailman/listinfo/python-list







From lluang at northwestern.edu  Thu Mar  8 16:52:14 2001
From: lluang at northwestern.edu (Louis Luangkesorn)
Date: Thu, 08 Mar 2001 15:52:14 -0600
Subject: cursor.execute fails under mxODBC
References: <mailman.983993842.26448.python-list@python.org> <tUzp6.29925$ML1.1507874@e420r-atl2.usenetserver.com>
Message-ID: <3AA7FF0E.50E6D70F@northwestern.edu>

I have the same problem.  Below is my output.  I will first use the odbc
that comes included with ActivePython, then the mxODBC module  (by the way,
I've tried to rename the odbc.pyc file, that did not help, except break
odbc)

PythonWin 2.0 (#8, Oct 19 2000, 11:30:05) [MSC 32 bit (Intel)] on win32.
Portions Copyright 1994-2000 Mark Hammond (MarkH at ActiveState.com) - see
'Help/About PythonWin' for further copyright information.
>>>
>>> import dbi,odbc
>>> import pprint
>>> channel = odbc.odbc('sqltutor')
>>> ch1 = channel.cursor()
>>> ch1.execute('SELECT FirstName, LastName, Address, City, State FROM
EmployeeAddressTable')
0
>>> test1 = ch1.fetchall()
>>> pprint.pprint(test1)
[('Joe', 'Smith', '83 First Street', 'Howard', 'Ohio'),
 ('Mary', 'Scott ', '842 Vine Ave.', 'Losantiville', 'Ohio '),
 ('Sam', 'Jones ', '33 Elm St.', 'Paris', 'New York '),
 ('Sarah', 'Ackerman ', '440 U.S. 110', 'Upton', 'Michigan ')]
>>> ch1.close()
>>> channel.close()
>>> import mx.ODBC
>>> channel = mx.ODBC.Windows.connect('sqltutor')
>>> ch1 = channel.cursor()
>>> ch1.execute('SELECT FirstName, LastName, Address, City, State FROM
EmployeeAddressTable')

at this point the whole thing dies.  No error messages.  I didn't see any
log files anywhere either.

Louis

--
K Louis Luangkesorn
lluang at northwestern.edu  http://pubweb.nwu.edu/~kll560  PGP:0xF3E2D362
Whatsoever things are true, ... honest, ... just, ... pure, ... lovely, ...
of good report; if there be any virtue, and if there be any praise, think on
these things.- motto - Northwestern University




From nde at comp.leeds.ac.uk  Thu Mar 22 07:39:44 2001
From: nde at comp.leeds.ac.uk (Nick Efford)
Date: Thu, 22 Mar 2001 12:39:44 GMT
Subject: Please rate these Python books
References: <20010322.1445209@mis.configured.host> <dAeu6.144721$__6.26735546@typhoon.southeast.rr.com>
Message-ID: <GALn69.CA1@leeds.ac.uk>

On Thu, 22 Mar 2001 03:46:17 GMT, Don Tuttle <tuttledon at hotmail.com> wrote:

> All 3 are excellent books.  You won't go wrong with any of them.  (Althought
> you might want to wait til June for Beazley's 2nd edition.) For more info

I find Beazley's book indispensible once you've learned the language.
Good to hear that a 2nd edition is imminent...


N.



From cfelling at iae.nl  Fri Mar 16 18:38:53 2001
From: cfelling at iae.nl (Carel Fellinger)
Date: 17 Mar 2001 00:38:53 +0100
Subject: I come to praise .join, not to bury it...
References: <mailman.983485923.1136.python-list@python.org> <97p7iq$kk2$1@nntp6.u.washington.edu> <97rka002k9@news1.newsguy.com> <3AAF7E47.294617E1@sage.att.com> <98qeco02hn7@news1.newsguy.com> <98rmod$c8p$1@animus.fel.iae.nl> <98tacr0q5s@news1.newsguy.com>
Message-ID: <98u86d$6p9$1@animus.fel.iae.nl>

Alex Martelli <aleaxit at yahoo.com> wrote:
> "Carel Fellinger" <cfelling at iae.nl> wrote in message
> news:98rmod$c8p$1 at animus.fel.iae.nl...
...
>> but if you get too blinded by this substance thing, you might easily
>> overlook that in the end you have to weight things and there again the
>> seasoned sense of aesthetics of the gifted ones is what makes the
>> difference.

> I appreciate "form follows function" as an aesthetic principle.

Here we agree, and clearly Rococo, though having a beauty of its own,
has nothing to do with the aesthetic principles that guided the
Bauhaus people, so aesthetics in it self as the sole guiding light has
it pitfals, to say the least:)  But we have discussed this before,
aesthettics as a design guide only works for the `Enlightend Ones'.

...

> You seem to imply some moderation (have to weight things) is
> key to YOUR aesthetics -- fine, Aristotle will agree, as will

I don't know why you think that moderation is key to my aesthetics,
surely simplicity is, but moderation?  People tent to descripe me
rather as an extremist.

The weighting you seem to refer to is the weighting you yourself
mentioned earlier on, i.e. that Guido has to weight the relative
importance of several sometimes conflicting substantial goodies or
baddies that come from e.g. introducing an extension to the print
command.

> In my own profession, I make certain choices which I perceive
> as guided by 'aesthetics' when I cannot afford to articulate
> in full detail the path back from the practical example up to
> the guiding-principle; my experience (and 'gift', if any) may
> help this synthesis (a productivity plus) by helping me
> perceive as "elegant" an architecture which will prove helpful
> to the design-task at hand.  If and when challenged, though, I
> can and will fall back to the time-consuming, perhaps boring,
> but often enlightening task of detailed analysis of the
> constituent elements -- there is nothing 'inherently ineffable'
> in this.  I see architects of manufacts with more psychological

And here we part. In my experience, and looking to history not mine
alone, it isn't allways possible to come to a satisfactory and full
analysis.  Not to say that we shouldn't strive to, and strive again
later if we have a better understanding of the things at hand!  But
even *if* you'll have a substantial analysis at hand, in the end it's
a personal judgement whether the analysis is satisfactory and complete
enough.  Very much like in logic class, when I was confronted with
some proof forms I didn't grog, but others had no problem with.  I've
to admit that in those days I was pleased to learn that I was not the
only one objecting to those proofs (IIRC it was called reduction ad
absurbum, not sure, my brain isn't functioning like it used to before
my illness), it turned out there was a whole School of Logic (The
Amsterdam School) based on rejecting them.

...

> Somebody who's unable to do anything but talk of aesthetics
> when discussing a choice in the architecture of a software
> component is simply not contributing to the discussion.  It

Here we agree again:) If one has nothing more to say then
express one's own aesthetics feelings, the discussion ends.
Though, in practice, for me the real quest starts:

  *Why* do/don't I like this.

So let me end telling you that I appreciate most of your
contributions to this newsgroup very much, especially
because they make me more aware of the beauty of Python
and above all becuase they make me better understand the
nature of that beauty.
-- 
groetjes, carel


From danielk at aracnet.com  Sun Mar  4 12:44:39 2001
From: danielk at aracnet.com (Daniel Klein)
Date: Sun, 04 Mar 2001 09:44:39 -0800
Subject: Subclassing builtin types
Message-ID: <i6u4atslg4euv732pvaujk6u036kemcidf@4ax.com>

One of the classes I have to write  is merely a wrapper around a 'list' object.
A very abreviated example,

class MyParams:
      def __init__(self):
            self.parameters = []
      def append(self, object):
            self.parameters.append(object)
      def insert(self, i, object):
            self.parameters.insert(i, object)
      def del(self, i):
            del self.parameters[i]
etc, etc...

This would allow me to do things like

p = MyParams()
p.append('foo')

Since I know you are going to ask :-), there are two main reasons I need to do
this:

1) I need to ensure that a certain 'capacity' is not exceeded. I will probably
have class variable which is checked in all methods which add objects to
'parameters' and then raise an exception if 'capacity' is exceeded.

2) I can only allow certain types of objects to be added to the list, so these
too will be checked where necessary and appropriate exceptions raised.

What I would _REALLY_ like to do is subclass list something like

	class MyParams(List):

so that I can inherit all of the list methods and only override the ones I need
to. In Smalltalk this would be accomplished simply with

	OrderedCollection subclass: #MyParams

I had actually thought of creating a generic List class that could be reused in
cases like this.

So two questions:

1) Are wrapper classes for the builtin types a feasible (and pythonic) thing to
do?

2) Are there another ways to accomplish what I'm trying to do here without
creating my own class hierarchy?

Thanks for any and all help, you people have been a wonderful lot to deal with
during this new-language learning period.

Daniel Klein
Portland OR USA


From hernan.foffani at iname.com  Sun Mar 18 08:41:34 2001
From: hernan.foffani at iname.com (Hernan M. Foffani)
Date: 18 Mar 2001 13:41:34 GMT
Subject: Alternative documentation formats?? (chm and info)
References: <m3elvw2inm.fsf@master.athome>
Message-ID: <20010318084134.922$G8@newsreader.com>

Andrew Markebo <flognat at flognat.myip.org> wrote:
> Hi!
>
> Uhm anyone out there thrown the Python docs through some magic filters
> and got them chm and/or info-formatted??
>

You can get it at:
http://alldunn.com/python/
or
http://www.orgmf.com.ar/condor/pytstuff.html

Regards,
-Hernan M Foffani
www.orgmf.com.ar/condor/

-- 
-------------------- http://NewsReader.Com/ --------------------
                       Usenet for the Web


From ryan at ryanflynn.com  Sun Mar 18 17:33:15 2001
From: ryan at ryanflynn.com (Ryan)
Date: Sun, 18 Mar 2001 17:33:15 -0500
Subject: Get Python2 Win32 working with Apache
Message-ID: <rKat6.3985$GD.2280188@typhoon.snet.net>

Any hints?

--
Ryan Flynn
Practical web development/design
www.ryanflynn.com






From guido at digicool.com  Mon Mar 19 11:32:33 2001
From: guido at digicool.com (Guido van Rossum)
Date: Mon, 19 Mar 2001 11:32:33 -0500
Subject: Python T-shirts
Message-ID: <200103191632.LAA26632@cj20424-a.reston1.va.home.com>

At the conference we handed out T-shirts with the slogan on the back
"Python: programming the way Guido indented it".  We've been asked if
there are any left.  Well, we gave them all away, but we're ordering
more.  You can get them for $10 + S+H.  Write to Melissa Light
<melissa at digicool.com>.  Be nice to her!

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




From jimhill at meldrick.swcp.com  Sun Mar 11 23:57:34 2001
From: jimhill at meldrick.swcp.com (Jim Hill)
Date: 12 Mar 2001 04:57:34 GMT
Subject: Python In A Nutshell
References: <3AAE7E0D@MailAndNews.com>
Message-ID: <slrn9aolph.ccf.jimhill@meldrick.swcp.com>

(Piggybacking)

>>===== Original Message From Hamish Lawson <hamish_lawson at yahoo.co.uk> =====
>>
>>For the third-party modules I'd nominate the following, grouped by
>>desirability of inclusion:
>>
>>    ESSENTIAL
>>
>>    NICE
>>       Numerical Python

I'd put this one in the ESSENTIAL column.


Jim
-- 
jimhill at swcp.com                         http://www.swcp.com/~jimhill/

   "Let me preface my remarks by saying that I have absolutely no love
            for Barry Manilow in my heart."  --  Jason Lempka


From aleaxit at yahoo.com  Thu Mar  1 12:20:04 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 1 Mar 2001 18:20:04 +0100
Subject: Collection interfaces
References: <3a90fa97.0@news1.mweb.co.za> <3A933830.7F98D118@ix.netcom.com> <CgJk6.515$v77.66568@newsread2.prod.itd.earthlink.net> <220220010813598042%cmh@bDistributed.com> <ud7cakvi8.fsf@roundpoint.com> <MPG.150312ceb68f1ada98969f@news.earthlink.net> <8Gem6.473353$U46.14105062@news1.sttls1.wa.home.com> <MPG.1503548620e632919896a4@news.earthlink.net> <MCjm6.474652$U46.14163288@news1.sttls1.wa.home.com> <MPG.150394195f7f04a19896a7@news.earthlink.net> <3A9A86B1.9A8C69A2@ix.netcom.com> <MPG.1504d14563dce6499896af@news.earthlink.net> <97fjla128ir@news2.newsguy.com> <MPG.1505a76831c34b749896b3@news.earthlink.net> <97h5s40vhc@news2.newsguy.com> <MPG.1506007591cf818b9896bc@news.earthlink.net> <slrn99opop.gr.amk@mira.erols.com> <MPG.1506265ebf6d934d9896be@news.earthlink.net> <97ibmc0k47@news1.newsguy.com> <MPG.1506cf3293b094b49896c1@news.earthlink.net>
Message-ID: <97m0c7080c@news2.newsguy.com>

"Topmind" <topmind at technologist.com> wrote in message
news:MPG.1506cf3293b094b49896c1 at news.earthlink.net...
>
> > Before you knock
> > it, TRY it -- pick a pilot project of suitable size, a team
> > ditto, and have a go at it with eXtreme Programming
>
> Sure, hack-it-til-it-works indeed does work after enough

XP is _not_ "hack it til it works".  Do you know _anything_
about it, and are just being deliberately dense, or do you
enjoy speaking of what you know not...?

> time. But whether it produces a better system in the
> long run is another issue. Without better metrics, it is
> just Yet Another Methodology to add to the heap.

I've often noticed that demanding metrics is the last
line of defense of the conservatives in our field against
innovation -- since nobody's really funded to run serious
(and hugely costly) controlled experiments, such demanders
can rest reasonably tranquil that they won't be faced with
such metrics.  Unfortunately, the world has this habit of
passing them by anyway.


> How about you try heavy table-oriented programming for a
> few years.

Been there, done that, more or less whatever you mean by
"table-oriented programming" -- after all, I _have_ been
programming for well over a quarter of a century, in a
reasonably vast variety of paradigms.  From the jump table
of well-constructed, speedy machine-code, to the SQL Table
of relational-DB programming, I've done more tables than
the average Parisian cafe has on the sidewalks on a bright
and warm spring day...


> > code test-first, *refactor mercilessly*,
>
> I would rather use task-orientation and avoid
> excessive refactoring to begin with.

...then your resulting sources will never be as good,
clear, clean, maintainable, and high-quality, as those
of a better programmer, who DOES refactor every time
the occasion affords.

> "Refactoring" is a euphemism for "OO is not
> change-friendly after all".

Refactoring is not necessary _only_ in OO, of course.

And, depending on the axis of change, any paradigm
can result in code that is hard to change to meet
the new/changed needs at equally high quality -- the
dimensionality of the space of requirements is so
high, that, unless you do anticipate the principal
axes of requirement-changes roughly correctly, you
do have work ahead of you.  The best a paradigm and/or
methodology can do for you is let you have a few more
dimensions -- a few more axes of possible change --
among those you can meet more easily.  _A few_ are
the operative words here.

You still need design ability, flair, and a decent
amount of domain-specific experience somewhere in
the team, to enjoy a substantial chance of delivering
LARGE software-systems that present all the desired
quality attributes, including of course the ability
to change to meet the changing challenges of the
application domain.  Increasing that chance is all
one can ask of any language/paradigm/methodology --
there STILL is no silver bullet.


> > This seems to show what blind alleys classical approaches can
> > lead one to -- ```Reorganization can be risky. Many managers do
> > not want programs altered except to add or fix features. If an
> > significant error occurs due to reorganization, then heads roll.
> > I have been told many times, "If it ain't broke, please don't
> > fix it."''' being the key quote from that paper.
>
> Well, they are the boss. Sell rework to them, not me.

Fortunately, given the current market for IT professionals (and
this comment applies to the last 25 years), no competent worker
in the field ever needs to be stuck for long under management
that is totally impervious to common sense and reason -- even
assuming that firm is not about to go bankrupt (which seems a
possibility, if they can't listen to reason:-), there are sure
to be dozens of lucrative and professionally interesting job
offers on one's desk.  "Think of it as evolution in action".


> > (But the ills are many indeed; the
> > "lava flow" anti-pattern, for example, is likely to be quite
> > pervasive in a development shops to which this article could
> > apply).
>
> What is this "lava flow" thing? I keep hearing it mentioned.
> More OO lore that nobody can keep up with. Scriptures
> build fast when the agnostics are silenced.

Brown, Malveau, McCormick and Mowbray, "AntiPatterns",
Wiley Computer Publishing, 1998, ISBN 0-471-19713-0.  As
it's been out three years, it's hard to accept your
classifying it as a recent work.  Note also that, while
published as part of the "Object Technology" series by
Wiley (which probably makes you allergic to it ipso facto:-),
most of it is perfectly applicable to software developed
under other paradigms (indeed, quite a few antipatterns
exhibit typical problems of _legacy_ software, which a
consultant is called in to solve).


> I have shown in great detail why procedural seems better able
> to deal with non-boundaried differences, and I have yet to

You have done no such thing (at least in your posts in
this thread, and the URL you mentioned here).

> see counter-arguments other than "you get used to code
> rework if you just give it a fancy name."
> I won't sell out to wasteful effort.

Effort that is wasteful, by definition, is well eschewed.
But changing code *and design* so that changing requirements
are well-reflected and optimally met, again by definition
of 'well' and 'optimally', is no waste:-).


> But the easy-to-componentize stuff is already for sale. Why

Some of it -- that which (in the conception of the backers,
venture capitalists, entrepreneurs, etc) enjoys a large
enough and profitable enough potential market.  Nowhere
like _all_ (nor even a majority) of the actual markets
have been tapped, much less saturated.

> would they bring in programmers if plug-and-play solutions
> were already available? (Report writers, sorters, DB's,
> GUI IDE's, communications, etc.)

There are no doubt databases for sale, but I doubt
that's what you're referring to, because it's not
software that's being sold, then, but data -- you
probably mean, database _management systems_ (DBMS's).

If some solution IS for sale in a given area, and
some capitalists and entrepreneurs are still desirous
to develop and sell programs in that area, they must
think that a better mousetrap can be developed to meet
those needs -- and they may well be right (my current
employer, for example, is busy experimentally testing
this thesis regarding 3D mechanical engineering design,
and ancillary applications such as product-data
management [PDM]).

It's not particularly different from the business
decision of opening a new restaurant in a town that
already has quite a few -- you'll win (make money)
if the market is not in fact saturated, and/or your
new restaurant can offer diners something they value
enough to pay you enough money for it.

It does not imply that your new restaurant cannot
possibly make profitable use of some existing recipes,
you know -- coming up with new recipes is just one
possible competitive differentiator, and there is a
lot of potential space in terms of execution and
delivery even if your fundamental recipes happen to
be very close to the competition's.


> What is left is stuff that has a jillion interconnections
> to other things such that putting a clean box around it
> is not practical. Components usually work for things where
> the algorithm is complex but the interfaces are small.
> In biz stuff the interfaces/links to many entities is the
> largest part.

...and interconnection-and-gluing components work wonderfully
well for _that_ kind of issues (scripting language interpreters
being a good example:-).

> Besides, you can characterize my development philosophy
> as "task componentization". AKA, a procedural module.

A philosophy is a wonderful thing to have -- it can no
doubt console you splendidly when you're rotting in jail,
for example (cfr. Boethius' "De consolatione philosophiae").
Or, in your case, apparently provide a high-faluting way
to justify your unshakable desire to keep working as you
always have (and, I guess, your grand-dad before you...?).
But, given the marketplace success that's likely to greet
your efforts, I'm sure the "consolation" part will come
in very handy for you, too, pretty soon:-).


> > Adopting the 'software that happens to appear to work when
> > not thoroughly tested is sacred and must not be touched' idea
> > is parallel to keeping a kid always on his/her side of the
> > street (except perhaps for emergencies such as schools and
> > doctors:-).  Not exactly conducive to optimal development
> > (be it of social skill and a healthy life, OR good software).
>
> I am not against re-arrangement. What I am complaining about
> is paradigms that *increase* the amount of re-arrangement that
> is required. OO appears to do such, partly because of the
> boundary/granularity issues.

Maybe "appears" is the operative word -- it LOOKS (to you) AS IF
object-oriented analysis/design/coding increases your 'rework'
efforts -- maybe because you lack the experience to do it well.

> The whole refactoring thing is side-effect of OO. The paradigm

No way -- just the _term_, I believe, tends to be used
originally in OO circles; and the reusable _patterns_ of
refactoring (just as those of other aspects of our jobs)
have mostly been identified and described inside an OO
worldview (OO admittedly does make it easier to handle
and describe design & other patterns -- a neat plus for it:-).

But -- without calling it that way, without the productivity
that reusable patterns or good tools can give me -- I _was_
doing refactoring 25 years ago (as were many other before
then); I've always loved the design and code I was turning
out, and strived for them to be clean, clear, high-quality,
and well attuned to the actual needs.  That's what you need
to refactor for.

> is starting to barf, and now the same people who sold you
> the barfers are now also selling barf bags.

Funny enough, I mostly use free tools -- books are, by far,
what I invest most money (and time) in, and they cover a
very wide range (I'm just as likely to pick up a new book
on functional programming, or more, than a new one on OO,
no matter that I don't currently use FP in my everyday
job but I do use OO -- exactly because I know less about
it, I desire books more, for example).


Alex





From echuck at mindspring.com  Thu Mar  1 08:39:07 2001
From: echuck at mindspring.com (Chuck Esterbrook)
Date: Thu, 01 Mar 2001 08:39:07 -0500
Subject: [ANNOUNCE] Webware for Python 0.5
Message-ID: <5.0.2.1.0.20010301083728.051a36f0@mail.mindspring.com>

Webware for Python is a suite of software components for developing 
object-oriented, web-based applications. The suite uses well known design 
patterns and includes popular features such as:

   * A fast, multi-threaded, easy-to-use application server
   * Servlets
   * Python Server Pages (PSP, similar to ASP, PHP and JSP)
   * An object-relational mapper
   * Task scheduling
   * User management and authentication
   * A CGI wrapper

Home:     http://webware.sourceforge.net/
Download: http://download.sourceforge.net/webware/Webware-0.5.tar.gz
Docs:     http://webware.sourceforge.net/Webware/Docs
Discuss:  mailto:webware-discuss at lists.sourceforge.net


Highlights of 0.5:

* New kits:
   * UserKit:   User management and authentication
   * TaskKit:   Scheduling
   * MiddleKit: Object-relational mapping
   * COMKit:    Enable multi-threaded COM support in the app server
* WebKit enhancements
   http://webware.sourceforge.net/Webware/WebKit/Docs/RelNotes-0.5.html
* MiscUtils enhancements
   http://webware.sourceforge.net/Webware/MiscUtils/Docs/RelNotes-0.5.html
* PSP enhancements
   http://webware.sourceforge.net/Webware/PSP/Docs/RelNotes-0.4.html


http://webware.sourceforge.net/




From nas at python.ca  Fri Mar 30 17:25:33 2001
From: nas at python.ca (Neil Schemenauer)
Date: Fri, 30 Mar 2001 14:25:33 -0800
Subject: list vs tuple
In-Reply-To: <71q9ct4le63anefmiajj5cbh3mg04qu1ga@4ax.com>; from Mike.Prager@noaa.gov on Fri, Mar 30, 2001 at 03:25:14PM -0500
References: <hGlw6.782$p5.3210@news1.rivrw1.nsw.optushome.com.au> <3AC20CD3.265A2D94@alcyone.com> <ru69cto61jtmrqsf809e38hb3pbm2lhi7g@4ax.com> <bC1x6.4319$4N4.905946@newsc.telia.net> <71q9ct4le63anefmiajj5cbh3mg04qu1ga@4ax.com>
Message-ID: <20010330142533.A2204@glacier.fnational.com>

Michael Prager wrote:
> That behavior is not in agreement with the "mental model" used
> by many other languages, so I would think it a major pitfall for
> beginners -- at least, it was for this beginner.

Have you used other languages before coming to Python?  I would
think it would be confusing to people who were used to languages
that have a reference/value inconsistency.  Python is totally
consistent (everything is a object, all names refer to objects).

  Neil



From tim.one at home.com  Sat Mar 24 02:23:16 2001
From: tim.one at home.com (Tim Peters)
Date: Sat, 24 Mar 2001 02:23:16 -0500
Subject: New PEP: The directive statement
In-Reply-To: <99fnif$gsq$1@news.mathworks.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCKEAFJIAA.tim.one@home.com>

[Joshua Marshall]
> Please, no more keywords.  It's unnecessary in this case.
> How about an interpreter switch:
>
>   python -directive nested_scopes fu.py

Impractical.  For a hint of *how* impractical, note that a few std library
modules in the 2.1 release require nested_scopes:  you would have to specify
the names of those modules on the cmdline every time you ran Python for
anything.  Now picture your own .py files on top of that, and twenty
libraries you get from other people.  Pure nightmare.

PEP 236 handles this stuff without a new keyword.  But it's limited to future
statements, and Python probably does need a general directive facility too.

new-keywords-are-better-than-syntax-contortions-over-the-long-run-ly
    y'rs  - tim




From cabrera at wrc.xerox.com  Thu Mar 22 16:42:03 2001
From: cabrera at wrc.xerox.com (jicman)
Date: 22 Mar 2001 21:42:03 GMT
Subject: ASCII encoding error: ordinal not in range(128)
References: <91o5ih$2kv$1@news.wrc.xerox.com> <20001219172933.B29727@ActiveState.com> <006201c0b184$a60b9430$5aa4af0d@dssc.mc.xerox.com> <mailman.985126108.22009.python-list@python.org>
Message-ID: <99drjb$eul$2@news.wrc.xerox.com>

paulp at ActiveState.com says...

>I'm going to presume that "code" is a Unicode string. In that case, you
>have to say HOW you want it encoded. What do you want to do with the
>non-ASCII characters? Probably what you want is:
>
>"abc".encode("Latin-1")
>
>That means that characters between 128 and 255 will be encoded as the
>equivalent byte-codes. Characters above 255 will simply not be
>representable. I'm presuming that you won't run into any of those...if
>you will be, then you need to think about representation some more.

thank you very much.

jos?



From syver at NOSPAMcyberwatcher.com  Tue Mar 27 16:16:44 2001
From: syver at NOSPAMcyberwatcher.com (Syver Enstad)
Date: Tue, 27 Mar 2001 23:16:44 +0200
Subject: Python scripts as Drop targets in Windows Explorer
References: <99qogo$o6g$1@troll.powertech.no> <99qqq3$qt7$0@216.39.170.247>
Message-ID: <99r043$qqr$1@troll.powertech.no>

Oh, sorry it was a unicode file,  I'll post an ascii version of it instead.


"Dave LeBlanc" <whisper at oz.net> wrote in message
news:99qqq3$qt7$0 at 216.39.170.247...
> How about using the export option in regedit for a human readable
> version of this please?

That's what I did, but the default on Win2k is to save as unicode. I guess
you're using NT or Win9x?


begin 666 python filedrop.reg
M4D5'141)5#0-"@T*6TA+15E?0TQ!4U-%4U]23T]47%!Y=&AO;BY&:6QE7'-H
M96QL97A<1')O<$AA;F1L97)=#0I /2)[.#9#.#8W,C M-#)!,"TQ,#8Y+4$R
513 at M,#@P,#)",S S,#E$?2(-"@T*
`
end



From mjbjr at best.com  Thu Mar  1 21:39:42 2001
From: mjbjr at best.com (Martin Brown)
Date: 2 Mar 2001 02:39:42 GMT
Subject: python irc channel?
Message-ID: <97n15e$9g3$1@nntp1.ba.best.com>

So, is there a preferred irc channel where python people hangout?

Thank you.
--
                            - Martin J. Brown, Jr. -                           
                            - mjbjr at beaudesign.com -                           
                                                                               
    PGP Public Key ID: 0xCED9BD8A  Key Server: http://www.keyserver.net/en/


From Eugene.Leitl at lrz.uni-muenchen.de  Tue Mar 27 05:07:52 2001
From: Eugene.Leitl at lrz.uni-muenchen.de (Eugene Leitl)
Date: Tue, 27 Mar 2001 12:07:52 +0200 (MET DST)
Subject: running code from "Scripting The Web with Python"
Message-ID: <Pine.GSO.4.03.10103271157060.10313-100000@sun1.lrz-muenchen.de>

I've snipped the following code from Guido's
http://www.w3j.com/6/s3.vanrossum.html presentation:

import StringIO import SimpleHTTPServer

BaseClass = SimpleHTTPServer.SimpleHTTPRequestHandler

CounterTemplate = """ <H1>Server
Statistics</H1>

This <A HREF=/index.html>server</A> has been accessed
<b>%d</b> times.  """

count = 0

class MyRequestHandler(BaseClass):

    def do_GET(self):
        global count count = count + 1 BaseClass.do_GET(self)

    def send_head(self):
        if self.path == "/counter.html":
            return self.send_counter() else:
            return BaseClass.send_head(self)

    def send_counter(self):
        self.send_response(200)
        self.send_header("Content-type",
        "text/html") self.end_headers() text =
        CounterTemplate % count return StringIO.StringIO(text)

def test():
    SimpleHTTPServer.test(MyRequestHandler)

test() 

I can't run it due to syntax errors (Python 2.0 and 1.5). I've tried
hacking it a bit, but not knowing Python syntax I've been unable to make
it run. Any ideas?




From db3l at fitlinxx.com  Tue Mar 20 14:49:14 2001
From: db3l at fitlinxx.com (David Bolen)
Date: 20 Mar 2001 14:49:14 -0500
Subject: Little red lines in PythonWin!
References: <mailman.985107113.4466.python-list@python.org>
Message-ID: <uu24ojk5h.fsf@ctwd0143.fitlinxx.com>

"Bill Bell" <bill-bell at bill-bell.hamilton.on.ca> writes:

> What, in general, do you guys do about distributing sourcecodes in 
> a language like Python that depends on indenting and where tabs 
> can cause chaos. (And, no, you can't possibly shoot all of those of 
> us who get caught this way, so that's not a solution.)
> 
> Please point me towards the accepted approach. Thanks.

The safest and most robust approach is probably just to untabify
anything that is going to be released to the public, to minimize the
potential for confusion.  How to do that differs by tool (most editors
support the idea, e.g., M-x untabify in Emacs, and I'm sure there are
standalone utilities around via a quick web search).

In theory you really shouldn't get into trouble using tabs as long as
you follow Python's parser definition of what a tab is (treat as
moving to next multiple of 8 - Language Reference 2.1.7).  The trouble
is mostly when an editor is using an explicit tab character but
internally adjusting its dimension to something other than that rule.
Even that might work as long as only tabs are used for indentation,
but if spaces get mixed in it's bound to cause problems.  So in
general that's nasty, but as long as you untabify in that same
editor/IDE before distribution, since it knows its own rules, the
untabified version should be fine.

Of course, not using tabs (or definitely not mixing tabs and spaces)
in the source is probably best, given prior experience with confusion
on how to handle them compatibly cross-toolsets.  Most editors and
IDEs will let you configure such that pressing the TAB key indents you
appropriately, but is really using spaces to do so.

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


From fredp at mygale.org.nospam  Thu Mar 22 03:18:05 2001
From: fredp at mygale.org.nospam (Fred Pacquier)
Date: 22 Mar 2001 08:18:05 GMT
Subject: Accessing Oracle from Python remotely ?
References: <Xns906AB9C901807PaCmAnRDLM@161.48.76.20> <eE3u6.776$9%.203510@newsr1.maine.rr.com> <99bj2o$u91$1@nntp9.atl.mindspring.net>
Message-ID: <Xns906C5EC922152PaCmAnRDLM@161.48.76.20>

"Andreas Jung" <andreas at andreas-jung.com> said :

> There are currently two different DCOracle versions available. DCOracle
>1.3.x
>supports is mostly build on top of Oracle 7 but I works although with
>Oracle 8i
>except features like BLOBs and CLOBs. When you need complete support
>for Oracle 8i take a look at Matt Kromers new version of DCOracle 2. It
>is still alpha so be warned :-) You can find DCO2 on www.zope.org  -
>search for DCO2.

Thanks Andrea - the target database is O7 so probably the "old" one will 
do...

-- 
YAFAP : http://www.multimania.com/fredp/


From martin.skott at propylon.com  Tue Mar  6 11:58:49 2001
From: martin.skott at propylon.com (Martin Skott)
Date: 06 Mar 2001 16:58:49 +0000
Subject: Processing XML
References: <mailman.983853727.31962.python-list@python.org>
Message-ID: <hf166dh2.fsf@propylon.com>

Timothy Grant <tjg at exceptionalminds.com> writes:

> Hi all,
> 
> For all of you at P9, I hope you are having a grand time, I
> wish I could be there!

Yeah, lucky bastards. Ireland is cold this time of year...
> I have Sean McGrath's book on XML Processing with Python, and
> have played a bit with Pyxie. At the moment I can accomplish
> more with Pyxie than I can with PyXML on its own. However, for
> some reason I'm a bit uncomfortable with it for some reason.

I felt the same way, but after having worked with Sean (and Pyxie)
since september some of it has gone away. Pyxie is in many ways a
state of mind.
> Part of my discomfort comes from the fact that part of my
> interest in XML is that it is so broadly useful. I am curious
> as to whether I will come to a point where I find Pyxie to be
> limiting the usefulness of my use of XML.

Pyxie does have some limits (and some ugly bugs have shown up lately),
but it takes some special documents/situations to reach them.
> If Pyxie is a good choice, how do I add nodes to a document
> stored as an xTree?

Use the String2xTree function to create a mini tree of your sting and
then use the paste functions to insert it into your main tree.

-- 
Martin Skott
Software Developer
Propylon - Enabling Universal Mobility
http://www.propylon.com
Tel: +353 096 37245
Mobile: +353 087 9680370


From SBrunning at trisystems.co.uk  Thu Mar 29 08:22:37 2001
From: SBrunning at trisystems.co.uk (Simon Brunning)
Date: Thu, 29 Mar 2001 14:22:37 +0100
Subject: Defining class files
Message-ID: <31575A892FF6D1118F5800600846864D78BA4E@intrepid>

> From:	Neil Benn [SMTP:neil.benn at cambridgeantibody.com]
> I'm wokring through the tutorial and have reached the section on classes
> in
> Python.  Working through the example on classes I have entered the class
> using the interpreter:-
> 
> class MyClass:
>     "A simple example class"
>     i = 12345
>     def f(x):
>         return 'hello world'
> 
>     This worked fine when I typed:-
> 
> >>> x=MyClass()
> >>> x.f()
> 'Hello World'
> 
>     However, I then tried to write the class as an external text file,
> using:-
> 
> class YourClass:
>     "A simple example class"
>     i = 12345
>     def f(x):
>         return 'hello world'
> 
>     I then imported the class, tried to assign the class and ivoke a
> method
> :-
> 
> >>> y = YourClass()
> Traceback (innermost last):
>   File "<console>", line 1,
> TypeError: call of non-func
> 
>     This seemed a bit strange, so I investigated the Your class and it
> seemed to be a module.
> 
> >>> YourClass
> <module YourClass at 2584541>
> 
>     The text file is saved as YourClass.py - is this the problem, should
> class files have different terminaters in their filename??
 
Assuming that you imported the module like this:

import YourClass

you need to qualify the class name, as so:

y = YourClass.YourClass()

the unqualified label YourClass refers to the imported module, not the class
within it.

HTH

Cheers,
Simon Brunning
TriSystems Ltd.
sbrunning at trisystems.co.uk




-----------------------------------------------------------------------
The information in this email is confidential and may be legally privileged.
It is intended solely for the addressee. Access to this email by anyone else
is unauthorised. If you are not the intended recipient, any disclosure,
copying, distribution, or any action taken or omitted to be taken in
reliance on it, is prohibited and may be unlawful. TriSystems Ltd. cannot
accept liability for statements made which are clearly the senders own.



From wcanopy at covad.net  Wed Mar 14 19:15:49 2001
From: wcanopy at covad.net (Wiwih Gunadi)
Date: Wed, 14 Mar 2001 18:15:49 -0600
Subject: wxPython UI refresh
References: <98jku0$ifi$1@newstest.laserlink.net> <slrn9avc13.p8n.zeitlin@seth.lpthe.jussieu.fr>
Message-ID: <98p123$ls$1@newstest.laserlink.net>

"Vadim Zeitlin" <zeitlin at seth.lpthe.jussieu.fr> wrote in message
news:slrn9avc13.p8n.zeitlin at seth.lpthe.jussieu.fr...

>  Try calling Win32 ::UpdateWindow() directly on the embedded window
(assuming
> you can find its HWND somehow) and/or sending it WM_PAINT explicitly
(again,
> you need the HWND).

How do I call UpdateWindow in wxPython?? isn't it encapsulated in the
wxWindow.Refresh() method?  I tried it, it didn't work.



>
> >BTW, Excel is not an ActiveX control is it? i.e. it's impossible to
"wrap"
> >Excel like IE with the activexwrapper, right?
>
>  Wrong, you can embed it, of course. Try putting an Excel table inside a
Word
> document and editing it.

Yes, but how to do it in wxPython??
I tried to load the Excel object library, there isn't the equivalent of
WebControl object of IE in Excel.

When I tried to MakeActiveXClass() the Excel.Application, it complains
because it can't find the UI part which leads me to believe that Excel can't
be treated as an ActiveX control.

-Will--



>
>  Regards,
> VZ
>
> --
> GCS/GM d? H+ s++:-- p2 au--- a- w+ v C+++ UBLS+++ P- L++ N++ E--- W++++ M?
V--
> -po+ R++ G`` !tv b+++ D--- e++++ u++ h--- f+ r++ n- y?




From francois_forest at yahoo.fr  Fri Mar 16 06:19:23 2001
From: francois_forest at yahoo.fr (Francois Forest)
Date: Fri, 16 Mar 2001 11:19:23 GMT
Subject: How do I get Win32 to locate DLLs elsewhere (not really a Python 
 thing)?
References: <mailman.984506597.12469.python-list@python.org>
Message-ID: <3AB1F697.485A776D@yahoo.fr>

I got the same kind of problem on WinNT and Tcl80.dll,
I did the same tinkering you did and saw no improvement either.
Finally, I cleaned up the registry of all useless path (a poor install
program had set some C:\RECYCLER\123456xxx\... in many paths).
Do not forget to reboot (actually I logged off).
I suspect one PATH variable to have reached a maximum length.
Anyway, the PATH variables on NT are totally unclear to me

Skip Montanaro wrote:

> I'm trying for the first time to use Python (2.0) on Win98 with external
> DLLs (GTK/GDK/...).  I installed all the various bits and pieces for
> GIMP/GTK and PyGTK, but three DLLs that come with GIMP/GTK are not where
> Windows is searching for them, so I get import errors.  (GIMP runs
> fine.) Using a tool called the dependency walker (something from MS) I
> am able to extend its module search directories to include C:\Program
> Files\Common Files\GNU so it will resolve those DLLS.
>
> How do I perform that trick so that Microsoft's equivalent of ld.so can
> find them at run-time?  I assume there is some key in the registry I can
> set, but I haven't stumbled on anything obvious so far.  I did extend
> the registry key HKEY_LOCAL_MACHINE\Software\Python\PythonCore\2.0
> \PythonPath with the above directory, but that had no discernable
> effect.  I also checked to see what GIMP does.  It has a Path value
> associated with gimp.exe that refers to the DLL-containing directory, so
> I tried adding a string named Path with C:\Program Files\Common
> Files\GNU as its value to python.exe.  That didn't work either.  (Is
> this the equivalent of a Unix environment variable?) Is this another one
> of those "you must reboot for the changes to take effect" things?
>
> Thanks,
>
> unix-weenie-ly y'rs,
>
> Skip Montanaro
> skip at pobox.com



From nelson at crynwr.com  Fri Mar 30 11:01:40 2001
From: nelson at crynwr.com (Russell Nelson)
Date: Fri, 30 Mar 2001 11:01:40 -0500
Subject: Linux GPS program
References: <3AC01637.4EF3A6F2@crynwr.com>    <slrn9c17k4.cfc.jra@dorothy.msas.net> <3AC0D915.83C2C0C3@crynwr.com>    <slrn9c1p9t.d3g.jra@dorothy.msas.net> <99suho$m7a$2@newsserver.rrzn.uni-hannover.de> <slrn9c7ejl.sv.jra@dorothy.msas.net>
Message-ID: <3AC4ADE4.DD629457@crynwr.com>

"Jay R. Ashworth" wrote:
> No, actually, I'd bet some cash that the largest collection by size is
> the combined efforts of the US Govenment.  That's (almost) all free.

So, do you have any suggestions on what vector file format and dataset I
should aim at first?

-- 
-russ nelson will be speaking at http://www.osdn.com/conferences/brie/
Crynwr sells support for free software  | PGPok | Watch out!  He's got
an
521 Pleasant Valley Rd. | +1 315 268 1925 voice | opinion, and he's not
Potsdam, NY 13676-3213  | +1 315 268 9201 FAX   | afraid to share it!


From jafo at tummy.com  Fri Mar  2 11:51:03 2001
From: jafo at tummy.com (Sean Reifschneider)
Date: Fri, 2 Mar 2001 09:51:03 -0700
Subject: [Distutils] Re: CPAN functionality for python - requirements
In-Reply-To: <20010302091545.DD722373C95@snelboot.oratrix.nl>; from jack@oratrix.nl on Fri, Mar 02, 2001 at 10:15:45AM +0100
References: <jafo@tummy.com> <20010302091545.DD722373C95@snelboot.oratrix.nl>
Message-ID: <20010302095103.C7084@tummy.com>

On Fri, Mar 02, 2001 at 10:15:45AM +0100, Jack Jansen wrote:
>Right, but an IDE can do better. For instance, if an import fails it can offer 
>to do a lookup on the packages site to see whether such a package is 

The client program is in python, so I don't forsee any problems with
integrating the functionality in an IDE or whatnot.

Sean
-- 
 Good idea: Slaves Girls of Gor
 Bad idea: Slave Girls of Al Gore.
Sean Reifschneider, Inimitably Superfluous <jafo at tummy.com>
tummy.com - Linux Consulting since 1995. Qmail, KRUD, Firewalls, Python



From wware at world.std.com  Thu Mar 29 23:41:28 2001
From: wware at world.std.com (Will Ware)
Date: Thu, 29 Mar 2001 23:41:28 -0500
Subject: latitude/longitude/mapping Python script
Message-ID: <3AC40E78.D57EB9BE@world.std.com>

import string, urllib, re, os

JUST_THE_US = 0

def xerox_parc_url(marklist):
    avg_lat, avg_lon = max_lat, max_lon = marklist[0]
    marks = "%f,%f" % marklist[0]
    for lat, lon in marklist[1:]:
        marks = marks + ";%f,%f" % (lat, lon)
        avg_lat = avg_lat + lat
        avg_lon = avg_lon + lon
        if lat > max_lat: max_lat = lat
        if lon > max_lon: max_lon = lon
    avg_lat = avg_lat / len(marklist)
    avg_lon = avg_lon / len(marklist)
    if len(marklist) == 1:
        max_lat, max_lon = avg_lat + 1, avg_lon + 1
    diff = max(max_lat - avg_lat, max_lon - avg_lon)
    D = {'height': 4 * diff,
         'width': 4 * diff,
         'lat': avg_lat, 'lon': avg_lon,
         'marks': marks}
    if JUST_THE_US:
        url = ("http://pubweb.parc.xerox.com/map/db=usa/ht=%(height)f" +
               "/wd=%(width)f/color=1/mark=%(marks)s/lat=%(lat)f/" +
               "lon=%(lon)f/") % D
    else:
        url = ("http://pubweb.parc.xerox.com/map/color=1/ht=%(height)f" +
               "/wd=%(width)f/color=1/mark=%(marks)s/lat=%(lat)f/" +
               "lon=%(lon)f/") % D
    return url

def findcity(city, state):
    Please_click = re.compile("Please click")
    url = ("""http://www.astro.ch/cgi-bin/atlw3/aq.cgi?expr=%s&lang=e"""
           % (string.replace(city, " ", "+") + "%2C+" + state))
    lst = [ ]
    inf = urllib.URLopener().open(url)
    for x in inf.readlines():
        quitsoon = 0
        if Please_click.search(x) != None:
            lst = [ ]
            quitsoon = 1
        L = [ ]
        for y in string.split(x[:-1], '<'):
            L = L + string.split(y, '>')
        L = filter(lambda x: len(x) > 0, L)
        lst.append(L)
        if quitsoon and len(lst) == 20:
            break
    inf.close()
    a_href = re.compile("a href")
    okay = 0
    for x in lst:
        for y in x:
            if a_href.search(y) != None:
                okay = 1
                break
        if okay:
            break
    assert okay
    lat, lon = x[6], x[10]
    def getdegrees(x, dividers):
        if string.count(x, dividers[0]):
            x = map(eval, string.split(x, dividers[0]))
            return x[0] + (x[1] / 60.)
        elif string.count(x, dividers[1]):
            x = map(eval, string.split(x, dividers[1]))
            return -(x[0] + (x[1] / 60.))
        else:
            raise "Bogus result", x
    return getdegrees(lat, "ns"), getdegrees(lon, "ew")

def showcities(citylist):
    marklist = [ ]
    for city, state in citylist:
        lat, lon = findcity(city, state)
        print ("%s, %s:" % (city, state)), lat, lon
        marklist.append((lat, lon))
    url = xerox_parc_url(marklist)
    os.system('netscape "%s"' % url)

citylist = (("Natick", "MA"),
            ("Rhinebeck", "NY"),
            ("New Haven", "CT"),
            ("King of Prussia", "PA"))

citylist = (("Cancun", "Mexico"),
            ("Merida", "Mexico"),
            ("Chichen Itza", "Mexico"),
            ("Tulum", "Mexico"))

showcities(citylist)
-------------- next part --------------
A non-text attachment was scrubbed...
Name: wware.vcf
Type: text/x-vcard
Size: 386 bytes
Desc: Card for Will Ware
URL: <http://mail.python.org/pipermail/python-list/attachments/20010329/97eb397e/attachment.vcf>

From guido at beopen.com  Sun Mar 18 17:41:16 2001
From: guido at beopen.com (Guido van Rossum)
Date: Sun, 18 Mar 2001 22:41:16 GMT
Subject: What has become of PEP224 ?
References: <3ab4f320@nntp.server.uni-frankfurt.de> <992tb4$qf5$1@panix2.panix.com>
Message-ID: <gQat6.188849$B6.41790081@news1.rdc1.md.home.com>

[On c.l.py]
"Aahz Maruch" <aahz at panix.com> wrote in message
news:992tb4$qf5$1 at panix2.panix.com...
> [cc'd to Barry Warsaw in case he wants to comment]

(I happen to be skimming c.l.py this lazy Sunday afternoon :-)

> In article <3ab4f320 at nntp.server.uni-frankfurt.de>,
> Michael 'Mickey' Lauer  <mickey at Vanille.de> wrote:
> >
> >Hi. If I remember correctly PEP224 (the famous "attribute docstrings")
> >has only been postponed because Python 2.0 was in feature freeze
> >in August 2000. Will it be in 2.1 ? If not, what's the reason ? What
> >is needed for it to be included in 2.1 ?
>
> I believe it has been essentially superseded by PEP 232; I thought
> function attributes were going to be in 2.1, but I don't see any clear
> indication.

Actually, the attribute docstrings PEP is about a syntax for giving
non-function objects a docstring.  That's quite different than the function
attributes PEP.

The attribute docstring PEP didn't get in (and is unlikely to get in in its
current form) because I don't like the syntax much, *and* because the way to
look up the docstrings is weird and ugly: you'd have to use something like
instance.spam__doc__ or instance.__doc__spam (I forget which; they're both
weird and ugly).

I also expect that the doc-sig will be using the same syntax (string
literals in non-docstring positions) for a different purpose.  So I see
little chance for PEP 224.  Maybe I should just pronounce on this, and
declare the PEP rejected.

Unless Ping thinks this would be a really cool feature to be added to pydoc?
(Ping's going to change pydoc from importing the target module to scanning
its surce, I believe -- then he could add this feature without changing the
Python parser. :-)

--Guido van Rossum




From ghughes at lynda.com  Fri Mar 16 00:17:32 2001
From: ghughes at lynda.com (Graham Hughes)
Date: Fri, 16 Mar 2001 05:17:32 GMT
Subject: Who's minister of propaganda this week?
References: <B98041329020E007.7DA84CAD9D0562AB.817BBC572F480869@lp.airnews.net> <98nu1m$12e@dispatch.concentric.net> <660ACCA1E4816F7E.C359CDF4168EF3FB.3B6B05538D619770@lp.airnews.net> <3AAFA13D.76E32C3F@divalsim.it> <98oa4o01i4v@news2.newsguy.com> <mailman.984619039.7856.python-list@python.org> <98q9h60122v@news2.newsguy.com> <mailman.984678561.24064.python-list@python.org>
Message-ID: <87wv9qi978.fsf@walnut.treepeople.pedantic.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

>>>>> "Don" == Don Dwiggins <dwig at advancedmp.net> writes:

    Don> Alex Martelli writes:
    >> Another way to express it: an interface is not just, not even
    >> _mostly_, about existence and signature of methods -- it's
    >> mostly about prereq's, post-conditions, and invariants; and
    >> nobody can check those at compile-time in enough cases to make
    >> a difference to your software's reliability.

    Don> I agree whole-heartedly with the first clause, but I wonder:
    Don> if you made the same statement over on comp.lang.eiffel, what
    Don> kind of response would you get from folks who use the
    Don> language extensively?  Are Eiffel's prereq's,
    Don> post-conditions, and invariants mostly ignored in practice,
    Don> or are they commonly and effectively used?

They are run time--Alex is talking about compile time.  I'm not aware
of anyone who has a compiler that can check that the output of sqrt
times itself is the input.
- -- 
Graham Hughes <ghughes at lynda.com>
(defun whee (n e) (subseq (let ((c (cons e e))) (nconc c c)) 0 n))
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.4 (GNU/Linux)
Comment: Processed by Mailcrypt 3.5.5 and Gnu Privacy Guard <http://www.gnupg.org/>

iD8DBQE6saHo8i4DkPg349kRAn3JAJ9Tru1pnnFnulnqdsqt1G+MHjLdKACeLW4A
h9tVYwUkF+C2FcLVqSJIqbg=
=Kgjo
-----END PGP SIGNATURE-----


From echuck at mindspring.com  Wed Mar 14 13:03:15 2001
From: echuck at mindspring.com (Chuck Esterbrook)
Date: Wed, 14 Mar 2001 13:03:15 -0500
Subject: [ANNOUNCE] Using Mix-ins with Python
Message-ID: <5.0.2.1.0.20010314130304.02166440@mail.mindspring.com>

There is a Python article in the April 2001 issue of the Linux Journal, 
"Using Mix-ins with Python". It covers the general topic of mix-in 
programming with some comments on various languages, and then dives into 
Python specifics including examples.

I used these concepts when creating the MiddleKit component of Webware 
(http://webware.sourceforge.net/), which includes a mix-in function in 
Webware.MiscUtils.MixIn.


-Chuck
--
Get web dev at http://webware.sourceforge.net/ 




From loredo at spacenet.tn.cornell.edu  Mon Mar  5 17:18:04 2001
From: loredo at spacenet.tn.cornell.edu (Tom Loredo)
Date: Mon, 05 Mar 2001 17:18:04 -0500
Subject: PIL for the Mac?
References: <3A9C6FDF.3F9422C@alcyone.com> <ty%m6.15923$Qb7.2598968@newsb.telia.net>
Message-ID: <3AA4109C.D13BEE26@spacenet.tn.cornell.edu>

PS: Questions about this are best submitted to the Python-Mac SIG:

http://www.python.org/pipermail/pythonmac-sig/

-Tom Loredo


From will at nospam.misconception.org.uk  Fri Mar 16 20:58:15 2001
From: will at nospam.misconception.org.uk (Will Newton)
Date: Sat, 17 Mar 2001 01:58:15 +0000
Subject: Deleting dictionary items
Message-ID: <98ug6n$q9d$1@news6.svr.pol.co.uk>

What's the best way to delete dictionary items?

I have a dictionary with tuples as keys, and I would like to delete all 
keys with a certain element. First I wrote:

                for item in td.data.keys():
                        if item[attr] != value:
                                del td.data[item]


But will this work?


From max at alcyone.com  Sat Mar 10 13:34:39 2001
From: max at alcyone.com (Erik Max Francis)
Date: Sat, 10 Mar 2001 10:34:39 -0800
Subject: Private class and instance methods?
Message-ID: <3AAA73BF.9CDCF3B2@alcyone.com>

How often do people use methods beginning with __ in order to get some
amount of privacy with the automatic name mangling that occurs, to get
some kind of weak protection, with, say, accessor methods:

    class C:
        def __init__(self, x):
            self.__x = x

        def getX(self): return self.__x
        def setX(self, x): self.__x = x

[which would of course allow C.getX and C.setX to do something different
in the future]?

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ Nothing is true -- all is permissible.
\__/ Hassan i Sabbah
    Product's Quake III Arena Tips / http://www.bosskey.net/
 Tips and tricks from the absolute beginner to the Arena Master.


From tuttledon at hotmail.com  Sat Mar 10 13:36:35 2001
From: tuttledon at hotmail.com (Don Tuttle)
Date: Sat, 10 Mar 2001 18:36:35 GMT
Subject: tkinter, activepython, win32
References: <OHiq6.4422$tP3.71012@news1.rdc1.bc.home.com>
Message-ID: <Tuuq6.63414$__6.12668586@typhoon.southeast.rr.com>

"Luke" <anduguid at NOThome.com> wrote in message
news:OHiq6.4422$tP3.71012 at news1.rdc1.bc.home.com...
> python newbie!
> cannot get tkinter programs to run...
>
> did 'normal' install of activepython 2.0 build 203
> installed Tcl/Tk 8.3 for Windows, Binary Distribution
>
> am i missing something here?
No.  It's broken.

If you wan't to play with TK right now, download PythonWare's distribution.
http://www.pythonware.com/products/python/index.htm

It's doesn't change anything in your system's environment so it's safe to
install beside ActivePython.

And while you're there, check out the free draft of Fredric's upcoming
Tkinter book:
http://www.pythonware.com/people/fredrik/tkintrobook.htm

Fredric is Python's chief TK champion. ;-)

Don





From robin at stop.spam.alldunn.com  Mon Mar 26 22:00:49 2001
From: robin at stop.spam.alldunn.com (Robin Dunn)
Date: Mon, 26 Mar 2001 19:00:49 -0800
Subject: warrior geeks
References: <3ABF72F1.98CB78F@sage.att.com>
Message-ID: <znTv6.762$cgz.2752875@news.randori.com>

> it seemed like an odd coincidence, but it made me wonder: is there an
> unusually high percentage of MA practitioners among the python
> community?
> or the open source community?  or software in general?

I don't know about Martial Arts, but almost every other person named Robin
that I know are in the Python community.  I've always thought that there is
some sort of cosmic thing going on there...


--
Robin Dunn
Software Craftsman
robin at AllDunn.com       Java give you jitters?
http://wxPython.org      Relax with wxPython!





From Laura_01 at MailAndNews.com  Sun Mar 11 23:22:52 2001
From: Laura_01 at MailAndNews.com (Laura Lewin)
Date: Sun, 11 Mar 2001 23:22:52 -0500
Subject: Python In A Nutshell
Message-ID: <3AAE7E0D@MailAndNews.com>

Hi,
This is very helpful feedback.  Thanks a lot!
Laura
LLewin at oreilly.com

>===== Original Message From Hamish Lawson <hamish_lawson at yahoo.co.uk> =====
>I think "Perl in a Nutshell" serves as a good model: fast-track
>tutorial on the language, reference on standard modules, and a
>reference on the more common third-party modules.
>
>In the standard modules section I'd hope to see nearly all the modules
>other than perhaps those that are obsolescent, platform-specific, or
>somewhat esoteric; one of the things I miss in David Beazley's "Python
>Essential Reference" is coverage on htmllib, sgmllib and telnetlib. I'd
>like to see some brief documentation of Tkinter and more extended
>coverage of PyXML than is given in the online library reference.
>
>For the third-party modules I'd nominate the following, grouped by
>desirability of inclusion:
>
>    ESSENTIAL
>       DB-API
>       mxDateTime
>       Python Imaging Library
>       pyunit
>       win32all
>
>    GOOD
>       HTMLgen
>       mxTextTools
>       ReportLab
>       wxPython
>       xmlrpc
>       soaplib
>
>    NICE
>       mod_python / mod_snake
>       Numerical Python
>       freeze/packaging
>
>
>Hamish Lawson
>
>
>____________________________________________________________
>Do You Yahoo!?
>Get your free @yahoo.co.uk address at http://mail.yahoo.co.uk
>or your free @yahoo.ie address at http://mail.yahoo.ie



From SBrunning at trisystems.co.uk  Tue Mar 27 03:23:43 2001
From: SBrunning at trisystems.co.uk (Simon Brunning)
Date: Tue, 27 Mar 2001 09:23:43 +0100
Subject: how to write a text file search & replace script
Message-ID: <31575A892FF6D1118F5800600846864D78BA38@intrepid>

The script below might be of some use. To use it, set up a text file with
all your from values and to values in it (one pair per line, tab separated),
and run it over your files.

#!/usr/bin/env python
# Module     : multirep.py
# Synopsis   : Multiple replacements
# Programmer : Simon Brunning - sbrunning at bigfoot.com
# Date       : 20/09/2000
# Notes      : Thanks to Fredrik Lundh for the MultiReplace class, which I
have butchered.
'''Perform multiple replacements.
Takes a list of delimited from and to values from a replacements file,
and replaces the from values with the to values in the target file(s).
The target files are not changed - new version(s) of the target file(s)
are written, with the filename prefixed.

Usage  : Python multirep.py [options] replacementsfile targetfile(s)
Options: -h = help
         -d = delimiter (defaults to tabs, s = spaces, c = commas)
         -p = prefix for new files (defaults to 'New')
         -c = case insensitive match
         -w = replace whole words only'''

def multirep(arguments):
    import getopt, glob, operator
    
    # Split arguments list into options and arguments
    options, arguments = getopt.getopt(arguments, '?hd:p:wc')

    # Set defaults
    delimiter = '\t'
    prefix = 'New'
    wholeWords = None
    caseInsensitive = None

    # Options - override defaults ond show help    
    for option, value in options:
        if option[-1] in '?h':
            print; print __doc__
        elif option[-1] == 'd': # Specify delimiter
            if value == 's':
                delimiter = ' '
            elif value == 'c':
                delimiter = ','
        elif option[-1] == 'p': # Specify prefix
            prefix = value
        elif option[-1] == 'w': # Whole words
            wholeWords = 1
        elif option[-1] == 'c': # case insensitive
            caseInsensitive = 1

    # Build replacement function from replacements file
    try:
        replacer = MultiReplacer(arguments[0], delimiter, wholeWords,
caseInsensitive)
    except IndexError:
        print; print __doc__
        return
    except ValueError:
        print; print 'Invalid replacements file.'
        return
    
    # Expand remaining arguments into target file list
    try:
        targetFiles = reduce(operator.add, map(glob.glob, arguments[1:]))
    except TypeError:
        print; print __doc__
        return

    # Perform replacement on each file
    for file in targetFiles:
        replaceFile(file, replacer, prefix)

def replaceFile(infile, replacer, prefix='New'):
    import os

    # Build outfile name    
    outfile = os.path.join(os.path.dirname(infile), ''.join((prefix,
os.path.basename(infile))))

    # Read from infile, replace values, and write to outfile    
    open(outfile, 'wb').write(replacer(open(infile, 'rb').read()))

class MultiReplacer:
    def __init__(self, replacements, delimiter='\t', wholeWords=None,
caseInsensitive=None):

        # Build replacements dictionary - may come in as a mapping or as a
file         
        self.replacements = {}
        try:
            # replacements is a mapping
            self.replacements.update(replacements)
        except TypeError:
            # replacements is a file
            for line in open(replacements, 'r').readlines():
                fromValue, toValue = line.split(delimiter)[:2] # Split line
                
                while toValue[-1] in '\r\n': # Strip newlines
                    toValue = toValue[:-1]

                self.replacements[fromValue] = toValue # Add to dictionary
  
        # Build char to char mapping...
        self.charMap = None
        if not wholeWords:
            charMap = map(chr, range(256))
            for fromValue, toValue in self.replacements.items():
                if len(fromValue) <> 1 or len(toValue) <> 1:
                    break
                if caseInsensitive:
                    charMap[ord(fromValue.upper())] = toValue
                    charMap[ord(fromValue.lower())] = toValue
                else:
                    charMap[ord(fromValue)] = toValue
            else:
                self.charMap = "".join(charMap)
                return

        # String to string mapping - use a regular expression
        import re
        fromVals = replacements.keys().sort()

        # Build regexp pattern
        if not wholeWords:
            rePattern = '|'.join(map(re.escape, fromVals))
        else:
            rePattern = r'\b(' + '|'.join(map(re.escape, fromVals)) + r')\b'
        
        # Compile regexp
        if caseInsensitive: 
            self.rePattern = re.compile(rePattern, re.I)
        else:
            self.rePattern = re.compile(rePattern)

    def __call__(self, string):
        # apply replacement to string
        
        # Char to char mapping
        if self.charMap: 
            return string.translate(self.charMap)

        # String to string mapping        
        return self.rePattern.sub(self.__replaceMatch, string)
    
    def __replaceMatch(self, match):
        item = match.group(0)
        return self.replacements.get(item)
        
if __name__ == '__main__':
    import sys
    multirep(sys.argv[1:])

Cheers,
Simon Brunning
TriSystems Ltd.
sbrunning at trisystems.co.uk





-----------------------------------------------------------------------
The information in this email is confidential and may be legally privileged.
It is intended solely for the addressee. Access to this email by anyone else
is unauthorised. If you are not the intended recipient, any disclosure,
copying, distribution, or any action taken or omitted to be taken in
reliance on it, is prohibited and may be unlawful. TriSystems Ltd. cannot
accept liability for statements made which are clearly the senders own.



From anbu at adsc.com  Wed Mar 14 19:56:52 2001
From: anbu at adsc.com (Anbuselvam Palusamy)
Date: Wed, 14 Mar 2001 16:56:52 -0800
Subject: Package
Message-ID: <001c01c0acea$d33c6810$9071fea5@ANBU>

Can any one explain how to create a packages in python?

Thanks
Anbu
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20010314/a3061f65/attachment.html>

From Dan.Rolander at marriott.com  Fri Mar  2 08:11:47 2001
From: Dan.Rolander at marriott.com (Rolander, Dan)
Date: Fri, 2 Mar 2001 08:11:47 -0500 
Subject: How can I get the logged in users username on Win32?
Message-ID: <6176E3D8E36FD111B58900805FA7E0F80CCF63EC@mcnc-mdm1-ex01>

It's pretty easy to use Python with WSH. Install the win32all extensions,
which will register the ActiveX scripting engine. Then just create a python
script with a .pys extension. Here's an example that will display the
machine name and currently logged on user:

  # wsh.pys
  # A Windows Scripting Host file that uses Python
  objNet = WScript.CreateObject("WScript.Network")
  WScript.Echo(objNet.ComputerName + " - " + objNet.UserName)

To run it use the command: wscript "W:\My Documents\Source
Code\WSH\PythonScripts\wshdemo.pys"

There is a chapter on Active Scripting in the book Python Programming on
Win32 by Mark Hammond & Andy Robinson. I highly recommend you get it.

HTH,
Dan

-----Original Message-----
From: nomad***@***freemail.absa.co.za
[mailto:nomad***@***freemail.absa.co.za]
Sent: Friday, March 02, 2001 2:56 AM
To: python-list at python.org
Subject: [Q]: How can I get the logged in users username on Win32?


Howzit,

After searching through the Python Docs, I found that on Un*x I could
use getpass.getuser(), but there is no equivalent for the Win32
platform.

In other words, how can I get the username in Windows?

As a side note, how can I enable Windows Scripting Host support for
python - there is mention of python in the original WSH docs - and if
I can use python scripts in WSH how would I call the built-in
Wscript.<methods>

TIA

nomad
-- 
http://mail.python.org/mailman/listinfo/python-list



From tshumway at transafari.com  Fri Mar  2 14:07:08 2001
From: tshumway at transafari.com (Terrel Shumway)
Date: Fri, 02 Mar 2001 14:07:08 -0500
Subject: [Webware-discuss] Re: [ANNOUNCE] Webware for Python 0.5
References: <5.0.2.1.0.20010301205419.047bfd90@mail.mindspring.com> <3A9FB474.A57363A4@NameConnector.com>
Message-ID: <3A9FEF5C.7B8465BA@transafari.com>

Geoff Talvola wrote:

> Chuck Esterbrook wrote:
>
> > "Don Tuttle" <tuttledon at hotmail.com> wrote in message
> > news:<fFCn6.13480$Z8.1902263 at typhoon.southeast.rr.com>...
> > >"Chuck Esterbrook"
> > > >    * Python Server Pages (PSP, similar to ASP, PHP and JSP)
> > >I 'm trying to get a handle on why I would want to use PSP on WinNt with
> > >IIS.  What's the difference/advantage to using PSP compared to using ASP
> > >with embedded Python script?
> > >Don
>
> I recall that ASP with Python places limitations on what you can store in session variables.  For example, if you try to place a dictionary into a session variable you'll get an error like "TypeError: Objects of type 'dictionary' can not be converted to a COM VARIANT".  Generally, only "simple" types like integers, floats, strings, and lists of them can be placed into a session variable.  WebKit/PSP has no such restrictions.
>

You can solve this problem easily with pickle.

The larger problem with using Python in ASP is that the ASP processor removes all "non-significant" white space before it sends the code to the interpreter.  Python's idea of "non-significant" white space is much more restrictive than other interpreters.  The easiest way to fix this would be to modify the python scanner to use non-white tokens to represent indents and outdents, then use your custom interpreter for the ASP.


> On the flip side, one possible advantage of ASP with Python is that you might be able to take advantage of 3rd-party extensions that are designed to only work with ASP.  I don't know if such things exist.  I happen to be using a 3rd-party COM component in my web pages to draw dynamic graphs.  The component was designed for use with Visual Basic and Active Server Pages, but it works great with WebKit too.
>

Third party extensions designed to work with ASP?  I think the model MS encourages is to to put such logic into COM components, which are just as easily accessible to Python as they are from VBScript.  In general, as your app grows you would put more of it into separate components.  This is a Good Thing for PSP too.  Keep as much logic out of *SP as possible.  The most complicated thing you should put into a template page is a simple loop to build a table.  Most of it should be HTML with an occassional <%=object.property%> or
<%=object.method()%>. The advantage Python gives you here is that it is much easier to do such refactoring if you can create components without messing with COM.

> ASP/Python also has the dubious advantage of being an easier sell within a corporation than WebKit/PSP, which may or may not be important to you...
>

It will be interesting to see how well Webware could integrate into IIS with a new IISAdapter (currently not even vaporware).  Python for .NET will give it some marketing muscle.  I can see a lot of VB developers moving to python when that gets solid.

    HTH
    -- Terrel





From michael at stroeder.com  Sun Mar  4 17:30:08 2001
From: michael at stroeder.com (Michael =?iso-8859-1?Q?Str=F6der?=)
Date: Sun, 04 Mar 2001 23:30:08 +0100
Subject: Concurrency and shelve
References: <3AA26D5E.2FD28B40@stroeder.com> <kuk865l6e1.fsf@lasipalatsi.fi>
Message-ID: <3AA2C1F0.980A23F7@stroeder.com>

Erno Kuusela wrote:
> 
> In article <3AA26D5E.2FD28B40 at stroeder.com>, Michael Str?der
> <michael at stroeder.com> writes:
> 
> | Can anybody summarize issues with shelve (based on anydbm) and
> | concurrent access?
> 
> afaik only gdbm does locking, the others break.

Hmm, it seems to me that gdbm does prevent other processes from
opening the database at all.

Python 2.0 (#1, Oct 18 2000, 18:49:22) 
[GCC 2.95.2 19991024 (release)] on linux2
Type "copyright", "credits" or "license" for more information.
>>> import gdbm
>>> d1=gdbm.open('/home/michael/test.dbm','c')
>>> d2=gdbm.open('/home/michael/test.dbm','c')
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
gdbm.error: (11, 'Die Ressource ist zur Zeit nicht verf\374gbar')
>>>

BTW: I'm trying to use shelve to store web session data. How does
others solve this problem with the the standard lib without being to
DB-specific? Unpossible?

Ciao, Michael.


From bedge at troikanetworks.com  Thu Mar 15 11:38:33 2001
From: bedge at troikanetworks.com (Bruce Edge)
Date: Thu, 15 Mar 2001 08:38:33 -0800
Subject: building python for multiple targets
Message-ID: <3AB0F009.E27D9AD7@troikanetworks.com>

I need to build Python for multiple targets from one host.

One target is the host, the other is for embedded Linux, with it's own
compiler, libs, include, etc, all of which are installed in say
/usr/local/rtlinux/...

The host build is simple, ./configure, make, make install.

The target one is ugly, currently I'd doing:
  ./configure --include=<rtpath> --bindir=/usr/local/rtlinux/bin
--sbindir=</usr/local/rtlinux/sbin ....<every other damn ./configure
option there is>

Is there a better way?


From qrczak at knm.org.pl  Fri Mar  9 03:48:22 2001
From: qrczak at knm.org.pl (Marcin 'Qrczak' Kowalczyk)
Date: 9 Mar 2001 08:48:22 GMT
Subject: Can anyone recomend a good intoduction to C...
References: <mailman.983894288.19965.python-list@python.org> <ewwp6.1466$y6.241005@ruti.visi.com> <slrn9aeblb.cg1.scarblac@pino.selwerd.nl> <djagat4kk3cs9hg0c3k39cudk6goblhkg6@4ax.com> <989mbk$24nuv$1@fido.engr.sgi.com>
Message-ID: <slrn9ah66m.k09.qrczak@qrnik.zagroda>

9 Mar 2001 04:31:48 GMT, Paul Jackson <pj at sgi.com> pisze:

> Bjorn's goal, clearly, was to support a number of styles,
> of which O-O is just one.

It completely fails for the functional style: no garbage collection, so
it's hard to return by reference and share substructures, no functions
as first class values (no nested functions and no partial application),
no algebraic types (tagged union of products), no laziness (and hard
to emulate it because of lack of nested functions).

I know C++ well and dislike it so much. Fortunately I don't have to
use it.

I dislike C too, but at least there is an area where it's a good
choice: low-level programming, i.e. OS interface, OS itself, runtime
systems for other languages, time-critical modules for other languages.
C++ doesn't fit there, is unnecessarily too complicated for these
tasks, and it isn't a high level language either.

IMHO it's used mostly because people knew C before, needed to use a
higher level language, didn't want to change C's habits or lose C's
libraries, and failed to realize that there is plenty of real high
level languages out there. Then C++ became popular enough to live
its own life.

Designers of high level languages probably have paid too little
attention to interoperability with other languages. Especially with C,
as the practice shows - system-level libraries are almost always in C,
at least under Unix. C is easy to use from C++; you don't even need
a separate manual for a C++ interface to a C library.

Language interoperability is fun.

working-on-Haskell<->Python-binding-ly y'rs
-- 
 __("<  Marcin Kowalczyk * qrczak at knm.org.pl http://qrczak.ids.net.pl/
 \__/
  ^^                      SYGNATURA ZAST?PCZA
QRCZAK


From emile at fenx.com  Thu Mar  8 16:08:31 2001
From: emile at fenx.com (Emile van Sebille)
Date: Thu, 8 Mar 2001 13:08:31 -0800
Subject: gah! I hate the new string syntax
References: <mailman.983615170.23313.python-list@python.org> <tN5o6.2746$eH3.14913@NewsReader> <Koco6.7518$1D5.244602@e420r-atl1.usenetserver.com> <mailman.983997252.5628.python-list@python.org> <m2vgpkyfjo.fsf@mycroft.actrix.gen.nz> <mailman.984063199.18779.python-list@python.org> <ffQp6.9$Tg.2012@ruti.visi.com>
Message-ID: <988tmr$t10b$1@ID-11957.news.dfncis.de>


"Grant Edwards" <grante at visi.com> wrote in message
news:ffQp6.9$Tg.2012 at ruti.visi.com...
<snip>

> I've always boggled at Americans' habit of claiming
foriegn
> nationality based on where one or two of their great,
great,
  ~~~~~~~~~~~

I suspect that nationality is confused with heritage.
Whether it's intentional or not has always boggled me!  ;-)


--

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





From grante at visi.com  Tue Mar 27 10:54:44 2001
From: grante at visi.com (Grant Edwards)
Date: Tue, 27 Mar 2001 15:54:44 GMT
Subject: Why not Tcl/Tk?
References: <4211ctgvusgcma0c6ok1cpmjou98lv9714@4ax.com>
Message-ID: <8J2w6.3336$SB2.606281@ruti.visi.com>

In article <4211ctgvusgcma0c6ok1cpmjou98lv9714 at 4ax.com>, Alexander Kluge wrote:
>Here are some very general newbie questions and a plea for your opinion.
>
>Having some outdated hobby hacking experience (Turbopascal, Modula2,
>DBase) on AtariST and PC, and some recent experience with a 70k perl cgi
>script consecutivly changed by different persons (@!?$@*!!!), I think
>that I actually don't want to use neither C, C++ or Java (overkill) nor
>perl (see above). I confess having used the try and error method of
>learning, therefore my autodidactic abilities stopped every time dynamic
>arrays were needed :). 
>I don't want neither spend much time in learning the language nor
>writing professional strength applications but I rather look for a
>replacement of very easy languages as e.g. pascal/basic.
>
>I'm interested in your opinions concerning the following points:
>
>-How would you describe the major strengths and flaws of Tcl vs. Python?

Tcl 
 Flaws:  
     Only one data type: string
     Incomprehensible quoting semantics
     Roll-your-own control flow
     No OO support
 Strengths:
     Tk integration
     Lots of code out there
     Roll-your-own control flow
     
Python:
 Flaws:
     No roll-your-own control flow
     Lots of GUI options
 Strengths:
     No roll-your-own control flow
     Good OO support
     Lots of GUI options
     Simple syntax and semantics
     Library modules 

>-Why do you prefer the one you use over the other?

I gave up on Tcl after one program and switch to Scheme.  Now I
use Python a lot more than Scheme.

[I don't know much about eithr on Win32, so I'll let others
answer those questions.]

-- 
Grant Edwards                   grante             Yow!  As President I
                                  at               have to go vacuum my coin
                               visi.com            collection!


From scarblac at pino.selwerd.nl  Tue Mar 20 04:15:53 2001
From: scarblac at pino.selwerd.nl (Remco Gerlich)
Date: 20 Mar 2001 09:15:53 GMT
Subject: Functionalism, aesthetics Was:(RE: I come to praise .join,  not
References: <mailman.985052844.17993.python-list@python.org> <996vss013mp@news2.newsguy.com>
Message-ID: <slrn9be7pb.b0k.scarblac@pino.selwerd.nl>

Alex Martelli <aleaxit at yahoo.com> wrote in comp.lang.python:
> So why did the elitist design look better?  Perhaps because it's a
> common subconscious mechanism to prefer the luxury good, even
> when in actual fact one won't be able to take advantage of it, as
> a wishful-thinking byproduct.  Remember that human beings are
> quite prone to a few well-known systematic perception errors --
> and it takes LOTS of training to teach one of us to stop making
> them.  Thus, a design which happens to hit squarely on the button
> one of those common perception errors may well be quite pleasing,
> superficially, to the untrained masses -- only highly experienced
> professionals really stand a chance to smell something fishy with
> it. 

"For every problem, there is a solution that is simple, elegant and wrong",
right?

-- 
Remco Gerlich


From ralph at inputplus.demon.co.uk  Thu Mar 22 05:47:35 2001
From: ralph at inputplus.demon.co.uk (Ralph Corderoy)
Date: 22 Mar 2001 10:47:35 -0000
Subject: I'm sure there's a good reason!
References: <mailman.985214563.25964.python-list@python.org>
Message-ID: <99cl87$3nq$1@inputplus.demon.co.uk>

> if-one-string-gimmick-were-enough-for-all-purposes-there-wouldn't-
>     be-two<wink>-ly y'rs  - tim

A PEP for Hollerith constants anyone?


Ralph.



From malte.forkel at berlin.de  Fri Mar 30 01:48:34 2001
From: malte.forkel at berlin.de (Malte Forkel)
Date: Fri, 30 Mar 2001 08:48:34 +0200
Subject: Python 2.0 vs 68k Mac
Message-ID: <9a1agj$o55$1@news.cs.tu-berlin.de>

Has anyone build Python 2.0 for 68k Macs? Please cc: your reply to me
(malte.forkel at berlin.de).

Thanks, Malte

----------------------------------------------------------------------------
Malte Forkel                                             Tel +49 30 21018926
Kurfuerstenstrasse 127                                   Fax +49 30 21018927
10785 Berlin
Germany                                        mailto:malte.forkel at berlin.de




From max at alcyone.com  Sat Mar 10 21:22:14 2001
From: max at alcyone.com (Erik Max Francis)
Date: Sat, 10 Mar 2001 18:22:14 -0800
Subject: Private class and instance methods?
References: <3AAA73BF.9CDCF3B2@alcyone.com> <B6D000FE.4D14%f8dy@diveintopython.org>
Message-ID: <3AAAE156.79E64ABF@alcyone.com>

Mark Pilgrim wrote:

> There is a better way to do this in Python, using the __getattr__ and
> __setattr__ special methods.

Ah, good point.  That didn't occur to me right away.  Thanks to you and
Paul Foley in private email for the suggestion.  (Note that I didn't
actually _plan_ on using private methods and accessor methods, I was
just thinking about the Python analogue.)

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ Love is the wisdom of the fool and the folly of the wise.
\__/ Samuel Johnson
    Interstelen / http://www.interstelen.com/
 A multiplayer, strategic, turn-based Web game on an interstellar scale.


From jarioksa at cc.oulu.fi  Thu Mar  8 04:17:11 2001
From: jarioksa at cc.oulu.fi (Jari Oksanen)
Date: Thu, 08 Mar 2001 11:17:11 +0200
Subject: Can anyone recomend a good intoduction to C...
References: <3aa366c6_2@nnrp1.news.uk.psi.net> <8gNo6.440$y6.128357@ruti.visi.com> <3aa6551e.212870101@news.laplaza.org> <Taup6.1292$y6.235409@ruti.visi.com>
Message-ID: <3AA74E17.740EFDF@cc.oulu.fi>

Grant Edwards wrote:
> 
> In article <3aa6551e.212870101 at news.laplaza.org>, Mats Wichmann wrote:
> 
> >(oops, I smell a flood of "I can top that" comments such as "you had a
> >terminal? I learned on a Teletype", "you had a Teletype? I had to
> >toggle code in using switches", "you had switches?  I had to use
> >Forth!".  Consider that joke already played out, eh).
> 
> You had 1's?  In my day we had to write everything using nothign but 0's!
> 
> [That's out of some Dilbert strip I believe]
> 
I think that's out of Russel's and Whitehead's "Principia mathematica."

jo.
-- 
Jari Oksanen -- Oulu, Finland


From fredrik at pythonware.com  Fri Mar  2 02:40:17 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 02 Mar 2001 07:40:17 GMT
Subject: gah! I hate the new string syntax
References: <mailman.983485923.1136.python-list@python.org>
Message-ID: <B7In6.16281$Qb7.2658902@newsb.telia.net>

Sean 'Shaleh' Perry wrote:
> return ";".join(["%s=%s" % (k, params[k]) for k in params.keys()])
>
> every day python seems to be moving closer to the line noise aspect of
> coding.

you're using list comprehensions and the wrong dictionary method,
and then you're complaining about string methods? ;-)

("items()" returns a key/value tuple that you can pass directly to
the formatting operator.  and nobody will complain if you stick to
string.join for stuff like this...)

Cheers /F




From sholden at holdenweb.com  Thu Mar 22 22:32:22 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 22 Mar 2001 22:32:22 -0500
Subject: Why Python/Jython?
References: <99cl9e$b2b$1@Merlin.et.bocholt.fh-gelsenkirchen.de> <mailman.985260089.28444.python-list@python.org> <1901D50DC30D30E4.16551C1AB0FB4363.C0D72E208759F8A7@lp.airnews.net> <99dkg301894@news1.newsguy.com> <3dy9txzed1.fsf@ute.cnri.reston.va.us>
Message-ID: <uwzu6.12885$_46.477797@e420r-atl2.usenetserver.com>

"Andrew Kuchling" <akuchlin at mems-exchange.org> wrote in message
news:3dy9txzed1.fsf at ute.cnri.reston.va.us...
> "Alex Martelli" <aleaxit at yahoo.com> writes:
> > *blink* -- I'm hardly a Java-ite, myself, but
> > "install and configure a minimal JDK" I would
> > estimate as typically a 10-minute job (on, say,
> > NT -- is it an order of magnitude harder on,
> > say, Linux?!).  Maybe I misunderstand what
> > you mean...?
>
> A co-worker of mine once took an hour to install the JDK on Solaris;
> most of this time was spent lost on Sun's horribly organized Web site
> trying to find the exact version required.
>
However this is simply due to Java's immaturity. As biotechnologists might
say of light bulbs, a correctly-configured installer will work out what it
needs, download itself, run, and ask you whether you want to reboot or load
Linux.

just-jokingly y'rs  - steve






From mike.mcleod at bt.com  Fri Mar 30 05:11:28 2001
From: mike.mcleod at bt.com (mike mcleod)
Date: Fri, 30 Mar 2001 11:11:28 +0100
Subject: HELP urllib POST can't get it to work
Message-ID: <9a1m91$fdg$1@pheidippides.axion.bt.co.uk>

My code:

import urllib
params = urllib.urlencode({'address': '07', 'identity': 'me',
                           'resultpg':'html/sendresult.html', 'message':
'test',
                           'fPage':'true',
                           })
f = urllib.urlopen("http://www.samemachine.uk/bin/afunc.pl?", params)
print f.read()

It looks like the order of the params come out different in the urlopen.
But because the server will only accept in the order shown it fails.
I have tried to do this using basic sockets but don't know how to get that
work without lots of reading.

Help please.
Mike McLeod.
My thoughts only....






From cfelling at iae.nl  Tue Mar  6 19:06:13 2001
From: cfelling at iae.nl (Carel Fellinger)
Date: 7 Mar 2001 01:06:13 +0100
Subject: polymorphjsm &c (was Re: I come to praise .join, not to bury it...)
References: <mailman.983485923.1136.python-list@python.org> <97p7iq$kk2$1@nntp6.u.washington.edu> <97rka002k9@news1.newsguy.com> <9811a8$k5k$1@nntp6.u.washington.edu> <982cja0v5c@news1.newsguy.com>
Message-ID: <983u1l$6mk$1@animus.fel.iae.nl>

once Alex Martelli wrote:
...
> that's a different flamewar ("form follows function" crusaders,
> this IS a call to arms!-).

so you wanna fight he,

and else were he wrote
... bunch of disgustingly true ramblings on polymorphisme snipped


and now he comes with
> class MixedJoiner:
...
> Thanks to the fully general polymorphism that joiner.join affords,
> we can then use a *special* joiner-object to do fancy joining.

So let's reverse things:

>>> class MixedSplitter:
...     def __init__(self, splitter1, splitter2):
...         self.splitter1, self.splitter2 = splitter1, splitter2
...     def split(self, sequence):
...         s = self.splitter1(sequence)
...         return s[:-1] + self.splitter(s[-1])
... 
>>> ms = MixedSplitter(', ', ' and ')
>>> ms.split('1, 2, 3 and 4')
some nasty errormessage snipped


This won't work with Python2.
Ofcourse we could ammend MixedSplitter's split method like:

>>> def split(self, sequence):
...     s = sequence.split(self.splitter1)
...     return s[:-1] + s[-1].split(self.splitter2)
...
>>> MixedSplitter.split = split
>>> ms.split('1, 2, 3 and 4')
['1', '2', '3', '4']

But the real evil is that no code out there will allow us to easily
use the polimorphic nature of this splitter.


The moral being that we shouldn't argue about the uglines joiner.join
but of the lack of a tru polymorphic splitter.split!
-- 
groetjes, carel


From danielk at aracnet.com  Thu Mar  8 09:20:42 2001
From: danielk at aracnet.com (Daniel Klein)
Date: Thu, 08 Mar 2001 06:20:42 -0800
Subject: Who am I: can a class instance determine its own name?
References: <mailman.984034150.9635.python-list@python.org>
Message-ID: <fr4fatgt9n6134sjcnhmnd7amog2lnin7b@4ax.com>

On Thu, 08 Mar 2001 17:48:24 +1100, Tim CHURCHES <TCHUR at doh.health.nsw.gov.au>
wrote:

>...can an instance of a class determine the name of the variable to which it is assigned?

It might help if you could tell us why you need to know this? I recently posted
the same question to this group but found that using the object's natural
identity, id(object), proved the best way to manage this.

Daniel Klein
Portland OR USA


From gReGlAnDrUm at earthlink.net  Fri Mar 16 17:42:23 2001
From: gReGlAnDrUm at earthlink.net (Greg Landrum)
Date: Fri, 16 Mar 2001 14:42:23 -0800
Subject: Extending using Numeric, a quest for patterns
Message-ID: <98u4q9$kip$1@slb6.atl.mindspring.net>

Hi all,

I'm in the process of converting some of the time-critical sections of my
code to C extension modules.  I use Numeric heavily, so I'm working with a
lot of Numeric arrays.  I've figured out how to do most everything I need to
do, but I'm hung up a bit on one particular piece.  So I'm looking for some
pointers towards either more documentation or suggestions for patterns to
use.

The core of the Python for the bit I'm interested in converting is:
          posArray = array(pos1+pos2)
          avgVect = sum(posArray)/len(posArray)
          devs = posArray-avgVect
          devs *= devs
          e = sum(sum(devs))
where pos1 and pos2 are python lists.

What I think would make a lot of this easier is if I could figure out how to
use the ufuncs from C code.  I pored over the source and the docs, but I was
pretty much totally unsuccessful in this undertaking.

Is there a straightforward way of doing this?

Thanks,
-greg




From dbrueck at edgix.com  Mon Mar 19 10:02:47 2001
From: dbrueck at edgix.com (Dave Brueck)
Date: Mon, 19 Mar 2001 08:02:47 -0700
Subject: Killing threads
In-Reply-To: <98u6pt$6j$1@panix6.panix.com>
Message-ID: <BLEOJGEMBNBNEAKMIFDHGEIGPMAA.dbrueck@edgix.com>

> -----Original Message-----
> From: python-list-admin at python.org
> [mailto:python-list-admin at python.org]On Behalf Of Aahz Maruch
> Sent: Friday, March 16, 2001 4:15 PM
> To: python-list at python.org
> Subject: Re: Killing threads
>
>
> In article <mailman.984769881.20374.python-list at python.org>,
> Dave Brueck <dbrueck at edgix.com> wrote:
> >
> >Or just a simple flag. Here's a class I use in designs that use
> work flows
> >(work objects pulled off one queue, processed, and moved onto the next
> >queue). If you just need an always-processing background thread you can
> >eliminate all the stuff with the semaphore.
>
> I'm confused.  Is there some reason you're not using Queue.Queue()?

Hi Aahz,

Queue is overkill for simple one-to-one workflows - it's built to be
multi-consumer and multi-producer so every get/put/check requires acquiring
and releasing the Queue's mutex (granted, in most cases that probably won't
make too big of a different, but...).

Anyway, the point here is killing separate threads and my WorkQueue class
was that was the first example that came to mind. Note that whether or not
you use Queue.Queue makes little difference in solving this problem:

class WorkQueue:
    'Waits (on a background thread) for work to appear on a queue and then
dispatches it to a handler'

    def __init__(self, handler):
        self.handler = handler
        self.keepRunning = 0
        self.isRunning = 0
        self.workQ = Queue.Queue(0)

    def Start(self):
        'Start processing'
        self.keepRunning = 1
        if not self.isRunning:
            threading.Thread(target=self.thread).start()

    def Stop(self):
        'Shutdown'
        self.keepRunning = 0
        self.workQ.put_nowait(0) # Wake up thread
        while self.isRunning:
            time.sleep(0.01)

    def AddWork(self, work):
        self.workQ.put_nowait(work) # Wake up the thread

    def thread(self):
        self.isRunning = 1
        workQ = self.workQ
        handler = self.handler
        try:
            while self.keepRunning:
                work = workQ.get() # Wait for work to appear
                if not self.keepRunning:
                    break # Someone called Stop so quit

                handler(work)
        finally:
            self.isRunning = 0

-Dave




From ws-news at gmx.at  Fri Mar  9 05:38:46 2001
From: ws-news at gmx.at (Werner Schiendl)
Date: Fri, 9 Mar 2001 11:38:46 +0100
Subject: learning C
References: <mailman.983972829.23481.python-list@python.org> <983975751.18833@newsmaster-04.atnet.at> <slrn9adklr.bv9.scarblac@pino.selwerd.nl> <984047775.592522@newsmaster-04.atnet.at> <9896hk$ig8$1@verence.demon.co.uk>
Message-ID: <984134612.781779@newsmaster-04.atnet.at>

> I think you have a different definition of "embedded" from me
> (admittedly it's a broad term). C is still in wide use in the industry I
> work in (which produces small, hard real time systems).

We have a large existing codebase for that systems in C and still a lot of
new stuff is done in C.

Yes, and we have one more proprietary platform that only supports C with
some very limited (and of course non-standard) libraries.
We ask for C++ support every other week ;-)

> This is unlikely
> to change in the near future: C++ is perceived as being slow and unsafe.

Why should C++ be slower than C if you restrict yourself to features that
perform well.
If you do dynamic_casts everywhere, pass huge objects by value it gets slow.
But this is simply poor design.

I admit that it is far less obvious in C++ than in C than something will
become slow.
But sometimes it is really neat to be able to copy objects in a defined
manner (without having to invoke a helper function for that stuff).

Unsafe is somewhat the same topic I think: It's a question of disciplined
use.
C++ is not a toy language like VB may be for some guys - you need to know
what you are doing.

> There's the feeling that you never quite know what it's going to do.  By
> contrast, C compilers are pretty efficient and what they'll end up
> producing is more or less in line with what you'd expect from looking at
> the code. C itself can be unsafe, of course, but by using restricted
> subsets of the language you can help with this somewhat (banning dynamic
> allocation is common, for example).

You can have the same in C++ (using one or more static objects, or construct
the objects just once when booting)
This is nothing more difficult than in C.

One of the particular unsafe things in C is error handling, at least for a
larger application.
C++ exceptions at least make sure the error _is_ handled somewhere and does
not cause strange behaviour that is hard to debug.

>
> I don't think these systems are simple. If they were, writing them would
> be a lot easier.
>

This is what I said, and why I think some well designed objects can ease the
stuff.

> >I personally use embedded platforms and C++ is readily available there
> >(VxWorks, to be precise).
>
> ObFourYorkshiremen: You had an OS? You were lucky! In my day we had some
> interrupts and timers, and we liked it! (Actually, my current project
> does have an OS: such luxury :-)

If there is no C++ compiler you cannot use it, nothing to add...

regards
werner





From Mike.mellor at tbe.com  Tue Mar  6 15:22:49 2001
From: Mike.mellor at tbe.com (Mellor, Mike)
Date: Tue, 6 Mar 2001 14:22:49 -0600 
Subject: FW: Need help with Tkinter bind statement
Message-ID: <A978D5EFFD32D211B00600A0C9B6B8890193B657@exhsv02.tbe.com>


-----Original Message-----
From: Mellor, Mike 
Sent: Tuesday, March 06, 2001 2:41 PM
To: python-list at yahoogroups.com
Subject: Need help with Tkinter bind statement


Hello - 
I have been working on an RPN calculator program for a while now, and 
I am having problems with some bindings.  Here is a code snippet of 
where I'm having trouble:

# Bindings
xStack.bind('<+>', lambda event, fxn = 'y+x':Calc(fxn))
xStack.bind('<->', lambda event, fxn = 'y-x':Calc(fxn)) # This is 
causing an error
xStack.bind('<*>', lambda event, fxn = 'y*x':Calc(fxn))
xStack.bind('</>', lambda event, fxn = 'y/x':Calc(fxn))

where xStack is an entry widget and Calc is a function that does the 
calculation.  

My 1st problem is that binding to the "-" key generates this error:

Traceback (most recent call last):
  File "D:\Python21\Projects\MCalc7.py", line 98, in ?
    xStack.bind('<->', lambda event, fxn = 'y-x':Calc(fxn)) # This is 
causing an error
  File "d:\python21\lib\lib-tk\Tkinter.py", line 899, in bind
    return self._bind(('bind', self._w), sequence, func, add)
  File "d:\python21\lib\lib-tk\Tkinter.py", line 854, in _bind
    self.tk.call(what + (sequence, cmd))
TclError: no event type or button # or keysym

but the other bindings work fine.  The second problem is that when I 
use any of these keys, the keystroke shows up in the entry widget.  
What I want to happen is that pressing the "+" key, for example, 
should add the values in the xStack and yStack (which it does) and 
then put the result in the yStack entry widget (which it also does).  
The problem is that the "+" is left in the xStack.  The contents of 
the xStack are deleted by the Calc function, so I thought it would 
also delete the "+" as well.  

Thanks.

Mike







From gem at hsv.crc.com  Wed Mar  7 08:23:24 2001
From: gem at hsv.crc.com (Glen Mettler)
Date: Wed, 7 Mar 2001 07:23:24 -0600
Subject: Can Python 1.5.2 code run on Python 2.x?
Message-ID: <985c9m$3gg$1@hobbes2.crc.com>

Can code written in Python 1.5.2 run without modification on Python 2.x?
If not, what modification in code is required?

Glen





From corour01 at motorola.com  Wed Mar 21 04:08:40 2001
From: corour01 at motorola.com (O'Rourke Clodagh-corour01)
Date: Wed, 21 Mar 2001 09:08:40 -0000
Subject: Tkinter Problem - NameError
Message-ID: <13EE655665F4D311AB4D0008C789498A01EE2C8A@zei02exm02.cork.cig.mot.com>

> Hi all,
> 
I am creating a multi-windowed GUI system to put a front end on GSM network Environment Variables.  I have created a class to represent each window.  So there are these four windows classes.  AuditResync.py*   GuiSelection.py*  System.py* Cell.py*  InitialWin.py*   EnvVar.py*.  Alll of them except EnvVar just contain button boxes of selections to navigate through the application.  That is the application starts with InitialWin.py and the user selects a group of variables and a new window opens and the old one is destroyed.  

If I run python AuditResync.py the window appears perfectly.  On selection of the button which launchs an EnvVar window it runs perfectly again.  
If I run python InitialWin.py it window appears fine and on selection of the AuditResync button the same window as we displayed earlier appears fine too.  The problems arises on selection of the button which launchs an EnvVar window.  

Here I get the following NameError:

Exception in Tkinter callback
Traceback (innermost last):
  File "/usr/vob/omc_reference/ThirdParty/scriptSources/lib/python1.5/lib-tk/Tkinter.py", line 764, in __call__
    return apply(self.func, args)
  File "AuditResync.py", line 64, in newAudit
    self.newEnvWin('audit')
  File "AuditResync.py", line 61, in newEnvWin
    EnvVar(g, self.parent)
NameError: EnvVar

Even though I was able to open an EnvVar Win from teh AuditResync window no problem earlier I could not understand why it would not open through the AuditResync window opened through the Initial win.
The same error happens a lot with all the classes.

> I would be very grateful if anyone could spot any possible errors.
> 
> Thanks a lot,
> Clodagh O' Rourke
> 



From aleaxit at yahoo.com  Thu Mar 22 02:54:42 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 22 Mar 2001 08:54:42 +0100
Subject: Good Editor?
References: <3ab9a32b$1_1@news2.one.net>
Message-ID: <99cb42013df@news2.newsguy.com>

"fjs" <fjs6 at usa.net> wrote in message news:3ab9a32b$1_1 at news2.one.net...
> Can someone recomend a good editor to write python in?  Im just getting
> started learning python and dont know which ones support python.

I mostly use VIM (www.vim.org), which among other things
supports the use of Python as a macro-language.  However,
VIM is a development of vi -- some of us love its modal
basic operations (it minimizes fingermovement wrt any
alternative -- highly ergonomic), some other detest it
(I think it's because they try to grasp it with their
brains rather than just let their _fingers_ learn it).

A good alternative is provided by integrated development
systems for Python, such as IDLE, which comes with the
standard Python distribution, and others which are either
platform-specific or cost some money.

Back to general-purpose editors, Emacs lovers find their
favourite perfectly good for Pythoning (after all, if it's
good for reading mail and news, managing files, brewing
coffee and bringing peace to the Middle East, how can it
be bad for editing whatever kind of files, right?).  Most
of my colleagues currently seem to swear by 'UltraEdit',
but I haven't tried it myself, nor do I know how well it
supports Python specifically.


Alex





From leo at iems.nwu.edu  Wed Mar  7 19:55:34 2001
From: leo at iems.nwu.edu (Leonardo B Lopes)
Date: Wed, 07 Mar 2001 18:55:34 -0600
Subject: pipe problems
Message-ID: <3AA6D886.4DD8D3A4@iems.nwu.edu>

I use a program that doesn't have readline compiled in. So I figured, it
would be easy to add a python wrapper around it. So I wrote the wrapper,
using 'cat' as a test command. With 'cat' it works fine. But when I try
using another program, such as 'python', the program just hangs at
readline(). Anybody know how to accomplish this? Here is the code:

#!/usr/local/bin/python

import sys,os
import cmd

class Amplcmd(cmd.Cmd):

        def __init__(self):
                cmd.Cmd.__init__(self)
                self.cin,self.cout = os.popen2('/usr/local/bin/python')

        def default(self,line):
                self.cin.write(line+'\n')
                self.cin.flush()
                self.cout.flush()
                print self.cout.readline(),

        def do_EOF(self,line):
                sys.exit()

mycmd = Amplcmd() 
mycmd.cmdloop()



-- 
=======================================================================
Leonardo B. Lopes                                      leo at iems.nwu.edu 
Ph.D. Student                                             (847)491-8470
IEMS - Northwestern University             http://www.iems.nwu.edu/~leo


From owen at astrono.junkwashington.emu  Tue Mar 20 13:29:06 2001
From: owen at astrono.junkwashington.emu (Russell E. Owen)
Date: Tue, 20 Mar 2001 10:29:06 -0800
Subject: Tkinter : resizing Text widget with main window
References: <3AB11827.469F6C0C@mega-nerd.com> <slrn9bdls6.noo.rickp@tc.niof.net> <3AB70C35.1B02410E@mega-nerd.com>
Message-ID: <9987hk$nkq$1@nntp6.u.washington.edu>

In article <3AB70C35.1B02410E at mega-nerd.com>,
 Erik de Castro Lopo <nospam at mega-nerd.com> wrote:

>Rick Pasotto wrote:
>> 
>> On Thu, 15 Mar 2001 19:38:12 GMT in comp.lang.python, Erik de Castro Lopo 
>> wrote:
>> > Hi all,
>> >
>> > How does one force a Tkinter.Text widget to be resized whenever
>> > the main window is resized?
>> 
>> t=Text(width=30,height=12) # sets the initial size
>> t.pack(expand=1,fill='both')
>
>That may work for widgets which are pack()ed, but what about
>widgets which are place in the parent window using grid ()?

For grid use

t.grid(sticky="news")

(Note: if you have multiple gridded widgets then you may also want to 
use the "weight" option of grid, to apportion how the various cells of 
the grid resize. This would be in addition to "sticky", not instead of 
it).

-- Russell


From grante at visi.com  Thu Mar  8 15:18:43 2001
From: grante at visi.com (Grant Edwards)
Date: Thu, 08 Mar 2001 20:18:43 GMT
Subject: Get terminal size in perl and python
References: <20010308194028.A1616@bork.demon.nl> <mailman.984079092.10641.python-list@python.org>
Message-ID: <DORp6.25$Tg.3395@ruti.visi.com>

In article <mailman.984079092.10641.python-list at python.org>, Kalle Svensson wrote:
>Sez Egbert Bouwman:
>> My perl correspondent uses:
>>     use Term::ReadKey;
>>     ($cols,$lines)=GetTerminalSize()
>> What is the python equivalent ?
>
>I'd use os.environ["COLUMNS"] and os.environ["LINES"], but I've no idea
>about how portable they are. 

In most Unixes you'd be better off using the ioctl() call (the
details of which I can't remember at the moment -- I thought it
was in the termios struct, but I can't find it).  The
environment variables are frozen at the time your program is
invoked and won't reflect any window resizing done since then.
If you start with the os.environ values you should then catch
SIGWINCH signals that are sent when the window is resized.

-- 
Grant Edwards                   grante             Yow!  My LIBRARY CARD
                                  at               expired...
                               visi.com            


From danielk at aracnet.com  Tue Mar  6 09:46:15 2001
From: danielk at aracnet.com (Daniel Klein)
Date: Tue, 06 Mar 2001 06:46:15 -0800
Subject: Static class methods
References: <rdzo6.154$gc4.80445@news.pacbell.net> <aqa7at0b6cbqjaq23mli0flmumlp8n6s9s@4ax.com> <7SZo6.7063$qv3.3125634@nnrp5-w.sbc.net> <qbr9atklivqc9mphck04suggdsbeubopqi@4ax.com>
Message-ID: <q0u9at847ajcbulqethlhr32p086sfbfki@4ax.com>

Oops, I meant TQPB. :-(

On Tue, 06 Mar 2001 06:10:30 -0800, Daniel Klein <danielk at aracnet.com> wrote:

>On Mon, 5 Mar 2001 20:40:29 -0800, "Alex Shindich" <alex at shindich.com> wrote:
>
>>There are indeed many ways to simulate the behavior. But shouldn't such a
>>strong OOP language like Python have support for static methods?
>
>It's not really necessary  to be honest. The approach outlined in TQJB was just
>a 'suggestion' cos you seemed 'distressed' you couldn't define a class method
>associated to a specific class.  When I need such a method I simply define it
>as a 'top level' function in the module and call if from the class instances.
>The fact that it isn't 'bound' to a particular class does not bother me. That's
>what doc strings are for. :-) On occasion there are situations where one of
>these 'top level' methods becomes useful to more than one class in the module
>and then I reap the benefits of reuse.
>
>Daniel Klein
>Portland OR USA



From aleaxit at yahoo.com  Thu Mar 29 09:06:03 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 29 Mar 2001 16:06:03 +0200
Subject: boolean values in COM / VB / Python
References: <Pine.SOL.4.30.0103252251210.18627-100000@mimosa.csv.warwick.ac.uk>
Message-ID: <99vfhj02f6f@news2.newsguy.com>

"John J. Lee" <phrxy at csv.warwick.ac.uk> wrote in message
news:Pine.SOL.4.30.0103252251210.18627-100000 at mimosa.csv.warwick.ac.uk...
>
> ie.Document.execCommand("SaveAs", false, pathname)
>
> where ie is an Internet Explorer object, pops up a dialogue box, no matter
> what value for false I try, from (1==0, 0, "0", None). According to the MS
> docs at

Yep, the docs do say what you say they say.  But, do they say truth?
Try this small VBScript script...:


WScript.Echo "Start"

Set ie = WScript.CreateObject("InternetExplorer.Application")
WScript.Echo "ie obj created"

ie.Navigate "file://D:\pybridge\bah.htm"
WScript.Echo "Navigated ok"

Set doc = ie.Document
WScript.Echo "Doc gotten"

doc.execCommand "SaveAs", 0, "D:\pybridge\foo.htm"
WScript.Echo "Saved!"


On my box, it hangs just after writing "Doc gotten"... with
some effort (minimize the dosbox and restore it) you get to
see why -- a modal "save file" dialog is being displayed --
you have to clic save or cancel to exit it.

So, apparently, it's not about how to represent Boolean
values in COM/VB/Python -- it's about this .execCommand
not following its documentation/specifications.


After a few years working with MS tools, one learns that
experimental verification of their docs' assertions with
a few different tools is a good idea (...if one hasn't
already learned that previously about ALL docs...:-).


Alex





From cdwom at mpinet.net  Wed Mar 28 19:53:33 2001
From: cdwom at mpinet.net (Corey Woodworth)
Date: Thu, 29 Mar 2001 00:53:33 GMT
Subject: Recommendations for Python Books
References: <3ac2582c$1@news.starhub.net.sg>
Message-ID: <hIvw6.10727$ka1.879919@news2.aus1.giganews.com>

"Wong Da Qiang" <noen at none.com> wrote in message
news:3ac2582c$1 at news.starhub.net.sg...
> Just a newbie to programming and python, need some recommendations on
books
> to buy

I'm a newbie too, and I've been reading Sam's Teach Yourself Python in 24
hours. Its very well written. I like it tons.


--
So long, and thanks for all the fish!
Corey

Check out my Webpage at: schitzomedia.y0ru.net
and vote for the top 10 games of all time!!!




From paulp at ActiveState.com  Thu Mar 15 11:49:15 2001
From: paulp at ActiveState.com (Paul Prescod)
Date: Thu, 15 Mar 2001 08:49:15 -0800
Subject: Raise exception in caller's stack frame
References: <3AB0BF02.BA0DB5E8@ActiveState.com>
Message-ID: <3AB0F28B.691B5DCC@ActiveState.com>

Paul Prescod wrote:
> 
> Can someone remind me how to raise an exception in my caller's stack
> frame? I'll add it as a recipe to the Python cookbook (or you can do it
> if you want the fame and...well, just fame).

I don't know if this made sense to anyone after I posted it in the
middle of the night. What I want to do is throw a exception but not have
*my* function body in the stack trace:

def throwException(condition):
    if condition:
        raise AssertionError in parent_stackframe

def foo():
    throwException(i==1)

I'm thinking that the trick has something to do with setting up a trace
function or maybe it just isn't possible. I've done similar tricks
before but perhaps not this particular one...

-- 
Take a recipe. Leave a recipe.  
Python Cookbook!  http://www.activestate.com/pythoncookbook



From kalle at gnupung.net  Wed Mar  7 19:58:25 2001
From: kalle at gnupung.net (Kalle Svensson)
Date: Thu, 8 Mar 2001 01:58:25 +0100
Subject: learning C
In-Reply-To: <slrn9adklr.bv9.scarblac@pino.selwerd.nl>; from scarblac@pino.selwerd.nl on Thu, Mar 08, 2001 at 12:32:47AM +0000
References: <mailman.983972829.23481.python-list@python.org> <983975751.18833@newsmaster-04.atnet.at> <slrn9adklr.bv9.scarblac@pino.selwerd.nl>
Message-ID: <20010308015825.A1237@apone.network.loc>

A non-text attachment was scrubbed...
Name: msg.pgp
Type: application/pgp
Size: 712 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/20010308/e394f91d/attachment.bin>

From aleaxit at yahoo.com  Wed Mar 28 11:16:01 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Wed, 28 Mar 2001 18:16:01 +0200
Subject: How does one set a value via win32com
References: <mailman.985764036.26774.python-list@python.org>
Message-ID: <99t2o002sgd@news2.newsguy.com>

"Stefan Migowsky" <smigowsky at dspace.de> wrote in message
news:mailman.985764036.26774.python-list at python.org...

"""
you ran on a problem with dynamic dispatch. It can't use properties
with parameters. In a makepy generated wrapper properties with parameters
are translated in methods. So you must use a wrapper to use such parameters.
To ensure a wrapper use the gencache module, in your case
"""

Yes, but... that's what I thought I'd advise, too, but then I
tried out a simple (?) case and just can't make it work.  I.e.:

D:\Python20\Lib>python
Python 2.0 (#8, Oct 16 2000, 17:27:58) [MSC 32 bit (Intel)] on win32
Type "copyright", "credits" or "license" for more information.
>>> from win32com.client.gencache import EnsureDispatch as disp
>>> dic=disp("Scripting.Dictionary")
>>> dic
<win32com.gen_py.Microsoft Scripting Runtime.IDictionary>
>>> dic.Item("foo","bar")
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
TypeError: too many arguments; expected 2, got 3
>>>

So how does one do the Python equivalent of VBS's
    dic.Item("foo")="bar"
i.e., set an indexed property?  Scripting.Dictionary
is nice in ALSO making a method (SetItem) available
duplicating the indexed-property put, but not all
COM objects with indexed properties will be as nice
as that -- how do I put the indexed property directly
then?  I thought I knew, but, it turns out I was wrong,
and I'm now wondering how (or whether) it's done...


Alex





From inv3rt3d0-nospam at centras.lt  Mon Mar  5 09:17:35 2001
From: inv3rt3d0-nospam at centras.lt (inverted zero)
Date: Mon, 5 Mar 2001 16:17:35 +0200
Subject: Open Source
References: <9806i4$2e97$1@news3.infoave.net>
Message-ID: <98072s$qk26m$1@ID-51035.news.dfncis.de>

Alton Luder III wrote:

> Is there any place on the web that offers open source python programs. It
> might be helpful for me to learn if I see other people's program's source
> code.
> Thanks in advance,
> Alton Luder III
> 
> 


hello,

http://www.vex.net/parnassus
 is a good  place to start.

regards, 

inv3rt3d0


From flognat at flognat.myip.org  Tue Mar 13 15:17:43 2001
From: flognat at flognat.myip.org (Andrew Markebo)
Date: Tue, 13 Mar 2001 20:17:43 GMT
Subject: Editing Python source as XML?
References: <mailman.983546704.10789.python-list@python.org>
Message-ID: <m3puflpgqa.fsf@master.athome>

You don't need to go xml, there are a mode for emacs, hideshow, that
can fold in and fold out depending on stuff like indentation, brackets
and so on.

You don't have a beautiful tree to navigate in, but it is quite
useful.

        /Andy

/ Steven Haryanto <steven at haryan.to> wrote:
| [...]
| What I would like is an editor that presents the source file as a
| tree outline, where I can collapse selected class or function
| definition body, certain long if's/for's, some section of code,
| view/hide the asserts, etc. So then I only have to deal with the
| section I am concentrating with and don't have look at other
| distracting parts (especially in C with all those #ifdef/#endif's,
| argh...).
| [...]


From fredrik at pythonware.com  Fri Mar 16 11:35:04 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 16 Mar 2001 16:35:04 GMT
Subject: How can install Tkinter on linux.
References: <98rnem$465$2@news2-2.kornet.net>
Message-ID: <Ygrs6.169$sk3.50068@newsb.telia.net>

"rambo" (now that's one lame alias) <rambo at linuxsecurity.co.kr> wrote:
> How can I install Tkinter on linux.
> I download Tkinter-2.0-8.0.tar.gz, and extract it ( tar xvzf
> Tkinter-2.0-8.0.tar.gz  )
>
>     Tkinter-2.0-8.0/
>     Tkinter-2.0-8.0/src/
>     Tkinter-2.0-8.0/src/_tkinter.c
>     Tkinter-2.0-8.0/src/tkappinit.c
>     Tkinter-2.0-8.0/setup.cfg
>     Tkinter-2.0-8.0/setup.py
>
>  to use Tkinter on linux, What should I do?
> and how to execute setup.py on linux prompt?

$ python setup.py

Cheers /F




From sheila at spamcop.net  Mon Mar 19 01:41:04 2001
From: sheila at spamcop.net (Sheila King)
Date: Mon, 19 Mar 2001 06:41:04 GMT
Subject: Problems with 2.0 cgi scripts
Message-ID: <fs9bbtg68qne71t853lp89l44uarbb5jk5@4ax.com>

My webhost has Python 1.5.1 installed as a core tool. However, I never could
get the smtplib module commands to work properly.

So, I installed Python-2.0 in my own web space. It has been fine for command
line tasks and I have also used it as a script for filtering e-mail invoked by
qmail. Great.

However, now I try to run some cgi scripts with it. Well, I do have a
functioning cgi script on my site with Python-2.0:
http://www.thinkspot.net/cgi-bin/pytestscript.py
(It's just "Hello, World")

So, I know I can run Python-2.0 cgi scripts.

Now comes the problem:
If I try to run a Python-2.0 script that needs to import a module, I get
internal server errors, and the log file reports something like this:

subscriberEmail=&command=%3CSelect+an+Item%3E&Submit=Submit+Changes
%response
%stderr
Could not find platform independent libraries <prefix>
Could not find platform dependent libraries <exec_prefix>
Consider setting $PYTHONHOME to <prefix>[:<exec_prefix>]
'import site' failed; traceback:
ImportError: No module named site
Python 2.0 (#1, Jan 15 2001, 01:09:04) 
[GCC 2.7.2.3] on linux2
Type "copyright", "credits" or "license" for more information.
Traceback (most recent call last):
  File "listsettings.py", line 5, in ?
    import cgi
ImportError: No module named cgi
# clear __builtin__._
# clear sys.path
# clear sys.argv

...(rest snipped)...

Yet, the same script will run from the command line during a telnet session
and find all the files.

So, basically, it cannot find the modules when I run it as a cgi-script. I
have all the paths set correctly, and the permissions set correctly. Grr, I'm
frustrated with this one by now...

Does anyone have any idea what could be causing behavior of this type?
(It isn't the script, because the script runs fine on my home machine under
Apache, and it runs fine from my web host at the command line. Unfortunately,
since this is a custom install, it is classified as a 3rd party script as far
as support goes: I'm on my own. :-(

If you want to look at the script itself, it is here:
http://www.thinkspot.net/test/listsettings.py

--
Sheila King
http://www.thinkspot.net/sheila/
http://www.k12groups.org/



From mlh at idi.ntnu.no  Fri Mar  2 16:34:37 2001
From: mlh at idi.ntnu.no (Magnus Lie Hetland)
Date: Fri, 2 Mar 2001 22:34:37 +0100
Subject: gah! I hate the new string syntax
References: <mailman.983551204.27573.python-list@python.org>
Message-ID: <97p3sa$bpq$1@tyfon.itea.ntnu.no>

"Sean 'Shaleh' Perry" <shaleh at valinux.com> wrote in message
news:mailman.983551204.27573.python-list at python.org...
[...]
> The point of all of this was 'join' especially,
> but a few of the others (it
> brother split) make for really ugly code.
>
> " ".split(list) is another I have seen recently.
>

What does *this* mean, then? (AFAIK it isn't valid
Python unless list is a character...)

--

  Magnus Lie Hetland          http://www.hetland.org

 "Reality is what refuses to disappear when you stop
  believing in it"                 -- Philip K. Dick





From ejr at cs.berkeley.edu  Tue Mar 13 18:30:40 2001
From: ejr at cs.berkeley.edu (Edward Jason Riedy)
Date: 13 Mar 2001 23:30:40 GMT
Subject: This math scares me
References: <3aad4062.1290021@News.CIS.DFN.DE> <jwbnews-7CC03F.19050312032001@corp.supernews.com> <yu993dcimjt8.fsf@europa.research.att.com> <98m5s0$gog$1@agate.berkeley.edu>
Message-ID: <98maj0$hk4$1@agate.berkeley.edu>

And I babble:
 - 
 - The next revision of IEEE 754 will require exact conversion, as
 - well.

Andrew Koenig graciously pointed out that my statement isn't quite 
right...

The actual changes are that all input and output conversions must 
be correctly rounded.  The current standard gives ranges where they
must be rounded correctly, and the changes extend that to the entire 
required conversion range.  I don't recall if the conversion range 
itself will be extended.  By and large, systems already fit this.

Jason, who realizes that if exact conversion were always possible, 
this thread wouldn't exist...
-- 


From cfelling at iae.nl  Sat Mar 10 11:33:27 2001
From: cfelling at iae.nl (Carel Fellinger)
Date: 10 Mar 2001 17:33:27 +0100
Subject: How to efficently build a nested dictionary
References: <3AAA2042.E69A53FC@gmx.de>
Message-ID: <98dl0n$30p$1@animus.fel.iae.nl>

Carsten Gaebler <cgaebler at gmx.de> wrote:
> Hi there!

> I wrote a script for FTP traffic accounting, i.e. for each "top level"
> FTP directory it counts the bytes transferred per day. The traffic is
> stored in a dictionary that has the following structure:

> traffic = {dir: {year: {month: {day: 42}}}}

Wouldn't it be more efficient if you would build it like this:

   traffic[dir,year,month,day] = traffic.get((dir,year,month,day), 0) + bytes

or use newer Python goodies like:

   m = traffic.setdefault(dir,{}).setdefault(year,{}).setdefault(month,{})
   m[day] = m[day] + bytes

or use intermediates like:

   if traffic.has_key(dir):
      d = traffic[dir]
   else:
      d = traffic{dir] = {}
   if d.has_key(year):
      y = d[year]
   else:
      y = d[year] = {}
   if y.has_key(month):
      m = y[month]
   else:
      m = y[month] = {}
   m[day] = m[day] + bytes

or use the sometimes more efficient try-except:

   try:
      d = traffic[dir]
   except:
      d = traffic[dir] = {}

   try:
      y = d[year]
   except:
      y = d[year] = {}

   try:
      m = y[month]
   except:
      m = y[month] = {}
   
   m[day] = m[day] + bytes
   
-- 
groetjes, carel
-- 
groetjes, carel


From walter.pleyer at euromatketing.at  Thu Mar  1 05:44:45 2001
From: walter.pleyer at euromatketing.at (Walter Pleyer)
Date: Thu, 01 Mar 2001 11:44:45 +0100
Subject: PIL setup troubles
Message-ID: <3A9E281C.33A6C693@euromatketing.at>

I'm using  Zope and I want to use TextImage. But somehow TextImage can't
find PIL.
I always get the error message:
Traceback (innermost last):
  File "/var/zope/lib/python/OFS/Application.py", line 397, in
import_products
    product=__import__(pname, global_dict, global_dict, silly)
  File "/var/zope/lib/python/Products/TextImage/__init__.py", line 5, in
?
    import TextImage
  File "/var/zope/lib/python/Products/TextImage/TextImage.py", line 12,
in ?
    import PIL.Image
ImportError: No module named PIL.Image


Does anyone know this problem, can anyone help me out here?
Any infos welcome,
Thanx
Walter



From robin at jessikat.fsnet.co.uk  Sun Mar 11 05:24:06 2001
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Sun, 11 Mar 2001 10:24:06 +0000
Subject: loading Python20.dll during run-time under Windows
References: <slrn9alouc.sor.zeitlin@seth.lpthe.jussieu.fr>
Message-ID: <bhG3$JAGJ1q6Ewxl@jessikat.fsnet.co.uk>

In article <slrn9alouc.sor.zeitlin at seth.lpthe.jussieu.fr>, Vadim Zeitlin
<zeitlin at seth.lpthe.jussieu.fr> writes
> Hello,
>
> I have an application which can use Python for scripting but it is not
>essential to its work, i.e. it works mostly fine without Python. So I'd
>like to be able to install it even on systems lacking Python installation
>but for this I need to [try to] load Python.dll during run-time instead
>of linking to it directly.
>
> Has anyone done this before and is there any (C) code I could reuse?
>Normally, there is no problem with doing what I want (especially
>after realizing that I need to have USE_DL_EXPORT defined...) but
>the perspective of having to manually write all the code needed to load
>the functions from the DLL is not very, hmm, encouraging just because there
>are quite a few of functions and so I'm looking for a slightly less painful
>way to do it.
>
> Thanks in advance for any help!
>VZ
>
Gordon McMillan's Installer software has cpp code that sets up and
launches python.

In essence the code does

1) set up a PYTHONPATH environment variable and the dllpath (where the
python dll is).
2) dll = LoadLibraryEx(dllpath, NULL, LOAD_WITH_ALTERED_SEARCH_PATH );
3) get the addresses of various key python variables and procedures from
the dll
4) Initialise many of the key variables and the python interpreter.
5) Run a predefined script.

If that sounds the right sort of thing then get the code from Installer-
3j/source at http://www.mcmillan-inc.com/builder.html


-- 
Robin Becker


From bos at hack.org  Tue Mar 27 10:55:18 2001
From: bos at hack.org (Rikard Bosnjakovic)
Date: Tue, 27 Mar 2001 17:55:18 +0200
Subject: Combining Lists
References: <99q887$hl9$1@ctb-nnrp2.saix.net>
Message-ID: <3AC0B7E6.FDC467C2@hack.org>

Mix wrote:

> Now how  does one get the following list
>  list3=['1','2','3','9','15']
> from a combing the two lists (i.e list1 and list2)?

<URL:http://www.python.org/doc/current/tut/node5.html>, read under
3.4.1.


-- 
Rikard Bosnjakovic - http://a214.ryd.student.liu.se/cv/ - ICQ: 1158217

Anyone sending unwanted advertising e-mail to my address will be
charged $250 for network traffic and computing time. By extracting my
address from this message or its header, you agree to these terms.


From sholden at holdenweb.com  Thu Mar 29 08:22:21 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 29 Mar 2001 08:22:21 -0500
Subject: Defining class files
References: <3ac33343$0$12248$ed9e5944@reading.news.pipex.net>
Message-ID: <WJGw6.20925$m04.864460@e420r-atl1.usenetserver.com>

"Neil Benn" <neil.benn at cambridgeantibody.com> wrote in message
news:3ac33343$0$12248$ed9e5944 at reading.news.pipex.net...
> Hello,
>
>             Sorry to ask such a basic question but -
>
> I'm wokring through the tutorial and have reached the section on classes
in
> Python.  Working through the example on classes I have entered the class
> using the interpreter:-
>
> class MyClass:
>     "A simple example class"
>     i = 12345
>     def f(x):
>         return 'hello world'
>
>     This worked fine when I typed:-
>
> >>> x=MyClass()
> >>> x.f()
> 'Hello World'
>
>     However, I then tried to write the class as an external text file,
> using:-
>
> class YourClass:
>     "A simple example class"
>     i = 12345
>     def f(x):
>         return 'hello world'
>
>     I then imported the class, tried to assign the class and ivoke a
method
> :-
>
> >>> y = YourClass()
> Traceback (innermost last):
>   File "<console>", line 1,
> TypeError: call of non-func
>
>     This seemed a bit strange, so I investigated the Your class and it
> seemed to be a module.
>
> >>> YourClass
> <module YourClass at 2584541>
>
>     The text file is saved as YourClass.py - is this the problem, should
> class files have different terminaters in their filename??
>
>     Any/all help would be most appreciated!!
>
The confusion here appears to be about namespaces. Since you now have a
YourClass module, which defines a YourClass class, the correct way to
instantiate (create an instance of) the class would be:

    y = YourClass.YourClass()

Here the module name is qualified by the name of the object.

Does this help?

regards
 Steve





From aleaxit at yahoo.com  Thu Mar 29 06:28:38 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 29 Mar 2001 13:28:38 +0200
Subject: Boost documentation.
References: <g0wv99x256.fsf@scumbag.ecs.soton.ac.uk> <g0u24dx087.fsf@scumbag.ecs.soton.ac.uk> <g0r8zhwynh.fsf@scumbag.ecs.soton.ac.uk>
Message-ID: <99v6aa025ou@news2.newsguy.com>

"Jacek Generowicz" <jmg at ecs.soton.ac.uk> wrote in message
news:g0r8zhwynh.fsf at scumbag.ecs.soton.ac.uk...
    [snip]
> OK, i was not initialized to zero here:
>
> >   for ( int i; i<tup.size(); i++ ) {
>
> (But surely this calls the default constructor for int, which should
> initialize it to zero ?)

No: in C++, the 'pseudo' default-constructors of built-in types
(such as int) are only implicitly called in very specific cases
(e.g., static variables -- NOT auto variables, which is what
you have here).  Your C++ compiler was well within its standard
obligations in not initializing i here (it WOULD have been nice
of it to NOTICE it was uninitialized and warn you about it, but
the standard doesn't mandate that warning).


Alex (Brainbench MVP for C++)





From skip at mojam.com  Thu Mar  8 19:29:32 2001
From: skip at mojam.com (Skip Montanaro)
Date: Thu, 8 Mar 2001 18:29:32 -0600 (CST)
Subject: Database with Python ??
In-Reply-To: <988vmn$t77$05$1@news.t-online.com>
References: <988vmn$t77$05$1@news.t-online.com>
Message-ID: <15016.9196.879956.629984@beluga.mojam.com>

    Thomas> Has Python a own database language and do someone know a link to
    Thomas> a Tutorial in database-working with python?

Try this:

    http://www.python.org/topics/database/

-- 
Skip Montanaro (skip at mojam.com)
Support Mojam & Musi-Cal: http://www.musi-cal.com/sponsor.shtml
(847)971-7098



From ngreco at softhome.net  Tue Mar 27 10:30:32 2001
From: ngreco at softhome.net (Nahuel Greco)
Date: Tue, 27 Mar 2001 12:30:32 -0300
Subject: win32: Pipes opened with popen3 cant be closed!
Message-ID: <20010327123032.3fa86c04.ngreco@softhome.net>

Hi, im doing the following thing in Windows 98 / Activestate Python 2.0:

	import win32pipe

	cmd = "c:\\zebedee\\zebedee.exe -d 127.0.0.1"
	pipes = win32pipe.popen3(cmd,'t')

	#Now.. if i do:

	pipes[0].close()
	pipes[1].close()
	pipes[2].close()  # in this line, the program hangs.


If i follow another closing order, the program stills hangs in the last
close() call .. i think that when all the pipes are closed.. win32pipe
try to close the child program.. and wait for that close.. but, the
child program never exit.

If i do not close the pipes, when my python script exits, the child
program is still running (i dont want that).

The zebedee.exe that im using is at  
http://www.winton.org.uk/zebedee/download.html

Zebedee.exe is a program that doesnt exit inmediatly, it wait for
tcp connections, and print the conections made to stderr (using -d
"detach" flag). You can emulate it with:


	import time
	import sys
 
	while 1:
		sys.stderr.write("hello\n")
		sys.stderr.flush()
		time.sleep(1)
 


Im dont use win32 usually.. im lost in the Gates dimension :).


_------------------------------------------------------------.
| Nahuel Greco                 Web Development - Open Source |
| http://www.codelarvs.com.ar  Game Programming - Research   |
| Freelance coding / sysadmin  Networking. The answer is 42. |
'------------------------------------------------------------'





From fellowsd at cs.man.ac.uk  Tue Mar 13 11:34:05 2001
From: fellowsd at cs.man.ac.uk (Donal K. Fellows)
Date: Tue, 13 Mar 2001 16:34:05 +0000
Subject: Einstein's Riddle
References: <oNPq6.396644$ge4.138479566@news2.rdc2.tx.home.com> <UdSq6.2184$54.2166@www.newsranger.com> <984380185.768920@news.aei.ca> <uP%q6.2375$54.2207@www.newsranger.com> <4w6r6.193645$df5.5192565@news1.crdva1.bc.home.com> <3aae4114.63604234@nntp.interaccess.com>
Message-ID: <3AAE4BFD.9A7B0D9C@cs.man.ac.uk>

Thaddeus L Olczyk wrote:
> I also figure the fish thing was simply a typo, and that this guy (
> who hasn't resdponded ) was probably given this as a homework
> assignement, and is simply looking for a cheap way to get it done.

That'd be a very lame homework assignment.  Not unless you had to write
a program in $FAVOURITE_LANGUAGE to compute the answer automatically
(which is much more interesting, especially if you make the program
start from the natural language expression of the problem!  :^)

Donal.
-- 
Donal K. Fellows    http://www.cs.man.ac.uk/~fellowsd/    fellowsd at cs.man.ac.uk
"I wouldn't just call you wrong.  I'd go further and call you an argumentative
 net-kook idiot who can't do his own research before opening his mouth and
 yammering bullshit."                -- Theo de Raadt <deraadt at zeus.theos.com>


From costas at meezon.com  Fri Mar 16 23:40:12 2001
From: costas at meezon.com (Costas Menico)
Date: Sat, 17 Mar 2001 04:40:12 GMT
Subject: Console printing
References: <3ab2c183.4460238@News.CIS.DFN.DE> <98ugf9$3kojv$1@ID-11957.news.dfncis.de>
Message-ID: <3ab2e942.920402@News.CIS.DFN.DE>

"Emile van Sebille" <emile at fenx.com> wrote:

>Invoke pythone with the -u option
>

How do you do this in Pythonwin? There is no place to define the
Python options that I can see. 


From aahz at panix.com  Mon Mar 19 12:53:57 2001
From: aahz at panix.com (Aahz Maruch)
Date: 19 Mar 2001 09:53:57 -0800
Subject: Why "from __future__" stinks; a counter-offer
References: <mailman.984975803.9006.python-list@python.org> <jwbnews-3793B2.08394319032001@corp.supernews.com> <slrn9bcg72.9q8.scarblac@pino.selwerd.nl>
Message-ID: <995h3l$2kp$1@panix6.panix.com>

In article <slrn9bcg72.9q8.scarblac at pino.selwerd.nl>,
Remco Gerlich <scarblac at pino.selwerd.nl> wrote:
>John W. Baxter <jwbnews at scandaroon.com> wrote in comp.lang.python:
>>
>> For the future, the time to complain about PEP contents is when the PEP 
>> is posted, not after it has been posted, discussed, implemented, and the 
>> result distributed.
>
>And the place to complain is the Python-DEV list, which is not public,
>although everyone can post to it and everyone can read it using some
>archives...

Officially, complaints need only be posted to c.l.py.  It's not clear to
me at this point what the True Practice is, but sending e-mail to the
PEP champion is supposed to work best of all.  (Not all PEP creators are
on python-dev.)
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

There's a difference between a person who gets shit zie doesn't deserve
and a person who gets more shit than zie deserves.  --Aahz


From fredrik at pythonware.com  Fri Mar 16 02:02:40 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 16 Mar 2001 07:02:40 GMT
Subject: Python2C
References: <3AAE98EA.2754A358@lindrum.de> <4gyr6.160$4d3.34825@newsb.telia.net> <3AB13BD5.5831BD4A@lindrum.de>
Message-ID: <kUis6.107$sk3.37367@newsb.telia.net>

Michael Lindrum wrote:
> Thanks, in the meantime I got the lib, but it needs libpython.a from Python
> 2.0

Python 2.0 is also available from www.python.org

(you're getting very close to the point where the right answer
is "if you cannot sort this out, python2c might not be the right
tool for you" ;-)

Cheers /F

> Fredrik Lundh schrieb:
>
> > Michael Lindrum wrote:
> > > I'm searching for the lib that is necessary to compile C-Code generated
> > > by Python2C.
> > > OS: SunOS
> > > Who knows where I can find it?
> >
> > assuming you mean libpython, all you have to do is to
> > download and build python itself:
> >
> >     http://www.python.org
> >
> > (if that's not it, consider telling us what library you think
> > you're missing)




From shaleh at valinux.com  Wed Mar 28 12:40:38 2001
From: shaleh at valinux.com (Sean 'Shaleh' Perry)
Date: Wed, 28 Mar 2001 09:40:38 -0800 (PST)
Subject: list vs tuple
In-Reply-To: <000e01c0b7ac$25f9f510$b503a8c0@activestate.ca>
Message-ID: <XFMail.20010328094038.shaleh@valinux.com>

On 28-Mar-2001 Brian Quinlan wrote:
> I'm not sure that this is a reasonable piece of self documentation since
> dir_list can change. It is the tuple ('/path', '/foo/bar', '/baz/bat/files')
> that cannot change. Would you say that x = 5 documents that x cannot change?
> 

dir_list is declared as a tuple, which is as close to a const one can find in
python.



From scarblac at pino.selwerd.nl  Tue Mar  6 02:57:38 2001
From: scarblac at pino.selwerd.nl (Remco Gerlich)
Date: 6 Mar 2001 07:57:38 GMT
Subject: the demise of 'from foo import * and its implications?
References: <3kVBAfATrDo6EwEM@jessikat.fsnet.co.uk> <mailman.983582584.8198.python-list@python.org> <z8dAGJAyqLo6EwlX@jessikat.fsnet.co.uk> <mailman.983746566.27513.python-list@python.org> <GqQAdIAXUuo6EwyY@jessikat.fsnet.co.uk> <mailman.983806269.21732.python-list@python.org> <HoAgdIAWV8o6Ewyd@jessikat.demon.co.uk>
Message-ID: <slrn9a9603.8jm.scarblac@pino.selwerd.nl>

Robin Becker <robin at jessikat.fsnet.co.uk> wrote in comp.lang.python:
> In article <mailman.983806269.21732.python-list at python.org>, Thomas
> Wouters <thomas at xs4all.net> writes
> >
> ....
> >> 
> >> doesn't from foo import * make things hard to locate at the global level
> >> as well? exec/eval/import can always do this surely.
> >
> >The global namespace is a 'last call' namespace. The compiler doesn't (and
> >can't, since you can modify global namespaces from outside the namespace)
> >keep track of global names because it isn't necessary. Neither is true for
> >function namespaces.
> 
> it isn't necessary in function namespaces either it's desirable to help
> the compiler and apparently now required to enforce the constant scope
> requirement.

Consider:

x=1

def whee():
   print x
   from spam import *

Does whee() give an error? Maybe. If there is a x in spam, then x is a
local variable, so 'print x' gives a NameError. If there isn't, then x is
the global, so print x is ok. Unfortunately it is not possible to know this
on time...

Causality loops. Some problems like this probably become even stranger with
nested scopes...

-- 
Remco Gerlich


From aleaxit at yahoo.com  Fri Mar  9 09:53:59 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 9 Mar 2001 15:53:59 +0100
Subject: I come to praise .join, not to bury it...
References: <mailman.983485923.1136.python-list@python.org> <97p7iq$kk2$1@nntp6.u.washington.edu> <97rka002k9@news1.newsguy.com> <mailman.983990532.16307.python-list@python.org>
Message-ID: <98aqqc021pt@news2.newsguy.com>

"D-Man" <dsh8290 at rit.edu> wrote in message
news:mailman.983990532.16307.python-list at python.org...
    [snip]
> | Python does not have multi-methods: it's a single-dispatch
>
> Where can I learn about multi-methods and their
> usefulness/applicability?

A Python implementation by Neel Krishnaswami was announced on
http://mail.python.org/pipermail/python-announce-list/2000-January/000419.ht
ml
and is available among other interesting Python snippets on
his page, http://www.sff.net/people/neelk/open-source/.
(Not sure if it still works with 2.0 or 2.1, but, roughly,
it should).  On the same page you'll find some Dylan pointers,
including one to a paper by Chambers and Chen on how to best
_implement_ multimethods.

If you're a member of the ACM, you can obtain classical papers
from ACM publications on the _use_ of multimethods -- e.g.,
http://dev.acm.org/pubs/citations/proceedings/oops/74877/p239-muller/
which describes the architecture of "Solo", a portable windowing
system written in Common LISP (CLOS) with multimethods -- key idea
from the abstract being "Generic dispatch on multiple arguments, a
feature unique to CLOS, allows a simpler and more flexible input
event dispatching protocol" (it's not "unique to CLOS" any more,
as Cecil, Dylan etc have also adopted it, but it was in 1989 when
the paper appeared).

I'm not aware of good papers freely available with similar
information on "what would we DO with multimethods [if we
had them]" -- any URLs would be appreciated!-)


> Ok, so instead of calling "join" a "string method" how about using the
> following Java-Python hybrid snippet to describe it :
>
> interface Joiner :
>     abstract Object join( Sequence s )
>
> class String implements Joiner :
>     String join( Sequence s ) :

Sure, if Python had 'interface' and 'implements' that IS no
doubt how it would look!  "ya got it".


> This does make sense to me.  It looks logical and reasonable.  You
> have convinced me Alex.  Your English (and writing) classes have paid
> off ;-).  You must have been a writer before computers took over ;-).

Actually, I'm not really old enough to remember the time "before
computers took over"... [actually, I am -- but you won't catch
me dead admitting I'm THAT old:-)].


Alex





From bdupire at seatech.fau.edu  Wed Mar 21 16:55:06 2001
From: bdupire at seatech.fau.edu (Benoit Dupire)
Date: Wed, 21 Mar 2001 16:55:06 -0500
Subject: learning python...
References: <3AB8F4E1.184DD9C@noaa.gov> <Z97u6.16733$Im6.1761313@newsread1.prod.itd.earthlink.net> <3AB91006.6D7E35F2@seatech.fau.edu> <3AB91997.3ECD003C@noaa.gov>
Message-ID: <3AB9233A.ACE63996@seatech.fau.edu>


Benjamin Altman wrote:

> Do you know how it compares to the first edition in terms of presentation and
> indexing?  I have heard from a few people that it is difficult to look up things in
> the 1st edition.  The only review on amazon of the 2nd ed. claims it is a complete
> rewrite despite his admitance that he has never read the book...
>
> Thanks,
> Ben
>

Yes, I also found "programming Python", 1st edition, from Mark Lutz, difficult to
read...
it was the most complete book at the time it was written. I still think it's the most
complete and the more advanced book about Python (I did not see any others.., tell me
if I am wrong).  The 2nd chapter, which was intended to be a 'sneak preview of Python'
or 'a view from 10.000 feet'  was very long, and tedious to understand.. I guess that
10000 feet, it's too high for my poor stomach. There were even backward references to
this chapter from many other chapters... The other pb, you mentionned it, was the
organization... but it was not intended to be a reference book...rather a 'linear
book'.

The second edition seems to be better organized, but once again, Ive just read a small
fraction of it.
Anyway, it is stated in the introduction that it is still not a reference book. The
book presents Python's libraries, tools and programming techniques. Its aim is "how to
use the Core language in applications." So, there are some big examples...

Anyway the book is  intended for those who already know the Core Language and want to
see some applications using it...
the book was refocused, since the 1st edition, and it is not aimed at beginners (it's
why you have to read the 'leaning Python' book first).
I was therefore happy to see that 'The view from 10000 feet' was removed, and chapter 2

is now fun to read, because it presents the libraries, so it's not so steep,....so,
good work, Mark!
The book is going to take me a few weeks to read, though, as it is 1200 pages long, but

seems very promising.. and yes, it was completely rewritten...







From jmarshal at mathworks.com  Fri Mar 23 09:54:33 2001
From: jmarshal at mathworks.com (Joshua Marshall)
Date: 23 Mar 2001 14:54:33 GMT
Subject: newbe   how to read in a string?
References: <3ABB4E70.54885C9F@noos.fr>
Message-ID: <99fo39$gsq$2@news.mathworks.com>

William Famy <william.famy at noos.fr> wrote:
> Hi.

> I am looking how to read in a string.

> Ex
> a='2000/12/28,24.3,87.1,65.4'

> I whant to extract the 87.1. In C i would code by a
> scanf("%s,%f,%f,%f\n"),...

> How can i do this in python?

I don't think there's any generic way.  It would be nice to see
something similar to scanf in Python.


From parkw at better.net  Sun Mar 11 16:45:09 2001
From: parkw at better.net (William Park)
Date: Sun, 11 Mar 2001 16:45:09 -0500
Subject: Chinese text GIF file generation
In-Reply-To: <98gr4i+r42i@eGroups.com>; from hungjunglu@yahoo.com on Sun, Mar 11, 2001 at 09:36:18PM -0000
References: <98gr4i+r42i@eGroups.com>
Message-ID: <20010311164509.A193@better.net>

On Sun, Mar 11, 2001 at 09:36:18PM -0000, hungjunglu at yahoo.com wrote:
> Is there anything in Python (or C library) that take as input a text 
> file in Chinese (BIG5/GB) and generates a GIF file? (Preferable in 
> Windows.) I would imagine someone already wrote something for this in 
> C...
> 
> If there is something similar for Vietnamese, I'd like to know, too.

Because there is so many variables (ie. font, style, size, ...), I don't
think there is package that does what you want.  However, you can use
LaTeX to produce Postscript format, and then convert it into GIF.

---William Park, Open Geometry Consulting, Linux/Python, 8 CPUs.



From mkc at mathdogs.com  Mon Mar 26 17:37:21 2001
From: mkc at mathdogs.com (Mike Coleman)
Date: 26 Mar 2001 16:37:21 -0600
Subject: fixing map(None, [1,2,3]) ?
Message-ID: <873dc040ny.fsf@mathdogs.com>

Shouldn't "map(None, [1,2,3])" return "[(1,), (2,), (3,)]"?

Is there any hope that this could be fixed?

--Mike

-- 
Mike Coleman, mkc at mathdogs.com
  http://www.mathdogs.com -- problem solving, expert software development


From rcameszREMOVETHIS at dds.removethistoo.nl  Sat Mar 31 17:29:15 2001
From: rcameszREMOVETHIS at dds.removethistoo.nl (Robert Amesz)
Date: Sat, 31 Mar 2001 22:29:15 GMT
Subject: Converting an integer base 10 to a binary number
References: <3AC1E898.D5929C44@fh-hof.de> <1er3z6w.13vo4oy1rq8rjmN%francois.granger@free.fr>
Message-ID: <Xns90763FD65212rcamesz@127.0.0.1>

Fran?ois Granger wrote:

>Peter St?hr <peter.stoehr at fh-hof.de> wrote:
>
>> Hi out there,
>> 
>> I know the int(x[,radix]) function that converts a string or a
>> number (for a give radix) to an integer. Is there a built-in
>> function that provides a conversion the other way round?
>> Something like
>>     bin(32) = "10000"
>
>Not fully polished, but works
>
># python
>"""
>create the binary rep of an integer as a string of '0' & '1'
>20010302    v 1.0
>"""
>def toBin(x):
>    a = ''
>    y = x
>    while x:
>        x, y = x / 2, y % 2
>        if y:
>            a = '1' + a
>        else:
>            a = '0' + a
>        y = x
>    return a
>
>if __name__ == '__main__'
>    l = [1, 2, 3, 4, 8, 16, 32, 64, 128, 255, 512, 1024, 2048, 4096,
>8192, 16384, 32768]
>    for x in l:
>        print toBin(x)
>


I'm afraid it doesn't work for x == 0, and you'll get into trouble when 
x is a float. (Strong typing does have its merits.) Also, for binary 
numbers we don't need the div-and-mod approach, a bitwise and (i.e. &) 
will do. (Incidentally, although I'm not using it here, there's a 
function called divmod() which is more efficient, presumably, than two 
seperate operations.) That leaves us with:


def toBin(x):
    x = int(x) # Just in case x is a float, 
    binstr = ''

    while 1: # Can't use while x, because x may be 0 initially
        if x & 1:
            binstr = '1' + binstr
        else:
            binstr = '0' + binstr
        x = x / 2
        if not x:
            return binstr


You can be even more concise (I don't know about the efficiency, 
though) by replacing the if/else statement by:

        binstr = string.digits[x & 1] + binstr
or:

        binstr = ('0', '1')[x & 1] + binstr


Robert Amesz


From fredrik at pythonware.com  Tue Mar 20 18:01:53 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Tue, 20 Mar 2001 23:01:53 GMT
Subject: re.match
References: <997skf$qoc$1@bmerhc5e.ca.nortel.com> <TrMt6.2490$4N4.318602@newsc.telia.net> <99874d$4tp$1@bmerhc5e.ca.nortel.com> <j4r8zsi3pw.fsf@informatik.hu-berlin.de> <998hhn$cep$1@bmerhc5e.ca.nortel.com>
Message-ID: <BjRt6.2586$4N4.345428@newsc.telia.net>

> My experience with regular expressions is completely with the Unix tools
> that use them (ie. grep, awk, sed, vi...) and Perl.

for some reason, this guy reminds me of the radio announcer who
uttered "bob dylan, best known from travelling wilburys"...

(or maybe even the work of two of last year's ignobel winners ;-)

anyone who knows the slightest bit about the theory behind regular
expressions should know that beyond the fundamental AB, A|B, and
A* forms, there's no such thing as a regular expression standard.

everything else is either sugar, irregularities, or both.

and don't get me started on the "regular expressions should always
search for a match" nonsense...

Cheers /F




From hinsen at cnrs-orleans.fr  Fri Mar 30 07:11:31 2001
From: hinsen at cnrs-orleans.fr (Konrad Hinsen)
Date: 30 Mar 2001 14:11:31 +0200
Subject: NumPy loop efficiency
References: <3ac44e27$0$16692$73bec57d@news.be.uu.net> <mailman.985951902.14535.python-list@python.org>
Message-ID: <m3g0fv78y4.fsf@chinon.cnrs-orleans.fr>

Carlos Alberto Reis Ribeiro <cribeiro at mail.inet.com.br> writes:

> The intermediate array is not really needed, but the python expresion 
> evaluator have no way to know it. However, something like the ufuncs could 
> be used. For instance, something like this could be done (the names are 
> just examples):

Something like that (but not entirely) is possible but little known:

     Numeric.add(a, b, c)

does the same as

     c = a + b

but storing the result in the preallocated array c (which must of
course have the right dimensions). And c can well be a or b, in which
case the data is overwritten. By using only these functions, many
temporary arrays can be avoided, but at the cost of producing
unreadable code.
-- 
-------------------------------------------------------------------------------
Konrad Hinsen                            | E-Mail: hinsen at cnrs-orleans.fr
Centre de Biophysique Moleculaire (CNRS) | Tel.: +33-2.38.25.56.24
Rue Charles Sadron                       | Fax:  +33-2.38.63.15.17
45071 Orleans Cedex 2                    | Deutsch/Esperanto/English/
France                                   | Nederlands/Francais
-------------------------------------------------------------------------------


From daniel769 at earthlink.net  Thu Mar 22 20:08:12 2001
From: daniel769 at earthlink.net (Daniel Chamberlin)
Date: Fri, 23 Mar 2001 01:08:12 GMT
Subject: Scrolling tkinter Canvas
Message-ID: <B6DFE101.52A%daniel769@earthlink.net>

I am unable to scroll a canvas.  could someone look at the code squib below
and point out what needs to be changed?
essentially what happens is that the Canvas fills up and stretches to fill
the screen, the scrollbar remains unchanged, and the remaining labels go
somewhere where they can't be seen.  what i would like to happen is for the
canvas's size to be limited, and for the scrollbar to control what is
viewed.  i have gotten scrollbars to work with text widgets, but i am unable
to get this to do what i want.

thanks for any help,
doug

from Tkinter import *
class contain:
    def __init__(self, parent):
        self.f = Frame(parent, width=30, height=30)
        self.f.pack()
        self.c=Canvas(self.f)
        self.s=Scrollbar(self.f, orient='horizontal')
        self.c.config(xscrollcommand=self.s.set)
        self.s.config(command=self.c.xview)
        for i in range(50):
            Label(self.c, text='label #' + str(i)).pack(side=LEFT)
        self.c.pack(expand=NO)
        self.s.pack(expand=YES, fill=X)

if __name__=='__main__':
    root = Tk()
    x  = contain(root)
    mainloop



From hinsen at cnrs-orleans.fr  Thu Mar 29 06:45:02 2001
From: hinsen at cnrs-orleans.fr (Konrad Hinsen)
Date: 29 Mar 2001 13:45:02 +0200
Subject: Gaussian line profile using numeric python
References: <3ac31698@zfree.co.nz>
Message-ID: <m3puf0re81.fsf@chinon.cnrs-orleans.fr>

"Geoff Low" <geoff at ou043085.otago.ac.nz> writes:

> I have an extensive set of xy data that I want to convolute with a gaussian
> profile.  Below is my algorithm,  which doesn't work.  Can anyone see where
> I'm being dumb?

To be honest, I am too lazy to look through this. However,

1) The function Numeric.convolve is certainly a lot more efficient
   than your code, and should make it simpler.

2) If you work with large data sets, you ought to use FFTs to
   compute the convolution, that's O(N*log(N)) in the size of the
   data set, instead of O(N**2) for the straightforward method.
   Any good book on FFTs should explain how this works in detail.

> P.S.  I'm no math wizard but this is my interpretation of the gaussian lineshape
> function
> G(v) = 1/(sqrt(2*PI)*FWHM)*exp(-(v-vo)^2/(2*FWHM^2))

Assuming that FWHM stands for full width at half minimum, no, it's not
correct. What you call FWHM is the variance sigma of the Gaussian, the
FWHM is around 2.35*sigma (quick calculation, no guarantee).
-- 
-------------------------------------------------------------------------------
Konrad Hinsen                            | E-Mail: hinsen at cnrs-orleans.fr
Centre de Biophysique Moleculaire (CNRS) | Tel.: +33-2.38.25.56.24
Rue Charles Sadron                       | Fax:  +33-2.38.63.15.17
45071 Orleans Cedex 2                    | Deutsch/Esperanto/English/
France                                   | Nederlands/Francais
-------------------------------------------------------------------------------


From scarblac at pino.selwerd.nl  Sat Mar 10 18:28:14 2001
From: scarblac at pino.selwerd.nl (Remco Gerlich)
Date: 10 Mar 2001 23:28:14 GMT
Subject: Python In A Nutshell
References: <nG5q6.54910$__6.11376601@typhoon.southeast.rr.com> <Ufcq6.140786$47.2078615@news.bc.tac.net> <Blfq6.60716$__6.11886688@typhoon.southeast.rr.com>
Message-ID: <slrn9ale0l.gvt.scarblac@pino.selwerd.nl>

Don Tuttle <tuttledon at hotmail.com> wrote in comp.lang.python:
> Planning stages?  We've all seen his posts (a.k.a. treatises) here.  I'd say
> he should have the final draft to you in 3 weeks tops. ;-)

Constructing the nutshell that can contain it will take years :-)

-- 
Remco Gerlich


From dmaas at dcine.com  Sun Mar 11 16:01:49 2001
From: dmaas at dcine.com (Dan Maas)
Date: Sun, 11 Mar 2001 21:01:49 GMT
Subject: supporting asyncronous messages
References: <mailman.983793067.17928.python-list@python.org> <j466hloqof.fsf@informatik.hu-berlin.de> <mailman.984312437.15426.python-list@python.org>
Message-ID: <1JRq6.240846$o91.31048136@typhoon.nyroc.rr.com>

> i'd like to use tristate logic to automatically
> wait for only the relevant events. idea being that if an expression
> evaluates to "unknown", it should block, but if it evaluates to true or
> false it should continue. it means that you only need to worry about
> a single expression to control flow and wait behaviour.

I see what you are trying to do here - basically get "as much work done as
possible" in an environment where input data for a given computation may not
be available immediately. I'm not sure working at the Python syntax level is
the best choice here; there are likely to be many unexpected problems (e.g.
how do you implement coroutines?).

I think a more practical approach would be to write a simple virtual machine
in Python that takes care of scheduling the execution of your coroutines. If
you don't need much flexibility, you could hard-code them or invent a really
simple interpreted language. If you really need full Python, I think you
could do it by breaking the routine up into individual statements. Have the
scheduler try to 'exec' each statement in turn -- if the input isn't ready
yet, throw a special exception. Then the scheduler can record what input was
missing, go take care of some other work, and come back later when the input
arrives...

Regards,
Dan




From charlottwilliams at mediaone.net  Thu Mar 22 12:21:39 2001
From: charlottwilliams at mediaone.net (Charlotte Williams)
Date: Thu, 22 Mar 2001 17:21:39 GMT
Subject: Help with creating Python COM
Message-ID: <Dwqu6.79766$Ch.10654381@typhoon.we.rr.com>

I am following a tutorial for creating a Python COM server that will
interact with a Visual Basic client.  The author included an example with a
set of instructions.  However, when I go to register the COM, it gives me an
error.  When I went back into the debugger and tried to debug the python
script, it tells me

ImportError: No module named ni

All the tutorials I have encountered tells me that these are the modules
that must be imported:

 import ni
 from win32com.server.exception import Exception
 import win32com.server.util

I am running PythonWin with Python 2.0 (BeOpen).  The link to the tutorial
is here:

http://www.hps1.demon.co.uk/users/andy/pyvb/index.htm


Any knowledge of what I am doing wrong?





From sheila at spamcop.net  Sun Mar 25 09:36:11 2001
From: sheila at spamcop.net (Sheila King)
Date: Sun, 25 Mar 2001 14:36:11 GMT
Subject: time to invest ???
References: <3abd3a17.7103358@news.vif.com> <99jfmh$1i58c$1@ID-11957.news.dfncis.de>
Message-ID: <rd0sbtg0n52kt100nj5lg78jp9t0f87eqb@4ax.com>

On Sat, 24 Mar 2001 16:52:23 -0800, "Emile van Sebille" <emile at fenx.com> wrote
in comp.lang.python in article <99jfmh$1i58c$1 at ID-11957.news.dfncis.de>:

:Pick a simple admin task, something you do regularly and feel you
:understand, work the tutorial, and then implement in python.
:
:Two ideal days.  If it takes a week, you've started with too big a project.
:Stop at two days and ask for help/clarification.

I would agree with the "two days" figure, if he already had some previous
programming experience. But as he says, "little as a programmer", it will
probably take longer, since he is learning programming at the same time.

Still, your advice to ask for help after two days is good advice.

I have (some) programming experience, and started with a task I wanted to do,
and it took me a few days, after having already spent a couple of days looking
through the tutorial and so forth.

--
Sheila King
http://www.thinkspot.net/sheila/
http://www.k12groups.org/

:---------
:"Yvon Boulianne" <mysticsguy at yahoo.ca> wrote in message
:news:3abd3a17.7103358 at news.vif.com...
:> Hello, i start to learn Python some day ago, with lot of knowledge as
:> a system admin but little as programmer and i wonder what is the
:> average time/hours someone have to invest to be able to use a language
:> like that easily.  i talk as well about devlopping the way of thinking
:> of a programmer and the knowledge of the language. please don't tell
:> me theres no average, at least take the idea of a normal guy, not a
:> genius and not stupid  ;-)
:>
:> Thanks verry much and sorry for my english, my first language is
:> french.
:>
:> Yvon Boulianne ; Saphi
:> mysticsguy at yahoo.ca
:> www.monunivers.net
:> Humour,Paradox & Changements
:



From robin at jessikat.fsnet.co.uk  Fri Mar 23 03:15:15 2001
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Fri, 23 Mar 2001 08:15:15 +0000
Subject: extending the built in dictionary type
References: <umfKujAz2iu6EwHa@jessikat.demon.co.uk>
 <6Rzu6.12951$_46.485491@e420r-atl2.usenetserver.com>
Message-ID: <BXz8sKATYwu6Ewg9@jessikat.fsnet.co.uk>

In article <6Rzu6.12951$_46.485491 at e420r-atl2.usenetserver.com>, Steve
Holden <sholden at holdenweb.com> writes
>"Robin Becker" <robin at jessikat.fsnet.co.uk> wrote in message
>news:umfKujAz2iu6EwHa at jessikat.demon.co.uk...
>> I'm interested in extending the built in dictionary. It's fairly simple
>> to wrap, but seems incredibly hard to chain the methods. Is there any
>> obvious way to use most of the built in methods whilst modifying a few.
>>
>> Seems that although it exists I can't chain the original mapp_methods of
>> dictobject as that is not visible outside pythonxx.dll. So is there an
>> easier way to get at these internal types?
>
>See FAQ 5.18: How do I define and create objects corresponding to
>built-in/extension types?
>
>regards
> Steve
perhaps I didn't make myself clear. I will use UserDict for a python
version, but if I want to accelerate to C I find fundamental problems in
getting useful access to the code of say dictobject; in particular
there's no obvious way to extend/modify in a small way; eg have a dict
with just one more method. I either have to just copy that code
wholesale or do rather silly things like changing my object's type
before and after the operation. As an example I'm forced to do this to
safely wrap d[x] and d[x]=v

static PyObject * MYDict_subscript(PyObject *self, register PyObject
*key)
{
        PyObject* r;
        self->ob_type=&PyDict_Type;
        r = dict_subscript(self,key);
        self->ob_type=&MyDictType;
        return r;
}

static int MyDict_ass_sub(PyObject *self, PyObject *v, PyObject *w)
{
        int r;
        self->ob_type=&PyDict_Type;
        r = dict_ass_sub(self,v,w);
        self->ob_type=&MyDictType;
        return r;
}

this because the internal methods dict_subscript, dict_ass_sub do type
checking which is a bit peculiar as I would expect the internal methods
to assume the correct type.

Worse I find that there's no standard way to get at or the chain of
methods for the built in types. Some lip service stuff is there in the
chain methods, but no way to get at the chain except by using some
rather dubious coding eg

        v = PyObject_GetAttrString(d,"has_key");
        dmapp_methods = (PyMethodDef*)(((PyCFunctionObject*)v)->m_ml);

"has_key" is the name of the first thing on the list and that could
change. So I can use this sneakiness to get at and build a chain. My
objects now work pretty much as expected; provided I wrap those methods
of dictobject which do PyDict_Check(). The "__methods__" special is
treated wrongly in that some methods appear twice in the list, but
that's a minor issue as the callchain is handled correctly.
-- 
Robin Becker


From tuttledon at hotmail.com  Thu Mar 22 11:17:25 2001
From: tuttledon at hotmail.com (Don Tuttle)
Date: Thu, 22 Mar 2001 16:17:25 GMT
Subject: Good Editor? 
References: <3ab9a32b$1_1@news2.one.net>
Message-ID: <pApu6.166124$Z8.35236583@typhoon.southeast.rr.com>

"fjs" <fjs6 at usa.net> wrote in message news:3ab9a32b$1_1 at news2.one.net...
> Can someone recomend a good editor to write python in?  Im just getting
> started learning python and dont know which ones support python.

And just to round out a couple of other 'free' choices:

1) If you're using Windows you need to get Mark Hammond's Win32 extensions
which includes the PythonWin editor.
http://www.activestate.com/Products/ActivePython/win32all.html (You already
have this if you use ActivePython)

2) Komodo, which is currently at beta 4 and runs on Windows & Linux.
http://www.activestate.com/Products/Komodo/index.html

Don




From m.wilson at bibliocraft.com  Mon Mar  5 22:43:05 2001
From: m.wilson at bibliocraft.com (Mark Wilson)
Date: Tue, 6 Mar 2001 03:43:05 -0000
Subject: MySQL-python build griefs
Message-ID: <006a01c0a5ef$92830240$17824ac3@mwdell1>

Will Ware:
>I finally did find a fix. There's a point in sysconfig.py around line
>195 or so where it says something like:

Hmmmm - a somewhat more subtle problem than the one I had :->

Glad you found a fix.

Mark




From corour01 at motorola.com  Tue Mar 27 08:15:31 2001
From: corour01 at motorola.com (O'Rourke Clodagh-corour01)
Date: Tue, 27 Mar 2001 14:15:31 +0100
Subject: Netscape and Python
Message-ID: <13EE655665F4D311AB4D0008C789498A01EE2CA5@zei02exm02.cork.cig.mot.com>

Hi all, 

Does anyone know how to launch Netscape from Python?

Thanks,
Clodagh




From dalfes at itu.edu.tr  Wed Mar  7 08:07:54 2001
From: dalfes at itu.edu.tr (Nuzhet Dalfes)
Date: Wed, 07 Mar 2001 15:07:54 +0200
Subject: How to installing NumPy?
References: <3AA57C0D.BBC5E956@itu.edu.tr> <9842vc$4bb$1@mothership.telocity.com>
Message-ID: <3AA632AA.7D15888C@itu.edu.tr>

Thanks for suggestions from Warren and Christopher, but I get some sort
of error related to some LONG_BIT issue. 

The output of my second trial of 'python setup.py install' is attached.

I'll appreciate some more help; is this a known problem?

Nuzhet


Warren Focke wrote:
> 
> In article <3AA57C0D.BBC5E956 at itu.edu.tr>,
> Nuzhet Dalfes wrote:
> >I am trying to install NumPy on a RedHat 7.0 running python2.0.
> >Everything is in the standard places (I think!) and I picked up and
> >unzipped http://download.sourceforge.net/numpy/NumPy1711_20.zip into
> 
> That's for windos.  You want to get Numeric-17.3.0.tar.gz, unpack it
> (not in site-packages, maybe in /usr/local/src or ~/src), cd to the
> directory it made, and follow the instructions there.  Probably
> something like:
> 
> /usr/local/bin/python2.0 setup.py install
> /usr/local/bin/python2.0 setup_all.py install
> 
> Warren Focke
> --
> Wrapping up her appearance with rubber, you can feel more comfortable.
>  -- directions for a set of Chinese binoculars
-------------- next part --------------
In file included from /usr/local/include/python2.0/Python.h:54,
                 from Src/_numpymodule.c:1:
/usr/local/include/python2.0/pyport.h:390:2: #error "LONG_BIT definition appears wrong for platform (bad gcc config?)."
running install
running build
running build_py
not copying Lib/ArrayPrinter.py (output up-to-date)
not copying Lib/Numeric.py (output up-to-date)
not copying Lib/Precision.py (output up-to-date)
not copying Lib/UserArray.py (output up-to-date)
not copying Lib/numeric_version.py (output up-to-date)
running build_ext
building '_numpy' extension
gcc -g -O2 -Wall -Wstrict-prototypes -fpic -IInclude -I/usr/local/include/python2.0 -c Src/_numpymodule.c -o build/temp.linux-i686-2.0/Src/_numpymodule.o
error: command 'gcc' failed with exit status 1
-------------- next part --------------
A non-text attachment was scrubbed...
Name: dalfes.vcf
Type: text/x-vcard
Size: 389 bytes
Desc: Card for Nuzhet Dalfes
URL: <http://mail.python.org/pipermail/python-list/attachments/20010307/7290c4dc/attachment.vcf>

From carribeiro at yahoo.com  Tue Mar 20 22:38:47 2001
From: carribeiro at yahoo.com (Carlos Ribeiro)
Date: Wed, 21 Mar 2001 00:38:47 -0300
Subject: Functionalism, aesthetics Was:(RE: I come to praise .join,
   not
In-Reply-To: <996vss013mp@news2.newsguy.com>
References: <mailman.985052844.17993.python-list@python.org>
Message-ID: <5.0.2.1.0.20010320235445.009e60d0@mail.inet.com.br>

At 08:11 20/03/01 +0100, Alex Martelli wrote:
>Back when "Concorde" and "Jumbo" first appeared on the scene, for
>example, non-experts found the former quite beautiful, the latter quite
>plain.  Sure the sleek, fast-looking lines of Concorde were superior to
>the goofy-looking fatness of Jumbo?

Are you sure that this is a good example? One thing is to be impressed by 
the look, and the other is to actually use the product. Opinions may (and 
will) change afterward. This is pretty much common in architecture also, 
where we have beautiful buildings that simply don't work. Also there is 
another important point in your example.

The Concorde was designed by comitte, and it shows. A lot of the reasons 
why the Concorde looks cool is that it was designed based on some purely 
objective assumptions. The comitee decided that people would pay more to go 
faster - so the supersonic speed. For a supersonic, you have to follow 
different design rules - so the cone nose and the shape of the wings was 
born. As the resulting airplane got large, the crew needed some way to see 
the ground - and the moving nose was designed. So the cool look. People 
were impressed by the visual design, but not by the solution. Thats why it 
failed.

The Jumbo is one of the best examples of "design patterns" that we can 
think of. It was an evolution of a line of airplanes, borrowing features 
from the great bombers of the World War II and the first commercial jets. 
It was a very conservative design, and while it keeps innovating, Boeing is 
extemily careful about introducing new features. It took them decades to 
make a fly-by-wire 747.

This is a great quote from Boeing website:

"One thing about the Boeing 747 is clear: the more things appear to stay 
the same with this magnificent giant, the more things change. While today's 
747 might look like the first 747, it is an entirely different airplane -- 
after all, an airplane that's accumulated 20 billion miles (32 billion km) 
of airborne experience is bound to evolve. "
http://www.boeing.com/commercial/747family/history.html


>We can endlessly discuss where computers and software
>fit on this scale, but surely the example of pharmaceutics
>does suffice to show conclusively that, with far too many
>modern technologies, it's quite likely that what appears to
>the untrained eye has *zero* relevance -- that, no matter
>how ideologically unpleasant this may feel, we DO have to
>rely on _expertise and training_ to be able to judge certain
>issues.  Vox populi is NOT necessarily vox Dei -- not when
>the underlying technology is too new (a few centuries) and/or
>not reflected in immediately apperceivable traits.

I'm not talking about completely untrained users here. Most doctors would 
appreciate the explanation given by the pharmaceutical industry. And while 
I agree that Vox Populi != Vox Dei, this has *nothing* to do with the 
technology maturity as measured by time alone. Our innate reaction against 
change play a big role here too.

> > To summarize it: a  *technically* optimal solution for a problem is not
> > always the *best* solution. If it does not look intuitively good for most
> > users, its probably a bad solution.
>
>So, you're arguing that the amount of artificial coloring should
>have been kept high (who cares about giving users cancer, right?),
>since that bright color was definitely what "looked intuitively
>good for most users" as proved by the endless carping...?

Ok, ok. ' Users' is misleading. However, for Python, 'users' usually means 
programmers, people with sufficient knowledge to make good questions about 
proposed changes.

> > Its unfortunate that in many cases the non-expert user keeps quiet,
> > assuming that whatever opinion the "master" haves, its going to be better
> > than theirs - after all they are the so-called experts.
>
>Ask the experts in any field whether untrained users keep too quiet,
>or not enough, in their field, and you'll get a different opinion (of
>course, self-serving conscious or unconscious biases are at work
>as well!).

The fact that many people complain without reason is not a valid excuse for 
ignoring all complaints.


Carlos Ribeiro


_________________________________________________________
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com




From gtalvola at nameconnector.com  Thu Mar 15 16:14:31 2001
From: gtalvola at nameconnector.com (Geoff Talvola)
Date: Thu, 15 Mar 2001 16:14:31 -0500
Subject: Python and ADO
References: <98r9qn$tbi$1@troll.powertech.no>
Message-ID: <3AB130B7.6A09C3F1@NameConnector.com>

Syver Enstad wrote:

> Has anyone considered making a DB API compliant wrapper around MS ADO? Or
> isn't this considered worthwhile?
>
> As ADO is the database api that I use, I could have a look at making a
> wrapper for it if there was any interest in it and if making such a thing is
> not too involved.

Michel Orengo already wrote a preliminary version of a DB-API wrapper for ADO but I don't think he ever finished it.  I have a copy of it, but I have never actually tried it.  If you can't find it in the newsgroup archives, drop me an email and I'll send you a copy.

And I do think this would be quite useful...

--


- Geoff Talvola
  Parlance Corporation
  gtalvola at NameConnector.com




From fredrik at pythonware.com  Wed Mar 28 13:57:11 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 28 Mar 2001 18:57:11 GMT
Subject: list vs tuple
References: <hGlw6.782$p5.3210@news1.rivrw1.nsw.optushome.com.au> <99stcb$483$1@panix6.panix.com> <99t9om$2o2q3$1@ID-77331.news.dfncis.de>
Message-ID: <buqw6.1999$sk3.675030@newsb.telia.net>

Mark Pilgrim wrote
> It was my understanding that only tuples containing immutable objects could
> be used as dictionary keys.  For instance, (1,2,3) can be, and ("a","b","c")
> can be, but ([1,2,3], ["a","b","c"]) can not, because the underlying objects
> could mutate and cause problems.  (Wow, that sounds like a bad sci-fi
> movie.)  Is this true?

no python interpreter within reach today?

>>> d = {}
>>> d[(1, 2, 3)] = 1
>>> d[("a", "b", "c")] = 2
>>> d[([1, 2, 3], ["a", "b", "c"])] = 3
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
TypeError: unhashable type

Cheers /F

PS. For some reason, nobody has mentioned that tuples and lists
tend to be *used* in different ways, also when you're not using
them as dictionary keys:
- Tuples are typically used to group heterogenous sets
  of data, just like structs/records in other languages.
- Lists are used as containers, to store homogenous sets
  of data, just like arrays/vectors in other languages.




From Andreas.Leitgeb at siemens.at  Wed Mar 14 07:54:09 2001
From: Andreas.Leitgeb at siemens.at (Andreas Leitgeb)
Date: Wed, 14 Mar 2001 12:54:09 +0000 (UTC)
Subject: Einstein's Riddle
References: <oNPq6.396644$ge4.138479566@news2.rdc2.tx.home.com> <slrn9apj4g.c54.Andreas.Leitgeb@pc7499.gud.siemens.at>
Message-ID: <slrn9auqu1.c54.Andreas.Leitgeb@pc7499.gud.siemens.at>

Andreas Leitgeb <Andreas.Leitgeb at siemens.at> wrote:
>based on this kind of riddles, I've created a  tcl/tk-prog, that
>helps one solving it. 

someone asked (per email), so here it is:

 <http://avl.enemy.org/MyWorks/TclTk/>  get  "logik.tgz"

PS: there is also some other stuff there ...



From fredrik at pythonware.com  Wed Mar 14 11:11:36 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 14 Mar 2001 16:11:36 GMT
Subject: main in mac?
References: <mailman.984497176.29067.python-list@python.org> <20010313123230.19201.00000153@ng-dh1.aol.com> <mailman.984569899.20412.python-list@python.org>
Message-ID: <YKMr6.256$4d3.54207@newsb.telia.net>

Doug Hellmann wrote:
> It's not that you're running on a Mac, it's that you're using an IDE.

footnote: not all IDE's have this problem, of course.

Cheers /F

<!-- pythonworks 1.2 coming soon:
http://www.pythonware.com/products/works/index.htm
-->




From johann at physics.berkeley.edu  Mon Mar  5 12:38:49 2001
From: johann at physics.berkeley.edu (Johann Hibschman)
Date: 05 Mar 2001 09:38:49 -0800
Subject: Little problem with the "+" operator
References: <983802310.155838463@news.t-online.de> <mailman.983804468.16428.python-list@python.org> <980i6q$crs$1@news.mathworks.com>
Message-ID: <mt1yscm7yu.fsf@astron.berkeley.edu>

Joshua Marshall writes:

>> This is infinitely more sane than:

>> % perl -e 'print 500 + "cool"'
>> 500

> Comparing Python to Perl here doesn't seem very fair.  They're very
> different languages.  In general, I'm surprised by the urgency with
> which people often jump to compare them.

Well, Perl's the only language that I know of that lets you add 500 to
"cool" without an error.  Well, I suppose tcl might.  But it's still a
fairly safe inference that the original poster was coming from Perl.

-- 
Johann Hibschman                           johann at physics.berkeley.edu


From imunoz at eso.org  Sat Mar 10 22:00:22 2001
From: imunoz at eso.org (Ivan Munoz)
Date: Sat, 10 Mar 2001 23:00:22 -0400
Subject: Setting Exception data.
Message-ID: <3AAAEA46.5922B1BD@eso.org>

Hi,

I have a C extension module called 'foo' and I define an Exception class
for it foo.error.

When raising a foo.error exception In my C code I set the error object
to some data (an object defined on my extension module, an integer and a
string) as follows:

 v=Py_BuildValue("Ois", e, errorNumber, errStr);
 if (v != NULL) {
   PyErr_SetObject(ErrorObject, v);
   Py_DECREF(v);
 }

so far so good, in Python then I catch this exception as:

except foo.error, (object, errnum, errstr):
    .....
    .....


Now what is the right way to acomplish the same from Python?, this does
not seem to work ....

    raise foo.error MyObject, errnum, 'You are toasted'


thanx in advance,
IvO



From bedge at troikanetworks.com  Tue Mar  6 11:06:19 2001
From: bedge at troikanetworks.com (Bruce Edge)
Date: Tue, 06 Mar 2001 08:06:19 -0800
Subject: absolute namespace for callbacks?
Message-ID: <3AA50AFB.83C7459F@troikanetworks.com>

Tcl has a func to get a fully qualified namespace from a context. This
is essential for registering callback functions.
How does one register a callback in Python, when the callback func
exists in a different namespace?

Is there any method to get the current absolute namespace?

Thanks, Bruce.


From scarblac at pino.selwerd.nl  Thu Mar  1 08:04:30 2001
From: scarblac at pino.selwerd.nl (Remco Gerlich)
Date: 1 Mar 2001 13:04:30 GMT
Subject: is and == (is not other_thread)
References: <3A9DD10F.C8ECF803@seebelow.org>
Message-ID: <slrn99si3q.72.scarblac@pino.selwerd.nl>

Grant Griffin <not.this at seebelow.org> wrote in comp.lang.python:
> I've had a few of my own little questions about "is" and "==", vis-a-vis
> strings:
> 
> 1) Are two strings that are equal in the "==" sense always guaranteed to
> be equal in the "is" sense?

No. Only when they have been interned. No guarantees about interning are
given, but you can use intern() yourself, I think string literals are
automatically interned, and maybe short strings as well. But no, you can't
in general use 'is' for string comparison.

> 2) Is "is" faster than "==" for strings?  Of hand, it would seem so. 
> That's why I'm tempted to use it for string equality.  (OTOH, if strings
> can be "==" equal without being "is" equal, than I guess that's wildly
> false economy. <wink>)

'is' may be slightly faster, but can't be used in general.

> 3) Internally, does "==" for strings always begin with an "is"
> operation?

'==' starts with an 'is' test for any argument.

Simple test:

class A:
   def __cmp__(self, other):
      return -1 # Compares different to anything

print A() == A() # Prints 0, two different instances
x = A()
print x == x # Prints 1, the 'is' test is used and __cmp__ is skipped


> (btw,-nobody-ever-answers-every-part-of-my-silly-multipart-questions,
>    -so-i-dare-you-folks-to-buck-the-trend-<wink>)-ly y'rs,

Oops. It was an accident, honest! I hadn't read this yet!

-- 
Remco Gerlich


From chrisl at gnuchina.org  Fri Mar 23 14:32:40 2001
From: chrisl at gnuchina.org (Christopher Li)
Date: Fri, 23 Mar 2001 11:32:40 -0800 (PST)
Subject: Yet another bison wrapper in pure python
Message-ID: <Pine.LNX.4.30.0103230953500.11741-100000@chris.cn.mvd>

Hi,

I have write yet another bison wrapper, yes I know there are many of them.
I like to use bison because it has a very nice output to help me debug
the grammer conflict in the .y file.

The idea is that
1. put the python statment in .y file just like C statment.
   it need to pay attention about the indent though.
2. invoke bison to generate the .tab.c file.
3. scan the .tab.c file to extract the tables(array) and the action
   statment, which is the python code.
4. put them together in another python program using the table.
   I write a  python version of the bison parser C code. It is just
   a python exactly remap of the C code.

I use this trick to write the chasm, and feel this tool is very handy
because do not need to compile the C module. Just the edit-run-debug
cycle. BTW, chasm is a C style Asm, it is in pure python,
http://sourceforge.net/projects/ch-asm/

I really enjoy python as a rapid phototype language.

I take the bison wrapper and some help function to do lexer in a tar ball.
http://gnuchina/~chrisl/pb.tgz

It include an example of the Multi-function Calc which is the python
version of the third Bison info example. I use it to debug the python
parser itself. The python version is smaller than the C version and
more powerful (It support all the math function in one blow)
-----------------cut--------------------
[root at chris pb]# python pbison -p bisonparser ptest2.y
ptest2.y contains 4 shift/reduce conflicts.
[root at chris pb]# python ptest2.py
>>>a = sin(1)+4-5
        -0.158529015192
>>>b = a+6
        5.84147098481
>>>
--------------------------------------
The code can be improved if there is a need. And if want to speed it up.
We can build a reusable C version of the parser module, table and
action still be the python. I guess the "PyBison" does similar things
but I can't download it, its link is broken.

BTW, seems the I have to use the python2.0 "pre" module as re. otherwise
scan big tables will have some problem.

And, is there a better way to do a long "switch" like statment?
Right now I map a deduce action as a hash of functions. and all the
state of parser is function too. anyway, it is not aim at run fast
at the first place.

Any idea and suggestion is welcome.

Chris




From egmanoj at yahoo.com  Wed Mar 28 08:10:09 2001
From: egmanoj at yahoo.com (egmanoj at yahoo.com)
Date: Wed, 28 Mar 2001 13:10:09 -0000
Subject: Using urlopen with a proxy server requiring authentication
Message-ID: <99snrh+6tpd@eGroups.com>

Hello,
I am trying to use the Python(1.5) urllib module's urlopen method to 
retrieve an xml file from a website. The request has to go through 
our LAN's proxy server and this needs a username and password.

How can I automatically supply the username and password while using 
this from a script? [There is no problem while using urlopen from the 
IDLE; the prompt comes up and I just have to type in the username and 
password]

Thanks in advance
Manoj E.G.




From pj at sgi.com  Wed Mar 28 15:30:58 2001
From: pj at sgi.com (Paul Jackson)
Date: 28 Mar 2001 20:30:58 GMT
Subject: Getting all the *files* from a directory -- A better way??
References: <mailman.985732718.3725.python-list@python.org> <99r6v7$4hef2$1@fido.engr.sgi.com> <slrn9c3jf1.qb3.scarblac@pino.selwerd.nl>
Message-ID: <99thm2$510ni$1@fido.engr.sgi.com>

Remco suggested:
|> dir="a/b/c"
|> files = [os.path.join(dir, f) for f in os.listdir(dir)
|>                               if os.path.isfile(f)]
|> 
|> If you do the map() with a list comprehension, why not the filter() as well.

I doubt this will work.  The condition:

    if os.path.isfile(f)

will fail to test the right path.  You have to test "a/b/c/f",
not "f".  Only by first changing directory to "a/b/c" would
the above work.
-- 
                          I won't rest till it's the best ...
                          Manager, Linux System Software
                          Paul Jackson <pj at sgi.com> 1.650.933.1373


From iansworld at thevortex.com  Tue Mar 20 10:13:18 2001
From: iansworld at thevortex.com (EEN)
Date: Tue, 20 Mar 2001 15:13:18 GMT
Subject: HELP - beginner to programing/python
References: <Vpxt6.416511$Pm2.6478704@news20.bellglobal.com>
Message-ID: <isKt6.423538$Pm2.6558222@news20.bellglobal.com>

Thanks for the help everyone! The links were very helpful, and some of them
start out quite "mickey mouse", which is waht i wanted. I found that the one
on the Pythom site jumps in too quick.

Thanks again, and i'll probably be back again!

EEN <iansworld at thevortex.com> wrote in message
news:Vpxt6.416511$Pm2.6478704 at news20.bellglobal.com...
> Hi there. I am a newbie and am hearing that python is the way to go to
start
> off. I am downloading it as i type this, but need to know how to use it :)
> of course :) Is there any site that has a good tutorial or beginners
guides,
> or do you have to buy books? If so, waht do you recommend?
>
> Thanks a lot for the help, i really appreciate any pointers!
>
>




From renzo.tomaselli at tecnotp.it  Fri Mar 16 12:51:47 2001
From: renzo.tomaselli at tecnotp.it (Renzo Tomaselli)
Date: Fri, 16 Mar 2001 18:51:47 +0100
Subject: relational/boolean operator overloading
Message-ID: <98tjq6$qre$1@news.flashnet.it>

Hi all,
        I'm trying to use a subset of Python for querying an object
database, so that I would like to use a relational operator between a set
and a const value; this would yield another set. Multiple relations would be
connected by boolean operators with a meaning of set operators and so on.
This approach would enable a straight variable binding.
However there seems no way to redefine such operators, only numeric binary
operators are mentioned in the ref. manual.
Any trick/suggestion/comment ?
Thanks,
                                             Renzo Tomaselli
-------------------------------------------------------------------------
TecnoTP s.n.c. Special Information System Design
Maso Pelauchi I38050 Ronchi Valsugana,  Trento TN  ITALY
Tel. +39 0461 773164      Fax. +39 0461 771514
e-mail: renzo.tomaselli at tecnotp.it
-------------------------------------------------------------------------





From boud at rempt.xs4all.nl  Mon Mar 26 00:46:57 2001
From: boud at rempt.xs4all.nl (Boudewijn Rempt)
Date: 26 Mar 2001 05:46:57 GMT
Subject: Good GUI for Python
References: <mailman.985314033.5120.python-list@python.org> <AADu6.436656$f36.14223632@news20.bellglobal.com>
Message-ID: <99ml4h$jm9$1@news1.xs4all.nl>

Bryan Mongeau <bryan at eevolved.com> wrote:
> For some reason, people tend to overlook FXPy, the python bindings for the 
> FOX GUI toolkit:

> fxpy.sourceforge.net

> There is also pyQT available from thekompany.com but the legality of the 
> windows licensing remains a thornbush.

Not really: you can buy a copy of Qt for Windows, and compile PyQt
yourself on Windows, in which case you can do with the result whatever
you want, or you can buy BlackAdder Business Edition (which includes
both Linux and Windows versions), and do with the application you've
developed with it whatever you want. So, yes, development for Windows
will cost you money: more if you intend to distribute your application
(the personal edition of BlackAdder is very cheap), but still not a
whole lot: $400,-.


-- 

Boudewijn Rempt  | http://www.valdyas.org


From sandipan at vsnl.com  Sat Mar 10 13:01:44 2001
From: sandipan at vsnl.com (Sandipan Gangopadhyay)
Date: Sat, 10 Mar 2001 23:31:44 +0530
Subject: CGISSL
Message-ID: <002201c0a98c$33a46020$020000c3@node02>

Is the cgissl module documentation available online ?

Thanks,

Sandipan




From stephen_purcell at yahoo.com  Wed Mar 21 11:50:17 2001
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Wed, 21 Mar 2001 17:50:17 +0100
Subject: os.system output
In-Reply-To: <B7C74E1A6FBAD411BFC60002B309DDF4E42F@mailbox.juilliard.edu>; from jsoons@juilliard.edu on Wed, Mar 21, 2001 at 11:32:39AM -0500
References: <B7C74E1A6FBAD411BFC60002B309DDF4E42F@mailbox.juilliard.edu>
Message-ID: <20010321175017.A28561@freedom.puma-ag.com>

Jonathan Soons wrote:
> I could do
> 
> >>>os.system('ls > file')
> >>>f = open('file')
> >>>dirs = f.readlines()
> 
> but it looks crude. Is there a better way?

Use the function 'popen' in the 'os' module.

>>> import os
>>> f = os.popen('ls')
>>> dirs = f.readlines()
>>> f.close()

Better still, for what you're trying to do:

>>> import os
>>> dirs = filter(os.path.isdir, os.listdir('.'))

-Steve

-- 
Steve Purcell, Pythangelist
Get testing at http://pyunit.sourceforge.net/
Any opinions expressed herein are my own and not necessarily those of Yahoo



From js.graham at home.com  Sun Mar 11 13:49:56 2001
From: js.graham at home.com (Steve Graham)
Date: Sun, 11 Mar 2001 18:49:56 GMT
Subject: Einstein's Riddle
Message-ID: <oNPq6.396644$ge4.138479566@news2.rdc2.tx.home.com>

Who would be interested in using his/her brain (and his computer) to solve
the following?


Steve Graham

===

Einstein's Riddle

Albert Einstein wrote this riddle this century [ed. 20th century].  He said
98% of the world could not solve it.

There are 5 houses in 5 different colors.  In each house lives a person with
a different nationality.  The 5 owners drink a certain type of beverage,
smoke a certain brand of cigar, and keep a certain pet.  No owners
have the same pet, smoke the same brand of cigar or drink the same beverage.

The question is: "Who owns the fish?"

Hints:

The Brit lives in the red house.

The Sweed keeps dogs as pets.

The Dane drinks tea.

The green house is on the left of the white house.

The green house's owner drinks coffee.

The person who smokes Pall Mall rears birds.

The owner of the yellow house smokes Dunhill.

The man living in the center house drinks milk.

The Norwegian lives in the first house.

The man who smokes Blends lives next to the one who keeps cats.

The man who keeps the horse lives next to the man who smokes Dunhill.

The owner who smokes Bluemasters drinks beer.

The German smokes Prince.

The Norwegian lives next to the blue house.

The man who smokes Blends has a neighbor who drinks water.











From jeremy at alum.mit.edu  Mon Mar 19 19:13:34 2001
From: jeremy at alum.mit.edu (Jeremy Hylton)
Date: Mon, 19 Mar 2001 19:13:34 -0500 (EST)
Subject: Why "from __future__" stinks; a counter-offer
In-Reply-To: <memo.20010319213204.8771C@digitig.compulink.co.uk>
References: <mailman.985029808.7088.python-list@python.org>
	<memo.20010319213204.8771C@digitig.compulink.co.uk>
Message-ID: <15030.41134.443827.108283@w221.z064000254.bwi-md.dsl.cnc.net>

>>>>> "TR" == Tim Rowe <digitig at cix.co.uk> writes:
  
  Jeremy:
  >Aahz:
  >> Officially, complaints need only be posted to c.l.py.  It's not
  >> clear to me at this point what the True Practice is, but sending
  >> e-mail to the PEP champion is supposed to work best of all.  (Not
  >> all PEP creators are on python-dev.)
  >
  > Discussion on comp.lang.python is fine, although it would usually
  > help to cc the PEP author.  If there are legitimate complaints
  > posted to c.l.py and the PEP author ignores them, I expect
  > someone will cry foul.  Tim reads c.l.py as devotedly as anyone.
  > Barry and I try to keep up, too.

  TR> But there's part of the problem; egoless PEP writing is as hard
  TR> as egoless programming, and having come up with a solution the
  TR> author of that solution is bound to have some emotional
  TR> involvement in that solution.  There have been (rightly IMHO)
  TR> howls of protest about __future__, but Tim has said it's ok for
  TR> him, routinely dismisses all protests, presumably because he
  TR> genuinely doesn't understand what our problem is, and he tells
  TR> us it's going in anyway. Can you see why we're not all convinced
  TR> that the PEP process is any help to us?

It's unlikely that Tim dismisses people because he doesn't understand
them.  It's likely that people complain and expect that their
complaints will cause the PEP to change.  Not likely.  We all agree
that overloading import is ugly, it's just less ugly than the
alternatives.

If you think it's ugly and you think you have a better solution, then
you can write a PEP, too.  None of the people who complained took the
time to write a PEP describing their solution.  If you want to
convince Guido that __future__ should disappear, you've got to
convince him that there is a better solution.

Jeremy



From aahz at panix.com  Tue Mar 20 23:33:44 2001
From: aahz at panix.com (Aahz Maruch)
Date: 20 Mar 2001 20:33:44 -0800
Subject: Yet Another PEP:  Query Protocol Interface or __query__
References: <Pine.LNX.4.21.0103201837570.21343-100000@clarkevans.com> <99917n$2jf$1@panix6.panix.com> <slrn9bgb7f.kam.neelk@alum.mit.edu>
Message-ID: <999av8$it1$1@panix6.panix.com>

In article <slrn9bgb7f.kam.neelk at alum.mit.edu>,
Neelakantan Krishnaswami <neelk at alum.mit.edu> wrote:
>On 20 Mar 2001 17:47:35 -0800, Aahz Maruch <aahz at panix.com> wrote:
>>In article <Pine.LNX.4.21.0103201837570.21343-100000 at clarkevans.com>,
>>Clark C. Evans <cce at clarkevans.com> wrote:
>>>
>>>    A new built-in method, __query__ is proposed.  This method has
>>>    a single argument, an interface protocol identifier, and either
>>>    returns an object supporting the given protocol, or throws an
>>>    unknown exception. [...]
>>
>> [...]
>>
>> Most importantly, how does one find out what interfaces an object *does*
>> support?
>
>Why is this important?
>
>If you are writing some client code, presumably all you care about is
>that the object you receive supports certain interfaces X, Y and Z
>that you are using. It wouldn't seem to matter at all what -other-
>interfaces it supports.

I had a horrified reaction to the idea of looping over X, Y, and Z
wrapped around a try/except.  Because Clark is now changing the
exception to None, my point is considerably less important.
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

"I won't accept a model of the universe in which free will, omniscient
gods, and atheism are simultaneously true."  -- M


From rickp at telocity.com  Mon Mar 19 23:07:34 2001
From: rickp at telocity.com (Rick Pasotto)
Date: Tue, 20 Mar 2001 04:07:34 GMT
Subject: Tkinter : resizing Text widget with main window
References: <3AB11827.469F6C0C@mega-nerd.com>
Message-ID: <slrn9bdls6.noo.rickp@tc.niof.net>

On Thu, 15 Mar 2001 19:38:12 GMT in comp.lang.python, Erik de Castro Lopo wrote:
> Hi all,
> 
> How does one force a Tkinter.Text widget to be resized whenever 
> the main window is resized?

t=Text(width=30,height=12) # sets the initial size
t.pack(expand=1,fill='both')

-- 
"So long as the power-seekers clung to the basic premises of the
welfare state, holding need as the criterion of rewards, logic forced
them, step by step, to champion the interests of the less & less
productive... until they reached the ultimate dead end of turning from
the role of champions of 'honest toil' to the role of champions of
open parasitism, parasitism on principle, parasitism as a 'right'."
		-- Ayn Rand "A PreView" _The Ayn Rand Letter_
		   Rick Pasotto email: rickp at telocity.com


From kevin_cazabon at hotmail.nospamplease!.com  Wed Mar 28 23:00:09 2001
From: kevin_cazabon at hotmail.nospamplease!.com (Kevin Cazabon)
Date: Wed, 28 Mar 2001 21:00:09 -0700
Subject: PIL-resize, aspect ratio
References: <3AC062B5.D5D0D471@hack.org>
Message-ID: <#iL1tVAuAHA.210@typhoon.i3s.net>

Unfortunately, the only way to "automatically" keep the aspect ratio is to
use the im.thumbnail(maxX, maxY) feature.  However, this uses NEAREST
resizing only.

Otherwise, you have to figure it out yourself, unfortunately.

However, I've asked Fred to add a "im.fit" method to PIL that I wrote a
while back that lets you resize, crop, etc. all at once.  He did at one
point say he was going to roll it into 1.1, however it would seem that it's
not in yet.

I've posted a zip of that stand-alone function, along with a separate module
that lets you rotate images an arbitrary amount, automatically adjusting the
canvas size to fit the entire rotated image (instead of cropping the output
to the input canvas size).

http://www.cazabon.com/python/downloads/PyImTools.zip


Fred:  any plans to incorporate these still?

Kevin Cazabon
kevin at cazabon.removethis.com





"Rikard Bosnjakovic" <bos at hack.org> wrote in message
news:3AC062B5.D5D0D471 at hack.org...
| Image.resize() takes a tuple with x,y when I resize, but how do I keep
| the aspect ratio? Do I need to calculate it on my own?
|
|
| --
| Rikard Bosnjakovic - http://a214.ryd.student.liu.se/cv/ - ICQ: 1158217
|
| Anyone sending unwanted advertising e-mail to my address will be
| charged $250 for network traffic and computing time. By extracting my
| address from this message or its header, you agree to these terms.




From moshez at zadka.site.co.il  Tue Mar 13 01:20:47 2001
From: moshez at zadka.site.co.il (Moshe Zadka)
Date: Tue, 13 Mar 2001 08:20:47 +0200
Subject: New list for Discussion of Python's Numeric Model
Message-ID: <E14ciAp-0005dJ-00@darjeeling>

After discussions in IPC9 one of the decisions was to set up a mailing
list for discussion of the numeric model of Python.

Subscribe here:

    http://lists.sourceforge.net/lists/listinfo/python-numerics

Or here:

    python-numerics-request at lists.sourceforge.net

I will post my PEPs there as soon as an initial checkin is completed.
Please direct all further numeric model discussion there.
-- 
"I'll be ex-DPL soon anyway so I'm        |LUKE: Is Perl better than Python?
looking for someplace else to grab power."|YODA: No...no... no. Quicker,
   -- Wichert Akkerman (on debian-private)|      easier, more seductive.
For public key, finger moshez at debian.org  |http://www.{python,debian,gnu}.org



From embed at geocities.com  Thu Mar 15 08:46:10 2001
From: embed at geocities.com (Warren Postma)
Date: Thu, 15 Mar 2001 08:46:10 -0500
Subject: Einstein's Riddle
References: <mailman.984369618.12493.python-list@python.org> <3AACBB41.5854666A@yahoo.com> <3AAF1885.FF28CC21@yahoo.com> <3AAF6352.726046A2@yahoo.com>
Message-ID: <XK3s6.9935$TW.39680@tor-nn1.netcom.ca>

"Steve" <smnordby at yahoo.com> wrote in message
news:3AAF6352.726046A2 at yahoo.com...
> Reliability is an indication of whether a measurement is repeatable:
> Does it give consistent results under the same conditions?

Ahh yes, because consistenly Wrong, or backwards, in certain circumstances,
is Perfectly Reliable.  The system will always fail, silently, if x = 5,  3
< y < 6, and it's tuesday. Every time.

;-)

Warren





From peter.stoehr at fh-hof.de  Wed Mar 28 08:35:20 2001
From: peter.stoehr at fh-hof.de (Peter =?iso-8859-1?Q?St=F6hr?=)
Date: Wed, 28 Mar 2001 15:35:20 +0200
Subject: Converting an integer base 10 to a binary number
Message-ID: <3AC1E898.D5929C44@fh-hof.de>

Hi out there,

I know the int(x[,radix]) function that converts a string or a number
(for a give radix) to an integer. Is there a built-in function that
provides a conversion the other way round?
Something like
    bin(32) = "10000"

Thanks in advance
    Peter
-- 
Fachhochschule Hof
Prof. Dr. Peter St?hr				Tel: 09281 - 409 485
Alfons-Goppel-Platz 1				Fax: 09281 - 409 477
95028 Hof / Germany


From ws-news at gmx.at  Fri Mar  9 09:41:11 2001
From: ws-news at gmx.at (Werner Schiendl)
Date: Fri, 9 Mar 2001 15:41:11 +0100
Subject: Foot in yourself shoot [was: learning C]
References: <mailman.983972829.23481.python-list@python.org> <983975751.18833@newsmaster-04.atnet.at> <3AA82422.53026D3D@cosc.canterbury.ac.nz> <20010308221519.A4367@harmony.cs.rit.edu> <mailman.984113052.19927.python-list@python.org> <984135226.858948@newsmaster-04.atnet.at> <mailman.984146352.13033.python-list@python.org> <eL5q6.45993$ML1.2000092@e420r-atl2.usenetserver.com>
Message-ID: <984149152.354104@newsmaster-04.atnet.at>

I particularly love the Unix example :-)

Steve Holden <sholden at holdenweb.com> wrote in message
news:eL5q6.45993$ML1.2000092 at e420r-atl2.usenetserver.com...
> "Chris Gonnerman" <chris.gonnerman at usa.net> wrote in message
> news:mailman.984146352.13033.python-list at python.org...
> > ----- Original Message -----
> > From: "Werner Schiendl" <ws-news at gmx.at>
> > Subject: Re: learning C
> [ ... ]
> > > Why not have a look at all of it and use what's suited for the
> application
> > > best?
> >
> > Like I said, been there, done that, couldn't afford the T-shirt.
Rewiring
> > my
> > brain for C++ is just too hard.  I'm glad that it works for you, and
many
> > others,
> > but there are also many like me that can't see why you are pushing so
> hard.
> > I
> > guess you just can't see our point of view, though I think I see yours
> just
> > fine.
> >
>
> I have for a long time been a pragmatist when it comes to language
choice --
> use the most appropriate language for a particular task, assuming the
> learning curve for a new language doesn't make it easier to use an
existing
> but somewhat less-appropriate one.
>
> The computing community has known for a long time that each language is
> better for some tasks than others, vide
>
>         http://www.cs.bgu.ac.il/~omri/Humor/shoot.html
>
> (quoted below). As with much good humo(u)r, there are real points of
> comparison available for those who care to look for them. I think the C++
> quote is especially relevant. To which I would only add:
>
> Python: After half an hour's design work you realise that foot is simply a
> special case of limb.extremity, so you use a standard library to solve
your
> problem!
>
> regards
>  Steve
>
> """
> The proliferation of modern programming languages (all of which seem to
have
> stolen countless features from one another) sometimes makes it difficult
to
> remember what language you're currently using. This handy reference is
> offered as a public service to help programmers who find themselves in
such
> a dilemma.
>
> TASK: Shoot yourself in the foot.
> C: You shoot yourself in the foot.
>
> C++: You accidentally create a dozen instances of yourself and shoot them
> all in the foot. Providing emergency medical assistance is impossible
since
> you can't tell which are bitwise copies and which are just pointing at
> others and saying, "That's me, over there."
> """
>
> [Many other languages omitted: follow the link if these are new to you].
>
>
>




From bsass at freenet.edmonton.ab.ca  Thu Mar  1 15:44:53 2001
From: bsass at freenet.edmonton.ab.ca (Bruce Sass)
Date: Thu, 1 Mar 2001 13:44:53 -0700 (MST)
Subject: [Distutils] Re: CPAN functionality for python - requirements
In-Reply-To: <20010228165449.A1781@tummy.com>
Message-ID: <Pine.LNX.4.33.0103011318280.13134-100000@bms>

On Wed, 28 Feb 2001, Sean Reifschneider wrote:

> On Wed, Feb 28, 2001 at 10:13:51AM +0100, Jack Jansen wrote:
> >Well, one of the things I'd like to do is integrate all this in MacPython IDE,
> >and the same probably holds for PythonWin and Idle. Then the "what do you
> >have" is pretty useful for building menus and such.
>
> I anticipate having a web-based interface to the database as well.  I'm
> sure that we can get the CGI to let the user do that sort of search, but
> at the moment it's restricted so that the user has to specify the name
> of the package they're looking for.  I figured dumping gigabytes across
> some user's line wouldn't be optimal.  ;-)

Gigabytes?

~$ avail=/var/lib/dpkg/available
~$ du -h $avail
3.4M    /var/lib/dpkg/available
~$ grep Package: $avail | wc -l
   4698

An entry in Debian's available DB looks like this...

Package: python-zlib
Priority: optional
Section: interpreters
Installed-Size: 80
Maintainer: Gregor Hoffleit <flight at debian.org>
Architecture: i386
Source: python
Version: 1.5.2-10
Depends: python-base (= 1.5.2-10), libc6 (>= 2.1.2), libz1
Filename: dists/potato/main/binary-i386/interpreters/python-zlib_1.5.2-10.deb
Size: 16578
MD5sum: 7226b0d0edf21374c1b09295c0d9497b
Description: A compression module for Python using zlib.
 A compression module for Python using the zlib library. zlib
 implements the deflate compression method used in gzip and
 pkzip.

...megabytes perhaps.  Of course if someone is querying for what is
available they should be expecting a lot of information.

Killing two birds with one stone...
It's not quite as bad as you think.
If the server can spit out a file containing the above info for each
package, you are partially apt-get-able; the other part is returning
the properly formatted package when you get a request for the
Filename: URI.


- Bruce




From whisper at oz.net  Sun Mar 11 14:02:45 2001
From: whisper at oz.net (Dave LeBlanc)
Date: 11 Mar 2001 19:02:45 GMT
Subject: Questions about no more nested imports
Message-ID: <98gi4l$n9n$0@216.39.170.247>

I'm a bit confused by this. 

It appears that if this is enforced, one can no longer dynamically
load extensions/whatever dynamically and all imports must be done at
the top of a module prior to anticipated use - whether used or not. If
this is true, then load times for python apps could go up drastically.

The other thing I can't figure out is how to dynamically load
extensions based on their discovery at run time - for example, a
plugins directory whose content might change from one invocation to
another. Since, as I understand it, I have to do all imports at the
top of the file, how am I going to discover and import such new
extensions without executing any code to do so?

No conditional loading either I guess? How about imports needed by "if
__main__" but not needed if used  as a libarary?

Dave LeBlanc


From bpurvy at packeteer.com  Mon Mar 12 18:55:45 2001
From: bpurvy at packeteer.com (Bob Purvy)
Date: Mon, 12 Mar 2001 15:55:45 -0800
Subject: (Port-induced?) memory leak in 2.0
Message-ID: <3AAD6201.8F549AD7@packeteer.com>

I've gotten 2.0 sort of working on pSOS, our embedded "operating
system."  Right now I'm seeing memory leaks after a call to Py_Main()
(so we can have a "enter Python shell" command in our command-line
interface), which I've verified in the debugger calls Py_Initialize()
once and Py_Finalize() twice

    (but the second call doesn't do anything since the initilize flag is
off).

I've replaced all the mem calls with kmalloc() / krealloc() / kfree(),
so I could check for leaks.  There seems to be a ton.

  i Type             Blocks  Allocs   Fails   Bytes HiWater   Limit
Percent
 62 python              174    1365       0   21320  139168 1048576
28.19

so there is 21K worth of stuff left over after the Py_Finalize().

I know this is almost certainly going to be something caused by my port,
but if the following list of free nodes gives anyone a hint about what
might be wrong, I'd sure appreciate it if you let me know.



data len: 988
0x03becaa0   00000000 03be8ad0 00000000 45455246   ............FREE
0x03becab0   45455246 03becaa4 45455246 45455246   FREE....FREEFREE
0x03becac0   03becab0 45455246 45455246 03becabc   ....FREEFREE....
0x03becad0   45455246 45455246 03becac8 45455246   FREEFREE....FREE
0x03becae0   45455246 03becad4 45455246 45455246   FREE....FREEFREE
0x03becaf0   03becae0 45455246 45455246 03becaec   ....FREEFREE....
0x03becb00   45455246 45455246 03becaf8 45455246   FREEFREE....FREE
0x03becb10   45455246 03becb04 45455246 45455246   FREE....FREEFREE
0x03becb20   03becb10 45455246 45455246 03becb1c   ....FREEFREE....
0x03becb30   45455246 45455246 03becb28 45455246   FREEFREE(...FREE
0x03becb40   45455246 03becb34 45455246 45455246   FREE4...FREEFREE
0x03becb50   03becb40 45455246 45455246 03becb4c   @...FREEFREEL...
0x03becb60   45455246 45455246 03becb58 45455246   FREEFREEX...FREE
0x03becb70   45455246 03becb64 45455246 45455246   FREEd...FREEFREE
0x03becb80   03becb70 45455246 45455246 03becb7c   p...FREEFREE|...
0x03becb90   45455246 45455246 03becb88 45455246   FREEFREE....FREE
snip

data len: 97
0x03bf5660   be49040a 03bfd4c0 03bfd498 c8f2da46   ..I.........F...
0x03bf5670   03bfd414 03bfbfe8 eccd053d 03bf7b6c   ........=...l{..
0x03bf5680   03bfd21c 00000000 00000000 00000000   ................
0x03bf5690   00000000 00000000 00000000 4f7b4c82   .............L{O
0x03bf56a0   03bfd174 03bfd120 502a8d9e 03bfd54c   t... .....*PL...
0x03bf56b0   03bfd524 00000000 00000000 00000000   $...............
0x03bf56c0   45455246 45455246 45455246 45455246   F

data len: 193
0x03bf5844   4771c4af 03bffc64 00618c08 502a8d9e   ..qGd.....a...*P
0x03bf5854   03bfd54c 03bffb1c eccd053d 03bf7b6c   L.......=...l{..
0x03bf5864   03bff914 00000000 00000000 00000000   ................
0x03bf5874   00000000 00000000 00000000 be49040a   ..............I.
0x03bf5884   03bfd4c0 03bffb50 4dafdc09 03bffd00   ....P......M....
0x03bf5894   00618c08 7a3ee938 03bffb9c 03bfeb8c   ..a.8.>z........
0x03bf58a4   51a7bed7 03bffccc 00618c08 00000000   ...Q......a.....
0x03bf58b4   00000000 00000000 3884f2b5 03bffc98   ...........8....
0x03bf58c4   00618c08 00000000 00000000 00000000   ..a.............
0x03bf58d4   00000000 00000000 00000000 4f7b4c82   .............L{O
0x03bf58e4   03bfd174 03bff9f0 00000000 00000000   t...............
0x03bf58f4   00000000 00000000 00000000 00000000   ................
0x03bf5904   45455246 45455246 45455246 45455246   F

data len: 12
0x03bf6b44   00000001 00618b1c 03bf7a9c 45455246   ......a..z..

data len: 35
0x03bf7528   0000002d 0061a328 0000000b ffffffff   -...(.a.........
0x03bf7538   00000000 6d75643c 6b20796d 003e7965   ....<dummy key>.
0x03bf7548   45455246 45455246 45455246 45455246   FRE

data len: 70
0x03bf76d4   00000001 0061a328 0000002e ffffffff   ....(.a.........
0x03bf76e4   00000000 65736142 616c6320 66207373   ....Base class f
0x03bf76f4   6120726f 73206c6c 646e6174 20647261   or all standard
0x03bf7704   68747950 65206e6f 70656378 6e6f6974   Python exception
0x03bf7714   45002e73 45455246 45455246 45455246   s..EFR

data len: 25
0x03bf7884   00000003 0061a328 00000001 d4cb6ede   ....(.a......n..
0x03bf7894   03bf7884 4545005f 45455246 45455246   .x.._.EEF

data len: 33
0x03bf7940   00000003 0061a328 00000009 9ec16a0f   ....(.a......j..
0x03bf7950   03bf7940 5f706d63 74617473 45450065   @y..cmp_state.EE
0x03bf7960   45455246 45455246 45455246 45455246   F

data len: 44
0x03bf7994   03bf7a90 0061eef4 45455246 00000001   .z....a.FREE....
0x03bf79a4   0061891c 00000096 0000006a 00000100   ..a.....j.......
0x03bf79b4   0000011d 03bfa850 004842e0 45455246   ....P....BH.

data len: 36
0x03bf79e8   00000004 0061a328 0000000c b8b08ac8   ....(.a.........
0x03bf79f8   03bf79e8 75625f5f 69746c69 5f5f736e   .y..__builtins__
0x03bf7a08   45455200 45455246 45455246 45455246   .REE

data len: 44
0x03bf7a90   03bfb8b8 03bf7994 45455246 00000002   .....y..FREE....
0x03bf7aa0   0061891c 00000059 00000059 00000100   ..a.Y...Y.......
0x03bf7ab0   0000011d 03c02168 004842e0 45455246   ....h!...BH.

data len: 32
0x03bf7ae4   00000005 0061a328 00000008 3cfa8ed9   ....(.a........<
0x03bf7af4   03bf7ae4 616e5f5f 5f5f656d 45455200   .z..__name__.REE

data len: 34
0x03bf7b18   00000003 0061a328 0000000a 2e1f7e1b   ....(.a......~..
0x03bf7b28   03bf7b18 6e697369 6e617473 45006563   .{..isinstance.E
0x03bf7b38   45455246 45455246 45455246 45455246   FR

data len: 31
0x03bf7b6c   0000000a 0061a328 00000007 eccd053d   ....(.a.....=...
0x03bf7b7c   03bf7b6c 6f645f5f 005f5f63 45455246   l{..__doc__.FRE

................ many more




From kevin_cazabon at hotmail.nospamplease!.com  Thu Mar 22 23:29:50 2001
From: kevin_cazabon at hotmail.nospamplease!.com (Kevin Cazabon)
Date: Thu, 22 Mar 2001 21:29:50 -0700
Subject: Puzzling input problem
References: <3AB9449C.A1D14738@monitor.net>
Message-ID: <d5#fcH1sAHA.187@typhoon.i3s.net>

How about using this instead:

answer = raw_input("prompt string: ")




"Stephen R. Figgins" <fig at monitor.net> wrote in message
news:3AB9449C.A1D14738 at monitor.net...
| I ran into a puzzling problem combining print with raw_input().
| Normally I would pass the prompt string to raw_input, but instead I
| tried something like this:
|
|   print "prompt string: ",
|   answer = raw_input()
|   print answer
|
| When it prints answer, it is indented one space.  Why is raw_input()
| putting a space in there?  Is it a bug, a feature, an unintended side
| effect?  input() does this too.   I am currently using Active Python 2.0
| build 202.
|
| When I tried it with Jython 2.0, I didn't get the extra space.
|
|
| Stephen




From loewis at informatik.hu-berlin.de  Tue Mar 20 15:36:46 2001
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 20 Mar 2001 21:36:46 +0100
Subject: File Descriptor Modes
References: <mailman.985118007.7016.python-list@python.org>
Message-ID: <j4lmq0i3dt.fsf@informatik.hu-berlin.de>

"O'Rourke Clodagh-corour01" <corour01 at motorola.com> writes:

> Does anyone where I would find a summary of the different modes it
> is possible to open a file descriptor in?

In the description of open:

http://www.python.org/doc/current/lib/built-in-funcs.html

> Do I need to explicitly find the end of the file and begin writing
> there or is there a mode that does this?

Yes :-)

Martin


From francois.granger at free.fr  Fri Mar 30 17:51:04 2001
From: francois.granger at free.fr (=?ISO-8859-1?Q?Fran=E7ois_Granger?=)
Date: Fri, 30 Mar 2001 22:51:04 GMT
Subject: Converting an integer base 10 to a binary number
References: <3AC1E898.D5929C44@fh-hof.de>
Message-ID: <1er3z6w.13vo4oy1rq8rjmN%francois.granger@free.fr>

Peter St?hr <peter.stoehr at fh-hof.de> wrote:

> Hi out there,
> 
> I know the int(x[,radix]) function that converts a string or a number
> (for a give radix) to an integer. Is there a built-in function that
> provides a conversion the other way round?
> Something like
>     bin(32) = "10000"

Not fully polished, but works

# python
"""
create the binary rep of an integer as a string of '0' & '1'
20010302    v 1.0
"""
def toBin(x):
    a = ''
    y = x
    while x:
        x, y = x / 2, y % 2
        if y:
            a = '1' + a
        else:
            a = '0' + a
        y = x
    return a

if __name__ == '__main__'
    l = [1, 2, 3, 4, 8, 16, 32, 64, 128, 255, 512, 1024, 2048, 4096,
8192, 16384, 32768]
    for x in l:
        print toBin(x)


-- 
Une faq de frj: http://faq.jardin.free.fr/, pr?s de l'encyclop?die
http://nature.jardin.free.fr/
MacPython beginners http://francois.granger.free.fr/MacPython/faq/


From sheila at spamcop.net  Sat Mar 31 21:40:50 2001
From: sheila at spamcop.net (Sheila King)
Date: Sun, 01 Apr 2001 02:40:50 GMT
Subject: PythonLabs Team Moves to ActiveState
References: <mailman.986089142.10423.python-list@python.org> <9a62cg$3it52$1@ID-11957.news.dfncis.de> <3AC69552.7D6FAD6A@monitor.net>
Message-ID: <475dcto0ii2nmeoipd6rl74n6v92nd2rhl@4ax.com>

On Sat, 31 Mar 2001 18:41:22 -0800, "Stephen R. Figgins" <fig at monitor.net>
wrote in comp.lang.python in article <3AC69552.7D6FAD6A at monitor.net>:

:This was a good one!  I was halfway through it before I realized it was
:almost word for word the same as the DC announcement.  When I got to the
:part about braces I about fell out of seat laughing. 

I went to the website, just after reading the article from Guido, and noticed
the startling similarity, myself.

:Check out the X-Apology header.

X-Apology: To the wonderful people at ActiveState and DC...gotcha! ;-)

Ah, peace of mind at last. Boy, that was a good April Fool's. (It ain't april
yet, in California, though...)

--
Sheila King
http://www.thinkspot.net/sheila/
http://www.k12groups.org/

:-Stephen



From rikbo716 at student.liu.se  Sun Mar 25 17:39:46 2001
From: rikbo716 at student.liu.se (Rikard Bosnjakovic)
Date: Mon, 26 Mar 2001 00:39:46 +0200
Subject: Newbie needs help to find bug in his first Perl program
References: <3ABCC757.6C6F17CA@earthlink.net>
Message-ID: <3ABE73B2.CAFCA177@student.liu.se>

Ron Stephens wrote:

> Can anyone see what is wrong with this? I have several lines like this
> and they all give me error messages.

*What* error messages? I don't think anyone want to *guess* what kind of
error messages you get.

There's no format-specifier named "%y".


-- 
Rikard Bosnjakovic - http://a214.ryd.student.liu.se/cv/ - ICQ: 1158217

Anyone sending unwanted advertising e-mail to my address will be
charged $250 for network traffic and computing time. By extracting my
address from this message or its header, you agree to these terms.


From db3l at fitlinxx.com  Fri Mar 30 13:03:59 2001
From: db3l at fitlinxx.com (David Bolen)
Date: 30 Mar 2001 13:03:59 -0500
Subject: Should I close after popen??
References: <m38zm6r8ih.fsf@master.athome> <fhOs6.510$Fb.777424@typhoon2.ba-dsg.net> <mailman.985919149.7252.python-list@python.org>
Message-ID: <uofujnng0.fsf@ctwd0143.fitlinxx.com>

pinard at iro.umontreal.ca (=?iso-8859-1?q?Fran=E7ois?= Pinard) writes:

> Explicit closing is probably not always good practice, and is not necessarily
> safer.  I would completely agree with the above for someone who wants to
> write Jython, or write Python that could be quickly turned into Jython.
> But if you stick with Python, you can safely rely on the fact that a file
> is automagically close as soon as you do not refer to it anymore.

I'd probably change Python in that sentence to CPython, since it's an
implementation issue and not a language one (the language makes no
such guarantees).

> In fact, if I remember well, Tim Peters confirmed (on this list) that
> Python, the C implementation, is very not likely to change on this.
> 
> Code like:
> 
>     open(logfile, 'a').write('Done!\n')
> 
> or maybe:
> 
>     for line in open(logfile).readlines():
>         process(line)
> 
> is very legible, and would rather be uselessly encumbered with explicit
> closes and intermediate variables.  It is not worth writing these in the
> style of other languages where closes are explicitely needed.

I can agree on the second point, but would caution that on the first
point you have no guarantees that your data actually got to the
logfile without doing the close and possibly handling any errors
there.

Ok, you may say the odds of failure are low and depending on the
criticality of the data it may or may not matter, but it's all too
easy to forget that there's any risk at all when looking at the code,
so I think it bears repeating in these sorts of discussions.

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


From claird at starbase.neosoft.com  Tue Mar 13 17:05:10 2001
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 13 Mar 2001 16:05:10 -0600
Subject: Who's minister of propaganda this week?
Message-ID: <B98041329020E007.7DA84CAD9D0562AB.817BBC572F480869@lp.airnews.net>

Sure; Bruce "Mr. Java" Eckel says Python's his favorite language.
That was last week's news.  What would you think if the former
editor of the *C++ report* gave his
    prediction for this decade: Keep an
    eye on languages like Python, Ruby,
    and Smalltalk. They are likely to
    become extremely important.
    <URL: http://forums.itworld.com/webx?14 at 139.drRfa36vgId^0 at .ee6eed4/11 >
?

I heard you wiseacres in the back there, with the, "But Python
and Smalltalk *already* are extremely important" comment.  
-- 

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


From sholden at holdenweb.com  Wed Mar  7 22:01:26 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Wed, 7 Mar 2001 22:01:26 -0500
Subject: class object access in class definition?
References: <Pine.LNX.4.21.0103061556510.2015-100000@zero> <slrn9ac2sa.ah1.scarblac@pino.selwerd.nl> <Pine.LNX.4.21.0103071835560.6130-100000@zero>
Message-ID: <CFCp6.30994$ML1.1565785@e420r-atl2.usenetserver.com>

"Lars Damerow" <lars at pixar.com> wrote in message
news:Pine.LNX.4.21.0103071835560.6130-100000 at zero...
> I figured that the my problem stems from when Python binds the class name
to
> the scope--when it encounters the "class" keyword, or when the class
definition
> completes. It seems like it happens when the class definition completes.
>
> What I'm trying to do is automatically register the derived classes of a
base
> class at compile time, like this:
>
> derivedClasses = []
>
> class BaseClass:
>     pass
>
> class DerivedClass(BaseClass):
>     derivedClasses.append(DerivedClass)
>
> This doesn't work because DerivedClass doesn't seem to be bound at the
time of
> the append() call. I can put the append() call in the DerivedClass's
> constructor, but that requires an instance to be created before the class
can
> be registered.
>
> If this isn't possible, it's no big deal; its goal is to avoid duplicating
> information in the code, nothing more.
>

Well, if you're prepared to be pragmatic about it, what's wrong with:

derivedClasses = []

class BaseClass:
    pass

class DerivedClass(BaseClass):
    pass

derivedClasses.append(DerivedClass)

Not elegant, but practical.

regards
 Steve





From paulp at ActiveState.com  Mon Mar 26 19:46:08 2001
From: paulp at ActiveState.com (Paul Prescod)
Date: Mon, 26 Mar 2001 16:46:08 -0800
Subject: warrior geeks
References: <B43D149A9AB2D411971300B0D03D7E8B0FBFE0@natasha.auslabs.avaya.com>
Message-ID: <3ABFE2D0.C82A0ADB@ActiveState.com>

"Delaney, Timothy" wrote:
> 
>...
> 
> This is a Good Thing(TM). I think a lot of people in the US don't realise
> that the "right to bear arms" was devised when "arms" were single-shot
> weapons and the US was attempting to secede from England. A lot of the
> attitude of Americans really scares me. I'd much rather be here with our
> multitude of venomous creatures than in the US. We've got anti-venine for
> almost every venom in almost every hospital in Australia. There's no
> anti-venine for a gunshot wound.

But the question is: Which of Perl versus Python is more appropriate for
shooting yourself in the foot.

At-least-that-flame-war-is-almost-on-topic 'ly yrs

-- 
Take a recipe. Leave a recipe.  
Python Cookbook!  http://www.activestate.com/pythoncookbook



From tjreedy at udel.edu  Mon Mar 19 10:22:58 2001
From: tjreedy at udel.edu (Terry Reedy)
Date: Mon, 19 Mar 2001 10:22:58 -0500
Subject: newbie - infinite loop
References: <985012513.18971.0.nnrp-12.c1c3e154@news.demon.co.uk>
Message-ID: <99589a$48$1@news.udel.edu>

Use a dictionary, not a list, to compile a 'list' of unique words.

words = {}
for word in text:
  if not hasattr(words, word):
    words[word] = None
wordlist = words.keys()

With only a little more effort and use of words.setdefault(), you can
easily count occurences.
Code recently posted.

Terry J. Reedy




From scarblac at pino.selwerd.nl  Thu Mar  8 02:32:22 2001
From: scarblac at pino.selwerd.nl (Remco Gerlich)
Date: 8 Mar 2001 07:32:22 GMT
Subject: UserDict and items()
References: <98656v$cud$1@news.rrz.Uni-Koeln.DE>
Message-ID: <slrn9aed8i.cg1.scarblac@pino.selwerd.nl>

Oliver Hofmann <a2619725 at smail.uni-koeln.de> wrote in comp.lang.python:
> Is there a more elegant way to handle items(), values() etc than
> overwriting each function?
> 
> One way would be to set data[key] to the required value during
> init(); however I'd like to postpone fetching data from the database
> until the time it is actually requested.
> 
> Any help would be appreciated,

It's early morning, but it seems to me that when items() is called, that is
when all the entries are actually requested, so that is the latest point at
which you can set all the data[key] things from the database values. Use
some sort of flag so that you only do this the first time items() is called?
(this won't work well when the database values change, but it seems they
don't since you considered the possibility of doing it in __init__).

Or maybe the intended behavior is to let items() returns only those values
that have been asked for already. You could make __getitem__ put the values
into the dict whenever something special is asked for, ie

   def __getitem__(self, key):
      if key == "special" and not self.data.has_key(key):
         self.data[key] = DatabaseCall(key)
      return self.data[key]

Of course to make that 'if' work you shouldn't initialize that value (why do
that anyway, you handle it with getitem).

The cuteness introduced in later Python versions would allow that to be:

   def __getitem__(self, key):
      return self.data.setdefault(key, DatabaseCall(key))

Ie, look up everything in the database that's not in the dictionary yet, put
it in, and return what's in the dict.

The last option is not to use items() :-)

-- 
Remco Gerlich


From xyzmats at laplaza.org  Sun Mar  4 20:59:15 2001
From: xyzmats at laplaza.org (Mats Wichmann)
Date: Mon, 05 Mar 2001 01:59:15 GMT
Subject: [Python-Dev] Re: Bug fix releases
References: <mailman.983681106.14778.python-list@python.org>
Message-ID: <3aa2e9c3.210139114@news.laplaza.org>

Whew...

Let me shift out of lurk mode and put in a couple of comments.

I think on the whole a large and varied user community such as
Python's would be well served by having the /possibility/ of
maintaining several release lines - with bug fixes and possibly minor
feature additions (controversial) being issued for an older line.

Some folks DO clearly want bug fixes without having to do a major
upgrade.  Sometimes you're kind of stuck: you're delivering an
application that's well tested on one version and the risk of
switching to another version just to pick up a fix for one annoying
bug is not worth it.  Sometimes there's a good reason why a fix can't
be backported into an older Python version, but often there isn't,
other than time/manpower/motivation/etc.  The rest of us should be
able to supply that if we really want something.

The main preparatory step happened when the source went into a public
CVS repository.  Now what's needed is enough demand == submitting the
fixes, and the maintainer who will make it happen.

I can't see that this will actually happen until there's a champion
who will handle the "boring stuff" - that is, babysit the older
released line (we don't all think that's boring work, but what the
heck).  If I can point at the Linux kernel development (don't anybody
gack now), Linus handles the "interesting stuff" - the latest branch;
when it reaches a certain degree of stability he hands over to Alan
Cox.  Did Aahz just volunteer to be our Alan Cox?

I think a little historical check is also in order:  the current state
may not be indicative of the future.  The licensing issue/GPL dispute
came at a time when record numbers of folks were using Python because
they got it "for free" with Linux distributions such as Red Hat and
Mandrake, who made it hard to upgrade by not packaging up new versions
of Python in the RPM format (which, of course, users of those systems
are told is "the" way to install software.  Not everybody who uses
Linux today is interested in rebuilding from source.  To insert a
subtle dig, the source tarball has tended to autoconfigure a
not-especially-useful configuration which you have to tweak a fair bit
in the modules directory. (Didn't I hear that 2.1 is taking a more
automated approach to working out what's supportable on the target?
Yay!  My recent Solaris/x86 build was a bit, er, painful).

Not to say that the rustling all comes from the Linux world.

On the whole, things have been remarkably compatible, but the long
life of 1.5.2 caused an artifical strange situation.


Do we need to limit to "no more than one bugfix release"?  Do we need
to have "official blessing from PythonLabs"?  For me, the answer is no
to both.  If you need a fix, you need a fix.  Having problems, if they
appear, taken care of in a timely manner isn't going to hurt Python's
reputation. If there's a maintainer who is trusted, known to be
careful about what patches are accepted, insures that the right test
cycle has happened before it's declared final, I don't think most
people would have a problem with it.  Such a maintainer isn't going to
let Python slip into a mode where releases have a dozen incremental
bugfix dot-releases (not that the high quality of Python seems to
require anything like that anyway). Feel free to disagree.


-- Mats Wichmann



From fdrake at cj42289-a.reston1.va.home.com  Thu Mar 29 08:01:26 2001
From: fdrake at cj42289-a.reston1.va.home.com (Fred Drake)
Date: Thu, 29 Mar 2001 08:01:26 -0500 (EST)
Subject: [development doc updates]
Message-ID: <20010329130126.C3EED2888E@cj42289-a.reston1.va.home.com>

The development version of the documentation has been updated:

	http://python.sourceforge.net/devel-docs/


For Peter Funk:  Removed space between function/method/class names and
their parameter lists for easier cut & paste.  This is a *tentative*
change; feedback is appreciated at python-docs at python.org.

Also added some new information on integrating with the cycle detector
and some additional C APIs introduced in Python 2.1 (PyObject_IsInstance(),
PyObject_IsSubclass()).




From aleaxit at yahoo.com  Wed Mar 14 18:14:42 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 15 Mar 2001 00:14:42 +0100
Subject: Who's minister of propaganda this week?
References: <B98041329020E007.7DA84CAD9D0562AB.817BBC572F480869@lp.airnews.net> <98nu1m$12e@dispatch.concentric.net> <660ACCA1E4816F7E.C359CDF4168EF3FB.3B6B05538D619770@lp.airnews.net> <3AAFA13D.76E32C3F@divalsim.it> <98oa4o01i4v@news2.newsguy.com> <3AAFBA5D.63AAF897@san.rr.com>
Message-ID: <98otur1nnf@news1.newsguy.com>

"Darren New" <dnew at san.rr.com> wrote in message
news:3AAFBA5D.63AAF897 at san.rr.com...
> Alex Martelli wrote:
> > So, all the compile-time checking is buying is catching (a small
> > subset of) the errors that would be caught in testing anyway, a
> > little bit earlier (thus, a little bit cheaper)
>
> Actually, if you do the XP thing, you write the tests before the code, so
> you're actually catching the type error before you even have something
that
> compiles. ;-)

No, because you catch the error when you *run* the tests, not
when you *write* them.  So, the error is indeed caught a bit
later than if the compiler did it -- could be a fraction of a microsecond,
which makes it irrelevant, or half an hour or more, from the time
you're writing the code to when you're testing it -- and it _is_ a
little bit cheaper to fix an error a minute after you've made it,
than half an hour later, since the context is marginally fresher
in your mind; I think the overall impact on coding productivity
can well come up to several percent points -- say 5%, ballpark.

5% is not 0% (in certain fields, a 5% competitive advantage is
easily large enough to make or break a firm), but it surely is no
panacea.  So, the _cost_ of moving checks ahead as much as
possible had better be kept very low.


Alex





From will at nospam.misconception.org.uk  Sun Mar 18 15:21:20 2001
From: will at nospam.misconception.org.uk (Will Newton)
Date: Sun, 18 Mar 2001 20:21:20 +0000
Subject: Deleting dictionary items
References: <98ug6n$q9d$1@news6.svr.pol.co.uk> <98vs4p$k9e$1@slb7.atl.mindspring.net>
Message-ID: <993570$71$1@newsg2.svr.pol.co.uk>

Andreas Jung wrote:

> #                What td.data ?  I assume td.data is your dictionary ?
> #                 item is now a tuple (x1,x2,x3...)
> 
>>                         if item[attr] != value:
> 
> #                  you can access the tuples elements by their position
> using the [] operator
> #                  e.g. item[0] gives x1, item[1] gives x2 ...
> #
>>                                 del td.data[item]
> 
> #                  This should work

This makes me think, would it be a sane way of doing things to use a list 
comprehension to modify a dictionary?

(I can't use it now, as I am stuck with 1.5.2 for the foreseeable future)

e.g.

filter(dict, lambda key: key == value)

or in my case a little more complex:

filter(dict, lambda key: key[attr] == value)

Is this possible? Desireable? Too stupid to contemplate?
 



From thomas at cintra.no  Mon Mar 12 08:05:52 2001
From: thomas at cintra.no (Thomas Weholt)
Date: Mon, 12 Mar 2001 14:05:52 +0100
Subject: Easiest way to get Berkley DB 3.x running on Windows
Message-ID: <SQ3r6.1961$NY6.31072@news3.oke.nextra.no>

Hi,

Do I have to compile Berkley DB to get an updated bsddb-module for Python on
win32?
I want to try the latest version available with python-bindings, cuz there
seems to be a memory leak
and a problem with deadloc(k)s in the one in the current Python lib.

Thanks for any pointers.

Thomas




From tdelaney at avaya.com  Wed Mar 21 18:38:07 2001
From: tdelaney at avaya.com (Delaney, Timothy)
Date: Thu, 22 Mar 2001 10:38:07 +1100
Subject: Yet Another PEP:  Query Protocol Interface or __query__
Message-ID: <B43D149A9AB2D411971300B0D03D7E8B0FBFBE@natasha.auslabs.avaya.com>

I think Huaiyu Zhu and I are thinking along very similar lines.

> Yes.  Very insightful.  It is easy to fall into this line of 
> thinking and not see the the traps.  I was heasitent when 
> I put in the "base class" stuff in the update to the PEP
> being promoted.... it just seemed expected.  Perhaps it
> isn't the right thing to do.  Once you have classes, 
> people like to treat class inheritance as interface
> inheritance... 
> 
> The rest of your post resonates deeply.  Is there some
> way to get a "simple" __query__ mechanism working 
> without having the type system in place?  

I think the major trap would be going with a system which did not have
sufficient expressiveness. One of the problems with passing a base class
object (or class instance) to __adapt__() is that there is no standard way
to specify requirements other than "interface has these methods".

In interface descriptor class is much more general, and much more easily
extended. For example, the first iteration could simply be used to add
method requirements. Then we could add method parameter and return type
signatures. Then pre- and post- conditions (such as parameter #3 <= 5). And
so on.

Huaiyu (I hope that's your given name - can't always tell with other
cultures whether the family name is first or last ;) appears to have gone
some way with what I envisage, but not all the way. I am so tempted to just
drop what I'm working on and get into this ...

An interface descriptor class itself would have to meet the interface
requirements for interface descriptors or it would be rejected by the
builtin adapt() function ;) I would actually propose a builtin "interface"
base class which contains all the required functionality from which specific
interface descriptor classes could be subclassed and/or instantiated. Note
that this would be a backwards-compatible change - it is simply adding a
name to the last namespace searched - likewise with the adapt() builtin
function.

I think the adapt stuff would work as follows (incomplete - doesn't really
do anything). Note: while working on this I discovered that both an
__adapt__() and __query__() method would be useful ... the __query__()
method could be used by a class to verify that it meets a single
requirement. Second note: This was coded up in about 15 minutes so there are
likely flaws.

class __requirement:
	""""""
	
	def __init__(self):
		""""""

	def verify (self, adaptor, __query__=None):
		""""""

		#	Allowed to pass in query to speed things up.
		#	If so, it *must* be the adaptor.__query__() method

		if not __query__:
			query = getattr(adaptor, '__query__', None)

		#	The __query__() method can return one of 3 types of
		#	values.
		#		
		#		True (e.g. 1): The object *definitely* does
		#		implement this requirement and should not be
		#		checked further
		#
		#		False (e.g. 0): The object *definitely* does
not
		#		implement this requirement and should not be
		#		checked further
		#
		#		None: The __query__() method doesn't know if
it
		#		implements this requirement and its
attributes
		#		should be checked to see if it does or not.

		if __query__:

			query = adaptor.__query__(self)

			if query:
				return adaptor
			elif query != None:
				return None

		#	For now, we will just verify that the object has the
		#	named attribute. Later this can be more complex
		#	(parameters, signatures, etc)

		attr = getattr(adaptor, self.getName(), None)

		if attr:
			return adaptor

		#	Not required, but explicit is better than implicit
;)
		return None

class interface:
	""""""
	
	def __init__(self):
		""""""

	def requirements (self):
		""""""
		#	return sequence of __requirement - requirements are
		#	generated automatically by factory methods in the
		#	interface class
		return ()

	def __adapt__(self, descriptor):
		""""""

		#	For now, only interface and its subclasses will be
		#	directly adapted. Later we will add a verification
		#	process directly here as the requirements() method
		#	should return an empty sequence unless requirements
		#	have been added by subclasses or instances

		if self == descriptor or isinstance(descriptor, interface):
			return self

__interface = interface()

class implements:
	""""""
	
	def __init__(self):
		""""""

	def __query__(self, requirement):
		""""""
		

def adapt(descriptor, object):

	#	Verify that the interface object actually meets
	#	the requirements for the interface descriptor interface

	if interface:
		assert adapt(None, descriptor)
	else:
		descriptor = __interface

	adapted = None
	__adapt__ = getattr(object, '__adapt__', None)

	if __adapt__:
		adapted = object.__adapt__(descriptor)

	if not adapted:

		adapted = object
		__query__ = getattr(object, '__query__', None)
		
		for r in descriptor.requirements():
			
			if not r.verify(adaptor, __query__):
				adapted = None
				break

	if adapted:
		return adapted
	else:
		raise AttributeError('Does not implement interface')

if __name__ == '__main__':
	print adapt(interface(), implements())



From ljohnson at resgen.com  Fri Mar 30 11:35:51 2001
From: ljohnson at resgen.com (Lyle Johnson)
Date: Fri, 30 Mar 2001 10:35:51 -0600
Subject: Application distribution with python
References: <mailman.985964564.7170.python-list@python.org>
Message-ID: <tc9db41hcu8417@corp.supernews.com>

> I need to install python for my client machine with my application.
>
> Can any one knows what are files to ship to client machine or
> list of distribution files.

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




From jkraska1 at san.rr.com  Sat Mar 31 17:56:19 2001
From: jkraska1 at san.rr.com (Courageous)
Date: Sat, 31 Mar 2001 22:56:19 GMT
Subject: list vs tuple
References: <mailman.986066384.30712.python-list@python.org> <d1rx6.1053$p5.4024@news1.rivrw1.nsw.optushome.com.au> <slrn9cciaj.3ug.qrczak@qrnik.zagroda> <lWrx6.1057$p5.4152@news1.rivrw1.nsw.optushome.com.au>
Message-ID: <edocctchjg1j333aum94s7qv1tjkmjg7nf@4ax.com>

>What happens to the objects is not the point, what happens to ...

IIRC, Python has been the subject of extensive usability tests
as part of DARPA's CP4E effort. The real point is that people
don't have any problem learning this aspect of Python; as far
as I can tell, your objection is tied wholly into your preconceptions.
Drop the preconceptions, and the problem will disappear.

Python is elegant and comprehensible.

Period.

C//



From kens at sightreader.com  Tue Mar 27 19:45:31 2001
From: kens at sightreader.com (Ken Seehof)
Date: Tue, 27 Mar 2001 16:45:31 -0800
Subject: Getting all the *files* from a directory -- A better way??
References: <mailman.985732718.3725.python-list@python.org> <99r6v7$4hef2$1@fido.engr.sgi.com>
Message-ID: <000901c0b720$64d39c00$04090a0a@upcast.com>

----- Original Message -----
From: "Paul Jackson" <pj at sgi.com>
Newsgroups: comp.lang.python
To: <python-list at python.org>
Sent: Tuesday, March 27, 2001 3:15 PM
Subject: Re: Getting all the *files* from a directory -- A better way??


> Quentin Crain asks:
> |> Here is how I get all the files from a directory:
> |>
> |>   dir="a/b/c"
> |>   files=filter(os.path.isfile,["%s/%s"%(dir,f) for f in os.listdir
> |> (dir)])
> |>
> |> Is this *REALLY* the Pythonic way to do this? (OR, why does
> |> os.listdir return only file names, and not the path given
> |> prepended?)
>
> Close - better to use os.path.join (more portable across
> operating environments with different path component
> separator characters, such as '\' in DOS/Windows).

Actually '/' is generally portable.  I use os.path.join for combining
variables, but I use '/' within hard-coded relative paths.

e.g.:
f = os.path.join(root, 'doc/chapter2/spam.html')
rather than
f = os.path.join(root, os.path.join('doc', os.path.join('chapter2',
'spam.html'))))

This is fine because open() and all other file functions automagically
normalize the path to the operating system.  Warning: if you want
to use the result in a command line (e.g. os.popen()) be sure to
normalize the path with os.path.normpath().

> Try:
>
>     dir="a/b/c"
>     files = filter(
> os.path.isfile,
> [os.path.join(dir, f) for f in os.listdir(dir)]
>     )
>
> I'd say that the os.listdir() method only returns the
> directory entries (by simple basename), not with any
> path prefixed, because it is intended to be basic
> operation on a directory, and directories have just
> the simple name entries, not full pathnames, in them.
>
> What one wants to do with those directory entries, after
> extracting them with os.listdir(), can vary all over the
> lot, and may or may not involve joining the directory path.
> So that should be left up to the caller of listdir().
>
> --
>                           I won't rest till it's the best ...
>                           Manager, Linux System Software
>                           Paul Jackson <pj at sgi.com> 1.650.933.1373
> --
> http://mail.python.org/mailman/listinfo/python-list
>




From MarkH at ActiveState.com  Sun Mar 11 17:34:02 2001
From: MarkH at ActiveState.com (Mark Hammond)
Date: Sun, 11 Mar 2001 22:34:02 GMT
Subject: Pythonwin default directory init
References: <3aab04bc.2359756@News.CIS.DFN.DE>
Message-ID: <3AABFC9E.3070706@ActiveState.com>

Costas Menico wrote:

 > I use Pythonwin and I would like it if the File/Open or File/Save  would
 >  default to a directory of my choice when I start it up.    I have
 > tried making a startup module to set the paths of sys but that  doesn't
 >  seem to work. I have looked thru the .py files trying to  figure
 > out where I could fix this and can't find it.    Also setting it up
 >  as a shortcut on the Desktop defaults to the  Desktop.     Does
 > anyone know how to do this?

Assuming you are on Win2k (or possibly Win98), explorer itself remembers 
the directory to open these dialogs in.  It remembers the last directory 
you navigated to, even between runs of the application.  This can be 
prevented by internally passing an explicit directory, but Pythonwin 
chooses not to do this.   In Windows NT/95, the current directory is 
always used.

Mark.



From pj at sgi.com  Tue Mar 27 18:15:51 2001
From: pj at sgi.com (Paul Jackson)
Date: 27 Mar 2001 23:15:51 GMT
Subject: Getting all the *files* from a directory -- A better way??
References: <mailman.985732718.3725.python-list@python.org>
Message-ID: <99r6v7$4hef2$1@fido.engr.sgi.com>

Quentin Crain asks:
|> Here is how I get all the files from a directory:
|> 
|>   dir="a/b/c"
|>   files=filter(os.path.isfile,["%s/%s"%(dir,f) for f in os.listdir
|> (dir)])
|> 
|> Is this *REALLY* the Pythonic way to do this? (OR, why does
|> os.listdir return only file names, and not the path given
|> prepended?)

Close - better to use os.path.join (more portable across
operating environments with different path component
separator characters, such as '\' in DOS/Windows).

Try:

    dir="a/b/c"
    files = filter(
	os.path.isfile,
	[os.path.join(dir, f) for f in os.listdir(dir)]
    )

I'd say that the os.listdir() method only returns the
directory entries (by simple basename), not with any
path prefixed, because it is intended to be basic
operation on a directory, and directories have just
the simple name entries, not full pathnames, in them.

What one wants to do with those directory entries, after
extracting them with os.listdir(), can vary all over the
lot, and may or may not involve joining the directory path.
So that should be left up to the caller of listdir().

-- 
                          I won't rest till it's the best ...
                          Manager, Linux System Software
                          Paul Jackson <pj at sgi.com> 1.650.933.1373


From phrxy at csv.warwick.ac.uk  Tue Mar 27 13:42:37 2001
From: phrxy at csv.warwick.ac.uk (John J. Lee)
Date: Tue, 27 Mar 2001 19:42:37 +0100
Subject: Netscape and Python
In-Reply-To: <mailman.985699776.29012.python-list@python.org>
References: <13EE655665F4D311AB4D0008C789498A01EE2CA5@zei02exm02.cork.cig.mot.com> <mailman.985699776.29012.python-list@python.org>
Message-ID: <Pine.SOL.4.30.0103271941460.10600-100000@mimosa.csv.warwick.ac.uk>

On Tue, 27 Mar 2001, Kalle Svensson wrote:

> Sez O'Rourke Clodagh-corour01:
> > Hi all,
> >
> > Does anyone know how to launch Netscape from Python?
>
> Use the webbrowser module in 2.0, otherwise look at os.system, os.exec*
> os.spawn*.

or COM, on windows?  As you can tell from a message I just posted, I'm new
to that, so there may not even be a COM interface for all I know.


John



From howard at brazee.net  Tue Mar 13 16:54:56 2001
From: howard at brazee.net (Howard Brazee)
Date: Tue, 13 Mar 2001 14:54:56 -0700
Subject: Einstein's Riddle
References: <oNPq6.396644$ge4.138479566@news2.rdc2.tx.home.com> <wi6elw1pg5y.fsf@merlin.irisa.fr>
Message-ID: <3AAE9730.5A7A4249@brazee.net>

This is cross-posted.  Please tell us which language you are using.

Yoann Padioleau wrote:

> "Steve Graham" <js.graham at home.com> writes:
>
> > Who would be interested in using his/her brain (and his computer) to solve
> > the following?
> here it is:
>
> %----------------------------------------------------------------
> % a sicstus program
> :- use_module(library(clpfd)).
>
> relation(Unknown) :-
>         Unknown =
>         [[Color1, Nationality1, Drink1, Smoke1, Pet1]
>         ,[Color2, Nationality2, Drink2, Smoke2, Pet2]
>         ,[Color3, Nationality3, Drink3, Smoke3, Pet3]
>         ,[Color4, Nationality4, Drink4, Smoke4, Pet4]
>         ,[Color5, Nationality5, Drink5, Smoke5, Pet5]
>         ]
>         , Colors =        [Color1, Color2, Color3, Color4, Color5]
>         , Nationalities = [Nationality1, Nationality2, Nationality3, Nationality4, Nationality5]
>         , Drinks =        [Drink1, Drink2, Drink3, Drink4, Drink5]
>         , Smokes =        [Smoke1, Smoke2, Smoke3, Smoke4, Smoke5]
>         , Pets   =        [Pet1, Pet2, Pet3, Pet4, Pet5],
>
>         Red = 1, Green = 2, Yellow = 3, Blue = 4, White = 5,
>         Brit = 1, Sweed = 2, Dane = 3, Norvegian = 4, German = 5,
>         Tea = 1, Coffee = 2, Milk = 3, Beer = 4, Water = 5,
>         Pall_mall = 1, Dunhill = 2, Blends = 3, Bluemaster = 4, Prince = 5,
>         Dogs = 1, Birds = 2, Cats = 3, Fish = 4, Horse = 5,
>
>         domain([Color1, Nationality1, Drink1, Smoke1, Pet1
>                ,Color2, Nationality2, Drink2, Smoke2, Pet2
>                ,Color3, Nationality3, Drink3, Smoke3, Pet3
>                ,Color4, Nationality4, Drink4, Smoke4, Pet4
>                ,Color5, Nationality5, Drink5, Smoke5, Pet5], 1, 5)
>         , all_different(Colors)
>         , all_different(Nationalities)
>         , all_different(Drinks)
>         , all_different(Smokes)
>         , all_different(Pets)
>         ,
>
> % cant do the member, cos member need instantiated things!! and if enum and then test member
> %  we gain nothing cos always big search space, the labeling must be at the end !!
>
>          A1 in 1..5, element(A1, Nationalities, Brit),  element(A1,Colors,Red),
>          A2 in 1..5, element(A2, Nationalities, Sweed), element(A2,Pets, Dogs),
>          A3 in 1..5, element(A3, Nationalities, Dane),  element(A3,Drinks, Tea),
>          A4 in 1..5, element(A4, Colors,        Green),
>          A5 in 1..5, A5 #= A4+1, element(A5, Colors, White),
>          A6 in 1..5, element(A6, Colors, Green),        element(A6, Drinks, Coffee),
>          A7 in 1..5, element(A7, Smokes, Pall_mall),    element(A7, Pets,  Birds),
>          A8 in 1..5, element(A8, Colors, Yellow),       element(A8, Smokes, Dunhill),
>          element(3, Drinks, Milk),
>          element(1, Nationalities, Norvegian),
>          A9 in 1..5, element(A9, Smokes, Blends),
>          A10 in 1..5, ((A10 #= A9-1) #\/ (A10 #= A9+1)), element(A10, Pets, Cats),
>          A11 in 1..5, element(A11, Pets, Horse),
>          A12 in 1..5 , ((A12 #= A11-1) #\/ (A12 #= A11+1)), element(A12, Smokes, Dunhill),
>          A13 in 1..5, element(A13, Smokes, Bluemaster),  element(A13, Drinks, Beer),
>          A14 in 1..5, element(A14, Nationalities, German), element(A14, Smokes, Prince),
>          A15 in 1..5, element(A15, Nationalities, Norvegian),
>          A16 in 1..5 , ((A16 #= A15-1) #\/ (A16 #= A15+1)), element(A16, Colors, Blue),
>          A17 in 1..5, element(A17, Smokes, Blends),
>          A18 in 1..5 , ((A18 #= A17-1) #\/ (A18 #= A17+1)), element(A18, Drinks, Water),
>
>         labeling([],
>                  [Color1, Nationality1, Drink1, Smoke1, Pet1
>                  ,Color2, Nationality2, Drink2, Smoke2, Pet2
>                  ,Color3, Nationality3, Drink3, Smoke3, Pet3
>                  ,Color4, Nationality4, Drink4, Smoke4, Pet4
>                  ,Color5, Nationality5, Drink5, Smoke5, Pet5
>                  , A1, A2, A3, A4, A5, A6, A7, A8, A9, A10, A11, A12, A13, A14, A15, A16, A17, A18
>                  ]).
>
> %relation(X).
> %  X = [[3,4,5,2,3],[4,3,1,3,5],[1,1,3,1,2],[2,5,2,5,4],[5,2,4,4,1]] ? ; corresponding to
> %  [Yellow, Norvegian, Water, Dunhill, Cats]
> %  [Blue,   Dane,       Tea,  Blends, Horse]
> %  [Red,    Brit,      Milk,  Pall_mall, Birds]
> %  [Green,  German,  Coffee,  Prince,    Fish]
> %  [White,  Sweed,   Beer,  Bluemaster,  Dogs]
>
> %----------------------------------------------------------------
>
> >
> >
> > Steve Graham
> >
> > ===
> >
> > Einstein's Riddle
> >
> > Albert Einstein wrote this riddle this century [ed. 20th century].  He said
> > 98% of the world could not solve it.
> >
> > There are 5 houses in 5 different colors.  In each house lives a person with
> > a different nationality.  The 5 owners drink a certain type of beverage,
> > smoke a certain brand of cigar, and keep a certain pet.  No owners
> > have the same pet, smoke the same brand of cigar or drink the same beverage.
> >
> > The question is: "Who owns the fish?"
> >
> > Hints:
> >
> > The Brit lives in the red house.
> >
> > The Sweed keeps dogs as pets.
> >
> > The Dane drinks tea.
> >
> > The green house is on the left of the white house.
> >
> > The green house's owner drinks coffee.
> >
> > The person who smokes Pall Mall rears birds.
> >
> > The owner of the yellow house smokes Dunhill.
> >
> > The man living in the center house drinks milk.
> >
> > The Norwegian lives in the first house.
> >
> > The man who smokes Blends lives next to the one who keeps cats.
> >
> > The man who keeps the horse lives next to the man who smokes Dunhill.
> >
> > The owner who smokes Bluemasters drinks beer.
> >
> > The German smokes Prince.
> >
> > The Norwegian lives next to the blue house.
> >
> > The man who smokes Blends has a neighbor who drinks water.
> >
> >
> >
> >
> >
> >
> >
> >
> >
>
> --
> Yoann  Padioleau,  INSA de Rennes, France,   http://www.irisa.fr/prive/padiolea
> Opinions expressed here are only mine. Je n'?cris qu'? titre personnel.
> **____   Get Free. Be Smart.  Simply use Linux and Free Software.   ____**



From bos at hack.org  Wed Mar 28 23:22:17 2001
From: bos at hack.org (Rikard Bosnjakovic)
Date: Thu, 29 Mar 2001 06:22:17 +0200
Subject: KeyBoardInterrupt error
References: <mailman.985802077.12779.python-list@python.org>
Message-ID: <3AC2B879.E2946AA0@hack.org>

"Jason Straw." wrote:

> What can trigger a Keyboard Interrupt?

>>> raise KeyboardInterrupt
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
KeyboardInterrupt


-- 
Rikard Bosnjakovic - http://a214.ryd.student.liu.se/cv/ - ICQ: 1158217

Anyone sending unwanted advertising e-mail to my address will be
charged $250 for network traffic and computing time. By extracting my
address from this message or its header, you agree to these terms.


From qrczak at knm.org.pl  Sat Mar  3 14:16:15 2001
From: qrczak at knm.org.pl (Marcin 'Qrczak' Kowalczyk)
Date: 3 Mar 2001 19:16:15 GMT
Subject: Opposite of PyLong_FromString
Message-ID: <slrn9a2gnv.hs6.qrczak@qrnik.zagroda>

What is the best way to convert a Python's Long to a C string
(matching "-?[0-9]+")?

-- 
 __("<  Marcin Kowalczyk * qrczak at knm.org.pl http://qrczak.ids.net.pl/
 \__/
  ^^                      SYGNATURA ZAST?PCZA
QRCZAK


From twanvds at xs4all.nl  Mon Mar 19 06:22:03 2001
From: twanvds at xs4all.nl (Twan van der Schoot)
Date: Mon, 19 Mar 2001 12:22:03 +0100
Subject: How to increase PythonWin v2.0: COM browser limits?
Message-ID: <994psh$so5$1@news1.xs4all.nl>

I'm trying to use the COM Browser in Pythonwin version 2.0.  But I run into
limitations of some kind.  As a Python newby I dutifully scanned FAQs and
newsgroups, but alas!

To reproduce the problem: On a heavily used NT 4.0 system with a lot of
registered COM stuff, start Pythonwin version 2.0 go to the menu entry: COM
Browser to start the COM browser and expand the "Registered Categories"
branch in the tree view.

Pythonwin shows in the interactive window a backtrace on an exception.  The
culprit raising this exception was the response:

File "D:\Python20\win32com\client\util.py", line 53, in __GetIndex
    result = self._oleobj_.Next(1)
pywintypes.com_error: (-2147024882, 'Not enough storage is available to
complete this operation.', None, None)
win32ui: Exception in OnNotify() handler

Something suggests that the number of registered objects is too large for
the COM browser.  As I've -of course- the sources available, I was wonder if
there is some parameter which I can increase to solve the problem.  Or am I
running into limitations of the treeview control of MS Windows?

TIA

Twan van der Schoot
twanvds at NOSPAM.xs4all.nl




From fredrik at pythonware.com  Thu Mar  1 01:26:58 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Thu, 01 Mar 2001 06:26:58 GMT
Subject: is and == (is not other_thread)
References: <3A9DD10F.C8ECF803@seebelow.org>
Message-ID: <SYln6.16081$Qb7.2628916@newsb.telia.net>

Grant Griffin wrote:
> 1) Are two strings that are equal in the "==" sense always
> guaranteed to be equal in the "is" sense?

no.

> 2) Is "is" faster than "==" for strings?

marginally.  "==" executes a bit more C code before it
gets to the identity test.

> 3) Internally, does "==" for strings always begin with an "is"
> operation?

yes

for the details, see cmp_outcome in Python/ceval.c, and
PyObject_Compare in Objects/Object.c.

Cheers /F

<!-- (the eff-bot guide to) the standard python library:
http://www.pythonware.com/people/fredrik/librarybook.htm
-->




From bedge at troikanetworks.com  Tue Mar  6 11:33:08 2001
From: bedge at troikanetworks.com (Bruce Edge)
Date: Tue, 06 Mar 2001 08:33:08 -0800
Subject: absolute namespace for callbacks?
References: <3AA50AFB.83C7459F@troikanetworks.com> <983895797.296667@newsmaster-04.atnet.at>
Message-ID: <3AA51144.3DD3FC3F@troikanetworks.com>

My callback func is defined in one module, and the callback is executed
from another module which does not know in advance what module the
callback func resides in. So, by default, the caller cannot find the
callback func in it's own namespace.
The callback registration procedure needs to specify the namespace where
the callback func exists.

OTOH, maybe I'm missing something here :-)

-Bruce.

Werner Schiendl wrote:
> 
> What's the matter?
> 
> You do not pass the name, but a reference to the function
> (the function is an object for python like anything else)
> It does not matter where the function is defined...
> 
> or did I miss anything in your post?
> 
> hth
> werner
> 
> Bruce Edge <bedge at troikanetworks.com> wrote in message
> news:3AA50AFB.83C7459F at troikanetworks.com...
> > Tcl has a func to get a fully qualified namespace from a context. This
> > is essential for registering callback functions.
> > How does one register a callback in Python, when the callback func
> > exists in a different namespace?
> >
> > Is there any method to get the current absolute namespace?
> >
> > Thanks, Bruce.
> >


From bedge at troikanetworks.com  Mon Mar  5 10:38:44 2001
From: bedge at troikanetworks.com (Bruce Edge)
Date: Mon, 05 Mar 2001 07:38:44 -0800
Subject: create new statement?
References: <3A9FE7C2.1F5525D2@troikanetworks.com> <003101c0a349$d5c76ad0$a325fea9@his> <3A9FF567.920D1027@troikanetworks.com> <007801c0a429$b53e3060$a325fea9@his>
Message-ID: <3AA3B304.7CBFFBB3@troikanetworks.com>

Ken Seehof wrote:
> 
> Instead of messing with the python interpreter, maybe you could make a very
> simple command translator that converts a command line into python syntax.
> Since you want to mix this with python syntax, you could have rules to
> determine which lines to apply the translator.
> 
> The translator would leave the line alone unless it meets the following
> conditions:
> - The line starts with one of your commands.
> - The line doesn't contain punctuation that is incompatible with the command
> syntax.
> 
> Example:
> chew spam 2
> 
> translates to:
> chew('spam', 2)

OK, perfect. Now, where does one hook the translator in? I'm using
readline and a derived ldcompleter to handle cmd line editing. I suppose
I could wedge it in with the completer, but I can't help wondering if
there's a better place.

Thanks, Bruce.

> 
> In this example, you would support space delimited integers, floats, and
> unquoted strings in your command syntax. Strings would be anything without
> whitespace that does not parse as integer or float; or if that's not
> acceptable, use quoted strings and delimit with comments.  The details
> depend on what you want the command syntax to look like.
> 
> I've done exactly the opposite of that for a company I used to work for.  I
> wrapped an existed command interpreter in python by generating python
> callable objects that generate the command syntax and invoke the existing
> command interpreter.
> 
> ----- Original Message -----
> From: "Bruce Edge" <bedge at troikanetworks.com>
> Newsgroups: comp.lang.python
> To: "Ken Seehof" <kens at sightreader.com>
> Sent: Friday, March 02, 2001 11:32 AM
> Subject: Re: create new statement?
> 
> > Ken Seehof wrote:
> > >
> > > Sorry, not without rebuilding python.
> > >
> > > What is the actual problem that you are trying to solve?  Maybe there's
> > > another approach.
> >
> > I'm creating a command line interface to a product.
> > I want to have access to python as a scripting tool, and be able to
> > access my own commands.
> > Ideally, this could be used by the end user. So, I'd like to not require
> > the my_func(args) syntax,
> > as that may look a bit kludgey/unfinished from a users standpoint. I'd
> > like to be able to use "my_command args" from the python cmd line.
> > Everything else I need is already there, readline line editing,
> > customizable tab-completer, python language support.
> >
> > My expertise is in Tcl, where this is trivial. I'm trying out python as
> > there are others here that already use it, and it's well, somewhat
> > syntactically cleaner.
> >
> >
> > -Bruce.
> >
> >
> > >
> > > ----- Original Message -----
> > > From: "Bruce Edge" <bedge at troikanetworks.com>
> > > Newsgroups: comp.lang.python
> > > To: <python-list at python.org>
> > > Sent: Friday, March 02, 2001 10:34 AM
> > > Subject: create new statement?
> > >
> > > > Can you add a new statement to python?
> > > > AFAICT the only type of commands that don't require () or . after them
> > > > are statements.
> > > >
> > > > I want to add my own command to the interp, but not have to use object
> > > > calling syntax
> > > > eg:
> > > >     create obj opt
> > > > not
> > > >     create(obj, opt)
> > > >
> > > > How do I define new statements, or, a way of calling functions using
> the
> > > > former syntax?
> > > >
> > > > Thanks, Bruce.
> > > > --
> > > > http://mail.python.org/mailman/listinfo/python-list
> > > >
> >


From tbryan at python.net  Sat Mar 10 23:32:37 2001
From: tbryan at python.net (Thomas A. Bryan)
Date: Sat, 10 Mar 2001 23:32:37 -0500
Subject: Python 2.0 and Tcl/Tk 8.3.2?
References: <3A997A0A.67CA82E1@UTS.Itron.com>
Message-ID: <3AAAFFE5.15A7A878@python.net>

Tom Bryan wrote:
> 
> I am upgrading our Python installation at work (on Tru64 UNIX boxes).  We
> were still running Python 1.5.2 and Tcl/Tk 8.0.5.  I'm installing Python-2.0,
> but I'm not sure which version of Tcl/Tk I should grab for Tkinter.  After
> some searching on Google, I saw some mention of problems with using Tcl/Tk 8.3
> with Python.
> 
> Any suggestions?  Should I stick with an 8.2 release of Tcl/Tk, or is the
> most recent Tcl/Tk 8.3 release stable enough to use with Python-2.0?

For the archives, I thought that I should note that I received e-mail that 
doesn't seem to have been sent to the mailinglist reporting that Tcl/Tk 8.3 
seems to work with Python 2.0.  I'll post again if I run into any serious 
problems.

If Python works as well with Tcl/Tk 8.3 as it does with Tcl/Tk 8.0, I wonder 
why the default Python-2.0 still defaults to the older version.  Does this 
line get updated to the latest version of Tcl/Tk that was known to work at 
the time of release, or is it frozen at 8.0 for the time being?

>From Modules/Setup.in:
# *** Uncomment and edit to reflect your Tcl/Tk versions:
#       -ltk8.0 -ltcl8.0 \

---Tom


From max at alcyone.com  Sat Mar 31 14:33:58 2001
From: max at alcyone.com (Erik Max Francis)
Date: Sat, 31 Mar 2001 11:33:58 -0800
Subject: I want to learn PYTHON!
References: <1iqx6.3744$aD4.234302@news2.atl>
Message-ID: <3AC63126.8EF56136@alcyone.com>

KEVIN wrote:

> Where do I start.
> 
> I am not a programmer but have an intermediate level of computer
> knowledge
> in Windows only.

Check out the tutorial on http://www.python.org/doc/ for starters.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ I like young girls.  Their stories are shorter.
\__/ Thomas McGuane
    Kepler's laws / http://www.alcyone.com/max/physics/kepler/
 A proof of Kepler's laws.


From Eugene.Leitl at lrz.uni-muenchen.de  Tue Mar 27 07:20:50 2001
From: Eugene.Leitl at lrz.uni-muenchen.de (Eugene Leitl)
Date: Tue, 27 Mar 2001 14:20:50 +0200 (MET DST)
Subject: running code from "Scripting The Web with Python"
In-Reply-To: <20010327103855.23536.qmail@web206.mail.yahoo.com>
Message-ID: <Pine.GSO.4.03.10103271409400.11563-100000@sun1.lrz-muenchen.de>

I haven't snipped the rest of the post, to preserve content. The program
works nicely (it is serving a Java applet page right now), and I'm very
much impressed with it. Not with the program, that is, with the language.

It solves several of my problems: I can have a single binary containing my
entire application, instead of littering the client's filespace with
diverse versions of Python and/or large frozen binaries, or even little
ugly fewmets in the cgi-bin and html directories. Perfect.

Are the handlers asynchronous? Or will the server hang if processing
a lengthy request? If yes, are Python threads painless, or are
there diverse caveats?

TIA,
Eugene

On Tue, 27 Mar 2001, Hamish Lawson wrote:

> Eugene Leitl wrote:
> 
> > I've snipped the following code from Guido's
> > http://www.w3j.com/6/s3.vanrossum.html presentation:
> > ... I can't run it due to syntax errors
> 
> 
> Wow, they really messed up the formatting on that! Below is the
> properly formatted version.
> 
> Hamish Lawson
> 
> ---
> 
> import StringIO
> import SimpleHTTPServer
> 
> BaseClass = SimpleHTTPServer.SimpleHTTPRequestHandler
> 
> CounterTemplate = """ <H1>Server
> Statistics</H1>
> 
> This <A HREF=/index.html>server</A> has been accessed
> <b>%d</b> times.  """
> 
> count = 0
> 
> class MyRequestHandler(BaseClass):
> 
>     def do_GET(self):
>         global count
>         count = count + 1
>         BaseClass.do_GET(self)
> 
>     def send_head(self):
>         if self.path == "/counter.html":
>             return self.send_counter()
>         else:
>             return BaseClass.send_head(self)
> 
>     def send_counter(self):
>         self.send_response(200)
>         self.send_header("Content-type", "text/html")
>         self.end_headers()
>         text = CounterTemplate % count
>         return StringIO.StringIO(text)
> 
> def test():
>     SimpleHTTPServer.test(MyRequestHandler)
> 
> test() 
> 
> 
> ____________________________________________________________
> Do You Yahoo!?
> Get your free @yahoo.co.uk address at http://mail.yahoo.co.uk
> or your free @yahoo.ie address at http://mail.yahoo.ie
> 
> -- 
> http://mail.python.org/mailman/listinfo/python-list
> 

______________________________________________________________
ICBMTO  : N48 10'07'' E011 33'53'' http://www.lrz.de/~ui22204 
57F9CFD3: ED90 0433 EB74 E4A9 537F CFF5 86E7 629B 57F9 CFD3




From jcm at bigskytel.com  Tue Mar 13 10:25:08 2001
From: jcm at bigskytel.com (David Porter)
Date: Tue, 13 Mar 2001 08:25:08 -0700
Subject: main in mac?
In-Reply-To: <20010313095339.19264.00000243@ng-dh1.aol.com>; from park997@aol.comnotospam on Tue, Mar 13, 2001 at 02:53:39PM +0000
References: <20010313095339.19264.00000243@ng-dh1.aol.com>
Message-ID: <20010313082508.A17857@bigskytel.com>

* Park997 <park997 at aol.comnotospam>:

> I am using Python 2.0 on a PPC mac. I have a module
> that is set up as follows:
> 
> from Numeric import *
> 
> class Epower:
> 
> 	3 class functions defined
> 
> if __name__ == '__main__':
> 
> 	test class and print output
> 
> This code does nothing if the run all button is pressed
> or if it is imported into interactive python.

I don't know what the run all button is (not on mac), but the whole reason
for this idiom is so it wont 'do anything' when imported. If a file named
x.py is imported, then it's __name__ is x. If it is being run as a program,
then it's __name__ is __main__. 


David



From zamurai at gmx.net  Mon Mar  5 10:19:58 2001
From: zamurai at gmx.net (Zamurai)
Date: Mon, 5 Mar 2001 16:19:58 +0100
Subject: Thank you!!
References: <mailman.983803627.13972.python-list@python.org>
Message-ID: <983805598.905582867@news.t-online.de>

Thank you!!


From shaleh at valinux.com  Tue Mar 20 14:23:22 2001
From: shaleh at valinux.com (Sean 'Shaleh' Perry)
Date: Tue, 20 Mar 2001 11:23:22 -0800 (PST)
Subject: Can Python execute a compiled C program
In-Reply-To: <3AB7A784.6C2E2BDA@lucent.com>
Message-ID: <XFMail.20010320112322.shaleh@valinux.com>

On 20-Mar-2001 Carl Moser wrote:
> Can Python execute a compiled C-program?
> 

import os
os.system('ls') # run "ls"
fp = os.popen("ls") # run ls and make the output look like a file I can read

and there are other ways to, most of the C exec functionality is in python.



From fredrik at pythonware.com  Sat Mar 31 04:21:58 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Sat, 31 Mar 2001 09:21:58 GMT
Subject: list vs tuple
References: <hGlw6.782$p5.3210@news1.rivrw1.nsw.optushome.com.au> <3AC20CD3.265A2D94@alcyone.com> <ru69cto61jtmrqsf809e38hb3pbm2lhi7g@4ax.com> <bC1x6.4319$4N4.905946@newsc.telia.net> <71q9ct4le63anefmiajj5cbh3mg04qu1ga@4ax.com>
Message-ID: <Wkhx6.4422$4N4.950567@newsc.telia.net>

Michael Prager wrote:
> I believe it's important enough to be mentioned pretty close to
> Page 1, and in practical terms understandable by those more interested
> in getting work done than in semantics of programming languages
> as an independent field.

agreed.

> >more info here:
> >
> >    http://effbot.org/guides/python-objects.htm
>
> Thanks!  Very nice!

it's derived from the introductory parts in our training
courses.  in my experience, explaining the object model
first tends to minimize the "mutability" confusion (you
may still get surprised when you mess up the references,
but you'll quickly figure out what's going on)

I've seen Python books that don't discuss this at all,
except for some carefully arranged "everything you
know is wrong" traps in the exercise sections...

Cheers /F




From paul.moore at uk.origin-it.com  Wed Mar  7 06:10:16 2001
From: paul.moore at uk.origin-it.com (Paul Moore)
Date: Wed, 7 Mar 2001 12:10:16 +0100 
Subject: urllib.urlopen and proxies
References: <vse4atgcgcfafth7c5tq3po7srd3l32f3h@4ax.com> <008301c0a57d$1331cde0$a100000a@local> <mailman.983804894.17415.python-list@python.org> <BfakOm9OYyn2V0Hg9h1SSS=a7t7Y@4ax.com> <mailman.983901970.11969.python-list@python.org>
Message-ID: <vBKmOkB47iH8JfPSAvl1kJShzae+@4ax.com>

On Tue, 6 Mar 2001 19:05:35 +0100 , Siggy Brentrup <bsb at winnegan.de>
wrote:

>The authinfo stuff is for remote sites requiring authentication, as
>long as you visit unprotected sites only you can ignore it.
>
>Modify the docstring example as follows:
>
>--------
[...]
>--------
>
>Proxy authorization still untested, please report if it works.

Unfortunately, it still fails, with a traceback

>>> f = urllib2.urlopen('http://www.python.org/')
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
  File "c:\applications\python\lib\urllib2.py", line 137, in urlopen
    return _opener.open(url, data)
  File "c:\applications\python\lib\urllib2.py", line 325, in open
    '_open', req)
  File "c:\applications\python\lib\urllib2.py", line 304, in
_call_chain
    result = func(*args)
  File "c:\applications\python\lib\urllib2.py", line 747, in http_open
    raise URLError(err)
urllib2.URLError: <urlopen error host not found>

I'll see if I can understand what is going on when I get a chance, but
I don't know how far I'll get...

Paul.



From jmarshal at mathworks.com  Fri Mar 16 14:47:23 2001
From: jmarshal at mathworks.com (Joshua Marshall)
Date: 16 Mar 2001 19:47:23 GMT
Subject: Killing threads
References: <3ab23ea7.328575630@News.CIS.DFN.DE> <mailman.984767303.16093.python-list@python.org>
Message-ID: <98tqkb$o4r$2@news.mathworks.com>

Steve Purcell <stephen_purcell at yahoo.com> wrote:
> costas at meezon.com wrote:
>> I have th program below. How does one go about kill a thread once
>> started? Why is there no method for killing threads? I thought del
>> would do it but obviously it doesn't.

> Thread cancellation is generally considered a bad thing. If I'm not mistaken,
> the argument against allowing thread cancellation is concerned with
> correctly releasing locks shared between threads.  (I can trust Mr. Martelli
> to set me straight on this if I'm wrong :-)

> You should think about using a synchronisation mechanism
> (threading.Condition, threading.Semaphore, Queue.Queue) so that the worker
> thread can exit itself when told to do so by the main thread changing the
> synchronising object.

> -Steve

Yes, thread cancellation can get you in trouble.  But I'd still like
to have this option.  Specifically, I'd like to be able to raise a
particular exception in a given thread.


From seandc at att.net  Sun Mar 25 10:06:10 2001
From: seandc at att.net (Sean)
Date: Sun, 25 Mar 2001 15:06:10 GMT
Subject: Python Mega Widgets rpm spec file?
References: <vwgv6.1085$107.1233792@typhoon.ne.mediaone.net>
Message-ID: <3ABE089A.91DBA2F3@att.net>

Python Megawigets is a library of .py files which
can be found at:

    http://pmw.sourceforge.net/

I've used it "as is" on Linux and Win9x.  No problems!

Sean
====

rodrigc at mediaone.net wrote:
> 
> Hi,
> 
> I am using John Grayson's book "Python and Tkinter Programming"
> to learn how to develop GUI's under Python.
> I developing under Redhat Linux 7.0 which comes with
> Python 1.52 and Tkinter.
> 
> Grayson's book refers to Python Mega Widgets (Pmw) quite
> a bit, but this does not come standard with Redhat distributions.
> Has anyone made an rpm spec file for Pmw, so that it
> can be installed as an rpm package?
> 
> I like to work on Redhat systems, and would like to deploy
> my GUI to other Redhat systems, without having to package
> up 3rd party libraries which are not included as part of
> the base distribution.
> 
> Thanks for any help.
> --
> Craig Rodrigues
> http://www.gis.net/~craigr
> rodrigc at mediaone.net


From hgg9140 at cola.ca.boeing.com  Wed Mar 14 11:26:45 2001
From: hgg9140 at cola.ca.boeing.com (Harry George)
Date: Wed, 14 Mar 2001 16:26:45 GMT
Subject: Who am I: can a class instance determine its own name?
References: <mailman.984034150.9635.python-list@python.org> <DYV4H3fqAHA.187@typhoon.i3s.net> <mailman.984324974.25480.python-list@python.org> <vZJg10qqAHA.210@typhoon.i3s.net>
Message-ID: <xqxzoeol3ju.fsf@cola.ca.boeing.com>

class MyName:
    #--------------------------
    def init(self,name):
        self.name=name
        try:
            raise None
        except:
            c=sys.exc_info()[2].tb_frame.f_back.f_back
        c.f_locals[name]=self
...

MyName("me")
print me.name

"Kevin Cazabon" <kevin_cazabon at hotmail.nospamplease!.com> writes:

> The reason I say it's not elegant is that there is the possibility (read
> "probability") that the assigned name and the actual name can differ.
> 
> We're assuming that the programmer assigns exactly the same name to both,
> which doesn't seem like a hard task, but it introduces the possibility of
> problems, and probably hard ones to debug too.
> 
> If there were an automatic method, it would be MUCH more reliable.
> 
> Just MHO.
> 
> Kevin.
> 
> 
> 
> "Paul Prescod" <paulp at ActiveState.com> wrote in message
> news:mailman.984324974.25480.python-list at python.org...
> | Kevin Cazabon wrote:
> | >
> | > ...
> | > You could also
> | > make it an initialization parameter.
> | >
> | > FooBar = Foo(name="FooBar")
> | >
> | > Just trying to help... not elegantly though.
> |
> | I disagree Kevin. This is not only the canonical way to give an instance
> | a name, it is also the elegant way. It gives you a lot more flexibility
> | because the name it has at runtime is unrelated to the variable name you
> | happened to have given it in your code. If you need to change one or the
> | other for any reason, you can do so safely.
> |
> |  Paul Prescod
> |
> 
> 

-- 
Harry George                E-mail:  harry.g.george at boeing.com
The Boeing Company          Renton:  (425) 237-6915
P. O. Box 3707  02-CA       Everett: (425) 266-3868
Seattle, WA 98124-2207      Page:    (425) 631-8803  


From alf at leo.logilab.fr  Mon Mar  5 05:51:25 2001
From: alf at leo.logilab.fr (Alexandre Fayolle)
Date: Mon, 5 Mar 2001 10:51:25 +0000 (UTC)
Subject: Editing Python source as XML?
References: <mailman.983546704.10789.python-list@python.org> <j44rxcjgnw.fsf@informatik.hu-berlin.de> <97vqq4$2ap7$1@norfair.nerim.net>
Message-ID: <97vr3d$2ap7$2@norfair.nerim.net>

Alexandre Fayolle <alf at leo.logilab.fr> wrote:
> of Narval. We call it pypasax. It uses the parser module to extract 
> information about classes and methods and generates an XML tree from this.
> We are working on XSLT to generate XMI files so that we can import the data 
> in some UML tool such as ArgoUML.

It an be found at http://www.logilab.org/pypasax/

Alexandre Fayolle
-- 
http://www.logilab.com 
Narval is the first software agent available as free software (GPL).
LOGILAB, Paris (France).


From missive at frontiernet.net  Thu Mar 29 16:56:44 2001
From: missive at frontiernet.net (Lee Harr)
Date: 29 Mar 2001 21:56:44 GMT
Subject: Zope: dtml-with and dtml-sqltest
Message-ID: <9a0b2s$16to$1@node17.cwnet.frontiernet.net>

I have an SQL method (named sqm) like:

SELECT
	*
FROM
	table
WHERE
	<dtml-sqltest aaa type="int">


and I want to use it from a dtml method like:

<dtml-let aaalist="[1,2,3,4]">
<dtml-let aaa=sequence-item>

<dtml-in sqm>
<dtml-var variousdata>
</dtml-in>

</dtml-let>
</dtml-let>


but I am getting:

Error Type: Bad Request
Error Value: ['aaa']


Any help is appreciated.



From bdupire at seatech.fau.edu  Wed Mar 21 09:35:01 2001
From: bdupire at seatech.fau.edu (Benoit Dupire)
Date: Wed, 21 Mar 2001 09:35:01 -0500
Subject: Access Issues
References: <mailman.985182567.26602.python-list@python.org>
Message-ID: <3AB8BC15.1F2911DE@seatech.fau.edu>


O'Rourke Clodagh-corour01 wrote:

> Hi all,
>
> I wonder if anyone can help.
>
> I am trying call the constructor of the EnvVar class from the newEnvWin function of the Cell class.
> But I am getting the following error
>
> Exception in Tkinter callback
> Traceback (innermost last):
>   File "/usr/vob/omc_reference/ThirdParty/scriptSources/lib/python1.5/lib-tk/Tkinter.py", line 764, in __call__
>     return apply(self.func, args)
>   File "Cell.py", line 63, in newCellMgt
>     self.newEnvWin('cellmgt')
>   File "Cell.py", line 60, in newEnvWin
>     instance = EnvVar.__init__(g, self.parent)
> AttributeError: __init__
>

your pb seems a namescope pb: try
#Cell.py
<snip>
def newEnvWin(self, g):
      newParent = Toplevel()
      instance = Envar.EnvVar(g, self.parent)



>
> I have include some extracts of the code.
> Thank you,
> Clodagh.
>
> **********************************************
> import EnvVar, GuiSelection, AuditResync, System
> from shellgui import EnvVarSite
> import Pmw
> from Tkinter import *
>
> class Cell(EnvVarSite):
>     def __init__(self, parent=None, **kw):
>           ......
>
>     def newCellMgt(self):
>         self.newEnvWin('cellmgt')
>
>     def newEnvWin(self, g):
>       newParent = Toplevel()
>       instance = EnvVar.__init__(g, self.parent)
>
> ********************************************
>
> import informixdb, os, sys, string, Pmw
> from Tkinter import *
> from shellgui import EnvVarSite
> import GuiSelection, AuditResync, Cell, System
>
> class EnvVar(EnvVarSite):
>
>     def __init__(self, g, parent=None):
>         self.sels = None
>       self.parent = parent
>         self.group = g
>         self.edit = 1
>         self.GetDBValues()
>         EnvVarSite.__init__(self,parent)
>
> ************************************************



From jota at telia.com  Sun Mar  4 06:35:09 2001
From: jota at telia.com (Jörgen Tapani)
Date: Sun, 04 Mar 2001 11:35:09 GMT
Subject: How to broadcast socket
Message-ID: <NLpo6.21534$AH6.3012743@newsc.telia.net>

I'm new to Python, and I have achived this
def test():
 s = socket(AF_INET, SOCK_DGRAM)
 s.setsockopt(SOL_SOCKET,SO_BROADCAST,1)
 s.sendto('Broadcast test',INADDR_BROADCAST)
 s.close()

And it do not run!
Help!
Jorgen Tapani






From sheila at spamcop.net  Fri Mar 23 23:44:34 2001
From: sheila at spamcop.net (Sheila King)
Date: Sat, 24 Mar 2001 04:44:34 GMT
Subject: Installing python on a virtual server account.
References: <3ABB8549.1987BA68@sewardweb.com>
Message-ID: <f69obt857fm756leasvciph2k6l6j26130@4ax.com>

On Fri, 23 Mar 2001 11:18:01 -0600, Lloyd Sommerer <lsommerer at sewardweb.com>
wrote in comp.lang.python in article <3ABB8549.1987BA68 at sewardweb.com>:

: My hosting
:company has python 1.5.1 installed and says they will not be upgrading

Sounds a lot like my web host. Well, I guess mine will upgrade eventually, but
don't mark it down on your calendar any time soon.

:(what they actually said was, "We are removing all references to Python
:from our website and are offically no longer supporting it. When we
:tried to upgrade to the newest version, it failed.").

Well, at least I know that you are not hosting where I host. They are still
supporting Python. 1.5.1/1.5.2, depending upon which server your site resides
on.

:Does anyone have experience installing 2.0 in a shared hosting
:situation? 

Yea, verily. That was my "challenge for the week", last week.

:Even telling me that I can't do it and that I should give up
:would  be helpful at this point.

No, it can be done.

:I am on a SunOS 5.7 machine.  Both configure and make seem to run okay.
:Make install dies with the message attached at the bottom of this
:message.  I'm assuming that is because I don't have the needed
:permission.

Well, I am not the one to ask about permissions, or any type of error messages
on a *nix type system. But I can point you to the documented version of my
adventure. My site is hosted on a Red Hat Linux server. I did successfully
install Python-2.0 in my own server space, with MUCH assistance from the folks
on the Python Tutor list. My (mis)adventure is detailed in the archives of the
Tutor list:
http://mail.python.org/pipermail/tutor/2001-March/thread.html

Under the subject:
Help with 500 error on cgi script 

As for whether you have permissions to install your own software, I guess
you'd have to ask your host about that.

--
Sheila King
http://www.thinkspot.net/sheila/
http://www.k12groups.org/

:Thanks for any advice.
:
:Lloyd Sommerer
:
:=====================================================
:gcc   python.o \
:          ../libpython2.0.a -ldb  -lpthread -lsocket -lnsl -ldl
:-lthread -lm  -
:o python
:Undefined                       first referenced
: symbol                             in file
:Py_Main                             python.o
:ld: fatal: Symbol referencing errors. No output written to python
:*** Error code 1
:make: Fatal error: Command failed for target `link'
:Current working directory /home/gatheri/python/Modules
:*** Error code 1
:make: Fatal error: Command failed for target `python'



From shaleh at valinux.com  Fri Mar  2 17:39:02 2001
From: shaleh at valinux.com (Sean 'Shaleh' Perry)
Date: Fri, 02 Mar 2001 14:39:02 -0800 (PST)
Subject: the demise of 'from foo import * and its implications?
In-Reply-To: <LNBBLJKPBEHFEDALKOLCGEIKJDAA.tim.one@home.com>
Message-ID: <XFMail.20010302143902.shaleh@valinux.com>

On 02-Mar-2001 Tim Peters wrote:
> [Sean 'Shaleh' Perry]
>> ## as an example of what happens
>> ## and sure, I could try to list every function I want to import,
>> ## but man does that get ugly
>> ...
> 
> What gave you the idea that "import *" is going away?
> 

/me needs to take a week off apparently.  Either that or stop believing
anything he reads.

I would have SWORN I read it here, and a fellow python hacker said something to
that affect as well.  Whichever, I will promptly use a gag as I am just digging
myself a hole today.



From dsh8290 at rit.edu  Fri Mar  9 11:22:38 2001
From: dsh8290 at rit.edu (D-Man)
Date: Fri, 9 Mar 2001 11:22:38 -0500
Subject: gah! I hate the new string syntax
In-Reply-To: <98an7i01tgp@news2.newsguy.com>; from aleaxit@yahoo.com on Fri, Mar 09, 2001 at 02:52:53PM +0100
References: <mailman.983615170.23313.python-list@python.org> <tN5o6.2746$eH3.14913@NewsReader> <Koco6.7518$1D5.244602@e420r-atl1.usenetserver.com> <mailman.983997252.5628.python-list@python.org> <m2vgpkyfjo.fsf@mycroft.actrix.gen.nz> <mailman.984063199.18779.python-list@python.org> <98an7i01tgp@news2.newsguy.com>
Message-ID: <20010309112238.F4546@harmony.cs.rit.edu>

On Fri, Mar 09, 2001 at 02:52:53PM +0100, Alex Martelli wrote:
| "D-Man" <dsh8290 at rit.edu> wrote in message
| news:mailman.984063199.18779.python-list at python.org...
| > that I am not really American since none of my ancestors came from
| > America.  "Real" Americans would be the so-called "Indians", or more
| 
| Why?  *THEIR* ancestors came from Asia over the Bering Straits,
| after all (and following the line a while longer, one gets back
| to Africa).

So we think, but no one has such records.  Actually, we are all
"people" and came from the Garden of Eden to take the line back to its
origin.  :-)

-D




From kalle at gnupung.net  Sun Mar  4 19:40:36 2001
From: kalle at gnupung.net (Kalle Svensson)
Date: Mon, 5 Mar 2001 01:40:36 +0100
Subject: [ANN] xmlrpccgi.py 0.1
Message-ID: <20010305014035.A15767@apone.network.loc>

[Not PGP signed according to c.l.py.announce posting guidelines]

xmlrpccgi.py is a CGI XML-RPC server, for those of us not lucky enough to be
able to run our own daemons.

This is a first release, expect bugs.  I'd really appreciate feedback.
There's no documentation yet, but check the test() function.
Any questions?  Don't hesitate to ask.

It uses xmlrpclib by Secret Labs, which can be found on
http://www.pythonware.com/products/xmlrpc/

URL:       http://gnupung.net/programming.php#xmlrpccgi
Download:  http://gnupung.net/download/xmlrpccgi.py
Demo:      http://gnupung.net/cgi-bin/xmlrpccgi.py
License:   GPL

About XML-RPC:
  XML-RPC is, as the name suggests, a protocol for remote procedure calls.
  It uses XML over e.g. HTTP.  Design goals were:
   * Cross platform compatibility.
   * Human readability.
   * Easy implementation.
  More info: http://www.xml-rpc.com/

Peace,
  Kalle
-- 
Email: kalle at gnupung.net     | You can tune a filesystem, but you
Web: http://www.gnupung.net/ | can't tune a fish. -- man tunefs(8)
PGP fingerprint: 0C56 B171 8159 327F 1824 F5DE 74D7 80D7 BF3B B1DD



From kelley at bioreason.com  Fri Mar 16 11:25:18 2001
From: kelley at bioreason.com (Brian Kelley)
Date: Fri, 16 Mar 2001 16:25:18 +0000
Subject: Yet another C++ binding (using Extension Classes)
Message-ID: <3AB23E6E.FF0234BC@bioreason.com>

About a year ago, I had to bind a C++ template class into objects that
could be used from within python.  I made a wrapper for Extension
Classes that extended the basic structure into a C++ class.  Any objects
that subclassed this class would automagically become valid python
objects and could be extended in much the same way as Extension Classes.

With the advent of swig/boost/sip and the like I didn't think there
would be much use outside our organization for this.  Anyway, I was
recently cajoled into asking whether anyone wants this code.  If I get
even one hit I'll write documentation for it and post it on my website
and submit it to the Vaults of Parnassus.

That said, the boost python library is very cool.

--
Brian Kelley          Bioreason, Inc
505 995-8188 ext 207  150 Washington St Suite 303
kelley at bioreason.com  Santa Fe, NM, 87501




From hungjunglu at yahoo.com  Mon Mar 12 10:19:47 2001
From: hungjunglu at yahoo.com (hungjunglu at yahoo.com)
Date: Mon, 12 Mar 2001 15:19:47 -0000
Subject: Chinese text GIF file generation
Message-ID: <98ipej+uq35@eGroups.com>

OK, got some help. Here are some more pointer from a friend:

- c2gif (listed at http://www.armed.net/how/pg003248.htm , at
  ftp://ftp.ifcss.org/pub/software/unix/convert/c2gif001.tar.gz
  http://www.math.ncu.edu.tw/~luors/c2gif/

- if you are using a Mac, you can just use AppleScript to copy the 
text,
  paste it into SimpleText or AppleWorks, paste in the text, and save 
it
  as a graphic file.  It works with TrueType fonts so it scales.

- http://www.ocrat.com/reaj/gif.html
  This is in Java and works on bitmapped Chinese fonts.

- http://www.ok88.com/go/svc/big52img.html
  It is a website that does it, but I don't know if you can get their
  script.

- http://www2.fhy.net/editor/JiZhang/software/gb2gif/win98.html
  runs on DOS..

Hung Jung





From ljohnson at resgen.com  Thu Mar  8 10:22:20 2001
From: ljohnson at resgen.com (Lyle Johnson)
Date: Thu, 8 Mar 2001 09:22:20 -0600
Subject: News from Python9!!!
References: <NUCp6.65206$Z8.11697516@typhoon.southeast.rr.com>
Message-ID: <taf8p52d86kd72@corp.supernews.com>

Since the links seem to be broken, here's a direct link to ActiveState's
news releases page:

    http://www.activestate.com/Corporate/Communications/News_Releases.html

Hope this helps,

Lyle




From tdelaney at avaya.com  Wed Mar 21 18:41:58 2001
From: tdelaney at avaya.com (Delaney, Timothy)
Date: Thu, 22 Mar 2001 10:41:58 +1100
Subject: New PEP: The directive statement
Message-ID: <B43D149A9AB2D411971300B0D03D7E8B0FBFBF@natasha.auslabs.avaya.com>

> > There are however, two things against this PEP.
> 
> > 1. It allows for one incompatible language change 
> (directive as keyword can
> > interfere with binding to name "directive" as the first 
> line of a file).
> 
> Again (and parser-engineering aside), this does not need to cause any
> backward-incompatibility.  The proposed syntax is not ambiguous.

I agree - the syntax *to us* is not ambiguous, but the PEP allows for
ambiguity as the parser will not be changed to distinguish the two cases.



From tuttledon at hotmail.com  Sat Mar 24 00:10:16 2001
From: tuttledon at hotmail.com (Don Tuttle)
Date: Sat, 24 Mar 2001 05:10:16 GMT
Subject: New Tkinter and PIL documentation (free)
References: <99en42$2vl6$1@newshost.nmt.edu>
Message-ID: <Y_Vu6.179074$Z8.37733056@typhoon.southeast.rr.com>

"John Shipman"
> At this site you will find my newly written reference document
> for Tkinter:
>
>     http://www.nmt.edu/tcc/help/pubs/lang.html
>

I downloaded the Tkinter PDF, but apparently I don't have the same font you
used.  (The text is very fuzzy)  Assuming this is the problem, would you
consided remaking the PDFs with a more standard font? Thanks.

Don




From william.famy at noos.fr  Fri Mar 23 08:24:00 2001
From: william.famy at noos.fr (William Famy)
Date: Fri, 23 Mar 2001 14:24:00 +0100
Subject: newbe   how to read in a string?
Message-ID: <3ABB4E70.54885C9F@noos.fr>

Hi.

I am looking how to read in a string.

Ex
a='2000/12/28,24.3,87.1,65.4'

I whant to extract the 87.1. In C i would code by a
scanf("%s,%f,%f,%f\n"),...

How can i do this in python?

Thanks for answer.

-- 
 			     \\\|///
                           \\  - -  //
                            (  @ @  )
+-------------------------oOOo-(_)-oOOo-------------------------+
|  Famy William  36 avenue des barattes  74000 Annecy  France   |
|  email:william.famy at noos.fr william.famy at mail-enseignant.com  |
+---------------------------------Oooo--------------------------+
                           oooO   (   )
                          (   )    ) /
                           \ (    (_/
                            \_)


From iansworld at thevortex.com  Mon Mar 19 19:23:17 2001
From: iansworld at thevortex.com (EEN)
Date: Tue, 20 Mar 2001 00:23:17 GMT
Subject: HELP - beginner to programing/python
Message-ID: <Vpxt6.416511$Pm2.6478704@news20.bellglobal.com>

Hi there. I am a newbie and am hearing that python is the way to go to start
off. I am downloading it as i type this, but need to know how to use it :)
of course :) Is there any site that has a good tutorial or beginners guides,
or do you have to buy books? If so, waht do you recommend?

Thanks a lot for the help, i really appreciate any pointers!




From bbollenbach at homenospam.com  Tue Mar  6 13:23:17 2001
From: bbollenbach at homenospam.com (Brad Bollenbach)
Date: Tue, 06 Mar 2001 18:23:17 GMT
Subject: Can Python replace Visual Basic? Should it?
References: <5DZo6.11126$hn5.1578410@news1.rdc1.mb.home.com> <3AA507D3.E384EF2@bigfoot.de>
Message-ID: <pW9p6.12818$hn5.1804827@news1.rdc1.mb.home.com>

"Gerhard H?ring" <gerhard.nospam at bigfoot.de> wrote in message
news:3AA507D3.E384EF2 at bigfoot.de...
[snip]
> Face it, most VB users aren't exactly star programmers. If I were
[snip]

That's a sweeping generalization, that I'd be very careful about making, to
say the least.

I'll certainly admit that I've only got about 7 months worth of experience
in "the industry", but of all the programmers I know of that had more than a
passing interest in VB (in other words, they liked and used it enough that
one might refer to them as "A VB Programmer(TM)"), at LEAST three (of the
six that I'm thinking of) were extremely bright and creative hackers. From
my perspective, I would be proud to be as good as they are!

The other three were also very very good, but maybe didn't "like" it enough
so much as they saw it as something that pays their bills. Even still, they
were very productive and useful, and found solutions quickly. In fact, the
only really VB lusers I've encountered seem to be via internet forms (like
IRC).

That said though, the stereotype of VB programmers all being idiots is
extremely common among Linux and Unix users, who, in many cases have almost
no knowledge of the tool, and would look silly themselves if throw into a
large project that used it ("silly" as in far from the creative geniuses
they would seem to indicate they are).


Brad




From bsass at freenet.edmonton.ab.ca  Thu Mar 22 03:53:20 2001
From: bsass at freenet.edmonton.ab.ca (Bruce Sass)
Date: Thu, 22 Mar 2001 01:53:20 -0700 (MST)
Subject: New PEP: The directive statement
In-Reply-To: <20010321215413.B25810@glacier.fnational.com>
Message-ID: <Pine.LNX.4.33.0103220004370.6632-100000@bms>

On Wed, 21 Mar 2001, Neil Schemenauer wrote:

> On Wed, Mar 21, 2001 at 10:21:50PM -0700, Bruce Sass wrote:
> > Why was (something along the lines of)...
> ...
> > use:
> >     __future__.nested_scopes
> >     # code that relies on nested scopes
> > except FutureError:
> >     # code that does not use nested scopes
> >
> > ...rejected?  [It looks like a natural to me <shrug>...
>
> "from __future__ import x" is a compile time switch.

Sorry, careless example.
"use:" would be a compile time switch (it would have to be),...

use:
    __future__.nested_scopes

...would be the same as...

from __future__ import nested_scopes


use:
    from __future__ import nested_scopes

...would have conveyed some of what I was after, but is overly
verbose.

The use/except construct may only make sense to a different
(hmmm) "class" of beta feature, or if you allow the user to override
the use of experimental code (in which case the programmer could
provide an alternative, if they are thoughtful).

> May I
> suggest reading the PEP:
<...>

Thanks for the URL... didn't help.  From what I read... what I'm
thinking of would probably be considered beyond the scope of the PEP;
the try/except Q: is kinda what I'm thinking of, my counter-question
would be:  So why not implement a similar mechanism at the compile
level (let's call it "use:")?

IMHO,
the PEP you referred to could be for the case where
"use: __future__.feature" has been used, there could be PEPs for
"use: __transitional__.whatever", etc.


- Bruce




From tanzer at swing.co.at  Wed Mar  7 01:26:20 2001
From: tanzer at swing.co.at (Christian Tanzer)
Date: Wed, 07 Mar 2001 07:26:20 +0100
Subject: Can Python replace Visual Basic? Should it?...and generalities : ) 
In-Reply-To: Your message of "Tue, 06 Mar 2001 20:41:34 GMT."
             <2Ybp6.12845$hn5.1818206@news1.rdc1.mb.home.com> 
Message-ID: <m14aXOv-000wcEC@swing.co.at>

"Brad Bollenbach" <bbollenbach at homenospam.com> wrote:

> But then, as seems to be often the case, people seem to blur the
> differentiation between "easy programming languages" and "easy programming
> tasks". How many people have written 30,000 lines out of _any_ programming
> language and still refer to it as being simple to work with?

Solving a task in different programming languages might result in
surprisingly different line counts. Therefore changing from an
unproductive language to a more productive one might reduce a big
solution to a much smaller one which might easier to work with. Of
course, this depends on the readability of the language giving the
smaller solution.

> As, you mentioned in a reply to the other thread of this thread (heh),
> coding really is the most insignificant part of a large programming project
> in a coporate environment. In theory at least, you should have a clear idea
> of what you're going to do before you do it, in which case the coding
> (almost irregardless of the language) is often times just grunt work.

You know the difference between theory and practice, do you? <wink>

No matter how clear an idea you have of what you're going to do, the
final code is not insignificant and it is the one thing which will be
there to be maintained for a long time. I have a lot of sympathy for
the XP position that the code is the only thing that really counts.

-- 
Christian Tanzer                                         tanzer at swing.co.at
Glasauergasse 32                                       Tel: +43 1 876 62 36
A-1130 Vienna, Austria                                 Fax: +43 1 877 66 92




From aahz at panix.com  Fri Mar 16 09:53:46 2001
From: aahz at panix.com (Aahz Maruch)
Date: 16 Mar 2001 06:53:46 -0800
Subject: list.sort(cmpfunc) question
References: <m1ae6lg5qt.fsf@pclabs188.labs.trema.com>
Message-ID: <98t9dq$qvu$1@panix6.panix.com>

In article <m1ae6lg5qt.fsf at pclabs188.labs.trema.com>,
Harri Pasanen  <harri at trema.com> wrote:
>
>The list.sort(cmp) docs for Python 2.0 state:
>
>   The sort() method takes an optional argument specifying a
>   comparison function of two arguments (list items) which should
>   return -1, 0 or 1 depending on whether the first argument is
>   considered smaller than, equal to, or larger than the second
>   argument.

Note that "should" carefully.  The current implementation of sort() only
checks for a negative value, but that's not a behavior that should be
relied on in your coding.  You can be sure that any anamolies in the
behavior of cmp() will be fixed if the sort() algorithm ever changes.
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

Three sins: BJ, B&J, B&J


From eye at niksula.hut.fi  Tue Mar 27 10:01:39 2001
From: eye at niksula.hut.fi (Petri Mikael Kuittinen)
Date: 27 Mar 2001 18:01:39 +0300
Subject: Regular expressions and non-standard character set
References: <ka8y9u01htt.fsf@cheerful.cs.hut.fi> <0XMt6.2503$4N4.321668@newsc.telia.net>
Message-ID: <ka8hf0fp86k.fsf@cheerful.cs.hut.fi>

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

> >>> import locale
> >>> locale.setlocale(locale.LC_ALL, "")
> 'Swedish_Sweden.1252'

I tried to find information about setting locale. The Linux man pages
nor the standard Python were not helpful on this matter.

I want to set the character set manually e.g. to ISO-8859-1 (ISO Latin
1). I tried something like:

locale.setlocale(locale.LC_CTYPE, "iso_88591_1")
locale.setlocale(locale.LC_CTYPE, "en_US.ISO8859-1")

Is there any good tutorial on locale? The best I could find is
http://www.uni-ulm.de/~s_smasch/Locale/

> >>> import re
> >>> re.findall(r"\b...\b", "spam, egg, bacon, and ???")
> ['egg', 'and']
> >>> re.findall(r"(?L)\b...\b", "spam, egg, bacon, and ???")
> ['egg', 'and', '???']

I tried the above. It didn't work under Python 2.0 under Windows 2000,
but it worked using pre instead of re (just like you said). The
problem is that is still doesn't solve my problem. ??? work, because
they are part of Finnish LC_CTYPE, but how do I get all the other
national characters work (german umlaut etc.) at the same time?


Petri

-- 
<(O)> Petri Kuittinen, also known as Eye, Dj Eye or Peku               <(O)>
<(O)> ADDRESS: Postipuuntie 10 A 14, FIN-02600 Espoo, Finland          <(O)>
<(O)> EMAIL: eye at iki.fi WWW: http://www.iki.fi/~eye/ PHONE: 09-5472380 <(O)>
~Steckel's Rule to Success: Good enough is never good enough. 



From phd at phd.pp.ru  Wed Mar 21 05:03:52 2001
From: phd at phd.pp.ru (Oleg Broytmann)
Date: Wed, 21 Mar 2001 13:03:52 +0300 (MSK)
Subject: OT: Re: re.match
In-Reply-To: <BjRt6.2586$4N4.345428@newsc.telia.net>
Message-ID: <Pine.LNX.4.30.0103211300400.15322-100000@fep132.fep.ru>

offtopic...

On Tue, 20 Mar 2001, Fredrik Lundh wrote:
> for some reason, this guy reminds me of the radio announcer who
> uttered "bob dylan, best known from travelling wilburys"...

   But I, personnaly, still do not know Dylan outside of TW :) And I
started to hear Roy Orbisson later, after Travelling Wilburys :))) (I mean
his earlier songs, of course :)

Oleg.
----
     Oleg Broytmann            http://phd.pp.ru/            phd at phd.pp.ru
           Programmers don't die, they just GOSUB without RETURN.




From thomas at xs4all.net  Fri Mar  2 19:47:25 2001
From: thomas at xs4all.net (Thomas Wouters)
Date: Sat, 3 Mar 2001 01:47:25 +0100
Subject: Intersesting Questions involving extending and embedding
In-Reply-To: <3aa0a4b8.44768661@news.digital.net>; from dev_null@email.com on Fri, Mar 02, 2001 at 01:52:15PM +0000
References: <3a9c6f89.9613214@news.digital.net> <0D1n6.15927$Qb7.2600092@newsb.telia.net> <3a9f0af7.37387355@news.digital.net> <508n6.15966$Qb7.2607706@newsb.telia.net> <3a9dc27e.1325053@news.digital.net> <SYln6.16080$Qb7.2629142@newsb.telia.net> <3a9f00e8.2826131@news.digital.net> <A7In6.16279$Qb7.2658942@newsb.telia.net> <3aa0a4b8.44768661@news.digital.net>
Message-ID: <20010303014725.T9678@xs4all.nl>

On Fri, Mar 02, 2001 at 01:52:15PM +0000, Dev_NuLL wrote:

> >Read the docs again, and look as the samples shipped with the
> >source distribution.

> Thanks for your responses, but they have been useless. From reading
> them it seems you didn't read my entire post or you could have
> answered the main problem I am having:

> " how to get the new module to directly connect to the 'Blender'
> module that is builtin the Blender executable."

> In other words how to write an extended module in C that can call
> functions in other compiled extended modules also in C.

Read the docs again. Really. Specifically, read
http://www.python.org/doc/ext/using-cobjects.html
which is the chapter headed
1.12 Providing a C API for an Extension Module

It contains everything you need to know. Use 'grep' in the modules dir and
you'll find three more examples.

-- 
Thomas Wouters <thomas at xs4all.net>

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



From kalle at gnupung.net  Thu Mar 22 16:09:28 2001
From: kalle at gnupung.net (Kalle Svensson)
Date: Thu, 22 Mar 2001 22:09:28 +0100
Subject: Python and emacs
In-Reply-To: <99dkg711894@news1.newsguy.com>; from aleaxit@yahoo.com on Thu, Mar 22, 2001 at 08:36:41PM +0100
References: <mailman.985287273.17896.python-list@python.org> <99dkg711894@news1.newsguy.com>
Message-ID: <20010322220928.A984@apone.network.loc>

Sez Alex Martelli:
> "Sean 'Shaleh' Perry" <shaleh at valinux.com> wrote in message
> news:mailman.985287273.17896.python-list at python.org...
> >
> > On 22-Mar-2001 Don Dwiggins wrote:
> > > A semi-idle question: has anyone looked into embedding Python as an
> > > extension language in Emacs, alongside Elisp?  I'm toying with the idea
> of
> > > doing some Emacs enhancements/modifications, and given the choice, I'd
> much
> > > rather do them in Python.  (One such enhancement would be to add
> > > Pythonwin-style "intellisense" popups to python-mode.)
> >
> > An editor that spoke python would not be a bad thing.  However, trying to
> get
> 
> Whaddya mean, 'would'?  VIM does that right now, and has for years
> (see www.vim.org)...

He said "an editor", not "a brain-damaged pile of user interface madness,
unusable for even the most simple text manipulation".  Flamewar!!! <wink>

Peace,
  Kalle
-- 
Email: kalle at gnupung.net     | You can tune a filesystem, but you
Web: http://www.gnupung.net/ | can't tune a fish. -- man tunefs(8)
PGP fingerprint: 0C56 B171 8159 327F 1824 F5DE 74D7 80D7 BF3B B1DD
 [ Not signed due to lossage.  Blame Microsoft Outlook Express. ]



From tim.one at home.com  Sat Mar 17 04:50:30 2001
From: tim.one at home.com (Tim Peters)
Date: Sat, 17 Mar 2001 04:50:30 -0500
Subject: Pep 238 Why new // operator (Non-integer Division)
In-Reply-To: <98v77t01qcg@news2.newsguy.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCIEEJJGAA.tim.one@home.com>

[Alex Martelli]
> So what's so terrible about non-infix notation, a builtin div(a,b) [and
> mod(a,b) too since we're at it...?].  Sure, infix can be handy --

Infix is also a speed issue in Python, since syntax is fixed but builtin
names are not reserved; so the speed freaks will surely argue for syntax.

> but then this might be the occasion to steal a leaf from Haskell's
> book... let any 2-operand function be used as an infix operator
> (possibly with some stropping if that is needed to disambiguate
> syntax).

I'll let someone else channel Guido on this one.  Who's up to the challenge?

    x delattr "attr"
    lamdba x: x%2 filter x
    "channel.txt" open "w"
    3 complex ("beef" int 16)

it's-often-not-pretty-in-haskell-either-ly y'rs  - tim




From cthurner at web.de  Tue Mar 27 04:02:43 2001
From: cthurner at web.de (Christian Thurner)
Date: Tue, 27 Mar 2001 09:02:43 +0000
Subject: HTTP redirect via Python?
References: <SNVv6.128157$tP3.2040832@news1.rdc1.bc.home.com>
Message-ID: <99pduj$30lp$1@news.nikoma.de>

John E. Barham wrote:

> Apologies if this is too basic a question, but I'm wondering how I can
> send a redirect/URL rewrite command to a browser via Python.
Just do the following:

print "Location: http://your.safe.url\n\n"

that's all

CU
Christian




From BrianQ at ActiveState.com  Wed Mar 28 12:25:55 2001
From: BrianQ at ActiveState.com (Brian Quinlan)
Date: Wed, 28 Mar 2001 09:25:55 -0800
Subject: list vs tuple
In-Reply-To: <XFMail.20010328081610.shaleh@valinux.com>
Message-ID: <000e01c0b7ac$25f9f510$b503a8c0@activestate.ca>

I'm not sure that this is a reasonable piece of self documentation since
dir_list can change. It is the tuple ('/path', '/foo/bar', '/baz/bat/files')
that cannot change. Would you say that x = 5 documents that x cannot change?

-----Original Message-----
From: python-list-admin at python.org
[mailto:python-list-admin at python.org]On Behalf Of Sean 'Shaleh' Perry
Sent: Wednesday, March 28, 2001 8:16 AM
To: deadmeat
Cc: python-list at python.org
Subject: RE: list vs tuple



On 28-Mar-2001 deadmeat wrote:
> Whats the practical difference between a list[] and a tuple() ?
>
> A tuple seems to be a very basic list, so unless it's faster, what's the
> point of it?
>

As others point out they are immutable.  This can lead to self documenting
code.

dir_list = ('/path', '/foo/bar', '/baz/bat/files')

dir_list shouldn't change and this documents it.

--
http://mail.python.org/mailman/listinfo/python-list




From aleaxit at yahoo.com  Mon Mar 19 01:48:53 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Mon, 19 Mar 2001 07:48:53 +0100
Subject: Who am I: can a class instance determine its own name?
References: <mailman.984971963.29821.python-list@python.org>
Message-ID: <994a1n01gtq@news2.newsguy.com>

"Carlos Ribeiro" <carribeiro at yahoo.com> wrote in message
news:mailman.984971963.29821.python-list at python.org...
>
> Alex Martelli wrote:
> >(My dream is actually of a collection of bits in an attribute...
>
> While I agree on the *principle*, I disagree on the details. Thinking
about

Yeah, yeah, you're right: the dream is actually about an attribute
that is a set of atoms; that they're representes as bits, or not, is
not essential to the dream.

> this as a bitmap is an artifact of historic optimization techniques. Worse

The isomorphism between sets on a universe of N atoms and N-bit
integers is hardly 'an artifact' -- it's a simple mathematical reality
which _enables_ certain 'optimizations' (e.g., enumerating all such
sets by enumerating all such integers).

> OTOH, having a set of attributes would make the code more readable and
> extensible. (the set of attributes is itself in a dictionary and would
also
> be subject to the same control).

Once we do start talking about implementation details, the issues
shift.  Python has no builtin 'set' type, so any set must be mapped
by some other type -- be it an integer, dictionary, string, whatever.
And I disagree with your contentions about what might be the
best mapping of this particular set.

I see no difference in readability between:

    adict.unchangeable |= 1

and

    adict.unchancheable[1] = 1

for "adding an element to the set" (in either case, of course, you
could use a symbolic name for the 1 that represents the atom --
no difference here).  When I want to add several elements at
the same time,

    adict.unchangeable |= ( 1 | 2 )

(or symbolic-named equivalent) has it all over alternatives a
dictionary (currently) affords.

Extensibility (to a set from a universe of N+n atoms) is clearly
equally feasible in either case.  And the 'nesting' whereby each
dictionary holds a dictionary (and so on ad infinitum) is complex
and tricky -- "dictionaries all the way down" can't be really
implemented, of course, so that some kind of magic would have
to be used to break the chain at some point; give me the utter
simplicity of the bitmap-representation of sets over that...!


Alex





From NoSpam at NoSpam.com  Sat Mar  3 11:38:00 2001
From: NoSpam at NoSpam.com (Tom)
Date: Sat, 03 Mar 2001 16:38:00 GMT
Subject: how to get the last interactive command enetered
References: <F2Tn6.15639$UZ4.4310299@news4.rdc1.on.home.com> <j4bsrjxjy3.fsf@informatik.hu-berlin.de>
Message-ID: <I59o6.19418$UZ4.5669336@news4.rdc1.on.home.com>

"Martin von Loewis" <loewis at informatik.hu-berlin.de> wrote in message
news:j4bsrjxjy3.fsf at informatik.hu-berlin.de...
> "Tom" <NoSpam at NoSpam.com> writes:
>
> > For example, say that I set a trace function.  I find the frame's code
> > object, look at co_filename and find that it is <stdin>.  But I want to
know
> > what the actual command was.  Is there anyway to do this (in C or
Python)?
>
> No. Once the code is executed, the source code is dropped. All you can
> get at is the binary code.

But when I catch a 'call' trace event the code has not been executed.  I
believe that it has been parsed and compiled but not executed (till I get
the 'return' trace event).

> Of course, "in C", you can rewrite the Python interpreter to behave
> differently. It takes more than an extension module to do so, though.

I am avoiding changing python itself.  I want to do this from 'the outside'.

Thanks for your response,

Tom.
> Regards,
> Martin
>




From frank at quantiva.com  Thu Mar  1 17:02:28 2001
From: frank at quantiva.com (Frank Stolze)
Date: Thu, 01 Mar 2001 17:02:28 -0500
Subject: [OT - JOB AD] Python / XML / Distributed Systems Developer
Message-ID: <5.0.2.1.0.20010301163601.00a29af0@pop3.norton.antivirus>

Sorry for the off-topic post. We are a well-funded, stealth mode startup in the
network service management field. We are building a novel, distributed system
and service that involves Internet protocol implementations (HTTP, SMTP, POP3,
DNS, ping, etc.), statistical analysis, some AI, database storage and 
reporting,
as well as issues such as routing, load balancing, fail-over, firewalls, etc.

Almost all of the implementation is being done in Python. We also use XML,
XML-RPC, HTTP tunneling and a few new things. We are looking for two
enthusiastic Python & networking gurus to join a small team of hands-on
people to help us implement a great vision!

The "official" job description is below.


Regards,
Frank

===============
Company Profile:

We're onto something really big in network service management.  We have
VC backing and a clear view of the future.  Our stealth situation hides our
long history in this area and lets us focus on the goal.  Join the team of
innovative, visionary, enthusiastic, and passionate people that will shape
our future.  Share the opportunity to work with industry leading experts on
the design and creation of a next generation system.

Quantiva is looking for people who can quickly grasp new concepts,
develop new, original solutions to existing problems, and in general can
"hit the ground running."

Quantiva is located in Princeton, New Jersey.

Please send resumes to techjobs at quantiva.com.


Job Description:

The Network Software Engineer will contribute to the design and development
Quantiva's network service management software. This position will involve
active participation in the design, architecture, and implementation of the 
product.


Requirements:
?       Strong software development skills in a Unix environment.
?       Strong network and distributed systems programming experience. Good 
knowledge of network protocols such as TCP/IP, HTTP, DNS, SMTP, POP3 and 
concepts such as firewalls, routing required.
?       4+ years of hands-on experience with scripting languages such as 
Perl, Python or Tcl. Python proficiency is required.
?       Applications programming experience in object-oriented languages 
such as C++ or Java.
?       DBMS experience including database schemas, SQL, and database 
programming.
?       Solid Unix experience (Solaris preferred) in a commercial environment.
?       XML experience is a plus.
?       BS/MS/PhD in CS, EE, CE or related field.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20010301/5b8b2a6a/attachment.html>

From bbollenbach at homenospam.com  Mon Mar  5 23:23:29 2001
From: bbollenbach at homenospam.com (Brad Bollenbach)
Date: Tue, 06 Mar 2001 04:23:29 GMT
Subject: Can Python replace Visual Basic? Should it?
Message-ID: <5DZo6.11126$hn5.1578410@news1.rdc1.mb.home.com>

Okay, sure this has been asked again and again, but I've never really seen a
clear answer to this question:

    Can Python replace Visual Basic? Should it?

As far as I can see, they both seem close to equal in their not-so-simple
ways of distributing apps to end-users, which VB programmers have seem to
overcome to some degree and so Python programmers should and probably do
likewise. What I want is a programming language that can make Windows fun
(like C or Perl does Unix), without having to shell out the big bucks for
full installs of VB or VC++. Though I don't believe in using price as an
important criteria in deciding what language I'll like (unless of course,
it's Powerbuilder, which is over $4000 :), I still have to admit that I like
what Python is about, and the benefits of an Open Source language are
obvious. As I work with Powerbuilder in my day job, I've also realized that
"languages" that are based on an IDE are not really my thing (I hate having
a 40,000 foot view of what I'm doing).

But really, very specifically, were YOU a former Visual Basic (or
Powerbuilder, or Delphi) programmer that is now using Python to write the
complex GUI's VB used to handle for you? I'm an on again, off again, student
of Python, who really wants to give this a go, if there's a good chance that
I'll get the return on my investment: a programming tool that can take over
where VB, Powerbuilder, and Delphi used to play.

Again, I'm much more interested in the real world experience of those who
used to use the big visual development tools in Windows like VB, Delphi or
Powerbuilder rather than theoretical answers.

Thanks!

Brad Bollenbach




From cce at clarkevans.com  Fri Mar 23 16:51:10 2001
From: cce at clarkevans.com (Clark C. Evans)
Date: Fri, 23 Mar 2001 16:51:10 -0500 (EST)
Subject: static methods
Message-ID: <Pine.LNX.4.21.0103231640070.5553-100000@clarkevans.com>

I was wondering if there is any rationale as to why 
Python classes do not have static methods?  

I was also wondering if it would make sence to allow
None to be used as self?  For example:

    class MyClass:
        def myStatic(static, str): print str
 
    MyClass.myStatic(None,'my-first-argument')

Why would you want to do this?  Let's say I want
a "check" method, which checks to see if an object
from another class can "pass off" as an object
in the current class (see Paul Prescod's work
on the type special interest group list).

    class MyClass:
        def func(str): print str
        def __check__(static,obj):
            if getattr(obj,"func"): return obj
            return None

    class LikeMyClass:
        def func(str): print str


    x = LikeMyClass()
    MyClass.__check__(None,x)

Anyway... how problematic is allowing None to be used
in "unbound" methods in order to 'simulate static methods?

Thanks!

Clark





From dkuhlman at rexx.com  Tue Mar 20 18:49:42 2001
From: dkuhlman at rexx.com (Dave Kuhlman)
Date: Tue, 20 Mar 2001 23:49:42 -0000
Subject: Functionalism, aesthetics Was:(RE: I come to praise .join, not
References: <mailman.985052844.17993.python-list@python.org>
Message-ID: <985132196.251112@rexx.com>

Carlos Ribeiro <carribeiro at yahoo.com> wrote:
> 
> 

[snip]

> 
> To summarize it: a  *technically* optimal solution for a problem is not 
> always the *best* solution. If it does not look intuitively good for most 
> users, its probably a bad solution.
> 
> Its unfortunate that in many cases the non-expert user keeps quiet, 
> assuming that whatever opinion the "master" haves, its going to be better 
> than theirs - after all they are the so-called experts.
> 
> Carlos Ribeiro
> 



It is very hard to predict the consequences of design changes to
complex things. Python is mature in the sense that it has learned
from these previous mistakes, rather than having to learn from
making them itself.

I believe that a study of the REXX programming languages might be
instructive here.  See:

    The REXX Language:A Practical Approach to Programming
    Michael Cowlishaw 

Cowlishaw claims in the first few chapters that the early design of
REXX was done by users and that many of these users were not
programmers.  I think it shows.  Concepts that we take for granted
were unknown, apparently to these users.  For example, parameters
are passed to functions as a string that must be parsed; functions
cannot be called recursively without tricks to "stack" parameters.

I'm sure that it has been said before on this newsgroup:
Programming language design is very hard.  Making something simple,
without losing power, is very hard.  People have worked very hard
on the design of Python, and it shows. On a graph that plots the
simplicity against the power of programming languages, Python is in
the upper right hand corner.

Users of Python are programmers.  They may even be, like me, users 
of a number of programming languages.  They might have, as I do, 
lots to say about features that they like and dislike and that are 
or are not it this or that language, *without* being experts on 
programming language *design*.  Asking *these* (programmer) users 
about language features is quite a bit different from asking non-
programmers.

[Don Cave says:]
> Anyway, as long as we're talking about things that we spend our
> days doing, we certainly do have the capacity for sound judgement.
> It doesn't mean our judgement is therefore infallibly sound, but
> the collective judgement of an established culture brings a kind
> of sense to bear that can dwarf a brilliant appeal to abstract
> principles.

I'd add that it can also dwarf the intuitions of non-users and non-
experts.  Non-users often favor things which, with a little 
experience and education, they would regret.

Designing a programming language for non-programmers (and their
intuitions) is as foolish as designing a house for cave-dwellers. 
Someone who has only lived in a cave has not learned what is
necessary in order to appreciate good house design.  (Hint: they
may not know what a bathroom is.) Since we have so few cave
dwellers among us, we sometimes forget this.  (But just in case I'm
about to be ripped for being politically uncorrect about cave
dwellers, ... Some of my best friends are cave dwellers. Cave
dwellers are as a group as smart and as good looking as non-cave
dwellers.  Etc.)

  - Dave

-- 
Dave Kuhlman
dkuhlman at rexx.com


From aahz at panix.com  Sat Mar  3 16:25:06 2001
From: aahz at panix.com (Aahz Maruch)
Date: 3 Mar 2001 13:25:06 -0800
Subject: Missing the functional mind set
References: <mlb2at8ptrr7sh5dgq8qlk2201kquh6tfl@4ax.com>
Message-ID: <97rnfi$37c$1@panix3.panix.com>

In article <mlb2at8ptrr7sh5dgq8qlk2201kquh6tfl at 4ax.com>,
Daniel Klein  <danielk at aracnet.com> wrote:
>The purpose of this code is to return a list of dictionary values sorted by
>key, resulting in 'templist'.
>
>>>> d = {3: 'd', 2: 'b', 1: 'a', 0: 'z'}
>>>> templist = []
>>>> for n in range(len(d)):
>	templist.append(d[n])

For a somewhat more general solution that's slightly less efficient, sort
d.items().
-- 
                      --- Aahz (Copyright 2001 by aahz at pobox.com)

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

Nostalgia just ain't what it used to be


From fdrake at acm.org  Sat Mar  3 14:30:13 2001
From: fdrake at acm.org (Fred L. Drake, Jr.)
Date: Sat, 3 Mar 2001 14:30:13 -0500 (EST)
Subject: [Python-Dev] Re: Bug fix releases
In-Reply-To: <200103031910.OAA21663@cj20424-a.reston1.va.home.com>
References: <LNBBLJKPBEHFEDALKOLCMEEPJBAA.tim.one@home.com>
	<mailman.982897324.9109.python-list@python.org>
	<200103031821.NAA24060@panix3.panix.com>
	<200103031910.OAA21663@cj20424-a.reston1.va.home.com>
Message-ID: <15009.17989.88203.844343@cj42289-a.reston1.va.home.com>

Guido van Rossum writes:
 > I wonder, does that extend to new library modules?  Is there also
 > resistance against the growth there?  I don't think so -- if anything,
 > people are clamoring for more stuff to become standard (while at the

  There is still the issue of name clashes; introducing a new module
in the top-level namespace introduces a potential conflict with
someone's application-specific modules.  This is a good reason for us
to get the standard library packagized sooner rather than later
(although this would have to be part of a "feature" release;).

 > Wait a minute!  Now you're making it too complicated.  Betas of bugfix
 > releases?  That seems to defeat the purpose.  What kind of

  Betas of the bugfix releases are important -- portability testing is
fairly difficult to do when all we have are Windows and Linux/x86
boxes.  There's definately a need for at least one beta.  We probably
don't need to lengthy, multi-phase alpha/alpha/beta/beta/candidate
cycle we're using for feature releases now.

 > It's not too late, as I mentioned.  We'll also do this for 2.1.

  Managing the bugfix releases would also be an excellent task for
someone who's expecting to use the bugfix releases more than the
feature releases -- the mentality has to be right for the task.  I
know I'm much more of a "features" person, and would have a hard time
not crossing the line if it were up to me what went into a bugfix
release.

 > BTW, See you at the conference!

  If we don't get snowed in!


  -Fred

-- 
Fred L. Drake, Jr.  <fdrake at acm.org>
PythonLabs at Digital Creations




From mark at chaos.x-philes.com  Tue Mar 20 16:36:35 2001
From: mark at chaos.x-philes.com (Mark Carroll)
Date: Tue, 20 Mar 2001 21:36:35 GMT
Subject: Standalone executables
Message-ID: <998igr$4al$1@niagara.nonet>

I'm investigating the possibility of using wxPython for development.
However, I need to be able to ship single standalone executables to
random users of 32-bit MS Windows and Linux. (It would be really nice
if I could compile both on Linux with mingw32 or suchlike. (-:)

What are the current options, and how about the near future? I'm not
having an easy time figuring out where things stand with freeze, etc.

Thanks.

-- Mark


From grante at visi.com  Tue Mar  6 11:53:45 2001
From: grante at visi.com (Grant Edwards)
Date: Tue, 06 Mar 2001 16:53:45 GMT
Subject: absolute namespace for callbacks?
References: <3AA50AFB.83C7459F@troikanetworks.com> <983895797.296667@newsmaster-04.atnet.at> <3AA51144.3DD3FC3F@troikanetworks.com>
Message-ID: <tC8p6.1048$y6.188875@ruti.visi.com>

In article <3AA51144.3DD3FC3F at troikanetworks.com>, Bruce Edge wrote:

>My callback func is defined in one module, and the callback is executed
>from another module which does not know in advance what module the
>callback func resides in. So, by default, the caller cannot find the
>callback func in it's own namespace.
>The callback registration procedure needs to specify the namespace where
>the callback func exists.
>
>OTOH, maybe I'm missing something here :-)

You want to register a callback function, but you want to do it
in a module that doesn't know what callback function you want
registered?

-- 
Grant Edwards                   grante             Yow!  When you get your
                                  at               PH.D. will you get able to
                               visi.com            work at BURGER KING?


From skip at mojam.com  Tue Mar  6 07:21:00 2001
From: skip at mojam.com (Skip Montanaro)
Date: Tue, 6 Mar 2001 06:21:00 -0600 (CST)
Subject: gah! I hate the new string syntax
In-Reply-To: <MWXo6.19412$ML1.974917@e420r-atl2.usenetserver.com>
References: <97ohh2$polt0$1@ID-77331.news.dfncis.de>
	<XFMail.20010302085406.shaleh@valinux.com>
	<mailman.983843650.6177.python-list@python.org>
	<MWXo6.19412$ML1.974917@e420r-atl2.usenetserver.com>
Message-ID: <15012.54828.258632.426985@beluga.mojam.com>

    >> The presence of join methods for strings and why that's where they
    >> landed has been hashed and rehashed several times.  Please check out
    >> FAQ 4.96 for details.

    Steve> This isn't really fair, since I wrote 4.96 in response to this
    Steve> thread.

My apologies.  I was out of town for a few days.  Upon seeing this thread I
decided I would create a FAQ entry, but an initial search turned up 4.96.
In my rush to wave it in front of people I failed to notice the date on it.

Skip



From aleaxit at yahoo.com  Thu Mar  1 11:29:48 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 1 Mar 2001 17:29:48 +0100
Subject: Unit testing data; What to test
References: <Pine.SOL.4.30.0102251735030.6152-100000@mimosa.csv.warwick.ac.uk>
Message-ID: <97ltdr02vf@news2.newsguy.com>

"John J. Lee" <phrxy at csv.warwick.ac.uk> wrote in message
news:Pine.SOL.4.30.0102251735030.6152-100000 at mimosa.csv.warwick.ac.uk...
>
> Sorry about the formatting etc. of this post -- I had to cut and paste
> from Google.

My sympathy.  I'll even forgive you're not mentioning that
I'm the guy whose comments you're responding to:-).

>
> > "John J. Lee" <phrxy at csv.warwick.ac.uk> wrote in message
> > news:Pine.SOL.4.30.0102160309090.7729-100000 at mimosa.csv.warwick.ac.uk...
> [...]
> > If you keep the (stimuli, expected_responses) sets separate from the
> > module being tested, you gain much the same benefits (and pay much
> > the same costs) as by keeping documentation separate from code for
> > other kinds of docs (think of tests as a form of executable docs...!).
> >
> > Keeping some tests/docs together with the code simplifies distribution
> > issues: anybody who has your code also has this minimal set of tests
> > (and other docs).
>
> Well, that's not a problem, is it (unless the tests really are huge)?
> They can be in the distribution without being mixed in with the code.

They can be, sure -- just like other documentation can be in your
distribution without being mixed with your code.  Again, I claim
the parallel is very strict.  Maybe people are *supposed* to only
ever distribute your code in the way you packaged it up -- but,
what happens if, e.g., some automated dependency finder picks up
your wonderful bleeppa.py, and stashes it into allyouneed.zip
*without* the accompanying bleeppa.doc and bleeppa_tests.py files,
for example?  Answer: whoever ends up rummaging some time later
in the unpacked allyouneed.zip WILL have your bleeppa.py but
none of the other files that you originally packaged with it.

So, if bleeppa.py itself contains some minimal/essential subset
of its own docs (as docstrings or comments) and unit-tests, you
are covering up for exactly such problems -- making your code
more usable when it happens to start going around without other
files that _should_ always accompany it.


> > However, in some cases, the total amount of tests
> > and other docs can easily swap the amount of code -- and in this case
> > keeping them together can easily be thought of as "too costly". The
> > code can become practically unreadable if there are an order of
> magnitude
> > more lines of docs (including tests) than lines of actual source in a
> .py.
>
> Agreed.  My tests are separate from the code -- it hadn't occurred to me
> to keep them in the same file.  It was the separation (or not) of test
> data and test code I was wondering about, though.

I guess I tend not to think of that because my typical 'test _code_'
IS separated from my typical 'test _data_' anyway -- Tim Peters'
doctest.py being the former:-).  So, I don't think of merging test
code and test data, any more than I think of distributing, say, Word
together with my .doc files:-).

Or, if you look at the docstrings which doctest.py runs as "code"
rather than "data", then I guess the issue I have not faced (in
Python) is that of _separating_ the test-data from the code.

When I _do_ face it in other languages (in some database-centered
components: the DB on which the tests are to run is unconscionably
BIG, if you think of it in any textual form in which it might be
feasible to merge it with the relatively small amount of code that
needs it), the testcode/testdata separation happens to be well
near inevitable.  The code carries with it the SQL code (as data)
that it sends to the DBMS and the results it expects, but not the
DB itself -- so, it's not really feasible to test the component
(run its standard unit tests, I mean) if you only have its executable
part without the accompanying files (since the starting DB needed
for the tests is just such an accompanying file).


> > My favourite approach (and I do wish I was more systematical in
> > practising what I preach:-) is to try and strike a balance: keep with
> > the code (docstrings, comments, special-purpose test functions) a
> > reasonable minimal amount of tests (& other docs) -- indicatively,
> > roughly the same order of magnitude as the source code itself;
> > but _also_ have a separate set of docs (& tests) for more extensive
> > needs.
> [...]
>
> This is all good advice, but it doesn't actually answer my question.  ;-)
> Anyway, I think I will try having some tests in the code, sounds like a
> great idea considering the tests-as-docs aspect.

Serendipity, pure serendipity:-).  I misread your question, or you
miswrite it (or 50-50 -- comes to the same thing in the end:-), yet
you still gain something useful from the resulting discussion...:-).


> > you release an application that's about 1,000 SLOC, it does not
> > really matter much if your process and approach are jumbled; at
>
> Let me assure you it _is_ possible to achieve an unmaintainable mess in
> ~1000 SLOC.  I have personally achieved this in, let's see (C-x C-f
> munge.pl RET) ... 832 SLOC.  This is one of many things that Perl makes

Oh, I'm sure I have beaten that, a LONG time ago back when I
wrote in APL -- but what I'm saying it's not really the _process_
that controls that, for a small-enough deliverable.  Good taste,
experience, and some common sense on the coder's part may suffice,
with just-about-nonexistent-process, to keep things under control
_at this level of code-size_.

> easy (this is completely unfair to Perl of course -- what really makes it
> easy is starting out thinking 'this is going to be a 100 line script' and
> then extending by cut and paste -- I've learned my lesson, honest!).
>
> > 10,000 SLOC, it's a problem; at 100,000, an utter nightmare, so
> > you HAVE to be precise in defining your process then (or, call it
>
> Yeah, of course, I understand that.  But there is still a place for lots
> of testing for smaller efforts, minus the heavyweight formal process.

Absolutely YES.  Testing ain't a bad thing even if done informally!-)


> > 100, 1000, 10,000 FP -- but it's really about SLOC more than it
> > is about FP, which is why higher-level languages help so much).
> >
> > Differently from what XP specifies, I think tests should be in two
> > categories (and, similarly, so should deliverables be, rather than
> > the just-one-deliverable which so characterizes XP -- that is a
> > tad too X for my conservative self, even though I buy into 70+%
> > of XP's tenets!).
>
> Which two categories?  External and internal?  How is this different from
> the XP unit test versus acceptance tests division?  Is your point just
> that the external / internal division applies on more levels than just
> final user / everything else -- which is what you seem to be talking about
> below?

Basically, yes, and I apologize for muddled expression.  The point is
that most often I build components that no 'final' user will ever
notice (unless something goes badly wrong:-) -- call it 'middleware'
or whatever.  But not-so-final users may need to get at parts of
them -- not just future maintainers/extenders, and current and future
re-users, but (at least when scripting is possible:-) current and
future _scripters_ (some 'power-users', 3rd party system integrators,
customer-support/application-engineers, ...).  What's "internal" and
what's "external" varies depending on the target audience.


> here's a snip explaining what XP people mean by acceptance tests for
> anybody that hasn't read about it:
>
> http://www.extremeprogramming.org/rules/functionaltests.html
    [snip, but the whole site IS a recommended read:-)]


Alex





From drew.csillag at starmedia.net  Wed Mar 21 18:27:16 2001
From: drew.csillag at starmedia.net (Drew Csillag)
Date: Wed, 21 Mar 2001 18:27:16 -0500
Subject: Simple regex that has me banging my head against a wall
In-Reply-To: <20010321143144.K24341@trufflehunter.avalongroup.net>; from tjg@exceptionalminds.com on Wed, Mar 21, 2001 at 02:31:44PM -0800
References: <20010321143144.K24341@trufflehunter.avalongroup.net>
Message-ID: <20010321182715.B1234@naxos.skunk.org>

On Wed, Mar 21, 2001 at 02:31:44PM -0800, Timothy Grant wrote:
> Hi again.
> 
> I actually thought I had this regex thing down pretty well, but
> I'm getting beat up by what I thought should be a simple one.
> 
> >>> x = '$1$asdfjke$lkjasdkjfsdasdfj'
> >>> re.sub(r'\$', r'\\\$', x)
> '\\\\$1\\\\$asdfjke\\\\$lkjasdkjfsdasdfj'
> 
> All I want to do is substitute \$ for $.

How about:
Python 2.1b1 (#1, Mar  4 2001, 11:31:25) 
[GCC 2.95.2 19991024 (release)] on linux2
Type "copyright", "credits" or "license" for more information.
>>> x = '$1$asdfjke$lkjasdkjfsdasdfj'
>>> import string
>>> print string.replace(x, '$', r'\$')
\$1\$asdfjke\$lkjasdkjfsdasdfj


-- 
Drew Csillag                                              
Acting Manager of Platform Engineering
drew.csillag at starmedia.net        Cell: (917)-679-8827  
StarMedia Network, Inc.           Land: (212)-905-8332 



From stephen_purcell at yahoo.com  Mon Mar 12 06:48:10 2001
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Mon, 12 Mar 2001 12:48:10 +0100
Subject: Problems using python.sty to document a module
In-Reply-To: <m31ys3qle8.fsf@vole.object-craft.com.au>; from djc@object-craft.com.au on Mon, Mar 12, 2001 at 10:25:35PM +1100
References: <m31ys3qle8.fsf@vole.object-craft.com.au>
Message-ID: <20010312124810.A16146@freedom.puma-ag.com>

Dave Cole wrote:
> Does anyone know how I can document a module which is named with a
> leading underscore?  I would prefer not to have to rename the module.

Presumably you can find out by looking at the source docs for the __main__
and __builtin__ modules.

-Steve

-- 
Steve Purcell, Pythangelist
Get testing at http://pyunit.sourceforge.net/
Any opinions expressed herein are my own and not necessarily those of Yahoo



From dmaas at nospam.dcine.com  Wed Mar 28 16:35:40 2001
From: dmaas at nospam.dcine.com (Dan Maas)
Date: Wed, 28 Mar 2001 21:35:40 GMT
Subject: Setting sleep interval less than 10 mS on a Linux PC?
References: <3AC22D31.FD4644B1@spammyraqia.com>
Message-ID: <MOsw6.7387$8R5.921141@typhoon.nyroc.rr.com>

In article <3AC22D31.FD4644B1 at spammyraqia.com>, "David Lees"
<DavidlNo at spammyraqia.com> wrote:

> I suspect this is a system question, but I will ask anyway.  I am using
> the Python sleep method (time.sleep) and on a PC running RedHat 7.0
> Linux,  it has a 10 millisecond lower limit.  Below that it still wakes
> up every 10 milliseconds, regardless of the argument.

This is a nasty problem - time.sleep() uses select() internally, and on
Linux select() cannot sleep for less than 10ms.

Easy way out - just use a busy loop:

start_time = time.time()
while (time.time() - start) < 0.001:
	pass

time.time() has microsecond accuracy so you will get very close to the
desired delay. Of course this burns CPU time, but with a 1ms polling
interval I assume you're not too worried about that.

Hard way out: find or write a python module that uses nanosleep()
internally; I *think* nanosleep() can give you <10ms delays without
busy-waiting. (you may need a kernel patch though :[...)

Regards,
Dan


From daniel.orlowski at bg-sys.de  Thu Mar  1 07:57:44 2001
From: daniel.orlowski at bg-sys.de (Daniel Orlowski)
Date: Thu, 01 Mar 2001 13:57:44 +0100
Subject: ignore - just a test
Message-ID: <3A9E4748.38C1F5ED@bg-sys.de>

it's just a test



From loewis at informatik.hu-berlin.de  Mon Mar  5 05:24:14 2001
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 05 Mar 2001 11:24:14 +0100
Subject: Can SAX parser return <!-- COMMENTS -->  ???
References: <MWxn6.15038$x27.906254@news1.rdc2.on.home.com> <j4d7c0jh03.fsf@informatik.hu-berlin.de> <wzwva5shuy.fsf@sunshine.cs.uu.nl>
Message-ID: <j48zmko6nl.fsf@informatik.hu-berlin.de>

Piet van Oostrum <piet at cs.uu.nl> writes:

> MvL> It is a matter of ongoing debate whether applications should even deal
> MvL> with comments; if an application needs XML comments for proper
> MvL> operation, something is wrong with that application.
> 
> What about a prettyprinter? Doesn't that count as an application?

Sure. But why does it have to preserve comments?

Regards,
Martin


From olczyk at interaccess.com  Mon Mar  5 08:13:19 2001
From: olczyk at interaccess.com (Thaddeus L. Olczyk)
Date: Mon, 05 Mar 2001 13:13:19 GMT
Subject: Python 3000.  Here we go again.
References: <Pine.GSO.4.21.0101142011080.14128-100000@y.glue.umd.edu>
Message-ID: <3aa690a9.398727515@nntp.interaccess.com>

On Sun, 14 Jan 2001 20:19:34 -0500, Roy Katz <katz at Glue.umd.edu>
wrote:

I would like to see aspects added. Maybe call it AspectPy?
If not that then at least full MOP features ( comes close, but not
really there ).


From loredo at spacenet.tn.cornell.edu  Mon Mar 12 18:29:03 2001
From: loredo at spacenet.tn.cornell.edu (Tom Loredo)
Date: Mon, 12 Mar 2001 18:29:03 -0500
Subject: import headaches/gripes
References: <3aaa3f94.15940553@News.CIS.DFN.DE> <tegkatsj8gdmmoctjqivgfdjq7gckfg41h@4ax.com>
Message-ID: <3AAD5BBF.5AF14631@spacenet.tn.cornell.edu>

Hi folks-

> I think the problem is that PythonWin sets up the Python environment
> in the beginning and everything runs under that environment.

The same annoying problem happens with IDLE and with the Mac IDE.  What
I'd love is a menu item or button that could "clean up" the environment,
so that the next time I ran the code, it was as if I had restarted the
IDE.  Is this inherently impossible?

-Tom Loredo


From fdrake at acm.org  Sat Mar  3 14:53:45 2001
From: fdrake at acm.org (Fred L. Drake, Jr.)
Date: Sat, 3 Mar 2001 14:53:45 -0500 (EST)
Subject: [Python-Dev] Re: Bug fix releases
In-Reply-To: <LNBBLJKPBEHFEDALKOLCEELHJDAA.tim.one@home.com>
References: <15009.17989.88203.844343@cj42289-a.reston1.va.home.com>
	<LNBBLJKPBEHFEDALKOLCEELHJDAA.tim.one@home.com>
Message-ID: <15009.19401.787058.744462@cj42289-a.reston1.va.home.com>

Tim Peters writes:
 > Note there was never a bugfix release for 1.5.2, despite that 1.5.2 had some
 > serious bugs, and that 1.5.2 was current for an unprecedentedly long time.
 > Guido put out a call for volunteers to produce a 1.5.2 bugfix release, but
 > nobody responded.  Past is prelude ...

  And as long as that continues, I'd have to conclude that the user
base is largely happy with the way we've done things.  *If* users want
bugfix releases badly enough, someone will do them.  If not, hey,
features can be useful!


  -Fred

-- 
Fred L. Drake, Jr.  <fdrake at acm.org>
PythonLabs at Digital Creations




From loewis at informatik.hu-berlin.de  Fri Mar  9 10:48:39 2001
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 09 Mar 2001 16:48:39 +0100
Subject: Problems with C API
References: <mailman.984144013.2277.python-list@python.org>
Message-ID: <j4n1avlz8o.fsf@informatik.hu-berlin.de>

David OGwynn <dog at ERC.MsState.Edu> writes:

> Everything works fine. But, when I try to run the exact same code
> through a script file (or through pdb) it segfaults on me. I'm
> totally lost. I really need this functionality built into my Octree
> object, given the large (30M or more) size of the binary files...

My guess is that you made a refcounting error somewhere; this is the
most frequent bug in C modules. Try running Python in a C debugger and
see where it crashes.

Regards,
Martin



From gerhard.nospam at bigfoot.de  Sat Mar 24 00:46:03 2001
From: gerhard.nospam at bigfoot.de (Gerhard =?iso-8859-1?Q?H=E4ring?=)
Date: Sat, 24 Mar 2001 06:46:03 +0100
Subject: Installing python on a virtual server account.
References: <3ABB8549.1987BA68@sewardweb.com>
Message-ID: <3ABC349B.2E85867@bigfoot.de>


Lloyd Sommerer wrote:
> 
> Greetings,
> 
> I am at a bit of a lose.  I've been using python for about a month to
> write scripts to handle CGI type things on a website.  My hosting
> company has python 1.5.1 installed and says they will not be upgrading
> (what they actually said was, "We are removing all references to Python
> from our website and are offically no longer supporting it. When we
> tried to upgrade to the newest version, it failed.").  They are leaving
> 1.5.1 in place, but I would prefer to use the current version for my
> next project.  I would rather not switch hosting companies.
> 
> Does anyone have experience installing 2.0 in a shared hosting
> situation?  Even telling me that I can't do it and that I should give up
> would  be helpful at this point.
> 
> I am on a SunOS 5.7 machine.  Both configure and make seem to run okay.
> Make install dies with the message attached at the bottom of this
> message.  I'm assuming that is because I don't have the needed
> permission.
> 
> Thanks for any advice.
> 
> Lloyd Sommerer
> 
> =====================================================
> gcc   python.o \
>           ../libpython2.0.a -ldb  -lpthread -lsocket -lnsl -ldl
> -lthread -lm  -
> o python
> Undefined                       first referenced
>  symbol                             in file
> Py_Main                             python.o
> ld: fatal: Symbol referencing errors. No output written to python
> *** Error code 1
> make: Fatal error: Command failed for target `link'
> Current working directory /home/gatheri/python/Modules
> *** Error code 1
> make: Fatal error: Command failed for target `python'

Hmm, this looks like there is a problem building Python. If I am not
mistaken, it fails *not* in the install stage, but when linking the
Python executable. Yesterday, I have built a Python 2.0 on Solaris 7, so
you might try my build. It's available at
http://sourceforge.net/project/showfiles.php?group_id=16151 It's
definitely possible to install your own Python interpreter. I have done
this twice. My first try would be to just unpack my archive in say
$HOME/opt/python. Then test in a shell if it works at all. Then try a
cgi script that calls the interpreter. The first line might look like
this:

#!/home/gatheri/opt/python/bin/python
# ...

If the webserver runs CGI scripts under a different uid, like a user
nobody's, you will have to make $HOME/opt/python readable and executable
by world.

Hope this helps a little. If it doesn't work out, just ask again (with
detailed error messages).

Gerhard
-- 
Sorry for the fake email, please use the real one below to reply.
contact: g e r h a r d @ b i g f o o t . d e
web:     http://highqualdev.com


From phd at phd.pp.ru  Fri Mar  2 12:13:44 2001
From: phd at phd.pp.ru (Oleg Broytmann)
Date: Fri, 2 Mar 2001 20:13:44 +0300 (MSK)
Subject: feature wish
In-Reply-To: <EZPn6.325779$f36.11597531@news20.bellglobal.com>
Message-ID: <Pine.LNX.4.30.0103022013050.7708-100000@fep132.fep.ru>

On Fri, 2 Mar 2001, Dan Parisien wrote:
> it would be cool if there was another method provided that allowed a
> programmer to overload the value returned by an instance as well.
>
> e.g.
>
> class example:
>         def __init__(self, name, data):
>                 self.name = name
>                 self.data = data
>
>         def __value__(self):
>                 return self.data
>
> eg = example("my name", (1,2,3))
>
> >>>print eg
> (1,2,3)

   Impossible: how do you render

eg.foobar()

   if eg returns tuple instaed of self?

Oleg.
----
     Oleg Broytmann            http://phd.pp.ru/            phd at phd.pp.ru
           Programmers don't die, they just GOSUB without RETURN.




From aahz at panix.com  Fri Mar 16 21:59:11 2001
From: aahz at panix.com (Aahz Maruch)
Date: 16 Mar 2001 18:59:11 -0800
Subject: ZPT, a next-generation template technology
References: <98ufbu$kkh$1@uranium.btinternet.com> <slrn9b5j52.i70.mgjv@martien.heliotrope.home>
Message-ID: <98ujtv$dcv$1@panix3.panix.com>

In article <slrn9b5j52.i70.mgjv at martien.heliotrope.home>,
Martien Verbruggen <mgjv at tradingpost.com.au> wrote:
>On Sat, 17 Mar 2001 01:43:50 -0000,
>	Hamish Lawson <hamish_lawson at btinternet.com> wrote:
>> 
>> * Zope is a leading open-source web application server and content-
>> management framework, written in Python by Digital Creations.
>
>So why did you post it to cll.perl.misc and cl.java? At least alt.php is
>slightly relevant to this, although you might piss them off as well. I
>don't know what their policy on offtopic posts is.
>
>However, cl.perl.misc gets enough offtopic crap as it is. Don't post
>future announcements here.

Ah, the friendliness of Perl programmers that we've all come to know and
love.
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

Three sins: BJ, B&J, B&J


From aahz at panix.com  Fri Mar  2 21:04:56 2001
From: aahz at panix.com (Aahz Maruch)
Date: 2 Mar 2001 18:04:56 -0800
Subject: the demise of 'from foo import * and its implications?
References: <mailman.983582584.8198.python-list@python.org>
Message-ID: <97pjg8$5nl$1@panix6.panix.com>

In article <mailman.983582584.8198.python-list at python.org>,
Tim Peters <tim.one at home.com> wrote:
>
>Would you prefer that we let you dope that out for yourself?  I'd be
>delighted to stick to the official rule, not catering at *all* to code that
>ignored the Ref Man.  Then we could replace the entire explanation with a
>one-line quote from the manual:  "you cheated; you got caught; tough luck".

To my mind, that's an excellent use for from __future__, particularly
combined with the new warnings.  So, yeah, I vote for extra-stringent.
-- 
                      --- Aahz (Copyright 2001 by aahz at pobox.com)

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

Nostalgia just ain't what it used to be


From vonWedel at lfpt.rwth-aachen.de  Wed Mar 21 03:34:52 2001
From: vonWedel at lfpt.rwth-aachen.de (Lars von Wedel)
Date: Wed, 21 Mar 2001 09:34:52 +0100
Subject: XLink/XPointer processing in Python
Message-ID: <3AB867AC.370C33F8@lfpt.rwth-aachen.de>

Hi,

I want to build an XML processing app which shall be
able to process "simple" links according to the XLink
spec (using SAX).

I haven't found any information on how these are handled
in the processing application, so I assume, I must handle
the xlink:href attribute content on 'my' own? Are there
any reusable pieces of software to do so?

Lars
-------------- next part --------------
A non-text attachment was scrubbed...
Name: vonWedel.vcf
Type: text/x-vcard
Size: 342 bytes
Desc: Card for Lars von Wedel
URL: <http://mail.python.org/pipermail/python-list/attachments/20010321/fd2e11a4/attachment.vcf>

From cce at clarkevans.com  Fri Mar 23 05:43:11 2001
From: cce at clarkevans.com (Clark C. Evans)
Date: Fri, 23 Mar 2001 05:43:11 -0500 (EST)
Subject: PartialInterfaces ? Re: Checking the type
In-Reply-To: <99acs601nsv@news2.newsguy.com>
Message-ID: <Pine.LNX.4.21.0103230526180.22896-100000@clarkevans.com>

> # safer
> def munge(alist):
>     # first extract all bound-methods we'll need
>     append = alist.append
>     extend = alist.extend
> 
>     # then check operations such as indexing
>     try: a[0]=a[0]
>     except IndexError: pass    # empty alist's OK
> 
>     # and finally operate -- no exceptions expected
>     append(23)
>     extend(range(5))
>     append(42)
>     alist[4] = alist[3]
>     extend(range(2))

This is a very cool, "protocol", a partial interface.
I was thinking about how this could be formalized as
a "checker/adapter"... here is very imperfect code)

class partial_list_checker:
      def __init__(self, append=0,extend=0,index=0,...
          self.check_append=append
          self.check_extend=extend
          self.check_index =index
          ...

      def check(self,alist)
          try:
              if self.check_append: append = alist.append
              if self.check_extend: extend = alist.extend
              if self.check_index:
                 try: alist[0] = alist[0]
                 except IndexError: pass
              ...
              return alist
          except AttributeError
             return None
          

What do you think?  Kinda slow, hunh?

> This doesn't interfere with polymorphism (any
> object supporting all needed operations with the
> right semantics will do), and yet is safer than
> the previous version _without_ being substantially
> slower or more complicated (well, a little, I guess).
> It's important to avoid overdoing the checks, but
> the assert statement can come in handy for checks
> we don't want to happen in production/optimized
> code (e.g, one may add "assert callable(append)"
> etc to the above example).
> 
> (Such versions of the look-before-you-leap idiom
> is popular in C++ templates, to get compiler
> errors that are clearer and earlier, but then,
> I've often said that C++ templates and Python
> functions show surprising parallels:-).

Yes, this was neat!

;) Clark






From cribeiro at mail.inet.com.br  Mon Mar 26 10:22:53 2001
From: cribeiro at mail.inet.com.br (Carlos Alberto Reis Ribeiro)
Date: Mon, 26 Mar 2001 12:22:53 -0300
Subject: Yet Another PEP:  Query Protocol Interface or __query__
In-Reply-To: <Pine.LNX.4.21.0103252253550.11867-100000@clarkevans.com>
References: <5.0.2.1.0.20010325210220.00a2cc00@mail.inet.com.br>
Message-ID: <5.0.2.1.0.20010326100519.02198bb0@mail.inet.com.br>

Clark,

The main point of my implementation was the idea of building the adapter 
from a empty 'Protocol' class, adding to it methods that are bound to the 
original object. This and a few helper functions solve the first part of 
the problem (the object returning an adapter).

Please also note that I made a clear division between an 'object instance' 
and a 'protocol instance'. The protocol instance, or adapter, is a proxy 
object. It's implemented as a normal Python object, but all its methods are 
redirected as described above. This approach has some advantages:

a) The proxy adapter is completely transparent to the object instance.
b) The adapter maps only the methods that are mandatory by the protocol 
specification. This leads to better encapsulation and abstraction (all 
details of the object are hidden from the protocol user).
c) Method names may be different in the object. The protocol does the name 
translation.
d) The protocol instance may point to methods in disparate objects, which 
leads to a interesting type of 'composition'.

Anyway, did you take a look at my example? Please disregard the explanation 
of the protocols in terms of interfaces. The details can be changed without 
changing the essence of the proposal.


Now some comments, in a different order from your original email; I moved 
the two-way adapter discussion to the end of the message.

At 23:08 25/03/01 -0500, Clark C. Evans wrote:
>On Mon, 26 Mar 2001, Carlos Ribeiro wrote:
> > 1) What is a protocol? The simplest definition that I cant think of is: a
> > protocol is just a bunch of methods. An object is said to support a
> > protocol if it implements all methods of the protocol specification. The
> > specification defines the names and parameters of all methods.
>
>This is what an interface is, and there is a proposal
>for this.  A protocol should be more broadly defined, no?

Hummm... I see your point. My intention is different - avoid mixing 
'protocols' and 'class definitions'. One of the reasons was the potential 
for name clashes, as shown in my example. A protocol does not need to be 
represented *exclusively* by a class in the hierarchy. A single class may 
implement multiple protocols, but some protocols may need a composition of 
objects to do the work. For example, a composite object may contain lots of 
objects that gets exposed through interfaces in the container.

Also I see no reason to avoid defining some kind of 'simple' interfaces in 
this PEP - in the end, protocols are better mapped to 'interfaces' than to 
'classes'. It's much more flexible in this way.

OTOH, it is possible to use 'protocol classes' instead of strings to 
describe the protocol. The implementation does not change very much. The 
catch here is to avoid the assumption that *implementing* a protocol is the 
same as inheriting from the abstract protocol class. For example:

class ProtocolSequence:
   """ Protocol to implement sequence-like objects """
   def __len__ (self): pass
   def __getitem__(self, key): pass
   def __setitem__(self, key, value): pass
   def __delitem__(self, key): pass

In this example, ProtocolSequence is the specification of the methods used 
for sequence type objects. Sequence objects does not need to be derived 
from ProtocolSequence; they may be written from the scratch, or be derived 
from UserList, etc. Again, keep in mind the potential of composition as a 
technique to build large systems. It is important to avoid the assumption 
that inheritance is *the only way* to build OO systems.

MakeProtocol would then use some getattr/setattr 'magic' to build the 
method list, and to 'rebind' all the methods to the ones in the object 
being adapted. I'm not checking the method signatures though; this would 
need some help from the compiler in order to work.

(note: MakeProtocol is not a good name. MakeAdapter, or 
MakeProtocolInstance, is better)

>IMHO, the goal was to have this dialogue...

The implementation that I sent March 25 is one-way only. It is somewhat 
simpler, because I dont need to define a separate protocol class - it's 
just a matter of building the adapter inside the __adapt__ method. Also, I 
did not see yours (Clark's) dialogue implemented in the last version of the 
draft (I have read the one from March 21, did I miss something)?

TWO-WAY IMPLEMENTATION

For a two-way implementation, I suggest the following dialogue (about the 
same as your):

   Hello protocol manager, do you know if this
   object X here supports the protocol Y?

   The protocol manager then does:

     Hello object X, do you support protocol Y?
     Please return an adapter if so.

     Object X says: yes, I do support Y, and here
     is the adapter. Otherwise, X would have nothing
     to give back to the protocol manager.

     if the object doesn't know about protocol Y,
     then the protocol itself is asked about X:

     Protocol Y, do you know about object X?

     Protocol Y says: Yes, I checked object X
     class/methods/protocols, and I think that I
     can make an adapter for you.

I initially had some concerns over the two-way implementation. The main 
reason is that only an object could possible know how to adapt itself to a 
given protocol. Now I see that this could be possible, specially if we 
provide some helper functions to hide some hacks that Y would need to do in 
order to check X structure. I see three things that could be checked by the 
protocol:

a) class inheritance, as in your example. However it does limit 'protocol' 
to 'root classes', and it does have a problem with multiple inheritance, 
namely because of potential name clashes. It also does not work if the 
object is built not by inheritance, but by *composition*.

b) methods, or 'interface signature'. In this case the protocol just checks 
if the object does support all methods required by the protocol.

c) protocol compatibility. The protocol may ask the object directly for 
some compatible protocol, and then built a new adapter. This could be done, 
for instance, if you have a newer version of a protocol with only small 
changes; the protocol itself provides the interoperability layer with the 
previous version.


Some helper functions could be provided to help in these cases, for example:

def TestMethods(object, list_of_method_names):
   """ check to see if object supports all methods
       listed on the second parameter
   """
   (...)

I'll try to build a new example with these points in mind.


Carlos Ribeiro





From aleaxit at yahoo.com  Wed Mar  7 03:03:58 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Wed, 7 Mar 2001 09:03:58 +0100
Subject: tokenizer.c crashes in 2.1b1
References: <3aa34d5c.1284101@news.newsguy.com> <ulmqiobyp.fsf@roundpoint.com> <3aa5d493.5922346@news.newsguy.com>
Message-ID: <984q1m0skr@news1.newsguy.com>

"Owen F. Ransen" <ransen_spam_me_not at nemo.it> wrote in message
news:3aa5d493.5922346 at news.newsguy.com...
> On 06 Mar 2001 12:54:06 -0800, Ben Hutchings
> <ben.hutchings at roundpoint.com> wrote:
>
> >> The debugger points to the guilty party:
> >> tokenizer.c, the fgets call at line 247 fails.
> >> My VC project uses Multithreaded Debug compiler settings,
> >> as does the Pythoncore project.
    [snip]
> Debug Multithreaded in my MFC app make settings
> Debug Multithreaded DLL in the Pythoncore DLL make settings
>
> Is this wrong?

Yes: VC++ puts strong constraints on the way you can
build modules (EXE's and DLL's) which you want to
interoperate "normally" (via the linker and loader).

In particular, they must use IDENTICAL settings for
the "Project/Settings.../C/C++/Code Generation/Use
run-time library" dropdown.  Otherwise, you'll run
into frequent, deep trouble -- different instances
of the C/C++ runtime libraries will be working at
the same time (it's technically possible to survive
that, if it never happens that something is allocated
by one module and freed by the other, no I/O happens
on the same file from both modules, etc, etc, but in
practice that is out of your control [particularly if
constructors/destructors, etc, enter the picture], so,
think of it as an amusing oddity if and when two
modules with mismatched runtime libraries happen
to work without crashing, and NEVER rely on it!-).

[As an aside, that's part of what's so USEFUL about
COM -- among other things, it freeds you from all
such constraints -- COM-interoperating modules may be
built any which way, including in completely different
languages/compiler releases/runtime library settings/
etc, because COM specifies exactly what's to happen
on the interface between them, _at a binary level_;
these COM constraints *on the language implementors*
turn into freedom *for the application programmers*!-)]


Alex





From tdelaney at avaya.com  Mon Mar 26 19:16:20 2001
From: tdelaney at avaya.com (Delaney, Timothy)
Date: Tue, 27 Mar 2001 10:16:20 +1000
Subject: warrior geeks
Message-ID: <B43D149A9AB2D411971300B0D03D7E8B0FBFE0@natasha.auslabs.avaya.com>

> >ESR's apparent fondness for Python. That sort of thing gives 
> me a fit of
> >the shudders (playing with guns, not ESR's fondness for Python). 
> 
> What are you?  Some kind of Commie?  Guns form the very foundation of
> the Pax Americana that gives you the freedom to use Python.

I hope there was a <wink> there that I missed ...

Australians in general are not overly keen on guns. Except for certain
security guards, non-military and non-police *cannot* get a license for a
handgun (and military and police cannot carry a gun unless on duty).
Likewise, it is almost impossible to buy ammunition for handguns.
Professional hunters can get a license for a semi-automatic rifle. I think
the most someone can get who is not a professional hunter can get is a
manual (bolt-action, level, etc) rifle or shotgun. In any case, I think
there is a mandatory 5-day cooling off period before you can actually get
your gun. However, I'm not sure on this as I've never even considered
getting a gun of any type. My father had a .22 bolt-action rifle when we
lived in the country for shooting foxes - it was kept in 3 pieces in various
parts of the house. Getting the bullets required taking a whole pile of
stuff out of the pantry, taking up a board in the floor and reaching down to
the bottom of a hole which an adult could only just reach.

This is a Good Thing(TM). I think a lot of people in the US don't realise
that the "right to bear arms" was devised when "arms" were single-shot
weapons and the US was attempting to secede from England. A lot of the
attitude of Americans really scares me. I'd much rather be here with our
multitude of venomous creatures than in the US. We've got anti-venine for
almost every venom in almost every hospital in Australia. There's no
anti-venine for a gunshot wound.

Tim Delaney

All recipient(s) of this email have permission to forward or reply to this
email, quoting this email in full or in part.



From NOstarfighterSPAM at freeuk.com  Tue Mar 20 14:01:52 2001
From: NOstarfighterSPAM at freeuk.com (Richard Townsend)
Date: Tue, 20 Mar 2001 19:01:52 -0000
Subject: Catching exceptions in Tkinter's mainloop
References: <%dws6.9947$Q47.2723350@news1.rdc1.tn.home.com> <drMs6.32482$HR6.3694604@nnrp4.clara.net> <owzt6.21061$Q47.5935326@news1.rdc1.tn.home.com>
Message-ID: <6QNt6.37618$g63.4509118@nnrp3.clara.net>

"Mike Callahan" <mcalla at home.com> wrote in message
news:owzt6.21061$Q47.5935326 at news1.rdc1.tn.home.com...
> I tried the following with no luck:
>

> import Tkinter as T
> import sys
>
> class Gui:
>     def __init__(self, master):
>         T.Button(master, text='Crash2', command=self.crash).pack()
>
>     def crash(self):
>         z = zz
>
> class Catcher:
>     def __init__(self, func, subst, widget):
>         self.func = func
>         self.subst = subst
>         self.widget = widget
>
>     def __call__(self, *args):
>         try:
>             if self.subst:
>                 args = apply(self.subst, args)
>             return apply(self.func, args)
>         except SystemExit, msg:
>             raise SystemExit, msg
>         except:
>             import traceback
>             traceback.print_exc()
>             self.widget.quit()
>
> root = T.Tk()
> f = open('test.log', 'w')
> sys.stderr = f
> T.CallWrapper = Catcher
> app = Gui(root)
> root.mainloop()
>
> What am I doing wrong?
>

I've just tried your code and it wrote the following to the 'test.log' file:

Traceback (most recent call last):
  File "catcher.py", line 21, in __call__
    return apply(self.func, args)
  File "catcher.py", line 9, in crash
    z = zz
NameError: There is no variable named 'zz'


Is that not what you wanted ??

Richard





From alex at shindich.com  Mon Mar  5 23:40:29 2001
From: alex at shindich.com (Alex Shindich)
Date: Mon, 5 Mar 2001 20:40:29 -0800
Subject: Static class methods
References: <rdzo6.154$gc4.80445@news.pacbell.net> <aqa7at0b6cbqjaq23mli0flmumlp8n6s9s@4ax.com>
Message-ID: <7SZo6.7063$qv3.3125634@nnrp5-w.sbc.net>

There are indeed many ways to simulate the behavior. But shouldn't such a
strong OOP language like Python have support for static methods?
"Daniel Klein" <danielk at aracnet.com> wrote in message
news:aqa7at0b6cbqjaq23mli0flmumlp8n6s9s at 4ax.com...
> On Sun, 4 Mar 2001 14:22:10 -0800, "Alex Shindich" <alex at shindich.com>
wrote:
>
> >Just out of my curiosity, have any of you ladies and gentlemen
experienced a
> >need for an equivalent of C++/Java static method?
> >Because I have, but it is impossible add them to Python classes -- all
class
> >methods are treated as instance methods.
>
> The Quick Python Book by Daryl Harms and Kenneth McDonald outlines a way
to
> accomplish this by breaking out the class into its own module and then
define
> the 'class methods' as top-level functions.
>
> myclass.py
>
> def myClassMethod():
> # do some class thing
>
> class Foo:
> def __init__(self):
>                   pass
> def bar(self):
> pass
>
> Daniel Klein
>




From edcjones at erols.com  Sat Mar 24 16:28:30 2001
From: edcjones at erols.com (Edward C. Jones)
Date: Sat, 24 Mar 2001 16:28:30 -0500
Subject: "object" vs "dynamic" programming styles
Message-ID: <3ABD117E.1E5492C4@erols.com>

Here are three methods for doing the same thing. When is it better to use
one of the approaches rather than another?

=============================================================
# Program 1:

# The constructor X(c) actually creates an A or B instance.
class X:
    def __init__(self, c):
        self.whoami = 'X'
        self.Xs = 'Xs'
        if c == 'a':
            self.__class__ = A
            A.__init__(self)
        elif c == 'b':
            self.__class__ = B
            B.__init__(self)

    def fun(self):
        print 'X.fun called'

class A(X):
    def __init__(self):
        X.__init__(self, '')
        self.whoami = 'A'
        self.As = 'As'

    def fun(self):
        print 'A.fun called'

class B(X):
    def __init__(self):
        X.__init__(self, '')
        self.whoami = 'B'
        self.Bs = 'Bs'

a = X('a')
print 'a.__class__:', a.__class__
print 'a.whoami:', a.whoami
print 'a.Xs:', a.Xs
print 'a.As:', a.As

a.fun()

b = X('b')
b.fun()

===================================================================

# Program 2

# Which attributes X(c) has depends on the value of c.
class X:
    def __init__(self, c):
        self.Xs = 'Xs'
        if c == 'a':
            self.set_A()
        elif c == 'b':
            self.set_B()

    def set_A(self):
        self.whoami = 'A'
        self.As = 'As'
        self.fun = self.fun_A

    def set_B(self):
        self.whoami = 'B'
        self.Bs = 'Bs'

    def fun_A(self):
            print 'A_fun'

a = X('a')
print 'a.__class__:', a.__class__
print 'a.whoami:', a.whoami
print 'a.Xs:', a.Xs
print 'a.As:', a.As

a.fun()

b = X('b')
try:
    b.fun()
except AttributeError, x:
    print x

===================================================================

# Program 3

# The function New_X returns an instance of either class A or class B.
def New_X(c):
    if c == 'a':
        return A()
    if c == 'b':
        return B()

class A:
    def __init__(self):
        self.whoami = 'A'
        self.Xs = 'Xs'
        self.As = 'As'

    def fun(self):
        print 'A.fun called'

class B:
    def __init__(self):
        self.whoami = 'B'
        self.Xs = 'Xs'
        self.Bs = 'Bs'

a = New_X('a')
print 'a.__class__:', a.__class__
print 'a.whoami:', a.whoami
print 'a.Xs:', a.Xs
print 'a.As:', a.As

a.fun()

b = New_X('b')
try:
    b.fun()
except AttributeError, x:
    print x

===================================================================




From winson at mdsserv.mds.com.tw  Sun Mar 11 22:52:42 2001
From: winson at mdsserv.mds.com.tw (=?big5?B?sWm4zq9x?=)
Date: Mon, 12 Mar 2001 11:52:42 +0800
Subject: (no subject)
Message-ID: <007c01c0aaa7$e6b8b960$340210ac@winson>

Hi All,

Can I find Apache module for Python ??

--
Best Regards,

Winson Chang

3F, 2, 150 Lane, Sec. 5, Hsin I Rd., Taipei, Taiwan, R.O.C.

Tel: +886-2-27225333 ext 376
Fax: +886-2-27222330

Email: winson at mdsserv.mds.com.tw
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20010312/b52d0e24/attachment.html>

From dgrisby at uk.research.att.com  Mon Mar 12 04:18:51 2001
From: dgrisby at uk.research.att.com (Duncan Grisby)
Date: 12 Mar 2001 09:18:51 GMT
Subject: Python and ObjectSpace Voyager
References: <4chnatkpmpvo29v515orlcjcfuoii5gcte@4ax.com>
Message-ID: <98i49r$94c$1@pea.uk.research.att.com>

In article <4chnatkpmpvo29v515orlcjcfuoii5gcte at 4ax.com>,
 kammie  <kammie at commie.com> wrote:

>Has anyone heard of writing Python CORBA apps for use with the Voyager
>orb?  It would be neat if that orb had its own Python mappings (I
>don't think so), otherwise IIOP would be nice.

I'm not sure quite what you're asking, but as long as Voyager is
properly CORBA compliant, you can talk to it with any of the Python
ORBs. Try

  http://www.omniorb.org/omniORBpy/
  http://www.fnorb.com/
  http://projects.sault.org/orbit-python/
  ftp://ftp.parc.xerox.com/pub/ilu/ilu.html

Cheers,

Duncan.

-- 
 -- Duncan Grisby  \  Research Engineer  --
  -- AT&T Laboratories Cambridge          --
   -- http://www.uk.research.att.com/~dpg1 --


From lluang at northwestern.edu  Sun Mar 18 22:37:48 2001
From: lluang at northwestern.edu (Louis Luangkesorn)
Date: Sun, 18 Mar 2001 21:37:48 -0600
Subject: stripping punctuation from tuples
Message-ID: <3AB57F0C.594CF3E9@northwestern.edu>

How do you strip the punctuation from tuples?  In particular, say I have
a tuple a[10] (which happened to come from a not very well designed
database)

>>> a[10]
('Alan Tsang',)

How do I get something to say 'Alan Tsang'  with out the parens or the
','

Or, what I'm really trying to do is make this so
>>>b = breakname(a[10])
>>>b[0], b[1]

'Alan','Tsang'

I've tried to do:
>>> d = split(str(a[0]))
>>> d
["('Aaron", "Fan',)"]
>>> d[0]
"('Aaron"
>>> d[1]
"Fan',)"

But that is not quite what I want.
Many thanks.
Louis
--
K Louis Luangkesorn
lluang at northwestern.edu  http://pubweb.nwu.edu/~kll560  PGP:0xF3E2D362

Whatsoever things are true, ... honest, ... just, ... pure, ... lovely,
... of good report; if there be any virtue, and if there be any praise,
think on these things.
     - motto - Northwestern University (Phil 4:8)


From paul.robinson at businesscollaborator.com  Wed Mar 14 04:07:46 2001
From: paul.robinson at businesscollaborator.com (Paul Robinson)
Date: Wed, 14 Mar 2001 09:07:46 +0000
Subject: Unpickling from strings
References: <qHxr6.19658$pR3.3830844@news2-win.server.ntlworld.com> <IHzr6.21597$Sl.962179@iad-read.news.verio.net>
Message-ID: <3AAF34E2.18A0EB93@businesscollaborator.com>

Doug Fort wrote:
> 
> Perhaps it is getting garbled in transmission.  We quote ours:
> 
>     urllib.quote_plus(cPickle.dumps(self._statsdict))
>
> "Marcin Chady" wrote:
>
> > I have pickled an object into a string and saved it in an hidden HTML form
> > field, so that I can recover the state when the form is resubmitted.

Is there not an inherent security risk from trusting what could
potentially be any kind of pickled object that you are retrieving from
the submission of an HTML form?

If you are going to do that then you should take appropriate
precautions.

Paul Robinson
Business Collaborator Development Manager
Enviros Software Solutions
WWW: http://www.businesscollaborator.com



From stephen_purcell at yahoo.com  Mon Mar 26 07:40:03 2001
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Mon, 26 Mar 2001 14:40:03 +0200
Subject: portable string localization and resource bundles...
In-Reply-To: <Pine.LNX.4.21.0103260619180.21288-100000@clarkevans.com>; from cce@clarkevans.com on Mon, Mar 26, 2001 at 06:23:32AM -0500
References: <Pine.LNX.4.21.0103260619180.21288-100000@clarkevans.com>
Message-ID: <20010326144003.A11987@freedom.puma-ag.com>

Clark C. Evans wrote:
> I was wondering if there is a solution to the 
> common problem of string localization 
> (menu and message box titles, for instance).


How about putting your strings in modules 'strings_en.py', 'strings_de.py'
then doing something simple like the following:

>> mylocale = 'de'
>> strings = __import__('strings_%s' % mylocale)
>> f = Frame(title=strings.main_frame_title)

With newer Pythons you can probably use the 'locale' module to handle the
locale strings nicely.

-Steve

-- 
Steve Purcell, Pythangelist
Get testing at http://pyunit.sourceforge.net/
Any opinions expressed herein are my own and not necessarily those of Yahoo



From paulp at ActiveState.com  Tue Mar 20 14:50:07 2001
From: paulp at ActiveState.com (Paul Prescod)
Date: Tue, 20 Mar 2001 11:50:07 -0800
Subject: Should __future__ statements ever be phased out?
References: <20010320173101.21650.qmail@web217.mail.yahoo.com>
Message-ID: <3AB7B46F.56ABDA85@ActiveState.com>

Hamish Lawson wrote:
> 
> ... However one
> day I give my program to a friend who (unknown to me) is still running
> Python 2.1. When he runs my program any print statements in it will
> behave differently from what I intended, but without any warning.

I see only two solutions to the problem you describe:

 1. Never make backwards incompatible changes

 2. Always have a version declaration or feature declaration at the top
of Python programs.

You already have the ability to implement 2 if you are paranoid. You can
check sys.version_info in your code.

-- 
Take a recipe. Leave a recipe.  
Python Cookbook!  http://www.activestate.com/pythoncookbook



From db3l at fitlinxx.com  Tue Mar 13 20:01:48 2001
From: db3l at fitlinxx.com (David Bolen)
Date: 13 Mar 2001 20:01:48 -0500
Subject: Going ZODB
References: <mailman.984324988.25505.python-list@python.org>
Message-ID: <uitld17ur.fsf@ctwd0143.fitlinxx.com>

Steven Haryanto <steven at haryan.to> writes:

> I am curious as to how many people in this list actually
> use ZODB in production environment and what is the typical
> uses of it (especially ZODB with relational/berkeleydb
> storage). Aside from along with Zope, of course.

We use it as the persistance store for an internal (it's production
internally, but not something shipped to customers) scheduling system
that handles execution of our network management tasks.  The main
scheduler is a process that can be attached to by any number of agents
on a collection of execution machines, and the scheduler dispatches
tasks as appropriate based on a variety of parameters, priorities,
resource requirements and so on.

The ZODB object store holds any of our objects related to the state of
the system.  We're not fancy - we commit changes fairly tightly with
where they are made, and we manage our own interlocked access to the
data from the multiple scheduler threads.  But it works great to
ensure that we can recover from failure at pretty much any point to
precisely the same scheduling state we were in before.  It's not all
that different to simply unpickling the information at startup and
pickling it at shutdown except that we have continuous persistence
(well, at the granularity of any significant state change) in the case
of failure.  Since we're not versioning our objects, we have a
periodic cycle where we pack the database to free space.

Due to this usage, moving away from ZODB if we had to isn't hard - we
just remove the commit stuff and read/write the state ourselves.  But
unless we had fairly rapid checkpoints, we'd be at more risk than with
what we do with ZODB now.  And we're also getting some memory benefits
with ZODB in terms of how it cleans out objects as appropriate and
then transparently brings them back in.  So when we start up, we don't
have all the history objects in memory - but they are brought in
transparently to the code should an administrative request need them.

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


From fredrik at pythonware.com  Wed Mar 21 05:31:40 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 21 Mar 2001 10:31:40 GMT
Subject: Tkinter Problem - NameError
References: <mailman.985165769.12848.python-list@python.org>
Message-ID: <gq%t6.864$sk3.285904@newsb.telia.net>

Clodagh O'Rourke wrote:
> Here I get the following NameError:
>
> Exception in Tkinter callback
> Traceback (innermost last):
>   File "/usr/vob/omc_reference/ThirdParty/scriptSources/lib/python1.5/lib-tk/Tkinter.py", line 764, in __call__
>     return apply(self.func, args)
>   File "AuditResync.py", line 64, in newAudit
>     self.newEnvWin('audit')
>   File "AuditResync.py", line 61, in newEnvWin
>     EnvVar(g, self.parent)
> NameError: EnvVar
>
> I would be very grateful if anyone could spot any possible errors.

Looks like recursive from-import problems [1]...

Can you post the import sections from InitialWin, EnvVar, and
AuditResync?

Cheers /F

1) http://effbot.org/guides/import-confusion.htm




From MarkH at ActiveState.com  Mon Mar 19 03:12:12 2001
From: MarkH at ActiveState.com (Mark Hammond)
Date: Mon, 19 Mar 2001 08:12:12 GMT
Subject: Using ActiveX Control from Python
References: <mailman.984965245.12168.python-list@python.org>
Message-ID: <3AB5C09D.8040402@ActiveState.com>

Hirsch, John wrote:

> Hi All
> I am trying to use an ActiveX Control from Python 2.0 with little success.

Python alone does not support true COM Controls - these are quite 
complex beasts and require significant "framework" support.

You have a couple of reasonable choices:
* Use Pythonwin - it inherits MFC's control container support.  Check 
out pywin\demos\ocx for examples.

* Check out some recent(ish) posts from Alex Martelli who describes in 
excellent detail how to use IE as your container, and still use Python 
to script it.

Mark.



From tim.one at home.com  Sat Mar 31 22:04:08 2001
From: tim.one at home.com (Tim Peters)
Date: Sat, 31 Mar 2001 22:04:08 -0500
Subject: Conditional operator in Python?
In-Reply-To: <3AC68CB8.22AAF7B8@alcyone.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCKEINJJAA.tim.one@home.com>

[Erik Max Francis]
> Okay, okay, but I still have to ask.
>
> Why isn't there a conditional operator (a ? b : c) in Python?  2.0 now
> has augmented assignments, and the FAQ even acknowledges that the
> conditional operator is sometimes quite convenient and gives some (very
> awkward) Python alternatives.
>
> I've been reading comp.lang.python for a while (since just before 2.0
> shipped) and I can't really remember much discussion about it.  Is it a
> perennial request and I'm just missing it?

Eric, there's no feature from *any* language that isn't regularly suggested
for Python.  This one is regularly requested too, but is down in the noise.
Its most persistent proponent today is Eric Raymond, and you might want to
nag him to finish his long-threatened PEP on the topic.

In the meantime, try if/then/else <wink>.

hey-it-works-for-me-ly y'rs  - tim




From hinsen at cnrs-orleans.fr  Thu Mar  8 09:20:39 2001
From: hinsen at cnrs-orleans.fr (Konrad Hinsen)
Date: 08 Mar 2001 15:20:39 +0100
Subject: NumPy's undocumented type code
References: <mailman.984037992.23917.python-list@python.org>
Message-ID: <m3wva0bavc.fsf@chinon.cnrs-orleans.fr>

Nadav Horesh <nadavh at envision.co.il> writes:

> Recently I encountered an unfamiliar NumPy array type code -- <tt><font size=+1>"O"</font></tt>.

That's the "general object" array, its elements are arbitrary Python
objects. It can be a very convenient for some applications, but,
unlike with the other array types, the execution speed for arithmetic
operations etc. is no better than equivalent operations with lists.
-- 
-------------------------------------------------------------------------------
Konrad Hinsen                            | E-Mail: hinsen at cnrs-orleans.fr
Centre de Biophysique Moleculaire (CNRS) | Tel.: +33-2.38.25.56.24
Rue Charles Sadron                       | Fax:  +33-2.38.63.15.17
45071 Orleans Cedex 2                    | Deutsch/Esperanto/English/
France                                   | Nederlands/Francais
-------------------------------------------------------------------------------


From aleaxit at yahoo.com  Tue Mar  6 03:13:02 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Tue, 6 Mar 2001 09:13:02 +0100
Subject: Can Python replace Visual Basic? Should it?
References: <5DZo6.11126$hn5.1578410@news1.rdc1.mb.home.com>
Message-ID: <98266k0p5h@news1.newsguy.com>

"Brad Bollenbach" <bbollenbach at homenospam.com> wrote in message
news:5DZo6.11126$hn5.1578410 at news1.rdc1.mb.home.com...
> Okay, sure this has been asked again and again, but I've never really seen
a
> clear answer to this question:
>
>     Can Python replace Visual Basic? Should it?

Yes, and no, respectively.

> "languages" that are based on an IDE are not really my thing (I hate
having

Then, FOR YOU, Python has it all over VB.  Others' tastes differ.

> But really, very specifically, were YOU a former Visual Basic (or
> Powerbuilder, or Delphi) programmer that is now using Python to write the
> complex GUI's VB used to handle for you? I'm an on again, off again,
student

Sorry, no 'former' here, nor any _complex_ GUIs.  I do some _modest_ use
of VB for reasonably simple tasks, mostly connected to ease of deployment
of stuff embedded in Office applications.  I share your distaste for
'wizards' that hide from me what they're doing (I don't mind tools that
generate some code for me, as long as they keep it all open for my
inspection & tweaking).

> Again, I'm much more interested in the real world experience of those who
> used to use the big visual development tools in Windows like VB, Delphi or
> Powerbuilder rather than theoretical answers.

Why do you focus on 'used to use'?  One may perfectly well keep using
some VB (&c) for tasks it's particularly well-suited to, while using
Python wherever feasible (and, indeed, using both in the same project
is hardly strange -- e.g., use VBA inside [say] Excel to provide some
minimal intrinsic event-handling to some special workbook, and use COM
Python objects, instantiated and invoker from that modest amount of VBA,
for all of the real 'business logic'/complex-GUI/etc aspects).


Alex





From aleaxit at yahoo.com  Thu Mar  1 09:09:56 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 1 Mar 2001 15:09:56 +0100
Subject: d = {}; d[0:1] = 1; d[0:1] = 2; print d[0:1]
References: <mailman.983419142.24027.python-list@python.org> <mailman.983429941.26766.python-list@python.org> <m3ae75rfl6.fsf@chinon.cnrs-orleans.fr>
Message-ID: <97ll7k02qsr@news2.newsguy.com>

"Konrad Hinsen" <hinsen at cnrs-orleans.fr> wrote in message
news:m3ae75rfl6.fsf at chinon.cnrs-orleans.fr...
> Robin Thomas <robin.thomas at starmedia.net> writes:
>
> > The question remains: should slice objects be hashable/comparable so
that
> >
> >  >>> slice(0,1) == slice(0,1)
> > 1
> >
> > I say YES YES YES. What do you say?
>
> Another YES YES YES. That's what anyone would expect in my opinion.

It definitely is what _I_ would expect, although, as a sample,
I guess I'm slightly skimpy.  What purpose is satisfied by
having this == test return 0...?


Alex





From m.hadfield at niwa.cri.nz  Tue Mar 13 15:42:34 2001
From: m.hadfield at niwa.cri.nz (Mark Hadfield)
Date: Wed, 14 Mar 2001 09:42:34 +1300
Subject: win32all buld 137 win32gui fails
References: <98l8qk$oeo$1@troll.powertech.no> <DYqr6.83020$__6.16207072@typhoon.southeast.rr.com>
Message-ID: <984516152.813137@clam-ext>

"Don Tuttle" <tuttledon at hotmail.com> wrote in message
news:DYqr6.83020$__6.16207072 at typhoon.southeast.rr.com...
> "Syver Enstad" <syver at NOSPAMcyberwatcher.com> wrote in message
> news:98l8qk$oeo$1 at troll.powertech.no...
> > I've just installed the win32all build 137 that contains the support for
> > Vtable interfaces in win32com. I've suddenly found that import win32gui
> > fails with the message:
> >
> > ImportError: DLL load failed: The specified module could not be found.
> >
> > This is mighty strange as the pyd file is lying there among the other
.pyd
> > files like win32help and so on. Maybe there are some other modules that
> > win32gui depends on that it didn't depend on before?
> >
> Don't know.  It imports fine for me.  Perhaps unistall 137, then
reinstall.

For me, it fails, and pops up a dialogue box with the following helpful
message:

"The DLL PyWinTypes21.dll could not be found in the specified path: ..."

There is a PyWinTypes20.dll in the Windows system32 directory, but no
PyWinTypes21.dll. This is not surprising as I have not installed any version
of Python 2.1 on my system.

win32all version is 137. This is advertised as being for Python 2.0 but
obviously there is a version mix-up.

As a work-around you could try making a copy of PyWinTypes20.dll and
renaming it PyWinTypes21.dll.

---
Mark Hadfield
m.hadfield at niwa.cri.nz  http://katipo.niwa.cri.nz/~hadfield
National Institute for Water and Atmospheric Research




From paul.moore at uk.origin-it.com  Tue Mar 20 11:22:14 2001
From: paul.moore at uk.origin-it.com (Paul Moore)
Date: Tue, 20 Mar 2001 17:22:14 +0100
Subject: Using ActiveX Control from Python
References: <mailman.984965245.12168.python-list@python.org> <3AB5C09D.8040402@ActiveState.com>
Message-ID: <N4S3OlRkH+C6dw9LF2cHWdmn2zp4@4ax.com>

On Mon, 19 Mar 2001 09:12:12 +0100, Mark Hammond <MarkH at ActiveState.com>
wrote:
>* Check out some recent(ish) posts from Alex Martelli who describes in 
>excellent detail how to use IE as your container, and still use Python 
>to script it.

Can you provide any pointers? Even subject lines would help.

Thanks,
Paul



From andreas at andreas-jung.com  Sat Mar 17 16:57:46 2001
From: andreas at andreas-jung.com (Andreas Jung)
Date: Sat, 17 Mar 2001 16:57:46 -0500
Subject: Accessing /dev/mixer under Linux 
Message-ID: <003f01c0af2d$62197ac0$9865fea9@SUXLAP>

Is there a module to read and write settings from /dev/mixer under Linux ?

Andreas




From dubois at users.sourceforge.net  Sat Mar 24 18:17:15 2001
From: dubois at users.sourceforge.net (Paul F. Dubois)
Date: Sat, 24 Mar 2001 23:17:15 GMT
Subject: Numeric 18.4 released
Message-ID: <%V9v6.2254$ad1.210638@news1.frmt1.sfba.home.com>

Numeric 18.4 is available, built with 2.1b2, at

http://sourceforge.net/projects/numpy

Numeric 18.4 requires Python 2.1. We have tested it with Python2.1b2a and
built the Windows version with Python 2.1b2.

Please read the notes and changelog on SourceForge.





From fgeiger at datec.at  Tue Mar 13 01:47:31 2001
From: fgeiger at datec.at (Franz GEIGER)
Date: Tue, 13 Mar 2001 07:47:31 +0100
Subject: Want to catch pywintypes.api_error
Message-ID: <98kftn$jf9$1@newsreaderm1.core.theplanet.net>

When I start win32api.FindFiles(filespec) with a non-existing filespec it
throws an exception of class pywintypes.api_error. But a "except
pywintypes.api_error, e:" does not work. Using "except Exception, e:" does
the job, because Exception is the base class. But how can I catch api_error?
What do I miss?

Thanks in advance and best regards
Franz GEIGER





From amused at webamused.com  Sat Mar 24 09:25:04 2001
From: amused at webamused.com (Joshua Macy)
Date: Sat, 24 Mar 2001 14:25:04 GMT
Subject: New PEP: The directive statement
References: <mailman.985418312.27803.python-list@python.org> <j43dc3o31m.fsf@informatik.hu-berlin.de>
Message-ID: <3ABCE601.D54C39B2@webamused.com>

Martin von Loewis wrote:

> 
> That is absolutely clear. I was asking you to predict the future: Do
> you think that people will prefer to write
> 
>   if __future__.nested_scopes.getOptionalRelease() >= sys.version_info:
> 
> instead of
> 
>   if sys.version_info < (2,1):
> 

  I certainly would, since the former communicates what's going on much
more clearly.  It spells out exactly why the optional release is
necessary for the code, while hiding the gory details of which optional
release is predicted to be the one satisfying it. In addition the same
__future__ approach would work for mandatory releases even if the final
version number where the feature really became mandatory changed,
whereas your version would require changing the magic number by hand in
every file if a mandatory release ever got bumped back--including in
files that I didn't write.  Even if all I checked the mandatory release
number for was emitting a warning to remove the from __future__, I don't
want to deal with a bunch of false positives.

  Clearer, more bulletproof code, yeah, I'd go for that.

  Joshua


From tdelaney at avaya.com  Mon Mar 26 19:44:48 2001
From: tdelaney at avaya.com (Delaney, Timothy)
Date: Tue, 27 Mar 2001 10:44:48 +1000
Subject: warrior geeks
Message-ID: <B43D149A9AB2D411971300B0D03D7E8B0FBFE2@natasha.auslabs.avaya.com>

> > This is a Good Thing(TM). I think a lot of people in the US 
> don't realise
> > that the "right to bear arms" was devised when "arms" were 
> single-shot
> > weapons and the US was attempting to secede from England. A 
> lot of the
> > attitude of Americans really scares me. I'd much rather be 
> here with our
> > multitude of venomous creatures than in the US. We've got 
> anti-venine for
> > almost every venom in almost every hospital in Australia. There's no
> > anti-venine for a gunshot wound.
> 
> But the question is: Which of Perl versus Python is more 
> appropriate for
> shooting yourself in the foot.
> 
> At-least-that-flame-war-is-almost-on-topic 'ly yrs

That's easy. With Python you can be sure that you have shot yourself in the
foot. With Perl, you think you've *probably* shot yourself in the foot, but
due to a difficult-to-find typo you have also shot several other people in
the foot.



From mikael.lexen at stop.spam.volvo.com  Fri Mar 16 06:38:45 2001
From: mikael.lexen at stop.spam.volvo.com (Mikael =?iso-8859-1?Q?Lex=E9n?=)
Date: Fri, 16 Mar 2001 12:38:45 +0100
Subject: Folding mode in win32all 138?
Message-ID: <3AB1FB45.4FD793D4@stop.spam.volvo.com>

What happened with the folding mode in win32all 138? I cant find it any
more. I miss it very much.

--
/Mikael

_________________________________________________
Mikael Lex?n
Volvo Information Technology AB
Dept 9514, HD2N
SE-405 08 G?teborg, Sweden



From owen at astrono.junkwashington.emu  Mon Mar  5 16:43:39 2001
From: owen at astrono.junkwashington.emu (Russell E. Owen)
Date: Mon, 05 Mar 2001 13:43:39 -0800
Subject: I come to praise .join, not to bury it...
References: <mailman.983485923.1136.python-list@python.org> <97p7iq$kk2$1@nntp6.u.washington.edu> <97rka002k9@news1.newsguy.com>
Message-ID: <9811a8$k5k$1@nntp6.u.washington.edu>

In article <97rka002k9 at news1.newsguy.com>,
 "Alex Martelli" <aleaxit at yahoo.com> wrote:

>"Russell E. Owen" <owen at astrono.junkwashington.emu> wrote in message
>news:97p7iq$kk2$1 at nntp6.u.washington.edu...
>    [snip]
>> However, I do agree that join is not intuitive. I believe the problem
>> (at least for the way I look at it) is that join should be a list
>> method, not a string method. I.e.:
>>
>>   joined_string = ['a', 'b'].join(', ')
>>
>> makes a lot of sense to me. The current join string method does not.
>
> (long discussion about polymorphism omitted).

I'm afraid I don't understand where the polymorphism comes in, so I 
cannot comment on that.

>...
>Say that I'm implementing a sequence-like object.  Would
>having .join as a part of the set of methods I must write
>be an _advantage_ to me?  As things stand now, to "be a
>sequence" I must just implement __getitem__ in the
>appropriate manner -- accepting an index from 0 to N-1,
>and raising IndexError for indices outside the range -- and
>maybe implement __length__ to return N.  With this small
>amount of effort, I gain the ability to be used as sequence
>in any existing context -- from the for statement onwards.

I'm afraid we disagree here. Collections can all inherit from a parent 
collection class, a perfectly reasonable requirement. You then define 
the few methods that are required to make your own kind of collection. 
The parent class should define things like join (which is, of course, a 
very simple method).

-- Russell


From corour01 at motorola.com  Tue Mar 20 14:52:48 2001
From: corour01 at motorola.com (O'Rourke Clodagh-corour01)
Date: Tue, 20 Mar 2001 19:52:48 -0000
Subject: File Descriptor Modes
Message-ID: <13EE655665F4D311AB4D0008C789498A01EE2C87@zei02exm02.cork.cig.mot.com>

Hi all,

Does anyone where I would find a summary of the different modes it is possible to open a file descriptor in?
I want to open a file and write data to it without writing over the existing contents of the file.
Do I need to explicitly find the end of the file and begin writing there or is there a mode that does this?
Thanks,
Clodagh



From parkw at better.net  Thu Mar  1 12:23:23 2001
From: parkw at better.net (William Park)
Date: Thu, 1 Mar 2001 12:23:23 -0500
Subject: Hypergeometric distribution ?
In-Reply-To: <y9vzof58mn0.fsf@genome.cbs.dtu.dk>; from thomas@cbs.dtu.dk on Thu, Mar 01, 2001 at 05:44:19PM +0100
References: <y9vzof58mn0.fsf@genome.cbs.dtu.dk>
Message-ID: <20010301122323.A732@better.net>

On Thu, Mar 01, 2001 at 05:44:19PM +0100, Thomas Sicheritz-Ponten wrote:
> Does anybody have a python function/class for calculating the
> hypergeometric distribution ?

Few links to search:
    SAL.KachinaTech.COM/B/
    www.netlib.org/cephes/
    www.nr.com
    www.python.org
    freshmeat.net

I remember Python wrapper advertised for 'cephes' which has
hypergeometric functions.  As last resort, you may have to transcribe
from C source to Python code.

---William Park, Open Geometry Consulting, Linux/Python, 8-cpu cluster



From claird at starbase.neosoft.com  Tue Mar 27 09:15:24 2001
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 27 Mar 2001 08:15:24 -0600
Subject: compiling questions, a la lisp
References: <mailman.985668094.3653.python-list@python.org>
Message-ID: <E634F2F1B5377E61.2CC39F5534B94719.AD658B57277BEA1E@lp.airnews.net>

In article <mailman.985668094.3653.python-list at python.org>,
Tim Peters <tim.one at home.com> wrote:
>[Pete Shinners]
>> ...
>> what is the biggest difference between lisp and python which makes
>> lisp more agreeable with native compiling?
>
>Legions of grad students in search of a thesis.
>
>progress-gets-made-on-things-that-are-worked-on-ly y'rs  - tim
>
>

SOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO true.
-- 

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


From thomas at xs4all.net  Sun Mar  4 22:27:50 2001
From: thomas at xs4all.net (Thomas Wouters)
Date: Mon, 5 Mar 2001 04:27:50 +0100
Subject: A counter-proposal to __future__ in PEP 236
In-Reply-To: <m3ae71dd22.fsf@atrus.jesus.cam.ac.uk>; from mwh21@cam.ac.uk on Sun, Mar 04, 2001 at 10:49:09AM +0000
References: <j47l29lpo8.fsf@informatik.hu-berlin.de> <97ll2u02qmp@news2.newsguy.com> <3A9FBE5C.95615241@sewardweb.com> <mailman.983578147.29040.python-list@python.org> <m3ae71dd22.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <20010305042750.W9678@xs4all.nl>

On Sun, Mar 04, 2001 at 10:49:09AM +0000, Michael Hudson wrote:
> Thomas Wouters <thomas at xs4all.net> writes:

> > On Fri, Mar 02, 2001 at 09:38:04AM -0600, Lloyd Sommerer wrote:
> > 
> > > I should also point out that the "from  __future__ import..." is also
> > > confusing to a newbie, but others have said that already.
> > 
> > And that's exactly as it should be: you shouldn't use the experimental
> > feature(s) if you're a newbie :)

> Bzzt.  If your new to Python, you should always use nested scopes, so
> you don't have to unlearn the old behaviour when the new behaviour
> becomes the default in 2.2.

Bzzt. If you're new to Python, chances are (nowadays) that you are using a
book. None of the books will cover 2.1 until after 2.2 is out.

-- 
Thomas Wouters <thomas at xs4all.net>

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



From vlouie at telusplanet.net  Sat Mar 24 20:35:43 2001
From: vlouie at telusplanet.net (Victor Louie)
Date: Sun, 25 Mar 2001 01:35:43 GMT
Subject: printing columns question
Message-ID: <PXbv6.52892$Lm2.6031361@news0.telusplanet.net>

Hi,

Just looking at some sample code...

        for i in trip[dest][1][1:] :
            print '\t%-15s%-15s%4d' % (prev, i, citymap[prev][i][0]),
            print '%5d:%02d' % (citymap[prev][i][1], citymap[prev][i][2])
            prev = i
        print "\t-------------------------------------------"
        print '\t%-30s%4d' % ("total:", trip[dest][0][0]),
        h = trip[dest][0][1] + trip[dest][0][2] / 60
        m = trip[dest][0][2] % 60
        print '%5d:%02d' % (h, m)

...i was just wondering if anybody could explain to me how the print
worked...
    ie. print '\t%-15s%-15s%4d' % (prev, i, citymap[prev][i][0]),

I understand that \t is a tab....but i don't get anything after that

If anyone could help it would be great.

Thanks
Victor





From ShaolinNOSPAM at btinternet.com  Tue Mar 13 17:19:07 2001
From: ShaolinNOSPAM at btinternet.com (P4R0XYSM)
Date: Tue, 13 Mar 2001 22:19:07 -0000
Subject: Einstein's Riddle
References: <oNPq6.396644$ge4.138479566@news2.rdc2.tx.home.com> <UdSq6.2184$54.2166@www.newsranger.com> <7z9r6.2576$54.2837@www.newsranger.com>
Message-ID: <98m6f3$n1q$1@plutonium.btinternet.com>

The statement you posted suggested that no owners have the same, well,
taking this logically I am assuming that this means they cannot share the
same pet, the statement does not explicitly specify that no two people can
own a fish, they could own different fishes, this would not be the same pet,
perhaps as a consequence of this, uncertainty comes into play and everyone
owns the fish at the same time.  I wonder if Einstein saw the quantum
implications behind his riddle.  Hehe

Only Joking

Ashley Mills
Gregory Jorgensen <gregj at pobox.com> wrote in message
news:7z9r6.2576$54.2837 at www.newsranger.com...
> I just looked at a bunch of web sites I turned up with Google searching
for
> "einstein riddle fish." All of them conclude that the German owns the
fish. I
> didn't find any confirmation that Einstein actually conceived the riddle
or said
> that only 2% of the population could solve it. Nor could I determine if
the
> original author wrote the riddle in English, German, or some other
language.
>
> Clearly my statement that the puzzle can't be solved puts me in a
minority. If
> you assume that SOMEONE keeps fish the puzzle can be solved with a little
bit of
> logic. Perhaps the intent of the riddle is to show how few people can
devise a
> useful and logical methodology.
>
> Greg Jorgensen
> Deschooling Society
> Portland, Oregon, USA
> gregj at pobox.com




From bencr at bigpond.com  Tue Mar 20 02:33:32 2001
From: bencr at bigpond.com (Ben)
Date: Tue, 20 Mar 2001 18:33:32 +1100
Subject: HELP - beginner to programing/python
References: <Vpxt6.416511$Pm2.6478704@news20.bellglobal.com>
Message-ID: <jqCt6.37019$v5.81476@newsfeeds.bigpond.com>

Try www.andamooka.org ... there is a free book posted there which will give
you a good start ... also read the docs :)
"EEN" <iansworld at thevortex.com> wrote in message
news:Vpxt6.416511$Pm2.6478704 at news20.bellglobal.com...
> Hi there. I am a newbie and am hearing that python is the way to go to
start
> off. I am downloading it as i type this, but need to know how to use it :)
> of course :) Is there any site that has a good tutorial or beginners
guides,
> or do you have to buy books? If so, waht do you recommend?
>
> Thanks a lot for the help, i really appreciate any pointers!
>
>




From superaaron at 263.net  Sat Mar 17 00:37:20 2001
From: superaaron at 263.net (Aaron)
Date: Sat, 17 Mar 2001 13:37:20 +0800
Subject: How can I get perl module for python?
Message-ID: <20010317133515.7C27.SUPERAARON@263.net>

I need to use "import perl"
but python tell me no module named perl found
where can I get it?
I'm using ActivePython 2.0 on win2k



From cribeiro at mail.inet.com.br  Wed Mar 28 12:33:48 2001
From: cribeiro at mail.inet.com.br (Carlos Alberto Reis Ribeiro)
Date: Wed, 28 Mar 2001 14:33:48 -0300
Subject: Speed problems with Python vs. Perl
In-Reply-To: <1znw6.1969$sk3.665386@newsb.telia.net>
References: <99sv33$8hk$1@news.cineca.it>
Message-ID: <5.0.2.1.0.20010328142623.021be9a0@mail.inet.com.br>

At 15:37 28/03/01 +0000, Fredrik Lundh wrote:
>for a backwards-compatible version of xreadlines, use
>the double-loop pattern from:
>
>http://effbot.org/guides/readline-performance.htm

Looking at your code there, a saw readline-example-3.py:

# readline-example-3.py
file = open("sample.txt")
while 1:
   lines = file.readlines(100000)
     if not lines:
       break
     for line in lines:
       pass # do something

Then I got to the library reference manual:

readlines ([sizehint])
   Read until EOF using readline() and return a list containing the
   lines thus read. If the optional sizehint argument is present,
   instead of reading up to EOF, whole lines totalling approximately
   sizehint bytes (possibly after rounding up to an internal buffer
   size) are read. Objects implementing a file-like interface may choose
   to ignore sizehint if it cannot be implemented, or cannot be
   implemented efficiently.


I think that this must be new on Python 2.0; I really dont remember it from 
my 1.5.2 days (maybe I overlooked it). So I have two questions:

1) What platforms do implement this? DO I need to check the code to have 
this information?
2) There is any documentation about buffer sizes for different platforms?


Carlos Ribeiro





From nas at arctrix.com  Sun Mar  4 02:47:14 2001
From: nas at arctrix.com (Neil Schemenauer)
Date: Sat, 3 Mar 2001 23:47:14 -0800
Subject: Help with finding how much memory a variable is using
In-Reply-To: <87n1b2vzke.fsf@pulsar.wlfdle1.on.wave.home.com>; from ksaranat@yahoo.com on Sun, Mar 04, 2001 at 06:05:38AM +0000
References: <JAwl6.17037$5n4.338201@news6-win.server.ntlworld.com> <87n1b2vzke.fsf@pulsar.wlfdle1.on.wave.home.com>
Message-ID: <20010303234714.A14614@glacier.fnational.com>

"Rich Somerfield" <rich_somerfield at tertio.com> writes:
> I am generating a huge, huge list of data (cant think of a
> decent way to reduce the required storage size and have it in a
> usable form).  Everytime i try to keep this list for future
> derivations I get a memory problem from Windows (effectively
> terminating my python script).

Depending on the data in the list the array module may help (eg
for storing an array of integers).

  Neil



From loewis at informatik.hu-berlin.de  Sun Mar 11 15:15:27 2001
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 11 Mar 2001 21:15:27 +0100
Subject: XML Module problems
References: <j4y9uhnb1m.fsf@informatik.hu-berlin.de> <memo.20010309203450.37329B@digitig.compulink.co.uk> <3aa93f65.699574094@news.iol.ie>
Message-ID: <j4hf10m59c.fsf@informatik.hu-berlin.de>

sean at digitome.com (Sean Mc Grath) writes:

> Yes. The original pyexpat used a list for attributes on the start
> element handler. A later version of pyexpat changed to a dictionary.
> The code on http://www.pyxie.org was changed to worth with both
> flavours of pyexpat.

The current version of pyexpat supports both: if the
ordered_attributes attribute of the Parser object is zero (which is
the default), it returns attributes; if it is nonzero, it returns
lists.

Regards,
Martin


From ddb at crystal.uwa.edu.au  Wed Mar 21 22:57:59 2001
From: ddb at crystal.uwa.edu.au (Douglas du Boulay)
Date: Thu, 22 Mar 2001 11:57:59 +0800
Subject: JPython1.1 jpythonc2 applet problems
Message-ID: <3AB97847.BBEB4021@crystal.uwa.edu.au>

Hi, 
I built a java applet using the jpythonc2 compiler and it 
runs perfectly well as a stand alone application. 

Unfortunately when I put it on a web server I get web browser 
security   member access violations as follows:

netscape.security.AppletSecurityException: security.member access
  at
netscape.security.AppletSecurity.checkMemberAccess(AppletSecurity.java:1158)
  at
netscape.security.AppletSecurity.checkMemberAccess(AppletSecurity.java:1146)
  at java.lang.Class.checkMemberAccess(Class.java:698)
  at java.lang.Class.getDeclaredConstructors(Class.java:575)
  at org.python.compiler.ProxyMaker.addConstructors(ProxyMaker.java:500)
  at org.python.compiler.ProxyMaker.build(ProxyMaker.java:614)
  at org.python.compiler.JavaMaker.build(JavaMaker.java:134)
  at org.python.core.MakeProxies.makeProxy(MakeProxies.java:77)
  at org.python.core.PyClass.init(PyClass.java:84)
  at org.python.core.Py.makeClass(Py.java:1306)
  at jstar.main$66(jstar.java:916)
  at jstar$_PyInner.call_function(jstar.java:313)



The original python code in question is:
# file: jstar.py
from star import VStar  # where star is a java package and
			# VStar is a java class in the package
class JSParse(VStar):



and this is translated into  the following java:
//file: jstar.java
903 private static PyObject main$66(PyFrame frame) {  
 ...
915     frame.setglobal("VStar",
imp.load("star").__getattr__("VStar"));   
916     frame.setglobal("JSParse", Py.makeClass("JSParse", new
PyObject[] {
                            frame.getglobal("VStar")}, c$47_JSParse,
null));


The question: Is is possible that this is an artifact of the "star" java 
package being imported via the wrong route (sys.path and not python.path
or something?), and is  there a way that jpythonc2 can actually create
the 
relevant proxy classes at compile time rather than at run time in the 
web browser which requires java reflection class introspection and its
associated security monitoring? 

Thanks 
Doug


From aahz at panix.com  Fri Mar 23 11:58:34 2001
From: aahz at panix.com (Aahz Maruch)
Date: 23 Mar 2001 08:58:34 -0800
Subject: sharing data between modules
References: <20010323.073801.76065818.26070@mead.troikanetworks.com>
Message-ID: <99fvbq$mmg$1@panix3.panix.com>

In article <20010323.073801.76065818.26070 at mead.troikanetworks.com>,
Bruce Edge <bedge at troikanetworks.com> wrote:
>
>As a C++ guy I tend to put each class in a different file. In python this
>gives me a different module for each class. My app uses this group of
>classes, ie: modules, what's the best way to share data between them?
>
>Currently I register the classes with each other using class static data, 
>so they all have a reference to each other. Ideally, I'd like to be able
>to have a few globals that every module can see.  

If you're sharing instance data, references are the best way to go.  If
it's truly global data, one of the advantages of Python modules is that
they *are* namespaces.  So either create a globals.py if variables are
shared across multiple classes or set variables in your class modules if
it's specific to a single class.
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

"I won't accept a model of the universe in which free will, omniscient
gods, and atheism are simultaneously true."  -- M


From db3l at fitlinxx.com  Tue Mar 13 20:47:16 2001
From: db3l at fitlinxx.com (David Bolen)
Date: 13 Mar 2001 20:47:16 -0500
Subject: This math scares me
References: <3aad4062.1290021@News.CIS.DFN.DE> <slrn9aqkth.aki.scarblac@pino.selwerd.nl> <3aad6ffe.3692710@News.CIS.DFN.DE>
Message-ID: <ug0gh15qz.fsf@ctwd0143.fitlinxx.com>

costas at meezon.com (Costas Menico) writes:

> Fortunately I discovered I can use "print 5.01 + 5.54" and it does the
> job. It has the extra intelligence built into it to autoformat

In case you're interested in why it has this difference, Python
generally has two ways of getting printable representations of
objects, which you can also access with the built-in functions str()
and repr().

str() is intended to provide a string of a "nice" or printable form of
the object, and generally avoids extra quoting, and in the case of
floating point numbers will perform some implicit rounding (or
truncation - well, one of the two).

repr() is intended to provide a string representation of the object
itself, such that as much as possible, you could pass the string to
eval() to recreate an instance of the object.

When you evaluate an expression at the interactive prompt (in recent
Python releases), Python will display the result object from the
expression using repr().  However, the print statement uses the str()
method on its arguments.

You can also obtain the same output as str() by using the "%s"
operator within a format string or by calling it directly.  

Note that as pointed out by many others, the repr() output is actually
more precise for what floating point values you are really storing in
the computer.  While print may make the output look nicer, you need to
be aware that the values are not truly as exact as they appear when
formatted with str(), which can be critical if you intend to use the
floating point values for any serious computations.

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


From moshez at zadka.site.co.il  Mon Mar 19 17:49:54 2001
From: moshez at zadka.site.co.il (Moshe Zadka)
Date: Tue, 20 Mar 2001 00:49:54 +0200
Subject: Python-related standards
In-Reply-To: <3d7l1l1jip.fsf@ute.cnri.reston.va.us>
References: <3d7l1l1jip.fsf@ute.cnri.reston.va.us>, <3dae6h21h8.fsf@ute.cnri.reston.va.us> <mailman.985022189.18857.python-list@python.org>
Message-ID: <E14f8TK-0006k3-00@darjeeling>

On 19 Mar 2001 17:27:58 -0500, Andrew Kuchling <akuchlin at mems-exchange.org> wrote:
> Moshe Zadka <moshez at zadka.site.co.il> writes:
> > FWIW, these could be "informational" PEPs.
> 
> OK.  Then the only thing missing from PEP 1 are 'Supersedes' and
> 'Superseded-by' headers, for replacing obsolete versions of documents.
> I'll send Barry a note.
 
Yes, these are needed for Standards Track PEPs too. 
(E.g., PEP-0228 is superseded by PEP-0237 - PEP-0240)
-- 
"I'll be ex-DPL soon anyway so I'm        |LUKE: Is Perl better than Python?
looking for someplace else to grab power."|YODA: No...no... no. Quicker,
   -- Wichert Akkerman (on debian-private)|      easier, more seductive.
For public key, finger moshez at debian.org  |http://www.{python,debian,gnu}.org



From t*b*r*u*m*l*e*y at f*u*s*e.n*e*t  Wed Mar 21 08:33:22 2001
From: t*b*r*u*m*l*e*y at f*u*s*e.n*e*t (Troy Brumley)
Date: Wed, 21 Mar 2001 08:33:22 -0500
Subject: Collection interfaces
References: <3a90fa97.0@news1.mweb.co.za>   <MCjm6.474652$U46.14163288@news1.sttls1.wa.home.com> <MPG.150394195f7f04a19896a7@news.earthlink.net> <3A9A86B1.9A8C69A2@ix.netcom.com> <MPG.1504d14563dce6499896af@news.earthlink.net> <MPG.1505a76831c34b749896b3@news.earthlink.net> <97h5s40vhc@news2.newsguy.com> <MPG.1506007591cf818b9896bc@news.earthlink.net> <slrn99opop.gr.amk@mira.erols.com> <MPG.1506265ebf6d934d9896be@news.earthlink.net> <97ibmc0k47@news1.newsguy.com> <MPG.1506cf3293b094b49896c1@news.earthlink.net> <97m0c7080c@news2.newsguy.com> <MPG.15085517debf4b649896ca@news.earthlink.net> <zb%n6.31033$o%6.586026@news1.rdc1.bc.home.com> <MPG.150b383cea226d409896d0@news.earthlink.net> <kNko6.6986$eq3.707121@news.uswest.net> <MPG.150c46db9cec283e9896d4@news.earthlink.net> <3AA2A898.450B659E@mail.com> <gRCt6.12224$Im6.1308560@newsread1.prod.itd.earthlink.net> <3AB755F8.67E84126@mail.com> <%a%t6.17245$227.1601569@newsread2.prod.itd.earthlink.net> <3AB8A730.4BE13D22@mail.com>
Message-ID: <B6DE17D2.D6B0%t*b*r*u*m*l*e*y@f*u*s*e.n*e*t>

in article 3AB8A730.4BE13D22 at mail.com, James A. Robertson at
jarober at mail.com wrote on 3/21/01 8:06 AM:

> topmind wrote:
>> 
> 
>>> http://www.cincom.com/smalltalk
>>> 
>> 
>> I only see a front page, not a parsing example.
>> 
> 
> Follow the tutorial link - it's right there on that page.
> 
> 
>> Besides, what are you comparing it to so say
>> it is better than a procedural/relational approach?
> 
> It's vastly easier than similar code done in C or basic, for example

User testimonial here ... last week I had to rip thorugh a pile of IIS logs
to get some statistics for some people, and I used the original (non-gui)
tutorial code as a starting point. I was once again impressed with how much
easier it was to do several common data processing tasks in Smalltalk than
it was in Assembly language or C or Pascal.

I'm a long time procedural type, learning to really think in and use
Smalltalk has been an interesting experience.

I've also done a lot of database "fixing" and Smalltalk has usually beat out
raw SQL for anything but the most trivial tasks.

All that said, I think it's the environment (interactive) that improves the
experience as much as the language itself. If I had a reasonable
interpretive BASIC environment available, I could probably do many of the
things I did in Smalltalk almost as easily ... certainly more easily than
with C.


> 
>> 
>>>> 
>>>>> [snip]
>>>> 
>>>> -tmind-
>>>> oop.ismad.com
>>> 
>>> --
>>> James A. Robertson
>>> Product Manager (Smalltalk), Cincom
>>> jarober at mail.com
>>> <Talk Small and Carry a Big Class Library>
>>> 



From dnew at san.rr.com  Sat Mar 31 19:26:40 2001
From: dnew at san.rr.com (Darren New)
Date: Sun, 01 Apr 2001 00:26:40 GMT
Subject: list vs tuple
References: <hGlw6.782$p5.3210@news1.rivrw1.nsw.optushome.com.au><3AC20CD3.265A2D94@alcyone.com><ru69cto61jtmrqsf809e38hb3pbm2lhi7g@4ax.com><bC1x6.4319$4N4.905946@newsc.telia.net><71q9ct4le63anefmiajj5cbh3mg04qu1ga@4ax.com><mailman.985991202.5578.python-list@python.org><7Bfx6.989$p5.3791@news1.rivrw1.nsw.optushome.com.au><Ukhx6.4421$4N4.950603@newsc.telia.net> <mailman.986049281.32365.python-list@python.org> <uGqx6.1052$p5.3997@news1.rivrw1.nsw.optushome.com.au>
Message-ID: <3AC675C1.EF9EB38B@san.rr.com>

deadmeat wrote:
> but thats not the point - which when the statement a = b copies the data
> across for one type it's expected to do it for others.  It doesn't, so it's
> inconsistant, contrary to the original claim.

Python doesn't copy integers either.
In python, you'd write
a = 5
b = a
b = 9
and a would still be 5 and b would still be 9.

The equivalent in Pascal would be

a, b, five, nine : ^integer;
five^ = 5; { ignoring "new" for now }
nine^ = 9;

a = five;
b = a;
b = nine;

What's the problem?

-- 
Darren New / Senior MTS & Free Radical / Invisible Worlds Inc.
San Diego, CA, USA (PST).  Cryptokeys on demand.
        schedule.c:7: warning: assignment makes calendar_week 
                          from programmer_week without a cast.


From pod at internode.on.net  Tue Mar 20 08:46:54 2001
From: pod at internode.on.net (PoD)
Date: Wed, 21 Mar 2001 00:16:54 +1030
Subject: Tkinter - catching window delete
Message-ID: <3AB75F4E.ED19B20F@internode.on.net>

I've been playing around with Tkinter for about a few days now and I've gotten a small app
to the point where I want it to bring up a "Do you want to save this file?" dialog when I
quit.
However I haven't been able to find how to catch the main window being closed by the window
manager.
The only doc that I have is 'An Introduction to Tkinter'.
Can anyone give me a clue please?

PoD.


From aahz at panix.com  Tue Mar 27 10:03:27 2001
From: aahz at panix.com (Aahz Maruch)
Date: 27 Mar 2001 07:03:27 -0800
Subject: Threading legality/morality
References: <fsDv6.1247$NR.100201@news3.oke.nextra.no> <99nnbv$phs$1@panix6.panix.com> <h0%v6.1842$NR.144463@news3.oke.nextra.no>
Message-ID: <99qa3v$6e0$1@panix6.panix.com>

In article <h0%v6.1842$NR.144463 at news3.oke.nextra.no>,
Syver Enstad <syver at NOSPAMcyberwatcher.com> wrote:
>
>I have seen some strange behaviour in some multi threaded programs I have
>made in python, where they seem to lockup, but a push on the the return key
>(or another key, I don't know) keeps them going. These are console programs,
>running under Win2000 server and professional, Python 2.0.
>
>As I was writing this I checked out my program that has been running for
>about 24 hours, downloading files from the internet via urllib, and using
>ADO via pythoncom. It has been running ever so happily until about 10 clock
>this morning when I inspected some output it had given. Then it froze for
>about three hours, when I pushed a key on the keyboard it continued again.
>Strange.

There is a misfeature in urllib such that the standard FancyURLOpener
class prompts for input if it gets an HTTP 401 (authorization needed).
You probably pressed ENTER.
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

"Boost the stock market -- fire someone"


From sandipan at vsnl.com  Thu Mar 22 21:28:54 2001
From: sandipan at vsnl.com (Sandipan Gangopadhyay)
Date: Fri, 23 Mar 2001 07:58:54 +0530
Subject: CGI for generating xml pages
Message-ID: <013b01c0b341$02438740$020000c3@node02>

Is text/xml in your mime-types on the server ?
Are you reporting text/xml correctly in the header ?

Regards,

Sandipan
----- Original Message ----- 
From: <lance.hill at olsinc.net>
Newsgroups: comp.lang.python
To: <python-list at python.org>
Sent: Friday, March 23, 2001 6:20 AM
Subject: CGI for generating xml pages


> 
> I have a python script which creates xml. If I redirect the output to
> a file, when I launch those pages from my browser all is well.
> Unfortunately, I can't seem to get it so I can write directly to my
> browser from the script. Instead, I get an internal server error.
> 
> Can someone kindly give me a clue? Do I need to do some processing
> prior to printing to standard output?
> 
> Thanks in advance,
> L. Hill
> -- 
> http://mail.python.org/mailman/listinfo/python-list
> 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20010323/c1a4c8ca/attachment.html>

From aleaxit at yahoo.com  Thu Mar 15 14:28:57 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 15 Mar 2001 20:28:57 +0100
Subject: Who's minister of propaganda this week?
References: <B98041329020E007.7DA84CAD9D0562AB.817BBC572F480869@lp.airnews.net><98nu1m$12e@dispatch.concentric.net><660ACCA1E4816F7E.C359CDF4168EF3FB.3B6B05538D619770@lp.airnews.net><3AAFA13D.76E32C3F@divalsim.it> <98oa4o01i4v@news2.newsguy.com><mailman.984619039.7856.python-list@python.org><98q9h60122v@news2.newsguy.com> <uy9u69b4b.fsf@advancedmp.net>
Message-ID: <007401c0ad87$965f6bc0$0300a8c0@arthur>

"Don Dwiggins" <dwig at advancedmp.net> writes:

> Alex Martelli writes:
> > Another way to express it: an interface is not just, not even
> > _mostly_, about existence and signature of methods -- it's mostly
> > about prereq's, post-conditions, and invariants; and nobody can
> > check those at compile-time in enough cases to make a difference
> > to your software's reliability.
>
> I agree whole-heartedly with the first clause, but I wonder: if you made
the
> same statement over on comp.lang.eiffel, what kind of response would you
> get from folks who use the language extensively?  Are Eiffel's prereq's,
> post-conditions, and invariants mostly ignored in practice, or are they
> commonly and effectively used?

AFAIK, the latter -- I don't know anybody who regularly posts on c.l.e,
but we do have ex-Eiffelists on our technical staff, and they were such
effective evangelists for the effectiveness of 'contracts' that we invested
considerable effort in preparing and using a C++ 'programming-as-
contracting' framework that is regularly used for our software production
(a constant grumble heard about COM, to which we are more and more
switching for all inter-component connectivity, is that there is no such
support -- we haven't expended the needed effort to prepare it, and
are now dissuaded from so doing by the perspective of moving to .NET,
which _does_ have some contract-support infrastructure already).

But contract-checking is done _at runtime_ (and often disabled for
production runs/releases, for those components whose need for speed
_justifies_ the use of fast-code-generating languages such as Eiffel
or C++...).


Alex



_________________________________________________________
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com




From gregj at pobox.com  Sat Mar 10 22:47:56 2001
From: gregj at pobox.com (Gregory Jorgensen)
Date: Sun, 11 Mar 2001 03:47:56 GMT
Subject: Problems writting a wrapper around linux' ssh command
References: <31rd89.lki.ln@127.0.0.1> <fdwq6.1906$54.1822@www.newsranger.com> <p9ae89.imr.ln@127.0.0.1>
Message-ID: <MzCq6.2024$54.2158@www.newsranger.com>

I'm not an ssh expert; I just know that it can authenticate without prompting
for a password, but only when the server and the client are configured correctly
with RSA keys. There are several versions of ssh, too. For more info I refer you
to:

http://www.ssh.com/

http://www.openssh.com/

http://www.oreilly.com/catalog/sshtdg/

I've seen quite a few web site updating/mirroring tools, but I don't know any
that use ssh/scp offhand. You may want to poke around freshmeat.

http://freshmeat.net/

Good luck.


In article <p9ae89.imr.ln at 127.0.0.1>, gradha at iname.com says...
>Er... I am new to ssh, so I don't even know what other methods there are...
>So far, I want to write a wrapper on top of sitecopy (a program which allows
>updating remote web sites through ftp) using ssh and scp to upload files
>and make directory modifications. If there's any other way of doing this,
>any help is very welcome. The only restriction is that the program I write
>from commandline or in background, accepts password only once, and logs
>everything to files, so I thought wrapping around ssh and scp would be the
>easiest way.

Greg Jorgensen
Deschooling Society
Portland, Oregon, USA
gregj at pobox.com


From loewis at informatik.hu-berlin.de  Sat Mar 24 05:20:27 2001
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 24 Mar 2001 11:20:27 +0100
Subject: New PEP: The directive statement
References: <200103201042.LAA20248@pandora.informatik.hu-berlin.de> <wzr8zpwn4e.fsf@sunshine.cs.uu.nl> <3d1yrp1osi.fsf@ute.cnri.reston.va.us> <3ABBA837.DEB8AB1B@san.rr.com> <99g9t7$r9j$1@panix6.panix.com>
Message-ID: <j4ae6bo4d0.fsf@informatik.hu-berlin.de>

aahz at panix.com (Aahz Maruch) writes:

> >What is the Python mechanism for ensuring a particular (or minimal) version
> >of Python is being used? Is there one?  I'm talking about something like
> >Tcl's [package require Tcl 8.3] or some such.
> 
> assert (sys.version_info[0] >= 2) and (sys.version_info[1] > 0)

That will fail on Python 3.0, so you better write

  assert sys.verion_info >= (2,0)

It also works for specific modules, e.g.

  assert xml.version_info >= (0,6,4)

Regards,
Martin


From aahz at panix.com  Tue Mar 20 11:08:23 2001
From: aahz at panix.com (Aahz Maruch)
Date: 20 Mar 2001 08:08:23 -0800
Subject: python suitability for large critical run forever apps
References: <3AB77CE4.78378862@troikanetworks.com>
Message-ID: <997v9n$76$1@panix3.panix.com>

In article <3AB77CE4.78378862 at troikanetworks.com>,
Bruce Edge  <bedge at troikanetworks.com> wrote:
>
>Multi-threading, and mutual exclusion. Are these available, reliable,
>and debuggable?

Not entirely sure what you mean by "debuggable", but threads and locks
are available and reliable.  Note that there's no good facility in
Python for cleaning up a thread that goes haywire (e.g. stuck on some
kind of blocking call).
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

There's a difference between a person who gets shit zie doesn't deserve
and a person who gets more shit than zie deserves.  --Aahz


From winkler at abaqus-sn.com  Thu Mar 29 12:45:16 2001
From: winkler at abaqus-sn.com (david winkler)
Date: Thu, 29 Mar 2001 17:45:16 GMT
Subject: how query function args
Message-ID: <3AC374BC.F90A2894@abaqus-sn.com>

I need to query the names and values of a function's keyword arguments.
I
suspect that there is an easy way to do this but cannot find it.  Would
someone care to make a suggestion?



From sholden at holdenweb.com  Mon Mar  5 20:54:05 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Mon, 5 Mar 2001 20:54:05 -0500
Subject: iis and python
References: <F5622749FED6DA5D.AF4594816A4865C9.16CD98F1C1FAC3F6@lp.airnews.net> <98191m$88j$1@bob.news.rcn.net>
Message-ID: <NuXo6.24167$1D5.970024@e420r-atl1.usenetserver.com>

"Satheesh Babu" <vsbabu at erols.com> wrote in message
news:98191m$88j$1 at bob.news.rcn.net...
> Hi,
>
> ASP with Python works really well. Additionally, Python's vast collection
of
> modules
> make a short work of many tasks ASP/VBScript pundits would need a compiled
> COM object for.
>
> We've some ASP/VB code. To extend it to connect to some network services
> I started using Python, mixed with VB and it rocks. When we get time,
> I'm certainly going to convert all VB stuff to Python.
>
> I  use ActiveState Python..
>
I find Python ASP to be somewhat faster than Python CGI (at least under
personal web server), as the interpreter startup time appears to dominate
performance on the short CGIs I have so far written.

However, neither of them hold a candle to Xitami's LRWP (log-running web
processes), which run like shit off the prverbial shovel, having done all
their database connection and initialization stuff before the web server
ever asks them to produce a line of HTML.

Next step will be to benchmark Xitami + LRWP against Apache + FastCGI.

if-only-there-were-time-ly y'rs  -steve





From fjs6 at usa.net  Thu Mar 22 02:00:58 2001
From: fjs6 at usa.net (fjs)
Date: Wed, 21 Mar 2001 23:00:58 -0800
Subject: Good Editor? 
Message-ID: <3ab9a32b$1_1@news2.one.net>

Can someone recomend a good editor to write python in?  Im just getting
started learning python and dont know which ones support python.

thanks





From paulp at ActiveState.com  Thu Mar 15 08:08:11 2001
From: paulp at ActiveState.com (Paul Prescod)
Date: Thu, 15 Mar 2001 05:08:11 -0800
Subject: Cameron Laird Quote on www.unixinsider.com -- Probably old news to 
 y'all, but new to me.
References: <bJOr6.9481$TW.38047@tor-nn1.netcom.ca> <3AB00593.D1C4BBBF@fxtech.com> <3AB08001.9532A3F0@letterror.com>
Message-ID: <3AB0BEBB.E7F57F5D@ActiveState.com>

Just van Rossum wrote:
> 
>...
> 
> Yeah, I was going to say: there's an awfully large application area _between_
> scripting and interactive 3d graphics apps! Python already does quite well
> with interactive 2d graphics, though: see eg. Sketch.

There were video game guys at the conference who say they use Python on
the client side for the AI. On the other hand, even the most basic
script depends heavily on parts of the interpreter and the OS written in
C. So the vast majority of apps will profit from a mix of languages. But
each year the proportion will shift from towards Python and away from C.
Embedding Python will become less interesting than embedding (and
inlining!) C, just as C inlines assembly.

-- 
Take a recipe. Leave a recipe.  
Python Cookbook!  http://www.activestate.com/pythoncookbook



From fredrik at pythonware.com  Mon Mar 12 16:57:19 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Mon, 12 Mar 2001 21:57:19 GMT
Subject: This math scares me
References: <3aad4062.1290021@News.CIS.DFN.DE>
Message-ID: <3Dbr6.17913$Qb7.3116049@newsb.telia.net>

costas at springmail.com wrote:
> Ok, I can see maybe division having problems. But why does addition of
> the two numbers below:
>
> 5.01+5.54
>
> give me this?
>
> 10.550000000000001

see

http://www.python.org/cgi-bin/moinmoin/FrequentlyAskedQuestions

    Q. Floating point is broken!

    A. Binary floating point cannot represent decimal
    fractions exactly, so some rounding always occurs

and

http://www.python.org/cgi-bin/moinmoin/RepresentationError

    ... some (most, actually) decimal fractions cannot
    be represented exactly as binary (base 2) fractions.
    This is the chief reason why Python (or Perl, C, C++,
    Java, Fortran, ...) often won't display the exact
    decimal number you expect ...

Cheers /F

<!-- pythonworks 1.2 coming soon:
http://www.pythonware.com/products/works/index.htm
-->




From tim.one at home.com  Fri Mar  2 17:24:22 2001
From: tim.one at home.com (Tim Peters)
Date: Fri, 2 Mar 2001 17:24:22 -0500
Subject: Unresolved symbols in MSVCRT.DLL when running a py2exe packaged program.
In-Reply-To: <xmTn6.15713$UZ4.4329504@news4.rdc1.on.home.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCGEIMJDAA.tim.one@home.com>

[Tom, on System File Protection in recent flavors of Windows]
> I didn't mean to suggest 'fooling' it - I agree that safeguards
> like SFP are needed - I meant that one had to do it according to
> the rules.

What rules?  Only MS can replace system files now -- there is no provision
for "one" to *ever* do it anymore, if one != MS (no API, no magic dance, not
even an option to turn SFP off).  One of the links I posted before suggests
that they *may* make an exception for developers of new hardware.

> As for not being able to update the DLL's, this would be a problem.  If a
> service pack for VC6 updates the MFC or CRT DLL's (which SP's usually do),
> and you ship an app build with it, then you must update these DLL's on your
> target system.  Perhaps this is where the new ability to run 'personal'
> instances of DLL's comes in: ie. you can't update the system32 version the
> DLL, but you can install your own personal copy.

That's my understanding, yes.  As one of the links pointed out, .inf files
don't have conditionals, so MS expects you to ship different infs for
different MS OSes now (depending on whether you can or cannot exploit private
system DLLs on the OS in question).

Python hasn't needed to deal with that yet.  Wrestling with Windows esoterica
isn't really a passion of mine, so I'm not looking forward to the day when it
must.

luckily-i'm-told-everyone-will-be-on-linux-then-anyway-ly y'rs  - tim




From amitp at Xenon.Stanford.EDU  Mon Mar 12 10:41:54 2001
From: amitp at Xenon.Stanford.EDU (Amit Patel)
Date: 12 Mar 2001 15:41:54 GMT
Subject: gah! I hate the new string syntax
References: <mailman.983485923.1136.python-list@python.org> <97sudi01i26@news1.newsguy.com> <98elrk$li8$1@nntp.Stanford.EDU> <98fc070thh@news1.newsguy.com>
Message-ID: <98iqo2$56o$1@nntp.Stanford.EDU>

 Alex Martelli <aleaxit at yahoo.com> wrote:
| "Amit Patel" <amitp at Xenon.Stanford.EDU> wrote in message
| news:98elrk$li8$1 at nntp.Stanford.EDU...
|     [snip]
| > What I mean to say is that I cannot write new methods on strings.  So
| 
| Perfectly true, just as you cannot on dictionaries, lists, tuples,
| file objects, and so on.  Only classes and extension-classes
| are extensible (by inheritance and overriding), not types.

Even if strings were a class, and I could extend it, it is NOT the
thing to do to add new functionality like capitalize().  Suppose
strings do not have a capitalize method, and I want to add one.  And
suppose strings are classes and can be extended.  If I wrote:

  class MyString(string):
     def capitalize(self):
        ...


then I can only capitalize strings I created!  I can't take
' '.join(['lister', 'is', 'a', 'smeghead'] and run .capitalize() on
it -- because it's an ordinary string and not a MyString.

The advantage of method functions like string.capitalize is that they
do not require me to change something I can't change (the string
type).  I can write capitalize(x) when x is a string.  I can't write
x.capitalize() when x is a string.

Defining methods is like running things as root.  Yes, you sometimes
need to do it, but it's best to limit yourself to only those things
that need superuser access.  Run as much as you can as non-root.  And
IMO that goes for methods and functions too -- string type methods are
"root" (where "root" = "Guido & pals") and have access to the
internals of a string, whereas functions on strings are user-level and
go through the abstractions.

    - Amit


P.S.  Thanks for reminding me about type != class :-)
-- 
--
Amit J Patel, Computer Science Department, Stanford University
http://www-cs-students.stanford.edu/~amitp/


From shichang at icubed.com  Tue Mar 27 22:32:13 2001
From: shichang at icubed.com (szhao)
Date: Wed, 28 Mar 2001 03:32:13 GMT
Subject: Where is the DBI module?
References: <MPG.152ab2bb88e6482d9896a4@news.pacbell.net>
Message-ID: <1Xcw6.344$j_3.253409@typhoon2.ba-dsg.net>

Python/win32/dbi.dll, Python/win32/odbc.pyd?

"Owen Ap Owen" <no at thank.you> wrote in message
news:MPG.152ab2bb88e6482d9896a4 at news.pacbell.net...
> I'm confused about something that probably has an obvious answer.
>
> Where is the DBI module?
>
> I can "import DBI" but when I try to find the actual module in order to
> examine it I can't. Am I missing something obvious?




From shaleh at valinux.com  Wed Mar 28 18:39:37 2001
From: shaleh at valinux.com (Sean 'Shaleh' Perry)
Date: Wed, 28 Mar 2001 15:39:37 -0800 (PST)
Subject: i need help on builtin function math
In-Reply-To: <3AC27472.AC25F54B@gmx.de>
Message-ID: <XFMail.20010328153937.shaleh@valinux.com>

On 28-Mar-2001 Marcus Konermann wrote:
> hello !
> 
> i want to use the sqrt() function of the builtin funtion math in that
> kind:
> 
> import math
> ...
> a=0
> a=sqrt(whatever)
> 
> and than a NameError: sqrt occured and i don?t know why ? Is it
> depending on the OS (i use Windows98 and the pywin
> editor). Other functions, like pow(), from the math module are working.
> Can anyone help me with my problem ?
> 

if you 'import math' then you have to call it as math.sqrt().



From rapto at arrakis.es  Thu Mar 22 11:18:48 2001
From: rapto at arrakis.es (Marcos Sįnchez Provencio)
Date: Thu, 22 Mar 2001 17:18:48 +0100
Subject: Transparent GIFs in PIL
Message-ID: <newscache$m8xlag$gk9$1@proxy.bitmailer.com>

Is it possible to set a transparent color in a GIF file (originally not
transparent) in PIL?




From erno-news at erno.iki.fi  Fri Mar 30 21:15:05 2001
From: erno-news at erno.iki.fi (Erno Kuusela)
Date: 31 Mar 2001 05:15:05 +0300
Subject: Speedup hint for users of viewcvs package
References: <m14j45Y-000CnEC@artcom0.artcom-gmbh.de> <190551356909.20010330211141@buz.ch> <mailman.985990600.4176.python-list@python.org>
Message-ID: <kuelveu1jq.fsf@lasipalatsi.fi>

In article <mailman.985990600.4176.python-list at python.org>, Neil
Schemenauer <nas at python.ca> writes:

| I have no idea why more people don't know about FastCGI.

one reason is probably that mod_fastcgi doesn't come with apache
(because of the license?) and then it's just as easy to download
and compile and configure mod_python or mod_snake.

  -- erno


From ws-news at gmx.at  Thu Mar 22 10:54:43 2001
From: ws-news at gmx.at (Werner Schiendl)
Date: Thu, 22 Mar 2001 16:54:43 +0100
Subject: Good Editor?
References: <3ab9a32b$1_1@news2.one.net>
Message-ID: <985276778.471557@newsmaster-04.atnet.at>

I like the PythonWin IDE that comes with the ActivePython distribution from
ActiveState for experimenting. You have everything in one place, immediate
Window, your scripts, the results from running these, a debugger is also
included.

Don't know how well it bahves for large projects...

regards
werner

"fjs" <fjs6 at usa.net> wrote in message news:3ab9a32b$1_1 at news2.one.net...
> Can someone recomend a good editor to write python in?  Im just getting
> started learning python and dont know which ones support python.
>
> thanks
>
>
>
>




From zeitlin at seth.lpthe.jussieu.fr  Wed Mar 14 12:50:00 2001
From: zeitlin at seth.lpthe.jussieu.fr (Vadim Zeitlin)
Date: 14 Mar 2001 17:50:00 GMT
Subject: loading Python20.dll during run-time under Windows
References: <slrn9alouc.sor.zeitlin@seth.lpthe.jussieu.fr> <5v3oatsgb21fu9ds1ci4dsuk87apbb1arm@4ax.com>
Message-ID: <slrn9avbq7.p8n.zeitlin@seth.lpthe.jussieu.fr>

On Sun, 11 Mar 2001 23:56:01 +0000, Paul Moore <gustav at morpheus.demon.co.uk> wrote:
>Look at the source code for Vim 6

 Ah, thanks a lot! I really should have thought about myself, being a vim-only
user but for some reason I forgot about this.

>Basically, it's simple, but tedious

 Yes, exactly my point - and why I didn't want to type all this in myself.

 Thanks again for the hint!
VZ
-- 
GCS/GM d? H+ s++:-- p2 au--- a- w+ v C+++ UBLS+++ P- L++ N++ E--- W++++ M? V--
-po+ R++ G`` !tv b+++ D--- e++++ u++ h--- f+ r++ n- y?


From fredrik at pythonware.com  Mon Mar 26 13:37:00 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Mon, 26 Mar 2001 18:37:00 GMT
Subject: portable string localization and resource bundles...
References: <mailman.985605333.14558.python-list@python.org>
Message-ID: <g%Lv6.1691$sk3.586837@newsb.telia.net>

Clark C. Evans wrote:
> I was wondering if there is a solution to the
> common problem of string localization
> (menu and message box titles, for instance).

here's one:
http://www.python.org/doc/current/lib/module-gettext.html

Cheers /F




From kens at sightreader.com  Wed Mar 28 16:05:35 2001
From: kens at sightreader.com (Ken Seehof)
Date: Wed, 28 Mar 2001 13:05:35 -0800
Subject: list vs tuple
Message-ID: <005a01c0b7ca$d5a01990$04090a0a@upcast.com>

From: "Aahz Maruch" <aahz at panix.com>
Newsgroups: comp.lang.python
To: <python-list at python.org>
Sent: Wednesday, March 28, 2001 11:35 AM
Subject: Re: list vs tuple


> In article <buqw6.1999$sk3.675030 at newsb.telia.net>,
> Fredrik Lundh <fredrik at effbot.org> wrote:
> >Mark Pilgrim wrote
> >>
> >> It was my understanding that only tuples containing immutable objects
> >> could be used as dictionary keys.  For instance, (1,2,3) can be, and
> >> ("a","b","c") can be, but ([1,2,3], ["a","b","c"]) can not, because
> >> the underlying objects could mutate and cause problems.  (Wow, that
> >> sounds like a bad sci-fi movie.)  Is this true?
> >
> >no python interpreter within reach today?
> >
> >>>> d = {}
> >>>> d[(1, 2, 3)] = 1
> >>>> d[("a", "b", "c")] = 2
> >>>> d[([1, 2, 3], ["a", "b", "c"])] = 3
> >Traceback (most recent call last):
> >  File "<stdin>", line 1, in ?
> >TypeError: unhashable type
>
> Sure, but someone could create a class with __hash__ and do some weird
> stuff that way.  Not Recommended.
> --

Actually it's not that bad if you follow the rules.  You should make sure
that __hash__ only uses private members (starting with "__") that are only
modified in __init__.  You also have to overload __cmp__ so that it also
depends on the same members as __hash__ (i.e. hashes should be equal if
instances are equal).  Or you can make __hash__ return id(self) if all
instances are considered unique.

If you mess up, only the offending items will become inaccessible in your
dictionary.  Nightmares of the hash table becoming completely corrupt are
exagerated.  Cross-links won't happen since a compare will verify the hash
hit, so bugs won't get overly nasty.  The following will safely check the
integrity of a dictionary (and raise a KeyError on damaged hash values):

for k in d.keys(): d[k]

- Ken Seehof
kseehof at neuralintegrator.com
www.neuralintegrator.com



-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20010328/657a3639/attachment.html>

From MarkH at ActiveState.com  Fri Mar 23 18:06:38 2001
From: MarkH at ActiveState.com (Mark Hammond)
Date: Fri, 23 Mar 2001 23:06:38 GMT
Subject: Commercial Products in Python?
References: <3abb86ee.142862695@localhost>
Message-ID: <3ABBD84B.9020404@ActiveState.com>

Victor Muslin wrote:

> A friend of mine is considering developing a shrink-wrapped commercial
> product. Python seems like a possible choice, but it is not clear
> whether there are any precedents (though probably similar concerns
> apply to a Perl-based product).
> 
> I would very much appreciate an advice in this area and possibly
> answers to some of the following questions:
> 
> 1. Are there any existing shirnk-wrap commercial products written in
> Python?
> 
> 2. If so, what are they? 

http://www.80-20.com/ has shrink-wrap apps written significantly in Python.

> 3. What is the best way to go about it? Is it better to have customers
> install Python and give them bytecode files or use one of the programs
> that creates an executable?

If it is truly a commercial app you should have your own professional installer that installs Python and the application.  Using a tool that creates an executable may be nice, but not a requirements IMO.


> 4. There is an easy shrink-wrap install for Python (at least on
> Windows NT). Are there such on Unix/Linux platforms? (Typically
> customers don't like having to compile software from source).

Sure - your installer would do basically what the Python installer itself does.

> 6. Are there any licensing issues to be aware of?
IANAL - read the license :)

> Anything else that comes to mind in this direction would be very
> helpful.

Just do it (tm) :)

Mark.



From erno-news at erno.iki.fi  Thu Mar  1 22:52:14 2001
From: erno-news at erno.iki.fi (Erno Kuusela)
Date: 02 Mar 2001 05:52:14 +0200
Subject: python irc channel?
References: <97n15e$9g3$1@nntp1.ba.best.com>
Message-ID: <kur90gq13l.fsf@lasipalatsi.fi>

In article <97n15e$9g3$1 at nntp1.ba.best.com>, mjbjr at best.com (Martin
Brown) writes:

| So, is there a preferred irc channel where python people hangout?

there is a bridge bot joining the #python channels of
ircnet, openprojects and efnet.

  -- erno


From fredrik at pythonware.com  Thu Mar  1 14:28:10 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Thu, 01 Mar 2001 19:28:10 GMT
Subject: A counter-counter-proposal for PEP 236: #pragma( nested_scopes)
References: <j47l29lpo8.fsf@informatik.hu-berlin.de> <mrwn6.16173$Qb7.2637930@newsb.telia.net> <0Swn6.3324$TW.17061@tor-nn1.netcom.ca>
Message-ID: <epxn6.16181$Qb7.2640372@newsb.telia.net>

Warren Postma wrote:
> I'm willing to attempt a patch if anybody else likes this idea.  The pragma
> thing can be useful for other things as well.  As python grows, this
> mechanism can be used instead of lots of other little cumulatively uglier
> hacks.

But Python isn't supposed to grow lots of pragma-style settings.
There shouldn't be any "other things".  There Should Only Be One
Language.

We're talking about two kinds of "directives" here:

-- Source code encoding, which needs to be specified at the at
the top of the source file, since it affects the tokenizer/parser
(cf. XML encodings).  As specified, "directive" doesn't cut it.

-- Experimental compiler features, which if successful will become
standard in the next release.  Again, "directive" doesn't work, since
it makes "experimental" stuff look "optional".

And using "pragma" instead of "directive" doesn't really change a
thing; it's still hypergeneralization.

Cheers /F




From digitig at cix.co.uk  Sun Mar 18 17:35:00 2001
From: digitig at cix.co.uk (Tim Rowe)
Date: Sun, 18 Mar 2001 22:35 +0000 (GMT Standard Time)
Subject: Thoughts about extensions to the Python language
References: <mailman.984609740.11496.python-list@python.org>
Message-ID: <memo.20010318223555.39809B@digitig.compulink.co.uk>

In article <mailman.984609740.11496.python-list at python.org>, 
jeremy at alum.mit.edu (Jeremy Hylton) wrote:

> Guido did a poll at developers day last week.  The question was: "Is
> Python development moving too fast or at just the right speed?"  (He
> was being dictatorial and didn't even consider votes from the
> more-faster crowd.)
>
> I believe the results were 42 too fast and 50 just right.  I noted
> that two of the more vocal proponents of the too fast camp proceeded
> to suggest new features.  So at least some of the "too fast" people
> probably meant "too fast unless it's a feature I want" :-).

<snip>

The speed of development isn't (directly) the issue. Features that fit 
cleanly and "Pythonically" into the language can come as fast as they 
like as far as I am concerned. I'm not fussed whether they're useful to me 
or not, just as long as they don't louse up the language. The problem is 
when the speed of development is such that misfeatures get shoehorned in 
without time for proper consideration of whether they fit -- the 
monstrosity of "//" springs immediately to mind. If this is not nipped in 
the bud /very/ quickly I predict that Python will be a dinosaur existing 
only for legacy code and programmers within two years. Sad :-(


From jafo at tummy.com  Thu Mar  1 00:51:23 2001
From: jafo at tummy.com (Sean Reifschneider)
Date: Wed, 28 Feb 2001 22:51:23 -0700
Subject: Performance when working directory is slow network drive
In-Reply-To: <983411179.898268@clam-ext>; from m.hadfield@niwa.cri.nz on Thu, Mar 01, 2001 at 02:45:25PM +1300
References: <983411179.898268@clam-ext>
Message-ID: <20010228225123.B12659@tummy.com>

On Thu, Mar 01, 2001 at 02:45:25PM +1300, Mark Hadfield wrote:
>If the working directory is on one of my hard drives the elapsed times is
>approx 0.1 seconds; if I then to a slow network drive (a Samba share to a
>Unix machine that mounts another machine's disks via NFS!) it takes 2
>seconds.

Have you tried removing the current directory from sys.path?  That'd be
the one that's empty.

Sean
-- 
 -rw-rw-rw-:  File permissions of the beast.  (Chuck Phillips)
 mknod /dev/beast c 6 66
Sean Reifschneider, Inimitably Superfluous <jafo at tummy.com>
tummy.com - Linux Consulting since 1995. Qmail, KRUD, Firewalls, Python



From vincent_a_primavera at netzero.net  Wed Mar 28 11:55:18 2001
From: vincent_a_primavera at netzero.net (Vincent A. Primavera)
Date: Wed, 28 Mar 2001 11:55:18 -0500
Subject: General CGI advice...
Message-ID: <01032811551700.00576@tos1>

Hello,
	Looking for some general guidance...  I am relatively new to Python and 
pretty much totally green to CGI programming.  However, I would like to start 
learning more about CGI using Python as my means.  Do I need to focus on 
generic CGI concepts first or are there any really good resources available 
to integrate the two immediately?

	Thank You,
	Vincent A. Primavera



From scarblac at pino.selwerd.nl  Mon Mar 26 07:32:05 2001
From: scarblac at pino.selwerd.nl (Remco Gerlich)
Date: 26 Mar 2001 12:32:05 GMT
Subject: Is this an inconsistency?
References: <slrn9bud88.2le.apardon@trout.vub.ac.be>
Message-ID: <slrn9budgl.mbh.scarblac@pino.selwerd.nl>

apardon at trout.vub.ac.be <apardon at trout.vub.ac.be> wrote in comp.lang.python:
>   map(None , list)
> 
> Doesn't turn into a list of singletons but
> just returns the list.
> 
> This seems a bit odd to me because if you
> use apply with a single parameter function
> the second argument has to be either a singleton
> or a list with one element.
> 
> 
> My question now: Is this behaviour changed in
> the following versions? If not, wouldn't it be
> more consistent if it was?
 
It's inconsistent if you see it this way, yes. On the other hand, if you
think of different functions to call map with that take one argument, it's
pretty nice that map(None, l) just returns l unchanged. It would be weird if
the 'None' function made tuples. (Some people would see that an exception or
returning a list of Nones would be more consistent, but that would be
useless).

With multiple lists, returning the list of tuples is the closest thing you
can do to returning the arguments to the 'None-function' unchanged.

In 2.0 and up, there is the zip() function, which does exactly what you want.

>>> zip(range(3))
[(0,), (1,), (2,)]
>>> zip(range(3),"abc")
[(0, 'a'), (1, 'b'), (2, 'c')]

-- 
Remco Gerlich


From no at thank.you  Tue Mar 27 17:23:28 2001
From: no at thank.you (Owen Ap Owen)
Date: Tue, 27 Mar 2001 14:23:28 -0800
Subject: Where is the DBI module?
Message-ID: <MPG.152ab2bb88e6482d9896a4@news.pacbell.net>

I'm confused about something that probably has an obvious answer.

Where is the DBI module?

I can "import DBI" but when I try to find the actual module in order to 
examine it I can't. Am I missing something obvious?


From scarblac at pino.selwerd.nl  Sat Mar 10 18:40:45 2001
From: scarblac at pino.selwerd.nl (Remco Gerlich)
Date: 10 Mar 2001 23:40:45 GMT
Subject: UserDict and items()
References: <98656v$cud$1@news.rrz.Uni-Koeln.DE> <slrn9aed8i.cg1.scarblac@pino.selwerd.nl> <98anga01trm@news2.newsguy.com>
Message-ID: <slrn9aleo5.h09.scarblac@pino.selwerd.nl>

Alex Martelli <aleaxit at yahoo.com> wrote in comp.lang.python:
> "Remco Gerlich" <scarblac at pino.selwerd.nl> wrote in message
> news:slrn9aed8i.cg1.scarblac at pino.selwerd.nl...
>     [snip]
> > The cuteness introduced in later Python versions would allow that to be:
> >
> >    def __getitem__(self, key):
> >       return self.data.setdefault(key, DatabaseCall(key))
> >
> > Ie, look up everything in the database that's not in the dictionary yet,
> put
> > it in, and return what's in the dict.
> 
> Unfortunately, this approach is not lazy -- DatabaseCall is here
> being called also for keys which ARE already among those of self.data.
> A more explicit expression is needed to allow the laziness.

Wow, what an oversight. Of course, that defeats the whole purpose of the
code, thanks for catching it...

-- 
Remco Gerlich


From claird at starbase.neosoft.com  Thu Mar 15 10:27:42 2001
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 15 Mar 2001 09:27:42 -0600
Subject: Lexical synchronization (was: Who's minister of propaganda this week?)
References: <B98041329020E007.7DA84CAD9D0562AB.817BBC572F480869@lp.airnews.net> <98nu1m$12e@dispatch.concentric.net> <98ph26$5a5@dispatch.concentric.net>
Message-ID: <4CFA6A6DFAD0A2CD.DCFFF1727C3312C8.157CACA372ADC451@lp.airnews.net>

In article <98ph26$5a5 at dispatch.concentric.net>,
Phlip  <phlip_cpp at my-deja.com> wrote:
			.
			.
			.
>> an object is data with code attached
>> a closure is code with data attached
>
>Correct. But a closure declares implicitly, on the spot. That's why you 
>don't have to interrupt your own flow to orchestrate one from scratch, as 
>an object, the way you would in Python or (gods help us) C++.
			.
			.
			.
I've always called this quality "anonymity".  Is
that a private vocabulary, or does anyone else
understand me when I talk that way?
-- 

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


From kens at sightreader.com  Fri Mar 16 17:58:42 2001
From: kens at sightreader.com (Ken Seehof)
Date: Fri, 16 Mar 2001 16:58:42 -0600
Subject: Cookies
References: <mailman.984707839.22839.python-list@python.org> <lmq6m07j.fsf@propylon.com> <mailman.984773062.25819.python-list@python.org> <uzoel8ed8.fsf@ctwd0143.fitlinxx.com>
Message-ID: <003401c0ae6e$274f2df0$04090a0a@upcast.com>

----- Original Message -----
From: "David Bolen" <db3l at fitlinxx.com>
Newsgroups: comp.lang.python
To: <python-list at python.org>
Sent: Friday, March 16, 2001 5:48 PM
Subject: Re: Cookies


> "Ken Seehof" <kens at sightreader.com> writes:
>
> > Does anyone know of another mailing list more specific to http
questions?
> >
> > 1. Client downloads an execuatable (frozen python application).
> > 2. Client downloads a web page.
> > 3. Web page javascript generates a statistically unique random
identifier.
> > 4. Web page sends identifier to the server, which uses it as database
key.
> > 5. User runs python application
> > 6. *** Python application somehow gets the identifier from somewhere ***
> > 7. Now the python application and the server both have the same
identifier.
>
> Because of the restrictive client Javascript environment you'll
> probably be faced with, crossing the boundary between inside the
> browser and outside is not trivial.

Yup.

> Is there a reason that the identifier has to only be generated by 3?
> It's not clear if your Python application is going to connect to the
> server anyway, but if so couldn't it just do the same processing
> you're putting into the javascript and generate a unique identifier
> (that it thus knows about) at that point?

Yes, the application connects to the server.  Specifically, it queries
the server with the identifier as the key.  The question is: what is
the processing that can be done by both the javascript and the
python application that will generate the same number?

> Or, if it isn't going to connect to the server, how about just letting
> the Python application open up it's own web page internally to run
> appropriate code on your web server to generate the identifier (which
> could run as a server script), and then Python could parse it right
> from the resulting HTML information.

Again, it's the same catch-22.  I need the server to generate the same
id that the javascipt generated earlier, but I need each client to have
a unique id.

> > The only question I am asking is how to do step 6.
> >
> > Alternative:
> >
> > 1. Client downloads an execuatable (frozen python application).
> > 2. Client downloads a web page.
> > 3. *** Web page javascript makes a unique identifier for the client. ***
> > 4. Web page sends identifier to the server, which uses it as database
key.
> > 5. User runs python application
> > 6. *** Python application somehow generates the same identifier ***
> > 7. Now the python application and the server both have the same
identifier.
> >
> > Again the only question I am asking is how to do steps 3 and 6 to get
the
> > same number.
>
> This would probably prove even harder.  Algorithmic generation of a
> truly unique identifier will almost always include some time based
> deviation, so I'm not sure about an algorithm that will both be
> globally unique and re-computable at different times - at least not
> without still passing some information between the two compute engines
> which puts you back at your same problem.
>
> --
> -- David
> --
> /-----------------------------------------------------------------------\
>  \               David Bolen            \   E-mail: db3l at fitlinxx.com  /
>   |             FitLinxx, Inc.            \  Phone: (203) 708-5192    |
>  /  860 Canal Street, Stamford, CT  06902   \  Fax: (203) 316-5150     \
> \-----------------------------------------------------------------------/
> --
> http://mail.python.org/mailman/listinfo/python-list
>




From jdavis at dynworks.com  Tue Mar 13 05:18:04 2001
From: jdavis at dynworks.com (Jeff Davis)
Date: Tue, 13 Mar 2001 02:18:04 -0800
Subject: module importing and variable syntax
Message-ID: <3AADF3DC.1030003@dynworks.com>

I was previously a perl user, but now I am taking an interest in python. 
I am used to common perl methods such as using "variable variables" 
where the accessed variable depends on the situation (perl: "print 
${$x};"). This is not especially important because that is essentially a 
dictionary (in python terms), but combined with the module importing 
(the way python does it is different than perl) I am having some problems.

Module importing doesn't take a string argument, but just a "bare word" 
(in perl terms). This means I have to know the name of the file in 
advance, and it needs to be in the right directory, or I need to use eval().

What I am trying to do is have an arbitrary string with an arbitrary 
type (type with respect to my program, not python) access the function 
associated with the string in the module associated with the type.

For example: someone writes an extension to my program creating a new 
object of name "O". Then some other part of the program (i.e. another 
extended module) tries to square an instance of that object, like:
o = O()
o.square()

However, my program is completely unaware of either "O" or "square" 
until someone writes the module(s) and registers them with the program 
(not writing to the main program's source).

It seems like I could get away with it if I used a lot of eval() 
statements, but talk about unreadable code (especially since you have to 
concatenate the variables in python to form a string)!

I think I am having problems understanding the way python works in this 
regrard, which is why this question was probably way more complex than 
it needed to be.

I would really appreciate some clarification. I like a lot about the way 
python is made and I think it has a lot of potential. Right now the 
concatenation of strings can be annoying compared to just (perl) 
'$string = "a $anotherstring be";', but for a lot of programs it is well 
worth the tradeoff.

Thanks,
	Jeff Davis



From tjg at exceptionalminds.com  Sun Mar 25 20:01:40 2001
From: tjg at exceptionalminds.com (Timothy Grant)
Date: Sun, 25 Mar 2001 17:01:40 -0800
Subject: So what's wrong with __future__?  (Was Re: Why "from __future__" stinks ...)
In-Reply-To: <LNBBLJKPBEHFEDALKOLCAEDKJIAA.tim.one@home.com>; from tim.one@home.com on Sun, Mar 25, 2001 at 04:10:47AM -0500
References: <slrn9bdka2.3j6.hzhu@rocket.knowledgetrack.com> <LNBBLJKPBEHFEDALKOLCAEDKJIAA.tim.one@home.com>
Message-ID: <20010325170140.A29222@trufflehunter.avalongroup.net>

On Sun, Mar 25, 2001 at 04:10:47AM -0500, Tim Peters wrote:
> [Huaiyu Zhu]
> > I don't quite see what the complaint about __future__ is really about.
> > So let me try to analyze a little bit.  Maybe this could encourage
> > someone to write a PEP to clarify.
> >
> > What's wrong with import __future__?
> > ...
> 
> Oh, I don't think it's hard to grasp the primary complaint:  it's a hack!
> Nobody in their right mind would *expect* an import statement to have
> profound effects on the syntax and/or semantics of the module containing the
> import, and it's un-Pythonically surprising to abuse import statements this
> way.

I'm glad that it appears that even the Python gods are
implementing this hideous travesty under protest. It is truly
the ugliest hack ever--far worse than lambda, which to some
extent could have been expected.
> 
> OTOH, it was the best hack we could dream up under the constraints that (a)
> we had to dream up *something*; (b) it had to be 100% backward compatible, no
> exceptions no way no how; and, (c) since it was late in the release cycle,
> any form of new syntax was a non-starter (there was no conceivable way to
> make time to adjust all the tools we ship, let alone pull that trick  with
> such scant warning on other people supplying Python tools).

OK, here's where I show my ignorance. Why did it have to be
100% backward compatible? Things in this industry get changed
every day that are not 100% backward compatible. I tend to
think that this was one of Microsoft's biggest problems with
their OS. They insisted on backward compatability. I often
wonder how much long-term pain (crappy OS stability) could have
been avoided if MS had convinced their users to accept a bit of
short term pain when Win95 was introduced.

Heck, I'm still on 1.5.2, but only because I'm lazy. If my code
breaks when 2.1 is the only available version I'll stop being
so lazy and install it. I realize that there are people with
installed bases, it would be a bigger problem, but not bigger
than making a new rpm or deb or tar.gz available.

> 
> It will probably go away by the time 2.2 ships.  In the meantime, it should
> do the job it was designed to do, for the three highly vocal people who are
> afraid of nested_scopes damage they won't actually suffer <0.9 wink>.

I certainly hope so.

-- 
Stand Fast,
    tjg.

Timothy Grant                         tjg at exceptionalminds.com
Chief Technology Officer              www.exceptionalminds.com
HyperLINq Technologies, Inc.          <><       (503) 246-3630
>>>>>>>>>>>>>Linux, because rebooting is *NOT* normal<<<<<<<<<
>>>>This machine was last rebooted:  68 days  4:56 hours ago<<



From mcfletch at home.com  Fri Mar 16 16:10:04 2001
From: mcfletch at home.com (Mike C. Fletcher)
Date: Fri, 16 Mar 2001 16:10:04 -0500
Subject: Functionalism, aesthetics Was:(RE: I come to praise .join, not to bury it...)
Message-ID: <000101c0ae5d$7968b270$a9807018@cr706570a>

And here I was going to avoid getting into this conversation again...

Recap (for those who have not been following along):
	Objections raised to the current definition of a "joiner" and the resulting
placement of the join method within the string type (as opposed to making it
a method on all sequence objects). I won't address this particular issue, it
has been sufficiently (excessively) explored for a decided issue. Suffice
that people like methods on objects, people like the idea of healing the
type/class split, one particular method might require a lookup in the
documentation for some people.

	Objections raised to the syntax of ".".join( something )
	Point raised that the . glyph is used primarily because of a correspondence
between the semantic meaning of "joined, element of" and the physical
connection afforded by the . character.  [Unchallenged]
	Point raised (and seconded if I recall correctly) that the physical joining
of the object (scanned as a single element) and method/attribute name is
reduced and potentially obviated when the object being accessed is expressed
as a short string literal (as in the case with ".".join). [Unchallenged]
	Objections raised to raising objections about the syntax of ".".join
without rigorous proof of functional degradation (aesthetic considerations
declared invalid for the purposes of discussion).  Request made to all those
unwilling to subscribe to solely functionalist criteria to just stop
bothering these decent people ;o) .
	Alternatives proposed:
		"." . join( )
		"."..join( ) &c
none of these was considered particularly satisfactory (obviation of the
physical connection of the glyphs, unfamiliarity of the user, need for
special casing), though the first item, by virtue of being usable today, is
less odious than the other options. [Unchallenged]
	Topic largely beaten to death, given that the language has already moved
beyond the question.  [It is a wart, we will deal with it]


The place of aesthetics...

Some Definitions:
	Functionalism -- movements within design theory exploring the use of
functional requirements as a generative and limiting mechanism.  "Nothing
can be added or taken away without detriment to the whole."  Includes many
"baroque" structures, where functional requirements lead to extremely
complex (and aesthetically pleasing) structures generated from simple first
principles.
	Minimalism -- movements within design theory exploring the simplification
and abstraction of designs. Often eliminates and/or minimizes expression of
functional structures in order to achieve a particular aesthetic
sensibility. The International Style (Corbusier &c) is an example of
minimalism.
	Eclecticism -- movement within design theory at the close of the
19th-century focusing on the "proper application of style", particularly
focusing on the social and semantic interpretation of stylistic elements
within a context.  Due to the rhetoric of the turn of the 20th-century
"functionalist" movement, generally seen as the antithesis of functionalism.
	Semantic Analysis -- Postmodern architects' reinterpretation of eclecticism
to fit within the orthodoxy of modern architecture. Reintroduces social and
semantic interpretation of particular structures into the "requirements" for
a structure.  Recognizes the fulfilment of deeper needs (social, spiritual,
emotional, philosophical) as part of the program for any given structure.
Tends to produce very simplistic and surface-oriented fulfilment of these
needs (they are still learning).
	Japan -- country with advanced and sophisticated aesthetic understanding.
Expresses itself (among other ways) in a juxtaposition of sensual and
minimalist design elements.  Strong tradition of "discovered" and
"wonder-full" elements introduced to provide pleasure and experiential
stimulation for the user.  A very sophisticated "aesthetic shorthand".


	The reason I felt it necessary to jump in here again is the implicit
creation of a "inner circle" of the "initiated" who are considered "worthy"
to contribute to a discussion about the evolution of the language.  The
users of a system, whether it is a building, a language, a piece of
software, or a chair, need to be heard.  A user complaining that something
is "ugly", "bad" or "wrong" needs the nature of their discomfort
ascertained, and that knowledge integrated into further design decisions.
	Designers who ignore their users, assuming that their own understanding of
the system is correct, and that anyone who is unable to speaking the dialect
of designers cannot possibly contribute to a design, will create
monstrosities which are next to impossible to inhabit.  Functionalism only
works when the understanding of "function" is broadened to include the
entire process of inhabitation, not just the parti-sketch with which the
architect is fascinated.  The modernist movement proved this time and again
with awe-inspiring buildings which failed to be good places to live.  The
postmodernists continue to prove it with each building they inflict upon the
world.

Your friendly neighbourhood designer signing off...
Mike

PS...
	Design is about solutions.  The problem is aesthetic.  Aesthetic solutions
are possible.  For instance, build an editor which allows for kerning or
alternate glyphs between the " and . glyphs such that the connection is
physically/graphically re-established between the two items.


-----Original Message-----
From: Alex Martelli [mailto:aleaxit at yahoo.com]
Sent: Friday, March 16, 2001 10:10 AM
To: python-list at python.org
Subject: Re: I come to praise .join, not to bury it...


"Carel Fellinger" <cfelling at iae.nl> wrote in message
news:98rmod$c8p$1 at animus.fel.iae.nl...
    [snip]
> > some people (just as it is going to attract others), so,
> > it's not a clear "win" either way, even if you just frame
> > it as a popularity-race.  Focus on *substance*, on the
>
> but if you get too blinded by this substance thing, you might easily
> overlook that in the end you have to weight things and there again the
> seasoned sense of aesthetics of the gifted ones is what makes the
> difference.

I appreciate "form follows function" as an aesthetic principle.

Clearly, given such masterpieces of functional spareness as
Bauhaus and other artistic schools of this century have been
able to produce, I'm not alone in this (and the spare functional
beauty of Medieval Francescan churches, etc, stands to prove
that this is not by any means an idea new to this century; also,
of course, it's NOT by any means an idea limited to Western
culture, as for example a certain Zen-inspired style of Japanese
furniture shows).

Just as clearly, given such masterpieces of rich ornamentation
as great Baroque architects (etc, etc -- again, this is not
limited to one century, or one continent!) have produced, it
is anything but a _universally accepted_ guiding principle for
the fashioning of beauty in human artefacts.

You seem to imply some moderation (have to weight things) is
key to YOUR aesthetics -- fine, Aristotle will agree, as will
no doubt others, but much breathtaking beauty is instead in
works which pushed their guiding principles to absolute
extremes (Bach's "Art of the Fugue", Picasso's "Guernica",
Olbrich's "Sezession Haus", Leopardi's "Infinito" -- where
is any 'moderation' or 'weighting' in these and many other
masterpieces across all arts and centuries and continents?!).

In my own profession, I make certain choices which I perceive
as guided by 'aesthetics' when I cannot afford to articulate
in full detail the path back from the practical example up to
the guiding-principle; my experience (and 'gift', if any) may
help this synthesis (a productivity plus) by helping me
perceive as "elegant" an architecture which will prove helpful
to the design-task at hand.  If and when challenged, though, I
can and will fall back to the time-consuming, perhaps boring,
but often enlightening task of detailed analysis of the
constituent elements -- there is nothing 'inherently ineffable'
in this.  I see architects of manufacts with more psychological
interactions, such as GUI's or buildings, perform similarly --
a sense of beauty to guide their everyday work (as reasoning
back to first principles would take forever and slow things
down unacceptably), but solid, explainable reasons they can
trace back if and when needed... *IF* they're worth a dime as
building-architects and/or GUI-architects.  And for the only
'art per se' I _do_ know a little bit about, cinema, I see a
very similar pattern again -- yet farther removed, OK, but a
great movie can be fully analyzed (and IS, over and over again,
in all sorts of debates and critical publications).

That doesn't mean it's possible to make somebody _without_
appropriate gifts into a great movie director, GUI designer,
building architect, OR software architect, of course -- but,
it DOES mean each of these professionals need not fall back
on vague generalities and empty hand-waving if and when their
design choices get challenged.

Somebody who's unable to do anything but talk of aesthetics
when discussing a choice in the architecture of a software
component is simply not contributing to the discussion.  It
looks good to you, it looks bad to him, it looks indifferent
to her, yawn, great debate, now let's get beyond this to the
REAL meat of the issue, please -- the technical implications
of this, that, or the other choice in a specific matter.

If you _can't_ get down to concrete specifics, then either
you're not investing enough time & energy to contribute
(in which case, staying out of a debate WOULD be a nice
matter of courtesy), or you lack some specific (experience,
culture, professional background, whatever) to be able to
analyze and express your surface impressions in the issue --
in which case, it's quite possible that they're misfounded,
and, IF the huge investment of time and energy doesn't
faze you, in-depth analysis to see if they ARE might well
be warranted (and that is basically what I've been doing
throughout this thread -- striving to help others articulate
their objections, in great detail, by first laying out in
just as much details my reason for NOT objecting at all to
the architecture Guido has chosen for .join; it won't of
course have any direct repercussion on how Python behaves
in the matter, but it MAY have positive implications on
technical growth of the people involved, and thus, very
indirectly, on future efforts on their part).


> > other hand, is going to be a win for all who _are_ involved
> > with Python -- not just those who happen to be on the
> > "I like it" side vs the "I like it not" one (and I've
> > never seen any argument about _aesthetics_ [as opposed to
>
> I hope you'll see the argument in the above:)
>
> > _usability_ -- human-factors studies DO have a substantial
> > side too:-)] that doesn't boil down to such trifles]).

Nope: it _still_ boils down to aesthetics being AT BEST
a mental and verbal shorthand to avoid the considerable
investment required to articulate the concrete underlying
realities in detail.  AT WORST (and, I suspect, more often,
particularly in Usenet debates) a way for somebody to keep
spewing words on a subject on which they actually have
nothing concrete, constructive, and useful to say.


Alex



--
http://mail.python.org/mailman/listinfo/python-list




From Laura_01 at MailAndNews.com  Thu Mar 15 18:08:05 2001
From: Laura_01 at MailAndNews.com (Laura Lewin)
Date: Thu, 15 Mar 2001 18:08:05 -0500
Subject: Python Programming on Win32 (was: Magnitude of the wx* market (was: Python In A Nutshell - suggestions))
Message-ID: <3AB2828E@MailAndNews.com>

OK, glad you like the Python Cookbook, and please do contribute!  About the 
need for "more specific topics on Python Development"...we're working on it.
Thanks for the added feedback on the Win32 book.
Laura
LLewin at oreilly.com

>===== Original Message From "Don Tuttle" <tuttledon at hotmail.com> =====
>"Laura Lewin" <Laura_01 at MailAndNews.com> wrote in message
>news:3AB16A79 at MailAndNews.com...
><snip>
>> To try to address your other concerns (your mini rant), we are planning
>> application-oriented books in the Python space, but these will be for
>> developers.  I hear your concerns about books that cover Python apps, but
>> ORA's main focus has always been for the developer, and that's where
>you'll
>> see most of our books.  I can't speak for all of the other
>publishers/authors
>> out there.
>
>Hmm... I guess I didn't make myself clear.  I'm not really interested in
>Python apps.  I'm interested in more specific topics on Python Development.
>In short, don't make me learn VB, Perl, or C++ just so I can THEN figure out
>how to do it in Python.  (This is a general statement to all publishers, not
>just O'Reilly.)
>
>The void in how to program in Python has been filled by several good books.
>The basic language is no longer an issue.  How to do usefull tasks is. This
>is where I have to go to current books written for other languages (And yes,
>I just got my copy yesterday Programming Python 2E. ;-)
>
>In Python Programming On Win32's case, it suffers from too much breadth, not
>enough depth. It would be much better to chop the cookbook sections up into
>a seperate series of books.  As for .NET, if the Common Language Runtime
>becomes cross platform, you won't have to call this a Win32 series anymore!
>
>> Please do check out The Python Cookbook
>> (http://pythoncookbook.activestate.com/)though.  You mention an interest
>in
>> system administration, databases, network programming.  These are all
>> sections > in the Cookbook.  Feel free to contribute recipes, or take any
>with you!!
>
>It's about time! <wink> I was complaining about a need for a Python Cookbook
>here several months ago!
>
>Don
>



From dbenson334 at earthlink.net  Tue Mar 13 19:26:18 2001
From: dbenson334 at earthlink.net ({USER_FIRSTNAME} {USER_LASTNAME})
Date: Wed, 14 Mar 2001 00:26:18 GMT
Subject: smtplib
Message-ID: <KUyr6.8614$6p5.760693@newsread1.prod.itd.earthlink.net>

I am trying to write a simple smtp mail form for a website.  It seems that
the code that i wrote is working fine but I keep getting the following error
from the Python Shell:

SMTPRecipientsRefused: {'dbenson at nothingspecific.com': (550, '5.1.1
<dbenson at nothingspecific.com>... User unknown')}

When I try and use the same code as an external object in Zope, I get a
"host not found" error.

Any suggestions??







From gustav at morpheus.demon.co.uk  Sun Mar 11 18:56:01 2001
From: gustav at morpheus.demon.co.uk (Paul Moore)
Date: Sun, 11 Mar 2001 23:56:01 +0000
Subject: loading Python20.dll during run-time under Windows
References: <slrn9alouc.sor.zeitlin@seth.lpthe.jussieu.fr>
Message-ID: <5v3oatsgb21fu9ds1ci4dsuk87apbb1arm@4ax.com>

On 11 Mar 2001 02:32:45 GMT, zeitlin at seth.lpthe.jussieu.fr (Vadim
Zeitlin) wrote:

> Hello,
>
> I have an application which can use Python for scripting but it is not
>essential to its work, i.e. it works mostly fine without Python. So I'd
>like to be able to install it even on systems lacking Python installation
>but for this I need to [try to] load Python.dll during run-time instead
>of linking to it directly.
>
> Has anyone done this before and is there any (C) code I could reuse?

Look at the source code for Vim 6 (from ftp.vim.org, in
/pub/vim/unreleased/{unix,extra}). The latest versions have dynamic
loading of Python - it's in if_python.c. Basically, it's simple, but
tedious - load the DLL, and then get the addresses of all the
functions you want, casting to the appropriate signatures.

Paul


From sebastien.libert at comexis.com  Wed Mar 14 03:09:52 2001
From: sebastien.libert at comexis.com (Sébastien Libert)
Date: Wed, 14 Mar 2001 09:09:52 +0100
Subject: file - codecs - unicode ??? 
References: <3aace9da$0$382$446d705c@news.skynet.be>
Message-ID: <3aaf25bd$0$612$5a62924d@news.skynet.be>

Hello,

Big thaks for your help !!!!!!!!!

Now i have found a  sulotion that work nice ( for me ! ).
If this solution can help - use it !

import codecs

def ConvertUTF16ToLATIN1(inputFile,outputFile)
    """
        ConvertUTF16ToLATIN1 : convert UTF-16 file in LATIN-1 file

        Parameters :
            inputFile : file in UTF-16 format
            outputFile : file in LATIN-1 format generated from the inputFile

    """

    f=codecs.open(inputFile,'rb','utf-16')
    g=open(outputFile,'w')

    ffile=f.read()
    ffile.encode('latin-1')
    g.write(ffile)

    f.close
    g.close


"S?bastien Libert" <sebastien.libert at comexis.com> wrote in message
news:3aace9da$0$382$446d705c at news.skynet.be...
> Hi,
>
> here is a sample of what i do :
>
> >>> h=open("c:\\myfile.log")
> >>> line=h.readline()
> >>> line
>
'\377\3760\000.\0000\0000\000\011\000i\000n\000f\000o\000\011\000L\000o\000g
>
\000_\000S\000t\000a\000n\000d\000a\000r\000d\000\011\000n\000g\000L\000o\00
> 0g\000\015\000\012'
>
> What can i do with this kind of thing ?????
> I want to convert thid in ascii.... how can i do thid in python ?
>
> I use python 2.0 on win32 !!!
>
> thx to answerd ;-)
>
>




From vze2f978 at mail.verizon.net  Sat Mar 10 11:05:38 2001
From: vze2f978 at mail.verizon.net (Lloyd Hugh Allen)
Date: Sat, 10 Mar 2001 16:05:38 GMT
Subject: help (nonEnglish)
References: <3aa3a6cd$0$16795@reader2>
Message-ID: <3AAA5123.922C33EE@mail.verizon.net>

patrick wrote:
> 
> I need a dutch tutorial for python 2.1
> Please send it to ppvisser at hotmail.com
> or just give me an download link or a good site

There is currently no dutch info on the page
<http://www.python.org/doc/NonEnglish.html>, but if there's something
that you like better than English, it may be there.

If you /do/ find something in Dutch, make sure that that page finds
out...


From carribeiro at yahoo.com  Thu Mar 22 07:16:10 2001
From: carribeiro at yahoo.com (Carlos Ribeiro)
Date: Thu, 22 Mar 2001 09:16:10 -0300
Subject: New PEP: The directive statement
In-Reply-To: <j4hf0mywbn.fsf@informatik.hu-berlin.de>
References: <mailman.985085006.24236.python-list@python.org>
 <3AB82078.8060005@caltech.edu.is.my.e-mail.address>
Message-ID: <5.0.2.1.0.20010322090509.021c6ec0@mail.inet.com.br>

At 10:41 22/03/01 +0100, Martin von Loewis wrote:
> > My second comment is that you may want to consider adopting a form of the
> > __future__ proposal's runtime semantics for directive statements.  Perhaps
> >
> >     directive transitional.nested_scopes
> >
> > could create a "_Feature object" called nested_scopes identical to the one
> > proposed in "back to the __future__".
> >
> >     print nested_scopes.getOptionalRelease()   # prints "(2, 1, 0, 
> 'beta', 1)"
>
>I have my problems with that part of PEP 236 also. While it is
>certainly possible to provide the release when a feature was first
>released, it is not possible to give an accurate answer to the
>question when it will become mandatory. The intent to make it
>mandatory in the future should go into the documentation, IMO.
>
>Furthermore, what is the use of that information to a program?

That's indeed a good question. For a program it's not so good as runtime 
information. It *could* be potentially useful for the interpreter prompt 
though, as a form of 'introspection' - but is nothing that could not be 
done by means of documentation. One different way to do it it would be 
through a dictionary called __directive__, returning a data structure 
similar to the one proposed by the 'import __future__' PEP.

It *could* be useful also in another circumstance - what a directive is 
supposed to do in a multi-module program? Will we need to set a directive 
globally (for every module)? I'm not sure of the semantics in this case. 
The transitional.nested_scopes does not seem to pose many problems for this 
particular case, but I'm yet to see more development in this area.




_________________________________________________________
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com




From fellowsd at cs.man.ac.uk  Mon Mar 12 11:20:46 2001
From: fellowsd at cs.man.ac.uk (Donal K. Fellows)
Date: Mon, 12 Mar 2001 16:20:46 +0000
Subject: Einstein's Riddle
References: <oNPq6.396644$ge4.138479566@news2.rdc2.tx.home.com> <slrn9apj4g.c54.Andreas.Leitgeb@pc7499.gud.siemens.at>
Message-ID: <3AACF75E.F53E913@cs.man.ac.uk>

Andreas Leitgeb wrote:
>   -) because Tk cannot yet write vertical texts,
>       it is not possible to put label for the columns, which
>       makes it tough to remember  what each column represents.
>       Actually, I haven't even realized the horizontal row-labels, yet :-)

Stick to all-caps and then just place one (unrotated) character above the
next, and everything will come out looking OK.  Simple?  Yes.  Easy?  Yes.
Exactly the effect you're after?  No, but Good Enough For The Lazy.  :^)

Donal.
-- 
Donal K. Fellows    http://www.cs.man.ac.uk/~fellowsd/    fellowsd at cs.man.ac.uk
-- There are worse futures that burning in hell. Imagine aeons filled with
   rewriting of your apps as WinN**X API will change through eternity...
                                           -- Alexander Nosenko <nae at titul.ru>


From paul.moore at uk.origin-it.com  Wed Mar 14 11:10:05 2001
From: paul.moore at uk.origin-it.com (Paul Moore)
Date: Wed, 14 Mar 2001 17:10:05 +0100
Subject: [Python-Dev] Revive the types sig?
References: <jeremy@alum.mit.edu> <15020.9404.557943.164934@w221.z064000254.bwi-md.dsl.cnc.net> <200103120711.AAA09711@localhost.localdomain> <mailman.984439040.24419.python-list@python.org> <s44tatolisnipdfm4r4p1aqt2e09e458hk@4ax.com> <mailman.984581571.10100.python-list@python.org>
Message-ID: <1pWvOlsMJ9Zxk0d4s9BhFmMCSusy@4ax.com>

On Wed, 14 Mar 2001 15:55:38 +0100, Paul Prescod <paulp at ActiveState.com>
wrote:
>Paul Moore wrote:
>> My view on all this is that type annotations are likely to be counterproductive,
>> in the face of Python's high degree of polymorphism. For example, what is the
>> type of
>> 
>> def fac(n):
>>     if n <= 1: return 1
>>     return n * fac(n-1)
>> 
>> ?
>> 
>> Obvious answer, integer -> integer. And that's likely what the programmer would
>> say.
>> 
>> But why can't this function be used with *any* type which supports comparison
>> with the number 1, multiplication, and subtraction of 1?
>
>Sure it can. But as a programmer, I probably haven't really thought
>through the meaning of this code for users who pass in strings or lists
>of SocketObjects. Therefore I should not be blamed when someone passes
>in a random type and gets strange behavior.

Sure. But once again, what "type" would you give this function? Would you
restrict it to integer? Would you allow long integers? What about gmpy.mpz
(another long integer type)?

If you gave me a library with this function in, and it didn't work with
gmpy.mpz values, _simply because you enforced that restriction via a type_, I
would be (justifiably, IMHO) cross.

>Anyhow, this kind of totally polymorphic code is exceedingly rare.

I wasn't suggesting "total" polymorphism (although I did overstate my case a
bit). What I was pointing out was essentially that it is possible to define
totally "plug-compatible" replacements for builtin types, and have them work
everywhere the built-in type does. With the exception of code that does "if
type(x) != type(1)".

Type annotations, as described so far, sound like a sanitised and formalised
version of this:

    def f(n):
        assert type(n) == type(1)

I can imagine *very* few places where an assert like this is of genuine
practical use.

Of course, if type annotations will *actually* formalise and support
interfaces in some form, that's entirely different - I'm mostly neutral, about
something like this.

    def f(n):
        assert <n supports the "integer" interface>

>Type annotations of any type are a known trade-off between raw
>flexibility, safety and error reporting. C++ and ML are way on one end
>of the spectrum. Python is way on the other end. There is a middle
>ground.

My concern is that over-use of assertions like this limit "creative use" of
functions, in contexts which the author may never have envisioned.

The best example to date in Python is "file-like" objects. There are *lots* of
library functions which take such objects. Not just files, but also
user-defined objects which act like files. And many such functions explicitly
state that they only require a subset of the full file interface. (Often not
requiring seekability, sometimes requiring nothing more than readline).

How would you handle this? (Remembering that files are a built-in type, so
saying "files are outside the scope" seems unreasonable).

Paul.



From daniel.orlowski at bg-sys.de  Wed Mar 21 08:38:34 2001
From: daniel.orlowski at bg-sys.de (Daniel Orlowski)
Date: Wed, 21 Mar 2001 14:38:34 +0100
Subject: [Newbie] line separators in file-output
Message-ID: <3AB8AEDA.2D2BD105@bg-sys.de>

If I in Python uder Windows do smth like this:

>>> File=open("test","w")
>>> File.write("a line"+"\n")
>>> File.close()

I have data in file test:
'a line\015\012'

How can I prevent output of  '\015'?

Thanks in advance
Daniel




From paul.moore at uk.origin-it.com  Mon Mar  5 08:48:28 2001
From: paul.moore at uk.origin-it.com (Paul Moore)
Date: Mon, 5 Mar 2001 14:48:28 +0100 
Subject: Name of the Script
References: <20010304.185354.2114937732.8277@localhost.localdomain> <NJCo6.6720$qv3.2834377@nnrp5-w.sbc.net>
Message-ID: <FpmjOjvwe5bm0Bg09LDTMPhh5vji@4ax.com>

On Mon, 5 Mar 2001 03:21:29 +0100 , "Alex Shindich"
<alex at shindich.com> wrote:

>try the following:
>
>import sys
>print sys.argv[0]
>....
>The first argument in the argument list is the name of the script that got
>invoked.

This doesn't always include full pathname information. On Windows,

--- a.py ---
import sys
print sys.argv[0]
---

python a.py

just prints "a.py". Is there a way of getting the *full* script name.
For example, so that the script can locate a data file which is held
in the same directory as the script itself.

(sys.argv[0] is probably fine in 99.9% of cases, but there are subtle
borderline cases which I can imagine could cause problems).

Paul.



From corour01 at motorola.com  Fri Mar 23 06:15:47 2001
From: corour01 at motorola.com (O'Rourke Clodagh-corour01)
Date: Fri, 23 Mar 2001 11:15:47 -0000
Subject: Tkinter - Widget Colours
Message-ID: <13EE655665F4D311AB4D0008C789498A01EE2C9E@zei02exm02.cork.cig.mot.com>

Sorry all - spellcheckers!! 
should have been bg(background)/fg(foreground)

-----Original Message-----
From: O'Rourke Clodagh-corour01
[mailto:Clodagh_O'Rourke-COROUR01 at email.mot.com]
Sent: 23 March 2001 11:10
To: 'python-list at python.org'
Subject: Tkinter - Widget Colours


Hi all,

I was wondering does anyone know how the bag/fag colours can be changed after the widget is packed.  

Thanks,
Clodagh

-- 
http://mail.python.org/mailman/listinfo/python-list



From tuttledon at hotmail.com  Wed Mar 21 22:59:23 2001
From: tuttledon at hotmail.com (Don Tuttle)
Date: Thu, 22 Mar 2001 03:59:23 GMT
Subject: learning python...
References: <3AB8F4E1.184DD9C@noaa.gov> <Z97u6.16733$Im6.1761313@newsread1.prod.itd.earthlink.net> <3ab90559_6@goliath.newsfeeds.com> <99b44s$m5j$1@panix6.panix.com> <3AB9182A.8FEBCF3E@noaa.gov> <N9bu6.66$Hp.3347@typhoon.aracnet.com>
Message-ID: <vMeu6.144753$__6.26762056@typhoon.southeast.rr.com>

"Daniel Klein"
> Pay no attention to Amazon reviews. You are ALWAYS going to get people who
> love it and those who hate and everything inbetween. If you want to get
> information, this is the best place to get it.

BTW, I complained to Amazon a week ago about all the old first edition
reviews being included in the 2nd edition's listing.  They have now created
a different page for the 2nd edition.  So it's a clean slate.  Why not post
a review yourself?  I intend to as soon as I've finished reading it.  (Which
might be awhile given my penchant for reading 5 or more books at once.  ;-)

Don




From bwalker at earth1.net  Sat Mar 24 05:00:43 2001
From: bwalker at earth1.net (Bill Walker)
Date: Sat, 24 Mar 2001 05:00:43 -0500
Subject: How 2 get rid of console
Message-ID: <tborq16qctlm42@corp.supernews.com>

I'm using Tkinter to build a GUI application. How do I get rid of the
distracting dos console under the Tk window?




From aleaxit at yahoo.com  Fri Mar  9 09:39:19 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 9 Mar 2001 15:39:19 +0100
Subject: I come to praise .join, not to bury it...
References: <mailman.983485923.1136.python-list@python.org> <97p7iq$kk2$1@nntp6.u.washington.edu> <97rka002k9@news1.newsguy.com> <3AA2F4B3.A7496E76@cosc.canterbury.ac.nz> <97vmoh061p@news2.newsguy.com> <3AA46134.D385208B@cosc.canterbury.ac.nz> <98292h0s0g@news1.newsguy.com> <3AA58BD0.3324C709@cosc.canterbury.ac.nz> <9850ge013ns@news1.newsguy.com> <gRup6.364652$ge4.128679630@news2.rdc2.tx.home.com>
Message-ID: <98apuj020km@news2.newsguy.com>

"Rainer Deyke" <root at rainerdeyke.com> wrote in message
news:gRup6.364652$ge4.128679630 at news2.rdc2.tx.home.com...
> "Alex Martelli" <aleaxit at yahoo.com> wrote in message
> news:9850ge013ns at news1.newsguy.com...
> > in fact, just about any test for 'IS object X of exact type
> > Y' you can find in the Python sources, unless they are
> > shortcut-like 'accelerators' for a special case _before_
> > the more-general request 'please o mr X try to behave like
> > an Y, are you able to?', can be seen as implementation
> > defects, in my personal opinion.
>
> I'm curious: how do you feel about the automagic conversions of functions
> (and only true function, not function-like objects) into unbound methods,
> and from unbound methods into bound methods?  I'm thinking that the former
> could be eliminated entirely, and the latter replaced by a __bind__ magic
> method that allows the mechanism to apply to function-like objects.

I'll admit I have not digged deep enough into this part of
the Python internal mechanisms to have any deep feelings
either for or against.  I find the _results_ pretty good,
although not perfect.  If the special-casing could in fact
be eliminated, and further polymorphism gained at no great
cost, through the modifications you're thinking of, then I
think they'd be swell.


Alex





From cribeiro at mail.inet.com.br  Sat Mar 31 17:09:53 2001
From: cribeiro at mail.inet.com.br (Carlos Alberto Reis Ribeiro)
Date: Sat, 31 Mar 2001 19:09:53 -0300
Subject: list vs tuple
In-Reply-To: <iAqx6.1051$p5.4277@news1.rivrw1.nsw.optushome.com.au>
References: <MVhx6.1021$p5.4157@news1.rivrw1.nsw.optushome.com.au>
 <hGlw6.782$p5.3210@news1.rivrw1.nsw.optushome.com.au>
 <3AC20CD3.265A2D94@alcyone.com>
 <ru69cto61jtmrqsf809e38hb3pbm2lhi7g@4ax.com>
 <bC1x6.4319$4N4.905946@newsc.telia.net>
 <71q9ct4le63anefmiajj5cbh3mg04qu1ga@4ax.com>
 <mailman.985991202.5578.python-list@python.org>
 <7Bfx6.989$p5.3791@news1.rivrw1.nsw.optushome.com.au>
 <Ukhx6.4421$4N4.950603@newsc.telia.net>
 <mailman.986035544.5572.python-list@python.org>
Message-ID: <5.0.2.1.0.20010331184327.022231e0@mail.inet.com.br>

Ok, I'll try again. Let me try to get the problem from your point of view. 
Please, have some patience (it's always good to ask first...)

The problem that you are pointing is not with the assignment. It's about 
types, and about the difference between a pointer and the other types. 
Putting it in Pascal terms,

var a, b: ^integer;
begin
   new(a);
   /* copying the reference */
   b := a;
   /* copying the value */
   b^ := a^;
end.

As you can see, what changes is not the assignment, but the hints that you 
give to the compiler, by explicitly dereferencing the pointer when you want 
to copy the value, and not the reference.

Now let's see Python. For the sake of simplicity, I'm going to use 
analogies from Pascal, and I think I'm going to annoy the purists <wink>. 
In Python, objects are accessed always through reference. So when you do 
the assignment, you are copying the reference (or "name-binding", which 
conceptually is about the same - the difference is that we don't talk about 
explicit memory addresses).

What is missing, in your opinion, (if I got you right) is a de-reference 
operator in objects to do the copy, as for in this example:

a = [1,2,3]
b = a       # copy the reference
b^ = a^     # copy the value

Why this is not needed? There are a lot of reasons, and the experts in the 
theoretical aspects of language design may give you a full course on this. 
I can tell you about my personal experience and knowledge. The need to 
remember to reference/dereference variables is error prone. It makes code 
bigger and harder to read for no good reason. I can say to you that it's 
fairly easy to get used to Python, and it is quite effective.

In Delphi, classes are also always treated as references, and it works very 
similar to Python. Look at this example:

var p, q: TStringList;
begin
   p := TStringList.Create;
   p.Add('hello');  // in 'pure' Pascal, it should be p^.Add
   q := p;          // both p and q point to the same list
   p.Add('world');  // both lists are updated (in fact are the same)
end.

There is no need to dereference p in the "p.Add" call. But in fact, p is a 
pointer to the object, and the semantics of the assignment are just that of 
a pointer assignment. Now this is not consistent, but still is very useful, 
and somewhat easy to live with.

Also note that in Python, the constant [1,2,3] is not simply a value, but 
also a full fledged object, as it can be proved by the following snippet:

 >>> dir([1,2,3])
['append', 'count', 'extend', 'index', 'insert', 'pop', 'remove', 
'reverse', 'sort']

So when you write "[1,2,3]", you are in fact instantiating a new object.


So my dear deadmeat, I could go on this for hours on end; but I feel that 
we are going way too far trying to explain it. It's up to you now.


Carlos Ribeiro





From john at cobra.nmt.edu  Fri Mar 23 00:31:46 2001
From: john at cobra.nmt.edu (John Shipman)
Date: Fri, 23 Mar 2001 05:31:46 +0000 (UTC)
Subject: New Tkinter and PIL documentation (free)
Message-ID: <99en42$2vl6$1@newshost.nmt.edu>

At this site you will find my newly written reference document
for Tkinter:

    http://www.nmt.edu/tcc/help/pubs/lang.html

This document is available in PostScript or PDF format and is 84
pp. long.  It is not utterly comprehensive but I have tried to
put in everything that people use in the majority of Tkinter
applications, without relying on external documents such as the
Tcl/Tk manual pages or the X programming manual.

This document includes quite a bit of information missing from
the reference section of John Grayson's ``Python and Tkinter
Programming'' book, including:

  -  The effect of every option to every widget is described
     separately without reference to a common options section.

  -  A discussion of .grid_propagate().

  -  How to make a scrollable Entry widget.

  -  An expanded table of keysyms, keycodes, and keysym_num
     values for the U.S. 101-key keyboard.

  -  More complete treatment of the X options database and
     its relation to widget pathnames.

  -  The Menu.add() method and its options.

  -  Text index modifiers.

  -  Text.image_create() and its options.

  -  Text.tag_configure() options.

  -  Text.window_create() options.

  -  The shared options used by objects on canvases, such
     as the width option for .create_line().

I would very much appreciate any suggestions for improving this
material.

-- 
John Shipman (john at nmt.edu), Applications Specialist, NM Tech Computer Center,
Speare 128, Socorro, NM 87801, (505) 835-5950, http://www.nmt.edu/~john
  ``Let's go outside and commiserate with nature.''  --Dave Farber


From scarblac at pino.selwerd.nl  Thu Mar  8 09:52:28 2001
From: scarblac at pino.selwerd.nl (Remco Gerlich)
Date: 8 Mar 2001 14:52:28 GMT
Subject: Who am I: can a class instance determine its own name?
References: <mailman.984034150.9635.python-list@python.org>
Message-ID: <slrn9af71n.d6p.scarblac@pino.selwerd.nl>

Tim CHURCHES <TCHUR at doh.health.nsw.gov.au> wrote in comp.lang.python:
> This is probably an elementary question and the answer is probably writ large in multiple places in the Python documentation, but...
> 
> ...can an instance of a class determine the name of the variable to which it is assigned? For example:
> 
> ###########################
> class Foo:
>     def whoami(self):
>          return "You are a Foo() but I do not know your name"
> 
> FooBar = Foo()
> 
> print FooBar.whoami()
> ###########################
> 
> 
> How does one define the method whoami() so that it returns "FooBar"? This
> sort of navel gazing is formally called introspection, I think (therefore I
> am)?

An instance may have no name, one name, or many names.

print Foo().whoami()  # No name

FooBar = [Foo()]*3 # Name? FooBar[0], FooBar[1] and FooBar[2]...

a = b = c = Foo() # a, b and c

So it's not possible. What do you need it for?

-- 
Remco Gerlich


From shaleh at valinux.com  Tue Mar 20 12:54:06 2001
From: shaleh at valinux.com (Sean 'Shaleh' Perry)
Date: Tue, 20 Mar 2001 09:54:06 -0800 (PST)
Subject: Regular expressions and non-standard character set
In-Reply-To: <ka8y9u01htt.fsf@cheerful.cs.hut.fi>
Message-ID: <XFMail.20010320095406.shaleh@valinux.com>

On 20-Mar-2001 Petri Mikael Kuittinen wrote:
> I want to match word boundaries using the special sequences \b and \B
> of regular expressions. They work OK when using the "standard"
> alphanumeric set [a-zA-Z0-9_]. But I would like them to work with
> character set which also contains various "national characters"
> e.g. ?, ?, ?, ?, ?, ?, ? etc. and their uppercase equivalents.

I know this is not exactly what you want to here, but I personally avoid \b
like the plague.  It has been my experience that what the regex engine thinks
is a word boundary does not match what I want to be a word boundary.  I am
trying to remember examples of this, but I know that it has bitten me a few
times. 



From ws-news at gmx.at  Tue Mar 13 07:21:50 2001
From: ws-news at gmx.at (Werner Schiendl)
Date: Tue, 13 Mar 2001 13:21:50 +0100
Subject: This math scares me
References: <3aad4062.1290021@News.CIS.DFN.DE> <20010312170356.B3513@better.net> <mailman.984480856.18097.python-list@python.org>
Message-ID: <984486395.407379@newsmaster-04.atnet.at>

I do not see an advantage of using decimal floating point arithmetic.
As you mentioned, it slows things down for no obvius gain.

The most useful approach is to round the numbers to 15 digits for display,
because this is the precision double values can really correctly represent.
The drawback is, that the rounded value does not discover all the
information actually stored in the variable.

Tim Peters post explains, why 17 digit representation is used in Python: To
be able to get the representation and evaluate it back to the internal
information without loss of information.

regards
werner

Ken Seehof <kens at sightreader.com> wrote in message
news:mailman.984480856.18097.python-list at python.org...
> You're not alone, but there is a good explanation for what you see, and it
> is not generally considered a bug in python, though some would like to
> change python to give more visually correct results.
>
> This is also true of many other programming languages, such as C and C++.
> The reason is that the computer stores your number in binary.  Turns out
> that 10.55 (or 0.55) does not have an exact representation in binary,
> therefore you get a "rounding error".  To give you an idea of what I am
> talking about, consider that in decimal, you can't represent 1/3 exactly
> (you get 0.33333333...).
>
> Let's take a simpler example:
>
> >>> 0.2
> 0.20000000000000001
>
> What happened here?  Well 0.2 = 1/5, which works fine in decimal.  When
you
> convert that to binary, it looks like this:
> 0.2 (decimal)  =
0.0011001100110011001100110011001100110011001100110011...
> (binary).
>
> The reason your desk calculator gets the right result is that it
represents
> numbers in decimal internally.  If python did the same, it would run
> significantly slower than it does because most computers prefer to do math
> in binary.  However, switching to decimal floating point is a proposed
> change for a future version of python, along with making 1/2 = 0.5 (or
some
> other equivalent answer (right now, 1 / 2 = 0 (try it) ) ) and other
stuff.
>
> Keep in mind that any floating point system will have rounding errors.
> Doing math in decimal merely hides the wierd unexpected ones like the one
> you found (at the cost of slowing things down).  Some people propose that
> python keeps fractional numbers in rational form (i.e. A/B).  This would
> completely solve the rounding error problem (for division and
> multiplication, that is), but would make python really really really slow.
> There are various other compromises but there is unfortunately no good
> solution to this dilemma at all.
>
> These numerical issues cause endless debates among python people.  As you
> will see, there will probably be a stream of endless innane babble in
> response to this message.  I think there's a SIG or a PEP or something
> (couldn't find it) where this debate is supposed to be instead of here.
>
> ----- Original Message -----
> From: "William Park" <parkw at better.net>
> To: <costas at springmail.com>
> Cc: <python-list at python.org>
> Sent: Monday, March 12, 2001 2:03 PM
> Subject: Re: This math scares me
>
>
> > On Mon, Mar 12, 2001 at 09:34:58PM +0000, costas at springmail.com wrote:
> > > Ok, I can see maybe division having problems. But why does addition of
> > > the two numbers below:
> > >
> > > 5.01+5.54
> > >
> > > give me this?
> > >
> > > 10.550000000000001
> >
> > It's called 'Round-off error'.
> >
> > ---William Park, Open Geometry Consulting, Linux/Python, 8 CPUs.
> >
> > --
> > http://mail.python.org/mailman/listinfo/python-list
> >
>
>




From stephen_purcell at yahoo.com  Sun Mar 18 04:56:13 2001
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Sun, 18 Mar 2001 10:56:13 +0100
Subject: standalone
In-Reply-To: <Nu_s6.244274$v.26337040@typhoon.nyroc.rr.com>; from vdanovich@empireone.net on Sun, Mar 18, 2001 at 08:39:09AM +0000
References: <Nu_s6.244274$v.26337040@typhoon.nyroc.rr.com>
Message-ID: <20010318105613.B7546@freedom.puma-ag.com>

Vyacheslav Danovich wrote:
> Hi!
> I'm pretty new to Python.
> I need to run an executable Python script in Windows environment which does
> not have Python library installed.
> My questions are: can it be done and if "yes" what steps should I take

Yes.

Take a look at Thomas Heller's Py2Exe:

  http://starship.python.net/crew/theller/py2exe/

-Steve

-- 
Steve Purcell, Pythangelist
Get testing at http://pyunit.sourceforge.net/
Any opinions expressed herein are my own and not necessarily those of Yahoo



From root at mailhost.stg.rt.daimlerchrysler.com  Tue Mar 13 04:12:42 2001
From: root at mailhost.stg.rt.daimlerchrysler.com (root)
Date: Tue, 13 Mar 2001 10:12:42 +0100
Subject: Subject with SMTPLIB?
Message-ID: <3AADE48A.5D8AA43@daimlerchrysler.com>

Hello!

I need to write some mail for my fax-log evaluating script. I'm
currently using
the smtplib to simply send some mails. But how can I get (without
studying RFCs
and papers on SMTP for years)  a SUBJECT for my mails? It must be
this easy, but the library reference in the python documentation package
tells
me nothing about it.  I was disappointed that the documentation on the
smtplib
module didn't tell me that.

Thanks in advance for any help,

Alex

--
Alexander Stirzel
alexander.stirzel at daimlerchrysler.com /
alexander.stirzel at informatik.uni-ulm.de
*** welcome to hell, here's your copy of windows ***


From mysticsguy at yahoo.ca  Thu Mar 22 21:24:37 2001
From: mysticsguy at yahoo.ca (Yvon Boulianne)
Date: Fri, 23 Mar 2001 02:24:37 GMT
Subject: python, perl, lisp, tcl / got a decision to make, can you give me a view point please ? and a joke for python users :)
References: <3aba3076.14395287@news.vif.com>
Message-ID: <3abab2ba.14983504@news.vif.com>

just to let you know, i pass the day on python, doing the intro
tutorial and i'm making fun.  it's the first language i use from the
old day basic on my radio shack color computer that i  find really
good.  this language is structured, the syntax is clear, i love the
way to work with tabs and not {} and i'm only a newbies on it ;-)
so maybe it's too simple to be mind expanding but maybe with a great
language like that i will be able to find new things to do with a
puter, we'll see :))

anyway this is not comparable to tcl or perl of vb, that's easy and
it's ... mmm  what can i say, I'm in Love ;-)

Take care.
With Love & Counsciousness
Yvon



Yvon Boulianne ; Saphi
mysticsguy at yahoo.ca
www.monunivers.net
Humour,Paradox & Changements


From tim.one at home.com  Sat Mar 24 01:04:16 2001
From: tim.one at home.com (Tim Peters)
Date: Sat, 24 Mar 2001 01:04:16 -0500
Subject: New PEP: The directive statement
In-Reply-To: <j4k85gkiye.fsf@informatik.hu-berlin.de>
Message-ID: <LNBBLJKPBEHFEDALKOLCIEACJIAA.tim.one@home.com>

[Martin von Loewis]
> ...
> That concerns me about PEP 236: The absolute certainty in which people
> predict the future. My experience is that the future is unpredictable
> when it comes to software releases. It may turn out that there are
> unforeseen problems with nested scopes, or that foreseen problems turn
> out to be more serious than expected.

The entire point of the PEP is to give incompatible language changes a trial
run of at least one full release cycle before saying "that's it".  In what
respect does that come off as absolute certainty to you?  And in what respect
does your PEP differ in relation to this point?  If you have a better idea,
change your PEP to spell it out.

> ...
> Tim claims that I invented the statement to declare file encodings,
> e.g.
>
> directive encoding "koi8-r"

Gimme a break:  At least that differs from the

    directive source_encoding "latin-1"

you put in your patch comment <wink>:

    http://sourceforge.net/tracker/index.php?func=detail&
        aid=404997&group_id=5470&atid=305470

That would be a fine use for "directive"!  I don't grasp what you're after in
repeating this "Tim claims" business:  getting a way to specify file
encodings is an obvious agenda of yours, and there's nothing wrong with it.






From glevenson at mediaone.net  Tue Mar 13 18:35:05 2001
From: glevenson at mediaone.net (Gary)
Date: Tue, 13 Mar 2001 23:35:05 GMT
Subject: deleting IMAP messages using imaplib
References: <LUwr6.30$YP3.125512@typhoon.ne.mediaone.net> <98m9af$ivm$0@216.39.151.169>
Message-ID: <J8yr6.41$YP3.163942@typhoon.ne.mediaone.net>

Thanks!

"Donn Cave" <donn at oz.net> wrote in message
news:98m9af$ivm$0 at 216.39.151.169...
> Quoth "Gary" <garyl at directhit.com>:
> | Can anyone tell me how to delete messages in an IMAP mailbox? The doc
says
> | delete is used to delete a mailbox, while expunge is used to clear out
> | deleted items. How do I delete a message itself?
>
> It's only a flag, to identify the message that as you noted
> will be actually deleted later by an expunge.
>
>  ok, error = imapobject.store(number, 'FLAGS', '(\Deleted)')
>
> Donn Cave, donn at oz.net




From fredrik at pythonware.com  Fri Mar 16 11:35:06 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 16 Mar 2001 16:35:06 GMT
Subject: OO-Browser
References: <mailman.984739039.5214.python-list@python.org>
Message-ID: <_grs6.171$sk3.49961@newsb.telia.net>

Eugene Leitl wrote:
> I'm setting up a Python/C development environment, so I decided to go with
> XEmacs/gdb/ddd/pdb and OO-Browser. However, I'm only running into dead
> links when attempting to access BeOpen.
>
> What's up with them?

they're dead.

also see:
http://hem.passagen.se/eff/2000_10_01_bot-archive.htm#1201078

Cheers /F




From ctavares at develop.com  Sat Mar 31 16:17:13 2001
From: ctavares at develop.com (Chris Tavares)
Date: Sat, 31 Mar 2001 21:17:13 GMT
Subject: list vs tuple
References: <hGlw6.782$p5.3210@news1.rivrw1.nsw.optushome.com.au><3AC20CD3.265A2D94@alcyone.com><ru69cto61jtmrqsf809e38hb3pbm2lhi7g@4ax.com><bC1x6.4319$4N4.905946@newsc.telia.net><71q9ct4le63anefmiajj5cbh3mg04qu1ga@4ax.com><mailman.985991202.5578.python-list@python.org><7Bfx6.989$p5.3791@news1.rivrw1.nsw.optushome.com.au><Ukhx6.4421$4N4.950603@newsc.telia.net> <mailman.986049281.32365.python-list@python.org> <uGqx6.1052$p5.3997@news1.rivrw1.nsw.optushome.com.au>
Message-ID: <tPrx6.18597$Uw.1388730@newsread1.prod.itd.earthlink.net>

"deadmeat" <root@[127.0.0.1]> wrote in message
news:uGqx6.1052$p5.3997 at news1.rivrw1.nsw.optushome.com.au...
> > - There is an difference between "copy" and "deep copy". What happens if
> > you make copy a structure that contains pointers in it?
>
> then you got a problem :)
>
> but thats not the point - which when the statement a = b copies the data
> across for one type it's expected to do it for others.  It doesn't, so
it's
> inconsistant, contrary to the original claim.
>

Here's the root of your problem. a = b does NOT, I repeat, does NOT do any
data copying!!!!!

When you do this:

a = 5

You've now got a name, a, that points to an integer object out on the heap.
Similarly, when you do:

b = 9

b now points to an integer object out on the heap. And then, when you do:

a = b

a and b now point to the same integer object 9, and the integer 5 gets
garbage collected at some unknown point in the future.

So, when you do this with lists:

a = [ 1, 2, 3 ]  <--- we'll call this list1. It lives on the heap.

b = a             <- b now also points to list1.

b[1] = 5        <- This is a method call on the thing b points to, which of
course is the same thing as what a points to!

>>> a
[ 5, 2, 3]

Did this help?

-Chris





From julian at zeus.com  Mon Mar  5 06:12:39 2001
From: julian at zeus.com (Julian Midgley)
Date: Mon, 05 Mar 2001 11:12:39 GMT
Subject: Test Your Web Apps - article at LinuxQuality
References: <3AA34C6B.EC12B25B@goingware.com>
Message-ID: <HwKo6.10392$mt.1713196@news2-win.server.ntlworld.com>

In article <3AA34C6B.EC12B25B at goingware.com>,
Michael D. Crawford <crawford at goingware.com> wrote:
>Tonight I posted:
>
>Use Validators and Load Generators to Test Your Web Applications
>http://linuxquality.sunsite.dk/articles/webapptesting/
>
>The article promotes the idea that one should check the HTML
>generated by a web application with a validator to ensure that
>it conforms to W3C standards, and that you should use a load
>generator to stress-test your server during development to
>detect performance problems early.
>
>Links to some load generators and validators are given, with
>tips on using them.

For performance testing of websites, I would recommend httperf and
ApacheBench above most of the tools you've listed.  ApacheBench is
distributed with Apache (look for 'ab.c'), and is a useful tool for
simple benchmarking.  

ApacheBench is particular nice because it is itself extremely
efficient.  It uses a single non-threaded process and non-blocking
IO to simulate concurrency, instead of the multithreaded or
multi-process implementations of most of the alternatives.  It's one
of the few tools which will usually bottleneck after the server, rather
than before it.

httperf allows more sophisticated testing, including the generation of
simulated 'real-world' workloads. It's available from:
http://www.hpl.hp.com/personal/David_Mosberger/httperf.html

Julian Midgley
-- 
Julian Midgley
Principal Consultant                    jmidgley at zeus.com
Zeus Technology Ltd                     http://www.zeus.com


From Mike_B at T-Online.de  Mon Mar 26 12:02:27 2001
From: Mike_B at T-Online.de (Michael Bauer)
Date: Mon, 26 Mar 2001 19:02:27 +0200
Subject: daemon app in win32
References: <99nk49$a2q$02$1@news.t-online.com> <mailman.985619314.10299.python-list@python.org>
Message-ID: <99nsn3$q39$06$1@news.t-online.com>

Steve Purcell wrote:

> Michael Bauer wrote:
>> i am writing a daemon application that runs both under windows and unix.
>> Since i am not very experienced in win32 - programming, i really would
>> appreciate some hints how to register my app in win32 that i can do some
>> cleanup-functions when windows is shut down.
>> 
>> At the moment it is just a standard console application that simply gets
>> killed when windows shuts down. The app does not have/need a GUI, it
>> should just run in the background.
> 
> 
> If you're using NT, take a look at the win32service API. I think it
> provides hooks like SvcStop that Windows will call for you (if you ask it
> nicely).

Hmm, seems like this is only available under NT. Is there a more general 
way to accomplish this task independently of the underlying Windows-version?

But maybe i should just get a life and write different wrappers for Win95, 
Win98, WinME, Win200, WinNT... :-(
 
still hoping... Mike


From ws-news at gmx.at  Tue Mar  6 11:18:41 2001
From: ws-news at gmx.at (Werner Schiendl)
Date: Tue, 6 Mar 2001 17:18:41 +0100
Subject: absolute namespace for callbacks?
References: <3AA50AFB.83C7459F@troikanetworks.com>
Message-ID: <983895797.296667@newsmaster-04.atnet.at>

What's the matter?

You do not pass the name, but a reference to the function
(the function is an object for python like anything else)
It does not matter where the function is defined...

or did I miss anything in your post?

hth
werner

Bruce Edge <bedge at troikanetworks.com> wrote in message
news:3AA50AFB.83C7459F at troikanetworks.com...
> Tcl has a func to get a fully qualified namespace from a context. This
> is essential for registering callback functions.
> How does one register a callback in Python, when the callback func
> exists in a different namespace?
>
> Is there any method to get the current absolute namespace?
>
> Thanks, Bruce.
>




From fuess at att.net  Sun Mar  4 13:19:20 2001
From: fuess at att.net (David Fuess)
Date: Sun, 04 Mar 2001 18:19:20 GMT
Subject: CGI & Python (error)
References: <3A9F5FF5.34E01148@usa.net> <pkNn6.4257$c7.1344814@news3.rdc1.on.home.com> <3AA1A066.D4F67A16@cs.utwente.nl>
Message-ID: <0m15atgoagbroulvhjk615ec6bk8mic40n@4ax.com>

Debugging CGI can be a trying task. You might want to enclose the
entire CGI in a try-except structure so you can trap all errors and
return a reasonable result to the server and write an error log for
yourself.

Dave

On Sun, 04 Mar 2001 02:54:46 +0100, Richard van de Stadt
<stadt at cs.utwente.nl> wrote:

>Colin Meeks wrote:
>> 
>> > When accessing  the  script I get the error:
>> >
>> > File "getdate2.py", line 12, in ?
>> >     day=form["day"].value
>> >   File "/usr/lib/python1.5/cgi.py", line 907, in __getitem__
>> >     raise KeyError, key
>> > KeyError: day
>> >
>> > The script works from the command line.
>> >
>> > My problems is:  the script I am trying to run is based on another one.
>> > This other script works. I can't  figure out the what is causing the
>> problem
>> > as the 2 scripts are identical  in every aspect execpt for 1 variable.
>> Both
>> > scripts are run in a page which has all the required parameters passed to
>> it.
>> > Help. Please...
>> If this is what I think it is (sorry it's not immediately clear), but when I
>> try
>> to parse anything from CGI, if nothing is entered in one of the boxes, this
>> seems to be especially for radio buttons, then the variable isn't
>> initialised.
>> 
>> The thing I always do is something like
>> 
>> try:
>>     day=form["day"].value
>> except:
>>     day=1 #or what ever you want the day variable initialised as
>> 
>> I hope this is of some help.
>
>You can also use this:
>
>if form.has_key("day") and form["day"].value) <> "":
>	has_day_action(form["day"].value)
>else:
>	doesnot_have_day_action()
>
>Note that neither of these options work in case the webform field
>is a checkbox list. In such cases you shouldn't use ".value" directly,
>but initially only the field name, like this:
>
>if aForm.has_key("checkbox_list") and aForm["checkbox_list"] <> "":
>	has_checkbox_list_action ()
>
>To get the values of the selected items of a checkbox list (after you
>tested if the fields exist like state above), do this:
>
>checkbox_items =  aForm["checkbox_list"]
>if type(checkbox_items) <> type ([]):
>	checkbox_items = [checkbox_items]
>
>After this, you will have the item(s) of the checkbox list in an array.
>To get the values of the items, walk through the array, and *then* use
>".value", e.g.:
>
>checkbox_item_values = []
>for item in checkbox_items:
>	checkbox_item_values.append(item.value)
>
>Cheers,
>
>Richard.



From aahz at panix.com  Sat Mar  3 07:40:19 2001
From: aahz at panix.com (Aahz Maruch)
Date: 3 Mar 2001 04:40:19 -0800
Subject: cataloging words in text file
References: <3AA0124E.74A39783@comm.mot.com> <3AA08D8F.255748A0@bigfoot.de>
Message-ID: <97qonj$gni$1@panix2.panix.com>

In article <3AA08D8F.255748A0 at bigfoot.de>,
Gerhard =?iso-8859-1?Q?H=E4ring?=  <gerhard.nospam at bigfoot.de> wrote:
>Stephen Boulet wrote:
>> 
>> I remember this homework assignment for my data structures (c++)
>> class: read in a large file, and create a data structure containing
>> every word in the file and the number of times it appears.
>
>Funny. This was the task I had to code in Java for a recent job
>application.

Don't you wish you could have coded it in Python?  ;-)
-- 
                      --- Aahz (Copyright 2001 by aahz at pobox.com)

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

Nostalgia just ain't what it used to be


From bsass at freenet.edmonton.ab.ca  Sat Mar 31 16:42:18 2001
From: bsass at freenet.edmonton.ab.ca (Bruce Sass)
Date: Sat, 31 Mar 2001 14:42:18 -0700 (MST)
Subject: list vs tuple
In-Reply-To: <4Lrx6.1055$p5.4170@news1.rivrw1.nsw.optushome.com.au>
References: <hGlw6.782$p5.3210@news1.rivrw1.nsw.optushome.com.au><3AC20CD3.265A2D94@alcyone.com><ru69cto61jtmrqsf809e38hb3pbm2lhi7g@4ax.com><bC1x6.4319$4N4.905946@newsc.telia.net><71q9ct4le63anefmiajj5cbh3mg04qu1ga@4ax.com><mailman.985991202.5578.python-list@python.org><7Bfx6.989$p5.3791@news1.rivrw1.nsw.optushome.com.au><Ukhx6.4421$4N4.950603@newsc.telia.net><mailman.986049281.32365.python-list@python.org><uGqx6.1052$p5.3997@news1.rivrw1.nsw.optushome.com.au>
 <mailman.986072259.10548.python-list@python.org>
 <4Lrx6.1055$p5.4170@news1.rivrw1.nsw.optushome.com.au>
Message-ID: <Pine.LNX.4.33.0103311431230.7534-100000@bms>

On Sat, 31 Mar 2001, deadmeat wrote:

> > But that is not what is happening!
>
> What is actually happening is not my point, the end result that the
> programmer sees is, and that is not consistant with what is typed.

Sure it is, once you realize that the "1" in "a = 1" is not an
integer with the value of 1, it is the object whose name is "1".


- Bruce




From claird at starbase.neosoft.com  Thu Mar 15 10:37:01 2001
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 15 Mar 2001 09:37:01 -0600
Subject: Cameron Laird Quote on www.unixinsider.com -- Probably old news to 
 y'all, but new to me.
References: <bJOr6.9481$TW.38047@tor-nn1.netcom.ca> <3AB00593.D1C4BBBF@fxtech.com> <3AB08001.9532A3F0@letterror.com> <mailman.984661520.22665.python-list@python.org>
Message-ID: <F33F155144667EE2.FBBB63AA76E7CAA5.B077320C6CF70D40@lp.airnews.net>

In article <mailman.984661520.22665.python-list at python.org>,
Paul Prescod  <paulp at ActiveState.com> wrote:
>Just van Rossum wrote:
>> 
>>...
>> 
>> Yeah, I was going to say: there's an awfully large application area _between_
>> scripting and interactive 3d graphics apps! Python already does quite well
>> with interactive 2d graphics, though: see eg. Sketch.
>
>There were video game guys at the conference who say they use Python on
>the client side for the AI. On the other hand, even the most basic
>script depends heavily on parts of the interpreter and the OS written in
>C. So the vast majority of apps will profit from a mix of languages. But
>each year the proportion will shift from towards Python and away from C.
>Embedding Python will become less interesting than embedding (and
>inlining!) C, just as C inlines assembly.
			.
			.
			.
Right.

Since my name is attached to this thread (through a
mistaken identification), I'll make a personal remark.
This is my position:
1.  If someone walked up to me on the street,
    and said, "What computing language should
    I learn?", I'd answer, "Without knowing
    anything about you and your true needs,
    the only safe thing I can say is that Py-
    thon is the one language applicable in
    the widest span of applications, and for
    the broadest spectrum of authors."
2.  It's my personal mission to help organi-
    zations feel safe using multiple
    languages--Python+C, C+assembler,
    Fortran+Tcl, Ruby+SQL+XML, ...
-- 

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


From db3l at fitlinxx.com  Thu Mar  1 20:08:41 2001
From: db3l at fitlinxx.com (David Bolen)
Date: 01 Mar 2001 20:08:41 -0500
Subject: Performance when working directory is slow network drive
References: <983411179.898268@clam-ext> <mailman.983425921.14373.python-list@python.org> <983480535.813990@clam-ext>
Message-ID: <uk869m0yu.fsf@ctwd0143.fitlinxx.com>

"Mark Hadfield" <m.hadfield at niwa.cri.nz> writes:

> 1. Why having the current directory in the search path affects the import
> time for the os module much more than it does for other modules I tested
> (like string). Is this because os is a package?

The os module isn't a package, but it does reference other modules.  I
can think of two possibilities for the performance difference:

* The os module itself imports various other things including the
system specific path module so there's just more going on and thus
more access to the current directory for import lookups.

* The os module also along the way imports the string module under
Windows (from within ntpath.py), so by the time your import of os
finishes, you already have the string module loaded, and thus if you
import os before you do string, the string import really doesn't do
much.

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


From hungjunglu at yahoo.com  Sun Mar 11 18:08:35 2001
From: hungjunglu at yahoo.com (hungjunglu at yahoo.com)
Date: Sun, 11 Mar 2001 23:08:35 -0000
Subject: Chinese text GIF file generation
Message-ID: <98h0hj+be0d@eGroups.com>

As I said, it's not hard in Java, and matter of fact it's done. 
Please visit http://www.ocrat.com/reaj/gif.html.

I am still looking for C or Python.

Hung Jung







From fjs6 at usa.net  Thu Mar 22 12:14:03 2001
From: fjs6 at usa.net (fjs)
Date: Thu, 22 Mar 2001 09:14:03 -0800
Subject: Good Editor?
References: <3ab9a32b$1_1@news2.one.net> <Xns906C5F868E6B7PaCmAnRDLM@161.48.76.20>
Message-ID: <3aba32d7$1_1@news2.one.net>

Sorry folks,

I use both windows and linux.  So, all the recomendations have been helpful.

Thanks


"Fred Pacquier" <fredp at mygale.org.nospam> wrote in message
news:Xns906C5F868E6B7PaCmAnRDLM at 161.48.76.20...
> "fjs" <fjs6 at usa.net> said :
>
> >Can someone recomend a good editor to write python in?  Im just getting
> >started learning python and dont know which ones support python.
>
> You don't mention the platform. Under Unix/Linux there are a number of
> options (including Vim, Emacs, etc.). Under Windows, I found WScite
> (http://www.scintilla.org) the easiest to get started with...
>
> --
> YAFAP : http://www.multimania.com/fredp/




From constant.dupuis at swing.be  Thu Mar 15 14:48:50 2001
From: constant.dupuis at swing.be (Furax)
Date: Thu, 15 Mar 2001 20:48:50 +0100
Subject: Dynamic append in a dictionnary object ...
Message-ID: <3ab11de3$0$389$446d705c@news.skynet.be>

Hi,
I'm pretty new to Python.
I cannot figure out, reading the documentation, how I can dynamicaly add a
new pair in a dictionnary ?

??

Thank's

Constant







From martin.skott at propylon.com  Fri Mar 16 14:03:15 2001
From: martin.skott at propylon.com (Martin Skott)
Date: 16 Mar 2001 19:03:15 +0000
Subject: Cookies
References: <mailman.984707839.22839.python-list@python.org> <lmq6m07j.fsf@propylon.com> <98t5f60lb7@news1.newsguy.com>
Message-ID: <hf0tmt8s.fsf@propylon.com>

"Alex Martelli" <aleaxit at yahoo.com> writes:

> "Martin Skott" <martin.skott at propylon.com> wrote in message
> news:lmq6m07j.fsf at propylon.com...
>     [snip]
> > You should first of all be aware that there is no common unique
> > identifier for a computer available from the net today[1]. IP
>     [snip]
> > [1]Why would somebody use cookies if there was?
> 
> Because cookies are more powerful -- they let you associate
> state with, e.g., several different _users_ on a single
> computer, rather than 'that computer as a whole'.  (That
> was a trick question, right?-)

Depends, if you by trick mean standing on my head with plates balanced
on my feet while typing the text I will have to disappoint you :-)
And I must admit that I never really have used cookies in that way
even though the documentation from Netscape includes some nice
examples in that direction. The times I have used cookies information
about state has been stored in other places.

-- 
Martin Skott
Software Developer
Propylon - Enabling Universal Mobility
http://www.propylon.com
Tel: +353 096 37245
Mobile: +353 087 9680370


From jsoons at juilliard.edu  Sat Mar 31 11:57:37 2001
From: jsoons at juilliard.edu (Jonathan Soons)
Date: Sat, 31 Mar 2001 11:57:37 -0500
Subject: Where are the examples/samples?
Message-ID: <B7C74E1A6FBAD411BFC60002B309DDF4E445@mailbox.juilliard.edu>

I recommend my method:
I bought a couple of books, read up on the basics,(DO NOT WASTE TIME BY
READING A WHOLE BOOK AND FORGETTING IT) then I embarked on a project
involving file opening, reading, text file processing, sending output to
new file and printer. I ran into many problems that the book solved by my
reading ahead, or the newsgroup solved.
In summary: Get yourself a project (HTML related??) and generate some
examples. Everything else will fall into place.

I now live in moderate seclusion, consider getting a pet, and occasionally
take a partial mental inventory of my obsolete computing hardware.




-----Original Message-----
From: lmhamric at aol.com
To: python-list at python.org
Sent: 3/30/01 2:13 AM
Subject: Where are the examples/samples?

After looking at every Python book in the local bookstore and at every
Python-related site I could find on the Internet I have to admit that
I'm
perplexed. I've seen a number of posts that have asked for examples of
various
things and when I follow the links - where are the examples? Lots of
theory,
lots computer-science indulgence, but no examples. Most recently tried
to get
started on using BaseHTMLProcessor.py (or HTMLLIB or SGMLIB or whatever)
and an
example called dialect.py was given to illustrate its use. The latter
translates things into Pig Latin, among other things! I can't tell what
part of
it involves HTML and what part is just fooling around with something
else. Does
anyone know of examples anywhere that illustrate usage at a bare bones
level
where you can simply see how the interfaces work? I suspect that
learning
Python would go a lot better if one could skip the theoretical stuff and
simply
get on wiht working examples. You can come back and pick up the theory
later.
Anyhow, if anyone else sees this as a problem and has found an answer
then I'd
appreciate hearing what it is.  
-- 
http://mail.python.org/mailman/listinfo/python-list



From justin at iago.org  Tue Mar  6 23:29:50 2001
From: justin at iago.org (Justin Sheehy)
Date: 06 Mar 2001 23:29:50 -0500
Subject: sorting on IP addresses
In-Reply-To: <3A7F26CB.BE85C96F@esec.com.au>
References: <3A7F26CB.BE85C96F@esec.com.au>
Message-ID: <871ysa9p6p.fsf@kitchen.iago.org>

Sam Wun <swun at esec.com.au> writes:

> Does anyone know what is the quickly way to sort a list of IP addresses?
> 
> ie. 203.21.254.89 should be larger than 203.21.254.9

Convert to longs and sort on those values?

Not all that quick, I suppose, but I don't know of a faster correct way.

I suppose you could do it one octet at a time, which might be faster
for cases like your example where the addresses are fairly close.

The actual code for either of those approaches is short and
straightforward.  Just write a little function to compare two
addresses using the method of your choice, and pass it to your list's
sort method.

-Justin

 



From tim.one at home.com  Fri Mar  2 20:22:46 2001
From: tim.one at home.com (Tim Peters)
Date: Fri, 2 Mar 2001 20:22:46 -0500
Subject: the demise of 'from foo import * and its implications?
In-Reply-To: <3kVBAfATrDo6EwEM@jessikat.fsnet.co.uk>
Message-ID: <LNBBLJKPBEHFEDALKOLCIEJGJDAA.tim.one@home.com>

[Robin Becker]
> it's an ambitious act to suppose that the new complicated
> __twilight_zone__ rules are better than the old simpler rules because of
> some optimisation.

The official rule was that import* at other than module scope has undefined
behavior.  Still is.  The complication comes from spelling out the accidents
of the 2.1 implementation that intersect with the accidents of the 2.0
implementation, for the benefit of whatever users may have been violating the
rules without realizing it.

Would you prefer that we let you dope that out for yourself?  I'd be
delighted to stick to the official rule, not catering at *all* to code that
ignored the Ref Man.  Then we could replace the entire explanation with a
one-line quote from the manual:  "you cheated; you got caught; tough luck".

couldn't-be-simpler-than-that!-ly y'rs  - tim




From dwig at advancedmp.net  Mon Mar 26 13:40:18 2001
From: dwig at advancedmp.net (Don Dwiggins)
Date: 26 Mar 2001 10:40:18 -0800
Subject: WinNT problem with Python and Cygwin Bash
In-Reply-To: Courageous's message of "Sun, 25 Mar 2001 03:40:16 GMT"
Message-ID: <ud7b45q7h.fsf_-_@advancedmp.net>

I've just installed Happydoc (nice!) on my WinNT box, and am having a bit of
a problem running the shell script generated by the installer under Cygwin
using Bash.  It looks as though Bash is passing Python a Unix-style pathname
(/e/Python20/Scripts/happydoc), but Python expects a Windows-style pathname,
so complains it can't find the file.  I can work around it by applying some
"Bash grease", but I'd like to know if anyone else has run across this, and
if there's a simple fix.

-- 
Don Dwiggins                    "Solvitur Ambulando"
Advanced MP Technology
dwig at advancedmp.net




From jamie at cctech.co.uk  Fri Mar 16 05:18:39 2001
From: jamie at cctech.co.uk (Jamie MacIsaac)
Date: Fri, 16 Mar 2001 10:18:39 -0000
Subject: Basic information requested
References: <MHks6.144$lC5.52172@news2-win.server.ntlworld.com> <98snjh$hsu$1@norfair.nerim.net>
Message-ID: <1Mls6.207$lC5.83390@news2-win.server.ntlworld.com>

Hi,

Thanks for the suggestion. I've been looking for that title in the bookshops
around here (Glasgow, Scotland), but I can't find it. I'm somewhat loathe to
order it online without actually knowing that it will cover what I'm after -
I've done that too often.

Do you have 'Python Programming on Win32'? Can you (or anyone else) verify
that it deals with mapping and disconnecting remote shares?

Thanks,

jMi.


"Alexandre Fayolle" <alf at leo.logilab.fr> wrote in message whatever:

<snip>

>You may want to purchase a copy of 'Python Programming on Win32' (O'Reilly,
> by Mark Hammond and Andy Robinson, ISBN 1-56592-621-8).
> Chapter 16 covers this point. You can download the code samples from
> http://starship.python.net/crew/mhammond/ppw32/
>
> Cheers
>
> Alexandre Fayolle
> --
> http://www.logilab.com
> Narval is the first software agent available as free software (GPL).
> LOGILAB, Paris (France).




From neelk at alum.mit.edu  Thu Mar 15 20:11:07 2001
From: neelk at alum.mit.edu (Neelakantan Krishnaswami)
Date: 16 Mar 2001 01:11:07 GMT
Subject: Who's minister of propaganda this week?
References: <mailman.984595938.11127.python-list@python.org> <98otuq0nnf@news1.newsguy.com> <slrn9b0cc2.gsa.neelk@alum.mit.edu> <98q6ro2uoa@news2.newsguy.com>
Message-ID: <slrn9b2u0r.j74.neelk@alum.mit.edu>

On Thu, 15 Mar 2001 11:26:52 +0100, Alex Martelli <aleaxit at yahoo.com> wrote:
>"Neelakantan Krishnaswami" <neelk at alum.mit.edu> wrote in message
>news:slrn9b0cc2.gsa.neelk at alum.mit.edu...
>>
>> What's a dynamic visitor? Is it a dictionary keyed on class? That's
>> what I usually do when emulating multimethods in Python.
>
> Sorry, the pattern-name is _Acyclic_ Visitor -- it's fully
> described by its author, Robert Martin, in the PDF at URL
> http://www.objectmentor.com/publications/acv.pdf for example.

Thanks for the link. 

> Keying a dict on class has the unfortunate side-effect of
> not allowing polymorphism by inheritance -- objects of classes
> _derived from_ the one used as the key will not match.
>
> You can preserve this basic idea by using as key, instead of the
> class, some ad-hoc attribute which will be inherited (if it IS
> acceptable to alter the class-objects of interest by adding an
> attribute) except where specifically overridden, an approach which
> can be used as a first-path (fast-path) to be followed by an
> isinstance loop (slow but sure) if and when the key attribute is not
> obtained (e.g. for types that are not classes, or, classes that
> can't be altered).

I actually have a function that walks the class tree checking it
against the dictionary, which solves this problem.  You don't need to
add a special attribute in Python since __class__ gives you what you
need for free. 

I only added it to my toolkit recently, though, sine I tend to use
classes like algebraic data types in ML or Haskell -- as tagged
variants rather than designed for user-extension. I find subclassing
friendly designs are a lot harder to design in a way that doesn't
reveal too much internal structure.

> I don't think the applicability cases of this approach to
> hand-simulation of multiple dispatch will overlap much
> with those of visitors (of any kind) or coercion approaches.

I honestly haven't used Visitors very much, since most of the OO
languages I've used have either had high dynamism (like Python) or
built-in multimethods. C++ and Java are alien worlds to me. :)


Neel



From aleaxit at yahoo.com  Wed Mar 14 18:23:58 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 15 Mar 2001 00:23:58 +0100
Subject: reading binary data
References: <3AAFE69D.FF01FEC8@yahoo.fr>
Message-ID: <98oulu0okh@news1.newsguy.com>

"Francois Forest" <francois_forest at yahoo.fr> wrote in message
news:3AAFE69D.FF01FEC8 at yahoo.fr...
> I would like to read and write xxx.wav files,
> this assume to alternatively read characters,  integers 16 bits and
> integers 32 bits.
> I found that I can read a binary file and get the result in a string,
> Is there a way to convert a string slice into an int16, int32, or
> floatting point ?
> (what is usually done in C through a cast).

The struct module offers exactly this facility.


> More generally, how can I browse through standard libraries ?

By reading the excellent library manual that comes with
Python and/or the well-commented sources.  Maybe I'm
missing something in your question...?


Alex





From hgg9140 at cola.ca.boeing.com  Thu Mar 15 13:56:56 2001
From: hgg9140 at cola.ca.boeing.com (Harry George)
Date: Thu, 15 Mar 2001 18:56:56 GMT
Subject: building python for multiple targets
References: <3AB0F009.E27D9AD7@troikanetworks.com> <3AB0F36A.F6B30A7F@troikanetworks.com>
Message-ID: <xqxae6mg8sn.fsf@cola.ca.boeing.com>

My expreience is with moving site-packages around.  Apparently when
you move dirtrees, the .py's work and the .so's work, but the .pyc's
do not.  So you may have to delete those and re-byte-compile them in
the new location.

Perhaps someone else can explain why.

Bruce Edge <bedge at troikanetworks.com> writes:

> Just a followup,
> 
> Is it possible to build python to be relacatable?
> So the same runtime tree can be placed in /usr/local/python, or in
> /home/anyone/target/usr/local/python?
> 
> 
> Bruce Edge wrote:
> > 
> > I need to build Python for multiple targets from one host.
> > 
> > One target is the host, the other is for embedded Linux, with it's own
> > compiler, libs, include, etc, all of which are installed in say
> > /usr/local/rtlinux/...
> > 
> > The host build is simple, ./configure, make, make install.
> > 
> > The target one is ugly, currently I'd doing:
> >   ./configure --include=<rtpath> --bindir=/usr/local/rtlinux/bin
> > --sbindir=</usr/local/rtlinux/sbin ....<every other damn ./configure
> > option there is>
> > 
> > Is there a better way?

-- 
Harry George                E-mail:  harry.g.george at boeing.com
The Boeing Company          Renton:  (425) 237-6915
P. O. Box 3707  02-CA       Everett: (425) 266-3868
Seattle, WA 98124-2207      Page:    (425) 631-8803  


From ben.hutchings at roundpoint.com  Tue Mar  6 18:59:23 2001
From: ben.hutchings at roundpoint.com (Ben Hutchings)
Date: 06 Mar 2001 15:59:23 -0800
Subject: Python and Printing
References: <3AA4C3DC.95EF7249@usa.net>
Message-ID: <upufula90.fsf@roundpoint.com>

Mix <mix77 at usa.net> writes:

> I have pictures (JPEG) which I get from a database and dump in an
> arbitrary directory.
> To print a picture I do the following:
> 
>       import os
>       os.system("lpr lena.jpg")
> 
> 
> My problem is that page orientation is always"landscape" instead
> "portatrait".  What switch is there to specify orientation for "lpr"
> if any?

Is 'lpr' part of Python?  No?  Then this isn't really a Python
question.

> P.S. I'm working under linux (redhat 7.0 distribution ).

So try asking in a Linux or Unix newsgroup such as
comp.os.linux.questions or comp.os.linux.redhat.  Or even searching
those groups first.

I think you're out of luck, actually.  You probably need to set up
separate printer queues for portrait and landscape documents.

-- 
Any opinions expressed are my own and not necessarily those of Roundpoint.


From dan at eevolved.com  Sun Mar 11 16:24:58 2001
From: dan at eevolved.com (Dan Parisien)
Date: Sun, 11 Mar 2001 21:24:58 GMT
Subject: random number between -maxint and +maxint
Message-ID: <K2Sq6.360684$Pm2.5406628@news20.bellglobal.com>

I can't figure out a good way to do it, does anyone know a better solution 
than

from whrandom import whrandom
rand = whrandom()
max = 2**30 + (2**30-1) #one less than 2**31
randomnumber = rand.randrange(0, max) - rand.randrange(0, max)

thank you for your time,
Dan


From aleaxit at yahoo.com  Thu Mar  1 10:01:05 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 1 Mar 2001 16:01:05 +0100
Subject: python performance
References: <J8fn6.756$me1.378895@news2.cableinet.net>
Message-ID: <97lo7h02u1l@news2.newsguy.com>

"bruce" <davidlam at cableinet.co.uk> wrote in message
news:J8fn6.756$me1.378895 at news2.cableinet.net...
> Can  anyone comment on the speed of Python programs under Windows?

"Pretty good" comes to mind, if you're wise in leaving the
low-level, time-consuming stuff in C modules, as you say:

> I am interested in using Python as a embedded game scripting language.
> Amongst other things, I want to doing AI, but not the low level routines
> like path finding.

Some very rough tests among interpreted languages on a Win32
box suggest that Python is roughly in the middle of the pack --
generally better than VBScript or JScript, not quite as fast
as Perl, Java, or Visual Basic (but still quite comparable, in
either direction).


To take a maybe-non-typical case -- one of my personal areas
of interest is in the theoretical underpinnings of contract
bridge (cfr my article "How Shape Influences Strength", in
the Jan and Feb 2000 issues of the Bridge World magazine; also,
http://www.fenx.com/alex/ for my vague ideas about 'PyBridge').

I am currently running a double-dummy simulation of play at
suit and no-trumps with strong balanced hands facing each
other; it runs in the background in an oldish box (Pentium II
266 MHz, Win2000, 192M RAM) and the current run has so far
simulated 182*24*20*2 = 174720 plays, in 249055 elapsed
[wall-clock] seconds (it's about 1/5th of the way through --
and as soon as it ends, I'll start yet another run; I need
many, MANY million simulated hand-plays for statistical
analysis of the results to mean anything).  Given how time
consuming it is, I did go through the trouble of carefully
instrumenting and profiling the h**k out of it, as well as
architecting it with checkpoints/restarts (wouldn't want a
power outage of something to set me back to ground zero...),
etc.

Which makes it easy for me to tell how much of that time
goes into the (irreducible) demands of the double dummy
simulation engine itself (part of the GIB commercial product --
I only wrote the interface letting Python control it [actually,
the interface exposes it as a COM object, and Python controls
it via win32com, but that's a minor difference]), and how much
into everything else -- generating and culling the hands,
periodically checkpointing, collecting and emitting statistics
about what is going on (including timing measurements and
reporting), etc.  Specifically: 248798 seconds were so far
taken up by the double-dummy simulation engine itself; _all_
of the other operations, together, therefore account for
257 seconds, a bit more than 1 part in 1000 of total elapsed
time.  So, _of course_, the engine (and the COM interface
that lets Python control it) are in carefully-coded, hand
optimized C/C++ -- and _all_ of the rest is in Python.  If
the Python parts were recoded and optimized by a factor of
10, or 100, still, by Amdahl's Law, I would not notice
(literally).

>From time to time I remote-copy the checkpoint and result-
archive files to another machine so I can run some Python
scripts to analyze and summarize some statistics so far.
Typical times for these analyses are a few seconds (with
pure-Python coded scripts) for a few dozen thousands
plays, for each kind of analysis I've tried (I'm still
looking for the ideal C-coded extension to use to do it
more speedily when I'll be running the analyses on many
millions of simulated plays -- taking a minute or so per
each analysis-run in pure-Python is acceptable, but if I
could cut it down to, say, 10 seconds, I wouldn't mind:-).


I would expect similar ratios on many apparently dissimilar
tasks -- simulations and raw number-crunching done in
C, C++, or Fortran, taking huge amounts of time; all of
the directing-intelligence, etc, done in Python, taking
a fraction of the time.


Alex





From phlip_cpp at my-deja.com  Mon Mar 12 12:54:13 2001
From: phlip_cpp at my-deja.com (Phlip)
Date: Mon, 12 Mar 2001 09:54:13 -0800
Subject: PyEval_CallObject and threads
Message-ID: <H58r6.9$697.25923@news.pacbell.net>

Thonpy:

My colleague came over and asked me to write this:

"We got multiple C++ pthreads calling PyEval_CallObject. How threadsafe is 
the Python API? A C++ module using Boost BPL creates multiple threads, 
which then call back into Python. When two threads simultaneously call 
PyEval_CallObject we go boom."

I will now go examine his serialization primitives. But if anyone has any 
better ideas...

-- 
  Phlip                          phlip_cpp at my-deja.com
============== http://phlip.webjump.com ==============
  --  "In my experience, the customer doesn't know what he wants
       until you don't give it to him." --David Brady  --


From hinsen at cnrs-orleans.fr  Wed Mar  7 08:13:59 2001
From: hinsen at cnrs-orleans.fr (Konrad Hinsen)
Date: 07 Mar 2001 14:13:59 +0100
Subject: Open Source Organic Chemistry software project
References: <taaseuid59k047@corp.supernews.com>
Message-ID: <m3hf15d8mg.fsf@chinon.cnrs-orleans.fr>

"mapMonkey" <jziniti at speakeasy.org> writes:

> The current code-base is geared to Organic Chemistry molecules 
> and the IUPAC nomenclature system and it works quite well.  It

That description is a bit vague, but it sounds interesting, so I do
express my interest. However, you'd probably get better feedback on
the Computational Chemistry List. Moreover, there is more use in
proselytizing Python among computational chemists than doing the
inverse ;-)
-- 
-------------------------------------------------------------------------------
Konrad Hinsen                            | E-Mail: hinsen at cnrs-orleans.fr
Centre de Biophysique Moleculaire (CNRS) | Tel.: +33-2.38.25.56.24
Rue Charles Sadron                       | Fax:  +33-2.38.63.15.17
45071 Orleans Cedex 2                    | Deutsch/Esperanto/English/
France                                   | Nederlands/Francais
-------------------------------------------------------------------------------


From fredrik at pythonware.com  Tue Mar 13 04:21:54 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Tue, 13 Mar 2001 09:21:54 GMT
Subject: Subject with SMTPLIB?
References: <3AADE48A.5D8AA43@daimlerchrysler.com>
Message-ID: <SElr6.2$4d3.460@newsb.telia.net>

"root" wrote:
> I was disappointed that the documentation on the smtplib
> module didn't tell me that.

look for the "From:" and "To:" headers in the SMTP example,
and add a third line for "Subject:".

    http://www.python.org/doc/current/lib/SMTP-example.html

Cheers /F

<!-- (the eff-bot guide to) the standard python library:
http://www.pythonware.com/people/fredrik/librarybook.htm
-->







From sholden at holdenweb.com  Mon Mar  5 00:55:27 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Mon, 5 Mar 2001 00:55:27 -0500
Subject: Need a free lancer python programmer - to do bots
References: <mailman.983753706.16462.python-list@python.org>
Message-ID: <4XFo6.19441$1D5.730654@e420r-atl1.usenetserver.com>

"Kevin Douglas" <kcocaine at GamesandTheory.com> wrote in message
news:mailman.983753706.16462.python-list at python.org...
> crap!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
>
> shoot me..
>
>
> -kev
>
Bang. Good luck.

Steve




From aleaxit at yahoo.com  Mon Mar  5 07:45:12 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Mon, 5 Mar 2001 13:45:12 +0100
Subject: Questions from a beginner
References: <1103_983792888@breakdown>
Message-ID: <9801op0iq4@news2.newsguy.com>

"Alison Myers" <squirmy at freeuk.com> wrote in message
news:1103_983792888 at breakdown...
> Hi. I'm a beginning programmer at a fairly late age and have been looking
for a language to learn.
> I have decided against starting with C or C++ because I think they will be
too complicated and they seem to be only really needed for writing device
drivers and operating systems, which is not what I wat to do.

Very good strategic decision.  The other application areas where C and C++
shine, and justify their complications, is software that must run at the
highest speed, yet remain portable enough that assembly-language is not
warranted.

> Is python a good language to learn first? Will my knowledge help me with
other languages later?

It's a VERY good language to learn first.  It will SPOIL you for other
languages, it's so good -- but, apart from that, the concepts you gain
in Python are indeed applicable to many other languages.

> I will be writing for win98 and Linux and I know Python is available on
both of these, but would the code be transportable, especially if I write
things using tcl (this is probably a long way off yet, but I'd like to
know).

Yes, TK Python applications can easily enough be kept portable between
Win98 and Linux (I assume you mean Tk/Tkinter -- TCL is another scripting
language, albeit the one for which Tk was originally developed).  The
platform-specific functionality is not all that much, and, with a few
exceptions (time.strptime being the outstanding one!), generally well
delineated/isolated into platform-specific modules.

> I have links to online tutorials, but where would I find good source code?

The Python sources of the Python standard libraries are quite good, and
well worth studying: they will teach you all sorts of nifty tricks, good
Python style, AND the wealth of functionality already available for your
reuse in the libraries.

> Do I need to buy any books to help me and if so which do you recommend?

Whether you learn better from books, online material, or experimentation,
is something only you can say.  Have a look at
    http://www.crosswinds.net/~agauld/
and/or, you can get substantially the same material in book form, from
Addison-Wesley; you might be pretty happy with it in either form (if it
were me, I'd go for the mashed-dead-tree version, but that is just MY
fetish -- I'm a confirmed book-lover; your mileage may vary!-).

> Is there a list of programming projects anywhere or would I get these from
a textbook?

I think the latter holds.

> I probably need to brush up on my maths, so what specific topics should I
look at?

Depends on your interests -- WHAT kind of maths do you want to handle
with Python?  If Statistics, for example, Stokburger's online "Introductory
Statistics" at http://www.psychstat.smsu.edu/sbk00.htm may be a good idea --
it's oriented to Psych majors, so, it goes light on unneeded maths and
includes a brief summary of those maths that are absolutely can't-live-
without (basically, a tiny subset of algebra).


Alex





From new_name at mit.edu  Thu Mar  1 23:44:40 2001
From: new_name at mit.edu (Alex)
Date: 01 Mar 2001 23:44:40 -0500
Subject: compile() source code from a string
References: <3a9ee600.391990@News.CIS.DFN.DE> <etdd7c1c8n5.fsf@pickled-herring.mit.edu> <3a9f01f5.7549826@News.CIS.DFN.DE>
Message-ID: <etdy9uobwzr.fsf@cutter-john.mit.edu>

Apologies for my careless reading, Costas.  The problem is that you need
to replace the '=' in 'elif x = 2:' with '=='.

The-dyslexic-leading-the-blind'ly yrs. :~)
Alex

-- 
Alex Coventry (finger alex_coventry at mit.edu)


From bpurvy at packeteer.com  Fri Mar 16 22:01:50 2001
From: bpurvy at packeteer.com (Bob Purvy)
Date: Fri, 16 Mar 2001 19:01:50 -0800
Subject: Newbie question about import and dir
Message-ID: <3AB2D39E.C78C8743@packeteer.com>

OK, I got this pSOS port sort of working for Python 2.0, and I thought
it would be a good idea to run the standard Python regression tests, to
verify that things are really correct.  But I'm getting a strange result
that I was hoping someone would understand:

>>> import string
>>> import regrtest
>>> dir(regrtest)
['__builtins__', '__doc__', '__file__', '__name__', 'sys']

why not any of the methods?

If I do the same thing on Linux on Python 1.5, dir(regrtest) shows all
the methods ('main', 'random', etc.), just as I expected.   So is there
some obvious reason why I'm not seeing those here?   Or is this some
obscure porting problem I just need to hunt down?

(since we don't have a real OS, I've embedded the whole thing within our
product and called Py_Main(), basically, so that's why I'm not just
running regrtest from the command line, as you'd normally do.)



From dsh8290 at rit.edu  Fri Mar 30 11:21:56 2001
From: dsh8290 at rit.edu (D-Man)
Date: Fri, 30 Mar 2001 11:21:56 -0500
Subject: Python GUI with Swing
In-Reply-To: <001e01c0b8ea$16a14660$340210ac@winson>; from winson@mdsserv.mds.com.tw on Fri, Mar 30, 2001 at 03:21:45PM +0800
References: <001e01c0b812$43f00760$340210ac@winson> <mailman.985889320.22860.python-list@python.org> <3AC3BE89.3F95C8DC@ou.edu> <001e01c0b8ea$16a14660$340210ac@winson>
Message-ID: <20010330112156.E13498@harmony.cs.rit.edu>

On Fri, Mar 30, 2001 at 03:21:45PM +0800, ??? wrote:
| Let me explain more about it. I mean my customer need Swing, but I need
                                                   ^^^^

How much is it really a _need_ vs. a _want_ or _like_?  Perhaps you
can convince them that a different toolkit would better suit their
needs?

| Is there a module for Python called PySwing, only use Swing's look and feel
| GUI ?? 

Swing's L&F ..., hmm ..., which one?  I personally like "Metal" but it
also has "Windows" and "Motif".  Supposedly Tk looks like Windows on
Windows and Motif on *nix.  wxWindows (wxPython) can look like Windows on
Windows, GTK or Motif on *nix and Mac (not sure if it is Classic or
the new OSX look) on MacOS (but AFAIK the Mac port isn't as
complete/stable).

| It maybe a big issue to rewrite Java's Swing library into C source
| and transfer into Python' s module.

I'm sure it would be.  Swing is quite large.  Perhaps someone with a
lot of interest could rewrite it in Python and devise some sort of
'Frame' replacement (since it needs to access the native windowing
system at some point).  It would be interesting, if not really nice,
to have a toolkit written (almost) entirely in Python.  The advantage
would be to not depend on any other "native" libraries, just a python
interpreter.

| Python to build a pilot system. Swing maybe slow but my customer can accept.
| If I use Swing thru Jython, it will be slower then directly using Java to
| develop the system. So I need a solution to use Swing in Python, that's all.

I kind of like Steve Purcell's suggestion of increasing minimum CPU
requirement to offset the overhead and appear just as fast.

Also, Java is faster than Jython (by definition pretty much) so
writing stuff that needs to be just a bit faster in Java is a pretty
good idea too.  (BTW, if you haven't tried Jython yet, you will be
quite amazed at its seamless integration with Java classes)

| Thanks all guys here.

No problem.  Glad to help.

-D




From kalle at gnupung.net  Thu Mar  8 14:24:36 2001
From: kalle at gnupung.net (Kalle Svensson)
Date: Thu, 8 Mar 2001 20:24:36 +0100
Subject: Get terminal size in perl and python
In-Reply-To: <20010308194028.A1616@bork.demon.nl>; from egbert@bork.demon.nl on Thu, Mar 08, 2001 at 07:40:29PM +0100
References: <20010308194028.A1616@bork.demon.nl>
Message-ID: <20010308202436.B6106@father>

Sez Egbert Bouwman:
> My perl correspondent uses:
>     use Term::ReadKey;
>     ($cols,$lines)=GetTerminalSize()
> What is the python equivalent ?

I'd use os.environ["COLUMNS"] and os.environ["LINES"], but I've no idea
about how portable they are.  Do you know how the perl function is
implemented?

Anyway, if you're getting all fancy, perhaps it is better to use curses?

> If I don't come up with one, python looses ten points.

Is that so?  You can keep your filthy perl points! <0.7 wink>

Peace,
  Kalle
-- 
Email: kalle at gnupung.net     | You can tune a filesystem, but you
Web: http://www.gnupung.net/ | can't tune a fish. -- man tunefs(8)
PGP fingerprint: 0C56 B171 8159 327F 1824 F5DE 74D7 80D7 BF3B B1DD
[Not signed due to braindamage.  Blame Microsoft Outlook Express.]



From paul.moore at uk.origin-it.com  Tue Mar  6 06:42:08 2001
From: paul.moore at uk.origin-it.com (Paul Moore)
Date: Tue, 6 Mar 2001 12:42:08 +0100 
Subject: I come to praise .join, not to bury it...
References: <mailman.983756766.27347.python-list@python.org> <97vnu3076q@news2.newsguy.com> <slrn9a89dk.2rm.hzhu@rocket.knowledgetrack.com> <DNXo6.24266$1D5.975603@e420r-atl1.usenetserver.com> <9829u70spn@news1.newsguy.com>
Message-ID: <hc2kOid4dW0ryYR+TAJ2+lefHxzl@4ax.com>

On Tue, 6 Mar 2001 10:16:55 +0100 , "Alex Martelli"
<aleaxit at yahoo.com> wrote:
>>>> uc=Sequence(3,u'bbb')
>>>> u'z'.join(uc)
>u'bbbzbbbzbbb'
>>>> 't'.join(uc)
>Traceback (most recent call last):
>  File "<stdin>", line 1, in ?
>TypeError: coercing to Unicode: need string or buffer, tuple found
>
>and a Control-Z now crashes/hangs the Python interpreter (2.0, on
>Win32).  OK, some kind of ugly bug in 2.0 (gotta get 2.1b1 and
>check if the bug is still there...!!!).

Yes, this is fixed in 2.1b1.

Paul.



From aleaxit at yahoo.com  Mon Mar  5 10:06:57 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Mon, 5 Mar 2001 16:06:57 +0100
Subject: Little problem with the "+" operator
References: <983802310.155838463@news.t-online.de>
Message-ID: <980a7i0rg0@news2.newsguy.com>

"Zamurai" <zamurai at gmx.net> wrote in message
news:983802310.155838463 at news.t-online.de...
> I want to make the following:
> variableOne = 500
> myVariable = variableOne + "cool"
>
> But I get the following error: "unsupported operand types for +"

Right -- you can't add an integer and a string.

> Does someone know how to solve the problem?

Sure -- make both variables into strings, e.g.

    myVariable = str(variableOne) + "cool"


Alex





From gradha at iname.com  Mon Mar 12 04:49:18 2001
From: gradha at iname.com (gradha at iname.com)
Date: Mon, 12 Mar 2001 10:49:18 +0100
Subject: Problems writting a wrapper around linux' ssh command
References: <31rd89.lki.ln@127.0.0.1> <kuhf0zj4iz.fsf@lasipalatsi.fi>
Message-ID: <u26i89.0bm.ln@127.0.0.1>

Erno Kuusela <erno-news at erno.iki.fi> wrote:
> do not embed the password in your python program. use a passphraseless
> rsa key instead, it is safer and less pain.

At the moment I am using the getpass module to retrieve the pass from the
user and feed it directly to ssh and scp. Is that secure enough?

-- 

 Grzegorz Adam Hankiewicz   gradha at iname.com - http://gradha.infierno.org
 Other web pages:  http://glub.ehu.es/  -  http://welcome.to/gogosoftware/


From zamurai at gmx.net  Mon Mar  5 10:46:54 2001
From: zamurai at gmx.net (Zamurai)
Date: Mon, 5 Mar 2001 16:46:54 +0100
Subject: Thank you!!
References: <983802310.155838463@news.t-online.de> <3AA3A75C.966D3B01@javanet.com>
Message-ID: <983807214.1209567466@news.t-online.de>

Raymond Hettinger <othello at javanet.com> wrote:
> Zamurai wrote:
> 
> > I want to make the following:
> >
> > variableOne = 500
> > myVariable = variableOne + "cool"
> >
> > But I get the following error: "unsupported operand types for +"
> >
> > Does someone know how to solve the problem?
> 
> Try:
> myVariable = str(variableOne) + "cool"
 
Thank you!!


From robin at stop.spam.alldunn.com  Thu Mar 15 00:59:35 2001
From: robin at stop.spam.alldunn.com (Robin Dunn)
Date: Wed, 14 Mar 2001 21:59:35 -0800
Subject: wxPython UI refresh
References: <98jku0$ifi$1@newstest.laserlink.net> <slrn9avc13.p8n.zeitlin@seth.lpthe.jussieu.fr> <98p123$ls$1@newstest.laserlink.net>
Message-ID: <tTYr6.39$yBO.17760287@news.randori.com>

"Wiwih Gunadi" <wcanopy at covad.net> wrote in message
news:98p123$ls$1 at newstest.laserlink.net...
>
> "Vadim Zeitlin" <zeitlin at seth.lpthe.jussieu.fr> wrote in message
> news:slrn9avc13.p8n.zeitlin at seth.lpthe.jussieu.fr...
>
> >  Try calling Win32 ::UpdateWindow() directly on the embedded window
> (assuming
> > you can find its HWND somehow) and/or sending it WM_PAINT explicitly
> (again,
> > you need the HWND).
>
> How do I call UpdateWindow in wxPython?? isn't it encapsulated in the
> wxWindow.Refresh() method?  I tried it, it didn't work.
>

Vadim was refering to the UpdateWindow function in the win32 api, effectivly
bypassing wxPython.  You can get the window's HWND by calling it's GetHandle
method.  So something like this:

    import win32gui
    win32gui.UpdateWindow(window.GetHandle())


>
> >
> > >BTW, Excel is not an ActiveX control is it? i.e. it's impossible to
> "wrap"
> > >Excel like IE with the activexwrapper, right?
> >
> >  Wrong, you can embed it, of course. Try putting an Excel table inside a
> Word
> > document and editing it.
>
> Yes, but how to do it in wxPython??
> I tried to load the Excel object library, there isn't the equivalent of
> WebControl object of IE in Excel.
>
> When I tried to MakeActiveXClass() the Excel.Application, it complains
> because it can't find the UI part which leads me to believe that Excel
can't
> be treated as an ActiveX control.
>

I've looked too.  None of the CoClass classes in the Excel type library will
instantiate as an ActiveX class.  I seem to remember seeing something in
MSDN about MS Office OLE needing contianer objects to implement IViewObject,
or something like that.  Maybe that's the problem...

There is a ray of hope however.  The WebBrowser ActiveX object (internet
explorer) can view excel workbooks just fine.  In the wxPython demo for
ActiveXWrapper_IE type the path of a .xls file into the combo box and it
should load it up fine.  I didn't look any further to see what else you
could do with it once it is there...  Perhaps this will be enough for your
app?


--
Robin Dunn
Software Craftsman
robin at AllDunn.com       Java give you jitters?
http://wxPython.org      Relax with wxPython!






From ben.hutchings at roundpoint.com  Wed Mar 28 17:26:17 2001
From: ben.hutchings at roundpoint.com (Ben Hutchings)
Date: 28 Mar 2001 14:26:17 -0800
Subject: Getting all the *files* from a directory -- A better way??
References: <mailman.985732718.3725.python-list@python.org> <99r6v7$4hef2$1@fido.engr.sgi.com> <mailman.985740096.16937.python-list@python.org>
Message-ID: <uvgot7com.fsf@roundpoint.com>

"Ken Seehof" <kens at sightreader.com> writes:

> "Paul Jackson" <pj at sgi.com> wrote:
<snip>
> > Close - better to use os.path.join (more portable across
> > operating environments with different path component
> > separator characters, such as '\' in DOS/Windows).
> 
> Actually '/' is generally portable.

Not to MacOS.

> I use os.path.join for combining
> variables, but I use '/' within hard-coded relative paths.
> 
> e.g.:
> f = os.path.join(root, 'doc/chapter2/spam.html')
> rather than
> f = os.path.join(root, os.path.join('doc', os.path.join('chapter2',
> 'spam.html'))))
> 
> This is fine because open() and all other file functions automagically
> normalize the path to the operating system.

No they don't.  Windows system calls that take file names accept '/'
as a name separator, so this works on both Windows and Unix (and
AmigaOS).  (For share names and registry key names, however, Windows
is not so liberal.)

> Warning: if you want to use the result in a command line
> (e.g. os.popen()) be sure to normalize the path with
> os.path.normpath().

Generally good practice, but in fact many command-line programs for
Windows accept '/' in file names.

-- 
Any opinions expressed are my own and not necessarily those of Roundpoint.


From loewis at informatik.hu-berlin.de  Fri Mar 23 03:06:33 2001
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 23 Mar 2001 09:06:33 +0100
Subject: New PEP: The directive statement
References: <mailman.985085006.24236.python-list@python.org> <slrn9bk7fk.fnm.scarblac@pino.selwerd.nl>
Message-ID: <j4k85gkiye.fsf@informatik.hu-berlin.de>

scarblac at pino.selwerd.nl (Remco Gerlich) writes:

> First the inconsistency: the syntax definition in (C) claims that some
> restrictions on the placement of directive may be in place, depending on
> the directive. However, (D) notes that directives must appear as the first
> non-string token in a file, for backwards compatibility. So in practice,
> directive can't be used anywhere in a file.

That's not inconsistent. There is only one directive defined so far,
to activate nested scopes, and it must appear at the top of the file.
In some later release of Python, directive will become a proper
keyword, at which time introducing directives with less strict
placement rules is possible.

> What I miss most in the 'directive' approach is the introspection. Can a
> program see which directives are on? List the current directives and their
> docstrings? Also, if I can turn on a directive further down a file, how can
> I turn it *off*?

Directives operate on a per-module basis, so from a program point of
view, they are neither on nor off (or both, in different modules).
Directives don't have docstrings (just as the print statement has no
docstring), so you can't list them. As for turning of directives: The
nested_scopes directive cannot be turned off, since the compiler needs
to know how variable binding works throughout the entire module. Other
directives may support being turned off, e.g.

directive omit_linenumbers 0 # or on
directive omit_linenumbers 1 # or off

> Transitional features are transitional for a hopefully short time. They are
> provided for people who need to test their existing software against an
> upcoming incompatible feature, and for early adopters who want the new
> feature now so they can tinker with it. The version in which the feature
> will become part of the language proper is already decided (2.2, in the case
> of nested scopes).

That concerns me about PEP 236: The absolute certainty in which people
predict the future. My experience is that the future is unpredictable
when it comes to software releases. It may turn out that there are
unforeseen problems with nested scopes, or that foreseen problems turn
out to be more serious than expected.

> The 'directive' idea still has the magical first line of the file bit, but
> it doesn't provide information to the running program about what is turned
> on.

There is nothing magical about it at all. Try placing a future import
as the second statement - it has exactly the same restriction. It's a
restriction imposed by the feature being activated, namely
nested_scopes.

> It's not possible to make a single tool once that removes unneeded
> 'directive' statements from files, or even prints a list of
> available statements.

Not true, it is certainly possible to build such a tool.

> And then the other half of the problem, use of the directive
> statement for other things. Is this vapor? I can't think of any
> useful applications at the moment

Tim claims that I invented the statement to declare file encodings,
e.g.

directive encoding "koi8-r"

However, source encodings are not implemented yet, so there is no
point in defining a directive for it. Also, propagating command line
options (-U, -O, -OO) on a per-module basis would be an application.

> But you can't toggle directives

Not true, that depends on the directive.

Regards,
Martin


From mcfletch at home.com  Wed Mar 28 08:48:25 2001
From: mcfletch at home.com (Mike C. Fletcher)
Date: Wed, 28 Mar 2001 08:48:25 -0500
Subject: diff lists
In-Reply-To: <3AC1DD4E.4BA7D421@aon.at>
Message-ID: <000001c0b78d$c3f59fe0$a9807018@cr706570a>

>>> a= ['a', 'b', 'c', 'd', 'e', 'f']
>>> b = ['e', 'c', 'f']
>>> [ element for element in a if element not in b ] # python 2.0+
['a', 'b', 'd']
>>> filter( lambda x, y=b: x not in y, a ) # any python
['a', 'b', 'd']
>>> 

That's not efficient for huge lists, but fine for short ones.

HTH,
Mike

-----Original Message-----
From: Oliver Vecernik [mailto:vecernik at aon.at]
Sent: Wednesday, March 28, 2001 7:47 AM
To: python-list at python.org
Subject: diff lists


Hi,

I've got following two lists:

['a', 'b', 'c', 'd', 'e', 'f']
['e', 'c', 'f']

I'd like to have the result:

['a', 'b', 'd']

The list need not to be ordered. ['d', 'a', 'b'] will also be ok. What
is the most effective way to achive that result?

Oliver
-- 
http://mail.python.org/mailman/listinfo/python-list




From robin at alldunn.com  Tue Mar 13 18:03:18 2001
From: robin at alldunn.com (Robin Dunn)
Date: Tue, 13 Mar 2001 15:03:18 -0800
Subject: How do I get Win32 to locate DLLs elsewhere (not really a Python thing)?
References: <mailman.984506597.12469.python-list@python.org>
Message-ID: <kHxr6.293$XlD.39518660@news.randori.com>

"Skip Montanaro" <skip at pobox.com> wrote in message
news:mailman.984506597.12469.python-list at python.org...
> I'm trying for the first time to use Python (2.0) on Win98 with external
> DLLs (GTK/GDK/...).  I installed all the various bits and pieces for
> GIMP/GTK and PyGTK, but three DLLs that come with GIMP/GTK are not where
> Windows is searching for them, so I get import errors.  (GIMP runs
> fine.) Using a tool called the dependency walker (something from MS) I
> am able to extend its module search directories to include C:\Program
> Files\Common Files\GNU so it will resolve those DLLS.
>
> How do I perform that trick so that Microsoft's equivalent of ld.so can
> find them at run-time?

I don't remember the search order off the top of my head, but DLLs are
normally searched for in the following locations:

1. The directory where the loading .exe is.  For Python that can be
python.exe or pythonw.exe, but it also works if the DLL is where the .pyd
file is at if it is the one that links to the DLL.

2. The SYSTEM directory.  (c:\windows\system or something similar)

3. The directories on the PATH environment variable.  Since you are on win98
you probaly have a c:\autoexec.bat that sets this variable.  If you edit
this you will need to reboot for it to take effect globally.

Perhaps the easiest is to copy the various gtk DLLs to the directory where
the _gtk.pyd file is located.  Perhaps better is to add the path where they
are already to the PATH and reboot.

--
Robin Dunn
Software Craftsman
robin at AllDunn.com       Java give you jitters?
http://wxPython.org      Relax with wxPython!







From jshREMankl at speakeasy.org  Sat Mar 17 15:58:53 2001
From: jshREMankl at speakeasy.org (John Shankland)
Date: Sat, 17 Mar 2001 20:58:53 -0000
Subject: How to get a frame's size
Message-ID: <906782E21jshanklspeakeasyorg@207.126.101.100>

I have a frame called IndexCard1 with some widgets packed in it. I want to 
make several of these and put them in a grid. I need to know how many I can 
fit on the screen, so I need the size of the frame. I'm able to get the 
dimensions of the screen okay, but IndexCard1.winfo_width() seems to always 
return 1. As do all of the other size related methods I could find. So my 
question is how do I get the size in pixels of IndexCard1?

Thanks for your time.

--
????
John


From clarence at netlojix.com  Sat Mar 10 23:27:32 2001
From: clarence at netlojix.com (Clarence Gardner)
Date: Sat, 10 Mar 2001 20:27:32 -0800
Subject: Can anyone recomend a good intoduction to C...
References: <mailman.984256456.30652.python-list@python.org>
Message-ID: <984284896.213412935@news.silcom.com>

On Sat, 10 Mar 2001, Tim Peters wrote:
>Right!  My automobile license plate reads
>
>    TOOWTDI
>
>Honest.  I will know that my work on Python is finally complete when the
>average American looks at my license plate and nods in silent agreement.  I'm
>not there yet.  So far, approximately 100 out of 100 look at it, ponder, then
>start making noises like some demented 170-pound bird with a broken beak.
>
>perhaps-i'd-have-better-luck-if-i-moved-to-california-ly y'rs  - tim

Come to Santa Barbara -- you'll get at least three nods (and one big
thumbs-up)


From kens at sightreader.com  Wed Mar 21 18:49:26 2001
From: kens at sightreader.com (Ken Seehof)
Date: Wed, 21 Mar 2001 15:49:26 -0800
Subject: I'm sure there's a good reason!
References: <memo.20010321220225.57471D@digitig.compulink.co.uk>
Message-ID: <011301c0b261$914a2060$04090a0a@upcast.com>

Does this answer your question?

>>> a = r"Am I using \"raw strings\" now?"
>>> print a
Am I using \"raw strings\" now?

If (odd) trailing backslashes were allowed, the example wouldn't
parse like that.

----- Original Message ----- 
From: "Tim Rowe" <digitig at cix.co.uk>
Newsgroups: comp.lang.python
To: <python-list at python.org>
Sent: Wednesday, March 21, 2001 2:02 PM
Subject: I'm sure there's a good reason!


> Can any of the assembled experts explain to me why:
> print r"\hello"
> is fine, but
> print r"hello\"
> isn't? In other words, why does the backslash escape quotes in a raw 
> string? At one level I'm going to guess (without reading the spec) that 
> it's because "that's what the spec says", but why? I thought the idea of 
> raw strings was that they /didn't/ handle escapes!
> 
> TIA,
> 
> Tim
> 
> -- 
> http://mail.python.org/mailman/listinfo/python-list
> 




From MarkH at ActiveState.com  Wed Mar 21 19:34:03 2001
From: MarkH at ActiveState.com (Mark Hammond)
Date: Thu, 22 Mar 2001 00:34:03 GMT
Subject: win32net, documentation
References: <3AB8CD29.C69CC1E2@Bibliothek.Uni-Augsburg.de>
Message-ID: <3AB949C4.1040501@ActiveState.com>

Sibylle Koczian wrote:

> Hello, all,
> 
> I want to do the Python equivalent of 
> 
> net use k: \\server\share
> 

I suggest you use win32wnet - that will work on Win9x too.

Look in win32\demos\win32wnet for some examples...

Mark.



From aleaxit at yahoo.com  Fri Mar 23 12:26:02 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 23 Mar 2001 18:26:02 +0100
Subject: How to use map
References: <slrn9bmq1g.pav.apardon@trout.vub.ac.be>
Message-ID: <99g0v902fj3@news2.newsguy.com>

<apardon at trout.vub.ac.be> wrote in message
news:slrn9bmq1g.pav.apardon at trout.vub.ac.be...
    [snip]
> I was wondering should k, l and m now be lists of lists
> of the same length and all sublists having the same
> length too, if it would be possible to eliminate the
> use of any loop and get the same result as
>
>   for i in range(len(m)):
>     for j in range(len(m[0])):
>       m[i][j] = f( k[i][j] , l[i][j] )

Depends on what you mean by 'loop' -- you can do
it with list comprehensions quite easily, of course:

m = [ [ f(ak, al) for ak, al in zip(lak, lal) ]
            for lak, lal in zip(k, l)
    ]

but one might argue that these _are_ just loops by
some other name (as, of course, another might
argue of map & friends).


Alex





From sdm7g at virginia.edu  Wed Mar 14 13:49:12 2001
From: sdm7g at virginia.edu (Steven D. Majewski)
Date: Wed, 14 Mar 2001 13:49:12 -0500 (EST)
Subject: Who's minister of propaganda this week?
In-Reply-To: <3AAFA13D.76E32C3F@divalsim.it>
Message-ID: <Pine.NXT.4.21.0103141255310.229-100000@localhost.virginia.edu>


On Wed, 14 Mar 2001, Nicola Musatti wrote:
> 
> But those things that the compiler used to check for you may well have
> become new tests for you to write: if the trade-off between the decrease
> in actual code size and the increase in test number is negative, even
> the better support for refactoring turns out to be an illusion. 

What sort of tests are you expecting to have to do ? 

 In a dynamically typed, polymorphic language (like Python,Lisp,Scheme...)
the language runtime does the type checking for you and in a object
oriented language, dispatching takes the place of most other type 
conditionals. 

( OK: not having multiple dispatch in Python means that some manual
dispatching is sometimes required, but usually this is burried deep
in some classes __coerce__ method. ) 

 If think you're expecting a file, what you are probably *really* 
expecting is a file-like object with a read or write method. Testing
if it is in fact a file is probably the wrong thing to do. ( This
has been termed "pedantic typechecking" in c.l.python ). If you try
to write to something that is non-writable, the runtime will signal
an error.  If you want a function that can take, for example, either
a filename or a file-like-object ( file, pipe, stringIO, socket, etc.)
then you need to distinguish between string-like and file-like args.
Probably the best method is to test getattr( obj, "write" ) [or "read"].

> >   Expressiveness ... is a far better aid to
> >   correctness ...
> 
> Here I'm puzzled, but maybe it's only a "language" problem. What do you
> mean by "Expressiveness"? I ask because until now I thought
> expressiveness was an advantage of strongly typed languages (namely,
> C++) over typeless ones.

You can read:
"On the Expressive Power of Programming Languages" by Matthias Felleisen
<http://citeseer.nj.nec.com/felleisen90expressive.html>
if you want a formal definition. 

I don't think static typing has much difference in expressiveness
one way or another. Polymorphism -- in either it's static or dynamic
variants is, I think, a plus. Otherwise, type annotation of any style
increases redundancy (this is good from a software engineering POV)
and verbosity, and thus reduce your expressiveness/verbosity ratio.


One terminological dispute (which has been endlessly debated on 
comp.lang.*, but may be an important distinction that is causing
some of your confusion. ):

 C++ is not usually considered "strongly typed" 

 There are two axis: 
   static vs. dynamic typing, and string vs. weak typing. 

    C++ is statically, but not strongly typed (certainly
     not compared to Haskell or ML) 

    Python, Lisp and Scheme are dynamically typed. 
      (Comparing whether one has a stronger type system than
       the other is probably beyond me(*), but by some definitions
       they are all considered to be "stronger" than C/C++ . ) 


-- Steve Majewski <sdm7g at Virginia.EDU> 

[(*) I think there's some argument to be made, especially in the
 light of a lot of c.l.python discussion about rationals, integer
 division, long int extensions, etc. that Python's numeric type
 system is, in some sense, weaker than Scheme's. But I don't have
 time to properly make that argument, so you can just take it as
 opinion. And mostly everyone seems to want to fix python to be
 "stronger" if it can be done without too much expense in processor
 time or code-breakage. ] 




From kalle at gnupung.net  Fri Mar 16 17:52:07 2001
From: kalle at gnupung.net (Kalle Svensson)
Date: Fri, 16 Mar 2001 23:52:07 +0100
Subject: Type mismatchings and the rank newbie
In-Reply-To: <ylws6.2639$227.274111@newsread2.prod.itd.earthlink.net>; from jgbrawley@earthlink.net on Fri, Mar 16, 2001 at 10:21:18PM +0000
References: <ylws6.2639$227.274111@newsread2.prod.itd.earthlink.net>
Message-ID: <20010316235207.A1082@apone.network.loc>

Sez John Brawley:
> Rank newbie here, trying to do something specific.
> I need to use a set of numbers (say, 1 to 25), and create 25 lists from the
> numbers (1 thru 25), which lists have a character "P" and a number (one of
> the 25).
> 
> The lists I want look like :
> P1=[x, y, z]
> P2=[x, y, z]
> P3=[x, y, z]
> (etcetera down to P25)
> 
> The user inputs the number that creates the lists, so I can't just write 25
> lists (the number could be anywhere from 13 to 500 or more).

Is there any special reason you can't use a list or a dictionary?

p = []
max = input("How many lists? ")
for i in range(max):
    p.append([x,y,z])
# note that this will create lists from
# p[0] to p[max-1], inclusive.

p = {}
max = input("How many lists? ")
for i in range(max):
    p["P" + str(i+1)] = [x,y,z]
# this will create lists from
# p["P1"] to p["Pmax"], inclusive.

HTH,
  Kalle
-- 
Email: kalle at gnupung.net     | You can tune a filesystem, but you
Web: http://www.gnupung.net/ | can't tune a fish. -- man tunefs(8)
PGP fingerprint: 0C56 B171 8159 327F 1824 F5DE 74D7 80D7 BF3B B1DD
 [ Not signed due to lossage.  Blame Microsoft Outlook Express. ]



From just at letterror.com  Tue Mar 13 10:15:54 2001
From: just at letterror.com (Just van Rossum)
Date: Tue, 13 Mar 2001 16:15:54 +0100
Subject: main in mac?
References: <20010313095339.19264.00000243@ng-dh1.aol.com>
Message-ID: <3AAE3980.D5E61F2A@letterror.com>

Park997 wrote:
> 
> Hi,
> 
> I am using Python 2.0 on a PPC mac. I have a module
> that is set up as follows:
> 
> from Numeric import *
> 
> class Epower:
> 
>         3 class functions defined
> 
> if __name__ == '__main__':
> 
>         test class and print output
> 
> This code does nothing if the run all button is pressed
> or if it is imported into interactive python.
> 
> when the above if statement is replaced by:
> 
> if 1:
> 
> the program runs properly with output. I thought that
> if __name__ == '__main__':
> is standard syntax. Does this need to be changed for mac
> Python?

No: if you look at the option popup menu -- which is very well hidden
at the top right corner of your edit window, it looks like an arrow --
there's an option called "run as __main__". This will set the __name__
variable in the script to "__main__".

Hope this helped,

Just


From djc at object-craft.com.au  Thu Mar  1 20:12:32 2001
From: djc at object-craft.com.au (Dave Cole)
Date: 02 Mar 2001 12:12:32 +1100
Subject: Lazy construction of Python proxies for C parse tree
References: <JAwl6.17037$5n4.338201@news6-win.server.ntlworld.com> <mailman.982950492.15473.python-list@python.org> <v3pm6.27666$5n4.596175@news6-win.server.ntlworld.com> <97dhh2$p9ul0$1@ID-11957.news.dfncis.de> <m33dd0juwy.fsf@vole.object-craft.com.au> <yw3n6.1025$sT1.130520@news2-win.server.ntlworld.com>
Message-ID: <m3wva9otxb.fsf@vole.object-craft.com.au>

>>>>> "Rich" == Rich Somerfield <rich_somerfield at tertio.com> writes:

Rich> That code would be gratefully received.
Rich> rich_somerfield at nO-sPaM.tertio.com

Oops - I somehow missed your response.

I wrote the code for someone else quite a while ago to demonstrate a
point.  They wanted to be able to load massive electronic schematic
files and make them available to Python.  The problem was similar to
yours, due the amount of the data they were loading, Python was just
too slow and too resource hungry.  They had a yacc grammar which they
were using and wanted to make the parse tree available to their Python
code.  I did not have any schematic files I could load, so I picked
the largest text file I could find on my computer; the Debian Packages
file.

Since I was only trying to demonstrate a technique, I took a few
shortcuts.  I did not bother trying to write a yacc grammar, I just
wrote a simple parser in C.  Instead of building my own collections, I
used those which come in GLIB.

There is too much code to post here, so I have placed it on our web
server.

        http://www.object-craft.com.au/projects/lazy/

There is a benchmark program which compares the speed of the raw
parser and the parser accessed through a Python module.  On my
machine, the Packages file is 121,111 lines and 4,471,834 bytes.  The
benchmark program prints the following:

Raw parser loads in 0.70 seconds
aalib1 1.2-18
aalib1 1.2
libc6 2.2.1
libgpmg1 1.14-16
slang1 1.3.0-0
xlibs 4.0.1-11
Wrapped parser loaded and used in 0.75 seconds

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
#!/usr/bin/env python

import os
import time

file = '/var/state/apt/lists/ftp.us.debian.org_debian_dists_unstable_main_binary-i386_Packages'

# Prime the buffer cache
os.system('./pkg_test %s' % file)

t0 = time.time()
os.system('./pkg_test %s' % file)
t1 = time.time()
print 'Raw parser loads in %.2f seconds' % (t1 - t0)

import dpkg
main = dpkg.load_file(file)
deps = main.packages['aalib-bin'].Depends
for dep in deps:
    print dep.Package, dep.Version
t2 = time.time()
print 'Wrapped parser loaded and used in %.2f seconds' % (t2 - t1)
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

The files are:

Makefile.pre.in -- used to build module
Setup.in        -- used to build module
benchmark.py    -- the above benchmark program
dpkg.c          -- the Python wrapper of the parser
pkg.c           -- simple (and incomplete) Packages file parser
pkg.h           -- parser interface
pkg_test.c      -- drives the parser from C

The key to the technique is that Python proxies for the parse tree
nodes are created on demand.  The Python attributes of each parse tree
node are created on demand.

dpkg.c is the interesting file as it demonstrates the technique.

The dpkg.c file should have probably been two files as it contains two
things; but what the hell, I wrote this for someone to demonstrate a
point.  It contains the building blocks for constructing the lazy
evaluation proxy objects, and it also contains the proxies for the
pkg.c parser.

If enough people are interested, I will do some more work on splitting
the code up to make it more general.

Oh, to build and test the code do this:

% make -f Makefile.pre.in boot
% make
% cc -g -Wall -o pkg_test pkg_test.c pkg.c `glib-config --cflags --libs`
% ./benchmark.py 

Traps for young players:

1- The data is read-only.

2- If you drop the reference to the top level parse tree object,
   things will go bad very quickly (main in the benchmark program).

- Dave

-- 
http://www.object-craft.com.au


From tuttledon at hotmail.com  Sat Mar 10 10:59:16 2001
From: tuttledon at hotmail.com (Don Tuttle)
Date: Sat, 10 Mar 2001 15:59:16 GMT
Subject: Python In A Nutshell
References: <3AB1700C@MailAndNews.com> <98cn2d$oad$1@newsreaderg1.core.theplanet.net>
Message-ID: <obsq6.62831$__6.12513082@typhoon.southeast.rr.com>

"Franz GEIGER"
> That's one of his qualities and should be greatly appreciated. Alex
EXPLAINS
> things and allways figures out the PYTHONIC way to do things. For me it
> always is a pleasure to read Alex' answers. It' s high time that Alex
writes
> a book, isn't it? Consider this: A book has to save you time. If one
writes
> too much you can pick out what you need for your problem at hand. If one
> writes too little, the book is a waste of time.

I agree with you. What Aahz and I are doing is sometime referred to as "just
pick'n".  Believe it or not, it's an American way of showing affection
towards a friend, something that might not be so obvious to an international
audience like this one.

So let be clear, I have no doubt's Alex's book will be a valuable addition
to any Python library.  I'm very much looking forward to it. (That's why in
my first post I was complaining about it NOT already being in the
bookstores!)

One last thought.  I wonder it the reason we haven't heard from Alex is that
there's no Unicode mapping for hand gestures? <wink>  (i.e. Just pick'n  ;-)

Don





From dav-nospam-id at westco-nospam-ntrol.com  Fri Mar 23 03:43:59 2001
From: dav-nospam-id at westco-nospam-ntrol.com (David Brown)
Date: Fri, 23 Mar 2001 09:43:59 +0100
Subject: Cross-platform serial port access
References: <99d274$ogv$1@news.netpower.no> <PRou6.1592$SB2.181656@ruti.visi.com> <99d7g3$qed$1@news.netpower.no> <PRqu6.1603$SB2.186660@ruti.visi.com>
Message-ID: <99f2c6$krk$1@news.netpower.no>

It looks like it will be a while before I can do much about this - maybe one
day.  Anyway, thanks for your help.


Grant Edwards wrote in message ...
>In article <99d7g3$qed$1 at news.netpower.no>, David Brown wrote:
>
>>>Currently, serial port stuff in Python is not portable between
>>>Mac/MS/Unix.
>>
>>I suppose it should be easy enough to make a small wrapper around the
>>different platforms' code.
>
>Yup -- somebody should do that someday...
>
>>>>Does anyone know if Linux supports RTS toggle handshaking
>>>>(i.e., the RTS line is high while transmitting, for convenient
>>>>connection to RS-485 convertors),
>>>
>>>Not with the standard PC serial port driver.  I hacked the
>>>16x50 driver to do RTS toggle several years ago, but it never
>>>worked reliably.  It's very difficult to do it reliably in
>>>software at high baud rates.
>
>Back when I tried to do it with the Linux driver, I was trying
>to use extra transmit data bytes (didn't work well) and later
>the shift-register empty bit to shut off RTS.  I should have
>done it using loop-back: enable loopback in the UART and shut
>off RTS in the receive handling code when the end of the tx
>message is seen.  That only works if loopback in the UART
>doesn't disable the transmit data pin...
>
>>>Your best bet is to buy a board that supports RTS toggle in
>>>hardware or use a 232-485 converter that handles it
>>>automatically by detecting the presence of data.
>
>>This is definitely a shame.  The RTS toggle works well under NT
>>(no idea about Win9x) - I have written a Delphi program that
>>uses it to run a RS-485 convertor at 57600.  I don't know for
>>sure how fast RTS is turned off after the last byte - the other
>>nodes on the bus wait 10 ms before answering anyway.
>
>That's the trick.  10ms is long enough that it can be done in
>software.  I had to get it shut off within 1 byte-time of the
>last stop bit.  At 57.6K that's 17us.
>
>>I guess under Linux I would have to toggle RTS manually.
>>Turning it on before sending is easy enough, it is turning it
>>off at the right time that is more of a hassle.
>
>Yup - it's extremely tricky.  I finally gave up and bought a
>board that did it in hardware.  If you've got a converter that
>loops back transmit data, you can use that: do a write(), then
>do a read() -- when you've seen the right number of bytes, shut
>off RTS.
>
>Some of the middle and upper level serial driver layers in
>Linux have a 10ms granularity -- so you might not be able do
>anything from user-space faster than 10ms.
>
>>So-called "intelligent" RS-485 convertors that read the
>>incomming data and change direction appropriately have their
>>own problems.  We have delivered systems that communicated at
>>300 baud, and "intelligent" convertors frequently do not work
>>at such slow rates.  Additionally, they generally need
>>configured with DIP-switches and the like, which is an extra
>>hassle if you need to change the baud rate.
>
>Yup.  The ones without dip-switches (fixed tx data timeout)
>will only work well at certain baud rates.
>
>>Simple RTS-controlled convertors are much easier in that
>>respect.
>
>I agree.
>
>>Although it is not really Python-specific, do you know if there
>>are serial drivers for Linux that simulate 9-bit UARTs (i.e.,
>>swapping the parity bit from space to mark as appropriate)?
>
>I don't think so.
>
>--
>Grant Edwards                   grante             Yow!  .. Do you like
>                                  at               "TENDER VITTLES?"?
>                               visi.com




From emile at fenx.com  Thu Mar 22 23:21:07 2001
From: emile at fenx.com (Emile van Sebille)
Date: Thu, 22 Mar 2001 20:21:07 -0800
Subject: Good Editor?
References: <3ab9a32b$1_1@news2.one.net> <99cv2t$f20g$1@ID-11957.news.dfncis.de> <sqzu6.12865$_46.475246@e420r-atl2.usenetserver.com>
Message-ID: <99ej5k$t8im$1@ID-11957.news.dfncis.de>

<OstrichMode>

What's unicode look like in hex?  How would I test it?

</OstrichMode>


Oh wait, from the Help page:

How to Work with Unicode Files
Overview:
TextPad automatically detects 16-bit Unicode and UTF-8 encoded characters,
when opening files. Unicode characters may be in "little endian" (Intel) or
"big endian" (RISC) order, and the order is preserved when a file is saved.

Internally, these files are converted to single or double byte characters
(DBCS), using the locale corresponding to the font script selected for the
document class. For example, if the screen font for the Text document class
is MS Mincho, with the script set to Japanese, Unicode characters in *.TXT
files will be converted to the corresponding DBCS characters in code page
932.



Emile van Sebille
emile at fenx.com


"Steve Holden" <sholden at holdenweb.com> wrote in message
news:sqzu6.12865$_46.475246 at e420r-atl2.usenetserver.com...
> "Emile van Sebille" <emile at fenx.com> wrote in message
> news:99cv2t$f20g$1 at ID-11957.news.dfncis.de...
> >
> > For windows, I like textpad http://www.textpad.com/ beause it uses
> > workspaces, has a hex display mode, runs code in a separate process,
> > supports other languages (programming and spoken), and generally behaves
> > well.
> >
> > Emile van Sebille
> > emile at fenx.com
> >
> Does it show Unicode string constants correctly in hex?
>
> regards
>  Steve
>
>
>




From lars at pixar.com  Tue Mar  6 19:03:14 2001
From: lars at pixar.com (Lars Damerow)
Date: Tue, 6 Mar 2001 16:03:14 -0800
Subject: class object access in class definition?
Message-ID: <Pine.LNX.4.21.0103061556510.2015-100000@zero>

Hi folks!

Is it possible to reference a class from within its own definition? Something
along the lines of:

class Foo:
    print Foo

This works for functions--that is,

>>> def foo():
...     print foo
...
>>> foo()
<function foo at 0x8111024>

works as expected, but the class definition above raises a NameError:

>>> class Foo:
...     print Foo
...
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
  File "<stdin>", line 2, in Foo
NameError: There is no variable named 'Foo'

Thanks for any pointers!
-lars

___________________________________________________________
lars damerow
senior unix administrator
pixar animation studios
lars at pixar.com

"Nauseous.  Nauseated.  The first means 'sickening to contemplate'; the second
means 'sick at the stomach.'  Do not, therefore, say, "I feel nauseous," unless
you are sure you have that effect on others."

- William Strunk Jr., "The Elements of Style"



From Eugene.Leitl at lrz.uni-muenchen.de  Thu Mar 29 09:07:24 2001
From: Eugene.Leitl at lrz.uni-muenchen.de (Eugene Leitl)
Date: Thu, 29 Mar 2001 16:07:24 +0200 (MET DST)
Subject: CGI XML-RPC (xmlrpccgi.py) demo?
In-Reply-To: <NHGw6.4158$4N4.847384@newsc.telia.net>
Message-ID: <Pine.GSO.4.03.10103291556460.23334-100000@sun1.lrz-muenchen.de>

On Thu, 29 Mar 2001, Fredrik Lundh wrote:

> sorry, it looks like I misread your original question -- I thought
> you were looking for a way to pass arguments (i.e. do RPC calls)
> to a server process via HTTP and CGI, not that you had to use
> a certain syntax.

Oh, I can use XML-RPC just fine, as my (apparently quite
buzzword-compliant) company wants to give their products in future a SOAPy
flavour (they kept mishearing me saying SOAP when I was saying Zope ;). I
just thought that CGI XML-RPC would also handle parsing POST/GET
parameters -- well, it would have been too easy then. Sigh.

I guess I'll have to parse the parameters by hand, and using XML-RPC
at a later stage of the processing pipeline. 

> under XML-RPC, you send XML-RPC requests, not FORMs.  see
> http://www.xmlrpc.com for more info.

Thanks, have been there (though got no T-shirt yet).

You as others been very helpful. Now if somebody could tell me where I can
get Bobo for Medusa...




From me at nospam.net  Mon Mar  5 22:11:21 2001
From: me at nospam.net (Scottie)
Date: Mon, 5 Mar 2001 19:11:21 -0800
Subject: Python 2.1 beta 1 documentation online--extending on Windows doc.
References: <mailman.983566565.21064.python-list@python.org>
Message-ID: <ta8latc8qa9098@corp.supernews.com>

"Fred L. Drake, Jr." <fdrake at acm.org> wrote in message
news:mailman.983566565.21064.python-list at python.org...
>
>   The documentation for Python 2.1 beta 1 is now online:
...

I note the "4.1 A Cookbook Approach" (windows version)
refers to Ascher's compile.py script from
http://starship.python.net/crew/da/compile/
which is not, unfortunately, available at that address.

http://starship.python.net/crew/da
claims the new home page is:
    http://starship.python.net:9673/crew/da
and the old contents are on:
    http://starship.python.net/crew/da/index_old.html
There you can find a link for
    An automatic extension compilation tool
which points to:
    http://starship.python.net/crew/da/compile


What's a poor boy to do?





From gilles.gosselin at t-online.de  Sat Mar 17 12:46:33 2001
From: gilles.gosselin at t-online.de (Gilles Gosselin)
Date: Sat, 17 Mar 2001 18:46:33 +0100
Subject: [Newbie]python,tkinter,tcl/tk 8.3
Message-ID: <pt7099.t45.ln@darkstar.example.net>

Hallo all,

I'm a very new user of Python language (< 2 weeks).
I've looked documentations etc.
For many examples in documentations, I must use tkinter.
python error is:
	ImportError: libtk8.0.so: cannot open shared object file
OK, it's normal I Have tcl/tk 8.3
I Have searched in /usr/lib/python1.5/config/Setup,I find a line like:
# *** Uncomment and edit to reflect your Tcl/Tk versions:
I modify it like
	-ltk8.3 -ltcl8.3 \

And Now, What I must do ?
I Have trying /usr/lib/python1.5/config/makesetup, etc nothing solve
my problem.

Thank's for help.

-- 
Gilles.Gosselin at T-Online.de
Moltkestrasse 19
77933 LAHR GERMANY


From mcfletch at home.com  Wed Mar 28 12:06:34 2001
From: mcfletch at home.com (Mike C. Fletcher)
Date: Wed, 28 Mar 2001 12:06:34 -0500
Subject: Python assembler
In-Reply-To: <3ac20dcc$0$15028$cc9e4d1f@news.dial.pipex.com>
Message-ID: <000001c0b7a9$7204d810$a9807018@cr706570a>

I _think_ (not really sure, as never used it) that bytecodehacks provides an
"assembler" language for Python.  Project is at:
http://bytecodehacks.sourceforge.net/

HTH,
Mike

-----Original Message-----
From: Eddie and Babs [mailto:kca17 at dial.pipex.com]
Sent: Wednesday, March 28, 2001 11:08 AM
To: python-list at python.org
Subject: Python assembler
...
Is the lack of a Python bytecode assembler a deliberate policy decision or
is it just that nobody has written one yet? Or have many tried, but none
lived to tell the tale?
...




From richard at starfighter.freeuk.com  Fri Mar 23 13:56:55 2001
From: richard at starfighter.freeuk.com (Richard Townsend)
Date: Fri, 23 Mar 2001 18:56:55 -0000
Subject: Catching exceptions in Tkinter's mainloop
References: <%dws6.9947$Q47.2723350@news1.rdc1.tn.home.com> <drMs6.32482$HR6.3694604@nnrp4.clara.net> <owzt6.21061$Q47.5935326@news1.rdc1.tn.home.com> <6QNt6.37618$g63.4509118@nnrp3.clara.net> <Qydu6.26214$Q47.7466405@news1.rdc1.tn.home.com> <oWru6.48063$HR6.5435948@nnrp4.clara.net> <hStu6.28419$Q47.7934789@news1.rdc1.tn.home.com>
Message-ID: <x1Nu6.44862$g63.5636850@nnrp3.clara.net>

"Mike Callahan" <mcalla at home.com> wrote in message
news:hStu6.28419$Q47.7934789 at news1.rdc1.tn.home.com...
> OK Richard,
>
> I will do some more digging on this end. Thanks for all your help. By the
> way, do you know the difference between mainloop() and mainloop(1)? What
is
> the optional argument for?
>
> Mike C.
>

Mike

I'm afraid I don't know what the option arg is for. The only reference I've
found to it is at:

http://mail.python.org/pipermail/edu-sig/2000-September/000651.html

but it didn't really me help much.

Good luck getting your program working.

Richard






From db3l at fitlinxx.com  Mon Mar 26 20:43:13 2001
From: db3l at fitlinxx.com (David Bolen)
Date: 26 Mar 2001 20:43:13 -0500
Subject: close(), exceptions and problems
References: <slrn9bm5u5.klj.kuntsi@myntti.helsinki.fi> <99i3j7$lqr$1@news.cybercity.dk>
Message-ID: <uzoe83s26.fsf@ctwd0143.fitlinxx.com>

erwin at andreasen.com (Erwin S. Andreasen) writes:

> What exactly do you mean by "interrupted" ?
> 
> close(), the system call, usually cannot fail (unless you try to close a fd
> that is not open).

Note that close() is often is the point at which buffered data is
flushed to a file, so there are a variety of ways in which that can
fail (I/O error, out of disk space, etc...) which will cause the
close() to fail.

I had a co-worker once who didn't check the results of a close() after
re-writing a new version of a config file following an update, and
didn't realize that it got truncated due to running out of disk space.

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


From jwbnews at scandaroon.com  Mon Mar 12 22:05:04 2001
From: jwbnews at scandaroon.com (John W. Baxter)
Date: Mon, 12 Mar 2001 19:05:04 -0800
Subject: This math scares me
References: <3aad4062.1290021@News.CIS.DFN.DE> <Hwbr6.1139$Tg.127571@ruti.visi.com> <3aad4485.2348499@News.CIS.DFN.DE>
Message-ID: <jwbnews-7CC03F.19050312032001@corp.supernews.com>

In article <3aad4485.2348499 at News.CIS.DFN.DE>, costas at springmail.com 
wrote:

> Joking aside, this should be fixed to display the number correctly. If
> I use 'print 5.01+5.54" then it gives the result correctly. It seems
> to happen only if you type it in the command line without a print
> command.

Many of us feel that the present truthful display is correct, and the 
former (before Python 2) display of 10.55 was wrong because it was a lie 
(some don't feel quite *that* strongly about it but still prefer the new 
form).

  --John

-- 
John W. Baxter   Port Ludlow, WA USA  jwbnews at scandaroon.com


From vonehr at ira.uka.de  Mon Mar 19 06:24:21 2001
From: vonehr at ira.uka.de (Markus von Ehr)
Date: Mon, 19 Mar 2001 12:24:21 +0100
Subject: Tkinter Canvas inheritance
Message-ID: <3AB5EC65.DF767FDC@ira.uka.de>

Hi,

I have a class inherited from Canvas. The create_line
command works well but when I use the create_image command, the
image isn't visible.

##################################################
from Tkinter import *
import thread

class Slider(Canvas):

    def __init__(self, parent = None, **kw):
        
	# Initialise the base class (after defining the options).
        Canvas.__init__(self, parent, kw)

	self.create_line(34, 52, 34, 127, width=2, fill="black")
	
        slider = PhotoImage(file="button.gif")
        img = self.canvas.create_image(25, 37, anchor=NW, image=slider)

	self.root = parent
	
if __name__ == '__main__':
    root = Tk()
    frame = Frame(root)
    frame.pack()
    
    c = Slider(frame, background="grey", width=50, height=141)
    
    c.pack(side=TOP)
    
    root.mainloop()



When I change the Code and create the image in the main function,
there's no problem, it works well. How can I do it in the constructor
and why doesn't it work????

############################################
class Slider(Canvas):

    def __init__(self, parent = None, **kw):
        
	# Initialise the base class (after defining the options).
        Canvas.__init__(self, parent, kw)

	self.create_line(34, 52, 34, 127, width=2, fill="black")
	
	self.root = parent
	
if __name__ == '__main__':
    root = Tk()
    frame = Frame(root)
    frame.pack()
    
    c = Slider(frame, background="grey", width=50, height=141)
    
    slider = PhotoImage(file="button.gif")
    c.create_image(25, 37, anchor=NW, image=slider)
    c.pack(side=TOP)
    
    root.mainloop()

Does anybody know where the rpoblem is?

Thanks,

Markus


From aleaxit at yahoo.com  Sun Mar  4 16:42:28 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sun, 4 Mar 2001 22:42:28 +0100
Subject: I come to praise .join, not to bury it...
References: <mailman.983485923.1136.python-list@python.org> <97p7iq$kk2$1@nntp6.u.washington.edu> <97rka002k9@news1.newsguy.com> <jwbnews-665CC3.11122604032001@corp.supernews.com>
Message-ID: <97ucs712v2k@news1.newsguy.com>

"John W. Baxter" <jwbnews at scandaroon.com> wrote in message
news:jwbnews-665CC3.11122604032001 at corp.supernews.com...
    [snip]
> > it is _just right_ that .join be a method on the joiner
> > object (e.g., a string) and that it take the sequence of
> > string to be joined as its argument.
>
> OK, thanks.  I'm convinced.  For the first time.  (Which won't

Glad my exposition helped someone.


> Which doesn't mean I have to *like* zero-based indexing...only that I

Of course not.  I don't *like* broccoli just because it's demonstrably
good for me (actually, it's a vegetable I definitely dislike).


Alex





From tino at wildenhain.de  Fri Mar  2 04:35:57 2001
From: tino at wildenhain.de (Tino Wildenhain)
Date: Fri, 02 Mar 2001 10:35:57 +0100
Subject: Newbie: Casting Int to string....
References: <3a9ed412$1@news1>
Message-ID: <3A9F697D.3EC4F75F@wildenhain.de>

Hi D.H.


try str() :-)


Digital Hyakugei wrote:
> 
> Ok, pretty simple, no? I can not find out how to do the following...
> 
> i've got a variable:
> 
> aNumber = 1
> 
> Now i have a function
> 
> nntp.stat( aVal )
> 
> where aVal has to be a string.
> 
> I cant do
> 
> nntp.stat( aNumber )
> 
> cuz it expects a string, not an int.
> 
> How, oh how, do i cast an int to a string?
> I'm sure its something that is forehead slapping obvious, but i just cant
> seem to find it...
> 
> Thanks very much!
> jos
> 
> --
> http://mail.python.org/mailman/listinfo/python-list



From timr at probo.com  Wed Mar 14 01:08:36 2001
From: timr at probo.com (Tim Roberts)
Date: Tue, 13 Mar 2001 22:08:36 -0800
Subject: This math scares me
References: <3aad4062.1290021@News.CIS.DFN.DE> <20010312170356.B3513@better.net> <mailman.984480856.18097.python-list@python.org> <984486395.407379@newsmaster-04.atnet.at> <slrn9asf1d.ja3.qrczak@qrnik.zagroda> <984518199.127140@newsmaster-04.atnet.at> <jwbnews-09D75F.13313313032001@corp.supernews.com> <984521930.27710@newsmaster-04.atnet.at>
Message-ID: <if2uatkmkihtmo9436hdju0kikkplivjuv@4ax.com>

"Werner Schiendl" <ws-news at gmx.at> wrote:
>
>I think Windows supports for a 'Currency' data type with some APIs but never
>used them...

Visual Basic (and products that embed it) does have a Currency data type,
but it is a (clever) hack.  Externally, it looks like a fixed-point value
wherein four decimal places can be stored and manipulated exactly.
Internally, it is just a double precision floating point number, scaled by
10,000.  So, if you do this:

  Dim x As Currency
  x = 1.23

x will actually contain the double precision value 12300.0.  The conversion
routines do the scaling automatically where required.
--
- Tim Roberts, timr at probo.com
  Providenza & Boekelheide, Inc.


From urner at alumni.princeton.edu  Sun Mar  4 13:53:13 2001
From: urner at alumni.princeton.edu (Kirby Urner)
Date: Sun, 04 Mar 2001 10:53:13 -0800
Subject: Missing the functional mind set
References: <mlb2at8ptrr7sh5dgq8qlk2201kquh6tfl@4ax.com> <mailman.983716566.6511.python-list@python.org>
Message-ID: <2j35at4v2u75l507v4djdqjv6if8rsq10u@4ax.com>

Steve Purcell <stephen_purcell at yahoo.com> wrote:

>The problem is that sorting a list can't be done in a functional way.
>

Maybe you didn't see the first responder's solution:

  >>> d = {3: 'd', 2: 'b', 1: 'a', 0: 'z'}
  >>> templist = [d[n] for n in range(len(d))]
  >>> templist
  ['z', 'a', 'b', 'd']

That's a good 2.0 and up solution.

>	>>> d = {3: 'd', 2: 'b', 1: 'a', 0: 'z'}
>	>>> stuff = d.items()
>	>>> stuff.sort()
>	>>> map(lambda i: i[1], stuff)
>	['z', 'a', 'b', 'd']
>
>The second is probably more efficient.
>
>-Steve

Kirby



From erno-news at erno.iki.fi  Sun Mar 11 17:51:44 2001
From: erno-news at erno.iki.fi (Erno Kuusela)
Date: 12 Mar 2001 00:51:44 +0200
Subject: Can anyone recomend a good intoduction to C...
References: <mailman.983894288.19965.python-list@python.org> <ewwp6.1466$y6.241005@ruti.visi.com> <slrn9aeblb.cg1.scarblac@pino.selwerd.nl> <djagat4kk3cs9hg0c3k39cudk6goblhkg6@4ax.com> <989mbk$24nuv$1@fido.engr.sgi.com> <slrn9ah66m.k09.qrczak@qrnik.zagroda> <98auhn0270j@news2.newsguy.com> <slrn9aiaq7.rmu.qrczak@qrnik.zagroda> <98dpoe12ckn@news1.newsguy.com> <3AAA85C1.81FB90AA@shore.net>
Message-ID: <kulmqbj4vz.fsf@lasipalatsi.fi>

In article <3AAA85C1.81FB90AA at shore.net>, "Kurt B. Kaiser"
<kbk at shore.net> writes:

| Alex Martelli wrote:
|| Your opinion is widespread, and self-perpetuating, but it has no
|| current technical basis for such tasks as Python extensions on Linux.
|| 
| My understanding is that the C++ compiler for Linux is still
| considered to be unstable and that that is a significant technical
| factor for sticking with C?

i suspect g++ as good as most proprietary compilers. but since
c++ is so complex and big, and the standard was only recently completed,
there are inevitably much more bugs and other "compiler issues" with
c++ than c. unix people need to worry about about the portability
of their software between different flavours of unix and different
c++ compilers, so these problems are aplified compared to people
working exclusively on m$ or macos. see for example
<URL:http://www.mozilla.org/hacking/portable-cpp.html>.

  -- erno


From MarkH at ActiveState.com  Thu Mar  1 19:25:45 2001
From: MarkH at ActiveState.com (Mark Hammond)
Date: Fri, 02 Mar 2001 00:25:45 GMT
Subject: Unresolved symbols in MSVCRT.DLL when running a py2exe packaged program.
References: <RKln6.6630$hn5.853203@news1.rdc1.mb.home.com>
Message-ID: <3A9EE7AA.2000204@ActiveState.com>

Parzival Herzog wrote:


> So, we attempted to replace the old MSVCRT, by installing a recent copy of
> MSVC++ 6.0.
> This indeed did replace the dll, replacing it with a version dated July 7,
> 2000. However,
> the problemwith the python app persisted exactly as before.

Try just copying the one that came with py2exe over the existing one.

Mark.



From dg at pearl.tao.co.uk  Mon Mar  5 12:22:38 2001
From: dg at pearl.tao.co.uk (David Given)
Date: Mon, 5 Mar 2001 17:22:38 +0000
Subject: Name of the Script
References: <20010304.185354.2114937732.8277@localhost.localdomain>
    <NJCo6.6720$qv3.2834377@nnrp5-w.sbc.net>
Message-ID: <u0i089.7it.ln@127.0.0.1>

In article <NJCo6.6720$qv3.2834377 at nnrp5-w.sbc.net>,
	"Alex Shindich" <alex at shindich.com> writes:
> try the following:
> 
> import sys
> print sys.argv[0]
> ....
> The first argument in the argument list is the name of the script that got
> invoked.

Note that it's not always accurate on Unix systems. There are a number of
ways scripts can be invoked, only some of which pass the name of the file
in argv[0]. (This is the same as a the generic Unix question: `How do I
find out where my executable lives?' It's basically not solveable.) I
don't know about Windows, though.

-- 
+- David Given ---------------McQ-+ "Opportunity is missed by most people
|  Work: dg at tao-group.com         | because it's dressed in overalls and looks
|  Play: dgiven at iname.com         | like work." --- Thomas Edison              
+- http://wired.st-and.ac.uk/~dg -+ 


From tim.one at home.com  Sun Mar 18 23:50:54 2001
From: tim.one at home.com (Tim Peters)
Date: Sun, 18 Mar 2001 23:50:54 -0500
Subject: Help Rewriting/Refactoring/Rethinking Parsing Algorythm
In-Reply-To: <000501c0b02d$8c35b200$a9807018@cr706570a>
Message-ID: <LNBBLJKPBEHFEDALKOLCOEJAJGAA.tim.one@home.com>

[Mike C. Fletcher]
> Not precisely what you're trying to do, but I do most of my
> programming with voice dictation (Dragon NS 5), and I find that a few
> command phrases are fairly useful and let me program in just about
> any editor: ...

Note that Joel Gould (Dragon NatSpeak's chief architect) developed a system
that lets you script NatSpeak from Python:

    http://www.synapseadaptive.com/joel/

He gave a very well-received talk about it at the Python Conference a couple
weeks ago.  The system is called NatLink.  Here's the root of its home on
Joel's site:

    http://www.synapseadaptive.com/joel/PythonMacroSystem.html

It's free for the taking, and lets you use the full power of Python to define
grammars and actions.

recognizingly y'rs  - tim




From akuchlin at mems-exchange.org  Fri Mar 23 11:46:26 2001
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 23 Mar 2001 11:46:26 -0500
Subject: Is setup.py a "good thing"? (long)
References: <3aba76e2.167083508@24.0.228.33> <3dvgp1zasw.fsf@ute.cnri.reston.va.us> <99fsdu$p2u$1@hop.gmd.de>
Message-ID: <3dpuf8zb4t.fsf@ute.cnri.reston.va.us>

woetzel at gmd.de (Gerd Woetzel) writes:
> Modules/Setup has buit zlibmodule.so and socketmodule.so which worked
> fine during the test phase, but setup.py has (secretly?) created
> zlib.so (socket.so) too! Finally "make install" has installed all of them.

That certainly shouldn't have happened.  Can you reproduce the problem?

(One thing to note is that "make clean" doesn't remove .so files from
the build/ tree, so if you did a make, then enabled zlib in
Modules/Setup and did "make clean ; make", the zlib.so is the build
tree would still be there.  Perhaps "make clean" should blow away the
build/ tree, even though that's inconsistent with what the clean
target does in Modules/.)

--amk



From pgf at globalreach.net  Sat Mar 17 12:35:24 2001
From: pgf at globalreach.net (Phil Fraering)
Date: 17 Mar 2001 11:35:24 -0600
Subject: How can install Tkinter on linux.
References: <98rnem$465$2@news2-2.kornet.net>
Message-ID: <87r8zwwb6r.fsf@globalreach.net>

"rambo" <rambo at linuxsecurity.co.kr> writes:

> Hello,
> 
> How can I install Tkinter on linux.
> I download Tkinter-2.0-8.0.tar.gz, and extract it ( tar xvzf
> Tkinter-2.0-8.0.tar.gz  )

Depending on what version of linux you're using, it's probably
available precompiled somewhere for you. If you're using redhat,
it's probably available somewhere at the redhat site; if you're
using debian, which is what I use, you'd use :

apt-get install python2-tk


-- 
Phil Fraering          "American-style iced tea is the perfect drink for
pgf at globalreach.net     a hot, sunny day. It's never really caught on in
                        the UK, probably because the last time we had a
			hot, sunny day was back in 1957." - Tom Holt
			


From kca17 at dial.pipex.com  Thu Mar 29 13:50:34 2001
From: kca17 at dial.pipex.com (Eddie and Babs)
Date: Thu, 29 Mar 2001 19:50:34 +0100
Subject: Quoted identifiers in Python?
References: <3ac3387f$0$12248$cc9e4d1f@news.dial.pipex.com>
	 <3AC35C84.FBED128@alcyone.com>
Message-ID: <3ac38552$0$15025$cc9e4d1f@news.dial.pipex.com>

Ahem... sorry, I missed the crucial word out there: "__dict__"!
I meant that

    x = obj."Ivor the engine"

...would be the same as:-

    x = obj.__dict__["Ivor the engine"]

The point I was really making was that since the object's attributes are
stored in a dictionary, there is no reason why you shouldn't have any string
you like as the key (you can, after all, do that with any other dictionary).
The ability to use arbitrary strings as attribute names (as opposed to just
the usual alpha-num-underscore characters) DOES have its uses when you are
allowed to re-define the attribute access "dot" operator.


----------
In article <3AC35C84.FBED128 at alcyone.com>, Erik Max Francis
<max at alcyone.com> wrote:


> Eddie and Babs wrote:
>
>> NOW... would this be a good thing for Python to have in the future? It
>> would
>> make element access consistent with subscripting:-
>>
>>         x = obj."Ivor the engine"
>>
>>             ...would be the same as
>>
>>         x = obj["Ivor the engine"]
>>
>> ..no?
>
> I would hope they would not be the same, since one invokes __getattr__
> while the other invokes __getitem__.  They're different things, they
> shouldn't be forced to be the same thing.
>
> --
>  Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
>  __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
> /  \ Traveler, there are no paths.  Paths are made by walking.
> \__/ Antonio Machado
>     REALpolitik / http://www.realpolitik.com/
>  Get your own customized newsfeed online in realtime ... for free!


From grante at visi.com  Thu Mar 15 10:04:07 2001
From: grante at visi.com (Grant Edwards)
Date: Thu, 15 Mar 2001 15:04:07 GMT
Subject: This math scares me
References: <20010312170356.B3513@better.net> <mailman.984480856.18097.python-list@python.org> <984486395.407379@newsmaster-04.atnet.at> <slrn9asf1d.ja3.qrczak@qrnik.zagroda> <984518199.127140@newsmaster-04.atnet.at> <jwbnews-09D75F.13313313032001@corp.supernews.com> <98ohuc$du9$1@news.udel.edu> <7bQr6.644$lz1.17686@ruti.visi.com> <98oku1$f6a$1@news.udel.edu> <erRr6.659$lz1.20662@ruti.visi.com> <mailman.984664520.26463.python-list@python.org>
Message-ID: <HR4s6.877$lz1.73512@ruti.visi.com>

In article <mailman.984664520.26463.python-list at python.org>, Fabio Olive Leite wrote:
>Hi there,
>
>On Wed, Mar 14, 2001 at 09:31:54PM +0000, Grant Edwards wrote:
>) No.  The binary floating point representation of many decimal
>) fraction values is not exact.  That's what started this whole
>) thread.  You can't represent 0.1 exactly in binary FP.  You can
>) in BCD.
>
>Just to make matters worse in this discussion, let me add some more noise.
>
>Since base 10 contains base 2 and base 5 (`factor 10`), every
>number coming from any of those bases is exactly represented in
>base 10. Conversely, base 2 contains only itself, and thus it
>cannot represent correctly some numbers that are exactly
>representable in other bases, like 0.1 in base 10.
>
>The usual 1/3 example is a problem for base 10, but not so for
>base 12, for example, since 12 contains 3 2 2. Thus:
>
>1_{12} / 3_{12} == 0.4_{12}
>
>And then base 30 would let us count exactly fractions of base
>2, 3, 5, 6, 10, 12, 15, 16 and all the others you can think of
>that have only 2, 3 and 5 as factors of the base. But since no
>one (usualy) has 30 fingers, that base is not terribly popular.
>:)

One step further is base-60, which is still used for many
things.  Time and angular measurement for example is partially
a base-60 system.  I believe that one or more of the ancient
Aztecs/Incans/Mayans used a base-60 number system.

-- 
Grant Edwards                   grante             Yow!  It's hard being
                                  at               an ARTIST!!
                               visi.com            


From Simon.Budig at unix-ag.org  Mon Mar 12 11:15:12 2001
From: Simon.Budig at unix-ag.org (Simon Budig)
Date: 12 Mar 2001 17:15:12 +0100
Subject: Apache and os.fork() in CGI-Binaries
References: <3aa7ded0@si-nic.hrz.uni-siegen.de> <3AA84BF3.E19873DB@ninenet.com>
Message-ID: <3aacf610@si-nic.hrz.uni-siegen.de>

Matt Harden <matth at ninenet.com> wrote:
> You need os.dup2().  Like this:
>    logfile = os.open("/path/to/logfile",os.O_CREAT|os.O_WRONLY)
>    os.lseek(logfile, 0, 2)  /* seek to end */
>    os.dup2(logfile, sys.stdout.fileno())
>    os.dup2(logfile, sys.stderr.fileno())
>    os.close(logfile)
> 
> I would also close stdin like this:
>    devnull = os.open("/dev/null", os.O_RDONLY)
>    os.dup2(devnull, sys.stdin.fileno())
>    os.close(devnull)

Thanks, this helped. Not nice, but it worked.

Bye,
        Simon
-- 
      Simon.Budig at unix-ag.org       http://www.home.unix-ag.org/simon/


From gs234 at cam.ac.uk  Thu Mar  8 06:43:43 2001
From: gs234 at cam.ac.uk (Gaute B Strokkenes)
Date: 08 Mar 2001 11:43:43 +0000
Subject: "!=" is bad form. Re: sorry....never mind
References: <DZkp6.18811$pF4.3015652@news1.telusplanet.net> <fdlp6.16982$go.2157953@news0.telusplanet.net> <3AA6E308.5B63ECC6@cosc.canterbury.ac.nz> <3AA72253.FB021815@Lugoj.Com>
Message-ID: <4ad7bsxz80.fsf@kern.srcf.societies.cam.ac.uk>

On Wed, 07 Mar 2001, JamesL at Lugoj.Com wrote:

> Greg Ewing wrote:
>> 
>> Victor Louie wrote:
>> >
>> > <> is obsolescent.
>> 
>> Really? I hope it's not taken away, because that will
>> break ALL of my code! I hate != (looks too much like
>> C), so I always use <>. Nobody ever told me I was doing
>> anything wrong!
> 
> Actually, for Python, "!=" IS totally bogus and should be considered
> the unusual variant. Unless the logical negation operation has been
> changed from "not" to "!" (shucks, version 1.5.2 still does all I
> need) then "!=" is not suggestive of anything. In "C" the "!=" was
> suggestive of:
> 
> !(expression1 == expression2)

I prefer != since it is one hell of a lot more reminiscent of the way
inequality is written in mathematics: as a equality sign with a
slightly slanted vertical line through it.  On the other hand, <>
doesn't look like anything even vaguely mathematical. 

PS |= wouldn't do, since it means something different entirely.

-- 
Gatue Strokkenes                        http://www.srcf.ucam.org/~gs234/
You mean now I can SHOOT YOU in the back and further BLUR
 th' distinction between FANTASY and REALITY?


From cyberian_bear at hotmail.com  Sun Mar 11 23:19:07 2001
From: cyberian_bear at hotmail.com (cyberian bear)
Date: Sun, 11 Mar 2001 23:19:07 -0500
Subject: From SETL to Python translation
Message-ID: <t7Yq6.12$N22.336@typhoon.nyu.edu>

Hi guys I have 2 programs below that I must have in Python but I have them
in SETL(whatever the hell that is) and I don't know anything about it. I
tried finding stuff about it on the web but all I found was two small,
irrelevant pages and no tutorials. The thing is, if at least I knew what the
'Euler path construction', or 'Prime factors' are, I could just program them
on my own without looking at the SETL code, but i don't. Basically I just
want to ask some good soul if he could translate these two programs for me
from SETL to Python. I know chances of finding anyone who knows SETL are
negligibe but i guess it's worth a try.
I just want to thank anyone in advance if someone decides to help me. If not
that's OK I understand it's a lot to ask for.
Well. Thanx a lot guys.
cb
PS: my e-mail is cyberian_bear at yahoo.com
----------------------------------------------------------------------------
-------------------
program Euler;    -- Eulerian path construction

    graph := {[1,2], [2,3], [3,4], [4,1], [4,2]};             -- a small
graph
    print(euler_path(graph + {[y, x]: [x, y] in graph}));        -- which is
undirected.

    procedure Euler_path(G);                -- constructs Eulerian path for
graph G

        nodes := domain(G);                        -- all nodes in the
graph.

        if #(odds := {x in nodes | odd(#G{x})}) > 2 then
            return OM;      -- since more than two nodes are
        end if;             -- touched by an odd number of edges

                        -- odds is the set of all nodes of G that
                        -- are touched by an odd number of edges
        x := arb(odds)?arb(nodes);      -- pick a node of odds if possible
                                        -- otherwise pick any node of G

        path := [x] + build_path(x,G);

        while exists z = path(i) | G{z} /= {} loop
            new_p := build_path(z, G); -- insert new section into path
            G -:= ({[y,x]: [x,y] in new_p} + {e: e in new_p});
            path := path(i..i - 1) + new_p + path(i..);
        end loop;

        return path;

    end Euler_path;

    procedure build_path(x, rw G);        -- builds maximal path section
starting at x,
                                          -- and deletes all edges traversed

        p := [ ];

         while (y := arb G{x}) /= OM loop
                 -- while there exists an edge leaving the last point
reached
            p with:= y;                 -- extend path to traverse the edge
            G -:= {[x, y], [y, x]};        -- delete the edge just traversed
            x := y;                        -- step to y
        end loop;

        return p;

    end build_path;

end euler;
----------------------------------------------------------------------------
---------------------
package body prime_factors_pak;

    procedure prime_factors(n);            -- get tuple  of prime factors of
n

        facts := [];        -- will collect

        while even(n) loop facts with:= 2; n /:= 2; end loop;        -- even
factors

        while exists k in [3,5..ceil(sqrt(float(n)))] | n mod k = 0 loop

            facts with:= k; n /:= k;        -- smallest remaining  factor,
up to sqrt(n)

        end loop;

       facts with:= n;            -- the final factor

        return facts;

    end prime_factors;

end prime_factors_pak;




From petasis at iit.demokritos.gr  Wed Mar 14 01:40:00 2001
From: petasis at iit.demokritos.gr (Petasis George)
Date: Wed, 14 Mar 2001 08:40:00 +0200
Subject: Einstein's Riddle
References: <oNPq6.396644$ge4.138479566@news2.rdc2.tx.home.com> <UdSq6.2184$54.2166@www.newsranger.com> <984380185.768920@news.aei.ca> <uP%q6.2375$54.2207@www.newsranger.com> <4w6r6.193645$df5.5192565@news1.crdva1.bc.home.com> <3aae4114.63604234@nntp.interaccess.com> <3AAE4BFD.9A7B0D9C@cs.man.ac.uk>
Message-ID: <3AAF1240.C446D19@iit.demokritos.gr>

"Donal K. Fellows" wrote:
> 
> Thaddeus L Olczyk wrote:
> > I also figure the fish thing was simply a typo, and that this guy (
> > who hasn't resdponded ) was probably given this as a homework
> > assignement, and is simply looking for a cheap way to get it done.
> 
> That'd be a very lame homework assignment.  Not unless you had to write
> a program in $FAVOURITE_LANGUAGE to compute the answer automatically
> (which is much more interesting, especially if you make the program
> start from the natural language expression of the problem!  :^)
> 
Well, try this:-)
Written by a friend of mine. Its amazing how small can programs be if you
know STL. To bad I don't know stl :-)
Of course it will need some time, as it randomly checks all combinations...

George

//A simple randomiser to find the solution to the riddle, by J. Y. Goulermas

#include <vector>
#include <ctime>
#include <algorithm>
#include <fstream>
#include <iomanip>
#include <iostream>

using namespace std;

enum element_type { nation,  colour,  pet,         drink,  cigs       };
enum nation_type  { english, swedish, danish,      german, norwegian  };
enum colour_type  { red,     blue,    yellow,      green,  white      };
enum pet_type     { cat,     bird,    horse,       dog,    fish       };
enum drink_type   { tea,     water,   milk,        beer,   coffee     };
enum cigs_type    { palmal,  dunhill, bluemasters, prince, blends     };

char* codes[][5] = { { "english", "swedish", "danish",      "german", 
"norwegian" },
                     { "red",     "blue",    "yellow",      "green",  
"white"     },
                     { "cat",     "bird",    "horse",       "dog",    "fish" 
      },
                     { "tea",     "water",   "milk",        "beer",   
"coffee"    },
                     { "palmal",  "dunhill", "bluemasters", "prince", 
"blends"    }
                   };

inline bool ValidConstraints(const vector< vector<int> >& p)
{
  return   p[nation][norwegian] == 0                   //Rule 9
         &&
           p[colour][blue] == 1                        //Rule 14
         &&
           p[drink][milk] == 2                         //Rule 8
         &&
           p[nation][english] == p[colour][red]        //Rule 1
         &&
           p[nation][swedish] == p[pet][dog]           //Rule 2
         &&
           p[nation][danish] == p[drink][tea]          //Rule 3
         &&
           p[colour][yellow] == p[cigs][dunhill]       //Rule 7
         &&
           p[pet][bird] == p[cigs][palmal]             //Rule 6
         &&
           p[drink][beer] == p[cigs][bluemasters]      //Rule 12
         &&
           p[nation][german] == p[cigs][prince]        //Rule 13
         &&
            p[colour][green] == p[drink][coffee]       //Rule 5
         &&
           abs(p[cigs][blends] - p[pet][cat]) == 1     //Rule 10
         &&
           abs(p[cigs][blends] - p[drink][water]) == 1 //Rule 15
         &&
           abs(p[pet][horse] - p[cigs][dunhill]) == 1  //Rule 11
         &&
           p[colour][green] < p[colour][white];        //Rule 4
}

void Output(const vector< vector<int> >& p,
                  unsigned               c
           )
{
  ofstream text("solution.txt");

  for (int i = nation; i <= cigs; ++i)
  {
    for (int j = 0; j < p[i].size(); ++j)
      text <<setw(20)
           <<codes[i][ find(p[i].begin(), p[i].end(), j) - p[i].begin() ]; 
//get inverse permutation
    text <<endl;
  }
  text <<"\n\nTotal attempts: " <<c;

  text.close();
}

void main(void)
{
  //srand( (unsigned) time(NULL) ); //optional RNG seeding

  int                   ramp[] = { 0, 1, 2, 3, 4 };
  vector< vector<int> > permutations;
  unsigned              counter(0);
  for (int i = nation; i <= cigs; i++ )
    permutations.push_back( vector<int>(ramp, ramp + 5) );

  do
  {
    for (int i = nation; i <= cigs; i++)
      random_shuffle( permutations[i].begin(), permutations[i].end() );
    if ( ! (++counter % 100000) )
      cout <<"\rRe-Randomisations: " <<counter;
  }
  while ( ! ValidConstraints(permutations) );

  Output(permutations, counter);
}


From sholden at holdenweb.com  Mon Mar  5 23:15:22 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Mon, 5 Mar 2001 23:15:22 -0500
Subject: Python 2.1 beta 1 documentation online--extending on Windows doc.
References: <mailman.983566565.21064.python-list@python.org> <ta8latc8qa9098@corp.supernews.com>
Message-ID: <XyZo6.19549$ML1.1002389@e420r-atl2.usenetserver.com>

"Scottie" <me at nospam.net> wrote
[about the beta docs]
> I note the "4.1 A Cookbook Approach" (windows version)
> refers to Ascher's compile.py script from
> http://starship.python.net/crew/da/compile/
> which is not, unfortunately, available at that address.
>
> http://starship.python.net/crew/da
> claims the new home page is:
>     http://starship.python.net:9673/crew/da
> and the old contents are on:
>     http://starship.python.net/crew/da/index_old.html
> There you can find a link for
>     An automatic extension compilation tool
> which points to:
>     http://starship.python.net/crew/da/compile
>
>
> What's a poor boy to do?
>
Well, if he felt concerned as to whether this was an endemic problem in that
part of the web world, he might do worse than checking the web as a whole.
The GUI-based version of webchecker (a neat tool that comes with this
language called Python ;-) says of http://starship.python.net/crew/da/:

523 total, 6 bad

<1% broken links is probably better than the average search engine output.

I am Cc'ing David Ascher on this mail, as I am sure he will be able to let
Fred have the correct reference (if he has not already).

feeling-my-own-23-total-1-bad-is-puny-by-comparison-even-though-the-bad-one-
        is-actually-good-javascript-ly y'rs  - steve




From robert at kasunic.de  Wed Mar 21 11:55:04 2001
From: robert at kasunic.de (Robert Kasunic)
Date: Wed, 21 Mar 2001 17:55:04 +0100
Subject: python and mysqldb
References: <mjia99.b14.ln@braveheart.dyndns.org> <99aj4a$digd$1@fu-berlin.de>
Message-ID: <3AB8DCE8.5040009@kasunic.de>

Hello Matthias,

Matthias Huening wrote:

> How did you try to insert the variable?
> The following should work:
> statement = "INSERT INTO test_table VALUES ('%s', 'test2', 'test3')" % temp

thank you for your quick help. The solution you proposed works well. 
Could you give me reference for further reading on this topic?

Robert





From gerhard.nospam at bigfoot.de  Thu Mar  1 23:00:02 2001
From: gerhard.nospam at bigfoot.de (Gerhard =?iso-8859-1?Q?H=E4ring?=)
Date: Fri, 02 Mar 2001 05:00:02 +0100
Subject: python irc channel?
References: <97n15e$9g3$1@nntp1.ba.best.com>
Message-ID: <3A9F1AC2.3FCD8B1F@bigfoot.de>

Martin Brown wrote:
> 
> So, is there a preferred irc channel where python people hangout?

There is a #python channel on irc.openprojects.net.

Gerhard
-- 
Sorry for the fake email, please use the real one below to reply.
contact: g e r h a r d @ b i g f o o t . d e
web:     http://highqualdev.com


From cce at clarkevans.com  Wed Mar  7 04:38:15 2001
From: cce at clarkevans.com (Clark C. Evans)
Date: Wed, 7 Mar 2001 04:38:15 -0500 (EST)
Subject: wxPython+py2exe+win98 ImportError: DLL load failed: A device attached
 to the system is not functioning 
In-Reply-To: <Pine.LNX.4.21.0103061336560.21280-100000@clarkevans.com>
Message-ID: <Pine.LNX.4.21.0103070420220.19255-100000@clarkevans.com>

Hello.  I've recently discovered py2exe and wanted to give
it a test run (I'm looking for a good software distribution
system for Python on Windoz).  

I was able to make an executable of a command-line python
program, and it workes just great on my Win2K development
box, and my Win98 install.  Further, I was able to create a
wxPython program (the first example from the wxpython.org
tutorial) and it works great on Win2K.  However, the
executable created gives the following error when I run it
on a fresh copy of Win98:                         

  Traceback (most recent call last):
    File "<string>", line 1, in ?
    File "imputil.pyc", line 88, in _import_hook
    File "<string>", line 20, in _import_top_module
    File "imputil.pyc", line 197, in import_top
    File "imputil.pyc", line 252, in _import_one
    File "<string>", line 95, in _process_result
    File "wxPython\__init__.pyc", line 20, in ?
    File "imputil.pyc", line 78, in _import_hook
    File "imputil.pyc", line 323, in _do_import
    File "imputil.pyc", line 248, in _import_one
    File "<string>", line 130, in get_code
  ImportError: DLL load failed: A device attached to
               the system is not functioning.

I was wondering if other people have the same problem
or if I'm doing something wrong here.  Thanks!  Clark Evans

..................................................
THE MINIMAL COMMAND LINE SCRIPT (I'm new to python)
.................................................
import sys
for line in sys.stdin.readlines():
  print line

..................................................
THE WXPYTHON SCRIPT
..................................................
from wxPython.wx import *

class MyApp(wxApp):
    def OnInit(self):
        frame = wxFrame(NULL, -1, "Hello from wxPython")
        frame.Show(true)
        self.SetTopWindow(frame)
        return true

app = MyApp(0)
app.MainLoop()

....................................................
THE SETUP SCRIPT
.....................................................

from distutils.core import setup
import py2exe

setup(name="myscript",
      scripts=["myscript.py"],
)

..................................................
THE WXPYTHON SCRIPT DIRECTORY CREATED BY PY2EXE
..................................................
03/07/2001  02:44a             199,458 myscript.exe
10/26/2000  01:00p           3,715,072 wxc.pyd
12/07/1999  04:00a             295,000 MSVCRT.dll
10/19/2000  11:30a             667,728 python20.dll
10/21/2000  08:52p           1,917,952 wx22_2.dll
12/07/1999  05:00a              77,878 MSVCIRT.dll

..................................................
DLLs LOADED BY THE COMMAND LINE EXE UNDER WIN2K
...................................................
C:\temp\dist\myscr\python20.dll 
F:\WINNT\system32\ADVAPI32.dll 
F:\WINNT\system32\COMCTL32.DLL 
F:\WINNT\System32\ctype.nls 
F:\WINNT\system32\GDI32.DLL 
F:\WINNT\System32\IMM32.DLL 
F:\WINNT\system32\KERNEL32.dll 
F:\WINNT\System32\locale.nls 
F:\WINNT\system32\MSVCRT.dll 
F:\WINNT\System32\ntdll.dll 
F:\WINNT\system32\RPCRT4.DLL 
F:\WINNT\system32\SHELL32.dll 
F:\WINNT\system32\SHLWAPI.DLL 
F:\WINNT\System32\sortkey.nls 
F:\WINNT\System32\sorttbls.nls 
F:\WINNT\System32\unicode.nls 
F:\WINNT\system32\USER32.dll 

..................................................
ADDITIONAL DLLs LOADED VIA THE wxPython EXE
..................................................

C:\temp\dist\myscript\wx22_2.dll 
F:\WINNT\System32\sbpcir32.dll 
C:\temp\dist\myscript\wxc.pyd 
F:\WINNT\System32\sbpcint4.dll 
F:\WINNT\System32\WS2HELP.DLL 
F:\WINNT\System32\WS2_32.dll 
F:\WINNT\system32\LZ32.DLL 
F:\WINNT\System32\WINMM.dll 
F:\WINNT\system32\VERSION.dll 
F:\WINNT\system32\ole32.dll 
C:\temp\dist\myscript\MSVCIRT.dll 

.....................................................
THE RUN OF THE PY2EXE SCRIPT
.....................................................

C:\temp>python setup.py py2exe
running py2exe
running build
running build_scripts
copying myscript.py -> build\scripts
running install_scripts
creating build\bdist.win32\winexe
creating build\bdist.win32\winexe\lib
creating build\bdist.win32\winexe\lib\Program Files
creating build\bdist.win32\winexe\lib\Program Files\Python
creating build\bdist.win32\winexe\lib\Program Files\Python\Scripts
copying build\scripts\myscript.py -> build\bdist.win32\winexe\lib\Program
Files\
Python\Scripts
copying build\scripts\myscr.py -> build\bdist.win32\winexe\lib\Program
Files\Pyt
hon\Scripts
copying build\scripts\myscrip.py -> build\bdist.win32\winexe\lib\Program
Files\P
ython\Scripts
warning: install: modules installed to
'build\bdist.win32\winexe\lib\Program Fil
es\Python\', which is not in Python's module search path (sys.path) --
you'll ha
ve to change the search path yourself
+------------------------------
| Processing script myscript.py
+------------------------------
creating build\bdist.win32\winexe\collect
creating build\bdist.win32\winexe\collect\myscript
creating build\bdist.win32\winexe\collect\myscript\Scripts
Searching modules needed to run 'myscript.py' on path:
['C:\\temp\\build\\bdist.win32\\winexe\\lib\\Program Files\\Python', '',
'c:\\te
mp', 'f:\\program files\\python', 'f:\\program files\\python\\installer',
'f:\\p
rogram files\\python\\wxpython', 'f:\\program files\\python\\pythonwin',
'f:\\pr
ogram files\\python\\win32', 'f:\\program files\\python\\win32\\lib',
'f:\\progr
am files\\python\\dlls', 'f:\\program files\\python\\lib', 'f:\\program
files\\p
ython\\lib\\plat-win', 'f:\\program files\\python\\lib\\lib-tk']
force_imports =
ext_mapping = {
 'wxPython.wxc': ('wxc.pyd', ('.pyd', 'rb', 3))
}
Resolving binary dependencies:
  F:\WINNT\System32\NTDLL.DLL
  f:\program files\python\wxPython\wxc.pyd
  F:\WINNT\System32\KERNEL32.dll
  F:\WINNT\System32\RPCRT4.DLL
  F:\WINNT\System32\MSVCRT.dll
  F:\WINNT\System32\WINMM.dll
  F:\WINNT\System32\USER32.dll
  F:\WINNT\System32\COMCTL32.DLL
  F:\WINNT\System32\SHLWAPI.DLL
  F:\WINNT\System32\python20.dll
  F:\WINNT\System32\SHELL32.dll
  F:\WINNT\System32\GDI32.DLL
  f:\program files\python\wxPython\wx22_2.dll
  F:\WINNT\System32\ADVAPI32.dll
  f:\program files\python\py2exe\run.exe
  F:\WINNT\System32\MSVCIRT.dll
copying f:\program files\python\py2exe\support.py ->
build\bdist.win32\winexe\co
llect\myscript\Scripts
byte-compiling f:\program files\python\wxPython\clip_dnd.py to
wxPython\clip_dnd
.pyc
byte-compiling f:\program files\python\wxPython\gdi.py to wxPython\gdi.pyc
byte-compiling f:\program files\python\wxPython\mdi.py to wxPython\mdi.pyc
byte-compiling f:\program files\python\wxPython\__version__.py to
wxPython\__ver
sion__.pyc
byte-compiling f:\program files\python\wxPython\__init__.py to
wxPython\__init__
.pyc
byte-compiling f:\program files\python\wxPython\frames.py to
wxPython\frames.pyc

byte-compiling f:\program files\python\wxPython\streams.py to
wxPython\streams.p
yc
byte-compiling f:\program files\python\wxPython\image.py to
wxPython\image.pyc
byte-compiling f:\program files\python\wxPython\cmndlgs.py to
wxPython\cmndlgs.p
yc
byte-compiling f:\program files\python\wxPython\filesys.py to
wxPython\filesys.p
yc
byte-compiling f:\program files\python\wxPython\misc.py to
wxPython\misc.pyc
byte-compiling f:\program files\python\wxPython\printfw.py to
wxPython\printfw.p
yc
byte-compiling f:\program files\python\wxPython\windows2.py to
wxPython\windows2
.pyc
byte-compiling f:\program files\python\wxPython\windows3.py to
wxPython\windows3
.pyc
byte-compiling f:\program files\python\wxPython\controls2.py to
wxPython\control
s2.pyc
byte-compiling f:\program files\python\lib\types.py to types.pyc
byte-compiling f:\program files\python\wxPython\utils.py to
wxPython\utils.pyc
byte-compiling f:\program files\python\wxPython\events.py to
wxPython\events.pyc

byte-compiling f:\program files\python\lib\imputil.py to imputil.pyc
byte-compiling f:\program files\python\wxPython\wx.py to wxPython\wx.pyc
byte-compiling f:\program files\python\wxPython\misc2.py to
wxPython\misc2.pyc
byte-compiling f:\program files\python\wxPython\sizers.py to
wxPython\sizers.pyc

byte-compiling f:\program files\python\wxPython\windows.py to
wxPython\windows.p
yc
byte-compiling f:\program files\python\lib\string.py to string.pyc
byte-compiling f:\program files\python\wxPython\controls.py to
wxPython\controls
.pyc
byte-compiling f:\program files\python\wxPython\stattool.py to
wxPython\stattool
.pyc
copying myscript.py ->
build\bdist.win32\winexe\collect\myscript\Scripts\__main_
_.py
changing into 'build\bdist.win32\winexe\collect\myscript'
zip -rq C:\temp\build\bdist.win32\winexe\myscript.zip .
creating 'C:\temp\build\bdist.win32\winexe\myscript.zip' and adding '.' to
it
changing back to 'C:\temp'
creating dist\myscript\myscript.exe
copying f:\program files\python\wxPython\wxc.pyd -> dist\myscript
copying F:\WINNT\System32\MSVCRT.dll -> dist\myscript
copying F:\WINNT\System32\python20.dll -> dist\myscript
copying f:\program files\python\wxPython\wx22_2.dll -> dist\myscript
copying F:\WINNT\System32\MSVCIRT.dll -> dist\myscript
warning: py2exe: ***************************************************************
**********
warning: py2exe: * The following modules were not found:
warning: py2exe: *   utilsc
warning: py2exe: *   miscc
warning: py2exe: *   windowsc
warning: py2exe: *   streamsc
warning: py2exe: *   gdic
warning: py2exe: *   framesc
warning: py2exe: *   sizersc
warning: py2exe: *   controls2c
warning: py2exe: *   printfwc
warning: py2exe: *   stattoolc
warning: py2exe: *   misc2c
warning: py2exe: *   controlsc
warning: py2exe: *   windows2c
warning: py2exe: *   eventsc
warning: py2exe: *   windows3c
warning: py2exe: *   clip_dndc
warning: py2exe: *   mdic
warning: py2exe: *   imagec
warning: py2exe: *   cmndlgsc
warning: py2exe: *   filesysc
warning: py2exe: ***************************************************************
**********
removing 'build\bdist.win32\winexe\collect\myscript' (and everything under
it)
removing 'build\bdist.win32\winexe' (and everything under it)





From pearu at cens.ioc.ee  Fri Mar 23 00:48:14 2001
From: pearu at cens.ioc.ee (Pearu Peterson)
Date: Fri, 23 Mar 2001 07:48:14 +0200
Subject: PEP 240 scares me
In-Reply-To: <EjAu6.13052$_46.499367@e420r-atl2.usenetserver.com>
References: <Pine.LNX.4.21.0103222338530.29360-100000@kev.ioc.ee> <99e2h61aao@news2.newsguy.com> <EjAu6.13052$_46.499367@e420r-atl2.usenetserver.com>
Message-ID: <Pine.LNX.4.21.0103230727440.31371-100000@kev.ioc.ee>

Hi!

Thanks for your replies. I now got it right what's in PEP 240.
The Proposed idea as it really is, is most surprising (not sure if 
it is good or bad thing). 
Seems that I have done too much numerics, so that I didn't realize that
typing 1.2 could then mean internal transformation rational(10*1.2,10).

Anyway, I take back my panic in my first message. 
Sorry for the confusion.

	Pearu

On Thu, 22 Mar 2001, Steve Holden wrote:

> "Alex Martelli" <aleaxit at yahoo.com> wrote in message
> news:99e2h61aao at news2.newsguy.com...
> > "Pearu Peterson" <pearu at cens.ioc.ee> wrote in message
> > news:Pine.LNX.4.21.0103222338530.29360-100000 at kev.ioc.ee...
> >     [snip]
> > > But making 1.2 to be a rational
> > >                                 1
> > >                                ---
> > >                                 2
> > > will most definitely surprise new (Python) programmers even more
> *because*
> >
> > Nobody's proposing THAT, AFAIK -- 1.2 would be the rational 6/5 (and I
> > gather the idea is that one uses 1.2e0 to get a float of similar value).
> >
> >
> Maybe Pearu complains about a typo which looks unlike an error?
> 
> regards
>  Steve
> 
> 
> 
> 



From ykingma at accessforall.nl  Wed Mar 28 16:58:59 2001
From: ykingma at accessforall.nl (Ype Kingma)
Date: Wed, 28 Mar 2001 22:58:59 +0100
Subject: Is inheritance broken?
References: <mailman.985808317.25055.python-list@python.org>
Message-ID: <3AC25E8F.33E06BB9@accessforall.nl>

Jonathan,

<snip>

> 
> If I really wanted this to work, guess one thing I could do is to cheat by
> creating a 'flat' list of classes and then manually creating my class tree.
> Something like this:
> 
> <code>
> 
> # 'flat' classes
> class parallelogram_flat:
>         def angle(self): return "variable"
>         def side(self): return "variable"
> 
> class rhombus_flat:
>         def side(self): return "same"
> 
> class rectangle_flat:
>         def angle(self): return "90 degree"
> 
> class square_flat:
>         pass
> 
> # 'real' classes
> class parallelogram(parallelogram_flat):
>         pass
> class rhombus(rhombus_flat, parallelogram_flat):
>         pass
> class rectangle(rectangle_flat, parallelogram_flat):
>         pass
> class square(square_flat, rhombus_flat, rectangle_flat, parallelogram_flat):
>         pass
> 
> # usage
> 
> s = square()
> print "squares have", s.angle(), "angles"
> print "squares have", s.side(), "sides"
> 
> </code>
> 
> Ugly and error prone, no?
>

This code follows a rule: never instantiate a class that
is inherited from. (In other languages your 'flat' classes
might be called 'abstract' in order to indicate that they cannot
be instantiated.)
IIRC this is a design pattern in class hierarchies.

In this particular case you avoid having a real square that
inherits from a real rectangle. In geometry a square
is a special case of rectangle. However this special case
from geometry is not implemented well by using OO inheritance.

The 'flat' and 'real' approach in the above class hierarchy
feels a lot better. It allows more choice of where to put
eventual attributes (eg. side length).

So, no it's not ugly at all, quite the contrary.
(Being non native in English, I don't know what a rhombus is,
so I might change my mind...)

It's also not really error prone: you only need to check
right to left, there is no need to go deeper into
the inheritance hierarchy.


Regards,
Ype Kingma


-- 
email at xs4all.nl


From othello at javanet.com  Sun Mar 18 01:56:53 2001
From: othello at javanet.com (Raymond Hettinger)
Date: Sun, 18 Mar 2001 01:56:53 -0500
Subject: PEP 42 and Py2.1
Message-ID: <3AB45C34.C8D1E82D@javanet.com>

I've checked out the Py2.1 beta and my most desired small feature didn't
make it.  To quote PEP 42:
    Generalize eval() to accept any mapping objects for locals and globals.

Is it too late to get this one in?
Is it difficult to implement?
Is there anyway I can help get it done?

Thanks,

Raymond
othello at javanet.com



From support at internetdiscovery.com  Thu Mar 22 17:08:24 2001
From: support at internetdiscovery.com (Mike Clarkson)
Date: Thu, 22 Mar 2001 22:08:24 GMT
Subject: Is setup.py a "good thing"? (long)
Message-ID: <3aba76e2.167083508@24.0.228.33>

setup.py was introduced in 2.1, presumably to help configure and
compile Python, so presumably it's a "good thing".  But I have an
uneasy feeling which I will try to explain. 

Perhaps it is just my lack of understanding of setup.py, or the
multitude of issues involved in building Python. If so, I apologize in
advance, and I welcome any light people can shed on the subject.

I know about configure and I know about make, and but I don't know
about setup.py - it's purpose is not clearly defined anywhere I can
see, so I worry that it's one of those things everyone can agree is a
"good thing" only because no one defined it's role, so it couldn't be
examined too closely. Because after all, the overlap between configure
and setup.py is significant, and that makes me uneasy.

It could be that setup.py is the first step in a programme for Python
to rule the world by gradually replacing configure. If so, I'm uneasy
because life is too short. Much as I hate configure (with a passion),
it's better to join it than to try and beat it. So I worry that
setup.py is duplicating what should really be in configure.

I believe a part of the reason for coming up with setup.py was to
facilitate the building of dynamically loaded extensions to Python,
which is a difficult problem for everyone: Tcl and Apache are two
other pieces of software that have faced this issue.

I'll frame my discussion in terms of adding a Tk extension to Tkinter,
which is what I was trying to do. I believe that the issues are not
Tkinter specific, and it serves as a concrete example of issues with
setup.py, and the path it leads us down. Or perhaps much of my
uneasiness is the way setup.py specifically deals with _tkinter.

Take for example compiling _tkinter, which needs to know where the tcl
libraries are. The cannonical way of doing this in Tcl would be to add
a --with-tcl configure option which lets you tell configure where the
Tcl files might be. Many modern Tcl programs use --with-tcl-config -
Tcl realized it was not just the locations of the includes and
libraries you needed to store, but also other compile settings like
--enable-shared  and --enable-stubs, because you must be coherent with
these settings too. So every tcl install generates, and every tcl
binary distribution comes with, a file tclConfig.sh which stores a
bunch of these things. Then extension configure scripts can be written
to look for these files (or be told with --with-tcl-config), and then
source them into its configure so the extension will be coherent.

setup.py uses a simplistic find_file which tries to find if the tcl
and tk libraries and include files are around. If it finds one, it
adds the include directory and library to the equivalent of
Modules/Setup and compiles _tkinter to be dynamically loading.
 
Problems:
#1)  It's black magic. I don't know where it looked or what it found.
#2) It's closed - I can't tell it where libtcl is unless I have the
courage and temerity to go hack on it (after all, maybe G wrote it!)
#3)  The algorithm is used to find things like libraries is not the
same as the one the OS will use to resolve dynamic linkages (ldd).
setup.py doesn't look at LD_LIBRARY_PATH or ld.so.cache for example.
#3b) The same is true of where Python will look to load .so files,
which I believe is partially determined by the linker -R flag, which
is set by configure, not setup.py.

Maybe you can see where I might be getting uneasy. 

Now assuming someone graciously explains to me all of the above, 
I'm still going to left feeling a little uneasy because:
#3c) The same is true of where *Tcl* will look to load .so files,
which I believe is partially determined by the linker -R flag, and
probably things like env(TCL_LIBRARY) and hence tclpkgPath.
#4) If setup.py finds a Tk extension library, it will add the WITH_FOO
flag to the compile, which will compile the extension statically, even
though the whole idea of setup.py was to facilitate the building of
dynamically loaded extensions to Python (this may just be a bug).
#5) setup.py is not dealing with one of the more important configure
options in Tcl --enable-stubs *with which is must be coherent.*
(BTW, I think Python should switch to the Tcl stubs API ASAP.)
#6) I'll have to foresake all of my setup.py hacks when I go back and
build again from scratch a static python to use freeze.py, for which I
must not use setup.py, but Modules/Setup instead. (It's problem
generic to any freezing, not just freeze.py in particular.)

Now all of this gets me thinking that the real reason was the setup.py
in the first place was the Tcl equivalent to the configure options
--enable-shared/--disable-shared, which at least makes it not black
magic because I know about 'configure --help' anyway. So I think I'd
feel less uneasy if:
#1) Someone documented what setup.py is supposed to do, and how it is
supposed to interact with configure and make.
#2) Python configure had the  configure options --enable-shared
--disable-shared so I could say explicitly what I want in the way of
static/dynamic. setup.py could then restrain itself to just looking in
the config.* files for anything it needs.  (See Apache for something
more elaborate on this point: configure --activate-module
--enable-module --disable-shared  unsw.).
#3) Python configure had the  configure options --with-tcl-config and
--with-tk-config so I could say explicitly where what I want is. This
allows Python configure to verify it is consistent with the Tcl
configure that was used to build the libraries I will be linking
against. This is relatively easy to do these days as almost all Tk
extensions have this, so that anyone who is good at configure (an
oxymoron?) can cut and paste one into the Python configure.
#4) While the oxymoron was at it, he/she/it could  cut and paste in
the section for --enable-stubs, which would be a "good thing".

Maybe I can ask the people who have a lot of experience with this
problem in Tcl like Jeff Hobbes and Cameron Laird to comment,
and hopefully lay my fears to rest.

Yours uneasi-ly,

Mike


From paul at fxtech.com  Sat Mar 17 09:53:35 2001
From: paul at fxtech.com (Paul T. Miller)
Date: Sat, 17 Mar 2001 08:53:35 -0600
Subject: embedding a Python debugger in an app?
Message-ID: <3AB37A6F.1AF538F2@fxtech.com>

Is it convenient to embed a Python debug in a C/C++ app that extends and
embeds Python as a scripting language?
Are there any examples that show how to do this?


From aahz at panix.com  Mon Mar 19 13:21:25 2001
From: aahz at panix.com (Aahz Maruch)
Date: 19 Mar 2001 10:21:25 -0800
Subject: What IDE are you'all using for Python 2.1 for Windows?
References: <995gai$2bv$1@news.chatlink.com>
Message-ID: <995in5$22s$1@panix2.panix.com>

In article <995gai$2bv$1 at news.chatlink.com>,
Luis Cortes <lcortes at aodinc.com> wrote:
>
>  Just wanted to know what "good" ide's are out there for python.  Please
>let me know which ones you prefer.

vim
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

There's a difference between a person who gets shit zie doesn't deserve
and a person who gets more shit than zie deserves.  --Aahz


From jarober at mail.com  Tue Mar 20 08:07:35 2001
From: jarober at mail.com (James A. Robertson)
Date: Tue, 20 Mar 2001 13:07:35 GMT
Subject: Collection interfaces
References: <3a90fa97.0@news1.mweb.co.za>   <MPG.1503548620e632919896a4@news.earthlink.net> <MCjm6.474652$U46.14163288@news1.sttls1.wa.home.com> <MPG.150394195f7f04a19896a7@news.earthlink.net> <3A9A86B1.9A8C69A2@ix.netcom.com> <MPG.1504d14563dce6499896af@news.earthlink.net> <97fjla128ir@news2.newsguy.com> <MPG.1505a76831c34b749896b3@news.earthlink.net> <97h5s40vhc@news2.newsguy.com> <MPG.1506007591cf818b9896bc@news.earthlink.net> <slrn99opop.gr.amk@mira.erols.com> <MPG.1506265ebf6d934d9896be@news.earthlink.net> <97ibmc0k47@news1.newsguy.com> <MPG.1506cf3293b094b49896c1@news.earthlink.net> <97m0c7080c@news2.newsguy.com> <MPG.15085517debf4b649896ca@news.earthlink.net> <zb%n6.31033$o%6.586026@news1.rdc1.bc.home.com> <MPG.150b383cea226d409896d0@news.earthlink.net> <kNko6.6986$eq3.707121@news.uswest.net> <MPG.150c46db9cec283e9896d4@news.earthlink.net> <3AA2A898.450B659E@mail.com> <gRCt6.12224$Im6.1308560@newsread1.prod.itd.earthlink.net>
Message-ID: <3AB755F8.67E84126@mail.com>

topmind wrote:
> 
> "James A. Robertson" <jarober at mail.com> wrote in message
> news:3AA2A898.450B659E at mail.com...
> > Topmind wrote:
> > >
> >
> > <snip>
> >
> > >
> > > Not exactly "typical custom business applications", the domain
> > > in which I focus most of my whining.
> > >
> > > The percentage of OO fans in the business domain seems lower
> > > than average. Perhaps this says something about OO and
> > > the business domain.
> > >
> >
> > 90% + of the customers using VisualWorks are writing 'typical' business
> > apps.
> >
> 
> Then why are most "see what OO can do" examples strangely non-business?

They are trying for a 'simplicity' I disagree with.  In any case, the
tutorials on our website use a straightforward example - parsing a log
left by website tools.

http://www.cincom.com/smalltalk

> 
> > [snip]
> 
> -tmind-
> oop.ismad.com

-- 
James A. Robertson
Product Manager (Smalltalk), Cincom
jarober at mail.com
<Talk Small and Carry a Big Class Library>


From hamish_lawson at yahoo.co.uk  Wed Mar  7 03:58:01 2001
From: hamish_lawson at yahoo.co.uk (=?iso-8859-1?q?Hamish=20Lawson?=)
Date: Wed, 7 Mar 2001 08:58:01 +0000 (GMT)
Subject: Ingres module?
Message-ID: <20010307085801.14290.qmail@web208.mail.yahoo.com>

Brian E Gallew wrote:

> I noticed (from searching the web site) that Jim Fulton had at one
> time talked about an Ingres module.  Has anyone actually seen such a
> beast?  Even parts of one?  I'm not actually looking forward to
> writing this myself, though I guess I will if I have to.

A module written by Holger Meyer can be downloaded from

http://freezope.nipltd.net/HamishLawson/PythonModules/ingmod/ingmod.tar.gz

Hamish Lawson
hamish_lawson at yahoo.co.uk


____________________________________________________________
Do You Yahoo!?
Get your free @yahoo.co.uk address at http://mail.yahoo.co.uk
or your free @yahoo.ie address at http://mail.yahoo.ie



From nick at olsen.ch  Wed Mar  7 03:20:43 2001
From: nick at olsen.ch (Nick Jacobs)
Date: Wed, 07 Mar 2001 08:20:43 GMT
Subject: Can anyone recomend a good intoduction to C...
References: <mailman.983894288.19965.python-list@python.org>
Message-ID: <3AA5EF59.507F1D91@olsen.ch>

Laura Creighton wrote:

> There are a very large number of people who think that C
> is a terrific language; extremely good at what it does, and well worth
> studying, in part to understand what it is that a language designer
> does when he *gets* *it* *right*, while also believing that studying C++ is
> of particular use when you want to see what happens when a language
> designer *gets* *it* *wrong*.  Or when forced to do so at gunpoint.

I don't think C deserves all the superlatives you apply to it, but I agree that
it's instructive to study C++ as an example of a language that gets it wrong.
Take a good language, C; add some good ideas; and end up with a
baroque mess. Fascinating.




From chrisl at gnuchina.org  Fri Mar 23 12:40:37 2001
From: chrisl at gnuchina.org (Christopher Li)
Date: Fri, 23 Mar 2001 09:40:37 -0800 (PST)
Subject: LONG_BIT problem
In-Reply-To: <3ABA93B8.8088F22D@sympatico.ca>
Message-ID: <Pine.LNX.4.30.0103230933300.11741-100000@chris.cn.mvd>

On Thu, 22 Mar 2001, Gil Hauer wrote:

> Hi,
>
> I've having a problem building Python 2.0 on RedHat 7.0.
>
[...]
>
> glibc versions are:
> 	glibc-2.2-12
> 	glibc-devel-2.1.92-14
^^^^^^^^^^^^^^^^^^^^^^^Here. Please upgrade to glibc-2.2-12.
If you grep the user group or mail list vs groups.google.com you will find
many answer like this.

BTW, "ldd" is missing from glibc-2.2-12. I am using glibc-2.2-5
with glibc-devel-2.2-12 seems fine so far.

Chris




From tim.one at home.com  Sat Mar  3 04:08:42 2001
From: tim.one at home.com (Tim Peters)
Date: Sat, 3 Mar 2001 04:08:42 -0500
Subject: A counter-proposal to __future__ in PEP 236
In-Reply-To: <3AA08F31.55CB6ADB@destiny.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCCEKDJDAA.tim.one@home.com>

Very quickly:

[Michael Chermside]
> ...
> It is *NOT* an actual import statement,

Not so.  Please read the PEP.

> and it does *NOT* scream "black-magic" -- just looks like another of
> those special "built-in" things; you know... the functions __cmp__ and
> __import__ and the modules __builtin__ and  __future__.

Yes, but names with double double-underscores are Python's way of saying
"black magic -- you're interfacing with language internals here".

> ...
> Through the creation of an (empty) "__future__" module, it becomes
> possible to write code that compiles and and runs fine in pre-2.1
> versions of python,

No.  Because it *is* an import statement, pre-2.1 versions of Python will
choke at runtime on the failure to find __future__.py, or the failure to
import the specific feature(s) named.  This is all explained in the PEP, and
indeed more than once.  BTW, double double-underscore names are reserved for
the language's use, so you've got no business creating your own __future__.py
anyway.

can-make-it-foolproof-but-not-damned-fool-proof<wink>-ly y'rs  - tim




From sholden at holdenweb.com  Wed Mar 28 12:29:03 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Wed, 28 Mar 2001 12:29:03 -0500
Subject: Python assembler
References: <3ac20dcc$0$15028$cc9e4d1f@news.dial.pipex.com>
Message-ID: <dfpw6.5687$m04.422805@e420r-atl1.usenetserver.com>

"Eddie and Babs" <kca17 at dial.pipex.com> wrote in message
news:3ac20dcc$0$15028$cc9e4d1f at news.dial.pipex.com...
> The Python library reference says of the 'dis' module: "Since there is no
> Python assembler, this module defines the Python assembly language."
>
> Is the lack of a Python bytecode assembler a deliberate policy decision or
> is it just that nobody has written one yet? Or have many tried, but none
> lived to tell the tale?
>
> I would consider such an assembler if there were no obvious reasons for it
> not to exist.

Probably the main reason is that to assemble the bytecodes you would
probably end up converting most of them into calls on the interpreter API,
and most of the interpreter would have to be loaded to support everything
one might possibly do in Python (eval, exec, etc.).

Since the interpreter's inner loop (fetch a byte code, execute it) is fairly
tightly coded, there would be little advantage to compiling.

regards
 Steve





From loredo at spacenet.tn.cornell.edu  Thu Mar 15 15:33:09 2001
From: loredo at spacenet.tn.cornell.edu (Tom Loredo)
Date: Thu, 15 Mar 2001 15:33:09 -0500
Subject: reading binary data
References: <3AAFE69D.FF01FEC8@yahoo.fr>
Message-ID: <3AB12705.5F53402C@spacenet.tn.cornell.edu>

Oops!  Sorry for posting that old announcement---I just noticed that an
announcement for a new version of Snack appeared just today!  Look for
it in another thread.

-Tom Loredo


From piet at cs.uu.nl  Wed Mar 21 07:04:28 2001
From: piet at cs.uu.nl (piet at cs.uu.nl)
Date: 21 Mar 2001 13:04:28 +0100
Subject: reversing a dictionary (newbie)
References: <mailman.984650604.21153.python-list@python.org>
Message-ID: <ud7bbjpkj.fsf@cs.uu.nl>

>>>>> "Christopher Brewster" <C.Brewster at dcs.shef.ac.uk> (CB) writes:

CB> def revdict(diction):
CB> 	revdict = {}
CB> 	for wordpair in diction.keys():
CB> 		#print wordpair
CB> 		list = []
Why start with an empty list? See below
CB> 		freq = diction[wordpair]
CB> 		if revdict.has_key(freq):
CB> 			list.extend(revdict[freq])
                        The list was empty, so list=revdict[freq] would be faster
CB> 			list.append(wordpair)
CB> 			revdict[freq] = list
                        But of course revdict[freq].append(wordpair) would
                        be even better instead of the 3 lines above.
CB> 		else:
CB> 			list.append(wordpair)
                        Here the same. list=[wordpair] would be the same
CB> 			revdict[freq] = list
                        But then revdict[freq]=[wordpair] would be the fastest.
CB> 	return revdict

-- 
Piet van Oostrum <piet at cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: P.van.Oostrum at hccnet.nl


From dkuhlman at rexx.com  Fri Mar  2 12:43:33 2001
From: dkuhlman at rexx.com (Dave Kuhlman)
Date: Fri, 02 Mar 2001 17:43:33 -0000
Subject: Editing Python source as XML?
References: <mailman.983546704.10789.python-list@python.org> <j44rxcjgnw.fsf@informatik.hu-berlin.de>
Message-ID: <983555015.38318@rexx.com>

SciTE also does folding of Python code.  The editors in both SciTE
and PythonWin are based on Scintilla.  You can find a link to SciTE
at:

   http://www.scintilla.org/

The following is not what you asked for but is related and you
might be interested.  SciTE enables you to "export to HTML", and
the exported code, when viewed in a Web browser has highlighting of
various kinds.

  - Dave


Martin von Loewis <loewis at informatik.hu-berlin.de> wrote:
> 
> Steven Haryanto <steven at haryan.to> writes:
> 
>> A recent experience of having to modify a long C source file has
>> led me to wonder whether there are any tool/effort towards editing
>> Python source code as XML or other form of structured document.
> 
> Not sure *why* you'd want to do that, but it is certainly possible to
> convert any script automatically to an XML document with the help of
> the parser module.
> 
>> What I would like is an editor that presents the source file as a
>> tree outline, where I can collapse selected class or function
>> definition body, certain long if's/for's, some section of code,
>> view/hide the asserts, etc. 
> 
> PythonWin, and the Emacs Python mode, both can do that today.
> 
> Regards,
> Martin

-- 
Dave Kuhlman
dkuhlman at rexx.com


From ljohnson at resgen.com  Tue Mar 27 11:16:06 2001
From: ljohnson at resgen.com (Lyle Johnson)
Date: Tue, 27 Mar 2001 10:16:06 -0600
Subject: stripping cr/lf, lf, cr
References: <Lp2w6.705$p5.3213@news1.rivrw1.nsw.optushome.com.au>
Message-ID: <tc1f21rq8tdud2@corp.supernews.com>

> Is there a platform independant function that will remove appropriate EOL
> characters from a string?  I don't want to use strip() since it will also
> remove other whitespace I want to keep.

The standard Python source code distribution includes two scripts:

    Tools/scripts/lfcr.py
    Tools/scripts/crlf.py

that convert back and forth between DOS and Unix line endings. I don't know
if there are *specific* library functions dedicated to these tasks but it
shouldn't be much of a stretch to mine these scripts for what you need.




From aon.912502367 at aon.at  Sun Mar 11 06:57:35 2001
From: aon.912502367 at aon.at (Gregor Lingl)
Date: Sun, 11 Mar 2001 12:57:35 +0100
Subject: true newbie
References: <3aa9a9e9$0$65277$38c90ffe@news1.twtelecom.net>
Message-ID: <3AAB682F.CB804361@aon.at>


.. and if you have learned just a bit, maybe, you would like to contribute
to

 http://www.lowerstandard.com/python/pythonsource.html

Gregor Lingl



From aleaxit at yahoo.com  Fri Mar 30 07:30:05 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 30 Mar 2001 14:30:05 +0200
Subject: diff lists
References: <mailman.985914639.5108.python-list@python.org>
Message-ID: <9a1u8m0306u@news1.newsguy.com>

"Steven D. Majewski" <sdm7g at Virginia.EDU> wrote in message
news:mailman.985914639.5108.python-list at python.org...
    [snip]
> On 29 Mar 2001, Michael Hudson wrote:
> > Tangentially, does anyone know of any good algorithms for "edit
> > distance" between two sequences?  E.g. if I have
> > "abcdef"
> > and want to get to
> > "abQUACKcde"
> > I want to get the answer back "insert 'QUACK' at position 3 and delete
> > a character at position 11".
> > "Good" means "pretty quick", here.
    [snip]
> http://www.ling.ohio-state.edu/~cbrew/795M/string-distance.html
>
> However, the algorithm for the distance does not give location, but
> maybe this info can be extracted from the intermediate matrix.

Thanks Steve for the good pointers -- the latter in particular
(by Chris Brew) is expressed in pretty-clear Pythonish pseudocode
(although incomplete and suffering a typo or two) so it was not
hard to transliterate -- here are both versions, distance only
and full-path.  I've CC'd Chris (as credit for using his page:-)

[Chris, the incompleteness is on the second half of what I
called edit_path, and the typo is in your line which goes
     d(i+1,0) = (sofar + delCost,DEL,d(0,j)) (**)
where it should be
     d(i+1,0) = (sofar + delCost,DEL,d(i,o)) (**)
but, mainly, thanks for that page!-)].


def atomDistance(char1, char2, substCost=1.0):
    if char1==char2: return 0
    return substCost

def edit_dist(string1, string2, delCost=1.0, insCost=1.0,
atomDistance=atomDistance):
    m = len(string1)
    n = len(string2)
    d = [ [0.0]*(n+1) for i in range(m+1) ]
    for i in range(m):
        d[i+1][0] = d[i][0] + delCost
    for j in range(n):
        d[0][j+1] = d[0][j] + insCost
    for i in range(m):
        for j in range(n):
            subst = atomDistance(string1[i], string2[j])
            d[i+1][j+1] = min(d[i][j]+subst,
                              d[i+1][j]+insCost,
                              d[i][j+1]+delCost)
    return d[m][n]

def edit_path(string1, string2, delCost=1.0, insCost=1.0,
atomDistance=atomDistance):
    m = len(string1)
    n = len(string2)
    d = [ [(0.0,'Start',None)]*(n+1) for i in range(m+1) ]
    for i in range(m):
        cost_so_far, move, tb = d[i][0]
        d[i+1][0] = cost_so_far+delCost, 'Del(%s)'%string1[i], d[i][0]
    for j in range(n):
        cost_so_far, move, tb = d[0][j]
        d[0][j+1] = cost_so_far+insCost, 'Ins(%s)'%string2[j], d[0][j]
    for i in range(m):
        for j in range(n):

            subst = atomDistance(string1[i], string2[j])
            if string1[i]==string2[j]:
                submove = 'Ok(%s)'%string1[i]
            else:
                submove = 'Sub(%s->%s)'%(string1[i],string2[j])
            cost_so_far, move, tb = d[i][j]
            cost_with_sub = cost_so_far + subst

            cost_so_far, move, tb = d[i+1][j]
            cost_with_ins = cost_so_far + insCost
            insmove = 'Ins(%s)'%string2[j]

            cost_so_far, move, tb = d[i][j+1]
            cost_with_del = cost_so_far + delCost
            delmove = 'Del(%s)'%string1[i]

            if cost_with_sub < cost_with_ins:
                if cost_with_sub < cost_with_del:
                    d[i+1][j+1] = cost_with_sub, submove, d[i][j]
                else:
                    d[i+1][j+1] = cost_with_del, delmove, d[i][j+1]
            else:
                if cost_with_ins < cost_with_del:
                    d[i+1][j+1] = cost_with_ins, insmove, d[i+1][j]
                else:
                    d[i+1][j+1] = cost_with_del, delmove, d[i][j+1]
    return d[m][n]

def unravel(chain):
    cost, move, previous = chain
    if previous is None:
        return [move]
    else:
        return unravel(previous) + [move]

def showedit(s1, s2):
    print "%s -> %s: %d" % (s1, s2, edit_dist(s1,s2))
    path = edit_path(s1, s2)
    print path
    print unravel(path)

if __name__=='__main__':
    import sys
    if len(sys.argv)==3:
        showedit(sys.argv[1],sys.argv[2])
    else:
        showedit("practical","rational")


Here is sample interactive use for Michel's question:

D:\pybridge>python
Python 2.0 (#8, Oct 16 2000, 17:27:58) [MSC 32 bit (Intel)] on win32
Type "copyright", "credits" or "license" for more information.
>>> import ed
>>> edit = ed.unravel(ed.sdistr('abcdef','abQUACKcde'))
>>> for move in edit:
...   print move
...
Start
Ok(a)
Ok(b)
Ins(Q)
Ins(U)
Ins(A)
Ins(C)
Ins(K)
Ok(c)
Ok(d)
Ok(e)
Del(f)
>>>


There are of course lots of little tweaks one could do to speed
this up in practice (the linked-list form of the 'chain' is neat
but probably costly, and it needs to be unraveled anyway -- and
unravel() itself might be non-recursive -- etc, etc) but I do
believe this version is quite readable and didactically valid.


Alex





From dsh8290 at rit.edu  Fri Mar 16 18:41:45 2001
From: dsh8290 at rit.edu (D-Man)
Date: Fri, 16 Mar 2001 18:41:45 -0500
Subject: Python style questions
In-Reply-To: <bess6.179$sk3.52516@newsb.telia.net>; from fredrik@pythonware.com on Fri, Mar 16, 2001 at 05:40:23PM +0000
References: <98tb91$9t9$1@saltmine.radix.net> <mailman.984759577.27755.python-list@python.org> <bess6.179$sk3.52516@newsb.telia.net>
Message-ID: <20010316184145.A16851@harmony.cs.rit.edu>

On Fri, Mar 16, 2001 at 05:40:23PM +0000, Fredrik Lundh wrote:
| "D-Man" <dsh8290 at rit.edu> wrote:
| > Curiosly enough, print's format string is supposed to be identical to
| > C's printf format string.  However the man page for printf says the %x
| > prints 0x then the number using A-F for 10-15.
| 
| better check that page again; printf should add 0x only if you
| use the # flag.

I missed that before.  Thanks.  (It's a pretty long page, and I only
wanted 1 piece of information ..., should always read carefully :-))

-D




From fredrik at pythonware.com  Sun Mar 18 04:02:42 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Sun, 18 Mar 2001 09:02:42 GMT
Subject: How to get a frame's size
References: <906782E21jshanklspeakeasyorg@207.126.101.100>
Message-ID: <SQ_s6.1301$4N4.160600@newsc.telia.net>

John Shankland wrote:
> I have a frame called IndexCard1 with some widgets packed in it. I want to
> make several of these and put them in a grid. I need to know how many I can
> fit on the screen, so I need the size of the frame. I'm able to get the
> dimensions of the screen okay, but IndexCard1.winfo_width() seems to always
> return 1. As do all of the other size related methods I could find. So my
> question is how do I get the size in pixels of IndexCard1?

winfo_width() returns the current managed size; it's updated
when Tk gets around to do geometry management.  you can
use update_idletasks() to force geometry management.

winfo_reqwidth() returns the current requested size.

Cheers /F




From pbrian at demon.net  Mon Mar 19 09:38:55 2001
From: pbrian at demon.net (Paul Brian)
Date: Mon, 19 Mar 2001 14:38:55 -0000
Subject: newbie - infinite loop 
Message-ID: <985012513.18971.0.nnrp-12.c1c3e154@news.demon.co.uk>

All,

I have solved the opriginal problem but am left with a serious question.

Before I discovered the keywords in and not in, I tried to count unique
instances of names in a file.
I devcided to create a (neaarly) empty list, compare each name in the file
with that list. If the name was not in, append to list, otherwise leave it
alone.

After the following was run (and crashed) a couple of times I started using
my brain and realised there must be some "in" operator. However the problem
still stands - a for condition seems completely mutable.

So, line 3 sets a condition, 4 compares "q" to "q" and (wrongly for my
needs) adds "q" to the end of the list.
But then it seems that line 3 realises it has another element in its list
and goes back and compares
the newly appended "q" to my lineand so on ad infinitum (I have to kill the
process each time) !

Surely this should not happen.  Surely the line 3 condition "for every
element in this list", should not constantly expand but should be set solid
till the next pass?

What have I done wrong - is it the mutable list v the immutable tuple
problem (but it should work..surely?) Or is this just the way things are -
"live with it and do not try anything like this again?"

thanks in advance
==============
uniqueList = ['q']               # create an nearl empyty list (seems to
ignore line 3 if truly empty)
line = 'q'                          # fake the retrieval of line from file

for listElement in uniqueList:             #3
    if listElement == line:                    #4
        uniqueList.append(line)            #5 at this point it seems to
think that there is now another list
                                                        #element to compare
and so loops for ever...help!

print len(uniqueList)
==============




From stevewilliams at wwc.com  Thu Mar 15 10:36:18 2001
From: stevewilliams at wwc.com (Steve Williams)
Date: Thu, 15 Mar 2001 15:36:18 GMT
Subject: news from python9!!!
References: <mailman.984074414.26983.python-list@python.org> <uJRp6.46512$__6.10501931@typhoon.southeast.rr.com> <mailman.984638183.20040.python-list@python.org>
Message-ID: <3AB0E18B.113BB8A0@wwc.com>

Sean Reifschneider wrote:

> On Thu, Mar 08, 2001 at 08:13:14PM +0000, Don Tuttle wrote:
> >Hey, you gotta run with what you can find, seeing how all you conventioneers
> >where having way too much fun to drop into the newsgroup and give us an
> >update! <wink>
>
> That was just because they didn't have *REAL* wireless networking at the
> show...  ;-)  For my release of the 2.1b1 RPMs, I built them in the bar
> while hanging out, then went to the terminal room to schlep them up to
> the server and send out the announcement...
>
> Sean

[chop]

128kb Ricochet wireless modems work just fine in the Los Angeles/Orange County
area--sometimes better than Verizon DSL, IMNSHO.

This is one of the most closely held secrets in the world.  Metricom's marketing
is even worse than Hewlett-Packard's, if that's possible.




From not.this at seebelow.org  Fri Mar 16 23:51:42 2001
From: not.this at seebelow.org (Grant Griffin)
Date: Fri, 16 Mar 2001 22:51:42 -0600
Subject: Pep 238 Why new // operator (Non-integer Division)
References: <mailman.984788150.18626.python-list@python.org>
Message-ID: <3AB2ED5E.A9B44B1E@seebelow.org>

Tim Peters wrote:
> 
> [Gordon Williams]
> > Why do we need a new // operator to do the divide operation.  If
> > we want to do an integer divide with two reals why wont people just
> > use int(a/b).  For
> 
> 1. int() truncates.  Current i/j returns the floor.
> 
> 2. There is no bound on the sizes of i and j; it's very easy to
>    get Python long ints that don't fit in a Python float.

BTW, I guess I'm not the first one to notice that this new operator
looks an awful lot like a C/C++ comment.  So maybe some other symbol
should be used.

Hey, isn't "$" still open? <wink>

just-tryin'-to-make-a-buck-ly y'rs,

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation            http://www.iowegian.com


From tim.one at home.com  Mon Mar 19 20:27:35 2001
From: tim.one at home.com (Tim Peters)
Date: Mon, 19 Mar 2001 20:27:35 -0500
Subject: Why "from __future__" stinks; a counter-offer
In-Reply-To: <slrn9bcg72.9q8.scarblac@pino.selwerd.nl>
Message-ID: <LNBBLJKPBEHFEDALKOLCMEMCJGAA.tim.one@home.com>

[Remco Gerlich]
> And the place to complain [about PEPs] is the Python-DEV list, which
> is not public, although everyone can post to it and everyone can read it
> using some archives...

Preference varies by PEP author.  Guido would generally rather keep it *off*
Python-Dev, and read the PEPs after they settle down.  Setting up a mailing
list on SourceForge is an alternative, currently being used to hash out
numeric-related PEPs:

    http://sourceforge.net/mail/?group_id=5470

I prefer to slug things out on c.l.py, like in "the good old days" --
although I'm not sure it's possible to carry out an extended rational debate
on c.l.py anymore without enormous duplication of effort (one guess as to why
Python-Dev got created <0.1 wink>).

> And counterproposals need their own PEP unless it's a minor change
> in the proposal.

Depends on the PEP and the author again, although as a general rule I expect
that's the best way to go.  The gross difficulty with trying to give "equal
treatment" to competing ideas in a single PEP is that at most one can win.
At that time, the losing proposals are supposed to be preserved in the
"rejected PEPs" section, so that people who missed the fight the first time
around can just read about it instead of burying c.l.py under endless fresh
rounds of repetition.  Speaking of which, somebody quick write a "case
statement" PEP so Guido can reject it once and for all <0.8 wink> ...




From tjg at exceptionalminds.com  Thu Mar  1 11:27:31 2001
From: tjg at exceptionalminds.com (Timothy Grant)
Date: Thu, 1 Mar 2001 08:27:31 -0800
Subject: long int behaviour
In-Reply-To: <LNBBLJKPBEHFEDALKOLCCEPDJCAA.tim.one@home.com>; from tim.one@home.com on Thu, Mar 01, 2001 at 12:09:25AM -0500
References: <20010228172937.L14262@trufflehunter.avalongroup.net> <LNBBLJKPBEHFEDALKOLCCEPDJCAA.tim.one@home.com>
Message-ID: <20010301082731.S14262@trufflehunter.avalongroup.net>

Thanks Tim,

Yours is the second excellent explanation of what's going on
and I appreciate it very much. I don't necessarily believe that
ints are 32 bits long, however, IP addresses definitely are,
and IP addressing has been my first need to play with longs,
hence the 32 bitedness of my question.

All of my math was working out correctly regardless of the
representation I was seeing, it was just that when it came to
debugging in my head it was much easier to do when I saw what I
expected to see.



On Thu, Mar 01, 2001 at 12:09:25AM -0500, Tim Peters wrote:
> [Timothy Grant]
> > ...
> > Given the following experimenting...
> >
> > >>> hex(4294965248L)
> > '0xFFFFF800L'
> > >>> hex(~4294965248L)
> > '-0xFFFFF801L'
> > >>>
> >
> > The first output is what I would expect, however, I would have
> > expected the second output to be
> >
> > 0x000007FFL
> >
> > It appears to be something to do with two's complement storage,
> > but I'm at a loss as to why my expectations were so incorrect.
> 
> Take heart!  Your expectations weren't born incorrect, they were made
> incorrect, by a world of evil languages promulgating their blasphemous
> treachery that integers are finite <wink>.
> 
> Python does store longs in 2's-comp form, but remember that Python longs are
> unbounded:  a long less than 0 has a (conceptually) infinite string of
> leading 1 bits, just as a long >= 0 has an infinite string of leading 0 bits.
> It's easy to show you an infinite string of leading 0 bits:  we just ignore
> them.  That infinite string of leading 1 bits isn't so easy to hide, though!
> 
> You've been further warped into believing that 32 bits is somehow a "natural"
> size for an int.  If that's what you believe, you can force the issue:
> 
> >>> hex(~4294965248L & (2L**32 - 1))
> '0x7FFL'
> >>>
> 
> That is, by masking your input with a string of 32 low-order 1-bits, the
> infinite string of leading 1 bits goes away, leaving the output you expected.
> Or, if you're lucky enough to believe that ints have 101 bits, similarly:
> 
> >>> hex(~4294965248L & (2L**101 - 1))
> '0x1FFFFFFFFFFFFFFFFF000007FFL'
> >>>
> 
> So that's the trick:  to represent the infinite by the finite, you either
> need a convention, or you need to throw an infinite amount of info away.  The
> "-" in '-0xFFFFF801L' is Python's convention for compressing the unbounded
> string of sign bits in such a way that no info is lost:
> 
> >>> eval(hex(~4294965248L)) == ~4294965248L
> 1
> >>>
> 
> next-step:-infinite-memory-ly y'rs  - tim
> 
> 
> -- 
> http://mail.python.org/mailman/listinfo/python-list

-- 
Stand Fast,
    tjg.

Timothy Grant                         tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.         <><       (503) 246-3630
>>>>>>>>>>>>>Linux, because rebooting is *NOT* normal<<<<<<<<<
>>>>This machine was last rebooted:  43 days 20:37 hours ago<<



From abrahams at mediaone.net  Fri Mar  9 18:40:50 2001
From: abrahams at mediaone.net (David Abrahams)
Date: Fri, 09 Mar 2001 23:40:50 GMT
Subject: [ANN] Boost.Python - C++ binding library updated
Message-ID: <6Sdq6.1786$G76.2947435@typhoon.ne.mediaone.net>

Revamped documentation and examples
Better Python 2.0 compatibility
New checks for pickling pitfalls thanks to Ralf Grosse-Kunstleve
NULL-Pointers and smart pointers are automatically converted to/from None

See www.boost.org/libs/python/doc/index.html for documentation
The boost libraries can be downloaded at www.boost.org





From sholden at holdenweb.com  Tue Mar  6 19:27:23 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Tue, 6 Mar 2001 19:27:23 -0500
Subject: Overriding UserList methods
References: <b566atck0ruu4tae8n1srrkntbmme26gor@4ax.com> <aUFo6.14917$ML1.731870@e420r-atl2.usenetserver.com> <qqu9atsitacnuacu29k25rvtrieofeg6nk@4ax.com>
Message-ID: <vjfp6.29406$1D5.1255147@e420r-atl1.usenetserver.com>

"Daniel Klein" <danielk at aracnet.com> wrote in message
news:qqu9atsitacnuacu29k25rvtrieofeg6nk at 4ax.com...
> On Mon, 5 Mar 2001 00:52:40 -0500, "Steve Holden" <sholden at holdenweb.com>
> wrote:
>
> >"Daniel Klein" <danielk at aracnet.com> wrote in message
> >news:b566atck0ruu4tae8n1srrkntbmme26gor at 4ax.com...
> >> I must be missing something, cos I can't figure out how to override a
> >method
> >> using the UserList module. Here is some code:
> >>
> >> import UserList
> >> class MyList(UserList.UserList):
> >>       __def __init__(self):
> >>         UserList.UserList.__init__(self)
> >>
> >While this would work, I don't see it's strictly necessary. Why not just
> >inherit the __init__ method from UserList?
>
> One of the reasons for subclassing UserList is to provide additional
instance
> variables. So I need my own __init__ method to initialize this, don't I?
>
Only if the instance variables have to be around from instance creation
onwards. They might otherwise be created by other methods. Perhaps you just
abbreviated your code?

[ ... ]

>>> class MyUL(UserList.UserList):
...  def addvar(self, val):
...   self.var = val
...  def showvar(self):
...   print self.var
...
>>> ul = MyUL([1,2,3,4,5])
>>> ul.addvar(42)
>>> ul.showvar()
42
>>> ul
[1, 2, 3, 4, 5]
>>>

regards
 Steve





From NoSpam at NoSpam.com  Tue Mar 27 12:40:24 2001
From: NoSpam at NoSpam.com (Tom)
Date: Tue, 27 Mar 2001 17:40:24 GMT
Subject: C++ and Python
References: <99q99l$g21$1@news.clinet.fi>
Message-ID: <cg4w6.91071$A6.20812039@news4.rdc1.on.home.com>

"Ahto" <ahto at jussila.as> wrote in message
news:99q99l$g21$1 at news.clinet.fi...
> I'm very new to Python and haven't quite figured out how to use c++ and
> python together on windows...if someone could send me an example program
of
> tutorial webpage where this is explained?!?
>
> Thanks

There is stuff in the docs (available from http://www.python.org/doc/) in
the 'extending and embedding' section.
If you use VC you might all be interested in Ken Seehof's page at
http://starship.python.net/crew/seehof/.

Tom.




From grante at visi.com  Wed Mar 14 15:06:27 2001
From: grante at visi.com (Grant Edwards)
Date: Wed, 14 Mar 2001 20:06:27 GMT
Subject: This math scares me
References: <3aad4062.1290021@News.CIS.DFN.DE> <20010312170356.B3513@better.net> <mailman.984480856.18097.python-list@python.org> <984486395.407379@newsmaster-04.atnet.at> <slrn9asf1d.ja3.qrczak@qrnik.zagroda> <984518199.127140@newsmaster-04.atnet.at> <jwbnews-09D75F.13313313032001@corp.supernews.com> <98ohuc$du9$1@news.udel.edu>
Message-ID: <7bQr6.644$lz1.17686@ruti.visi.com>

In article <98ohuc$du9$1 at news.udel.edu>, Terry Reedy wrote:

>> Bill Gates got this right in the early days, delivering two Basic
>> interpreters:  binary floating point for speed, and decimal floating
>> point for money and such
>
>If one counts pennies instead of dollars, then 32 bit ints (+- $20 million)
>will serve for many purposes.   Exact, faster, and more widely available
>than BCD

In fiancial calculations you also need to represent numbers
that aren't amounts of money.  Interest rates for example.

-- 
Grant Edwards                   grante             Yow!  BRILL CREAM is
                                  at               CREAM O' WHEAT in another
                               visi.com            DIMENSION...


From ws-news at gmx.at  Mon Mar 12 17:38:53 2001
From: ws-news at gmx.at (Werner Schiendl)
Date: Mon, 12 Mar 2001 23:38:53 +0100
Subject: This math scares me
References: <3aad4062.1290021@News.CIS.DFN.DE> <Hwbr6.1139$Tg.127571@ruti.visi.com> <3aad4485.2348499@News.CIS.DFN.DE>
Message-ID: <984437017.150042@newsmaster-04.atnet.at>

I think print has some built in rounding to show number in a 'nice' format.
When working with floating point numbers presented to the user, you need to
specify how many decimal places to show.
Otherwise you will end up with such behaviour, this is by design.

hth
werner

<costas at springmail.com> wrote in message
news:3aad4485.2348499 at News.CIS.DFN.DE...
> On Mon, 12 Mar 2001 21:50:31 GMT, grante at visi.com (Grant Edwards)
> wrote:
>
> >In article <3aad4062.1290021 at News.CIS.DFN.DE>, costas at springmail.com
wrote:
> >
> >>Ok, I can see maybe division having problems. But why does addition of
> >>the two numbers below:
> >>
> >>5.01+5.54
> >>
> >>give me this?
> >>
> >>10.550000000000001
> >
> >Sorry. That's how floating point numbers on computers work.
> >
> >Many numbers (including 10.01) can't be represented exactly
> >in the binary representation used by your computer.
> >
> >Therefore floating-point results are never going to be exact.
> >
> >>I hope its some setting that can take care of this. I am using Python
> >>under Windows 98.
> >
> >Use integer or fixed-point BCD math if you need exact results.
>
> I will also tell my customers that this is the new math. Maybe I will
> tell them to use a slide ruler too .
>
> Joking aside, this should be fixed to display the number correctly. If
> I use 'print 5.01+5.54" then it gives the result correctly. It seems
> to happen only if you type it in the command line without a print
> command.
>
> Costas
>
>




From aon.912502367 at aon.at  Sun Mar 11 06:24:40 2001
From: aon.912502367 at aon.at (Gregor Lingl)
Date: Sun, 11 Mar 2001 12:24:40 +0100
Subject: map - lambda - problem
Message-ID: <3AAB6078.EE50AE8D@aon.at>

I tried to program a function, that rotates a piece of a puzzle in the
following way:

I represented a (rectangular) piece to rotate

abc              cfil
def      ===>    behk
ghi              adgj
jkl

as a list:

>>> n = [['a', 'b', 'c'], ['d', 'e', 'f'], ['g', 'h', 'i'], ['j', 'k',
'l']]

In direct mode I got:

>>> r = []
>>> for i in range(len(n[0])-1,-1,-1):
        r.append(map(lambda x: x[i],n))

>>> r
[['c', 'f', 'i', 'l'], ['b', 'e', 'h', 'k'], ['a', 'd', 'g', 'j']]

as intended.

If i put this into a function:

def rotate(piece):
    r = []
    for i in range(len(piece[0])-1,-1,-1):
        r.append(map(lambda x: x[i], piece))
    return r

I get

>>> rotate(n)
[['a', 'd', 'g', 'j'], ['a', 'd', 'g', 'j'], ['a', 'd', 'g', 'j']]
>>>

... as if i always had the value 0.

Why, exactly, this behaviour. Is i now taken from some other namespace?
If so, which one and why is it initialized with 0?
How can I work around this?

Thanks, Gregor Lingl

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20010311/8f6c7e1d/attachment.html>

From n8gray at caltech.edu.is.my.e-mail.address  Thu Mar  8 01:24:01 2001
From: n8gray at caltech.edu.is.my.e-mail.address (Nathaniel Gray)
Date: Wed, 7 Mar 2001 22:24:01 -0800
Subject: I come to praise .join, not to bury it...
References: <mailman.983756766.27347.python-list@python.org> <97vnu3076q@news2.newsguy.com> <slrn9a89dk.2rm.hzhu@rocket.knowledgetrack.com> <982ae80t56@news1.newsguy.com>
Message-ID: <9878td$n6b@gap.cco.caltech.edu>

Alex Martelli wrote:

> [snip]
> I am not a native speaker of English, but . . .
> [snip]

When I read this statement, my first reaction was to laugh out loud.  I 
know innumerable native English speakers and only a select few of them can 
compose sentences with the clarity (much less verbosity) of Alex' posts.  
This has got to be some sort of joke.

Then I realized that of course it was the truth.  A robot's native language 
is binary, or perhaps some sort of hexadecimal machine language, but 
definitely not English.  It probably took a massive data set to train the 
Martelli-bot's advanced neural networks to communicate using the English 
language.  I'm sure he (it?) has to restrain itself (himself?) to keep from 
from posting in streams of Python bytecodes (bzip2 compressed for 
efficiency).  :-)

Luckily,-on-the-internet-nobody-knows-I'm-a-dog-ly y'rs,

-n8-dog

-- 
_.~'^`~._.~'^`~._.~'^`~._.~'^`~._.~'^`~._
             Nathaniel Gray
   California Institute of Technology
     Computation and Neural Systems
     n8gray <at> caltech <dot> edu
_.~'^`~._.~'^`~._.~'^`~._.~'^`~._.~'^`~._



From costas at meezon.com  Sun Mar 11 19:45:18 2001
From: costas at meezon.com (Costas Menico)
Date: Mon, 12 Mar 2001 00:45:18 GMT
Subject: Pythonwin default directory init
References: <3aab04bc.2359756@News.CIS.DFN.DE> <3AABFC9E.3070706@ActiveState.com>
Message-ID: <3aac185d.25472528@News.CIS.DFN.DE>

Mark Hammond <MarkH at ActiveState.com> wrote:

> > I use Pythonwin and I would like it if the File/Open or File/Save  would
> >  default to a directory of my choice when I start it up.    I have
> > tried making a startup module to set the paths of sys but that  doesn't
> >  seem to work. I have looked thru the .py files trying to  figure
> > out where I could fix this and can't find it.    Also setting it up
> >  as a shortcut on the Desktop defaults to the  Desktop.     Does
> > anyone know how to do this?
>
>Assuming you are on Win2k (or possibly Win98), explorer itself remembers 
>the directory to open these dialogs in.  It remembers the last directory 
>you navigated to, even between runs of the application.  This can be 
>prevented by internally passing an explicit directory, but Pythonwin 
>chooses not to do this.   In Windows NT/95, the current directory is 
>always used.

Mark Hi,

I am using Windows 98. Is there someplace in the .py code that I can
explicitly force on startup to default it to the directory from "Start
In". It is already loaded in os.getcwd(). So it should be a simple
matter. 

This is a great product. Its the reason I started using Python and
building my confidence to dump VB. The code completion (intellisense?)
makes using Python a pleasure.

Is there a planned upgrade coming up any time soon? And is there some
wish/comments list I can participate in?

Thanks

Costas






From hyakugei at nospam.theorganization.net  Thu Mar  1 22:24:38 2001
From: hyakugei at nospam.theorganization.net (Digital Hyakugei)
Date: Fri, 02 Mar 2001 03:24:38 GMT
Subject: Thank you for the replys!
References: <3a9ed412$1@news1> <3a9edb8a.95346080@news.uswest.net> <ug0gxm0lx.fsf@ctwd0143.fitlinxx.com>
Message-ID: <WnEn6.270545$Pm2.4254461@news20.bellglobal.com>

That both solves my problem and deepens my knowledge of Python!
thanks to you all!

jos


"David Bolen" <db3l at fitlinxx.com> wrote in message
news:ug0gxm0lx.fsf at ctwd0143.fitlinxx.com...
> joconnor at cybermesa.com writes:
>
> > Yes, it is :)
> >
> > You can  do either....
> >
> > nntp.stat (str(aNumber))
> >
> > or teh shorter
> > nntp.stat (`aNumber`)
> >
> > Notice that the single quotes are backwards
>
> Note that for an single solitary number, this will yield the same
> results, but under the cover, str() is using __str__ (or the
> equivalent for built-ins) whereas backquotes (``) are a short form of
> repr() which uses __repr__.  So while the two the methods will both
> yield a string, it may not be the same string value depending on the
> objects involved.
>
> --
> -- David
> --
> /-----------------------------------------------------------------------\
>  \               David Bolen            \   E-mail: db3l at fitlinxx.com  /
>   |             FitLinxx, Inc.            \  Phone: (203) 708-5192    |
>  /  860 Canal Street, Stamford, CT  06902   \  Fax: (203) 316-5150     \
> \-----------------------------------------------------------------------/




From scarblac at pino.selwerd.nl  Tue Mar 20 14:58:44 2001
From: scarblac at pino.selwerd.nl (Remco Gerlich)
Date: 20 Mar 2001 19:58:44 GMT
Subject: Sequence-spreading
References: <3AB7167F.2816AA21@student.liu.se> <99790u01c73@news2.newsguy.com> <99851p$ba6$2@neptun.beotel.net>
Message-ID: <slrn9bfdek.c3q.scarblac@pino.selwerd.nl>

gzeljko <gzeljko at sezampro.yu> wrote in comp.lang.python:
> 
> Alex Martelli <aleaxit at yahoo.com> wrote in message
> > offsets = [ [] for i in range(pm) ]
> > for i in range(len(arg)):
> >     offsets[i%pm].append(i)
> 
> Old syntax yet works:
> 
> offsets = [[]]*pm

>>> offsets = [[]]*2
>>> offsets[0].append(1)
>>> offsets
[[1], [1]]

oops

-- 
Remco Gerlich


From Paul.Moore at uk.origin-it.com  Fri Mar 30 07:32:05 2001
From: Paul.Moore at uk.origin-it.com (Moore, Paul)
Date: Fri, 30 Mar 2001 13:32:05 +0100
Subject: PEP: Use site-packages on all platforms
Message-ID: <714DFA46B9BBD0119CD000805FC1F53B01B5ADD2@ukrux002.rundc.uk.origin-it.com>

Attached is a first draft of a proposal to use the "site-packages" directory
for locally installed modules, on all platforms instead of just on Unix. If
the consensus is that this is a worthwhile proposal, I'll submit it as a
formal PEP.

Any advice or suggestions welcomed - I've never written a PEP before - I
hope I've got the procedure right...

Paul Moore

PEP: TBA
Title: Install local packages in site-packages on all platforms
Version $Revision$
Author: Paul Moore <gustav at morpheus.demon.co.uk>
Status: Draft
Type: Standards Track
Python-Version: 2.2
Created: 2001-03-30
Post-History: TBA

Abstract

    The standard Python distribution includes a directory Lib/site-packages,
    which is used on Unix platforms to hold locally-installed modules and
    packages. The site.py module distributed with Python includes support
for
    locating modules in this directory.

    This PEP proposes that the site-packages directory should be used
    uniformly across all platforms for locally installed modules.


Motivation

    On Windows platforms, the default setting for sys.path does not include
a
    directory suitable for users to install locally-developed modules. The
    "expected" location appears to be the directory containing the Python
    executable itself. Including locally developed code in the same
directory
    as installed executables is not good practice.

    Clearly, users can manipulate sys.path, either in a locally modified
    site.py, or in a suitable sitecustomize.py, or even via .pth files.
    However, there should be a standard location for such files, rather than
    relying on every individual site having to set their own policy.

    In addition, with distutils becoming more prevalent as a means of
    distributing modules, the need for a standard install location for
    distributed modules will become more common. It would be better to
define
    such a standard now, rather than later when more distutils-based
packages
    exist which will need rebuilding.

    It is relevant to note that prior to Python 2.1, the site-packages
    directory was not included in sys.path for Macintosh platforms. This has
    been changed in 2.1, and Macintosh includes sys.path now, leaving
Windows
    as the only major platform with no site-specific modules directory.


Implementation

    The implementation of this feature is fairly trivial. All that would be
    required is a change to site.py, to change the section setting sitedirs.
    The Python 2.1 version has

        if os.sep == '/':
            sitedirs = [makepath(prefix,
                                 "lib",
                                 "python" + sys.version[:3],
                                 "site-packages"),
                        makepath(prefix, "lib", "site-python")]
        elif os.sep == ':':
            sitedirs = [makepath(prefix, "lib", "site-packages")]
        else:
            sitedirs = [prefix]

    A suitable change would be to simply replace the last 4 lines with

        else:
            sitedirs = [makepath(prefix, "lib", "site-packages")]

    Changes would also be required to distutils, in the sysconfig.py file.
It
    is worth noting that this file does not seem to have been updated in
line
    with the change of policy on the Macintosh, as of this writing.

Notes

    1. It would be better if this change could be included in Python 2.1, as
       changing something of this nature is better done sooner, rather than
       later, to reduce the backward-compatibility burden. This is extremely
       unlikely to happen at this late stage in the release cycle, however.

    2. This change does not preclude packages using the current location -
       the change only adds a directory to sys.path, it does not remove
       anything.

    3. In the Windows distribution of Python 2.1 (beta 1), the
       Lib\site-packages directory has been removed. It would need to be
       reinstated.


Copyright

    This document has been placed in the public domain.



From chrisw at nipltd.com  Tue Mar 20 03:38:03 2001
From: chrisw at nipltd.com (Chris Withers)
Date: Tue, 20 Mar 2001 08:38:03 +0000
Subject: Serial Port on Unix
References: <3AB637A6.FDEAE8FF@nipltd.com> <3AB51BE6.64CF103A@wildenhain.de> <002701c0b0f2$26e526e0$020000c3@node02>
Message-ID: <3AB716EB.B859A192@nipltd.com>

Sandipan Gangopadhyay wrote:
> 
> GSMLib: http://www.pxh.de/fs/gsmlib/
> 
> Kannel: http://www.kannel.org/

Thanks, these look really cool :-)

Chris



From tim.one at home.com  Sat Mar 31 22:04:10 2001
From: tim.one at home.com (Tim Peters)
Date: Sat, 31 Mar 2001 22:04:10 -0500
Subject: PythonLabs Team Moves to ActiveState
In-Reply-To: <9a626p$3qrct$1@ID-11957.news.dfncis.de>
Message-ID: <LNBBLJKPBEHFEDALKOLCMEINJJAA.tim.one@home.com>

[Emile van Sebille]
> Guido and team:

Guido always runs away after posting one of these things.  So you're stuck
with me.

> I'm sorry things didn't work out with Digital Creations.

It was news to me too!  Quite a shock.  I had thought things were working out
great with DC, so long as they kept ignoring that we spent all our time on
Python.

> I had hoped that Python core development had found a natural home, and
> was looking forward to seeing the results of a combined effort.
> Somehow, imagining a future with enhanced acquisition is much more
> interesting than a future with curly braces.

Well, if Guido said that was needed for Python to become popular, I won't
argue.  A good whore goes where he's kicked.

> Can IFEND and DEFEND or even a fork be far behind? (stackless?) I
> think that the "build it and they will come" philosophy has worked
> nicely so far.

Luckily, __END__ is already a reserved name in Python thanks to the double
double underscores.

> Some clarification may help.  Perhaps you can reconcile the *optional*
> curly braces and the "from __future__ import" discrepancy.  It
> may just be my misunderstanding, but isn't __future__ designed to allow
> bleeding-edgers to write code for what will become python?

Yes, but also tied into a warning system to force non-bleeding-edgers to pay
attention to their code that *will* break.  This is covered in PEP 236.

> That is, won't all __future__ thingys become mandatory?

That's the presumption, yes.  It remains possible that a given __future__
gimmick will be withdrawn before becoming mandatory, though.

> I believe this is how it has been presented so far, and makes me think
> that *optional* curly braces won't be after a short transition period.

Correct.  If we're to believe Guido here, you'll get warnings for all code
blocks without curly braces in 2.1, and curly braces will become mandatory in
2.2.  However, I think there's more to this story than has been public yet!
Let's not jump the gun.

> It also occurs to me that ActiveState is already deep into .Net and
> python for .Net.  Couldn't you simply provide a 'python for perlites'
> obfustication mode?  ;-)

Since I'm told I'm now an employee of ActiveState, I'm afraid I can't speak
about their plans in public anymore.

> The good news is that I believe that ActiveState will do good things
> for Python,

Evidently Guido doesn't care what you think, so stuff it.

> and as you say, should provide a stable home for python development.
> You join a team of highly respected python people in this move, and the
> resulting products will be far from common, and that is exciting.

Emile!  I hadn't pegged you for such a shameless suck-up.  Good show.

> Good luck to you all,

who-needs-luck-when-you've-got-the-gross-national-product-of-canada-
    behind-you-ly y'rs  - tim




From claird at starbase.neosoft.com  Fri Mar 30 09:51:47 2001
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 30 Mar 2001 08:51:47 -0600
Subject: Creation d'un NG francophone sur Python
References: <1eqzyk6.gxd4jr1tkjykzN%francois.granger@free.fr> <9a1e6f011es@enews3.newsguy.com>
Message-ID: <D9305864F80B9ADF.9B74F9BE0F743AE2.DC045B55BF4E1CD8@lp.airnews.net>

In article <9a1e6f011es at enews3.newsguy.com>,
Michael Spencer <spencerm at cruzio.com> wrote:
>Pardon, mais pourquoi un NG pour Python en fran?ais?
>
>Python a cr?? par un nederlandais qui a ?crit en anglais - pas sa langue
>natif. Il a ?crit en anglais ? assurer la plus grande distribution de ses
>id?es. Pourquoi on marcherait en arri?re aux temps quand chaque groupe
>a maintenu sa <<independance>> par s'isoler des autres?
			.
			.
			.
C'est pas pour <<s'isoler>>; ce n'est pas un pas <<en arri?re>>.
C'est un choix, une opportunit?.

Ou d'appuyer ou de s'opposer f.c.l.p--il m?rite une discussion
plus nuanc?e.  Moi, je souhaite la bienvenue a f.c.l.p.
-- 

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


From matth at ninenet.com  Thu Mar  8 22:20:19 2001
From: matth at ninenet.com (Matt Harden)
Date: Thu, 08 Mar 2001 21:20:19 -0600
Subject: Apache and os.fork() in CGI-Binaries
References: <3aa7ded0@si-nic.hrz.uni-siegen.de>
Message-ID: <3AA84BF3.E19873DB@ninenet.com>

Simon,

You need os.dup2().  Like this:
   logfile = os.open("/path/to/logfile",os.O_CREAT|os.O_WRONLY)
   os.lseek(logfile, 0, 2)  /* seek to end */
   os.dup2(logfile, sys.stdout.fileno())
   os.dup2(logfile, sys.stderr.fileno())
   os.close(logfile)

I would also close stdin like this:
   devnull = os.open("/dev/null", os.O_RDONLY)
   os.dup2(devnull, sys.stdin.fileno())
   os.close(devnull)

If you do that, you don't need the close() calls anymore.  Also make
sure you call sys.stdout.flush() and sys.stderr.flush() before you
os.fork().  Otherwise you could get duplicate output in some situations.

Hope this helps!
Matt

Simon Budig wrote:
> 
> Hi all.
> 
> I am currently developing a CGI-Binary which starts a second
> (long running) process. The problem is: Apache keeps the
> connection open until the fork()'ed child exits.
> 
> Discussing this thing with friends we came to the conclusion
> that some open filedescriptors keep the connection between the forked
> child and the apache alive:
> 
> 1) Apache gets a request
> 2) Apache starts CGI-Script (Python). This is the parent process
> 3) Parent process forks and creates the child
> 4) Parent process exits
> 5) Child process exits
> 6) Apache closes the connection to the browser.
> 
> But I need the closed connection after step 4. Since the Child
> inherits all filedescriptors, this is probably the thing that
> tells apache to keep the connection open.
> 
> I tried various things:
> 
> sys.stdout.close()
> sys.stdin.close()
> sys.stderr.close()
> 
> then I added
> 
> sys.__stdout__.close()
> sys.__stdin__.close()
> sys.__stderr__.close()
> 
> and this did not help. Finally i tried:
> 
> os.close (sys.stdout.fileno ())
> os.close (sys.stdin.fileno())
> os.close (sys.stderr.fileno ())
> 
> But at that time the Child does not run properly to the end.
> Of course I do something a la
> 
> logfile = open (mebong.basedir + "/logfile", "a")
> sys.stdout = sys.stderr = logfile
> 
> immediately afterwards, but this does not help.
> 
> So - how can I release all filedescriptors inherited from the parent?
> 
> The relevant parts from my code are:
> 
> -------------------------------
> # this is from the CGI-Binary
> childpid = os.fork ()
> if not childpid:
>    os.execv ("/path/to/jobcontrol", ["jobcontrol", arg1, arg2])
> -------------------------------
> 
> -------------------------------
> # this is from jobcontrol
> os.close (sys.stdin.fileno ())
> os.close (sys.stdout.fileno ())
> os.close (sys.stderr.fileno ())
> 
> logfile = open ("/path/to/logfile", "a")
> sys.stdout = sys.stderr = logfile
> mypid = os.getpid ()
> print "%d started: %s" % (mypid, repr(sys.argv))
> -------------------------------
> 
> Thanks for hints!
> 
> Bye,
>         Simon
> 
> --
>       Simon.Budig at unix-ag.org       http://www.home.unix-ag.org/simon/


From kuntsi at cc.helsinki.fi.spam.no  Mon Mar 26 16:56:43 2001
From: kuntsi at cc.helsinki.fi.spam.no (Antti Kuntsi)
Date: 26 Mar 2001 21:56:43 GMT
Subject: close(), exceptions and problems
References: <slrn9bm5u5.klj.kuntsi@myntti.helsinki.fi> <99i3j7$lqr$1@news.cybercity.dk> <slrn9bpiej.a63.qrczak@qrnik.zagroda> <99l8f4$1fob$1@news.cybercity.dk> <slrn9buioe.3ta.jepler@potty.housenet>
Message-ID: <slrn9bveoq.r0r.kuntsi@myntti.helsinki.fi>

Jeff Epler <jepler at inetnebr.com> wrote:
> Well, *either*
> 	1) The system returns -EINTR on close() and the file is not closed.
> 	   This should cause fclose() to return EOF, and Python to raise
> 	   an IOError.
> 	2) The system restarts the close() syscall internally in fclose().
> 	   When the fclose() call returns, it must have succeeded.
> I don't see how anything else could happen.
> 
> However, I did just notice one potential problem when the system raises
> IOError on close.  Regardless of the returned value from f->f_close(f->f_fp),
> f->f_fp is set to NULL.  Thus, if close() returns -EINTR, the underlying fd
> is not closed, but the file is marked as closed, and in effect the file is
> "leaked".
> 
> Could this be the behavior the original poster was referring to?

Close, the problem is that the implementation in at least Python 1.5.2
does not check the return code of the close() and thus does not raise an
exception, at least not with sockets.

Antti

-- 
         /"\                           |    iki.
         \ /     ASCII Ribbon Campaign |    fi/
          X      Against HTML Mail     |    mickut
         / \


From emile at fenx.com  Thu Mar 29 15:06:24 2001
From: emile at fenx.com (Emile van Sebille)
Date: Thu, 29 Mar 2001 12:06:24 -0800
Subject: Printing mixtures in integers and strings
References: <3ac38cf6$0$12246$ed9e5944@reading.news.pipex.net>
Message-ID: <9a04pl$32gl2$1@ID-11957.news.dfncis.de>

Use ',' instead of '+'
--or--
print "Moved to Row - %d" % row

--

Emile van Sebille
emile at fenx.com

---------
"Neil Benn" <neil.benn at cambridgeantibody.com> wrote in message
news:3ac38cf6$0$12246$ed9e5944 at reading.news.pipex.net...
> Hello,
>
>             I'm trying to use something like this:-
>
> def move(row): #Moves the dispenser to the row on the plate
>     print "Moved to Row - " + row
>
>     Where row is expected to be a number.  However when I import this into
> teh interpreter I get an error message - stating that __add__ and __radd__
> have not been defined:-
>
> Traceback (innermost last):
>   File "<console>", line 1, in ?
>   File "c:\projects\scripting\TrialScripting\AcmeBRD.py", li
> TypeError: __add__ nor __radd__ defined for these operands
>
>     However when I enter text e.g. "BOB" into the function it works fine!
> Therefore I assume that I need to explicity convert the into to a string,
> does anybody know how to do that - I can;t find the command in teh docs!!!
>
>     Thanks, in advance, for your help.
>
> Cheers,
>
> Neil Benn
> Automation Informatics Scientist
> Cambridge Antibody Technology
>
>




From digitig at cix.co.uk  Wed Mar 21 17:02:00 2001
From: digitig at cix.co.uk (Tim Rowe)
Date: Wed, 21 Mar 2001 22:02 +0000 (GMT Standard Time)
Subject: The most important modules
References: <Pine.GSO.4.21.0103191732210.22234-100000@laplace>
Message-ID: <memo.20010321220222.57471B@digitig.compulink.co.uk>

In article <Pine.GSO.4.21.0103191732210.22234-100000 at laplace>, 
mrq at for.mat.bham.ac.uk (Martyn Quick) wrote:

> I've seen a few comments along the lines of "When you are a newbie it 
> will
> take a while to realize which the important modules are."  As someone 
> who
> is in the middle of wading through Chapter 3 of the Library Reference, 
> it
> would be very helpful to know which ones are particularly worth
> noting.  (Presumably  sys, (c)Pickle and shelve, from the ones I've so 
> far
> in the chapter - but what other treats are out there for me?)

<snip>

For me it's been:

sys
xml.parsers.expat
string

But that probably just reflects the fact that I've been doing a fair bit 
with command line utilities (using sys.argv, sys.stdin, sys.stdout and 
sys.stderr) that use XML! As somebody else said, it depends on what 
/you're/ doing.




From sholden at holdenweb.com  Sun Mar  4 12:35:47 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Sun, 4 Mar 2001 12:35:47 -0500
Subject: Hitting two targets:  OO + group theory (pedagogy)
References: <u2k0atkgkiatdp9ukj3mpd1ej0b648ldsm@4ax.com> <3aa1077a.1110803@nntp.sprynet.com> <0932atgt9v8d8aboq90evoi5ipj5c0sk8s@4ax.com> <3aa11a91.5998558@nntp.sprynet.com>
Message-ID: <p5vo6.10990$ML1.527923@e420r-atl2.usenetserver.com>

"David C. Ullrich" <ullrich at math.okstate.edu> wrote in message
news:3aa11a91.5998558 at nntp.sprynet.com...
[ ... ]
> Some people would say that the reference to Genesis
> is out of place here... probably you know that some
> people would say that. Might be regarded as just
> generally inappropriate - some other people might
> think that the bit about Abel is misleading, since
> "Abelian" _does_ refer to a man named Abel, but
> not the one in Genesis.
>
[ ... ]

What's purple and commutes?














A: an Abelian grape.

sorry
 Steve





From jeremy at alum.mit.edu  Sat Mar  3 17:18:34 2001
From: jeremy at alum.mit.edu (Jeremy Hylton)
Date: Sat, 3 Mar 2001 17:18:34 -0500 (EST)
Subject: the demise of 'from foo import * and its implications?
In-Reply-To: <z8dAGJAyqLo6EwlX@jessikat.fsnet.co.uk>
References: <3kVBAfATrDo6EwEM@jessikat.fsnet.co.uk>
	<mailman.983582584.8198.python-list@python.org>
	<z8dAGJAyqLo6EwlX@jessikat.fsnet.co.uk>
Message-ID: <15009.28090.89635.437023@w221.z064000254.bwi-md.dsl.cnc.net>

>>>>> "RB" == Robin Becker <robin at jessikat.fsnet.co.uk> writes:

  RB> my preference would be that statements legal in one context
  RB> should be legal in another where they make sense. I cannot see
  RB> how using import * or eval etc in nested scopes is somehow
  RB> obviously bad or different from the same usage in non-nested
  RB> scopes. The fact that the prohibition comes from making things
  RB> easier for the compiler adds to my confusion.

Using import * in function's is asking for trouble, because it makes
it impossible for someone reading the code to know what a variable
refers to.  Consider this hypothetical bit of code:

    from string import *

    def f(x):
        from amodule import *
        return strip(x)

What does strip refer to?  Feel free to assume Python 2.0 name
binding rules.  As a reader of the code, you'll have to know what
names are exported by string and amodule.  That makes reading the code
hard.  As a program evolves, it gets even worse.  If a new function is
added to amodule, the module using import * may break.  If you add a
function mumble() to your module and call it from f, it may break
because you forgot that amodule defines a mumble, too.

Python has always had static, but not-nested scopes.  The use of
import * and exec without in-clause in functions has always clashed
with the static nature of name binding.  Code that actually uses
import * or exec in functions can be quite confusing to figure out.
(Not always, but often.)

One other thing you're confused about is that the prohibition comes
from making things easier for the compiler.  That's bunk.  The
prohibition comes because there is no reasonable and intuitive
semantics for these features in the presence of nested scopes.  When
the issue first came up on python-dev, two seasoned Python programmers
explained what the obvious semantics where.  They chose completely
opposite semantics.  There's just no sensible meaning to assign to
this kind of code in most cases.

Jeremy



From sheila at spamcop.net  Sat Mar 31 21:37:30 2001
From: sheila at spamcop.net (Sheila King)
Date: Sun, 01 Apr 2001 02:37:30 GMT
Subject: Conditional operator in Python?
References: <3AC68CB8.22AAF7B8@alcyone.com>
Message-ID: <of4dct0nul0rc6ik758gb1io3i099q1bgo@4ax.com>

On Sat, 31 Mar 2001 18:04:40 -0800, Erik Max Francis <max at alcyone.com> wrote
in comp.lang.python in article <3AC68CB8.22AAF7B8 at alcyone.com>:

:Why isn't there a conditional operator (a ? b : c) in Python? 

This can be implemented in Python by using the 'and' and 'or' operators. I
wish I could remember where I saw the explanation, but I can't remember
whether it was in Chun's _Core Python Programming_ or the "Dive Into Python"
website. In any case, I can't find it right now.

>From Kernighan and Ritchie, page 51:

"""
The <i>conditional experssion<i>, written with the ternary operator "?:",
provides an alternate way to write this and similar constructions. In the
expression
	expr1 ? expr2 : expr3

the expression expr1 is evaluate first. If it is non-zero (true), then the
expression expr2 is evaluated, and that is the value of the conditional
expression. Otherwise, expr3 is evaluated, and that is the value. Only one of
expr2 and expr3 is evaluated. Thus to set z to the maximum of a and b,
	z = ( a > b ) ? a : b;  /* z = max(a,b) */
"""

I'm sure I saw this on the "Dive into Python" site...? (Ok, I finally found
the reference, here:
http://diveintopython.org/apihelper_andor.html  )
How can we do this in Python using 'and' and 'or'...

in Python:
a and b   returns the value of b if a is true, else returns the value of a
a or b    returns the value of a if a is true, else returns the value of b

So, to return the max of a or b I could do

return  ( a > b ) and a or b

Here is an interpreter session:

>>> a = 5
>>> b = 3
>>> (a > b) and a or b
5
>>> a = 2
>>> b = 12
>>> (a > b) and a or b
12
>>> 

--
Sheila King
http://www.thinkspot.net/sheila/
http://www.k12groups.org/

: 2.0 now
:has augmented assignments, and the FAQ even acknowledges that the
:conditional operator is sometimes quite convenient and gives some (very
:awkward) Python alternatives.
:
:I've been reading comp.lang.python for a while (since just before 2.0
:shipped) and I can't really remember much discussion about it.  Is it a
:perennial request and I'm just missing it?



From cg at schlund.de  Sat Mar 10 13:25:12 2001
From: cg at schlund.de (Carsten Gaebler)
Date: Sat, 10 Mar 2001 19:25:12 +0100
Subject: How to efficently build a nested dictionary
References: <3AAA2042.E69A53FC@gmx.de> <98d9f1$1hopn$1@ID-11957.news.dfncis.de>
Message-ID: <3AAA7188.266516E9@schlund.de>

I'd like to be able to print out the annual, monthly or daily traffic for
each directory. With nested dictionaries I could easily accomplish this by
using nested for loops. How would I do that with a compound key?

cg.


Emile van Sebille wrote:
> 
> Any particular reason not to do:
> 
> ky = (dir, year, month, day)
> p_ttl = traffic.get(key,0)
> traffic[ky] = p_ttl + bytes
> 
> --
> 
> Emile van Sebille
> emile at fenx.com
> -------------------
> 
> "Carsten Gaebler" <cgaebler at gmx.de> wrote in message
> news:3AAA2042.E69A53FC at gmx.de...
> > Hi there!
> >
> > I wrote a script for FTP traffic accounting, i.e. for each
> "top level"
> > FTP directory it counts the bytes transferred per day. The
> traffic is
> > stored in a dictionary that has the following structure:
> >
> > traffic = {dir: {year: {month: {day: 42}}}}
> >
> > for several dirs, years, months, days, of course.
> >
> > While parsing the FTP log files I am building up the
> dictionary in this
> > way:
> >
> > if not traffic.has_key(dir):
> >     traffic[dir] = {year: {month: {day: bytes}}}
> > elif not traffic[dir].has_key(year):
> >     traffic[dir][year] = {month: {day: bytes}}
> > elif not traffic[dir][year].has_key(month):
> >     traffic[dir][year][month] = {day: bytes}
> > elif not traffic[dir][year][month].has_key(day):
> >     traffic[dir][year][month][day] = bytes
> > else:
> >     traffic[dir][year][month][day] += bytes
> >
> >
> > Now, this looks a bit clumsy to me, especially if I
> consider
> > you-know-which-language-I-mean which would simply allow to
> write
> > $traffic{$year}{$month}{$day} += $bytes;
> >
> > Any ideas how to do this more elegantly?
> >
> > Regards
> > Carsten.


From sheila at spamcop.net  Sat Mar 31 21:20:46 2001
From: sheila at spamcop.net (Sheila King)
Date: Sun, 01 Apr 2001 02:20:46 GMT
Subject: PythonLabs Team Moves to ActiveState
References: <mailman.986086501.5576.python-list@python.org> <9a626p$3qrct$1@ID-11957.news.dfncis.de>
Message-ID: <kt3dctgg86boone9n8i8rtrl3t9mavglaj@4ax.com>

On Sat, 31 Mar 2001 17:58:49 -0800, "Emile van Sebille" <emile at fenx.com> wrote
in comp.lang.python in article <9a626p$3qrct$1 at ID-11957.news.dfncis.de>:

:I
:think that the "build it and they will come" philosophy has worked nicely so
:far.

I think so, too (for all the newbie that I am). Hey, I already converted a
former Perl script writer to Python in just a week of trying. I think it takes
a while to get the word out, and it isn't a popularity contest, anyhow. Perl
already has an established base, and it is hard to fight that. I think that
adding in curly braces really isn't going to convince people to switch to
Python.

:Some clarification may help.  Perhaps you can reconcile the *optional* curly
:braces and the "from __future__ import" discrepancy.  It may just be my
:misunderstanding, but isn't __future__ designed to allow bleeding-edgers to
:write code for what will become python?  That is, won't all __future__
:thingys become mandatory?  I believe this is how it has been presented so
:far, and makes me think that *optional* curly braces won't be after a short
:transition period.

This thought flit through my mind upon the first reading of the announcement,
also. I hope that that is very wrong. I HATE the curly braces. I sure love
being able to write code without it.

Does this mean, that if curly braces are used to define blocks, that
whitespace will not be significant, then?

Confused,

--
Sheila King
http://www.thinkspot.net/sheila/
http://www.k12groups.org/



From aahz at panix.com  Thu Mar  8 20:51:47 2001
From: aahz at panix.com (Aahz Maruch)
Date: 8 Mar 2001 17:51:47 -0800
Subject: Can anyone recomend a good intoduction to C...
References: <mailman.983894288.19965.python-list@python.org> <3AA5EF59.507F1D91@olsen.ch> <983958878.976344@newsmaster-04.atnet.at>
Message-ID: <989cvj$sba$1@panix3.panix.com>

In article <983958878.976344 at newsmaster-04.atnet.at>,
Werner Schiendl <ws-news at gmx.at> wrote:
>
>It supports for a real large number of different design approaches.

Many of us are using Python precisely because it has a design philosophy
that goes directly counter to TMTOWTDI.
-- 
                      --- Aahz (Copyright 2001 by aahz at pobox.com)

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

"..., and some of you may regard all women as evil traps that exist only
to tease, torture, and suck out your very soul."  --DrMax


From loewis at informatik.hu-berlin.de  Sat Mar 24 06:33:46 2001
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 24 Mar 2001 12:33:46 +0100
Subject: I want to dynamically reload a module...
References: <3ABB8520.D42D5E66@seatech.fau.edu>
Message-ID: <j4wv9fmmed.fsf@informatik.hu-berlin.de>

Benoit Dupire <bdupire at seatech.fau.edu> writes:

> My goal is to dynamically import  a module called 'func' and to get the
> references to the objects that 'func' created.
> It works like this:
> prog1.py  dynamically imports func.py.

[by doing]

> class foo:
>     def __init__(self, module):
>         self.aa=__import__(module)          # dynamically import func

That won't reload module if it is already loaded. If you want to be
sure that you have the latest reference to module all the time, you
should add

          self.aa = reload(self.aa)

Regards,
Martin


From carribeiro at yahoo.com  Tue Mar 20 21:53:21 2001
From: carribeiro at yahoo.com (Carlos Ribeiro)
Date: Tue, 20 Mar 2001 23:53:21 -0300
Subject: Functionalism, aesthetics Was:(RE: I come to praise .join,
   not
In-Reply-To: <998psi02tl5@news2.newsguy.com>
References: <mailman.985052844.17993.python-list@python.org>
 <996vss013mp@news2.newsguy.com>
 <998a11$q3e$1@nntp6.u.washington.edu>
Message-ID: <5.0.2.1.0.20010320225336.00a99c50@mail.inet.com.br>

At 00:39 21/03/01 +0100, Alex wrote:
>Whatever holds in the realm of architecture, therefore,
>has no necessary parallel in that of programming, where
>"collective judgment" is concerned.

Humm. I've read dozen of answers, and got the impression that my point was 
somewhat misunderstood. That one summarizes best what the mistake is about. 
And please Alex, have some mercy when replying to me :-)

The parallel itself is valid, and we have a lot of theory on Design 
Patterns that say so. It's just that we are pointing our analysis to the 
wrong topic. Design patterns do not focus on specific solutions, but on the 
*substance* that is behind the solution. Analysing lots of solutions we can 
spot these patterns, and they say a lot about this "collective judgement" 
that is the net result of individual decisions.

We are not expecting collective judgment to make the right decisions. We 
have plenty of proof that this can lead to bad decisions many times. As a 
sidenote, this is one of the main failures of the "design-by-comitte" 
method. New things need to be invented by someone, and many times they go 
against what is considered to be the stablished way of doing things.

What we do expect is to collectively analyse the proposal using our 
knowledge of the past. Is this a sureproof method? No. Some of the 
innovative proposals will look nice at first but end up showing to be bad 
decisions. This is part of the risk of being innovative. The main point is 
that our collective judgement may spot failures. And sometimes, we do spot 
these failures by intuition. Is the intuition a valid tool? I believe so - 
sometimes we just feel that something is not right, and it takes an awful 
lot of time to find why. If we are analysing a problem where we do HAVE 
previous experience, we'd better checking what our intuition tells us.

Design patterns have other nice property. Good solutions just look right, 
because they fit nicely in some well defined pattern. Sometimes is pretty 
hard to do it, we try a lot of things, but end up finding by chance 
something that fits much better than any of the previous solutions.

Problem is, we cant just sit down and wait for ideas to pour over our head. 
We need to do something. And then we have proposals - lots of them - to be 
carefully considered. Some of them will fit the problem better than the 
others. It is actually desirable to try to develop ideas using some 
well-defined method, and the method itself can be described as a proven 
pattern. Programming techniques - structured, OOP, and so on - are full of 
design patterns that we keep applying to solve our programming needs. But 
the solution *must* fit nicely inside the problem, and our previous 
collective experience tells us a lot about that, sometimes just as a "gut 
feeling".

Many of the extensions that we are analysing right now follow well know 
patterns, many of them borrowed from other languages. Some of them are best 
described as "anti-patterns" - things that were done before, with dubious 
results.

For instance, why many of us are using Python right now? Because of the 
nice and consistent environment it does offer. I believe that the 
"featuritis" anti-pattern is going against it all. Well thought out 
features have their place (string methods, list comprehensions...) but I'm 
sure that some of the current candidates aren't really needed.


Carlos Ribeiro


_________________________________________________________
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com




From jorendorff at zoho.com  Sun Mar 18 13:04:29 2001
From: jorendorff at zoho.com (Jason Orendorff)
Date: Sun, 18 Mar 2001 10:04:29 -0800
Subject: Why "from __future__" stinks; a counter-offer
Message-ID: <AD70ED184AFED311ADFD009027AC917F83617A@dmc1nt.zoho.com>

Reply-to: sender

Okay, nested scopes are a good idea.  Having the feature
"disabled by default" in 2.1 is a good idea.  But this
"from __future__" stuff isn't so hot.

First, it's ugly.

Second, owing mostly to its ugliness, it will be
*really* embarrassing to try and explain this to my
(non-Python-bitten) friends.

Third, it's extremely silly to have a module named
"__future__" that describes only the present and the past.

Fourth, Python 2.0 and earlier report an
    ImportError: No module named __future__
when they encounter a future_statement.  Surely a
compiler error, even a SyntaxError, would be more
appropriate and less confusing.

Lastly and most importantly, it's not too late to do
something saner.  I agree it's desirable to have the
code itself advertise that it uses a certain nascent
feature.  So, what if we do something like this:

  1. Pick a name.  Suggestions follow:
         require
         using
         pragma
         this_code_uses
         feature

  2. Add a new statement to Python's grammar:
         require <name> [, <name>]*
     Or:
         import feature <name>
     It's allowed in the places where a
     "future_statement" *would* be allowed, if they
     didn't *suck* so much.

  3. In Python 2.1, the chosen name is not a
     keyword.  This way, the feature does not break
     any old code.

It could be argued that adding a new statement that
uses a pseudo-keyword is ugly, but it's certainly less
ugly than "from __future__", and more intuitive, and
less humiliating to explain.  And there's a precedent
(the pseudo-keyword "as" in import statements.)

And there has to be a limit to what we'll put up with
to avoid adding a new keyword (or pseudo-keyword) to
the language.

I'll be happy to implement, if there's any interest.

-- 
Jason Orendorff
Please reply to jorendorff at zoho.com .



From clee at gnwy100.wuh.wustl.edu  Tue Mar  6 21:18:28 2001
From: clee at gnwy100.wuh.wustl.edu (Christopher Lee)
Date: 06 Mar 2001 20:18:28 -0600
Subject: How to installing NumPy?
References: <3AA57C0D.BBC5E956@itu.edu.tr>
Message-ID: <7u3dcqqq2z.fsf@gnwy100.wuh.wustl.edu>

>>>>> "Nuzhet" == Nuzhet Dalfes <dalfes at itu.edu.tr> writes:

    Nuzhet> Hi,

    Nuzhet> I am trying to install NumPy on a RedHat 7.0 running python2.0.
    Nuzhet> Everything is in the standard places (I think!) and I picked up
    Nuzhet> and unzipped
    Nuzhet> http://download.sourceforge.net/numpy/NumPy1711_20.zip into
    Nuzhet> /usr/local/lib/python2.0/site-packages. And set PYTHONPATH to
    Nuzhet> /usr/local/lib/python2.0/site-packages.

Numpy uses distutils for easy installation with python 2.0

1. become root at the command line
2. cd to the directory into which the Numeric package was unzipped.  It
   should be called something like Numeric-17.1.1 
3. type:
          python setup.py install

   (This installs the core Numeric modules)

4. Then type:
          python setup_all.py install

   (installs the additinal packages)

You should be good-to-go.  If you don't see the scripts setup.py and
setup_all.py you are either in the wrong place or you have a broken
distribution.  If you get the chance, you might want to download 17.3.3;
it has some bug fixes.


From aahz at panix.com  Mon Mar 19 11:35:56 2001
From: aahz at panix.com (Aahz Maruch)
Date: 19 Mar 2001 08:35:56 -0800
Subject: Python-related standards
References: <3dae6h21h8.fsf@ute.cnri.reston.va.us>
Message-ID: <995chc$hb4$1@panix3.panix.com>

In article <3dae6h21h8.fsf at ute.cnri.reston.va.us>,
Andrew Kuchling  <akuchlin at mems-exchange.org> wrote:
>
>It struck me today that it might be a good idea to create a series of
>Python-related standard documents that would be a counterpart to PEPs.
>We already have some documents that would fit into this series:
>
>        * The DB-API.
>        * The DO-SIG's CORBA mapping.
>        * The SAX and DOM mappings.
>
>One future topic might be standard interfaces for cryptographic
>functions (hashes, block ciphers, &c.)  Other topics might be metadata
>for Python packages (currently being done in PEP 241), or defining a
>structured text format for docstrings.  IMHO such topics aren't really
>suitable for a PEP, since they don't necessarily have anything to do
>with the core interpreter or language.

I agree with your idea.  I think, though, that it makes a certain amount
of sense to just reuse the PEP infrastructure rather than creating yet
another place to look for information.
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

There's a difference between a person who gets shit zie doesn't deserve
and a person who gets more shit than zie deserves.  --Aahz


From tim.one at home.com  Sat Mar 31 14:18:02 2001
From: tim.one at home.com (Tim Peters)
Date: Sat, 31 Mar 2001 14:18:02 -0500
Subject: list vs tuple
In-Reply-To: <MVhx6.1021$p5.4157@news1.rivrw1.nsw.optushome.com.au>
Message-ID: <LNBBLJKPBEHFEDALKOLCOEHOJJAA.tim.one@home.com>

[Fredrik, gently guiding the young]
> since you don't know what python does, why don't you just shut up?

[The young, in the guise of one "deadmeat", gratefully guided]
> Ooh, poor diddums got proven wrong. I blew the lid on your little baby.
> <cries>

That could perhaps have gone better.

> What python DOES is irrelevant - the way it's written (and therefore
> viewed by programmers coming from other languages) makes it
> inconsistant.

This depends on which "other languages" you're talking about.  For example,
Python's behavior is "obvious" to those coming from Java or Icon, which have
similarly simple rules.  I realize you're not eager to believe it, but the
Pascal rules you're coming from are too complicated to apply directly to
Python.

> The same statement (a = b) behaves entirely differently in each case.

The things you're forgetting in this claim (which, btw, is dead wrong as
written) are the source of the problem.  Fredrik is right, and you won't make
progress until you're willing to entertain that.  Like the first line of his
"Python Objects" at

    http://w1.132.telia.com/~u13208596/guides/python-objects.htm

says, "reset your brain".  This isn't Pascal.

when-all-you've-eaten-is-sauerkraut-the-taste-of-chocolate-
    may-be-a-shock<wink>-ly y'rs  - tim




From sandipan at vsnl.com  Mon Mar 19 22:59:22 2001
From: sandipan at vsnl.com (Sandipan Gangopadhyay)
Date: Tue, 20 Mar 2001 09:29:22 +0530
Subject: Serial Port on Unix
References: <3AB637A6.FDEAE8FF@nipltd.com> <3AB51BE6.64CF103A@wildenhain.de>
Message-ID: <002701c0b0f2$26e526e0$020000c3@node02>

Hi.

GSMLib: http://www.pxh.de/fs/gsmlib/

Kannel: http://www.kannel.org/

Both open source.

Regards,

Sandipan

----- Original Message -----
From: "Tino Wildenhain" <tino at wildenhain.de>
To: "Chris Withers" <chrisw at nipltd.com>
Cc: <python-list at python.org>
Sent: Monday, March 19, 2001 2:04 AM
Subject: Re: Serial Port on Unix


> Hi Chris,
>
> Chris Withers wrote:
> >
> > Hi,
> >
> > I need to do some interaction with a serial port on Linux (actually
getting an
> > SMS from a Mobile Phone attached via a PCMCIA card, if anyone has any
better
> > ideas of how to setup an SMS target, please let me know ;-)
> >
> > Anyway, I had a hunt on python.org without much luck. A few obscure
postings and
> > a module from the vaults that turned out to be a broken link. What
module should
> > I be using to do this and where can I get it from?
>
> Add me! Me too! AOL! ;)
>
> If you ever find something usefull, let me know ;)
> Found some usefull solutions for win32 but no luck with *nix so far :(
> May be some magic with IOCTL and open/popen?
>
> Regards
> Tino
>
> --
> http://mail.python.org/mailman/listinfo/python-list
>





From aleaxit at yahoo.com  Wed Mar  7 03:07:00 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Wed, 7 Mar 2001 09:07:00 +0100
Subject: Can Python replace Visual Basic? Should it?...and generalities : )
References: <mailman.983946668.9428.python-list@python.org>
Message-ID: <984q740ss3@news1.newsguy.com>

"Christian Tanzer" <tanzer at swing.co.at> wrote in message
news:mailman.983946668.9428.python-list at python.org...
    [snip]
> No matter how clear an idea you have of what you're going to do, the
> final code is not insignificant and it is the one thing which will be
> there to be maintained for a long time. I have a lot of sympathy for
> the XP position that the code is the only thing that really counts.

...as long as you include the TESTING code as well, of course (if you
don't have more SLOC's of testing code [including unit and acceptance
tests] in an XP project's deliverables, than you have of implementation,
you must be doing something pretty peculiar indeed...).


Alex





From kriol at wally.fnal.gov  Thu Mar  8 13:56:06 2001
From: kriol at wally.fnal.gov (Oleg Krivosheev)
Date: 08 Mar 2001 12:56:06 -0600
Subject: How to call C function in shared library?
Message-ID: <vi81ys8gke1.fsf@wally.fnal.gov>

Hi, All

Linux 2.2, python 1.5.2

i already have shared library
with simple function which takes double
and returns double. Is it simple way 
to call it directly from Python?

Or some python encapsulation for dlopen/dlsym/dlclose ?

thank you

OK


From twanvds at xs4all.nl  Wed Mar 21 17:04:50 2001
From: twanvds at xs4all.nl (Twan van der Schoot)
Date: Wed, 21 Mar 2001 23:04:50 +0100
Subject: How to increase PythonWin v2.0: COM browser limits?
References: <994psh$so5$1@news1.xs4all.nl> <uy9u06vji.fsf@roundpoint.com> <uhf0o6ptw.fsf@roundpoint.com>
Message-ID: <99b8qg$b0u$1@news1.xs4all.nl>

Hi Ben,

It's great to see that a good explaination has been found.

And although I just barely grasp the meaning of your diagnosis, I wonder how
to repair it without causing a 'domino-effect'.  As my in-depth knowledge of
Pythonwin and the Python support for win32 is probably not sufficient to
oversee this.

What to say about informing Mark Hammond about this?

thanks again,

Twan.


"Ben Hutchings" <ben.hutchings at roundpoint.com> wrote in message
news:uhf0o6ptw.fsf at roundpoint.com...
> I wrote:
>
> > The wrapper for COM enumerators, in true Pythonic style, does not
> > check the size of the collection in advance, but attempts to read
> > items until this yields an error.  I think that the category
> > enumerator implementation is reporting the no-more-items error in the
> > wrong way (returning E_OUTOFMEMORY rather than S_FALSE), so the
> > wrapper does not translate it into an IndexError exception which
> > would cause the loop to terminate successfully.  OLEView probably
> > just doesn't check for the error but displays what it already got.
>
> Let me revise that:
>
> A Python iterator for a COM 'container' is an instance of
> win32com.client.util.Enumerator, which wraps a PyIEnumVARIANT object
> that in turn wraps an IEnumVARIANT interface pointer obtained from the
> container.  When an IEnumVARIANT user requests more items than there
> are left, the implementation should return any items that are left,
> with a status code of S_FALSE (not an error code).
>
> In the case of iteration, Enumerator will request only 1 item each
> time.  PyIEnumVARIANT's wrapper function returns a list of items,
> which at the end of iteration will have length 0.  Enumerator checks
> for this case and throws an IndexError when it occurs, which is the
> normal way for iteration to end.  Unfortunately it appears that this
> specific implementation of IEnumVARIANT returns a status code of
> E_OUTOFMEMORY (an error code), which PyIEnumVARIANT turns into a
> win32types.com_error exception.
>
> OLEView probably treats this error in the same way as a normal end
> of enumeration, and displays the results it already obtained.
>
> --
> Any opinions expressed are my own and not necessarily those of Roundpoint.




From max at alcyone.com  Sat Mar 24 17:50:09 2001
From: max at alcyone.com (Erik Max Francis)
Date: Sat, 24 Mar 2001 14:50:09 -0800
Subject: Best Python documentation ->HTML?
References: <pg0qbt0e598g1ecnfj5lj0lpa1q1pqvagq@4ax.com>
Message-ID: <3ABD24A1.39A7D576@alcyone.com>

Brian Glendenning wrote:

> I am a Python newbie and am confused about what the best way to turn
> Python documentation into HTML is (i.e., a "javadoc" equivalent). If
> it could also make good printed documentation (e.g., page numbers) as
> well as HTML, all the better.

HappyDoc's not bad, though with the version I got (1.3) I had to do some
tweaking to get around some bugs (which I believe have been fixed in
1.4):

    http://happydoc.sourceforge.net/

The advantage there is that the pages for HappyDoc were created with
HappyDoc itself, so you can get an idea what it looks like ...

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ To be refutable is not the least charm of a theory.
\__/ Friedrich Nietzsche
    Erik Max Francis' bookmarks / http://www.alcyone.com/max/links/
 A highly categorized list of Web links.


From jepler at inetnebr.com  Tue Mar 20 22:38:28 2001
From: jepler at inetnebr.com (Jeff Epler)
Date: Wed, 21 Mar 2001 03:38:28 GMT
Subject: pwd behaviour when new users are added
References: <mailman.985033765.15773.python-list@python.org>
Message-ID: <slrn9bg8hj.8m9.jepler@potty.housenet>

On Mon, 19 Mar 2001 12:31:23 -0800, Timothy Grant
 <tjg at exceptionalminds.com> wrote:
>I just ran into an interesting feature of the pwd module
>
>say I have users fred, ethel, ricky and lucy on my system and I
>issue a 'pwd.pwgetall()' this returns a tuple with user
>information. Now, while my programme is running I issue a
>'useradd gilligan'
>
>if my programme now issues another 'pwd.pwgetall()' or a
>'pwd.pwgetnam()' or 'pwd.pwgetuid()' the new user name is not
>available.
>
>if I 'reload(pwd)' it still doesn't work if I 'del pwd' and
>then 'import pwd' it doesn't work.
>
>What am I doing wrong?

Calls to the underlying C library functions don't appear to re-open
the password file.

Take a look at the following session with 'strace -e open':
	>>> pwd.getpwnam("jepler")
open("/etc/nsswitch.conf", O_RDONLY)    = 3
open("/etc/ld.so.cache", O_RDONLY)      = 3
open("/lib/libnss_files.so.2", O_RDONLY) = 3
open("/etc/passwd", O_RDONLY)           = 3
('jepler', 'x', 405, 405, 'Jeff Epler', '/home/jepler', '/bin/zsh')
	>>> pwd.getpwnam("jepler")
open("/etc/passwd", O_RDONLY)           = 3
('jepler', 'x', 405, 405, 'Jeff Epler', '/home/jepler', '/bin/zsh')

Adding a call which would endpwent() may help (this doesn't exist in
the pwd module yet), because the next getpw* call will have to reopen
/etc/passwd

Jeff


From pinard at IRO.UMontreal.CA  Fri Mar 30 09:55:40 2001
From: pinard at IRO.UMontreal.CA (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 30 Mar 2001 09:55:40 -0500
Subject: Creation d'un NG francophone sur Python
In-Reply-To: claird@starbase.neosoft.com's message of "30 Mar 2001 08:51:47 -0600"
References: <1eqzyk6.gxd4jr1tkjykzN%francois.granger@free.fr>
	<9a1e6f011es@enews3.newsguy.com>
	<D9305864F80B9ADF.9B74F9BE0F743AE2.DC045B55BF4E1CD8@lp.airnews.net>
Message-ID: <vrk857qpar.fsf@bor.IRO.UMontreal.CA>

[Cameron Laird]

> In article <9a1e6f011es at enews3.newsguy.com>,
> Michael Spencer <spencerm at cruzio.com> wrote:
> >Pardon, mais pourquoi un NG pour Python en fran?ais?

> C'est un choix, une opportunit?.

I directly know some people that do not read English easily, and that
would be Python users if there were more French resources for Python.
I'm sure the same applies for people speaking many other languages.

English should not be necessary to use computers, and if Python is meant
to be a CP4E, it should have wide views about non-English people.

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



From sheila at spamcop.net  Mon Mar 19 20:21:48 2001
From: sheila at spamcop.net (Sheila King)
Date: Tue, 20 Mar 2001 01:21:48 GMT
Subject: Paring down diskspace required by a Python-2.0 installation
Message-ID: <6dbdbtsg6beh5p73cvlubeb192oocohfbb@4ax.com>

My webhost only has Python 1.5.1 as default install. So, in my own webspace, I
have installed Python 2.0

However, it takes up so much space, that I have had to purchase extra disk
space.

What I want to know, is what I can safely delete off of this Linux box (Red
Hat) that my site is running on.

So far, I copied the contents of the Demo/ and Tools/ directories to my home
machine, and deleted them off of my web host's machine. I also deleted the
Docs/ folder (and contents), since I have a copy on my home machine. And the
PC folders (both of them). I also removed the contents of the test/ folder to
my home machine, and deleted it from my web host.

What I am wondering if I can delete, now, is...well I guess it is the source
directories. Like Grammar/ and Modules/ which have only .c and .h and .o files
in them. These files are not used for running Python, only for creating the
executable, right? So can I safely delete them?

Is there anything else I can remove?

--
Sheila King
http://www.thinkspot.net/sheila/
http://www.k12groups.org/



From cmukai at pacbell.net  Sat Mar  3 01:49:36 2001
From: cmukai at pacbell.net (Conrad Mukai)
Date: Fri, 02 Mar 2001 22:49:36 -0800
Subject: no zlib in Python 2.0
Message-ID: <3AA09400.E5174688@pacbell.net>

I've recently installed Python 2.0 on a Mandrake 2.2.17 system. When I
invoke python I get the following version information:

Python 2.0 (#1, Feb 27 2001, 16:54:15)
[GCC 2.95.3 19991030 (prerelease)] on linux2

and uname -a returns:

Linux clearapp.clear-logic.com 2.2.17-21mdk #1 Thu Oct 5 13:16:08 CEST
2000 i686 unknown

The problem is when I run a script that formerly ran under Python 1.5.2
I get the following error message:

  File "/usr/local/lib/python2.0/gzip.py", line 9, in ?
    import zlib
ImportError: No module named zlib

Looking into the python2.0 installation, I can see that the lib-dynload
directory is empty. The 1.5 installation has the shared object
zlibmodule.so as well as other files. It would appear that the source
code is in the 2.0 package, but the install process is not compiling the
code. Does anyone have any hints as to what is wrong? Thanks in advance.

--
************************************************************
* Conrad Mukai          "Consistency is the last resort of *
* cmukai at pacbell.net     the unimaginative." - Oscar Wilde *
************************************************************





From kalle at gnupung.net  Thu Mar  8 17:16:34 2001
From: kalle at gnupung.net (Kalle Svensson)
Date: Thu, 8 Mar 2001 23:16:34 +0100
Subject: Get terminal size in perl and python
In-Reply-To: <DORp6.25$Tg.3395@ruti.visi.com>; from grante@visi.com on Thu, Mar 08, 2001 at 08:18:43PM +0000
References: <20010308194028.A1616@bork.demon.nl> <mailman.984079092.10641.python-list@python.org> <DORp6.25$Tg.3395@ruti.visi.com>
Message-ID: <20010308231634.A748@apone.network.loc>

Sez Grant Edwards:
> In article <mailman.984079092.10641.python-list at python.org>, Kalle Svensson wrote:
> >Sez Egbert Bouwman:
> >> My perl correspondent uses:
> >>     use Term::ReadKey;
> >>     ($cols,$lines)=GetTerminalSize()
> >> What is the python equivalent ?
> >
> >I'd use os.environ["COLUMNS"] and os.environ["LINES"], but I've no idea
> >about how portable they are. 
> 
> In most Unixes you'd be better off using the ioctl() call (the
> details of which I can't remember at the moment -- I thought it
> was in the termios struct, but I can't find it).

Ah, yes.  Some research gave me this (i686 Linux 2.2.17, Python 2.1a2):
###### START ######
import struct, fcntl, IOCTL, sys

s = struct.pack("HHHH", 0, 0, 0, 0)
while 1:
    if raw_input("") == "q":
        break
    lines, cols = struct.unpack("HHHH", fcntl.ioctl(sys.stdout.fileno(),
		                                    IOCTL.TIOCGWINSZ, s))[:2]
    print lines, "lines,", cols, "cols."
####### END #######
If you don't have a IOCTL module, don't want to make you own, and have the
same platform as I, replace IOCTL.TIOCGWINSZ with 0x5413.

Lovely beasts, those ioctls.

Peace,
  Kalle
-- 
Email: kalle at gnupung.net     | You can tune a filesystem, but you
Web: http://www.gnupung.net/ | can't tune a fish. -- man tunefs(8)
PGP fingerprint: 0C56 B171 8159 327F 1824 F5DE 74D7 80D7 BF3B B1DD
 [ Not signed due to lossage.  Blame Microsoft Outlook Express. ]



From aahz at panix.com  Tue Mar 20 16:26:58 2001
From: aahz at panix.com (Aahz Maruch)
Date: 20 Mar 2001 13:26:58 -0800
Subject: New PEP: The directive statement
References: <mailman.985117312.5806.python-list@python.org> <j4u24oi4e8.fsf@informatik.hu-berlin.de> <yHPt6.2553$4N4.334967@newsc.telia.net>
Message-ID: <998hv2$5ub$1@panix3.panix.com>

In article <yHPt6.2553$4N4.334967 at newsc.telia.net>,
Fredrik Lundh <fredrik at effbot.org> wrote:
>Martin von Loewis wrote:
>>
>> That is also why the PEP 236 proposal (__future__ imports) finds that
>> much resistance it uses an action statement to declare things.
>
>except that "import" already is a declaration statement, of course.

Depends how you define "already".  There is no current released version
of Python (alphas and betas do not count as "released versions") in
which "import" is a declaration.  Moreover, making "import" into both an
action statement and a declaration has some obvious warts about it.

I'm still not entirely sure that I think this directive idea is better,
but let's address it on its merits and not throw in the bogeyman of
compatibility issues.

>> P.S. If "directive transitional nested_scopes" is approved
>
>well, adding new, ugly, and utterly non-pythonic syntax for a one-
>shot workaround strikes me as a really bad thing.  if I want Tcl
>(random interpretation of barewords) or applescript (lame attemts
>to make things look like english), I know where to find it.

So you're saying that "from __future__" is only to be used for this one
issue of nested scopes?  I don't think so....
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

"I won't accept a model of the universe in which free will, omniscient
gods, and atheism are simultaneously true."  -- M


From sekter at matavnet.hu  Thu Mar  8 10:22:21 2001
From: sekter at matavnet.hu (Arpad Kiss)
Date: Thu, 08 Mar 2001 16:22:21 +0100
Subject: virtual hosting in Zope
References: <987lbo$119t$1@infocom.km.ua> <8zmg2xch.fsf@propylon.com>
Message-ID: <3AA7A3AD.3090304@matavnet.hu>

Yes, SiteAcces is included in the latest Zope setup.
Arpad

Martin Skott wrote:

> "Anton A. Zotov" <futurb at nh.km.ua> writes:
> 
> 
> 
> 
>> Urgent: I'd appreciate a person who would tell me how to implement virtual
> 
>> hosting under Zope/Medusa environment
> 
> 
> 
> What do you actually mean with virtual hosting?
> 
> Is your Zope server placed behind a webserver (eg. Apache) or is it
> 
> standalone Zope?
> 
> 
> 
> I think nomatter your setup you should take a look at the SiteAccess
> 
> product which you can find on www.zope.org. I have heard rumors that
> 
> it should be included in the latest Zope, but haven't checked myself.


-- 
"Don't try to teach a pig to sing;
it wastes your time and it irritates the pig."
Awareness by Anthony de Mello



From tjg at exceptionalminds.com  Sat Mar 17 15:39:50 2001
From: tjg at exceptionalminds.com (Timothy Grant)
Date: Sat, 17 Mar 2001 12:39:50 -0800
Subject: installing 2.x over the top of 1.5.2
Message-ID: <20010317123950.C24992@trufflehunter.avalongroup.net>

This problem has been bothering me for a while.

Python v1.5.2 is installed by default on RH6.2 boxes. 

My problem is that upgrading an RH6.2 box by using the 2.x rpms
now means that there are two versions of Python installed on my
box, and any extra modules I have installed have to be copied
to the new location for v2.x. I then need to uninstall the
v1.5.2 rpms.

I'm sure someone somewhere has come up with a clean way of
upgrading Python. As with all thing Pythonic, there must be an
easier way.

-- 
Stand Fast,
    tjg.

Timothy Grant                         tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.         <><       (503) 246-3630
>>>>>>>>>>>>>Linux, because rebooting is *NOT* normal<<<<<<<<<
>>>>This machine was last rebooted:  60 days     4 hours ago<<



From mwh21 at cam.ac.uk  Tue Mar  6 13:19:33 2001
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 06 Mar 2001 18:19:33 +0000
Subject: Can anyone recomend a good intoduction to C...
References: <mailman.983894288.19965.python-list@python.org> <983896344.701456@newsmaster-04.atnet.at>
Message-ID: <m3lmqibw0a.fsf@atrus.jesus.cam.ac.uk>

"Werner Schiendl" <ws-news at gmx.at> writes:

> Hi Laura,
> 
> The only reason I can imagine at this point to go with plain C is that there
> are a lot more platforms that support C than do support C++. This is mainly
> a topic in embedded development and realtime platforms.
[...] 
> Did I miss an important point?

Sprinkle IMHO's throughout what follows.

Yes, you're missing the fact that C++ sucks large rocks through very
small straws.  There's some merit in learning C for when low-levelness
is wanted, eg. when developing Python.  If you're writing an
application, then you might think that C++ would be a good language to
use.  You'd be wrong (FWIW, you'd be wrong to use C here, too).  Use a
sane programming language like Python or Common Lisp (and if you use
Python, then you can write portions in C - but leave the application
logic out of it!).

One of my more recent additions to my sig collection is:

  C++ is a siren song.  It *looks* like a HLL in which you ought to
  be able to write an application, but it really isn't.
                                       -- Alain Picard, comp.lang.lisp

I hope everyone noticed the:

Sprinkle IMHO's throughout what follows.

at the top of this article.  My feelings for C++ oscillate between
mild dislike and outright searing hatred - and I have programmed in
it, I've formed my opinion on experience, and I think it sucks.  You
may have a different opinion, but this is mine.

Cheers,
M.

-- 
  There's a difference between random people with stripy jumpers, and
  a respected scientist with a reputation.  
                                            -- Steve Kitson, ucam.chat


From brendhanhorne at bellsouth.net  Mon Mar 12 20:45:10 2001
From: brendhanhorne at bellsouth.net (Brendhan Horne)
Date: Mon, 12 Mar 2001 20:45:10 -0500
Subject: Very Cool
References: <g13r6.201$qt5.10920@news3.mia>
Message-ID: <OYer6.280$qt5.17754@news3.mia>

Everybody thank you for your help you may have broken the one barrier that
was causing me not to understand this. Now let explain what happened and
maybe you can clarify something even futher for me.
I went in to my Python 20 command line interpreter the DOS based one ( am I
saying that right )  And wrote in what you guys sent me.
Items in parenthsis () are notes about what I did or what happen nothing in
those parenthesis was ever typed in.

>>>while 1:    ( I wrote this in )(press enter key
....    print "Hello World"  ( the dots would appear as soon as I entered in
while 1:)
File "<stdin>", line 2
   print  "Hello World"
       (there is a control symbol under the t in print)
Indentation Error: expected an indented block



I tried typing it in several different ways and basically that was the error
message I got. So I tried it in the IDLE and it worked very cool. I saved
the file as test.py.  If I opened it in IDLE it would open another window
showing me the file code not actully running it.  If I click on it the
command line would open and run the program properly you would see hello
world running up the screen. If I hit CTRL+C it would just completly close
(while the command line version was running). I have set path in my
autoexec.bat for Python and I am running win98se. But in easy terms why the
differences. Why couldn't I type it in command line and get it to work
correctly, yet in IDLE no problem. Why when I clicked on the file would it
open in command and work ( I know about the path helping that but I guess I
get miffed when I type it in the command line and it doesn't  work for me).
Please forgive the lack of intellegence in my questions but I figure I will
learn more if I ask even the stupid questions.

--
Thanks,
Brendhan





From jp at NOSPAMdemonseed.net  Sat Mar 31 17:18:38 2001
From: jp at NOSPAMdemonseed.net (jason petrone)
Date: Sat, 31 Mar 2001 22:18:38 +0000
Subject: Know of Substantial Apps Written in Python?
References: <3ac4e4a8_1@news4.newsfeeds.com> <KE6x6.415$SR6.235983@typhoon1.ba-dsg.net> <dj7x6.9561$SB2.850896@ruti.visi.com>
Message-ID: <u3l5a9.fud.ln@demonseed.net>

Grant Edwards <grante at visi.com> wrote:
> Anybody get Grail to work lately?  Last time I tried it
> wouldn't run.

I use it occasionally.  It took about 3-5 minor changes for it to work with 
python 2.0.

-jason



From James_Althoff at i2.com  Mon Mar 26 13:49:31 2001
From: James_Althoff at i2.com (James_Althoff at i2.com)
Date: Mon, 26 Mar 2001 10:49:31 -0800
Subject: PEP 245: Python interfaces
Message-ID: <OF1F17CAE4.07231E40-ON88256A1B.0066234F@i2.com>

I would like to propose a change to PEP 245.  Instead of calling for the
"Interfaces package" to be turned into a standard Python "type" (as stated
in PEP 245), I think "interface objects" should be (i.e., remain)
instances of a true "Interface class" (_not_ a built in type) so that the
(unfortunate) split between "types" and "classes" in Python does not grow
any wider.  This would leave open the possibility for someone to subclass
the "Interfaces class" if necessary to add methods and fields to extend its
initial functionality.

Jim




                                                                                                                   
                    Michel Pelletier                                                                               
                    <michel at digicool.c        To:     python-list at python.org                                       
                    om>                       cc:                                                                  
                    Sent by:                  Subject:     PEP 245: Python interfaces                              
                    python-list-admin@                                                                             
                    python.org                                                                                     
                                                                                                                   
                                                                                                                   
                    03/22/01 11:02 PM                                                                              
                                                                                                                   
                                                                                                                   





Hi!

I'm the author of PEP 245, a framework and syntax for interface objects in
Python.

http://python.sourceforge.net/peps/pep-0245.html

It has just recently been made a PEP, and I'd like to solicite some more
input from the community (you).

For those of you who don't know the formal defintion of "interface", don't
worry (there isn't one!).  There are, however, a lot of agreed and
disagreed upon definitions out there in the world, and PEP 245 is one
particular take on the situation, based on pioneering work done by the
Type-sig and Jim Fulton over the last couple of years.

There is a Wiki (referenced in the PEP) for formal discussion and, of
course, this newsgroup, which I will hanging out in for the next few days
answering any questions about the proposal.

Thanks!

-Michel

--
http://mail.python.org/mailman/listinfo/python-list







From bos at hack.org  Wed Mar 28 23:24:08 2001
From: bos at hack.org (Rikard Bosnjakovic)
Date: Thu, 29 Mar 2001 06:24:08 +0200
Subject: Recommendations for Python Books
References: <3ac2582c$1@news.starhub.net.sg>
Message-ID: <3AC2B8E8.C3651AD0@hack.org>

Wong Da Qiang wrote:
 
> Just a newbie to programming and python, need some recommendations on books
> to buy

I used this one and I liked it a lot:

	http://www.oreilly.com/catalog/lpython/


-- 
Rikard Bosnjakovic - http://a214.ryd.student.liu.se/cv/ - ICQ: 1158217

Anyone sending unwanted advertising e-mail to my address will be
charged $250 for network traffic and computing time. By extracting my
address from this message or its header, you agree to these terms.


From sholden at holdenweb.com  Wed Mar  7 19:01:07 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Wed, 7 Mar 2001 19:01:07 -0500
Subject: Remote script execution with Python ...
References: <7Ybp6.7729$r4.7336@news.indigo.ie> <ulmqil8qu.fsf@roundpoint.com> <Lpwp6.8059$r4.7712@news.indigo.ie>
Message-ID: <U0Ap6.36361$1D5.1540871@e420r-atl1.usenetserver.com>

"Robert Gahan" <gahanr at gofree.indigo.ie> wrote in message
news:Lpwp6.8059$r4.7712 at news.indigo.ie...
> Thanks Ben.
>
> os.popen2 is not available to windows programmers but
> win32pipe came to the rescue !!. win32pipe I think is available
> from the following location
> http://www.activestate.com/Products/ActivePython/win32all.html
>
> Got something like the following to work ....
>
> import win32pipe
>
> stdin, stdout = win32pipe.popen2("rsh <hostname> -l <username>
> <scriptname>", "t")
> stdin.write("<whatever you want to write to stdin !!>")
> print stdout.readlines() # This will print the result.
>
> Regards,
> Robert.
>
> Ben Hutchings <ben.hutchings at roundpoint.com> wrote in message
> news:ulmqil8qu.fsf at roundpoint.com...
> > "Robert Gahan" <gahanr at gofree.indigo.ie> writes:
> >
> > > Hi there,
> > >
> > > I have installed python on a Windows PC and I am looking to
> > > remotely execute a script on a UNIX box from a python program.
> > > I need to be able to pipe a "y" to stdin also. Is there a windows
> > > equivalent of the unix "rexec" that could help me ?. What is the
> > > most appropriate way of doing this ?.
> >
> > You might be able to do this with something like:
> >
> >     script_in, script_out = os.popen2("rsh otherhost -c myscript")
> >     print >>script_in 'y'
> >     result = script_out.readlines()
> >
> > Windows 2000 includes an rsh client but I'm not sure about other
> > versions of Windows.  You might want to use ssh instead for greater
> > security.

For portability, Luz and Ascher recommend:

import sys
if sys.platform == "win32":
    try:
        import win32pipe
        popen = win32pipe.open()
    except ImportError:
        raise ImportError, "Could not find win32pipe module"
else:
    import os
    popen = os.popen()

And then use popen() to open pipes. There is an obscure bug in the Windows
os.popen() module on Windows, but this seems to be a suitable workaround.

regards
 Steve







From robin.thomas at starmedia.net  Wed Mar 14 21:33:14 2001
From: robin.thomas at starmedia.net (Robin Thomas)
Date: Wed, 14 Mar 2001 21:33:14 -0500
Subject: Package
In-Reply-To: <001c01c0acea$d33c6810$9071fea5@ANBU>
Message-ID: <4.3.1.2.20010314213057.00d419b0@exchange.starmedia.net>

At 04:56 PM 3/14/01 -0800, Anbuselvam Palusamy wrote:
>Can any one explain how to create a packages in python?
>

See section 6.4 of the Python Tutorial.

http://www.python.org/doc/current/tut/node8.html


--
Robin Thomas
Engineering
StarMedia Network, Inc.
robin.thomas at starmedia.net
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20010314/0e5fd615/attachment.html>

From aleaxit at yahoo.com  Sat Mar 17 03:45:18 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sat, 17 Mar 2001 09:45:18 +0100
Subject: I come to praise .join, not to bury it...
References: <mailman.983485923.1136.python-list@python.org> <97p7iq$kk2$1@nntp6.u.washington.edu> <97rka002k9@news1.newsguy.com> <3AAF7E47.294617E1@sage.att.com> <98qeco02hn7@news1.newsguy.com> <98rmod$c8p$1@animus.fel.iae.nl> <98tacr0q5s@news1.newsguy.com> <98u86d$6p9$1@animus.fel.iae.nl>
Message-ID: <98v88s01ri1@news2.newsguy.com>

"Carel Fellinger" <cfelling at iae.nl> wrote in message
news:98u86d$6p9$1 at animus.fel.iae.nl...
> Alex Martelli <aleaxit at yahoo.com> wrote:
> > "Carel Fellinger" <cfelling at iae.nl> wrote in message
> > news:98rmod$c8p$1 at animus.fel.iae.nl...
> ...
> >> but if you get too blinded by this substance thing, you might easily
> >> overlook that in the end you have to weight things and there again the
> >> seasoned sense of aesthetics of the gifted ones is what makes the
> >> difference.
>
> > I appreciate "form follows function" as an aesthetic principle.
>
> Here we agree, and clearly Rococo, though having a beauty of its own,
> has nothing to do with the aesthetic principles that guided the
> Bauhaus people, so aesthetics in it self as the sole guiding light has
> it pitfals, to say the least:)  But we have discussed this before,
> aesthettics as a design guide only works for the `Enlightend Ones'.

So you're saying Brunelleschi was not 'Enlightend' enough for you,
for example, because he ornamented aplenty?  This keeps sounding
to me as high-falutin' cover-up for "what _I_ like is good".


> > You seem to imply some moderation (have to weight things) is
> > key to YOUR aesthetics -- fine, Aristotle will agree, as will
>
> I don't know why you think that moderation is key to my aesthetics,

That mention of "you have to weight things" above.  Extremist
aesthetics does no 'weighing' of one thing against another -- it
applies its principles ruthlessly (that's why it's extremist).


> The weighting you seem to refer to is the weighting you yourself
> mentioned earlier on, i.e. that Guido has to weight the relative

No, it's from the quote from YOUR post which is again repeated
at the beginning of this one (2nd line of the quote at the start).


Alex





From aleaxit at yahoo.com  Wed Mar 21 10:54:13 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Wed, 21 Mar 2001 16:54:13 +0100
Subject: Rich comparison in derived class, but not in base (boost.python)
References: <Pine.LNX.4.21.0103211112040.1131-100000@kev.ioc.ee>
Message-ID: <99air301tpg@news2.newsguy.com>

"Pearu Peterson" <pearu at cens.ioc.ee> wrote in message
news:Pine.LNX.4.21.0103211112040.1131-100000 at kev.ioc.ee...
    [snip]
> I checked also the following approach:
> class expr:
>     def __init__(self,*args):
>         self.e = ex(*args)
>     def __cmp__(self,other):
>         raise NotImplementedError,"expr.__cmp__() should not be called"
>     def __lt__(self, other):
>         return relational(self.e, other, '<')
>     # also other richcomp methods are defined but not shown here
>
> that works fine but then I would have to re-define also all `ex' methods
> in `expr' --- there are more than 30 of them, so I would not want to do

As a temporary workaround, just delegate via __getattr__:

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


Alex





From tim.one at home.com  Fri Mar  2 20:41:25 2001
From: tim.one at home.com (Tim Peters)
Date: Fri, 2 Mar 2001 20:41:25 -0500
Subject: python-dev summary 2001-02-15 - 2001-03-01
In-Reply-To: <3AA045A6.3060408@caltech.edu.is.my.e-mail.address>
Message-ID: <LNBBLJKPBEHFEDALKOLCEEJIJDAA.tim.one@home.com>

> Greg Wilson set up an egroups (now Yahoo) discussion group for it, where
> debate has continued:
>
>     http://groups.yahoo.com/group/python-iter

[Nathan Gray]
> Any reason why non-members are excluded from reading the archives
> on this list?

Why are you asking this on c.l.py?  I don't know the answer, but if it were a
question I cared about I'd probably ask the list owner.

betting-the-answer-is-"huh!-guess-that's-the-default"-ly y'rs  - tim




From J.Jonkers at acriter.com  Thu Mar  8 02:38:49 2001
From: J.Jonkers at acriter.com (Johan Jonkers)
Date: Thu, 08 Mar 2001 08:38:49 +0100
Subject: about environment variable.
References: <9877ho$kqn$1@news2-2.kornet.net>
Message-ID: <3AA73709.2AB2F944@acriter.com>

What you need to do first is create a frame on which the button can be
created. Like this:

Python 1.5.2 (#6, Jun 19 2000, 15:37:01)  [GCC 2.95.2 19991024
(release)] on sunos5
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> from Tkinter import *
>>> r=Tk()
>>> b=Button(r, text="hello world!", command="exit")
>>> b.pack()

after b.pack(), the button will appear
-- 
    -- The Secret To Success Is Knowing Who To Blame For Your Failures.
--


From bdupire at seatech.fau.edu  Wed Mar 21 15:33:10 2001
From: bdupire at seatech.fau.edu (Benoit Dupire)
Date: Wed, 21 Mar 2001 15:33:10 -0500
Subject: learning python...
References: <3AB8F4E1.184DD9C@noaa.gov> <Z97u6.16733$Im6.1761313@newsread1.prod.itd.earthlink.net>
Message-ID: <3AB91006.6D7E35F2@seatech.fau.edu>


Simon Kesenci wrote:

> If you know programming, forget _Learning_ _Python_.  I tried _Programming_
> _Python_ last year, but it covered version 1.3.  A good book if they
> updated it since then.  As a programmer, you should find the online
> tutorial easy and helpful <http://python.org/doc/current/tut/tut.html>.

Programming Python was updated and now covers Python 2.0
I am reading it, actually, and, till now, I find it great.. (but I am only page
35! ;o) ).


This edition is intended to be the 'sequel' to " Learning Python"... so yes,
you can try to read the Python tutorial and this book...




From claird at starbase.neosoft.com  Thu Mar 22 12:59:06 2001
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 22 Mar 2001 11:59:06 -0600
Subject: Why Python/Jython?
References: <99cl9e$b2b$1@Merlin.et.bocholt.fh-gelsenkirchen.de> <mailman.985260089.28444.python-list@python.org>
Message-ID: <1901D50DC30D30E4.16551C1AB0FB4363.C0D72E208759F8A7@lp.airnews.net>

In article <mailman.985260089.28444.python-list at python.org>,
Steve Purcell  <stephen_purcell at yahoo.com> wrote:
>Andreas Freier wrote:
>> In several articles I read that Python/Jython imporoves the programmers
>> performance many times. Why?
>
>Simple: Python code is usually many times shorter than the corresponding
>Java code.
>
>Programmer performance is usually gauged by coding time and
>number of bugs, both of which are vaguely proportional to the number of
>lines of code in most sane languages*.
>
>* Perl is a noteworthy exception
>
>> I am considering to use Jython for my job as Java programmer. How should I
>> use it to achieve this performance?
>
>Just use it and see if it works for you.
			.
			.
			.
I'm going to repeat several of the same points, to
help ensure that they arrive safely:
1.  Jython is different from Java in that "Just
    use it and see if it works for you" is
    meaningful.  People do NOT start learning
    Java lightly; it often takes hours just to 
    install and configure a minimal JDK, let
    alone understand the language at all.
    Python (and Jython) are far lighter-weight.
    It's realistic to schedule two hours to
    locate, download, install, and begin in a
    productive way to program Python.
2.  You don't have to be a Java programmer to
    program with Java classes.  In fact, many
    of use believe it might even be suboptimal.
    The most productive way to program in Java--
    to develop applications, including GUIs,
    test routines, ...--is, we think, likely to
    be with a "very-high-level language" (VHLL)
    that understands how to manipulate Java
    classes.  Many of these languages <URL:
    http://starbase.neosoft.com/~claird/comp.lang.java/java_scripting.html >
    are already in industrial use.  Jython is
    arguably the most refined and advanced among
    these.
-- 

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


From cmfinlay at SPAMmagnet.com.au  Tue Mar 13 21:35:45 2001
From: cmfinlay at SPAMmagnet.com.au (cmfinlay at SPAMmagnet.com.au)
Date: Wed, 14 Mar 2001 13:35:45 +1100
Subject: Where can I find Docs on how to use the debugger please?
Message-ID: <3lltat8tgd86rknhgnqkij6bk9rkbqk824@4ax.com>

Where can I find Docs on how to use the debugger please?

I am using the Win32 version of Python 2.1

I would like to set a breakpoint
and then later step down the source code line by line.
and view the source code as it is executed.

I am using a different file for each module.

I have looked in some standard online docs and all I found was help on
the keyboard and menus.

Also is there a better debugger with a GUI for Python available
(I am willing to try other OSes.)

Thank you for your time in reading this.

Colin Finlay
cmfinlay at SPAMmagnet.com.au


From fdrake at acm.org  Fri Mar  2 15:49:09 2001
From: fdrake at acm.org (Fred L. Drake, Jr.)
Date: Fri, 2 Mar 2001 15:49:09 -0500 (EST)
Subject: Python 2.1 beta 1 documentation online
Message-ID: <15008.1861.84677.687041@localhost.localdomain>

  The documentation for Python 2.1 beta 1 is now online:

	http://python.sourceforge.net/devel-docs/

  This is the same as the documentation that will ship with the
Windows installer.
  This is the online location of the development version of the
documentation.  As I make updates to the documentation, this will be
updated periodically; the "front page" will indicate the date of the
most recent update.


  -Fred

-- 
Fred L. Drake, Jr.  <fdrake at acm.org>
PythonLabs at Digital Creations




From michael at stroeder.com  Tue Mar 20 09:01:27 2001
From: michael at stroeder.com (Michael =?iso-8859-1?Q?Str=F6der?=)
Date: Tue, 20 Mar 2001 15:01:27 +0100
Subject: Whois client?
References: <mailman.985090767.19146.python-list@python.org>
Message-ID: <3AB762B7.69295001@stroeder.com>

Bill Bell wrote:
> 
> Has anyone seen a Whois client coded in Python?
> 
> Thanks for any leads.

On first position when searching with Google:

  http://www.google.com/search?q=python+whois+module&hl=de&lr=

Ciao, Michael.


From db3l at fitlinxx.com  Thu Mar 15 17:57:46 2001
From: db3l at fitlinxx.com (David Bolen)
Date: 15 Mar 2001 17:57:46 -0500
Subject: reversing a dictionary (newbie)
References: <007201c0ad37$12571f80$6809a78f@dcs.shef.ac.uk> <mailman.984652340.26249.python-list@python.org>
Message-ID: <uk85qmyhh.fsf@ctwd0143.fitlinxx.com>

Steve Purcell <stephen_purcell at yahoo.com> writes:

> Expect total memory usage to be triple that of 'dict' alone towards the end
> of the loop, because of the 'items()' call.

Of course, "triple that of 'dict' alone" covers the object references
in the dictionary, but not triple the overall object space, since both
the items() list and the new dictionary just have references to the
same objects.  So you'll only have space allocated for the actual word
strings themselves once in the application.

Which can make this much less overhead than might otherwise be
considered at first glance.

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


From root at [127.0.0.1]  Sat Mar 31 14:59:22 2001
From: root at [127.0.0.1] (deadmeat)
Date: Sat, 31 Mar 2001 19:59:22 GMT
Subject: list vs tuple
References: <hGlw6.782$p5.3210@news1.rivrw1.nsw.optushome.com.au><3AC20CD3.265A2D94@alcyone.com><ru69cto61jtmrqsf809e38hb3pbm2lhi7g@4ax.com><bC1x6.4319$4N4.905946@newsc.telia.net><71q9ct4le63anefmiajj5cbh3mg04qu1ga@4ax.com><mailman.985991202.5578.python-list@python.org><7Bfx6.989$p5.3791@news1.rivrw1.nsw.optushome.com.au><Ukhx6.4421$4N4.950603@newsc.telia.net> <mailman.986049281.32365.python-list@python.org>
Message-ID: <uGqx6.1052$p5.3997@news1.rivrw1.nsw.optushome.com.au>

> - There is an difference between "copy" and "deep copy". What happens if
> you make copy a structure that contains pointers in it?

then you got a problem :)

but thats not the point - which when the statement a = b copies the data
across for one type it's expected to do it for others.  It doesn't, so it's
inconsistant, contrary to the original claim.





From nomad*** at ***freemail.absa.co.za  Fri Mar 30 08:15:16 2001
From: nomad*** at ***freemail.absa.co.za (nomad)
Date: Fri, 30 Mar 2001 13:15:16 GMT
Subject: Couple of simple questions
Message-ID: <3ac4854a.1170697@news.is.co.za>

Howzit,

I have a couple of (hopefully) simple questions:

Does anybody know what size python variabel types are?
Eg. how many bytes is and integer?  What about a long?  I have
searched through the docs but I cant find a reference to the byte
sizes.  The reason is that I'm battling with struct.py and some
complex C++ structs.  Any help here would greatly appreciated

Also on and win32 platform, how do I persistently add a custom path to
the pythonpath?  I have tried adding it to the pytho path in the
registry, but then when I view the class browser in IDLE, only the
default paths are there.  I can do a `sys.path.append.....' each time,
but I would prefer to have my custom module path there automatically.

Cheers and TIA.


From trace at reinventnow.com  Wed Mar 28 00:38:32 2001
From: trace at reinventnow.com (trace s. ruggles)
Date: Wed, 28 Mar 2001 05:38:32 GMT
Subject: Problems installing MySQLdb...
Message-ID: <B6E6D582.1026%trace@reinventnow.com>

Hi, I'm having the darndest time installing MySQLdb in Python 2.0 running on
FreeBSD.

Below is the interaction I get.  /distutils/sysconfig.py is looking for
config/Makefile which doesn't exist, but I've never had a directory named
"config" at the top-level.  No matter how many times I've tried to
re-install Python.

Does anyone out there have a clue about what's going on?

Additional info: I'm running a virtual private server through Verio and I'm
trying to install this on a local version of Python (hence the
/usr/home/axiomfir prefix on everything).

--Trace

--
Trace S. Ruggles :: trace at reinventnow.com
512/894.4402 :: 512/845.9354

..........

axiomfire% ../../python setup.py install
running install
Traceback (most recent call last):
  File "setup.py", line 103, in ?
    extra_objects=extra_objects,
  File "/usr/home/axiomfir/usr/local/lib/Python-2.0/Lib/distutils/core.py",
line 138, in setup
    dist.run_commands()
  File "/usr/home/axiomfir/usr/local/lib/Python-2.0/Lib/distutils/dist.py",
line 829, in run_commands
    self.run_command(cmd)
  File "/usr/home/axiomfir/usr/local/lib/Python-2.0/Lib/distutils/dist.py",
line 848, in run_command
    cmd_obj.ensure_finalized()
  File "/usr/home/axiomfir/usr/local/lib/Python-2.0/Lib/distutils/cmd.py",
line 112, in ensure_finalized
    self.finalize_options()
  File 
"/usr/home/axiomfir/usr/local/lib/Python-2.0/Lib/distutils/command/install.p
y", line 255, in finalize_options
    (prefix, exec_prefix) = get_config_vars('prefix', 'exec_prefix')
  File 
"/usr/home/axiomfir/usr/local/lib/Python-2.0/Lib/distutils/sysconfig.py",
line 372, in get_config_vars
    func()
  File 
"/usr/home/axiomfir/usr/local/lib/Python-2.0/Lib/distutils/sysconfig.py",
line 284, in _init_posix
    raise DistutilsPlatformError, my_msg
distutils.errors.DistutilsPlatformError: invalid Python installation: unable
to open /usr/home/axiomfir/usr/local/lib/Python-2.0/config/Makefile (No such
file or directory)
axiomfire% 



From aleaxit at yahoo.com  Wed Mar  7 03:21:38 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Wed, 7 Mar 2001 09:21:38 +0100
Subject: C/C++ extension: subclassing an extension type with Python code?
References: <ytjp6.25967$jg1.598879@news1.alsv1.occa.home.com>
Message-ID: <984r2s0tu0@news1.newsguy.com>

"Regular Expression" <xeger at usa.net> wrote in message
news:ytjp6.25967$jg1.598879 at news1.alsv1.occa.home.com...
> I'd like to create an extension type which can be further extended by end
> users of the type. Ideally, I'd like to define certain properties that are
> read-only, and others that cannot be occluded in "subclasses" of the type.
> I'd also like to define certain methods that cannot be overridden.
>
> Does anybody have an idea how to do this?

www.boost.org has an EXCELLENT start for you -- the Boost Python
Library lets you expose your C++ classes to Python in a way that
lets the Python users inherit from them (extend, override, &c).

It uses the deep and wondrous 'Beaudry Hook' for that purpose,
but, blessedly, you don't have to understand its intricacies to
just do what comes natural and Pythonic -- unimpeded subclassing.

You may have to dig very deep into the BPL internals to find ways
to selectively STOP overriding for certain things -- that's very
unPythonic, and liable to be a real mess.  But, at least you have
a start -- the BPL source code itself.  By just using BPL you'll
get all the real functionality -- 99% of your work will then be
in trying to make Python into what it isn't, a language which
protects the programmer against himself/herself (but, while you're
slaving away on that quixotic track, you may already release your
completely-functional code, "stopping" unintended overriding the
Pythonic way -- by DOCUMENTING that such-and-such method is not
to be overridden, naming it with a leading underscore if it's for
internal-use-only, and that's it).


Alex





From aleaxit at yahoo.com  Fri Mar 30 06:38:20 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 30 Mar 2001 13:38:20 +0200
Subject: list vs tuple
References: <mailman.985789357.19622.python-list@python.org> <3ac40acd$0$10576@wodc7nh0.news.uu.net>
Message-ID: <9a1r7k02spo@news1.newsguy.com>

"EuphoriaDj" <eddthompson at ssi.nospamhere.parlorcity.com> wrote in message
news:3ac40acd$0$10576 at wodc7nh0.news.uu.net...
>
> "Simon Brunning" <SBrunning at trisystems.co.uk> wrote in message
> news:mailman.985789357.19622.python-list at python.org...
> > > From: deadmeat [SMTP:root@[127.0.0.1]]
> > > Whats the practical difference between a list[] and a tuple() ?
> > >
> > > A tuple seems to be a very basic list, so unless it's faster, what's
the
> > > point of it?
> >
> > Off the top of my head - A tuple is immutable, and therefore hashable,
and
> > therefore can be used as a key to a dictionary. I'm sure that there are
> > other things...
> >
> what do you mean by hashable?

Uh, what about "able to be hashed", where "hashed" is the
past participle of verb "to hash"?  Python has a builtin
function "hash", which accepts only "hashable" objects,
and returns an integer:

>>> hash((1,2,3))
-821448277
>>> hash([1,2,3])
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
TypeError: unhashable type
>>>

As you see, you can hash a tuple (a tuple is hashable) but
not a list (it's unhashable).  As Simon goes on to say,
being hashable is equivalent to being usable as a key to
a dictionary (since the dictionary internally uses hash
on its keys).


Alex





From michael at stroeder.com  Thu Mar  8 14:36:36 2001
From: michael at stroeder.com (Michael =?iso-8859-1?Q?Str=F6der?=)
Date: Thu, 08 Mar 2001 20:36:36 +0100
Subject: Ldap / Python help
References: <mailman.984067391.2391.python-list@python.org>
Message-ID: <3AA7DF44.B5DDC75A@stroeder.com>

Pablo Pernot wrote:
> 
> Where can I find documentation about ldap/python

BTW: All the methods have __doc__ strings.

> >>> l.modify_s(dn,mylist)

Try:

print l.modify_s.__doc__

Ciao, Michael.


From cognite at zianet.com  Thu Mar  1 21:38:01 2001
From: cognite at zianet.com (editor@pythonjournal.com)
Date: Thu, 01 Mar 2001 21:38:01 -0500
Subject: PythonJournal.com becomes Interactive, new issue today
Message-ID: <E14YfSD-0003gZ-00@mail.python.org>

A new, _interactive_, issue of The Python Journal is appearing at
PythonJournal.com website 
(<A HREF="http://PythonJournal.com">PythonJournal.com>a/>) today. 

NB: Incorporating new ***interactive feature sections***, it
shows off (and tests) the python programming language's web application,
scripting, prototyping and graphics capabilities. Focus of all parts of
the PythonJournal centers on python. PythonJournal's new format is
designed to ease collaborating among people in the fast-growing and
far-flung python community, to complement and bolster existing activity.

This special transitional issue will be changing before your eyes. Gems
and coverage of the big Python 9 conference March 5-8 have been
promised. What with conference sessions running up to 3 deep in
parallel, PythonJournal hopes to post nightly updates -- maybe with
streaming -- to help anybody anywhere in the world know about new
capacities brought out at the conference. 

Major articles, known from previous issues to be of permanent interest,
will now be able to get refereed status through peer review. Invited
articles and contributed short pieces will have their place, and
BBS-style Leader position papers with Commentaries will provide a
provocative and hopefully efficient way of complementing existing python
activity. Code discussions will be up soon.

You can input 24x7 using various Interactive python apps on the
PythonJournal.com website, some light-duty (check out the python
proverbs section), and some heavy-duty, with capacity for images and
other goodies. Interesting to try them out.

Come parley with Pyxj, PythonJournal's new flying python icon <wink>.







<P>
A new, _interactive_, issue of The Python Journal is appearing today at 
website PythonJournal.com 
(<A HREF="http://PythonJournal.com">PythonJournal.com>a/>). 
Dialog apps (in python) supplement articles short, long, and
Leader+Commentaries style, which can now achieve refereed status.  Coverage
of Python9 (March 2001 conference in Long Beach CA) discoveries included.
(01-03-01)




From scarblac at pino.selwerd.nl  Tue Mar 13 20:57:02 2001
From: scarblac at pino.selwerd.nl (Remco Gerlich)
Date: 14 Mar 2001 01:57:02 GMT
Subject: Einstein's Riddle
References: <oNPq6.396644$ge4.138479566@news2.rdc2.tx.home.com> <wi6elw1pg5y.fsf@merlin.irisa.fr> <3AAE9730.5A7A4249@brazee.net>
Message-ID: <slrn9atjsc.5j.scarblac@pino.selwerd.nl>

Howard Brazee <howard at brazee.net> wrote in comp.lang.python:
> This is cross-posted.  Please tell us which language you are using.
> 
> Yoann Padioleau wrote:
> 
> > "Steve Graham" <js.graham at home.com> writes:
> >
> > > Who would be interested in using his/her brain (and his computer) to solve
> > > the following?
> > here it is:
> >
> > %----------------------------------------------------------------
> > % a sicstus program

   hint ^^^^^^^
	
	
Sicstus Prolog. Seems a nice choice for this sort of problem.

-- 
Remco Gerlich


From francois.granger at free.fr  Fri Mar  9 12:18:08 2001
From: francois.granger at free.fr (=?ISO-8859-1?Q?Fran=E7ois_Granger?=)
Date: Fri, 09 Mar 2001 17:18:08 GMT
Subject: Can anyone recomend a good intoduction to C...
References: <3aa366c6_2@nnrp1.news.uk.psi.net> <983887378.363030@newsmaster-04.atnet.at> <mailman.983894288.19965.python-list@python.org> <3AA5EF59.507F1D91@olsen.ch> <983958878.976344@newsmaster-04.atnet.at> <989cvj$sba$1@panix3.panix.com> <984132875.222868@newsmaster-04.atnet.at>
Message-ID: <1eq0a7k.1yanuaitkzmn4N%francois.granger@free.fr>

Werner Schiendl <ws-news at gmx.at> wrote:

> What's TMTOWTDI, please?

There is More Than One Way To Do It

wich is a Perl motto as far as I know.

-- 
"La vie est une maladie sexuellement transmissible  dont l'issue est
g?n?ralement fatale." - Inconnu, Signature de Noelle Adam


From sholden at holdenweb.com  Wed Mar  7 18:51:44 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Wed, 7 Mar 2001 18:51:44 -0500
Subject: UserDict and items()
References: <98656v$cud$1@news.rrz.Uni-Koeln.DE>
Message-ID: <NTzp6.29924$ML1.1507766@e420r-atl2.usenetserver.com>

"Oliver Hofmann" <a2619725 at smail.uni-koeln.de> wrote in message
news:98656v$cud$1 at news.rrz.Uni-Koeln.DE...
> Hello everyone!
>
>
> Been toying around with UserDict lately and stumbled into a problem.
> Here's
> the relevant code:
>
> ---
> from UserDict import UserDict
>
> class ClText(UserDict):
>     def __init__(self):
>         UserDict.__init__(self)
>
>         # Set initial value
>         self.data['test'] = ''
>
Since __getitem__() is checking for this value, why bother to add it to the
dictionary? Just so items() includes the key? As you see later, when you use
__getitem__() to fake a value for a given key it doesn't appear in items(),
which is just delegated to the underlying dictionary.

>
>     def __getitem__(self, key):
>         if key == 'test':
>             return 'found'
>
>         return UserDict.__getitem__(self, key)
> ---
>
> In short, whenever a few reserved keys are queried I'd like to return
> default values or fetch needed values from a database. This works
> quite nicely:
>
> >>> a = class.ClText()
>
wouldn't

a = ClText() be better? "class" is a reserved word.

> >>> a['test']
> 'found'
>
So you can see that your __getitem__() call is working.

> Unfortunately:
>
> >>> a.items()
> [('test', '')]
>
> Is there a more elegant way to handle items(), values() etc than
> overwriting each function?
>
> One way would be to set data[key] to the required value during
> init(); however I'd like to postpone fetching data from the database
> until the time it is actually requested.

Such laziness is fine, but it does imply that calls to items(), values() are
going to have to trigger the lazy fetch, and therefore cannot simply be
inherited, I suspect.

>
> Any help would be appreciated,
>
>
I have an idea that a further description of the surrounding application
might allow refactoring of your code to give rthe interface you want.

regards
 Steve




From cce at clarkevans.com  Tue Mar 20 00:55:56 2001
From: cce at clarkevans.com (Clark C. Evans)
Date: Tue, 20 Mar 2001 00:55:56 -0500 (EST)
Subject: Functionalism, aesthetics Was:(RE: I come to praise .join, not
 to bury it...)
In-Reply-To: <000101c0ae5d$7968b270$a9807018@cr706570a>
Message-ID: <Pine.LNX.4.21.0103200043050.21343-100000@clarkevans.com>

Alex Martelli [mailto:aleaxit at yahoo.com] wrote:
> it DOES mean each of these professionals need not fall back
> on vague generalities and empty hand-waving if and when their
> design choices get challenged.

I've been confronted on too many occasions where my 
"aesthetics" differed from another party's.  Perhaps
our functions were different.  But in any case, I was
able to clearly articulate what my functions were and
why the form I choose supported those functions.
The other party typically rests soley on dogma, 
giving in to "god", "universal patterns", or "beauty" 
that I obviously don't comprehend.  Oh well.  

> Somebody who's unable to do anything but talk of aesthetics
> when discussing a choice in the architecture of a software
> component is simply not contributing to the discussion.  It
> looks good to you, it looks bad to him, it looks indifferent
> to her, yawn, great debate, now let's get beyond this to the
> REAL meat of the issue, please -- the technical implications
> of this, that, or the other choice in a specific matter.

Most often those bringing up aesthetics need to go
and do their homework, detailing those functions that
they would expect or like to have from the software.
And then an analysis as to how each form satisfies those
particular functions. 

Another approach, is a user community survey, explaining 
both sides, the options and implications, and asking 
each user to think deeply before casting a vote.  Usually
if there is any core "aesthetic" it will pop out of
a stastical survey.  

> If you _can't_ get down to concrete specifics, then either
> you're not investing enough time & energy to contribute
> (in which case, staying out of a debate WOULD be a nice
> matter of courtesy), or you lack some specific (experience,
> culture, professional background, whatever) to be able to
> analyze and express your surface impressions in the issue --
> in which case, it's quite possible that they're misfounded,
> and, IF the huge investment of time and energy doesn't
> faze you, in-depth analysis to see if they ARE might well
> be warranted (and that is basically what I've been doing
> throughout this thread -- striving to help others articulate
> their objections, in great detail, by first laying out in
> just as much details my reason for NOT objecting at all to
> the architecture Guido has chosen for .join; it won't of
> course have any direct repercussion on how Python behaves
> in the matter, but it MAY have positive implications on
> technical growth of the people involved, and thus, very
> indirectly, on future efforts on their part).

Nice summary.

Clark 




From tim at vegeta.ath.cx  Tue Mar  6 02:30:15 2001
From: tim at vegeta.ath.cx (Tim Hammerquist)
Date: Tue, 06 Mar 2001 07:30:15 GMT
Subject: gah! I hate the new string syntax
References: <97ohh2$polt0$1@ID-77331.news.dfncis.de> <XFMail.20010302085406.shaleh@valinux.com> <mailman.983843650.6177.python-list@python.org>
Message-ID: <slrn9a94mq.2vi.tim@vegeta.ath.cx>

Skip Montanaro <skip at mojam.com> wrote:
> (In reality, I actually do my best to make my Perl code look as much like
> Python to my Python-infected mind as I can.)

You are probably selling yourself short as far as Perl is concerned.
While I will never say that Perl is better than Python (nor the other
way 'round, for that matter), trying to use Pythonic syntax in Perl can
make the code run hundreds of times slower than taking advantage of Perl
idioms.  If idioms are too much of an obstacle, maybe Python's a better
choice for you.

-- 
-Tim Hammerquist <timmy at cpan.org>

Big egos are big shields for lots of empty space.
	-- Diana Black


From slhath at home.com  Tue Mar  6 10:05:43 2001
From: slhath at home.com (Scott Hathaway)
Date: Tue, 06 Mar 2001 15:05:43 GMT
Subject: resize a tkinter window to the size of a gif
Message-ID: <b17p6.140400$bb.1468273@news1.rdc1.tx.home.com>

I have a tkinter window that I want to resize to the size of a background
image I am placing on it.  The window always seems to have a border of a few
pixels all the way around it.  Can anyone tell me how to get rid of this?
The code I am using is below:

frmMain = Pmw.initialise()
iPath = mainPath + themePath + '\\'
imgBG = PhotoImage(file=iPath + imgSettings[1])
frmMain.width = imgBG.width()
frmMain.height = imgBG.height()
theCanvas = Canvas(frmMain,width=frmMain.width,height=frmMain.height)
theCanvas.pack(expand=0)
theCanvas.create_image(0,0,anchor=NW,image=imgBG)

Any help would be appreciated.

TIA,
Scott




From mike.mcleod at bt.com  Fri Mar 30 06:14:14 2001
From: mike.mcleod at bt.com (mike mcleod)
Date: Fri, 30 Mar 2001 12:14:14 +0100
Subject: HELP urllib POST can't get it to work
References: <9a1m91$fdg$1@pheidippides.axion.bt.co.uk>
Message-ID: <9a1puq$gnf$1@pheidippides.axion.bt.co.uk>

Also if I try to add the params as as a string:

params = 'address==07&message=test&...
f = urllib.urlopen("http://www.samemachine.uk/bin/afunc.pl?", params)

I get:
<HTML><body><p><b>Error occured on server. Couldnt open result file,

Mike McLeod.




From bos at hack.org  Wed Mar 28 23:15:50 2001
From: bos at hack.org (Rikard Bosnjakovic)
Date: Thu, 29 Mar 2001 06:15:50 +0200
Subject: list vs tuple
References: <hGlw6.782$p5.3210@news1.rivrw1.nsw.optushome.com.au> <3AC1EB93.7BC87292@hack.org> <wEmw6.785$p5.3426@news1.rivrw1.nsw.optushome.com.au> <3AC1F93A.BAC9E34B@hack.org> <99teuu02p5f@news1.newsguy.com> <O4tw6.6798$m04.497167@e420r-atl1.usenetserver.com> <k835ctknej8rhhprvg7cr91cd63iu291qr@4ax.com>
Message-ID: <3AC2B6F6.B3220A75@hack.org>

Sheila King wrote:

> :So, tuples are not only immutable, they are also uncooperative.
> Tuples can be used as keys in dictionaries. Lists cannot.

You missed the pun.


-- 
Rikard Bosnjakovic - http://a214.ryd.student.liu.se/cv/ - ICQ: 1158217

Anyone sending unwanted advertising e-mail to my address will be
charged $250 for network traffic and computing time. By extracting my
address from this message or its header, you agree to these terms.


From dsavitsk at e-coli.net  Mon Mar 12 18:10:05 2001
From: dsavitsk at e-coli.net (dsavitsk)
Date: Mon, 12 Mar 2001 17:10:05 -0600
Subject: tkinter entry index
Message-ID: <iHcr6.27192$tV.5424761@newsrump.sjc.telocity.net>

how can i get the position of the cursor in an entry widget? .index and
.icursor don't seem to it. That is, if "foo" is in an entry, and the cursor
is between the "o" and the other "o", I could get a return of 2.

thanks,
doug

also, is there a tkinter FAQ? I feel like everytime I start a new project
these sorts of things come up.




From sandysj at asme.org  Tue Mar 27 11:21:14 2001
From: sandysj at asme.org (Jeff Sandys)
Date: Tue, 27 Mar 2001 16:21:14 GMT
Subject: stripping cr/lf, lf, cr
References: <Lp2w6.705$p5.3213@news1.rivrw1.nsw.optushome.com.au>
Message-ID: <3AC0BDFA.4C9F6E83@asme.org>

deadmeat wrote:
> 
> Is there a platform independant function that will remove appropriate EOL
> characters from a string?  I don't want to use strip() since it will also
> remove other whitespace I want to keep.

Here is how I do it in windows and unix:

def trimcr(str):
	if str == '':
		return ''
	if str[-1:] in '\n\r':
		return trimcr(str[:-1])
	else:
		return str

Thanks,
Jeff Sandys


From ylee12 at uiuc.edu  Thu Mar  1 14:20:25 2001
From: ylee12 at uiuc.edu (Young-Jin Lee)
Date: Thu, 1 Mar 2001 13:20:25 -0600
Subject: [Q] Question from Python tutorial
Message-ID: <Lgxn6.6964$dL4.91439@vixen.cso.uiuc.edu>

Hi, I have a question from Python tutorial.
The code segment in the page 60 (chapter 9. classes) is not executed.
I don't know what I did wrong. Here is the code and the error I got.

class MyClass:
    "A simple example class"
    i = 12345
    def f( x ):
        return 'hello world'

x = MyClass()

x.counter = 1
while x.counter < 10:
    x.counter = x.counter * 2
print x.counter
      ^
SyntaxError: invalid syntax

There is no typo, I have no idea of where this SyntaxError came from.

Thanks in advance. Any comment would be greatly appreciated.

Young-Jin




From chrisw at nipltd.com  Thu Mar  8 09:32:50 2001
From: chrisw at nipltd.com (Chris Withers)
Date: Thu, 08 Mar 2001 14:32:50 +0000
Subject: Playing with .zip or .tar.gz files
References: <3AA795FA.5C222E25@nipltd.com>
Message-ID: <3AA79812.17C0C886@nipltd.com>

Chris Withers wrote:
> 
> Can anyone recommend a library to me for playing with .zip or .tar.gz files.
> 'playing with' includes adding files to them, removing files from, and creating
> them.

To answer my own question:

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

*grinz*

I love python....

Chris



From bill-bell at bill-bell.hamilton.on.ca  Tue Mar 20 11:18:56 2001
From: bill-bell at bill-bell.hamilton.on.ca (Bill Bell)
Date: Tue, 20 Mar 2001 11:18:56 -0500
Subject: Little red lines in PythonWin!
In-Reply-To: <985099968.574.91219.l8@yahoogroups.com>
Message-ID: <3AB73CA0.22338.C08FBD@localhost>

Someone was kind enough to point me toward a whois client. I'm 
new to Python and this is one of my first experiences of 
downloading source.

When I view the whois client using PythonWin I see that lots of the 
Python indenting is lost and that numerous red lines have been 
introduced which are (presumably) meant to inform me that 
something is wrong.

Please, what's the proper way of getting hold of Python?! How does 
one deal correctly with those darned lines?

Thanks for your help.



From shaleh at valinux.com  Fri Mar  2 12:07:53 2001
From: shaleh at valinux.com (Sean 'Shaleh' Perry)
Date: Fri, 02 Mar 2001 09:07:53 -0800 (PST)
Subject: feature wish
In-Reply-To: <EZPn6.325779$f36.11597531@news20.bellglobal.com>
Message-ID: <XFMail.20010302090753.shaleh@valinux.com>

On 02-Mar-2001 Dan Parisien wrote:
> it seems all the special purpose overloadable methods are to provide type 
> emulation through classes
> e.g.
> __nonzero__
> __contains__
> __setitem__
> etc.
> 
> it would be cool if there was another method provided that allowed a 
> programmer to overload the value returned by an instance as well.
> 
> e.g.
> 
> class example:
>         def __init__(self, name, data):
>                 self.name = name
>                 self.data = data
> 
>         def __value__(self):
>                 return self.data
> 
> eg = example("my name", (1,2,3))
> 
>>>>print eg
> (1,2,3)
> 
> You could get to the instance behind the value by using a function like
> instance(eg)
> 

have you looked into __repr__()?



From lars at pixar.com  Wed Mar  7 21:40:45 2001
From: lars at pixar.com (Lars Damerow)
Date: Wed, 7 Mar 2001 18:40:45 -0800
Subject: class object access in class definition?
In-Reply-To: <slrn9ac2sa.ah1.scarblac@pino.selwerd.nl>
References: <Pine.LNX.4.21.0103061556510.2015-100000@zero> <slrn9ac2sa.ah1.scarblac@pino.selwerd.nl>
Message-ID: <Pine.LNX.4.21.0103071835560.6130-100000@zero>

I figured that the my problem stems from when Python binds the class name to
the scope--when it encounters the "class" keyword, or when the class definition
completes. It seems like it happens when the class definition completes.

What I'm trying to do is automatically register the derived classes of a base
class at compile time, like this:

derivedClasses = []

class BaseClass:
    pass

class DerivedClass(BaseClass):
    derivedClasses.append(DerivedClass)

This doesn't work because DerivedClass doesn't seem to be bound at the time of
the append() call. I can put the append() call in the DerivedClass's
constructor, but that requires an instance to be created before the class can
be registered.

If this isn't possible, it's no big deal; its goal is to avoid duplicating
information in the code, nothing more.

thanks!
-lars

On 7 Mar 2001, Remco Gerlich wrote:

| Lars Damerow <lars at pixar.com> wrote in comp.lang.python:
| > Is it possible to reference a class from within its own definition? Something
| > along the lines of:
| > 
| > class Foo:
| >     print Foo
| > 
| > This works for functions--that is,
| > 
| > >>> def foo():
| > ...     print foo
| > ...
| > >>> foo()
| > <function foo at 0x8111024>
| > 
| > works as expected, but the class definition above raises a NameError:
| > 
| > >>> class Foo:
| > ...     print Foo
| > ...
| > Traceback (most recent call last):
| >   File "<stdin>", line 1, in ?
| >   File "<stdin>", line 2, in Foo
| > NameError: There is no variable named 'Foo'
| 
| The difference is that the function is executed *after* you defined it, so
| that by then the name exists. In the case of the class, it is executed
| during its definition, and at that time the name doesn't exist yet.
| 
| I have no idea what you're trying to do, but maybe a workaround is to create
| an empty class of another name first, and inherit from that. You can then
| use references to that class, and be able to influence the current class.
| 
| But well, no idea what you're trying to do :)
| 
| -- 
| Remco Gerlich
| 

___________________________________________________________
lars damerow
button pusher
lars at pixar.com

Eat Snacky Smores!



From lex at cc.gatech.edu  Tue Mar 27 18:42:49 2001
From: lex at cc.gatech.edu (Lex Spoon)
Date: 27 Mar 2001 18:42:49 -0500
Subject: Collection interfaces
References: <3a90fa97.0@news1.mweb.co.za> <MPG.15085517debf4b649896ca@news.earthlink.net> <zb%n6.31033$o%6.586026@news1.rdc1.bc.home.com> <o1jv6.9635$P4.789083@newsread1.prod.itd.earthlink.net> <99l6s8$gpu$1@nyheter.chalmers.se>
Message-ID: <m38zlq4w34.fsf@chaos.resnet.gatech.edu>

matso at dtek.chalmers.se (Mats Olsson) writes:

> In article <o1jv6.9635$P4.789083 at newsread1.prod.itd.earthlink.net>,
> topmind <topmind at technologist.com> wrote:
> >
> >The last thing I need is yet more foriegn software on my PC making it crash
> >more and confusing other DLL's.
> 
>     Yea, use a Java applet instead. Guaranted not to fiddle with your
> DLL's :-)
> 


Oh great, so we get to fiddle with JAR's instead of DLL's!  At least
JAR is easier to pronounce.  :)


-Lex




From cribeiro at mail.inet.com.br  Mon Mar 26 14:13:21 2001
From: cribeiro at mail.inet.com.br (Carlos Alberto Reis Ribeiro)
Date: Mon, 26 Mar 2001 16:13:21 -0300
Subject: warrior geeks
In-Reply-To: <3ABF72F1.98CB78F@sage.att.com>
Message-ID: <5.0.2.1.0.20010326160822.021e5090@mail.inet.com.br>

At 16:48 26/03/01 +0000, Garry Hodgson wrote:
>it seemed like an odd coincidence, but it made me wonder: is there an
>unusually high percentage of MA practitioners among the python
>community?
>or the open source community?  or software in general?

Some time ago I've seen this same topic debated (I think that it was at 
Slashdot or some similar site). If I remember well, someone got to talk 
about some Open Source gurus that are Martial Arts pratictioners (I think 
that someone cited Eric S. Raymond and other well know guys -- my memory 
isn't that good, so take it with a large grain of salt <wink>)

Anyway I think that this is mostly related to the open, inquisitive nature 
of the geek culture. I myself have never practiced MA, but I was deeply 
involve in oriental philosophy for several years - Budhism, Zen, Tao, I 
Ching, tea ceremonies and other assorted things... And I would like to 
learn Kendo - if I could find a good teacher in my city.


Carlos Ribeiro





From tanzer at swing.co.at  Mon Mar 26 00:55:03 2001
From: tanzer at swing.co.at (Christian Tanzer)
Date: Mon, 26 Mar 2001 07:55:03 +0200
Subject: static methods 
In-Reply-To: Your message of "Sun, 25 Mar 2001 18:50:49 +0200."
             <99l7mu01ed6@news1.newsguy.com> 
Message-ID: <m14hPy3-000wcBC@swing.co.at>

"Alex Martelli" <aleaxit at yahoo.com> wrote:

> "Rick Lee" <rwklee at home.com> wrote in message 
> news:3ABD7673.9D447A32 at home.com...
> > Thanks, and thanks to Alex.  This is very useful, also very telling that
> it
> > is pretty obscure.  For me, static, or class-wide methods are very useful.
> > With this, you only have to pass the class reference to the caller.
> > Otherwise, you have to pass both the class reference and the module
> > reference to the caller.
> 
> Actually, the module reference will suffice; it will contain suitably
> named callables to:
>     -- generate instances of the class (the class-object itself and/or
>         suitable factory functions),
>     -- do whatever else you need to be done class-wise.
> 
> You can package up those callables any way you want; there is
> really no special reason they should be attributes of the class
> object as opposed to that of any other object, and a module
> object is typically handiest.

This seems to be the standard answer on c.l.py. I'm not convinced.

There are two cases:

- Modules defining a single class.

  In this case, most people would use the same name for both module
  and class. If you need the module object for scoping purposes, you
  end up with code using lots of `Bletch.Bletch' which I find
  atrocious.

  Rick also talked about passing around a reference to the class. If
  you pass a reference to a module instead, you need a convention how
  to refer to the class defined by the module.
  `getattr (module_reference, module_reference.__name__)' isn't any
  better than `Bletch.Bletch'. Another possibility would be to use
  `Module.Class' but that doesn't seem like a good idea, either (it 
  definitely smells of Ada-83 to me).

- Modules defining multiple classes. In this case, passing a module
  reference instead of a class reference won't fly, anyway.

  You could dismiss modules with multiple classes out of hand, but
  IMNSHO this would be rash (and inconsistent with your approval of
  Lakos' ideas about large scale design).
  
> The desire to use the class-object specifically, rather than any
> other, typically seems to have more to do with familiarity with
> other languages where classes must play certain special roles
> (C++, Java, Smalltalk, Eiffel) than with any actual programming
> needs.

I used languages with strong emphasis on the module as primary
construct (like Modula-2 and Ada) and languages favoring the
class as the primary construct (like Eiffel and C++). I found the
latter much better suited to my programming needs. That doesn't mean
that I shun modules, though. Fortunately, Python doesn't force either
direction (with some tricks to get class functions).

-- 
Christian Tanzer                                         tanzer at swing.co.at
Glasauergasse 32                                       Tel: +43 1 876 62 36
A-1130 Vienna, Austria                                 Fax: +43 1 877 66 92




From emile at fenx.com  Thu Mar  8 16:45:04 2001
From: emile at fenx.com (Emile van Sebille)
Date: Thu, 8 Mar 2001 13:45:04 -0800
Subject: Who am I: can a class instance determine its own name?
References: <mailman.984034150.9635.python-list@python.org> <slrn9af71n.d6p.scarblac@pino.selwerd.nl> <mailman.984081385.20244.python-list@python.org>
Message-ID: <988uj3$t5f1$1@ID-11957.news.dfncis.de>

Now that you know exactly why you shouldn't,


class Bar:
    pass

class Foo:
    " container class for Bars "
    def __init__(self):
        self.barList = []

    def addBar(self, bar):
        import traceback
        from string import split
        callingCode = traceback.extract_stack()[-2][3]
        name = split(split(callingCode, 'addBar(')[1],
')')[0]
        self.barList.append(name)

    def listBars(self):
        return self.barList

if __name__ == '__main__':
    Foo_One = Foo()
    Foo_Two = Foo()

    Bar_A = Bar()
    Bar_B = Bar()

    Foo_One.addBar(Bar_A)
    Foo_One.addBar(Bar_B)

    Foo_Two.addBar(Bar_B)

    print Foo_One.listBars()
    #...returns a sequence of Bar instance names:
['Bar_A','Bar_B']
    print Foo_Two.listBars()
    #   returns a sequence of Bar instance names: ['Bar_B']



use-at-your-own-risk-ly y'rs

--

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


"Tim Churches" <tchur at optushome.com.au> wrote in message
news:mailman.984081385.20244.python-list at python.org...
> Thanks to Gregory Jorgensen, Jeremy Reed, Steve Purcell,
Steve Holden,
> Matthias Oberlander, Daniel Klein and Remco Gerlich for
responding to my
> question: can a class instance determine the name of the
variable which
> refers to it? The answer appears to be "no, it can't". A
number of
> respondents pointed out that multiple variables may all
refer to the
> same class instance, so my question should have been: "can
a class
> instance determine the name(s) of the variable(s) which
refer(s) to
> it?". The answer is still no, and that the solution is to
explicitly
> assign each instance a name attribute at instantiation.
>
> A number of people asked why I needed to know. Consider
the problem of
> composition. You define a container class Foo which can
contain many
> instances of the Bar class, which are added to Foo with
the addBar
> method.
>
> ######################################
> Foo_One = Foo(name="Foo_One")
> Foo_Two = Foo(name="Foo_Two")
>
> Bar_A = Bar(name="Bar_A")
> Bar_B = Bar(Name="Bar_B")
>
> Foo_One.addBar(Bar_A)
> Foo_One.addBar(Bar_B)
>
> Foo_Two.addBar(Bar_B)
>
> Foo_One.listBars()
> ...returns a sequence of Bar instance names:
['Bar_A','Bar_B']
> Foo_Two.listBars()
> ...returns a sequence of Bar instance names: ['Bar_B']
> ######################################
> 
> I was just trying to avoid the inelegant redundancy of
having to name
> each Bar instance twice - once when specifying the name of
the variable
> which refers to the Bar instance, and once as an attribute
of that
> instance (the use of a keyword parameter to __init__ to
specify the name
> is just for clarity).
>
> What I was aiming at was this sort of functionality:
>
> ######################################
> Foo_One = Foo(name="Foo_One")
> Foo_Two = Foo(name="Foo_Two")
>
> Bar_A = Bar()
> Bar_B = Bar()
>
> Foo_One.addBar(Bar_A)
> Foo_One.addBar(Bar_B)
>
> Foo_Two.addBar(Bar_B)
>
> Foo_One.listBars()
> ...returns a sequence of Bar instance names:
['Bar_A','Bar_B']
> Foo_Two.listBars()
> ...returns a sequence of Bar instance names: ['Bar_B']
> ######################################
>
> One solution would be:
>
> ######################################
> Foo_One = Foo(name="Foo_One")
> Foo_Two = Foo(name="Foo_Two")
>
> Foo_One.addBar(Bar(name="Bar_A"))
> Foo_One.addBar(Bar(name="Bar_B"))
>
> Foo_Two.addBar(Bar(name="Bar_B"))
>
> Foo_One.listBars()
> ...returns a sequence of Bar instance names:
['Bar_A','Bar_B']
> Foo_Two.listBars()
> ...returns a sequence of Bar instance names: ['Bar_B']
> ######################################
>
> ...but that just creates a different form of redundancy...
>
> The solution I have settled on is to accept the minor
inelegance and
> move on to more important matters...
>
> Cheers,
>
> Tim C
>




From stephen_purcell at yahoo.com  Wed Mar 21 15:12:46 2001
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Wed, 21 Mar 2001 21:12:46 +0100
Subject: Substitute for join()
In-Reply-To: <Mo7u6.58$Hp.2896@typhoon.aracnet.com>; from DanielK@jBASE.com on Wed, Mar 21, 2001 at 11:35:53AM -0800
References: <Mo7u6.58$Hp.2896@typhoon.aracnet.com>
Message-ID: <20010321211245.A30720@freedom.puma-ag.com>

Daniel Klein wrote:
> I've got to make a script (written in version 2.0) to work on a machine
> running 1.5.2. The only incompatible line is one that uses the
> "".join(mylist) method. To 'get out of the woods' quickly, I substituted it
> with
> 
> s = ''
> for element in mylist: s = s + element
> 
> Is this the most efficient way to code this?

No. Instead, try:

    import string
    s = string.join(mylist, '')

-Steve

-- 
Steve Purcell, Pythangelist
Get testing at http://pyunit.sourceforge.net/
Any opinions expressed herein are my own and not necessarily those of Yahoo



From mlh at idi.ntnu.no  Sat Mar  3 16:54:17 2001
From: mlh at idi.ntnu.no (Magnus Lie Hetland)
Date: Sat, 3 Mar 2001 22:54:17 +0100
Subject: Python 2.1b1 released
References: <mailman.983573285.14721.python-list@python.org>
Message-ID: <97rp3p$l5a$1@tyfon.itea.ntnu.no>

"Guido van Rossum" <guido at digicool.com> wrote in message
news:mailman.983573285.14721.python-list at python.org...
[snip]
> - Nested Scopes (PEP 227)[*] are now optional.  They must be enabled
>   by including the statement "from __future__ import nested_scopes" at
>   the beginning of a module (PEP 236).  Nested scopes will be a
>   standard feature in Python 2.2.

Finally! Proof positive of the infamous time-machine -- and we
all get to use it!

Yay :)

--

  Magnus Lie Hetland          http://www.hetland.org

 "Reality is what refuses to disappear when you stop
  believing in it"                 -- Philip K. Dick





From urner at alumni.princeton.edu  Sat Mar  3 02:18:46 2001
From: urner at alumni.princeton.edu (Kirby Urner)
Date: Fri, 02 Mar 2001 23:18:46 -0800
Subject: Adding a method to an instance
References: <G2%n6.360714$JT5.11606234@news20.bellglobal.com> <6621ato2vknc667m3iuqvmm8vcj0h99qqk@4ax.com> <Sc0o6.361217$JT5.11613656@news20.bellglobal.com> <ej41atcongof4vpi5i2vkrilalbdbc06m4@4ax.com>
Message-ID: <pd61atcceals21f7kj157cfrq75bp71aq4@4ax.com>

Kirby Urner <urner at alumni.princeton.edu> wrote:

Or here's a variation you may prefer -- I don't know your
context, but in this one, we just hard-code the name of 
the instance and therefore don't even bother to pass it 
as a parameter:

  >>> class test:
	def prn(self,msg):
	   self.message = msg
	   print msg

	   
  >>> o = test()
  >>> o.prn("1-2-3 testing")
  1-2-3 testing
  >>> o.message
  '1-2-3 testing'
  >>> def newMethod(newtext):
	o.added = newtext
	return o.message + newtext

  >>> o.newMethod = newMethod
  >>> o.newMethod(" trying again")
  '1-2-3 testing trying again'
  >>> o.added
  ' trying again'
  >>> dir(o)
  ['added', 'message', 'newMethod']

I think I like the first example better.  Maybe others
will offer yet more ways to skin the snake.

Kirby



From aleaxit at yahoo.com  Wed Mar 21 10:48:19 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Wed, 21 Mar 2001 16:48:19 +0100
Subject: Yet Another PEP:  Query Protocol Interface or __query__
References: <Pine.LNX.4.21.0103201837570.21343-100000@clarkevans.com> <998r4j02vmb@news2.newsguy.com> <Pine.LNX.4.21.0103202032080.14634-100000@clarkevans.com> <slrn9bgprm.rv.hzhu@mars.localdomain> <slrn9bgrsn.rv.hzhu@mars.localdomain> <mailman.985168527.16577.python-list@python.org>
Message-ID: <99aig001tfs@news2.newsguy.com>

"Clark C. Evans" <cce at clarkevans.com> wrote in message
news:mailman.985168527.16577.python-list at python.org...
    [snip]
> The rest of your post resonates deeply.  Is there some
> way to get a "simple" __query__ mechanism working
> without having the type system in place?

A way to prototype 'query'/'adapt' in current Python
itself, you mean?  Sure, that should be easy.


Alex





From aleaxit at yahoo.com  Fri Mar 30 07:44:53 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 30 Mar 2001 14:44:53 +0200
Subject: htmllib samples
References: <3AC2F48E.4EF77C71@aon.at> <gKCw6.4130$4N4.837469@newsc.telia.net> <9a05uv$png$0@216.39.170.247>
Message-ID: <9a1v4l0313l@news1.newsguy.com>

"Dave LeBlanc" <whisper at oz.net> wrote in message
news:9a05uv$png$0 at 216.39.170.247...
    [snip -- moved here an effbot quote that Dave has afterwards]:
> >note that when you think you need the htmllib module, you
> >usually want the sgmllib module:
    [snip]
> Ok, i'll bite - why would you want to use the significantly greater
> complexity of SGML if you think you want to do HTML?

sgmllib only covers a small subset of SGML, so you do NOT
incur any "significantly greater complexity" whatsoever.


Alex





From whisper at oz.net  Thu Mar 29 15:12:21 2001
From: whisper at oz.net (Dave LeBlanc)
Date: 29 Mar 2001 20:12:21 GMT
Subject: Building "non-standard" standard extensions
References: <719vbt8tlkb05h4n1mcvh65l0f0i62tgsj@4ax.com> <j4u24dm4ig.fsf@informatik.hu-berlin.de>
Message-ID: <9a04v5$n2q$0@216.39.170.247>

Of course, the downside is that once you integrate GPL code into your
Python, then it's tainted by the GPL and anything you build with it is
subsequently GPL'd. I think this is so because there was a lot of not
using GCC until they LGPL'd the libraries so that resulting code
wouldn't be GPL.

Regards,

Dave LeBlanc

On 29 Mar 2001 09:13:43 +0200, Martin von Loewis
<loewis at informatik.hu-berlin.de> wrote:

>Paul Moore <gustav at morpheus.demon.co.uk> writes:
>
>> However, I do have a binary version of the readline library
>> available. Can I build the readline module for Windows usingh this?
>> Ideally, I'd like to do this *without* having to rebuild the whole
>> of Python for Windows (I'm happy with just using the normal build
>> for everything else). I've never built Python for myself, so
>> apologies if the answer to this is "obvious"...
>
>Building the readline module alone should be possible. I'm not sure
>whether the interactive interpreter will use it - it only tries if
>stdin isatty.
>
>> (BTW, this doesn't just apply to readline - I have various other libraries
>> such as a curses port, which I wouldn't mind trying to integrate as well...)
>
>That should work as well.
>
>Regards,
>Martin



From rich_somerfield at tertio.com  Mon Mar  5 04:59:32 2001
From: rich_somerfield at tertio.com (Rich Somerfield)
Date: Mon, 5 Mar 2001 09:59:32 -0000
Subject: Help with finding how much memory a variable is using
References: <JAwl6.17037$5n4.338201@news6-win.server.ntlworld.com> <87n1b2vzke.fsf@pulsar.wlfdle1.on.wave.home.com> <mailman.983691965.7856.python-list@python.org>
Message-ID: <csJo6.10304$mt.1687377@news2-win.server.ntlworld.com>

The data is being stored in an array of integers.

The problem at the moment is that it was only a guess that the storage of
this data was causing my application to crash.  However after a bit of
hacking it suggests that there are other problems.  I really wanted to watch
memory usage of variables and to see if they were causing problems.

As the data is stored local to a class, and I assign a global variable to
the local variable (breaks scoping rules I know!).  would this data be
stored on the stack / heap?  Or does this suggest a fundamental pointer
problem??  Maybe i should make the data type global and do some tests on
that.

Unfortunately I am discovering bugs I never knew existed (my code is 2 years
old and it seemed to work fine then!!), so cant really perform any tests on
a reliable base system.

Regards
Rich
Neil Schemenauer <nas at arctrix.com> wrote in message
news:mailman.983691965.7856.python-list at python.org...
> "Rich Somerfield" <rich_somerfield at tertio.com> writes:
> > I am generating a huge, huge list of data (cant think of a
> > decent way to reduce the required storage size and have it in a
> > usable form).  Everytime i try to keep this list for future
> > derivations I get a memory problem from Windows (effectively
> > terminating my python script).
>
> Depending on the data in the list the array module may help (eg
> for storing an array of integers).
>
>   Neil
>




From phd at phd.pp.ru  Thu Mar 22 12:14:17 2001
From: phd at phd.pp.ru (Oleg Broytmann)
Date: Thu, 22 Mar 2001 20:14:17 +0300 (MSK)
Subject: Transparent GIFs in PIL
In-Reply-To: <20010322180110.A18476@freedom.puma-ag.com>
Message-ID: <Pine.LNX.4.30.0103222013420.31302-100000@fep132.fep.ru>

Hi!

   Tried to send the patch to PIL authors?

On Thu, 22 Mar 2001, Steve Purcell wrote:
> Marcos S?nchez Provencio wrote:
> > Is it possible to set a transparent color in a GIF file (originally not
> > transparent) in PIL?
>
> With PIL 1.0, not directly. For later versions (if they exist), I don't know.
> You can patch the PIL 1.0 version of 'GifImagePlugin.py' as follows:

Oleg.
----
     Oleg Broytmann            http://phd.pp.ru/            phd at phd.pp.ru
           Programmers don't die, they just GOSUB without RETURN.




From craig at patterncraft.com  Fri Mar  9 11:28:33 2001
From: craig at patterncraft.com (Craig Holman)
Date: Fri, 9 Mar 2001 10:28:33 -0600
Subject: Memory leaks in VC6 2 line Python 2.1b1 apps?
References: <3aa88f52.5419185@news.newsguy.com> <hGPp6.53719$UZ4.13963947@news4.rdc1.on.home.com>
Message-ID: <98b0a0$2g8i$1@shadow.skypoint.net>

Greetings,

I'm afraid that both versions 2.0 and 2.1b1 leak memory for a single
Py_Initialize, Py_Finalize sequence in a
Visual C++ MFC application.

I created a simple VC 6 MFC application using the wizard as follows:

Dialog shared MFC-DLL application
Run-time library debug multithreaded DLL
Added a Test button to the dialog
The handler for the Test button calls Py_Initialize and then Py_Finalize.

I used this application to test for leaks in both Python 2.0 and Python
2.1b1.

Python 2.0 had 84 memory leaks that resulted in a leakage of a total of
15,936 bytes for a single pair of
Py_Initialize, Py_Finalize calls. The addition of a second Py_Intialize,
Py_Finalize pair resulted in an additional
leak of at least 2,300 bytes.  The memory leak dump listing in the Visual
Studio debug window was truncated after those additional 2,300 bytes,
so I'm not sure how large the additional leak was for the second call, nor
can I tell you how much is leaked for additional pairs.

Python 2.1b1 had 21 memory leaks that resulted in a leakage of a total of
6,899 bytes for a single pair of
Py_Initialize, Py_Finalize calls.  Additional Py_Intitialize, Py_Finalize
pairs did not result in any
additional leakage (10,000 pairs were tested).

Based upon these experiments, I'd suggest using 2.1b1.

Python development is apparently aware of some of these leaks.
I found the following comment in pythonrun.c, lines 254-258 near the end of
the Py_Finalize procedure:

 /* XXX Still allocated:
    - various static ad-hoc pointers to interned strings
    - int and float free list blocks
    - whatever various modules and libraries allocate
 */

I am going to continue testing Python 2.1b1 for leaks when its driven from a
VC++ 6 MFC application.

I'm hoping that some of these leaks can be addressed before the 2.1 release
is issued.

The remainder of this note is an annotated listing of memory leaks that were
produced when running the
application with Python 2.1b1.  It is incomplete, but Owen's note prompted
me to post it early.
The addresses in the "normal block at" lines may not match up with the
addresses in the runtime stack
captures that I provide - I captured this information over several runs of
the program.

Craig
____________________________________________________________________________
_
Python 2.1b1 memory leaks as reported by a debug Visual C++ 6 MFC
application

Dumping objects ->

{1037} normal block at 0x0079CE80, 39 bytes long.
 Data: <  y             > 20 EE 79 00 88 85 1F 1E 03 00 00 00 B8 17 20 1E

 Cause :  interned string not being released

    Breakpoint:

  classobject.c, line 544 :  delstr = PyString_InternFromString("__del__");

 Runtime stack:

  instance_dealloc(PyInstanceObject * 0x00947e04) line 544
  _Py_Dealloc(_object * 0x00947e04) line 1367 + 7 bytes
  fini_exceptions() line 1139 + 70 bytes
  Py_Finalize() line 242


{1020} normal block at 0x0094B960, 6144 bytes long.
 Data: <   w Hy         > FF BD 07 77 B8 48 79 00 00 00 00 00 00 00 00 00

 Cause :

  /* Delete some special variables first.  These are common
     places where user values hide and people complain when their
     destructors fail.  Since the modules containing them are
     deleted *last* of all, they would come too late in the normal
     destruction order.  Sigh. */  (source: PyImport_Cleanup()  (I believe))

 Breakpoint:

     dictobject.c, line 404 :  newtable = PyMem_NEW(dictentry, newsize);

 Runtime stack:

  malloc(unsigned int 6144) line 130
  dictresize(dictobject * 0x00794bc4, int 342) line 404 + 13 bytes
  PyDict_SetItem(_object * 0x00794bc4, _object * 0x0079d4a0, _object *
0x0079d4a0) line 491 + 18 bytes
  PyString_InternInPlace(_object * * 0x0012f6cc) line 3171 + 20 bytes
  PyDict_SetItemString(_object * 0x00798ca4, char * 0x1e1f24e0, _object *
0x1e1f84f0 __Py_NoneStruct) line 1271 + 9 bytes
  PyImport_Cleanup() line 253 + 20 bytes
  Py_Finalize() line 213

{436} normal block at 0x0079EE20, 40 bytes long.
 Data: <  y   y         > B0 EA 79 00 80 CE 79 00 03 00 00 00 B8 17 20 1E

 Cause :  interned string not being released

    Breakpoint:

        classobject.c, line 474 :  initstr =
PyString_InternFromString("__init__");

 Runtime stack:

  PyInstance_New(_object * 0x00946a3c, _object * 0x0079cfa4, _object *
0x00000000) line 474
  call_object(_object * 0x00946a3c, _object * 0x0079cfa4, _object *
0x00000000) line 2705 + 17 bytes
  PyEval_CallObjectWithKeywords(_object * 0x00946a3c, _object * 0x0079cfa4,
_object * 0x00000000) line 2673 + 17 bytes
  init_exceptions() line 1123 + 24 bytes
  Py_Initialize() line 148

{427} normal block at 0x0079EAB0, 43 bytes long.
 Data: <H y   y         > 48 EA 79 00 20 EE 79 00 03 00 00 00 B8 17 20 1E

 Cause :  interned string not being released

    Breakpoint:

        classobject.c, line 97 :  delattrstr =
PyString_InternFromString("__delattr__");

 Runtime stack:

  PyClass_New(_object * 0x0079cfa4, _object * 0x0079dc3c, _object *
0x0079e7f8) line 97
  make_Exception(char * 0x1e1e9cfc) line 345 + 15 bytes
  init_exceptions() line 1073 + 9 bytes
  Py_Initialize() line 148

{426} normal block at 0x0079EA48, 43 bytes long.
 Data: <  y   y         > E0 E9 79 00 B0 EA 79 00 03 00 00 00 B8 17 20 1E

 Cause :  interned string not being released

    Breakpoint:

        classobject.c, line 96 :  setattrstr =
PyString_InternFromString("__setattr__");

 Runtime stack:

  PyClass_New(_object * 0x0079cfa4, _object * 0x0079dc3c, _object *
0x0079e7f8) line 96
  make_Exception(char * 0x1e1e9cfc) line 345 + 15 bytes
  init_exceptions() line 1073 + 9 bytes
  Py_Initialize() line 148

{425} normal block at 0x0079E9E0, 43 bytes long.
 Data: <  y H y         > B0 DD 79 00 48 EA 79 00 03 00 00 00 B8 17 20 1E

 Cause :  interned string not being released

    Breakpoint:

        classobject.c, line 95 :  getattrstr =
PyString_InternFromString("__getattr__");

 Runtime stack:

  PyClass_New(_object * 0x0079cfa4, _object * 0x0079dc3c, _object *
0x0079e7f8) line 95
  make_Exception(char * 0x1e1e9cfc) line 345 + 15 bytes
  init_exceptions() line 1073 + 9 bytes
  Py_Initialize() line 148

{416} normal block at 0x0079DDB0, 42 bytes long.
 Data: <  y   y         > A0 DE 79 00 E0 E9 79 00 03 00 00 00 B8 17 20 1E

 Cause :  interned string not being released

    Breakpoint:

        classobject.c, line 29 :  modstr=
PyString_InternFromString("__module__");

 Runtime stack:

  PyClass_New(_object * 0x00000000, _object * 0x0079dc3c, _object *
0x0079e7f8) line 29
  make_Exception(char * 0x1e1e9cfc) line 345 + 15 bytes
  init_exceptions() line 1073 + 9 bytes
  Py_Initialize() line 148

{406} normal block at 0x0079DEA0, 40 bytes long.
 Data: <  y   y         > D0 DC 79 00 B0 DD 79 00 03 00 00 00 B8 17 20 1E

 Cause :  interned string not being released

    Breakpoint:

        import.c, line 1584 :  pathstr =
PyString_InternFromString("__path__");

 Runtime stack:

  get_parent(_object * 0x0079dc3c, char * 0x0012f494, int * 0x0012f490) line
1584
  import_module_ex(char * 0x0079daac, _object * 0x0079dc3c, _object *
0x0079dc3c, _object * 0x0079db6c) line 1518 + 23 bytes
  PyImport_ImportModuleEx(char * 0x0079daac, _object * 0x0079dc3c, _object *
0x0079dc3c, _object * 0x0079db6c) line 1563 + 21 bytes
  builtin___import__(_object * 0x00000000, _object * 0x0079dd2c) line 31 +
21 bytes
  call_cfunction(_object * 0x00794e68, _object * 0x0079dd2c, _object *
0x00000000) line 2740 + 11 bytes
  call_object(_object * 0x00794e68, _object * 0x0079dd2c, _object *
0x00000000) line 2703 + 17 bytes
  PyEval_CallObjectWithKeywords(_object * 0x00794e68, _object * 0x0079dd2c,
_object * 0x00000000) line 2673 + 17 bytes
  PyObject_CallObject(_object * 0x00794e68, _object * 0x0079dd2c) line 1529
+ 15 bytes
  PyObject_CallFunction(_object * 0x00794e68, char * 0x1e1f2dfc) line 1570 +
13 bytes
  PyImport_Import(_object * 0x0079da90) line 1954 + 33 bytes
  PyImport_ImportModule(char * 0x1e1e9d08) line 1494 + 9 bytes
  init_exceptions() line 1060 + 10 bytes
  Py_Initialize() line 148

{399} normal block at 0x00795558, 12 bytes long.
 Data: <        \   > FF FF FF FF 00 00 00 00 5C 00 00 00

 Cause :  allocated lock is not released

 Breakpoint:

        thread_nt.h, line 240 :  aLock = AllocNonRecursiveMutex() ;

 Runtime stack:

  PyThread_allocate_lock() line 240
  lock_import() line 154 + 5 bytes
  PyImport_ImportModuleEx(char * 0x1e1f2de8, _object * 0x00000000, _object *
0x00000000, _object * 0x00000000) line 1563
  PyImport_Import(_object * 0x0079da90) line 1936 + 16 bytes
  PyImport_ImportModule(char * 0x1e1e9d08) line 1494 + 9 bytes
  init_exceptions() line 1060 + 10 bytes
  Py_Initialize() line 148

{398} normal block at 0x0079DCD0, 39 bytes long.
 Data: <t y   y         > 74 DC 79 00 A0 DE 79 00 01 00 00 00 B8 17 20 1E

 Runtime stack:



{397} normal block at 0x007953E8, 4 bytes long.
 Data: <  y > D0 DC 79 00

 Runtime stack:



{396} normal block at 0x0079DC68, 36 bytes long.
 Data: <    P y       y > F0 BC 1E 1E 50 CF 79 00 CD CD CD CD 00 DC 79 00

 Runtime stack:



{395} normal block at 0x0079DC00, 44 bytes long.
 Data: <\ y t y         > 5C CF 79 00 74 DC 79 00 03 00 00 00 B8 17 20 1E

 Cause :  interned string not being released

    Breakpoint:

        import.c, line 1915 :  builtins_str =
PyString_InternFromString("__builtins__");

 Runtime stack:

  PyImport_Import(_object * 0x0079da90) line 1915
  PyImport_ImportModule(char * 0x1e1e9d08) line 1494 + 9 bytes
  init_exceptions() line 1060 + 10 bytes
  Py_Initialize() line 148

{365} normal block at 0x0079CF50, 36 bytes long.
 Data: <h y  Ly      Oy > 68 DC 79 00 D8 4C 79 00 CD CD CD CD E0 4F 79 00



{280} normal block at 0x00797590, 66 bytes long.
 Data: <.\DLLs;.\lib;.\l> 2E 5C 44 4C 4C 73 3B 2E 5C 6C 69 62 3B 2E 5C 6C



{85} normal block at 0x00794FE0, 42 bytes long.
 Data: <@Ny \ y         > 40 4E 79 00 5C CF 79 00 03 00 00 00 B8 17 20 1E

 Cause :  interned string not being released

    Breakpoint:

        import.c, line 1912 :  import_str =
PyString_InternFromString("__import__");

 Runtime stack:

  PyImport_Import(_object * 0x0079da90) line 1912
  PyImport_ImportModule(char * 0x1e1e9d08) line 1494 + 9 bytes
  init_exceptions() line 1060 + 10 bytes
  Py_Initialize() line 148

{81} normal block at 0x00794E40, 39 bytes long.
 Data: < Ly  Oy         > E4 4C 79 00 E0 4F 79 00 03 00 00 00 B8 17 20 1E

 Cause :  interned string not being released

    Breakpoint:

        classobject.c, line 24 :  docstr=
PyString_InternFromString("__doc__");

 Runtime stack:

  PyClass_New(_object * 0x00000000, _object * 0x0079dc3c, _object *
0x0079e7f8) line 24
  make_Exception(char * 0x1e1e9cfc) line 345 + 15 bytes
  init_exceptions() line 1073 + 9 bytes
  Py_Initialize() line 148

{78} normal block at 0x00794CD8, 52 bytes long.
 Data: <P y         pLy > 50 CF 79 00 F0 BC 1E 1E CD CD CD CD 70 4C 79 00



{77} normal block at 0x00794C70, 40 bytes long.
 Data: < Hy  Ly         > B8 48 79 00 E4 4C 79 00 04 00 00 00 B8 17 20 1E

 Cause :  interned string not being released

 Breakpoint:

        import.c, line 1579 :  namestr =
PyString_InternFromString("__name__");

 Runtime stack:

  get_parent(_object * 0x0079dc3c, char * 0x0012f494, int * 0x0012f490) line
1579
  import_module_ex(char * 0x0079daac, _object * 0x0079dc3c, _object *
0x0079dc3c, _object * 0x0079db6c) line 1518 + 23 bytes
  PyImport_ImportModuleEx(char * 0x0079daac, _object * 0x0079dc3c, _object *
0x0079dc3c, _object * 0x0079db6c) line 1563 + 21 bytes
  builtin___import__(_object * 0x00000000, _object * 0x0079dd2c) line 31 +
21 bytes
  call_cfunction(_object * 0x00794e68, _object * 0x0079dd2c, _object *
0x00000000) line 2740 + 11 bytes
  call_object(_object * 0x00794e68, _object * 0x0079dd2c, _object *
0x00000000) line 2703 + 17 bytes
  PyEval_CallObjectWithKeywords(_object * 0x00794e68, _object * 0x0079dd2c,
_object * 0x00000000) line 2673 + 17 bytes
  PyObject_CallObject(_object * 0x00794e68, _object * 0x0079dd2c) line 1529
+ 15 bytes
  PyObject_CallFunction(_object * 0x00794e68, char * 0x1e1f2dfc) line 1570 +
13 bytes
  PyImport_Import(_object * 0x0079da90) line 1954 + 33 bytes
  PyImport_ImportModule(char * 0x1e1e9d08) line 1494 + 9 bytes
  init_exceptions() line 1060 + 10 bytes
  Py_Initialize() line 148

{67} normal block at 0x007948B8, 43 bytes long.
 Data: <    pLy         > 88 85 1F 1E 70 4C 79 00 B4 00 00 00 B8 17 20 1E

 Cause :  string is not being released

    Breakpoint:

        dictobject.c, line 130 :  dummy = PyString_FromString("<dummy
key>");

 PyDict_New() line 130
 Py_Initialize() line 116 + 5 bytes

{65} normal block at 0x007947F8, 12 bytes long.
 Data: <        X   > FF FF FF FF 00 00 00 00 58 00 00 00

 Cause :  allocated lock is not released

 Breakpoint:

        thread_nt.h, line 240 :  aLock = AllocNonRecursiveMutex() ;

 PyThread_allocate_lock() line 240
 PyInterpreterState_New() line 36 + 14 bytes
 Py_Initialize() line 107 + 5 bytes

Object dump complete.

"Tom" <NoSpam at NoSpam.com> wrote in message
news:hGPp6.53719$UZ4.13963947 at news4.rdc1.on.home.com...
> "Owen F. Ransen" <ransen_spam_me_not at nemo.it> wrote in message
> news:3aa88f52.5419185 at news.newsguy.com...
> > Python 2.1b1
> >
> > I tried to create a simple VC MFC app using the
> > wizard as follows:
> >
> > 1) Multidoc shared MFC-DLL application.
> > 2) Run time library debug multithreaded DLL
> > 3) Added #include <python.h> to the main file
> > 4) Added
> >         Py_Initialize()
> >         Py_Finalize()
> >    at the end of the InitInstance()
> > 5) When I run I get memory leaks visible at
> >    exit.
> >
> > Any clues anyone?
>
> For comparison sake.
> I use VC6 and Python 2.0.  I create an EXE that calls the same same
> functions that you mention, and Python itself doesn't leak (I get various
> wierd leaks from other sources, but not Python so far as I can tell).
>
> Tom.
>
> >
> > (I can't yet report this via Source Forge because my account
> >  problems have not be sorted out yet.)
> >
> >
> > --
> > Owen F. Ransen
> > http://www.ransen.com/
> > Home of Gliftic & Repligator Image Generators
>
>




From p.magwene at snet.net  Tue Mar 27 10:55:41 2001
From: p.magwene at snet.net (Paul Magwene)
Date: Tue, 27 Mar 2001 10:55:41 -0500
Subject: Whats in your $PYTHONSTARTUP$ ?
Message-ID: <FJ2w6.5358$GD.2921338@typhoon.snet.net>

I've been programming in Python for several years now.  Recently I realized
that I've never really taken advantage of being able to specify a default
startup file.

Other than import statements, what little programming gems do you folks have
in your startup files?  I thought it would be useful to draw on the
collective knowledge of the Python community.  If we get enough good ideas
it might be worth sticking up on a web page, or even distributing (in the
demos or Tools section) with a later release.

Cheers,

Paul Magwene
paul.magwene at yale.edu


My contribution is the following little snippet that I found useful.  It
makes it easier to navigate and explore directories from the interactive
interpretter, using the familiar 'cd' and 'ls' (change it to 'dir' if you're
a long time DOS user).


#-------------------------------------------------------------------
import sys, os, os.path
class DirLister:

    def __getitem__(self, key):
        s =  os.listdir(os.getcwd())
        return s[key]

    def __getslice__(self,i,j):
        s =  os.listdir(os.getcwd())
        return s[i:j]

    def __call__(self, path=os.getcwd()):
        path = os.path.expanduser(os.path.expandvars(path))
        return os.listdir(path)

    def __repr__(self):
        return str(os.listdir(os.getcwd()))


class DirChanger:
    def __init__(self, path=os.getcwd()):
        self.__call__(path)

    def __call__(self, path=os.getcwd()):
        path = os.path.expanduser(os.path.expandvars(path))
        os.chdir(path)

    def __repr__(self):
        return os.getcwd()


ls = DirLister()
cd = DirChanger()
#-------------------------------------------------------------------



From jkraska1 at san.rr.com  Sat Mar 24 20:42:12 2001
From: jkraska1 at san.rr.com (Courageous)
Date: Sun, 25 Mar 2001 01:42:12 GMT
Subject: static methods
References: <mailman.985383828.2942.python-list@python.org> <slrn9bpqgo.15q.grante@tuxtop.visi.com> <mailman.985457492.29908.python-list@python.org> <7h1qbt8h8aos1parc090s8tqu2ooc04t4d@4ax.com> <slrn9bqc2r.7nq.grante@isis.visi.com> <4lgqbto46c4c6i307lnq0e614uujd2fsss@4ax.com> <slrn9bqoet.1bs.grante@tuxtop.visi.com>
Message-ID: <gdjqbtkdtuj6og1s83la3g0p8oqjhqk541@4ax.com>

>>>I'm confused -- are you saying that C doesn't have namespaces?
>>
>>Erm, well, unless C99 defined them, no.
>
>There's a global namespace, a "file" namespace, and each set of
>{} defines a namespace (they're all nested).  Somehow I don't
>think that's what you're talking about...

No; I was talking about python's modules, Java packages, or C++'s
namespace, all of which have similar intent. Python's modules are
my favorite, simply because they occur automatically and naturally,
as a result of the language. While it might be somewhat inconvenient
to not divide up a namespace amongst multiple files, the added
benefits far more than make up for this cost. Python's relative terseness
to other programming languages also helps, keeping even the most
ambitious of Python's modules small, more often than not.

C//



From Eugene.Leitl at lrz.uni-muenchen.de  Wed Mar 28 12:14:12 2001
From: Eugene.Leitl at lrz.uni-muenchen.de (Eugene Leitl)
Date: Wed, 28 Mar 2001 19:14:12 +0200 (MET DST)
Subject: General CGI advice...
In-Reply-To: <01032811551700.00576@tos1>
Message-ID: <Pine.GSO.4.03.10103281912320.16808-100000@sun1.lrz-muenchen.de>

Try 
	http://www.devshed.com/Server_Side/Python/CGI/
	http://www.cs.virginia.edu/~lab2q/

Strangely enough, the first two hits from Google ;)

On Wed, 28 Mar 2001, Vincent A. Primavera wrote:

> Hello,
> 	Looking for some general guidance...  I am relatively new to Python and 
> pretty much totally green to CGI programming.  However, I would like to start 
> learning more about CGI using Python as my means.  Do I need to focus on 
> generic CGI concepts first or are there any really good resources available 
> to integrate the two immediately?




From roy at panix.com  Tue Mar  6 15:46:19 2001
From: roy at panix.com (Roy Smith)
Date: Tue, 06 Mar 2001 15:46:19 -0500
Subject: Can anyone recomend a good intoduction to C...
References: <mailman.983894288.19965.python-list@python.org> <983896344.701456@newsmaster-04.atnet.at> <m3lmqibw0a.fsf@atrus.jesus.cam.ac.uk> <983911749.217622@newsmaster-04.atnet.at>
Message-ID: <roy-32CDC5.15461906032001@news.panix.com>

"Werner Schiendl" <ws-news at gmx.at> wrote:
> Years ago I had a lot of discussions why use C instead of Assembler.
> The arguments were mainly for performance.

I wouldn't be surprised if a good optimizing C compiler can actually 
produce faster code than hand-coded assembler in many cases.


From aleaxit at yahoo.com  Fri Mar 23 03:29:30 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 23 Mar 2001 09:29:30 +0100
Subject: automatic delegation (was Re: PEP 245: Python interfaces)
References: <mailman.985334130.2362.python-list@python.org>
Message-ID: <99f1h901dd8@news2.newsguy.com>

"Clark C. Evans" <cce at clarkevans.com> wrote in message
news:mailman.985334130.2362.python-list at python.org...
    [snip]
>    d) What is also missing from Python is
>       an "auto-delegation".  An analysis

Is it (missing, I mean)?  That isn't clear to me --
please see, for example:
http://pythoncookbook.activestate.com/pythoncookbook/community/data/98525811
4

Maybe you mean you miss some favourite _syntax sugar_
of yours for automatic delegation...?


Alex





From shredwheat at mediaone.net  Fri Mar 30 12:44:04 2001
From: shredwheat at mediaone.net (Pete Shinners)
Date: Fri, 30 Mar 2001 17:44:04 GMT
Subject: Use site-packages on all platforms
References: <mailman.985955621.21096.python-list@python.org>
Message-ID: <EB3x6.114662$Ch.15365680@typhoon.we.rr.com>

"Moore, Paul" <Paul.Moore at uk.origin-it.com> wrote
> Attached is a first draft of a proposal to use the "site-packages" directory
> for locally installed modules, on all platforms instead of just on Unix. If
> the consensus is that this is a worthwhile proposal, I'll submit it as a
> formal PEP.
>
> Any advice or suggestions welcomed - I've never written a PEP before - I
> hope I've got the procedure right...

i heartily agree with this proposal. let's keep user installed
packages/modules/scripts out of the executable directory. it's
unclean.

although it is out of the scope for your pep, i'd also like to
get hear some initial ideas for standard documentation/examples
locations and formats. my package comes with many tutorials and
examples, but i haven't found a consistent place to store them
on all platforms.








From TCHUR at doh.health.nsw.gov.au  Tue Mar 20 22:22:45 2001
From: TCHUR at doh.health.nsw.gov.au (Tim CHURCHES)
Date: Wed, 21 Mar 2001 14:22:45 +1100
Subject: Base 2 to long integers and back
Message-ID: <sab8b93d.007@dohprimo.ccf.health.nsw.gov.au>

Can anyone suggest a fast method of converting base2 strings (e.g. '0000101110') to Python long integers, and back? I couldn't locate any functions in the standard library to do this, but no doubt I am overlooking something obvious.

Cheers,

Tim Churches
Sydney, Australia





From fredrik at pythonware.com  Fri Mar 23 11:18:40 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 23 Mar 2001 16:18:40 GMT
Subject: File in a directory
References: <3ABB6E13.403B963B@marugg.net>
Message-ID: <AHKu6.3195$4N4.475205@newsc.telia.net>

Oliver Marugg wrote:
> Has anyone an idea how to solve that with the size or timestamp
> verification?

  os.listdir(dir) => list of files in directory dir
  os.path.getsize(file) => size of file
  os.path.getmtime(file) => time of last modification (in seconds since epoch)
  time.time() => current time (in seconds since epoch)
+ five to ten minutes of your time
-----------------------------
= solution

reference info here:
http://www.python.org/doc/current/lib/module-os.html
http://www.python.org/doc/current/lib/module-os.path.html
http://www.python.org/doc/current/lib/module-time.html

Cheers /F




From volucris at hotmail.com  Fri Mar 16 00:43:41 2001
From: volucris at hotmail.com (Voluciris)
Date: Thu, 15 Mar 2001 23:43:41 -0600
Subject: tempfile.tempdir
Message-ID: <3ab1a806$0$48743$6e49188b@news.goldengate.net>

If I change the value of tempfile.tempdir is it necessary to change it back
when I'm finished?  Does the change just happen inside Python, without
affecting any enviroment variables?

thanks
greg




From piet at cs.uu.nl  Wed Mar 21 06:39:30 2001
From: piet at cs.uu.nl (piet at cs.uu.nl)
Date: 21 Mar 2001 12:39:30 +0100
Subject: New PEP: The directive statement
References: <mailman.985117312.5806.python-list@python.org>
	<j4u24oi4e8.fsf@informatik.hu-berlin.de>
	<yHPt6.2553$4N4.334967@newsc.telia.net>
	<998hv2$5ub$1@panix3.panix.com>
	<oOQt6.2576$4N4.342378@newsc.telia.net>
Message-ID: <uhf0njqq5.fsf@cs.uu.nl>

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

FL> Aahz Maruch wrote:
>> >except that "import" already is a declaration statement, of course.
>> 
>> Depends how you define "already".  There is no current released version
>> of Python (alphas and betas do not count as "released versions") in
>> which "import" is a declaration.

FL> import declares things, in all existing Python versions (hint: it's
FL> an assignment statement)

On the other hand, the effect of import is the same as the calling of the
magical __import__ function, of which the definition says:
This function is invoked by the import statement (not declaration).
And it is done at runtime, not compiletime.
-- 
Piet van Oostrum <piet at cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: P.van.Oostrum at hccnet.nl


From rosemann at imkt.uni-hannover.de  Wed Mar  7 05:27:57 2001
From: rosemann at imkt.uni-hannover.de (Harald Rosemann)
Date: Wed, 07 Mar 2001 11:27:57 +0100
Subject: How to installing NumPy?
References: <3AA57C0D.BBC5E956@itu.edu.tr> <7u3dcqqq2z.fsf@gnwy100.wuh.wustl.edu>
Message-ID: <3AA60D2D.4E31AA50@imkt.uni-hannover.de>

Christopher Lee wrote:
> 
> >>>>> "Nuzhet" == Nuzhet Dalfes <dalfes at itu.edu.tr> writes:
> 
>     Nuzhet> Hi,
> 
>     Nuzhet> I am trying to install NumPy on a RedHat 7.0 running python2.0.
>     Nuzhet> Everything is in the standard places (I think!) and I picked up
>     Nuzhet> and unzipped
>     Nuzhet> http://download.sourceforge.net/numpy/NumPy1711_20.zip into
>     Nuzhet> /usr/local/lib/python2.0/site-packages. And set PYTHONPATH to
>     Nuzhet> /usr/local/lib/python2.0/site-packages.
> 
> Numpy uses distutils for easy installation with python 2.0
> 
> 1. become root at the command line
> 2. cd to the directory into which the Numeric package was unzipped.  It
>    should be called something like Numeric-17.1.1
> 3. type:
>           python setup.py install
> 
>    (This installs the core Numeric modules)
> 
> 4. Then type:
>           python setup_all.py install
> 
>    (installs the additinal packages)
> 
> You should be good-to-go.  If you don't see the scripts setup.py and
> setup_all.py you are either in the wrong place or you have a broken
> distribution.  If you get the chance, you might want to download 17.3.3;
> it has some bug fixes.


I tried to install Numeric-17.3.0 by:
        python setup.py install
on SuSE-Linux 7.1. 

Got the error message:
....
gcc -O2 -m486 -fpic -IInclude -I/usr/include/python2.0 -c
Src/_numpymodule.c -o build/temp.linux-i686-2.0/Src/_numpymodu
Src/_numpymodule.c:1: Python.h: No such file or directory
error: command 'gcc' failed with exit status 1
....


In Python-2.0 I found:
        Python-2.0/Include/Python.h
but that doesn't seem to work.

What's wrong here?
Many thanks!

Harald Rosemann


From jmurray at agyinc.com  Fri Mar 16 20:01:53 2001
From: jmurray at agyinc.com (Joe Murray)
Date: Fri, 16 Mar 2001 17:01:53 -0800
Subject: profile module behavior
Message-ID: <3AB2B781.647AD64A@agyinc.com>

Dear All,

Is this normal profile.run() standard report output?

         24057864 function calls in -3365.677 CPU seconds

   Ordered by: standard name

   ncalls  tottime  percall  cumtime  percall filename:lineno(function)
        1    0.010    0.010 -3365.697 -3365.697 <string>:1(?)
        1   26.440   26.440 -3365.707 -3365.707
findphrases.py:23(runScript)
     7843    0.090    0.000    0.090    0.000
findphrases.py:52(<lambda>)
        0    0.000             0.000          profile:0(profiler)
        1    0.020    0.020 -3365.677 -3365.677
profile:0(runScript(arglist[0][1]))
   388090   34.210    0.000 -3398.947   -0.009 sre.py:117(_subn)
.
.
.

What do negative numbers mean?  :)  Seems like my code was fast, but I
didn't know it was done, well, basically yesterday!

Hoping that the answer to my question is NOT obviously documented,

joe

-- 
Joseph Murray
Bioinformatics Specialist, AGY Therapeutics
290 Utah Avenue, South San Francisco, CA 94080
(650) 228-1146



From grante at visi.com  Fri Mar  9 15:07:02 2001
From: grante at visi.com (Grant Edwards)
Date: Fri, 09 Mar 2001 20:07:02 GMT
Subject: is it possible to use Tkinter on linux
References: <98714n$3p8$1@news2-2.kornet.net> <Xj_p6.189678$df5.5158755@news1.crdva1.bc.home.com>
Message-ID: <GJaq6.266$Tg.42959@ruti.visi.com>

In article <Xj_p6.189678$df5.5158755 at news1.crdva1.bc.home.com>, Wayne Izatt wrote:

>> someone knows that it is possible to use Tkinter on linux? if
>> it is, what program I should need to use Tkinter on linux, how
>> to install , which environment variable should I set.

>as i recall (this was with 1.5.2 anyway) tkinter could be
>enabled as part of the build process (by default it wasn't). it
>was quite well documented in the source code, so start there.

Tkinter wasn't enabled by default in the build process, but
don't most Linux's come with Tkinter already installed?  The
RH-derived ones all do.  THe others mostly all have it
available as an optional package somewhere on the CD.

-- 
Grant Edwards                   grante             Yow!  Now that I have my
                                  at               "APPLE," I comprehend COST
                               visi.com            ACCOUNTING!!


From emile at fcfw.fenx.com  Tue Mar 27 10:24:41 2001
From: emile at fcfw.fenx.com (Emile van Sebille)
Date: Tue, 27 Mar 2001 07:24:41 -0800
Subject: who is my caller?
References: <Pine.LNX.4.21.0103271010460.24308-100000@clarkevans.com>
Message-ID: <15d801c0b6d2$0c09a540$0a06a8c0@FENX.COM>

It appears you can get to the callers locals and globals.  Looking at
traceback, you find sys.exc_info(), which leads to [2].tb_frame which gets
you to f_locals and f_globals.


Emile van Sebille
emile at fenx.com

---------
----- Original Message -----
From: "Clark C. Evans" <cce at clarkevans.com>
To: "Emile van Sebille" <emile at fenx.com>
Cc: <python-list at python.org>
Sent: Tuesday, March 27, 2001 7:15 AM
Subject: Re: who is my caller?


> On Tue, 27 Mar 2001, Emile van Sebille wrote:
> > If I'm not mistaken, with 2.0 you can directly import traceback without
> > needing to generate an exception first.
>
> I was actually looking to do something like..
>
>     caller = traceback.call_stack.top()
>     try:
>       if '0h3kfd' = caller.check('key'):
>           # ok... you passed security
>     except:
>       pass
>
> A string print-out is useless... I'm after
> the function objects....
>
> This brings up another question, if I
> have a function can I get the module
> handle; or if I have a bound-function can
> I fetch the object that it is part of?
>
> Thanks!
>
> Clark
>
>
>
>




From NoSpam at NoSpam.com  Fri Mar  2 12:07:12 2001
From: NoSpam at NoSpam.com (Tom)
Date: Fri, 02 Mar 2001 17:07:12 GMT
Subject: Unresolved symbols in MSVCRT.DLL when running a py2exe packaged program.
References: <mailman.983539023.17143.python-list@python.org>
Message-ID: <4rQn6.14308$UZ4.4134344@news4.rdc1.on.home.com>

I disagree.

You can distribute system dll's.  You are expected to, and sometimes you
have to.  You must insure that the version you load is at least equal to the
version that you linked against.

I have done this (distribute system DLL's) on all windows except WinME and
the 16bit ones.  Unfortunately I can't say how I did it precisely, because I
just used InstallShield.  But I believe that, where SFP exists, you must use
specific and new API's to replace DLL's, otherwise the system will back out
your changes.

One solution is to specify the changes in an .INF file and let the system
worry about these details (but this has its limitations).

I assume that Python itself updates the MSVCRT.dll.  Might look at how it
does it on Win2K.

One way I remember getting the error being described was writing code for
the install process that used the CRT and that got executed before my
version of the CRT got installed.

Tom.

"Rolander, Dan" <Dan.Rolander at marriott.com> wrote in message
news:mailman.983539023.17143.python-list at python.org...
> I agree with Robin and Gordon's assessment. It's risky at best to
distribute
> system dll's with your application, and impossible to upgrade system dll's
> on Win2K because of system file protection.
>
> Dan
>
> -----Original Message-----
> From: Thomas Heller [mailto:thomas.heller at ion-tof.com]
> Sent: Friday, March 02, 2001 8:03 AM
> To: python-list at python.org
> Subject: Re: Unresolved symbols in MSVCRT.DLL when running a py2exe
> packaged program.
>
>
> Here is a comment about a similar problem in
> another py2exe-packed application, which
> I received from Jeff Kunce:
>
> >>I just tried runing a py2exe-packaged application on an older NT
machine,
> and
> >>it choked with an "Entry Point Not Found" error on MSVCRT.dll
> >
> >Well, I played with it some more and got it to work.
> >
> >I simply excluded all the DLLs that are normally found in NTs system32
> >directory (MPR.dll, MSVCIRT.dll, MSVCRT.dll, NETAPI32.dll, NETRAP.dll,
> >SAMLIB.dll) - and it works now on the older machine.
> >
> >It looks like there is no easy "good" solution to this problem. Here's a
> >comment by Gordon McMillan on the topic:
> >  http://www.wxpython.org/archive/2000-June/002469.html
>
> Thomas
>
>
> --
> http://mail.python.org/mailman/listinfo/python-list
>




From aleaxit at yahoo.com  Tue Mar 13 10:40:48 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Tue, 13 Mar 2001 16:40:48 +0100
Subject: Converting a text data file from positional to tab delimited.
References: <B6D37840.2124%lee@joramo.com>
Message-ID: <98lf1t02b1q@news1.newsguy.com>

"Lee Joramo" <lee at joramo.com> wrote in message
news:B6D37840.2124%lee at joramo.com...
> I am looking for suggestions to speed up the process of converting a large
> text data file from 'positional' layout to tab delimited. The data file is
> over 200MB in size containing over 40,000 lines which have over 600
fields.
>
> I suspect that the 'for' loop that splits each line into tab delimited,
> could be optimized. Perhaps it could be replaced with a regex or other
> technique.
    [snip]
> layout = [
>     ['STUDY', 0, 7]
>     ['MDLNO', 8, 12]
    [snip]
>         for field in layout:
>             #
>             #can this loop be improved??
>             #
>             fieldValue = line[field[1]:field[2]]

First of all, there seems to be a serious bug here: it looks like the
layout has a pair of numbers, the upper one of which is meant to be
*included*, but the a:b slice notation *excludes* b -- so, you most
likely want line[field[1]:field[2]+1].

>             delimitedLine = delimitedLine + delimit + fieldValue
>             delimit = "\t"
>         outFile.write(delimitedLine+"\n")

A small optimization is to use a _list_ of pieces (strings) and
output them with a single .writelines call (which does not just
write _lines_, but arbitrary strings).

As you know beforehand the number of fields in your layout, you
can prepare the list-of-pieces in advance:

list_of_pieces = ['\t'] * (2*len(layout)-1)
indexed_fields = zip(layout, range(0,len(layout),2))

and fill alternate pieces with other-than-tabs in the nested
loop:

    for field, index in indexed_fields:
        list_of_pieces[index] = line[field[1]:field[2]+1]
    outFile.writelines(list_of_pieces)

We can do a little more work in advance, outside of the loop:

indexed_fields = [ (i*2, layout[i][1], layout[i][2]+1)
    for i in range(len(layout)) ]

and the loop becomes the very-slightly-faster:

    for index, lower, upper in indexed_fields:
        list_of_pieces[index] = line[lower:upper]

I don't know if a substantially faster approach exists to
avoid the hundreds of calls to line[lower:upper] for each
line.  You could break the line into a list of characters
once just before the inner-loop then slice the list, but
I don't know if that would gain you anything (you'd have
to join the sublists at some point, or have a much bigger
list-of-pieces, and either approach seems of doubtful
gain).

Another possibility is avoiding some data-copy by having
the inner-loop's statement as:
        list_of_pieces[index] = buffer(line, offset, size)
you'll need to prepare the indexed_fields a bit differently:
indexed_fields = [ (i*2, layout[i][1], layout[i][2]-layout[i][1]+1)
    for i in range(len(layout)) ]
and the loop itself becomes:
    for index, offset, size in indexed_fields:

Not sure if .writelines supports a list of alternate
strings and buffers, or what the speed becomes then --
just suggesting alternatives for you to try...


Alex





From rambo at linuxsecurity.co.kr  Thu Mar  8 16:19:24 2001
From: rambo at linuxsecurity.co.kr (rambo)
Date: Thu, 8 Mar 2001 13:19:24 -0800
Subject: is it possible to use Tkinter on linux
Message-ID: <98714n$3p8$1@news2-2.kornet.net>

Hello, there.
someone knows that it is possible to use Tkinter on linux?
if it is, what program I should need to use Tkinter on linux,
how to install , which environment variable should I set.

thanks for your time.







From ransen_spam_me_not at nemo.it  Thu Mar  8 10:01:16 2001
From: ransen_spam_me_not at nemo.it (Owen F. Ransen)
Date: Thu, 08 Mar 2001 15:01:16 GMT
Subject: tokenizer.c crashes in 2.1b1
References: <3aa34d5c.1284101@news.newsguy.com> <ulmqiobyp.fsf@roundpoint.com> <3aa5d493.5922346@news.newsguy.com> <984q1m0skr@news1.newsguy.com>
Message-ID: <3aaa8fad.5510341@news.newsguy.com>

On Wed, 7 Mar 2001 09:03:58 +0100, "Alex Martelli" <aleaxit at yahoo.com>
wrote:


>> Debug Multithreaded in my MFC app make settings
>> Debug Multithreaded DLL in the Pythoncore DLL make settings
>> Is this wrong?
>Yes.
>Alex

Many thanks Alex, I can now run Python source files from
within an MFC app.

Get memory leaks though (the adventure continues!)

;)


--
Owen F. Ransen
http://www.ransen.com/
Home of Gliftic & Repligator Image Generators


From nomad*** at ***freemail.absa.co.za  Mon Mar 26 07:57:31 2001
From: nomad*** at ***freemail.absa.co.za (nomad)
Date: Mon, 26 Mar 2001 12:57:31 GMT
Subject: Help how do I read `word' variables from a binary file?
Message-ID: <3abf3b0b.1487830@news.is.co.za>

Howzit,

I'm having trouble with reading a binary file properly in Python 2.0
on win32.  I'm stating explicitly that I'm using Win32 as I know that
you have to state when opening the file that it is for `br' as in
binary reading (I'm think this is due to intels little-endian
approach, but I could be wrong).

I can open the file without problems, and I can read from the file
without a hitch, _BUT_ the trouble comes in that some of the data
records in the file have fields that would be a double word or even a
normal word in C++, so in effect the data may be 0x80002310, but
python reads the data directly from the file, and as such is reading
hex bytes as (in this example) `10 23 00 80' etc.  I'm sure that there
must be a simple way to read in data like this, but I can't figure it
out.

Any help would be greatfully recieved.

TIA.


From ws-news at gmx.at  Wed Mar  7 09:56:24 2001
From: ws-news at gmx.at (Werner Schiendl)
Date: Wed, 7 Mar 2001 15:56:24 +0100
Subject: Python script for deMSWordification?
References: <985h02$1pbm$1@swen.emba.uvm.edu> <983976916.181372@newsmaster-04.atnet.at>
Message-ID: <983977262.909062@newsmaster-04.atnet.at>

A resource that just came to my mind a couple of seconds too late to do it
in my original post is

www.wotsit.org

there seems to be quite a lot of information on that subject

good luck
werner

Werner Schiendl <ws-news at gmx.at> wrote in message
news:983976916.181372 at newsmaster-04.atnet.at...
> If you have it on your machine you can use itself with Automation to do
the
> job.
>
> otherwise
>
> .doc Files are basically OLE Compound Documents that represents an
IStorage.
> IStorage is an interface that models DCOM structured storage alongside
> IStream.
> Comapred to a file system IStorage is a directory, and IStream is a file.
> There are APIs to cope with that kind of file.
> With Visual Studio comes a tool called DocFile Viewer that shows the basic
> content of the file.
>
> But I guess you want to do that on non-MS platform.
> This sounds to me like big fun, so enjoy it!
>
> hth
> werner
>
>
> <kelvin.chu at uvm.edu> wrote in message
> news:985h02$1pbm$1 at swen.emba.uvm.edu...
> > Dear fellow Pythoneers;
> >
> > Numerous colleagues and administration types send memos, colloqium
> > announcements and other things around in Micro$oft word format.  Has
> anyone
> > written a filter in Python that can convert these things into normal
text?
> >
> > If not, is there a basic set of rules for decoding .doc format files
> > anywhere that I can look at?
> >
> > Cheers,
> >
> > -k
> >
> > --
> >
> >   kelvin.chu at uvm.edu      (802) 656-0064
> >   http://www.uvm.edu/~kchu/ FAX: (802) 656-0817
> >
>
>




From h-karls at telia.com  Mon Mar  5 13:23:59 2001
From: h-karls at telia.com (Håkan Karlsson)
Date: Mon, 05 Mar 2001 18:23:59 GMT
Subject: Graphical python development tool for Linux?
References: <3a93e509.1040040@news.earthlink.net>
Message-ID: <3RQo6.21838$AH6.3063511@newsc.telia.net>

> Would anyone be kind enough to recommend a graphical Python
> development tool, perhaps similar to Pythonwin for Windows, only for
> Linux?

You could always have a look at PythonWorks Pro at
www.pythonware.com/products/works

It's available for both Linux and Windows.

/Hakan




From jshaXXXnkland at insXXXpiration.com  Tue Mar  6 16:22:48 2001
From: jshaXXXnkland at insXXXpiration.com (John Shankland)
Date: Tue, 06 Mar 2001 21:22:48 -0000
Subject: FW: Need help with Tkinter bind statement
References: <mailman.983910277.5499.python-list@python.org>
Message-ID: <905C8F8F5jshanklandinspiratio@207.126.101.100>

I'm not sure about the <-> problem but for the other keys you need to
return "break", to stop Tkinter from going up the bindings chain. A good
doc for this would be
http://www.pythonware.com/library/tkinter/introduction/events-and-bindings
.htm 

--
????
John



Mike.mellor at tbe.com (Mellor, Mike) wrote in
<mailman.983910277.5499.python-list at python.org>: 

># Bindings
>xStack.bind('<+>', lambda event, fxn = 'y+x':Calc(fxn))
>xStack.bind('<->', lambda event, fxn = 'y-x':Calc(fxn)) # This is 
>causing an error
>xStack.bind('<*>', lambda event, fxn = 'y*x':Calc(fxn))
>xStack.bind('</>', lambda event, fxn = 'y/x':Calc(fxn))
>
>where xStack is an entry widget and Calc is a function that does the 
>calculation.  
>
-- snip --

>
>but the other bindings work fine.  The second problem is that when I 
>use any of these keys, the keystroke shows up in the entry widget.  
>What I want to happen is that pressing the "+" key, for example, 
>should add the values in the xStack and yStack (which it does) and 
>then put the result in the yStack entry widget (which it also does).  
>The problem is that the "+" is left in the xStack.  The contents of 
>the xStack are deleted by the Calc function, so I thought it would 
>also delete the "+" as well.  
>
>Thanks.
>
>Mike
>
>
>
>
>
>



From dev_null at email.com  Sat Mar  3 20:05:09 2001
From: dev_null at email.com (Dev_NuLL)
Date: Sun, 04 Mar 2001 01:05:09 GMT
Subject: Intersesting Questions involving extending and embedding
References: <mailman.983540226.21488.python-list@python.org>
Message-ID: <3aa194d9.3955493@news.digital.net>

Thanks for the help
>> bunk URL
>
>   http://www.mudlib.org/~rassilon/p2c, but th stuff is old - Aug 1999.
>


From craig at patterncraft.com  Sat Mar 10 12:17:11 2001
From: craig at patterncraft.com (Craig Holman)
Date: Sat, 10 Mar 2001 11:17:11 -0600
Subject: Memory leaks in VC6 2 line Python 2.1b1 apps?
References: <3aa88f52.5419185@news.newsguy.com> <hGPp6.53719$UZ4.13963947@news4.rdc1.on.home.com> <98b0a0$2g8i$1@shadow.skypoint.net> <3aa91ea5.645940@news.newsguy.com>
Message-ID: <98dngs$2cmm$1@shadow.skypoint.net>

I wish that Visual Studio's memory-leak reporter were smart enough to
display where the leaked blocks of memory were allocated.  It only puts out
pairs of lines like the following :

{1037} normal block at 0x0079CE80, 39 bytes long.
 Data: <  y             > 20 EE 79 00 88 85 1F 1E 03 00 00 00 B8 17 20 1E

The other lines in the annotated dump were added by me.  They contain the
results of my efforts to figure out where the leaked blocks were allocated.
After each of the pair of lines produced by the memory-leak reporter, I've
listed my best guess as to what kind of a block was leaked and why, the line
in the Python 2.1b1 source file that I placed a breakpoint at that allocates
the leaked memory block, and a snapshot of the relevant part of the runtime
stack at the point that the breakpoint was hit and execution was paused.  I
haven't identified where all of the leaked blocks are being allocated yet.

Hope this helps,
Craig

"Owen F. Ransen" <ransen_spam_me_not at nemo.it> wrote in message
news:3aa91ea5.645940 at news.newsguy.com...
> On Fri, 9 Mar 2001 10:28:33 -0600, "Craig Holman"
> <craig at patterncraft.com> wrote:
>
> >Greetings,
> >
> >I'm afraid that both versions 2.0 and 2.1b1 leak memory for a single
> >Py_Initialize, Py_Finalize sequence in a
> >Visual C++ MFC application.
>
> Phew! I thought it was just me!  ;)
>
> I hope someone understands the rest of your post, I'm too new
> to Python embedding to help. Also, for some reason, though I
> did rebuild from source, VC gives me the leak "number" but not
> the line in the library which caused it.
>
>
> --
> Owen F. Ransen
> http://www.ransen.com/
> Home of Gliftic & Repligator Image Generators




From william.famy at noos.fr  Sat Mar 24 12:12:31 2001
From: william.famy at noos.fr (William Famy)
Date: Sat, 24 Mar 2001 18:12:31 +0100
Subject: difference between script and prompt ! looking for the Problem
References: <3ABCCA03.90871133@noos.fr> <99iht4$1c7hg$1@ID-11957.news.dfncis.de>
Message-ID: <3ABCD57F.DC395F88@noos.fr>

Thanks.

but if i print in the script i have 3.0 4.5 insted of (3.0, 4.5)



Emile van Sebille wrote:
> 
> You need to print from within a script.  Interactive mode assumes it.
> 
> ie:
> 
> print c.r, c.i
> 
> HTH
> --
> 
> Emile van Sebille
> emile at fenx.com

-- 
 			     \\\|///
                           \\  - -  //
                            (  @ @  )
+-------------------------oOOo-(_)-oOOo-------------------------+
|  Famy William  36 avenue des barattes  74000 Annecy  France   |
|  email:william.famy at noos.fr william.famy at mail-enseignant.com  |
+---------------------------------Oooo--------------------------+
                           oooO   (   )
                          (   )    ) /
                           \ (    (_/
                            \_)


From fredrik at pythonware.com  Tue Mar 20 14:51:45 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Tue, 20 Mar 2001 19:51:45 GMT
Subject: Insertin **keywords into a class
References: <mailman.984971990.29857.python-list@python.org> <9960i1$tp0$1@nntp6.u.washington.edu> <99661m$scc$1@panix6.panix.com> <99877b$uss$1@nntp6.u.washington.edu> <mailman.985114465.27377.python-list@python.org>
Message-ID: <lxOt6.2532$4N4.329778@newsc.telia.net>

C. Porter Bassett wrote:
> Can you please tell me why the following code snippet doesn't update the
> value of radius?

only if you can tell me why you think it should update the
value of radius ;-)

> class myClass:
>    def __init__(self, *arguments, **keywords):
>       self.radius = 0.0

here you set radius to 0.0.

>       for kw in keywords.keys():
>          print kw, ":", keywords[kw]
>          self.kw = keywords[kw]

and here you're attempting to update a non-existent
attribute called "kw".  I assume you get an Attribute-
Error...

>       print "self.radius =", self.radius
>
> b = myClass(radius = 1.0)

maybe you were thinking of the __dict__ dictinary (which
holds instance variables)?  something like this would work:

       for kw in keywords.keys():
          print kw, ":", keywords[kw]
          self.__dict__[kw] = keywords[kw]

or, better:

       for k, v in keywords.items():
          setattr(self, k, v)

or, faster:

            self.__dict__.update(keywords)

or perhaps even:

 class myClass:
    def __init__(self, radius=0.0, *arguments, **keywords):
       self.radius = radius

(but that will of course not work if you really need to pass in
arbitrary attributes...)

Cheers /F




From aleaxit at yahoo.com  Fri Mar 16 08:58:06 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 16 Mar 2001 14:58:06 +0100
Subject: Basic information requested
References: <MHks6.144$lC5.52172@news2-win.server.ntlworld.com>
Message-ID: <98t65e0lri@news1.newsguy.com>

"Jamie MacIsaac" <jamie at cctech.co.uk> wrote in message
news:MHks6.144$lC5.52172 at news2-win.server.ntlworld.com...
> Hi,
>
> I've just started looking at Python and would like to use it to perform
some
> simple admin tasks on a Windows network. However, I'm having difficulty
> finding out how to accomplish elementary tasks.
>
> Can someone please point me to sample code that shows how to map a network
> share and copy some files? Is there a central resource of code samples?

win32net (part of the win32all extensions) does that sort of 'networking'
tasks (and, yep, Hammond's book does cover it inter alia).

If you have win32all installed (and if you don't, visit ActiveState
and download and install it!), there will be a directory such as
D:\Python20\win32\demos which has various Python demo/example
sources -- have a look at win32netdemo.py and you'll see some
example uses of the win32net and ancillary extension modules.


Alex





From aleaxit at yahoo.com  Fri Mar  9 11:04:11 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 9 Mar 2001 17:04:11 +0100
Subject: Can anyone recomend a good intoduction to C...
References: <mailman.983894288.19965.python-list@python.org> <3AA5EF59.507F1D91@olsen.ch> <983958878.976344@newsmaster-04.atnet.at> <989cvj$sba$1@panix3.panix.com>
Message-ID: <98autr027ef@news2.newsguy.com>

"Aahz Maruch" <aahz at panix.com> wrote in message
news:989cvj$sba$1 at panix3.panix.com...
> In article <983958878.976344 at newsmaster-04.atnet.at>,
> Werner Schiendl <ws-news at gmx.at> wrote:
> >
> >It supports for a real large number of different design approaches.
>
> Many of us are using Python precisely because it has a design philosophy
> that goes directly counter to TMTOWTDI.

Funny, since Python is SO great at supporting a large number
of different design approaches -- no fewer than C++ supports.

For example, you can code things up as instance-objects with
methods, but you don't HAVE to do it like that -- just as in
C++, and differently from (e.g.) Eiffel or Java, you can also
code 'real functions' -- don't have to make things into
methods if that isn't the best paradigm in certain cases.

Python goes further -- in most cases, for example, one can
choose between an explicit while-loop, or a call to the
standard function 'map' with some lambda, or the use of
a list comprehension; and it's anything but obvious that
one of these three approaches will be ideal in all cases.
(Similarly, C++ offers explicit loops as well as functions
such as std::copy that encapsulate typical uses thereof).

"One obviously-right way to do it" is a great philosophy
*for programming in the small*, but, even there, it's more
of an ideological stance than a Python reality.  And when
it comes to programming *in the large* -- design issues,
choices of paradigm[s], etc -- it's not even funny to claim
there can be just "one true way".


Alex





From rdsteph at earthlink.net  Sun Mar 25 23:24:14 2001
From: rdsteph at earthlink.net (Ron Stephens)
Date: Mon, 26 Mar 2001 04:24:14 GMT
Subject: Newbie needs help to find bug in his first Perl program
References: <3ABCC757.6C6F17CA@earthlink.net> <3ABE73B2.CAFCA177@student.liu.se>
Message-ID: <3ABD67C6.7B714D7A@earthlink.net>

Thanks, Richard, you solved my problem, despite my fuzzy question ;-))) and
to the rest of you, yes, I will try to be much more careful and precise from
now on when asking for help! Point acknowledged.


Thanks to all who answered.. By the way , here is a copy of my now debugged
and working program



Richard Bosnjakovic wrote:

> Ron Stephens wrote:
>
> > Can anyone see what is wrong with this? I have several lines like this
> > and they all give me error messages.
>
> *What* error messages? I don't think anyone want to *guess* what kind of
> error messages you get.
>
> There's no format-specifier named "%y".
>
> --
> Rikard Bosnjakovic - http://a214.ryd.student.liu.se/cv/ - ICQ: 1158217
>
> Anyone sending unwanted advertising e-mail to my address will be
> charged $250 for network traffic and computing time. By extracting my
> address from this message or its header, you agree to these terms.
-------------- next part --------------
print "Decision Analysis Beta Test 0.1"
print "To Help You Decide Amongst Various Alternatives"
print "A General Program Requiring User Input"

number = input ("type a numeral from 2 to 5 to indicate how many options do you want to decide amongst and then press enter:")


x = 1


	
	
option1 = raw_input ("what is option1:")

x = x + 1
option2 = raw_input ("what is option2:")
x = x + 1
while x:
	if x > number: break
	option3 = raw_input ("what is option3:")
	x = x + 1
	if x > number: break
	option4 = raw_input ("what is option4:")
	x = x + 1
	if x > number: break
	option5 = raw_input ("what is option5:")
	


criteria = input ("How many criteria do you want to use for this decision? Type a numeral from 3 to 6 and then press enter")
y = 1


	
	
criteria1 = raw_input ("what is criterium1:")

y = y + 1
criteria2 = raw_input ("what is criterium2:")
y = y + 1
while y:
	if y > criteria: break
	criteria3 = raw_input ("what is criterium3:")
	y = y + 1
	
	if y > criteria: break
	criteria4 = raw_input ("what is criterium4:")
	y = y + 1
	if y > criteria: break
	criteria5 = raw_input ("what is criterium5:")
	y = y + 1
	if y > criteria: break
	criteria6 = raw_input ("what is criterium6:")

z = 1
print "Of the criteria you have listed, please give each one a rank when prompted, with 1 being the most important criteria, 2 being the second most important, etc. In other words, after being prompted for each ranking criteria, enter a numeral from 1 to x where x is the total number of ranking criteria you are using, and then press enter."
while z:
	rank1 = input ("what is the rank for %s " %criteria1)
	z = z + 1
	rank2 = input ("what is the rank for %s " %criteria2)
	z = z + 1
	if z > criteria: break
	rank3 = input ("what is the rank for %s " %criteria3)
	z = z + 1
	if z > criteria: break
	rank4 = input ("what is the rank for %s " %criteria4)
	z = z + 1
	if z > criteria: break
	rank5 = input ("what is the rank for %s " %criteria5)
	z = z + 1
	if z > criteria: break
	rank6 = input ("what is the rank for %s " %criteria6)

q = 1
while q:
	
	a = 1
	while a:
	
		oneone = input ("where does %s rank for %s ?" % (option1, criteria1))
		a = a + 1
		onetwo = input ("where does %s rank for %s ?" % (option1, criteria2))
		a = a + 1
		if a > criteria: break
		onethree = input ("where does %s rank for %s ?" % (option1, criteria3))
		a = a + 1
		if a > criteria: break
		onefour = input ("where does %s rank for %s ?" % (option1, criteria4))
		a = a + 1
		if a > criteria: break
		onefive = input ("where does %s rank for %s ?" % (option1, criteria5))
		a = a + 1
		if a > criteria: break
						 

		onesix = input ("where does %s rank for %s ?" % (option1, criteria6))
	q = q + 1

	if q > number: break		
	
	b = 1
		
	while b:
	
	
		twoone = input ("where does %s rank for %s ?" % (option2, criteria1))
		b = b + 1
		twotwo = input ("where does %s rank for %s ?" % (option2, criteria2))
		if b > criteria: break
		b = b + 1
		twothree = input ("where does %s rank for %s ?" % (option2, criteria3))
		b = b + 1
		if b > criteria: break
		twofour = input ("where does %s rank for %s ?" % (option2, criteria4))
		b = b + 1
		if b > criteria: break
		twofive = input ("where does %s rank for %s ?" % (option2, criteri5))
		b = b + 1
		if b > criteria: break
		twosix = input ("where does %s rank for %s ?" % (option2, criteria6))
		b = b + 1

	q = q + 1

	if q > number: break		
	
	c = 1
	while c:
		threeone = input ("where does %s rank for %s ?" % (option3, criteria1))
		c = c + 1
		threetwo = input ("where does %s rank for %s ?" % (option3, criteria2))
		c = c + 1
		if c > criteria: break
		threethree = input ("where does %s rank for %s ?" % (option3, criteria3))
		c = c + 1
		if c > criteria: break
		threefour = input ("where does %s rank for %s ?" % (option3, criteria4))
		c = c + 1
		if c > criteria: break
		threefive = input ("where does %s rank for %s ?" % (option3, criteria5))
		c = c + 1
		if c > criteria: break
		threesix = input ("where does %s rank for %s ?" % (option3, criteria6))

	q = q + 1

	if q > number: break		


	d = 1
	while d:
		fourone = input ("where does %s rank for %s ?" % (option4, criteria1))
		d = d + 1
		fourtwo = input ("where does %s rank for %s ?" % (option4, criteria2))
		d = d + 1
		if d > criteria: break
		fourthree = input ("where does %s rank for %s ?" % (option4, criteria3))
		d = d + 1
		if d > criteria: break
		fourfour = input ("where does %s rank for %s ?" % (option4, criteria4))
		d = d + 1
		if d > criteria: break
		fourfive = input ("where does %s rank for %s ?" % (option4, criteria5))
		d = d + 1
		if d > criteria: break
		foursix = input ("where does %s rank for %s ?" % (option4, criteria6))

	q = q + 1

	if q > number: break		

	e = 1
	while e:
		
		fiveone = input ("where does %s rank for %s ?" % (option5, criteria1))
		e = e + 1
		fivetwo = input ("where does %s rank for %s ?" % (option5, criteria2))
		e = e + 1
		if e > criteria: break
		fivethree = input ("where does %s rank for %s ?" % (option5, criteria3))
		e = e + 1
		if e > criteria: break
		fivefour = input ("where does %s rank for %s ?" % (option5, criteria4))
		e = e + 1
		if e > criteria: break
		fivefive = input ("where does %s rank for %s ?" % (option5, criteria5))
		e = e + 1
		if e > criteria: break
		fivesix = input ("where does %s rank for %s ?" % (option5, criteria6))

	q = q + 1

	if q > number: break

	
	f = 1
	while f:
		sixone = input ("where does %s rank for %s ?" % (option6, criteria1))
		f = f + 1
		sixtwo = input ("where does %s rank for %s ?" % (option6, criteria2))
		f = f + 1
		if f > criteria: break
		sixthree = input ("where does %s rank for %s ?" % (option6, criteria3))
		f = f + 1
		if f > criteria: break
		sixfour = input ("where does %s rank for %s ?" % (option6, criteria4))
		f = f + 1
		if f > criteria: break
		sixfive = input ("where does %s rank for %s ?" % (option6, criteria5))
		f = f + 1
		if f > criteria: break
		sixsix = input ("where does %s rank for %s ?" % (option6, criteria6))

if criteria < 4:
    onefour = 0
    onefive = 0
    onesix = 0
    twofour = 0
    twofive = 0
    twosix = 0
    threefour = 0
    threefive = 0
    threesix = 0
    fourfour = 0
    fourfive = 0
    foursix = 0
    fivefour = 0
    fivefive = 0
    fivesix = 0

    rank4 = 0
    rank5 = 0
    rank6 = 0
    
    
elif criteria < 5:
    onefive = 0
    onesix = 0
    twofive = 0
    twosix = 0
    threefive = 0
    threesix = 0
    fourfive = 0
    foursix = 0
    fivefive = 0
    fivesix = 0
    sixfive = 0
    sixsix = 0

    rank5 = 0
    rank6 = 0
    
elif criteria < six:
    onesix = 0
    twosix = 0
    threesix = 0
    foursix = 0
    fivesix = 0
    sixsix = 0

    rank6 = 0    

else:
    print "These are the scores for each option. The highest score is the best option"
    
			
w = 1
while w:				
	valueoption1 = (oneone)*(7-rank1) + (onetwo)*(7-rank2) + (onethree)*(7-rank3) + (onefour)*(7 - rank4) + (onefive)*(7-rank5) +(onesix)*(7-rank6)
	w = w + 1
	if w > number: break
	valueoption2 = (twoone)*(7-rank1) + (twotwo)*(7-rank2) + (twothree)*(7-rank3) + (twofour)*(7 - rank4) + (twofive)*(7-rank5) +(twosix)*(7-rank6)
	w = w + 1
	if w > number: break
	valueoption3 = (threeone)*(7-rank1) + (threetwo)*(7-rank2) + (threethree)*(7-rank3) + (threefour)*(7 - rank4) + (threefive)*(7-rank5) +(threesix)*(7-rank6)
	w = w + 1
	if w > number: break
	valueoption4 = (fourone)*(7-rank1) + (fourtwo)*(7-rank2) + (fourthree)*(7-rank3) + (fourfour)*(7 - rank4) + (fourfive)*(7-rank5) +(foursix)*(7-rank6)
	w = w + 1
	if w > number: break
	valueoption5 = (fiveone)*(7-rank1) + (fivetwo)*(7-rank2) + (fivethree)*(7-rank3) + (fivefour)*(7 - rank4) + (fivefive)*(7-rank5) +(fivesix)*(7-rank6)
	w = w + 1
	if w > number: break
	valueoption6 = (oneone)*(7-rank1) + (sixtwo)*(7-rank2) + (sixthree)*(7-rank3) + (sixfour)*(7 - rank4) + (sixfive)*(7-rank5) +(sixsix)*(7-rank6)
r = 1
while r:
	print "%s is valued at %d" % (option1, valueoption1)
	r = r + 1
	if r > number: break
	print "%s is valued at %d" % (option2, valueoption2)
	r = r + 1
	if r > number: break
	print "%s is valued at %d" % (option3, valueoption3)
	r = r + 1
	if r > number: break
	print "%s is valued at %d" % (option4, valueoption4)
	r = r + 1
	if r > number: break
	print "%s is valued at %d" % (option5, valueoption5)
	r = r + 1
	if r > number: break
	print "%s is valued at %d" % (option6, valueoption6)
	r = r + 1
	if r > number: break


print "the option with the highest score is the best choice!!!"

print "Thank you for using DecisionAnalysis. Comments and suggestions are welcome and can be addressed to Ron Stephens at rdsteph at earthlink.net"





  



	


						   


    	







    
    
    





   


   

From jcea at argo.es  Mon Mar 12 07:46:45 2001
From: jcea at argo.es (Jesus Cea Avion)
Date: Mon, 12 Mar 2001 13:46:45 +0100
Subject: Pyro Mailing list
Message-ID: <3AACC535.EA2F7625@argo.es>

Is there any mailing list (in Spanish or english) about Pyro?.

About Pyro: http://www.xs4all.nl/~irmen/ap/pyro.html

-- 
Jesus Cea Avion                         _/_/      _/_/_/        _/_/_/
jcea at argo.es http://www.argo.es/~jcea/ _/_/    _/_/  _/_/    _/_/  _/_/
                                      _/_/    _/_/          _/_/_/_/_/
PGP Key Available at KeyServ   _/_/  _/_/    _/_/          _/_/  _/_/
"Things are not so easy"      _/_/  _/_/    _/_/  _/_/    _/_/  _/_/
"My name is Dump, Core Dump"   _/_/_/        _/_/_/      _/_/  _/_/
"El amor es poner tu felicidad en la felicidad de otro" - Leibniz


From robin at jessikat.fsnet.co.uk  Fri Mar 23 05:36:58 2001
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Fri, 23 Mar 2001 10:36:58 +0000
Subject: Please rate these Python books
References: <20010322.1445209@mis.configured.host>
 <_Meu6.8736$8O6.1818817@typhoon.austin.rr.com>
 <slrn9bjuha.4qv.grey@teleute.dmiyu.org>
 <blzu6.8974$G72.207211@e420r-atl1.usenetserver.com>
Message-ID: <PmBCOIAKdyu6Ewje@jessikat.demon.co.uk>

I like Beazley's  book. I like the online quick reference (1999/05/17,
upgraded by Richard Gruet, rgruet at ina.fr from V1.3 ref: 1995/10/30, by
Chris Hoffmann, choffman at vicorp.com)  as well. If this were upgraded to
2.0/2.1 it would be a real benefit.
-- 
Robin Becker


From gregjor at yahoo.com  Fri Mar 16 17:05:54 2001
From: gregjor at yahoo.com (Greg Jorgensen)
Date: Fri, 16 Mar 2001 14:05:54 -0800 (PST)
Subject: Cookies
In-Reply-To: <004d01c0ae3d$1cbfbde0$04090a0a@upcast.com>
Message-ID: <20010316220554.18816.qmail@web9307.mail.yahoo.com>

What about copy/paste? Or tell the user to save the file on their disk, then
let your Python script dig the ID out of it later. The cookie thing could be
made to work but you are right about the OS and browser incompatibilities.

Good luck,

Greg Jorgensen
gregj at pobox.com


--- Ken Seehof <kens at sightreader.com> wrote:
> Yeah, that's the closest I've come up with too, but the difficulty is making
> the system reliably support all browsers.  Unfortunately from the point of
> view of a standalone application, there is no standard for cookies.  Cookies
> are standardized from the point of view of servers and javascript, but that
> doesn't help me with a standalone app.
> 
> The best approach so far seems to be to have the server send back a page
> with a 10 digit unique identifier in a large font.  The python app would
> then ask the user to type in the number manually.  Seems kinda dumb that I
> need the users fingers as part of my protocol. :-)
> 
> - Ken
> 
> ----- Original Message -----
> From: "Greg Jorgensen" <gregj at pobox.com>
> Newsgroups: comp.lang.python
> To: <python-list at python.org>
> Sent: Thursday, March 15, 2001 11:35 PM
> Subject: Re: Cookies
> 
> 
> > A cookie is just a text file stored in a special place where the browser
> can
> > find it. Where exactly the cookies go depends on your OS and browser.
> >
> > Using a text file to store the key would seem the easiest solution, but
> > client-side Javascript can't create or write a file.
> >
> > I think you should generate unique IDs on the server, not on the client.
> > Here's one way to handle it:
> >
> > 1. Client browser sends form data to server.
> > 2. Server processes form, generates a unique ID, stores the form data
> keyed
> > by the unique ID.
> > 3. Server replies with a "thank you" or "database updated" page that also
> > writes a cookie.
> > 4. Python program starts up and looks for the cookie. Depending on your OS
> > and browser the location and name of the cookie will vary, but with a
> little
> > experimenting you can easily find the most likely locations.
> > 5. Python program reads the cookie. Open your cookie with a text editor to
> > see what's in there--it's just name: value pairs.
> > 6. Python program sends HTTP request to server with unique ID it got from
> > the cookie.
> >
> > Obviously this won't work if the client has disabled cookies. With a
> little
> > more trouble you can detect this and perhaps put the unique ID on the
> > response page and tell the user to save it or something. Or embed the
> unique
> > ID in the reply page and tell the user to save that page and give it a
> > specific name, then have your Python program open that text file and find
> > the unique ID in there.
> >
> > Depending on your requirements and security concerns, you can either have
> > your database program generate the unique ID (my preference), use the
> > Windows GUID functions (which are available to ASP on the server), or
> > implement one of the GUID functions you can easily find on the web. Many
> > programmers have solved this in one way or another; send email to me if
> you
> > want specific advice on generating unique IDs from your database engine.
> >
> > Good luck.
> >
> > <<<
> > "Ken Seehof" <kens at sightreader.com> wrote in message
> > news:mailman.984707839.22839.python-list at python.org...
> > Short version:
> >
> > Is it possible for a stand-alone python application to read a cookie
> without
> > the help of a browser?"
> >
> > Long version:
> >
> > I have a web page W and a client side python script P.  I want W to
> generate
> > a GUID (perhaps using javascript) which is then sent to a server S, where
> it
> > is to be used as a key in a database.  Now I run P (on the client, and not
> > in a browser) and I want P to somehow find the value of the GUID, so that
> it
> > can query the database on S.
> >
> > Alternate question:
> >
> > Is there a technique common to both javascript and python that will obtain
> a
> > unique identifier for a computer (e.g. IP address).  It must not be
> browser
> > specific.
> >
> > Motivation:
> >
> > I need to write a record of data from an html form, keyed by some kind of
> > unique client id, to the server and then be able to query that record from
> a
> > stand alone python script running on the client.  The missing link is
> > getting the html form and the python script to agree on a value for the
> id;
> > the rest is easy.
> > >>>
> >
> > --
> > Greg Jorgensen
> > PDXperts
> > Portland, Oregon, USA
> > gregj at pobox.com
> >
> >
> > --
> > http://mail.python.org/mailman/listinfo/python-list
> >
> 
> 


__________________________________________________
Do You Yahoo!?
Get email at your own domain with Yahoo! Mail. 
http://personal.mail.yahoo.com/



From aleaxit at yahoo.com  Fri Mar 23 07:30:06 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 23 Mar 2001 13:30:06 +0100
Subject: Tkinter - Widget Colours
References: <mailman.985346222.22848.python-list@python.org>
Message-ID: <99ffkj01sao@news2.newsguy.com>

"O'Rourke Clodagh-corour01" <corour01 at motorola.com> wrote in message
news:mailman.985346222.22848.python-list at python.org...
> Sorry all - spellcheckers!!
> should have been bg(background)/fg(foreground)

I liked them better as bag/fag -- much more pronounceable, as
well as more in line with timbot's stated policy of "Laugh at
Your Tools".


Alex





From gerhard.nospam at bigfoot.de  Tue Mar  6 10:52:51 2001
From: gerhard.nospam at bigfoot.de (Gerhard =?iso-8859-1?Q?H=E4ring?=)
Date: Tue, 06 Mar 2001 16:52:51 +0100
Subject: Can Python replace Visual Basic? Should it?
References: <5DZo6.11126$hn5.1578410@news1.rdc1.mb.home.com>
Message-ID: <3AA507D3.E384EF2@bigfoot.de>

Brad Bollenbach wrote:
> 
> Okay, sure this has been asked again and again, but I've never really seen a
> clear answer to this question:
> 
>     Can Python replace Visual Basic? Should it?

That's a difficult question. I took part in a VB6 project for about one
and a half year, and I *really* learnt to hate the language, the
environment and the habits VB programmers have. The next generation
Visual Basic, VB.NET will be quite another beast. But that's no excuse
for Microsoft for providing us with that sh*t for ten years.

IMO, VB is good for exactly one thing: integrating COM components. For
everything else, it's a mess.

I use Python for almost everything today (but I don't do much GUI work).
I only use VB as a test container for COM components. That doesn't mean,
however, that Python is a Visual Basic replacement for other people.
Face it, most VB users aren't exactly star programmers. If I were
sarcastic, I'd say they deserve their Visual Basic :->

Have you seen Boa Constructor? [1] I have played with the CVS version a
few weeks ago. A stable release of Boa wouldn't have to fear a
comparison to Visual Basic.

[1] http://boa-constructor.sourceforge.net/

> As far as I can see, they both seem close to equal in their not-so-simple
> ways of distributing apps to end-users, which VB programmers have seem to
> overcome to some degree and so Python programmers should and probably do
> likewise. What I want is a programming language that can make Windows fun
> (like C or Perl does Unix), without having to shell out the big bucks for
> full installs of VB or VC++. 

I wouldn't call C or Perl exactly fun.

> Though I don't believe in using price as an
> important criteria in deciding what language I'll like (unless of course,
> it's Powerbuilder, which is over $4000 :), I still have to admit that I like
> what Python is about, and the benefits of an Open Source language are
> obvious. As I work with Powerbuilder in my day job, I've also realized that
> "languages" that are based on an IDE are not really my thing (I hate having
> a 40,000 foot view of what I'm doing).
> 
> But really, very specifically, were YOU a former Visual Basic (or
> Powerbuilder, or Delphi) programmer that is now using Python to write the
> complex GUI's VB used to handle for you? I'm an on again, off again, student
> of Python, who really wants to give this a go, if there's a good chance that
> I'll get the return on my investment: a programming tool that can take over
> where VB, Powerbuilder, and Delphi used to play.
> 
> Again, I'm much more interested in the real world experience of those who
> used to use the big visual development tools in Windows like VB, Delphi or
> Powerbuilder rather than theoretical answers.

But your question was about spefific, real-world experiences. My answer
would be: no, not just now, but in a 6 - 12 months, maybe.

Gerhard
--
Sorry for the fake email, please use the real one below to reply.
contact: g e r h a r d @ b i g f o o t . d e
web:     http://highqualdev.com


From chris.gonnerman at usa.net  Fri Mar  2 08:24:24 2001
From: chris.gonnerman at usa.net (Chris Gonnerman)
Date: Fri, 2 Mar 2001 07:24:24 -0600
Subject: CGI & Python (error)
References: <3A9F5FF5.34E01148@usa.net>
Message-ID: <002101c0a31c$1e19e840$a100000a@local>

Please post your source(s) so we can inspect them.

----- Original Message ----- 
From: "Mix" <mix77 at usa.net>
Subject: CGI & Python (error)
> When accessing  the  script I get the error:
> 
> File "getdate2.py", line 12, in ?
>     day=form["day"].value
>   File "/usr/lib/python1.5/cgi.py", line 907, in __getitem__
>     raise KeyError, key
> KeyError: day
> 
> The script works from the command line.
> 
> My problems is:  the script I am trying to run is based on another one.
> This
> other script works. I can't  figure out the what is causing the problem
> as
> the 2 scripts are identical  in every aspect execpt for 1 variable. Both
> scripts
> are run in a page which has all the required parameters passed to it.
> Help. Please...
> 
> I hope I'm understandable!





From martin.franklin at westerngeco.com  Thu Mar  8 02:51:35 2001
From: martin.franklin at westerngeco.com (Martin Franklin)
Date: Thu, 08 Mar 2001 07:51:35 +0000
Subject: News from Python9!!!
References: <NUCp6.65206$Z8.11697516@typhoon.southeast.rr.com> <00ac01c0a785$efa17a00$a100000a@local>
Message-ID: <3AA73A07.C3448DDA@westerngeco.com>

If your mail client is anything like mine it wraped the URL so that you 
lost the .html at the end...




Chris Gonnerman wrote:
> 
> I've just got one problem...  When I go to the links given, I get this:
> 
> <html><head><title>Site Not Found</title></head>
> <body>No web site is configured at this address.</body></html>
> 
> Huh.  What happened?
> 
> ----- Original Message -----
> From: "Don Tuttle" <tuttledon at hotmail.com>
> Newsgroups: comp.lang.python
> To: <python-list at python.org>
> Sent: Wednesday, March 07, 2001 9:21 PM
> Subject: News from Python9!!!
> 
> > For those of you, like myself, who are NOT at the Python bash in Long
> Beach,
> > some very good news from there  has been posted at ActiveState.
> >
> > 1) The "Python Software Foundation" has been created
> >
> http://www.activestate.com/Corporate/Communications/Releases/Press983863928.
> > html
> >
> > 2)The Programmers' Choice for Python award winner is Christian Tismer
> >
> http://www.activestate.com/Corporate/Communications/Releases/Press983917761.
> > html
> >
> > 3) A "Python Cookbook" project has started.
> >
> http://www.activestate.com/Corporate/Communications/Releases/Press983583358.
> > html
> >
> > 4)And a bugfix release of Python 2.0 has been released. It includes PPM
> > (Programmer's Package Manager ) to make finding and installing Python
> > modules simple.
> >
> http://www.activestate.com/Corporate/Communications/Releases/Press983916629.
> > html
> >
> > Don
> 
> --
> http://mail.python.org/mailman/listinfo/python-list



From baisly at gis.net  Tue Mar 20 10:19:12 2001
From: baisly at gis.net (peter_douglass)
Date: Tue, 20 Mar 2001 10:19:12 -0500
Subject: Collection interfaces
References: <3a90fa97.0@news1.mweb.co.za>   <MPG.1503548620e632919896a4@news.earthlink.net> <MCjm6.474652$U46.14163288@news1.sttls1.wa.home.com> <MPG.150394195f7f04a19896a7@news.earthlink.net> <3A9A86B1.9A8C69A2@ix.netcom.com> <MPG.1504d14563dce6499896af@news.earthlink.net> <97fjla128ir@news2.newsguy.com> <MPG.1505a76831c34b749896b3@news.earthlink.net> <97h5s40vhc@news2.newsguy.com> <MPG.1506007591cf818b9896bc@news.earthlink.net> <slrn99opop.gr.amk@mira.erols.com> <MPG.1506265ebf6d934d9896be@news.earthlink.net> <97ibmc0k47@news1.newsguy.com> <MPG.1506cf3293b094b49896c1@news.earthlink.net> <97m0c7080c@news2.newsguy.com> <MPG.15085517debf4b649896ca@news.earthlink.net> <zb%n6.31033$o%6.586026@news1.rdc1.bc.home.com> <MPG.150b383cea226d409896d0@news.earthlink.net> <kNko6.6986$eq3.707121@news.uswest.net> <MPG.150c46db9cec283e9896d4@news.earthlink.net> <L9Go6.2720$hZ4.875436@news.uswest.net> <DWCt6.12229$Im6.1309770@newsread1.prod.itd.earthlink.net> <jWEt6.7867$3_.998377@news.uswest.net>
Message-ID: <997sf5$d19$1@taliesin.netcom.net.uk>

Randy A. Ynchausti wrote:

> Now -- where is that procedural code -- I have the OO code in hand?

If you want to put OO and procedural to a test, lets do it right.

1)  Your specifications are a bit "vague".  They need to be made more
precise.
2)  How will we measure the results?  Let's have some metrics.

If you meet those two requirements, I would be happy to write the
<<procedural>> code.

--PeterD




From alex at shindich.com  Sun Mar  4 21:21:29 2001
From: alex at shindich.com (Alex Shindich)
Date: Sun, 4 Mar 2001 18:21:29 -0800
Subject: Name of the Script
References: <20010304.185354.2114937732.8277@localhost.localdomain>
Message-ID: <NJCo6.6720$qv3.2834377@nnrp5-w.sbc.net>

try the following:

import sys
print sys.argv[0]
....
The first argument in the argument list is the name of the script that got
invoked.

I hope this helps.

Regards,

Alex Shindich
visit http://www.shindich.com/sources
"stadel" <stadel at excite.com> wrote in message
news:20010304.185354.2114937732.8277 at localhost.localdomain...
> What is the method for returning the file name of the running script?
>
> Thanks in advance.
> Stadel
>




From jra at dorothy.msas.net  Tue Mar 27 14:08:47 2001
From: jra at dorothy.msas.net (Jay R. Ashworth)
Date: Tue, 27 Mar 2001 19:08:47 GMT
Subject: Linux GPS program
References: <3AC01637.4EF3A6F2@crynwr.com> <slrn9c17k4.cfc.jra@dorothy.msas.net> <3AC0D915.83C2C0C3@crynwr.com>
Message-ID: <slrn9c1p9t.d3g.jra@dorothy.msas.net>

This one time, in band camp,
  Russell Nelson <nelson at crynwr.com> wrote:
> "Jay R. Ashworth" wrote:
> > > Right now, it does moving maps (UTM bitmaps), track recording, and
> > > playback. Will select the "best" of multiple maps.
> > 
> > No *schidt*.  Cool, Russ; thanks.  Any plans to do vector map data as
> > well?  (Yeah, I know, I know; "it's never enough". :-)
> 
> I'm pretty ignorant here.  What sources of vector information are there,
> how big is the dataset, and how do I render it so I can use lat/lon or
> easting/northing to get my location?

There are *stunning* amounts of vector data, all in different formats. 

I'm trying to (get time cleared up to) sort some of that all out even
as I type.

> > Did you use wxPython?
> 
> No.  I used pygtk, which is a set of Python bindings to the gtk
> toolkit.  I also used libglade to load up an XML file of widget
> definitions, and the python bindings to libglade.  The cool thing is
> that somebody could write their own glade file, and use it with pygps on
> a machine with a big screen.

Indeed.

> I ended up with a little Python gps program, a big NMEA parsing module,
> and a Lat/Lon to GPS converter module.

Hmmm.  Cool.
-- j

-- 
Jay R. Ashworth                                                jra at baylink.com
Member of the Technical Staff     Baylink
The Suncoast Freenet         The Things I Think
Tampa Bay, Florida        http://baylink.pitas.com             +1 727 804 5015


From aleaxit at yahoo.com  Mon Mar 12 06:52:56 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Mon, 12 Mar 2001 12:52:56 +0100
Subject: Extending Python: rewriting a single method in C
References: <g0wv9vxms0.fsf@scumbag.ecs.soton.ac.uk>
Message-ID: <98idtn013v3@news2.newsguy.com>

"Jacek Generowicz" <jmg at ecs.soton.ac.uk> wrote in message
news:g0wv9vxms0.fsf at scumbag.ecs.soton.ac.uk...
> I have an object-oriented numerical code written in Python. It has
> come to the stage where the speed of the code is seriously impeding
> progress. Profiling reveals (surprise, surprise) that most of the time
> is spent in only 2 methods.
>
> Is there any way of re-writing just these methods in C ?

You can write a function (in C) that takes the needed state
information as arguments, and your Python-level method will
delegate everything to that function.

Doing it with a "real" method would need C++ (and, e.g.,
Boost Python).


Alex





From steven at haryan.to  Tue Mar 20 01:15:17 2001
From: steven at haryan.to (Steven Haryanto)
Date: Tue, 20 Mar 2001 13:15:17 +0700
Subject: slightly OOT: packaging Python 2.1b1 in RPM
Message-ID: <4.3.2.7.2.20010320130831.00bb0dc0@202.53.255.51>

I am packaging Python 2.1b1 on RedHat 6.2, rpm 3.0.5, using
spec file modified from Jeremy's beopen-python.spec for
BeOpen-Python 2.0.

In the very latest step of %install, I generate an automatic
file list using find:

find $RPM_BUILD_ROOT/ -type f | grep -v 'file-list\.txt' | \
     sed -e "s|^$RPM_BUILD_ROOT||g" > file-list.txt
find $RPM_BUILD_ROOT/usr/lib/python2.1/ \
     $RPM_BUILD_ROOT/usr/include/python2.1/ -type d | \
     sed -e "s|^$RPM_BUILD_ROOT|%dir |g" >> file-list.txt

and feed it to %files:

%files -f file-list.txt

so I can be pretty sure that the package will not be missing
any file.

However, when rpm is scanning for files, /usr/man/man1/python.1
is not found because it has been compressed and becomes
/usr/man/man1/python.1.gz. This must happen after the %install,
since 'find' still lists the uncompressed one.

Any hint why/what program did this?

Thanks,
Steve




From topmind at technologist.com  Wed Mar 21 05:15:23 2001
From: topmind at technologist.com (topmind)
Date: Wed, 21 Mar 2001 10:15:23 GMT
Subject: Collection interfaces
References: <3a90fa97.0@news1.mweb.co.za>   <MPG.1503548620e632919896a4@news.earthlink.net> <MCjm6.474652$U46.14163288@news1.sttls1.wa.home.com> <MPG.150394195f7f04a19896a7@news.earthlink.net> <3A9A86B1.9A8C69A2@ix.netcom.com> <MPG.1504d14563dce6499896af@news.earthlink.net> <97fjla128ir@news2.newsguy.com> <MPG.1505a76831c34b749896b3@news.earthlink.net> <97h5s40vhc@news2.newsguy.com> <MPG.1506007591cf818b9896bc@news.earthlink.net> <slrn99opop.gr.amk@mira.erols.com> <MPG.1506265ebf6d934d9896be@news.earthlink.net> <97ibmc0k47@news1.newsguy.com> <MPG.1506cf3293b094b49896c1@news.earthlink.net> <97m0c7080c@news2.newsguy.com> <MPG.15085517debf4b649896ca@news.earthlink.net> <zb%n6.31033$o%6.586026@news1.rdc1.bc.home.com> <MPG.150b383cea226d409896d0@news.earthlink.net> <kNko6.6986$eq3.707121@news.uswest.net> <MPG.150c46db9cec283e9896d4@news.earthlink.net> <3AA2A898.450B659E@mail.com> <gRCt6.12224$Im6.1308560@newsread1.prod.itd.earthlink.net> <3AB755F8.67E84126@mail.com>
Message-ID: <%a%t6.17245$227.1601569@newsread2.prod.itd.earthlink.net>

"James A. Robertson" <jarober at mail.com> wrote in message
news:3AB755F8.67E84126 at mail.com...
> topmind wrote:
> >
> > "James A. Robertson" <jarober at mail.com> wrote in message
> > news:3AA2A898.450B659E at mail.com...
> > > Topmind wrote:
> > > >
> > >
> > > <snip>
> > >
> > > >
> > > > Not exactly "typical custom business applications", the domain
> > > > in which I focus most of my whining.
> > > >
> > > > The percentage of OO fans in the business domain seems lower
> > > > than average. Perhaps this says something about OO and
> > > > the business domain.
> > > >
> > >
> > > 90% + of the customers using VisualWorks are writing 'typical'
business
> > > apps.
> > >
> >
> > Then why are most "see what OO can do" examples strangely non-business?
>
> They are trying for a 'simplicity' I disagree with.  In any case, the
> tutorials on our website use a straightforward example - parsing a log
> left by website tools.
>
> http://www.cincom.com/smalltalk
>

I only see a front page, not a parsing example.

Besides, what are you comparing it to so say
it is better than a procedural/relational approach?


> >
> > > [snip]
> >
> > -tmind-
> > oop.ismad.com
>
> --
> James A. Robertson
> Product Manager (Smalltalk), Cincom
> jarober at mail.com
> <Talk Small and Carry a Big Class Library>
>




From moshez at zadka.site.co.il  Tue Mar 20 12:37:37 2001
From: moshez at zadka.site.co.il (Moshe Zadka)
Date: Tue, 20 Mar 2001 19:37:37 +0200
Subject: Should __future__ statements ever be phased out?
In-Reply-To: <20010320173101.21650.qmail@web217.mail.yahoo.com>
References: <20010320173101.21650.qmail@web217.mail.yahoo.com>
Message-ID: <E14fQ4f-0000Td-00@darjeeling>

On Tue, 20 Mar 2001, <hamish_lawson at yahoo.co.uk> wrote:

> The __future__ proposal envisages that future statements in a piece of
> code will be phased out once the language change becomes standard.
> (Many of the alternative proposals also envisage this.) Does this not
> pose possible problems with running post-change-aware code on an
> pre-change version of the interpreter?

Yes. That is what is meant when saying "it breaks backwards compatability".
Just like your hypothetical friend can't compile Python, because his
C compile only supports K&R C, not ANSI C.

-- 
"I'll be ex-DPL soon anyway so I'm        |LUKE: Is Perl better than Python?
looking for someplace else to grab power."|YODA: No...no... no. Quicker,
   -- Wichert Akkerman (on debian-private)|      easier, more seductive.
For public key, finger moshez at debian.org  |http://www.{python,debian,gnu}.org



From bkline at rksystems.com  Thu Mar 29 11:18:13 2001
From: bkline at rksystems.com (Bob Kline)
Date: Thu, 29 Mar 2001 11:18:13 -0500 (EST)
Subject: Converting an integer base 10 to a binary number
In-Reply-To: <d7j6ct0r6mu73c42jkfvucq87kl07khbe1@4ax.com>
Message-ID: <Pine.LNX.4.10.10103291107080.31480-100000@rksystems.com>

On Thu, 29 Mar 2001, Daniel Klein wrote:

> 
> def toBinary(dec):
> 	bin = ''
> 	for x in range(8):
> 		bin = str(dec % 2) + bin
> 		dec = int(dec/2)
> 	return bin
> 
> >>> toBinary(32)
> '00100000'
> 
> Of course this only works for 0 <= dec <= 255

No need for that limitation:

$ cat toBinary.py
def toBinary(dec):
    bin = ''
    while dec:
        bin = (dec % 2 and '1' or '0') + bin
        dec = long(dec/2)
    return bin or '0'

if __name__ == "__main__":
    import sys
    dec = len(sys.argv) > 1 and long(sys.argv[1]) or 2 ** 10
    print toBinary(dec)
$ toBinary.py
10000000000
$ toBinary.py 10000000000000000000000
10000111100001100111100000110010011011101010110010010000000000000000000000

You might - for the sake of efficiency - want to have an int-specific
version for cases in which you knew the value would fall within range.

-- 
Bob Kline
mailto:bkline at rksystems.com
http://www.rksystems.com




From jafo-pythonlist at tummy.com  Wed Mar  7 00:48:35 2001
From: jafo-pythonlist at tummy.com (Sean Reifschneider)
Date: Tue, 6 Mar 2001 22:48:35 -0700
Subject: ANNOUNCE: 2.1b1 SRPM
Message-ID: <20010306224835.A5636@tummy.com>

I've finally gotten the 2.1b1 SRPM built.  Sorry for the delay.

The SRPM is available at:

   ftp.tummy.com:/pub/tummy/RPMS/SRPMS

You will also need the "expat-1.1-3tummy.src.rpm" SRPM (available in the same
location) to build the binary above (it prevents a conflict with the
w2c-libwww-devel RPM).

By default, it builds a binary named "python2.1" -- this can be changed by
commenting out the "2.1" line at the very top of the spec file.  To build
a binary RPM from the SRPM above, do (for example):

   rpm --rebuild expat-1.1-3tummy.src.rpm
   rpm -i /usr/src/redhat/RPMS/i386/expat-1.1-3tummy.i386.rpm
   rpm --rebuild python-2.1b1-1tummy.src.rpm
   rpm -i /usr/src/redhat/RPMS/i386/python-2.1a1-tummy.i386.rpm
   rpm -i /usr/src/redhat/RPMS/i386/python-tkinter-2.1a1-tummy.i386.rpm
   rpm -i /usr/src/redhat/RPMS/i386/python-devel-2.1a1-tummy.i386.rpm

These commands will build binary RPMs tailored exactly to your system
(hopefully avoiding problems with 6.2 and 7.0 differences).

Note that these SRPMs are in RPM v4 format.  If you are running RedHat 6.2
or older without the errata applied (shame on you), you'll need to upgrade
to install version 3.0.5 of RPM, available at:

   ftp.valinux.com:/pub/mirrors/redhat/redhat/updates/6.2/i386

as:

   rpm-3.0.5-9.6x.i386.rpm
   rpm-build-3.0.5-9.6x.i386.rpm
   rpm-devel-3.0.5-9.6x.i386.rpm
   rpm-python-3.0.5-9.6x.i386.rpm

If you have any problems with the SRPM, let me know.  Hopefully it will
allow you to more easily test Python 2.1a1 and report problems with it
to the development team.

Sean
-- 
 Your e-mail has been returned due to insufficient voltage.
Sean Reifschneider, Inimitably Superfluous <jafo at tummy.com>
tummy.com - Linux Consulting since 1995. Qmail, KRUD, Firewalls, Python



From Jaeho.Lee at brooks.com  Mon Mar 12 15:27:15 2001
From: Jaeho.Lee at brooks.com (Lee, Jaeho)
Date: Mon, 12 Mar 2001 15:27:15 -0500
Subject: binary comparison of file object
Message-ID: <857F15D7E3D8D3118D290008C7CF058602C97525@mail-naeast1.brooks.com>

Hello,

I am trying to compare the contents of binary file. Checking if the files
are same or not is enough.
Does anybody has idea? Comparing file object in python always return FALSE.

Thanks,
/Jaeho



From root at [127.0.0.1]  Sat Mar 31 14:52:46 2001
From: root at [127.0.0.1] (deadmeat)
Date: Sat, 31 Mar 2001 19:52:46 GMT
Subject: list vs tuple
References: <MVhx6.1021$p5.4157@news1.rivrw1.nsw.optushome.com.au>, <hGlw6.782$p5.3210@news1.rivrw1.nsw.optushome.com.au> <3AC20CD3.265A2D94@alcyone.com> <ru69cto61jtmrqsf809e38hb3pbm2lhi7g@4ax.com> <bC1x6.4319$4N4.905946@newsc.telia.net> <71q9ct4le63anefmiajj5cbh3mg04qu1ga@4ax.com> <mailman.985991202.5578.python-list@python.org> <7Bfx6.989$p5.3791@news1.rivrw1.nsw.optushome.com.au> <Ukhx6.4421$4N4.950603@newsc.telia.net> <mailman.986035544.5572.python-list@python.org>
Message-ID: <iAqx6.1051$p5.4277@news1.rivrw1.nsw.optushome.com.au>

> The same statement behaves *exactly the same*.

No, it doesn't.

> It's the next statements
> (b[5] = 9, b = 9) that behave entirely differently.

Are you on crack?  a = b in the first example copies the data across. a = b
in the second points both a and b to the same space.  This is basic pointer
stuff in other languages - but since Python doesn't do that, one would
expect it to behave like it wasn't using pointers.

> b = 9 binds the variable
> b[5] = 9 is equivalent to b.__setitem__(5, 9)

Your point is what?





From greg at cosc.canterbury.ac.nz  Tue Mar  6 20:16:00 2001
From: greg at cosc.canterbury.ac.nz (Greg Ewing)
Date: Wed, 07 Mar 2001 14:16:00 +1300
Subject: I come to praise .join, not to bury it...
References: <mailman.983485923.1136.python-list@python.org> <97p7iq$kk2$1@nntp6.u.washington.edu> <97rka002k9@news1.newsguy.com> <3AA2F4B3.A7496E76@cosc.canterbury.ac.nz> <97vmoh061p@news2.newsguy.com> <3AA46134.D385208B@cosc.canterbury.ac.nz> <98292h0s0g@news1.newsguy.com>
Message-ID: <3AA58BD0.3324C709@cosc.canterbury.ac.nz>

Alex Martelli wrote:
> 
> No!  string_join (in stringobject.c) currently just _checks_
> PyString_Check on the items -- if any item fails that, it
> delegates to PyUnicode_Join (if PyUnicode_Check OK's that).

And that's somehow not a typeswitch? If it's a regular
string do this... if it's a Unicode string do that...
smells like a typeswitch to me...

You seemed to be arguing that dispatching on the joiner
somehow resulted in an efficiency gain by going straight
to the right code for the kind of strings being joined,
and I was pointing out that that's not the case. But
maybe I misunderstood what you were saying.

> You gain a clean approach, which the implementation could
> (and should) exploit to access the items in the sequence
> through the _appropriate_ standard interface

It's only clean according to one measure of cleanness.
It could be considered cleaner to treat all the arguments
uniformly, and access them *all* through a standard
interface.

> There are NO technical costs associated with these
> advantages.

I disagree. I've mentioned one already -- it bloats the
string object, and makes the Python core larger and
less modular. 

> Only technical advantages and costs
> can be fairly put on the balance

I disagree with that, too. I think it's quite possible
and appropriate for a small technical advantage to be
outweighed by a larger aesthetic one. For example, there
are technical advantages to braces instead of indentation:
the scanner is easier to implement, it's easier to write
programs which generate code, and there's no chance of
space-tab confusion. On the indentation side, there's the
advantage that indentation and block structure always
agree. If you're counting technical advantages, that's
one in favour of indentation and three against. But I
doubt Guido would switch on that basis.

> Note that I've seen NO arguments AT ALL for having
> .join be a method on SEQUENCE object ... despite this 
> being often mentioned on an irrational-purely-aesthetical 
> basis.

I don't think the basis is irrational at all. It
seems to me that there are two quite distinct possible
reasons for deciding to make something a method of a
particular object.

One is to get polymorphism on that object. The other
is because the object makes sense as the direct object
(in the English grammar sense) of the operation, so
that you can naturally read x.verb(y) as "verb x
with y" or "verb x using y".

When these two considerations pull in the same
direction, all is well. And with most of the string
methods they do, but in the case of join, they don't.
There are good reasons for not making it a method of
the sequence, but when it is made a method of the
joiner, the resulting expression sounds backwards[1].

So we are faced with two alternatives, neither
of which are entirely good. In such cases, my
preference is not to choose either alternative --
i.e. make join a function, not a method.

Footnotes:
[1] In English, anyway. Maybe it doesn't sound backwards
    in Dutch. Hmmm... is THAT why Guido accepted it?

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


From lsommerer at sewardweb.com  Fri Mar  2 11:02:43 2001
From: lsommerer at sewardweb.com (Lloyd Sommerer)
Date: Fri, 02 Mar 2001 10:02:43 -0600
Subject: newbie: meta question
Message-ID: <3A9FC423.CEA28D54@sewardweb.com>

With three weeks and one (more or less) complete program under my belt,
I have some worries.  I want to know if I'm doing things that make sense
in python.  Am I making good use of language features?  Is there
anything I've twisted to do something when there is a tool to do just
that thing?  Am I missing anything that would turn many lines of ugly
code into a few lines of pretty code?

Those are the things I'm worried about; my question is:  Is it
considered good etiquette here to ask people to look at a program and
comment on those types of things.  I've been reading the group for a few
weeks, and have not seen anyone do that, so I thought I'd ask.

If the answer turns out to be that it is okay, the next question would
be whether to post the program (29k) or post a link to it?

Thanks,

Lloyd Sommerer



From cce at clarkevans.com  Sun Mar 25 22:11:32 2001
From: cce at clarkevans.com (Clark C. Evans)
Date: Sun, 25 Mar 2001 22:11:32 -0500 (EST)
Subject: Yet Another PEP:  Interface Adapter Mechanism __adapt__
In-Reply-To: <Pine.LNX.4.21.0103251549140.8511-100000@server01.shindich.com>
Message-ID: <Pine.LNX.4.21.0103252208070.10200-100000@clarkevans.com>

On Mon, 26 Mar 2001, Alex Shindich wrote:
> It sounds like your PEP suggests adding Strategy pattern
> behavior to all Python objects.

I'm suggesting a bi-directional adapter pattern.  IMHO,
It has nothing to do with the strategy pattern.   Did you
read the proposal?  Evidently my explanation failed, hunh?

> Another way of looking at your PEP is to say that you are 
> asking for Python Java-like interfaces.

No, actually there are other PEPs that do that.

;) Clark




From aleaxit at yahoo.com  Wed Mar 14 12:35:26 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Wed, 14 Mar 2001 18:35:26 +0100
Subject: Who's minister of propaganda this week?
References: <B98041329020E007.7DA84CAD9D0562AB.817BBC572F480869@lp.airnews.net> <98nu1m$12e@dispatch.concentric.net> <660ACCA1E4816F7E.C359CDF4168EF3FB.3B6B05538D619770@lp.airnews.net> <3AAFA13D.76E32C3F@divalsim.it>
Message-ID: <98oa4o01i4v@news2.newsguy.com>

"Nicola Musatti" <objectway at divalsim.it> wrote in message
news:3AAFA13D.76E32C3F at divalsim.it...
> Hi,
> I'm a Python curious not yet sold to the advantages of typeless
> languages; I tend to think that typelessness is suited to a set of tasks
> that has only a limited overlap with the tasks for which static typing
> is preferable.

Indeed, for any given task, either static or dynamic typing
is 'preferable' -- not both (by definition of 'preferable');
so, the overlap between the set of tasks for which each is
best suited is nil...!-)  [Rather tautologically].


> > >Robert C. Martin meant two things. A> If you write wall-to-wall unit
tests
> > >when you code, if you use a typeless language you will add bullet-proof
> > >features faster than if you use a statically typed language, even
though
> > >the former has less built-in error checking at type conversion time.
The
> > >latter supports refactoring easier.
>
> But those things that the compiler used to check for you may well have
> become new tests for you to write:

Except this "may well" does not in fact happen -- or, at least, this
seems to be the common experience of several people known as being
pretty good at statically typed languages (Martin, Eckel, me, just to
name a few:-).  You do need to check via testing anyway that all
meaningful objects are "behaving as needed" -- and this turns out
to cover ``having the right type'' as a by-product.

Not surprising.  Say you're passing a reference to an abstract itf
which has method 'foo' taking one int parameter.  The compiler can
assure that, yes, all objects passed do satisfy that interface.  But
*what ensures the calls to theobj.foo(23) actually DO what they're
MEANT to do...*?!  Answer: the unit-tests.  But then, the unit tests
would ensure failure (at runtime, during the test) if theobj had no
foo method at all, or a method not callable with an int argument.

So, all the compile-time checking is buying is catching (a small
subset of) the errors that would be caught in testing anyway, a
little bit earlier (thus, a little bit cheaper) -- it's never the
case that one has to write a test which would not be needed at
all if type-checking was static, since "the object has the right
type" is a small subcase of "the object _behaves_ per the specs".


> >   Expressiveness ... is a far better aid to
> >   correctness ...
>
> Here I'm puzzled, but maybe it's only a "language" problem. What do you
> mean by "Expressiveness"? I ask because until now I thought
> expressiveness was an advantage of strongly typed languages (namely,
> C++) over typeless ones.

I'll pass on this one, as I don't know what the original quote
meant when seeing it out of context like this.  I consider the
expressiveness of C++ on a par with Python's -- actually, what
you can do in a C++ template is quite close to what you can do
in a Python function == the Python advantage is that you're not
limited to doing it at compile-time only, the C++ advantage one
of speed (doing more at compile time means faster-running code,
all other things being equal -- of course, sometimes they aren't).


Alex





From grante at visi.com  Mon Mar 12 17:53:19 2001
From: grante at visi.com (Grant Edwards)
Date: Mon, 12 Mar 2001 22:53:19 GMT
Subject: This math scares me
References: <3aad4062.1290021@News.CIS.DFN.DE> <Hwbr6.1139$Tg.127571@ruti.visi.com> <3aad4485.2348499@News.CIS.DFN.DE> <DZbr6.17917$Qb7.3117518@newsb.telia.net>
Message-ID: <zrcr6.1144$Tg.129254@ruti.visi.com>

In article <DZbr6.17917$Qb7.3117518 at newsb.telia.net>, Fredrik Lundh wrote:
>costas at springmail.com wrote:
>> I will also tell my customers that this is the new math.
>
>if you don't understand how floating point numbers work, you
>probably shouldn't use them in customer projects.

Fredrik is right.  Using floating point numbers requires that
you know what you're doing.  They are _not_ exact.  If your
customers are expecting exact results and you're using floating
point numbers, then you did a bad thing.

People should have to pass a test before they're allowed to use
floating point numbers.    1/2 ;)

Do _not_ use floating point numbers for currency.  Use integers
or fixed point BCD.  Use floating point numbers to represent
quantities that vary continuously and are not known exactly:
the temperature outside, the length of a piece of string, etc.

-- 
Grant Edwards                   grante             Yow!  Where do your SOCKS
                                  at               go when you lose them in
                               visi.com            th' WASHER?


From a2619725 at smail.uni-koeln.de  Thu Mar  8 17:28:52 2001
From: a2619725 at smail.uni-koeln.de (Oliver Hofmann)
Date: Thu, 8 Mar 2001 23:28:52 +0100
Subject: UserDict and items()
References: <98656v$cud$1@news.rrz.Uni-Koeln.DE> <slrn9aed8i.cg1.scarblac@pino.selwerd.nl>
Message-ID: <9890g1$e9k$1@news.rrz.Uni-Koeln.DE>

Thanks Steve, Remco!


Looks like I should have been more precise or given a better description.
As it is a lot of problems mentioned here do not show up in the actual
application.

>> class ClText(UserDict):
>>     def __init__(self):
>>         UserDict.__init__(self)
>>
>>         # Set initial value
>>         self.data['test'] = ''
>>
>Since __getitem__() is checking for this value, why bother to add it to the
>dictionary?

The dictionary is supposed to contain information from different tables of
a database. When it is initialized I set various keys with empty strings as
values since this allows a user to look at allowed keys; it also allows
iterating via 'for a in dict.keys():' etc

> wouldn't a = ClText() be better? "class" is a reserved word.

Absolutely, that's what I get for quickly typing up an example.

> It's early morning, but it seems to me that when items() is called, that
is
> when all the entries are actually requested, so that is the latest point
at
> which you can set all the data[key] things from the database values. Use

Sigh :-) This occured to me last night, about ten minutes after sending out
the mail. Can you believe that this kept me busy for more than hours earlier
on? Raah. Anyhow, items(), values() etc are just going to call a setall()
method in case not all keys have been filled with values yet. The database
is in read-only mode during this time.

>   def __getitem__(self, key):
>      return self.data.setdefault(key, DatabaseCall(key))

Need to take a good look at setdefault; didn't even consider this yet.

> The last option is not to use items() :-)

*lol* That was next on my list.

Thanks again to both of you!

 Oliver


--
Oliver Hofmann - University of Cologne - Department of Biochemistry
setar at gmx.de - o.hofmann at smail.uni-koeln.de - connla at thewell.com





From spamdropbox at myrealbox.com  Sun Mar  4 19:07:59 2001
From: spamdropbox at myrealbox.com (Walter Hofmann)
Date: Mon, 5 Mar 2001 01:07:59 +0100
Subject: CGI & Python (error)
References: <3A9F5FF5.34E01148@usa.net> <pkNn6.4257$c7.1344814@news3.rdc1.on.home.com> <3AA1A066.D4F67A16@cs.utwente.nl> <0m15atgoagbroulvhjk615ec6bk8mic40n@4ax.com>
Message-ID: <slrn9a5m6v.d5o.spamdropbox@frodo.uni-erlangen.de>

On Sun, 04 Mar 2001 18:19:20 GMT, David Fuess <fuess at att.net> wrote:
>
>Debugging CGI can be a trying task. You might want to enclose the
>entire CGI in a try-except structure so you can trap all errors and
>return a reasonable result to the server and write an error log for
>yourself.

I once wrote the attached module for debugging purposes. Start your CGI
with

import log
log.init ("/path/to/logfile", "nameofprogram")

and all error messages should end up in the log file, unless you have a
syntax error in the main module. You can also use the error, warning and
info functiond to log things.

Walter



#-------------------------------------------------------------
# This file contains the logging functions
#-------------------------------------------------------------

import sys, time, string

def error (message):
    file.write (time.asctime (time.gmtime (time.time ()))+' UTC [error] '+name+': '+message+'\n')
    file.flush ()

def warning (message):
    file.write (time.asctime (time.gmtime (time.time ()))+' UTC [warning] '+name+': '+message+'\n')
    file.flush ()

def info (message):
    file.write (time.asctime (time.gmtime (time.time ()))+' UTC [info] '+name+': '+message+'\n')
    file.flush ()

class logredirect:
    def __init__ (self):
        self.saved = ""
    def write (self, message):
        list = string.split (self.saved+message, "\n")
        self.saved = list[-1]
        del list[-1]
        for line in list:
            error (line)

def init (logfile, myname):
    global file, name
    file = open (logfile, "a")
    name = myname
    sys.stderr = logredirect()


From Dan.Rolander at marriott.com  Fri Mar 23 15:04:15 2001
From: Dan.Rolander at marriott.com (Rolander, Dan)
Date: Fri, 23 Mar 2001 15:04:15 -0500
Subject: Python Script to dial a VPN connection, automate VPN routing 
	table updates.
Message-ID: <6176E3D8E36FD111B58900805FA7E0F80CCF6491@mcnc-mdm1-ex01>

Have you looked at the win32ras module? There is a Dial method.

Also, I have Windows 2000 with a VPN connection to my office over DSL. I
don't have to "dial" to setup a connection. I just use my LAN. Not sure
what's going on with your setup. Perhaps it's due to your client. I use
CheckPoint SecuRemote.

Dan

-----Original Message-----
From: Warren Postma [mailto:warren-postma at home.com]
Sent: Friday, March 23, 2001 2:38 PM
To: python-list at python.org
Subject: Python Script to dial a VPN connection, automate VPN routing
table updates.


Sorry for the nearly-off-topic nature of this post, but ....

I work from home now, and there's something in Windows 2000 that's
annoyingly manual:

I have a permanent internet connection via @home cable modem, and I "dial"
through that to create a VPN connection to the office.  The problems are
two:

(1) Windows won't accept or save my userid, password and the domain name, so
I have to enter it manually.

(2) After connecting, I have two default gateways in my routing tables
(ROUTE PRINT) and I need to update it so only the traffic that is destined
to the office goes through that interface.  The IP addresses I get given are
different each time, making a batch file insufficient, so voila, a perfect
use for python.

So after solving #2 above in Python, I wondered, does anyone know how, in
Python win32 extensions, one might dial an RAS connection, and if it's
possible to do so  through the win32 API somewhere or through COM? Then I
could solve #1 above as well.

Anyone else think this might make a nice little Python applet if I build in
some more VPN/RAS stuff and maybe even Gui-fy it a bit in wxPython or
tkInter.

Warren Postma



-- 
http://mail.python.org/mailman/listinfo/python-list



From jkraska1 at san.rr.com  Tue Mar 13 22:22:57 2001
From: jkraska1 at san.rr.com (Courageous)
Date: Wed, 14 Mar 2001 03:22:57 GMT
Subject: Einstein's Riddle
References: <mailman.984369618.12493.python-list@python.org> <NI9r6.2583$54.2812@www.newsranger.com> <98kpaj0p6m@news2.newsguy.com> <mailman.984492395.16291.python-list@python.org>
Message-ID: <scptatsfsbfpf2926f4ohsl53od0sjald2@4ax.com>

>> Actually, I'm pretty sure she was called Wanda.
>
>Only if the german didn't have a fish licence for it.

Socrates was *not* Belgian.


C//



From martin.franklin at westerngeco.com  Fri Mar  2 03:34:20 2001
From: martin.franklin at westerngeco.com (Martin Franklin)
Date: Fri, 02 Mar 2001 08:34:20 +0000
Subject: Calling win32ui from Tkinter app......
Message-ID: <3A9F5B0C.2B8D2E44@westerngeco.com>

Hi,

Hope i'm not being too stupid....  I want to call a win32
print dialog from within a Tk(inter) application.  Can I do
this if so any examples?


Thanks in advance,
Martin.



From c.evans at clear.net.nz  Mon Mar 19 20:27:39 2001
From: c.evans at clear.net.nz (Carey Evans)
Date: 20 Mar 2001 13:27:39 +1200
Subject: PEP 241: Metadata for Python Software Packages
References: <mailman.985041900.16224.clpa-moderators@python.org>
Message-ID: <87elvtqlf8.fsf@psyche.dnsalias.org>

Andrew Kuchling <akuchlin at mems-exchange.org> writes:

[snip]

>         Artistic, BSD, DFSG, GNU PL, Lesser GNU PL, "MIT/X11", 
>         Mozilla PL, "public domain", Python, Qt PL, Zope PL, unknown,
>         nocommercial, nosell, nosource, shareware, other

"GNU PL" and "Lesser GNU PL" being the "GNU General Public License"
and "GNU Lesser General Public License" respectively, should probably
be abbreviated as "GNU GPL", and "GNU Lesser GPL" or "GNU LGPL".

DFSG could be defined as something like the following, since it might
not be immediately obvious to everyone what it means.

    The license conforms to the Debian Free Software Guidelines, but
    does not use one of the other DFSG conforming licenses listed.
    More information is available at:
    http://www.debian.org/social_contract#guidelines .

Credit to Graham Williams for compiling the similar license list for
CTAN would be nice.

-- 
	 Carey Evans  http://home.clear.net.nz/pages/c.evans/

	    "Quiet, you'll miss the humorous conclusion."


From qrczak at knm.org.pl  Thu Mar  8 11:20:53 2001
From: qrczak at knm.org.pl (Marcin 'Qrczak' Kowalczyk)
Date: 8 Mar 2001 16:20:53 GMT
Subject: "!=" is bad form. Re: sorry....never mind
References: <DZkp6.18811$pF4.3015652@news1.telusplanet.net> <fdlp6.16982$go.2157953@news0.telusplanet.net> <3AA6E308.5B63ECC6@cosc.canterbury.ac.nz> <3AA72253.FB021815@Lugoj.Com>
Message-ID: <slrn9afcb5.lou.qrczak@qrnik.zagroda>

Wed, 07 Mar 2001 22:10:27 -0800, James Logajan <JamesL at Lugoj.Com> pisze:

> The operations "<", ">", "<=", ">=", "==", "<>" form a nice
> consistent and predictable set.

Except "==". Using "=" instead would be consistent.

PS. Lisp, Ada, Eiffel and Haskell use "/=" for inequality. Icon uses
"~=" for numbers and "~==" for strings. Simula uses NE for value
inequality and "=/=" for identity inequality.  OCaml uses "=" and "<>"
for value (in)equality, and "==" and "!=" for identity (in)equality.

-- 
 __("<  Marcin Kowalczyk * qrczak at knm.org.pl http://qrczak.ids.net.pl/
 \__/
  ^^                      SYGNATURA ZAST?PCZA
QRCZAK


From donn at oz.net  Sat Mar  3 13:08:43 2001
From: donn at oz.net (Donn Cave)
Date: 3 Mar 2001 18:08:43 GMT
Subject: Thinking Pythonically (was Re: gah! I hate the new string   syntax)
References: <mailman.983485923.1136.python-list@python.org> <B7In6.16281$Qb7.2658902@newsb.telia.net> <97ohh2$polt0$1@ID-77331.news.dfncis.de> <97opsb$nhu$1@nntp6.u.washington.edu> <97oupq$qijrh$1@ID-77331.news.dfncis.de> <97p3mh$bkm$1@tyfon.itea.ntnu.no> <B6C58CBD.212F%f8dy@diveintopython.org> <97q14l$jbi$0@216.39.151.169> <7k31at0hbkuqgjtnsfts9nieskove1nli4@4ax.com> <B6C60DB9.48DA%f8dy@diveintopython.org>
Message-ID: <97rbvb$lrj$0@216.39.151.169>

I used to use map, filter, reduce, but a couple of years ago
I gave that up because I felt it would make my code harder
for other people to read.  An explicit loop is better.

	Donn Cave, donn at oz.net


From je at bton.ac.uk  Thu Mar  1 09:21:11 2001
From: je at bton.ac.uk (John English)
Date: Thu, 01 Mar 2001 14:21:11 +0000
Subject: [ANNOUNCE] pyAda, first public release
References: <3A9DB9E1.8DD0E36D@bigfoot.de> <Xvsn6.7$R4.940@news.clarkson.edu>
Message-ID: <3A9E5AD7.FAB962F5@bton.ac.uk>

Brad Clements wrote:
> 
> I thought one of the big points of using Ada was its verifiability -- making
> it more suitable for complex, "large-scale" (i.e, expensive) projects, like
> missile defense.
> 
> Plugging in a "non-verifiable" script plug-in kinda defeats this purpose..
> doesn't it?

You're asking for a pure Ada implementation of Python, then. Even
without this, being able to embed a Python (or Tcl) interpreter
into an existing system can be very useful, as it lets you provide
the system with a macro language as part of the user interface...

-----------------------------------------------------------------
 John English              | mailto:je at brighton.ac.uk
 Senior Lecturer           | http://www.it.bton.ac.uk/staff/je
 Dept. of Computing        | ** NON-PROFIT CD FOR CS STUDENTS **
 University of Brighton    |    -- see http://burks.bton.ac.uk
-----------------------------------------------------------------


From sulfugor at rocketmail.cum  Mon Mar 12 11:13:02 2001
From: sulfugor at rocketmail.cum (sulfugor)
Date: Mon, 12 Mar 2001 16:13:02 GMT
Subject: Einstein's Riddle
References: <oNPq6.396644$ge4.138479566@news2.rdc2.tx.home.com> <UdSq6.2184$54.2166@www.newsranger.com> <984380185.768920@news.aei.ca> <3AAC73A8.F12BC8B3@iit.demokritos.gr>
Message-ID: <984413582.192419@news.aei.ca>

Petasis George wrote in message <3AAC73A8.F12BC8B3 at iit.demokritos.gr>...
>sulfugor wrote:
>>
>> Gregory Jorgensen wrote in message ...
>>
>> But the fish is mentionned in the question before the clues are given. We
>> are able to find who owns each of the 4 pets mentionned in the hint list.
is
>> it so unreasonable to assume then that the last pet is a fish, since we
were
>> asked about it ? The question IMplied that one of the 5 people owned a
fish.
>>
>> sulfugor
>>
>> ps = by the way, how long is this supposed to take ( solving the riddle)
?
>>
>Yes, that the problem: it implies, it does not states :-)
>It took me about 20 minutes to solve it
>(with paper+pencil of-course :-))...


It took me about 30-35min in total (in 2 stages)  to do it. But i must
mention that i was in the metro and i was going to and returning from a
death metal concert, so i can be excused for the lousy times :)
By the way, i also doubt that only 2% of the population  could solve this.
It may be that only 2% of people can solve it under a certain number of
minutes.

It was fun to do !

sulfugor

>
>George




From Bill.Scherer at VerizonWireless.com  Tue Mar 20 12:53:48 2001
From: Bill.Scherer at VerizonWireless.com (Bill Scherer)
Date: Tue, 20 Mar 2001 12:53:48 -0500
Subject: Accessing Oracle from Python remotely ?
References: <Xns906AB9C901807PaCmAnRDLM@161.48.76.20>
Message-ID: <3AB7992C.3B4E9C85@VerizonWireless.com>

Fred - 
I've not used it with oracle v7, but DCOracle works very well with
Oracle 8x.
It is available as part of the ZOracleDA package for Zope
(http://www.zope.org).  You do not need to install zope to use it.  Just
get the ZOracleDA package and untar it.  Follow the directions inside to
build DCOracle, then put the DCoracle directory in your PYHTHONPATH. 
You will need a proper installation of the Oracle client to build
DCOracle.

Good luck.


Fred Pacquier wrote:
> 
> Hi,
> 
> I have an Oracle (7) database running on a Windows NT 4 server. Is it
> possible (easy, difficult ?) to read the data in those tables from a python
> script running on a remote Unix (Solaris) box ?
> 
> Sorry if the question is silly -- I do know a bit of python but I'm
> inexperienced in databases and their connectivity. Any pointers welcome
> though.
> 
> Thanks,
> fp
> 
> --
> YAFAP : http://www.multimania.com/fredp/
> --
> http://mail.python.org/mailman/listinfo/python-list

-- 
William K. Scherer
Sr. Member of Applications Staff - Verizon Wireless
Bill.Scherer_at_VerizonWireless.com



From bwinton at tor.dhs.org  Thu Mar 15 23:31:42 2001
From: bwinton at tor.dhs.org (Blake Winton)
Date: Fri, 16 Mar 2001 04:31:42 GMT
Subject: Little problem with the "+" operator
References: <983802310.155838463@news.t-online.de> <mailman.983804468.16428.python-list@python.org> <980i6q$crs$1@news.mathworks.com> <mt1yscm7yu.fsf@astron.berkeley.edu>
Message-ID: <slrn9b35m0.lkh.bwinton@tor.dhs.org>

On 05 Mar 2001 09:38:49 -0800, Johann Hibschman wrote:
>Joshua Marshall writes:
>>> This is infinitely more sane than:
>>> % perl -e 'print 500 + "cool"'
>>> 500
>> Comparing Python to Perl here doesn't seem very fair.  They're very
>> different languages.  In general, I'm surprised by the urgency with
>> which people often jump to compare them.
>Well, Perl's the only language that I know of that lets you add 500 to
>"cool" without an error.  Well, I suppose tcl might.  But it's still a
>fairly safe inference that the original poster was coming from Perl.

Two words.  Visual Basic.

Of course, in VB, you can also get the month of "3".
(It's January.  But the month of 1 is December.)

And don't get me started on default methods...  ;)

Later,
Blake.
-- 
9:40pm up 52 days, 21:07, 2 users, load average: 0.02, 0.09, 0.07


From root at [127.0.0.1]  Sat Mar 31 15:23:37 2001
From: root at [127.0.0.1] (deadmeat)
Date: Sat, 31 Mar 2001 20:23:37 GMT
Subject: list vs tuple
References: <mailman.986066384.30712.python-list@python.org>
Message-ID: <d1rx6.1053$p5.4024@news1.rivrw1.nsw.optushome.com.au>

> written) are the source of the problem.  Fredrik is right, and you won't
make
> progress until you're willing to entertain that.

Neither of you are reading what I'm writing.

The statement is the same in both cases, but what happens to your _data_ is
not the same. That is inconsistant.

> says, "reset your brain".  This isn't Pascal.

No, it's not specific.  In Pascal the programmer defines what is a pointer
and what is not, and the difference is specified in the syntax. In Python
does not. The difference is based on what the name refers to, and that
difference is not specified in the syntax - which is my point.






From digitig at cix.co.uk  Mon Mar 19 16:32:00 2001
From: digitig at cix.co.uk (Tim Rowe)
Date: Mon, 19 Mar 2001 21:32 +0000 (GMT Standard Time)
Subject: Why "from __future__" stinks; a counter-offer
References: <mailman.984966923.14984.python-list@python.org>
Message-ID: <memo.20010319213203.8771B@digitig.compulink.co.uk>

In article <mailman.984966923.14984.python-list at python.org>, 
carribeiro at yahoo.com (Carlos Ribeiro) wrote:

<snip>

> Part of my rants are about the name of the module - __future__.py is a 
> bad and misleading name in my opinion. One alternative would be to call 
> this module __compiler__.py. It's a better name because:

My problems are more fundamental than the name, but if it ends up being 
done this way how about "__break_code__.py"? :-)




From dev_null at email.com  Fri Mar  2 08:52:15 2001
From: dev_null at email.com (Dev_NuLL)
Date: Fri, 02 Mar 2001 13:52:15 GMT
Subject: Intersesting Questions involving extending and embedding
References: <3a9c6f89.9613214@news.digital.net> <0D1n6.15927$Qb7.2600092@newsb.telia.net> <3a9f0af7.37387355@news.digital.net> <508n6.15966$Qb7.2607706@newsb.telia.net> <3a9dc27e.1325053@news.digital.net> <SYln6.16080$Qb7.2629142@newsb.telia.net> <3a9f00e8.2826131@news.digital.net> <A7In6.16279$Qb7.2658942@newsb.telia.net>
Message-ID: <3aa0a4b8.44768661@news.digital.net>

>- If you're writing C or C++ code, you're *extending* Python.

No shit, and you could be embedding also.

>Read the docs again, and look as the samples shipped with the
>source distribution.

Thanks for your responses, but they have been useless. From reading
them it seems you didn't read my entire post or you could have
answered the main problem I am having:

" how to get the new module to directly connect to the 'Blender'
module that is builtin the Blender executable."

In other words how to write an extended module in C that can call
functions in other compiled extended modules also in C.


From robin at stop.spam.alldunn.com  Fri Mar 23 19:18:55 2001
From: robin at stop.spam.alldunn.com (Robin Dunn)
Date: Fri, 23 Mar 2001 16:18:55 -0800
Subject: wxImageList use in wxPanel
References: <mailman.985369532.6873.python-list@python.org>
Message-ID: <NJRu6.338$O_e.4849977@news.randori.com>

"Thys" <thys at klaribel.netsys.co.za> wrote in message
news:mailman.985369532.6873.python-list at python.org...
> Hi All,
>
> I need to display a few images on several places in a wxPanel, the obvious
> soluttion would be to use a wxImageList, especially as the panel changes
over
> time (a status screen of sorts).
>
> I cant seem to get it to work though, no compilation problems but the
graphics
> refuse to be displayed.
>

How are you drawing the images?  The wxImageList is a convenient way to
store and reuse images but it is not a control or widget that knows how to
draw itself and respond to events.  You have to catch the paint events, etc.
and draw the images where you want them using imageList.Draw.  Is this what
you are doing?  If so then please send a code sample or a more descriptive
explaination.


> The documentation seems to concentrate on wxListCtrl
> and wxtreeCtrl none of which is applicable in this case.

It's just that these classes use images lists as described above.  There's
nothing preventing you from doing the same, although for a custom class like
this it may be just as easier or easy to just keep track of and draw the
images yourself.

--
Robin Dunn
Software Craftsman
robin at AllDunn.com       Java give you jitters?
http://wxPython.org      Relax with wxPython!





From gregj at pobox.com  Mon Mar 12 14:46:53 2001
From: gregj at pobox.com (Gregory Jorgensen)
Date: Mon, 12 Mar 2001 19:46:53 GMT
Subject: Einstein's Riddle
References: <mailman.984369618.12493.python-list@python.org>
Message-ID: <NI9r6.2583$54.2812@www.newsranger.com>

So I'm stupid, is that it? I've taken lots of IQ tests and none of them had this
question on them; that's probably why I don't score so well! Actually I did
figure out that the German owned the fish. Furthermore I was able to determine
that the fish is named Eric. 

"I am not a looney! Why should I be tied with the epithet looney merely because
I have a pet halibut? I've heard tell that Sir Gerald Nabardo has a pet prawn
called Simon and you wouldn't call him a looney; furthermore, Dawn Pailthorpe,
the lady show-jumper, had a clam, called Stafford, after the late Chancellor,
Allan Bullock has two pikes, both called Chris, and Marcel Proust had an
haddock! So, if you're calling the author of 'A la recherche du temps perdu' a
looney, I shall have to ask you to step outside!"

-- Monty Python, "Fish License"


"If you use logic on me I'll lose my advantage." -- Michael Caine, in "Blame it
on Rio"


In article <mailman.984369618.12493.python-list at python.org>, Tim Peters says...

>It's a std feature of "advanced" IQ tests that the problem stmts are
>intentionally somewhat vague:  part of what they're testing is whether you
>can come up with the most reasonable assumptions necessary to make the
>problem *interesting*.  Since the problem statement here asked "Who owns the
>fish?", and the problem admits of a unique solution if you assume *someone*
>owns a fish, that's a more interesting assumption to make than to complain
>that the question is ill-defined; the ability to fill in gaps reasonably is
>as much a kind of intelligence as the ability to follow chains of reasoning.
>
>You could cover both cheeks by answering "if anyone owns a fish, it must be
>the German".  Then again, you may prefer to complain that "on the left of"
>doesn't necessarily imply adjacent to; or that the problem statement never
>says that a person who "lives in" a house is also its "owner"; or etc etc.
>
>reading-a-question-as-trivial-is-best-reserved-for-saints-ly y'rs  - tim

Greg Jorgensen
Deschooling Society
Portland, Oregon, USA
gregj at pobox.com


From tjg at exceptionalminds.com  Wed Mar 21 18:02:05 2001
From: tjg at exceptionalminds.com (Timothy Grant)
Date: Wed, 21 Mar 2001 15:02:05 -0800
Subject: Simple regex that has me banging my head against a wall
In-Reply-To: <20010321235241.D807@apone.network.loc>; from kalle@gnupung.net on Wed, Mar 21, 2001 at 11:52:41PM +0100
References: <20010321143144.K24341@trufflehunter.avalongroup.net> <20010321235241.D807@apone.network.loc>
Message-ID: <20010321150205.N24341@trufflehunter.avalongroup.net>

On Wed, Mar 21, 2001 at 11:52:41PM +0100, Kalle Svensson wrote:
> Sez Timothy Grant:
> > Hi again.
> > 
> > I actually thought I had this regex thing down pretty well, but
> > I'm getting beat up by what I thought should be a simple one.
> > 
> > >>> x = '$1$asdfjke$lkjasdkjfsdasdfj'
> > >>> re.sub(r'\$', r'\\\$', x)
> 
> One \ too many, it seems.  The second argument to re.sub is not a regex, so
> you don't have to escape the $.
> 
I don't believe it is that simple.

>>> re.sub(r'\$', '\$', x)
'\\$1\\$asdfjke\\$lkjasdkjfsdasdfj'

That's why I'm a bit confused.


-- 
Stand Fast,
    rhacer.

Timothy Grant, RHCE <><                              MIG #1433
tjg at exceptionalminds.com                          '00 Marauder
www.exceptionalminds.com/rhacer                   Craigelachie 
>>>>>>>>>>Linux, because rebooting is *NOT* normal.<<<<<<<<<<<
>>>>This machine was last rebooted:  64 days  3:17 hours ago<<



From grante at visi.com  Fri Mar 30 10:40:41 2001
From: grante at visi.com (Grant Edwards)
Date: Fri, 30 Mar 2001 15:40:41 GMT
Subject: Evaluating *EVERYTHING*
References: <eBLw6.3059$j_3.1383318@typhoon2.ba-dsg.net> <mailman.985935647.12028.python-list@python.org>
Message-ID: <ZN1x6.9223$SB2.838149@ruti.visi.com>

In article <mailman.985935647.12028.python-list at python.org>, Steve Purcell wrote:

>> Is there a way to get python to look through a whole file and
>> say "Whoa there bucky - I've never heard of FoobarExceptoin"
>
>You can never really track down all such possible problems,
>because the names used for objects in your python code could
>(in theory more often than in practice) be defined only at
>execution time. As an extreme example, look at the following:

Like you said, that doesn't happen as much in practice (at
least not in the programs I write).

>For that reason, I see things like this: in Python, there's no
>difference between a typo and a coding logic mistake, because
>you won't find either until you run that code in a test or 'for
>real'.

In almost all of my python programs, it would be possible to
detect typos like the OP described by static analysis. It's
possible that typos in difficult paths might not be detected
until long after the program is in use but could be easily
detected with a static analysis.  Not sure if the ROI justifies
the expenditure of writing a lint program for Python.

-- 
Grant Edwards                   grante             Yow!  ... I want a COLOR
                                  at               T.V. and a VIBRATING BED!!!
                               visi.com            


From aleaxit at yahoo.com  Sun Mar  4 03:29:18 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sun, 4 Mar 2001 09:29:18 +0100
Subject: gah! I hate the new string syntax
References: <mailman.983485923.1136.python-list@python.org> <97p7iq$kk2$1@nntp6.u.washington.edu> <97soak$qh9$1@nntp.Stanford.EDU>
Message-ID: <97sudi01i26@news1.newsguy.com>

"Amit Patel" <amitp at Xenon.Stanford.EDU> wrote in message
news:97soak$qh9$1 at nntp.Stanford.EDU...
    [snip]
> When I see split(a,b), I think, "Oh, this isn't an inherent part of a
> string-- *I* could write it, even though I didn't write the string
> object."
>
> When I see b.split(a), I think, "This is a basic part of a string, and
> I *cannot* write it."

Reasonable, but the latest assertion is a bit too strong -- tone down
to "cannot write it without typeswitching or other potential inefficiencies
since it doesn't do a general dispatch on b".

E.g., take the .read, .readline and .readlines method of file objects.

Clearly, given one of them, you might write the others as functions,
not as efficiently maybe, and maybe with some typeswitching.  It's
more efficient to let the file object supply them all, just like it's more
efficient to let the string object supply join and split methods (so
one doesn't have to typeswitch between singlebyte and Unicode, &c),
but it's an issue of practical effectiveness, not one of theoretical
purity (practicality trumps purity...).


> If I have a split-like function I'd like to write (perhaps
> re.split, or some custom split function), then what I'll see is:
>
>           function style              method style
>
>           string.split(a,b)           b.split(a)
>           re.split(a,b)               re.split(a,b)
>           mysplit(a,b)                mysplit(a,b)
>
> Note that with the function style, they're consistent -- MY function
> is on equal footing with string.split.  But with the new method style,
> the split method on strings is elevated to a privileged position that
> I cannot do for mysplit.  So now I have inconsistent syntax.

But not if the .split is a method on your objects -- re.split being a
good example, as it will be simpler (no need to cache, typeswitch,
etc) to use a bound method than a function.


Alex





From a2619725 at uni-koeln.de  Tue Mar 27 11:03:59 2001
From: a2619725 at uni-koeln.de (Oliver Hofmann)
Date: 27 Mar 2001 16:03:59 GMT
Subject: Uploading a file via POST ?
Message-ID: <99qdlf$a9r$1@news.rrz.Uni-Koeln.DE>

'lo everyone!


I am trying to write a small client in Python which submits a file
to a webpage via the POST method. Part of the .html reads:

----
<FORM METHOD=POST  ENCTYPE=multipart/form-data
   ACTION="/cgi-bin/tst.pl_batch" 
          onsubmit="return FSubmitValidation(this)">

<INPUT TYPE=text SIZE=40 NAME=Email_Address>
<INPUT TYPE=file NAME=UpLoad_File SIZE=50>  
----

I've taken a look at urllib.urlopen, however it seems to be limited
more or less to GET requests. An optional data-argument must be
in 'standard application/x-www-form-urlencoded format'; since the
form requires multipart form-data this is probably not going to 
work. 

Is there an easy way to upload a file and pass parameters to the
server script?


Thanks a lot,


	Oliver



From michael at stroeder.com  Sat Mar 10 08:09:02 2001
From: michael at stroeder.com (Michael =?iso-8859-1?Q?Str=F6der?=)
Date: Sat, 10 Mar 2001 14:09:02 +0100
Subject: Ldap / Python help
References: <mailman.984067391.2391.python-list@python.org>
Message-ID: <3AAA276E.BC9AB27A@stroeder.com>

Pablo Pernot wrote:
> 
> I found an ldap module.
> 
> Where can I find documentation about ldap/python

New docs were uploaded to:

  http://python-ldap.sourceforge.net/docs.php

Ciao, Michael.


From msoulier at nortelnetworks.com  Tue Mar 20 10:22:55 2001
From: msoulier at nortelnetworks.com (Michael P. Soulier)
Date: 20 Mar 2001 15:22:55 GMT
Subject: re.match
Message-ID: <997skf$qoc$1@bmerhc5e.ca.nortel.com>

    Hey people. Could someone explain the purpose of the match function in the
re module? If it's only difference from search is to match at the beginning of
the line, why does it exist? That's what the ^ in the regexp is for. 

    Thanks,

    Mike

-- 
Michael P. Soulier, TD12, SKY  Tel: 613-765-4699 (ESN: 39-54699)
Optical Networks, Nortel Networks, SDE Pegasus
"With sufficient thrust, pigs fly just fine. However, this is not necessarily a
good idea. It is hard to be sure where they are going to land, and it could be
dangerous sitting under them as they fly overhead." -- RFC 1925
Nortel Linux User's Group Ottawa: (internal) http://nlug.ca.nortel.com


From cwmoser at lucent.com  Tue Mar 20 13:55:01 2001
From: cwmoser at lucent.com (Carl Moser)
Date: Tue, 20 Mar 2001 13:55:01 -0500
Subject: Can Python execute a compiled C program
Message-ID: <3AB7A784.6C2E2BDA@lucent.com>

Can Python execute a compiled C-program?





From robin at jessikat.fsnet.co.uk  Fri Mar  2 19:29:07 2001
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Sat, 3 Mar 2001 00:29:07 +0000
Subject: the demise of 'from foo import * and its implications?
References: <XFMail.20010302143902.shaleh@valinux.com>
 <mailman.983574424.19418.python-list@python.org>
Message-ID: <3kVBAfATrDo6EwEM@jessikat.fsnet.co.uk>

In article <mailman.983574424.19418.python-list at python.org>, Tim Peters
<tim.one at home.com> writes
...
>It's quite an ambitious act of compression to read that as "import * is going
>away" <wink>.  And, according to the Reference Manual, "import *" at other
>than module scope has never been legitimate (nested functions or not).
...
it's an ambitious act to suppose that the new complicated
__twilight_zone__ rules are better than the old simpler rules because of
some optimisation.
-- 
Robin Becker


From pieter at nagel.co.za  Fri Mar 16 09:49:07 2001
From: pieter at nagel.co.za (Pieter Nagel)
Date: Fri, 16 Mar 2001 16:49:07 +0200
Subject: gah! I hate the new string syntax
In-Reply-To: <98t5bv0l8b@news1.newsguy.com>
References: <mailman.983485923.1136.python-list@python.org> <97sudi01i26@news1.newsguy.com> <98elrk$li8$1@nntp.Stanford.EDU> <98fc070thh@news1.newsguy.com> <98iqo2$56o$1@nntp.Stanford.EDU> <j4kr6.68739$ML1.3282814@e420r-atl2.usenetserver.com> <Pine.LNX.4.21.0103161339210.2809-100000@bast.jhb.equinox.co.za> <98t5bv0l8b@news1.newsguy.com>
Message-ID: <Pine.LNX.4.21.0103161647280.2809-100000@bast.jhb.equinox.co.za>

On Fri, 16 Mar 2001, Alex Martelli wrote:

> "Pieter Nagel" <pieter at nagel.co.za> wrote in message
> news:Pine.LNX.4.21.0103161339210.2809-100000 at bast.jhb.equinox.co.za...
>     [snip]
> > Of course you can. The technique is to modify the String class itself
> 
> It's not a class.  It's a non-class type, and does not support
> modification.

Yes. But the author I responded to said "assuming that String is a
class, not a type". My answer was made within the context of that
assumption.

-- 
     ,_
     /_)              /| /
    /   i e t e r    / |/ a g e l



From wware at world.std.com  Fri Mar  2 21:43:56 2001
From: wware at world.std.com (Will Ware)
Date: Sat, 3 Mar 2001 02:43:56 GMT
Subject: shortcut for 'self' wish
References: <3aa050ad.35416419@News.CIS.DFN.DE>
Message-ID: <G9Lox8.49o@world.std.com>

Costas Menico (costas at meezon.com) wrote:
> Since Python lets you declare your own self it maybe better to change
> the 'self' to a single letter standard.  Maybe m.property or m.method...
> def myMethod(self, arg1, arg2): ..... def myMethod(m, arg1, arg2):
> This would save a lot of typing and encourage using OOP more,

Nobody will have any complaints about you using this in your own code
but it may be an uphill battle to get it accepted as a general social
convention. The "self." convention has a lot of history behind it and
is self-explanatory (much more so than "m." is m-explanatory).

Another possibility might be to check whether your editor supports some
kind of keyboard macro, so that you can type one key and the editor
fills in "self." for you. I think you could pretty easily to this in
either Emacs or vi. Then you get to save typing without violating the
prevailing convention.

Another thought: write code with just the leading period, and write a
program to perform the translation for you, from '\.[a-zA-Z]' to 'self.x'.
That's a maintainability problem waiting to happen though; you're really
better off with the keyboard macro.

As far as encouraging the use of OOP, its efficacy as a means to manage
some kinds of complexity will be a much bigger decision factor than
saving a few keystrokes here and there.
-- 
-----------------------------------+---------------------
 22nd century: Esperanto, geodesic | Will Ware
 domes, hovercrafts, metric system | wware at world.std.com


From tjg at exceptionalminds.com  Fri Mar 16 17:19:33 2001
From: tjg at exceptionalminds.com (Timothy Grant)
Date: Fri, 16 Mar 2001 14:19:33 -0800
Subject: I come to praise .join, not to bury it...
In-Reply-To: <3AB28588.35EC06FD@sage.att.com>; from garry@sage.att.com on Fri, Mar 16, 2001 at 09:28:40PM +0000
References: <mailman.983485923.1136.python-list@python.org> <97p7iq$kk2$1@nntp6.u.washington.edu> <97rka002k9@news1.newsguy.com> <3AAF7E47.294617E1@sage.att.com> <98qeco02hn7@news1.newsguy.com> <98srjf$fov$1@tyfon.itea.ntnu.no> <98t5t40lle@news1.newsguy.com> <jwbnews-13D058.07293016032001@corp.supernews.com> <3AB28588.35EC06FD@sage.att.com>
Message-ID: <20010316141933.M3109@trufflehunter.avalongroup.net>

On Fri, Mar 16, 2001 at 09:28:40PM +0000, Garry Hodgson wrote:
> "John W. Baxter" wrote:
> 
> > I came to Python (and dragged my company along) by way of a flooded
> > Tower Books store [Bellevue, WA].  They had a [slightly] water damaged
> > "Programming Python" on the shelf at something like 60% off as part of
> > their flood sale.  [The bottoms of most of the pages are still stained
> > and oddly curled.]
> 
> you might want to arrange for another flood.  the 2nd edition
> is out, effectively a whole new book, and it looks to be
> *really* good.

Can you clarify this?

Is it really a new book? While I learned much from the first
edition, it was *so* poorly organized that I didn't learn as
much as I should have been able too.

If it has been completely changed, I will take another look and
send some more money to ORA.


-- 
Stand Fast,
    tjg.

Timothy Grant                         tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.         <><       (503) 246-3630
>>>>>>>>>>>>>Linux, because rebooting is *NOT* normal<<<<<<<<<
>>>>This machine was last rebooted:  59 days  2:33 hours ago<<



From ljohnson at resgen.com  Fri Mar 23 10:49:47 2001
From: ljohnson at resgen.com (Lyle Johnson)
Date: Fri, 23 Mar 2001 09:49:47 -0600
Subject: Good GUI for Python
References: <mailman.985314033.5120.python-list@python.org>
Message-ID: <tbms0mdmft0e91@corp.supernews.com>

I haven't confirmed this, but I don't imagine you're going to be able to use *any* GUI besides win32ui/MFC with PythonWin. They're going to be competing for who gets to be in charge of the event loop ;)

As an aside, your choice of GUI depends a good bit on the application(s) you're developing and the conditions under which you will distribute them, etc. For example, Tk is your only real choice (for the near future) if you need Macintosh support. There are licensing issues on the Windows platform for Qt (which you didn't mention yet).

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20010323/6733ba26/attachment.html>

From DSHIN at nc.rr.com  Thu Mar 29 22:31:00 2001
From: DSHIN at nc.rr.com (Daehyok Shin)
Date: Fri, 30 Mar 2001 03:31:00 GMT
Subject: [Q] Python wrapper for ArcView Shapelib?
Message-ID: <U5Tw6.209744$__6.36574947@typhoon.southeast.rr.com>

Is there a python wrapper for ArcView Shapelib?


Daehyok Shin (Peter)
sdhyok at email.unc.edu
Terrestrial Hydrological Ecosystem Modelers
Geography Department
UNC-CH





From ullrich at math.okstate.edu  Tue Mar 13 11:35:10 2001
From: ullrich at math.okstate.edu (David C. Ullrich)
Date: Tue, 13 Mar 2001 16:35:10 GMT
Subject: This math scares me
References: <3aad4062.1290021@News.CIS.DFN.DE> <mailman.984434913.16256.python-list@python.org> <t98satkcr7gg4d8g155mtu93k88ndvteui@4ax.com>
Message-ID: <3aae49f3.606381144@news.okstate.edu>

On Tue, 13 Mar 2001 13:37:46 GMT, David Fuess <fuess at att.net> wrote:

>On Mon, 12 Mar 2001 17:03:56 -0500, William Park <parkw at better.net>
>wrote:
>
>>On Mon, Mar 12, 2001 at 09:34:58PM +0000, costas at springmail.com wrote:
>>> Ok, I can see maybe division having problems. But why does addition of
>>> the two numbers below:
>>> 
>>> 5.01+5.54
>>> 
>>> give me this?
>>> 
>>> 10.550000000000001
>>
>>It's called 'Round-off error'.
>
>Well, truncation error actually 

Actually it's round-off error. Truncation error is something else.
(It happens a lot that we have something that _converges_
to the right answer but we don't have time to wait: We're
adding an infinite series but we actually only consider the
first N terms, or we have some iterative scheme to approximate
the solution to something and we stop after the first N
iterations, whatever. The truncation error is the error that
gets introduced because we only took finitely many terms -
it's the error that would be there even if the individual
steps were exact calculuations, the error induced by the
fact that the algorithm we're actually _using_ does not
quite give the right answer even theoretically. Otoh the 
round-off error is the error introduced by the fact that 
the individual steps are not exactly right. Here there
is no truncation error: the expression "5.01+5.54"
should theoretically give the exact value of the sum
of those two numbers.)

>But who's counting? 

Nobody, luckily.

>(The name derives
>from truncating the infinite sequence to the finite number of bits
>available for representing the floating point mantissa.)
>
>Dave
>>
>>---William Park, Open Geometry Consulting, Linux/Python, 8 CPUs.
>



From fcy at enteract.com  Fri Mar 23 17:53:43 2001
From: fcy at enteract.com (Fred Yankowski)
Date: 23 Mar 2001 22:53:43 GMT
Subject: Subject: Re: VPython/Tk app works on single processor but not dual
References: <mailman.985376269.19505.python-list@python.org>
Message-ID: <99gk5n$pcj$1@bob.news.rcn.net>

In article <mailman.985376269.19505.python-list at python.org>,
 <Arthur_Siegel at rsmi.com> wrote:
>The source distro at:
>http://virtualphoton.pc.cc.cmu.edu/projects/visual/developers.htm
>has a VC++ project file.

'D'oh!  How did I miss that?  Thanks.

>You will have something to share sometime soon?  

My one example so far is at <http://ontosys.com/src/residue.py>.

--
Fred Yankowski
fred at ontosys.com



From root at [127.0.0.1]  Sat Mar 31 05:01:16 2001
From: root at [127.0.0.1] (deadmeat)
Date: Sat, 31 Mar 2001 10:01:16 GMT
Subject: list vs tuple
References: <hGlw6.782$p5.3210@news1.rivrw1.nsw.optushome.com.au> <3AC20CD3.265A2D94@alcyone.com> <ru69cto61jtmrqsf809e38hb3pbm2lhi7g@4ax.com> <bC1x6.4319$4N4.905946@newsc.telia.net> <71q9ct4le63anefmiajj5cbh3mg04qu1ga@4ax.com> <mailman.985991202.5578.python-list@python.org> <7Bfx6.989$p5.3791@news1.rivrw1.nsw.optushome.com.au> <Ukhx6.4421$4N4.950603@newsc.telia.net>
Message-ID: <MVhx6.1021$p5.4157@news1.rivrw1.nsw.optushome.com.au>

> since you don't know what python does, why don't
> you just shut up?

Ooh, poor diddums got proven wrong. I blew the lid on your little baby.
<cries>

What python DOES is irrelevant - the way it's written (and therefore viewed
by programmers coming from other languages) makes it inconsistant.  The same
statement (a = b) behaves entirely differently in each case.





From dg at pearl.tao.co.uk  Mon Mar  5 12:18:00 2001
From: dg at pearl.tao.co.uk (David Given)
Date: Mon, 5 Mar 2001 17:18:00 +0000
Subject: Coroutines, and memory leaks
Message-ID: <8oh089.7it.ln@127.0.0.1>

Two questions here. Both this relate to a project I'm currently working
on, a Gnome MUA written in Python (see http://sqmail.sourceforge.net).

1. Does Python do coroutines?

Reason: at one part of the program I want to run a command-line-like
routine, that processes and emits log messages at intervals. I want to
display the result in a window. The most elegant way of connecting this
routine to GTK is to put them in different threads; but since I don't need
true concurrency, coroutines would be the simplest solution.

2. I have this massive memory leak. How can I find it?

The more I use the program, the more memory it uses. I've seen it use
>100MB before I killed it. I guess that I've got cyclic memory structures
in the message cache, so it's never freeing any of the mail I look at. Are
there any suggestions for how I can find these and kill them?

-- 
+- David Given ---------------McQ-+ "Opportunity is missed by most people
|  Work: dg at tao-group.com         | because it's dressed in overalls and looks
|  Play: dgiven at iname.com         | like work." --- Thomas Edison              
+- http://wired.st-and.ac.uk/~dg -+ 


From nils.hensel at online.de  Wed Mar 21 12:23:00 2001
From: nils.hensel at online.de (Nils Hensel)
Date: Wed, 21 Mar 2001 18:23:00 +0100
Subject: Stupid String Question
References: <3AB8327C.493C70A5@direct.ca>
Message-ID: <99ao02$7a2$1@news.online.de>

It's
a = '*' * 30

Nils

"Brian & Colleen" <greybria at direct.ca> schrieb im Newsbeitrag
news:3AB8327C.493C70A5 at direct.ca...
I know I've seen this somewhere but dejanews has not yielded the answer with
a variety of searches.
What is the pretty way to generate a string of 30 asterisks. I'm sure it's
not:
def fill(c,n):
    s=''
    for i in range(n):
        s=s+c
    return s
a=fill('*',30)

--
 Brian Smith
 greybria at direct.ca
 http://mypage.direct.ca/g/greybria





From root at rainerdeyke.com  Sat Mar 17 12:54:42 2001
From: root at rainerdeyke.com (Rainer Deyke)
Date: Sat, 17 Mar 2001 17:54:42 GMT
Subject: Python style questions
References: <98tb91$9t9$1@saltmine.radix.net> <98tdu1$cfk$1@panix2.panix.com> <1eqdo5u.1xxmzfi2kjk0qN%francois.granger@free.fr> <98u24p0a98@news2.newsguy.com> <1eqfb5l.10gfjcu1qi9xrwN%francois.granger@free.fr>
Message-ID: <CxNs6.432931$ge4.152635172@news2.rdc2.tx.home.com>

"Fran?ois Granger" <francois.granger at free.fr> wrote in message
news:1eqfb5l.10gfjcu1qi9xrwN%francois.granger at free.fr...
> I am currently on the following dispatch issue wich is far from corect:
>
> def game():
>     print 'Starting game.'
>     a = tictactoe() # create a new game
>     ans = raw_input('Who start (h/c) : ')
>     if  (ans[0]=='q' or ans[0]=='Q'):
>         return
>     human = userinput   # proc to handle human interaction
>     c = player('X') # create a machine player
>     computer = c.play
>     if (ans[0]=='h' or ans[0]=='H'):
>         first, second = human, computer
>     else:
>         first, second = computer, human
>     while 1:
>         if a.turn % 2:
>             move, token = second(a)
>         else:
>             move, token = first(a)
>         a.valid(move, token)
>         a.display()
>         win = a.won()
>         if win == 'tie':
>             print 'tie'
>             # call player with tie
>             # exit game
>         elif win == 'X':
>             print 'X'
>             # call player with X
>             # exit game
>         elif win == 'O':
>             print 'O'
>             # call player with O
>             # exit game
>         else:
>             continue

I would do something like this:

class Player:
  def __init__(self, game, side):
    self.game = game
    self.side = side

class HumanPlayer(Player):
  def move(self, game):
    ...

class GAPlayer(Player):
  def move(self, game):
    ...

class NNPlayer(Player):
  def move(self, game):
    ...

class Game:
  def __init__(self, player1, player2):
    self.players = [player1, player2]
    self.active_player = 0
  def run(self):
    while not self.game_over():
      self.apply_move(self.players[active_player].move(self))
      self.active_player = (self.active_player + 1) % 2

Game(HumanPlayer(), NNPlayer()).run()

--
Rainer Deyke (root at rainerdeyke.com)
Shareware computer games           -           http://rainerdeyke.com
"In ihren Reihen zu stehen heisst unter Feinden zu kaempfen" - Abigor




From cg at schlund.de  Sat Mar 10 13:26:57 2001
From: cg at schlund.de (Carsten Gaebler)
Date: Sat, 10 Mar 2001 19:26:57 +0100
Subject: How to efficently build a nested dictionary
References: <3AAA2042.E69A53FC@gmx.de> <98dgjb$iee$1@panix3.panix.com>
Message-ID: <3AAA71F1.7FC7597A@schlund.de>

Aahz Maruch wrote:
> >
> >traffic = {dir: {year: {month: {day: 42}}}}
> 
> I think Emile's suggestion to use a single compound key is a good one.
> If that's not sufficient *and* you can use Python 2.0, take a look at
> the dict.setdefault() method.

How could the setdefault() method help me out?

cg.


From deblNO at nospammytheworld.com  Tue Mar 27 23:48:08 2001
From: deblNO at nospammytheworld.com (David Lees)
Date: Wed, 28 Mar 2001 04:48:08 GMT
Subject: Help with regular expression backreferences
Message-ID: <3AC16D0F.6E33377E@nospammytheworld.com>

I would like to use regular expression back references as keys into a
dictionary when doing substitution.  So far all I have is:

s='This xxMYJunkkxx the string that xxYourJunkxx the thing'
p=re.compile(r'xx(\w+)xx')
new_s=re.sub(p,'junk',s)
print new_s

which substitutes a fixed expression for the matches.  In this example
there are 2 matches, one of which is 'MYJunk' and the other 'YourJunk'
but I can not figure out how to get at the backreference values in the
re.sub

Thanks in advance.

David Lees


From greybria at direct.ca  Tue Mar 20 23:47:57 2001
From: greybria at direct.ca (Brian & Colleen)
Date: Tue, 20 Mar 2001 20:47:57 -0800
Subject: Stupid String Question
Message-ID: <3AB8327C.493C70A5@direct.ca>

I know I've seen this somewhere but dejanews has not yielded the answer
with a variety of searches.

What is the pretty way to generate a string of 30 asterisks. I'm sure
it's not:

def fill(c,n):
    s=''
    for i in range(n):
        s=s+c
    return s

a=fill('*',30)


--
 Brian Smith
 greybria at direct.ca
 http://mypage.direct.ca/g/greybria


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20010320/e4a648a1/attachment.html>

From root at [127.0.0.1]  Sat Mar 31 16:16:54 2001
From: root at [127.0.0.1] (deadmeat)
Date: Sat, 31 Mar 2001 21:16:54 GMT
Subject: list vs tuple
References: <hGlw6.782$p5.3210@news1.rivrw1.nsw.optushome.com.au><3AC20CD3.265A2D94@alcyone.com><ru69cto61jtmrqsf809e38hb3pbm2lhi7g@4ax.com><bC1x6.4319$4N4.905946@newsc.telia.net><71q9ct4le63anefmiajj5cbh3mg04qu1ga@4ax.com><mailman.985991202.5578.python-list@python.org><7Bfx6.989$p5.3791@news1.rivrw1.nsw.optushome.com.au><Ukhx6.4421$4N4.950603@newsc.telia.net> <mailman.986049281.32365.python-list@python.org> <uGqx6.1052$p5.3997@news1.rivrw1.nsw.optushome.com.au> <slrn9cchb4.3oa.qrczak@qrnik.zagroda>
Message-ID: <aPrx6.1056$p5.4182@news1.rivrw1.nsw.optushome.com.au>

> It *never* copies the data across,

OK

> thus it is consistent.

The syntax is not.





From whisper at oz.net  Sun Mar 11 13:51:30 2001
From: whisper at oz.net (Dave LeBlanc)
Date: 11 Mar 2001 18:51:30 GMT
Subject: mod_python or MoinMoin problem.
References: <98fioq$qud$1@news2-2.kornet.net>
Message-ID: <98ghfi$lck$0@216.39.170.247>

I'm using MoinMoin on Apache 1.3.17 (and more recently, 1.3.19) and
Windows NT using cgi. I've tried to get MoinMoin to run under
mod_python, but I think the problem is an incompatibility with how
mod_python handles cgi compared to regular cgi.

It's also possible I just didn't write the handler correctly - there's
precious little doc about that from the mod_python crew.

If you get it working, let me know please :)

Dave LeBlanc

On Sun, 11 Mar 2001 19:16:48 +0900, "Seo ji won" <skidrow at nownuri.net>
wrote:

>I'm now using FreeBSD 4.2 release, apache 1.3.14, mod_python 2.6.3
>
>and trying to use MoinMoin(one of wiki clones, written in python, and is
>kind of CGI program I guess)
>
>but it doesn't work out.
>
>the apache log says,
>----------------------------------------------------------------------------
>-------------------------
>Mod_python error: "PythonHandler mod_python.cgihandler"
>
>Traceback (most recent call last):
>
>  File "/usr/local/lib/python2.0/site-packages/mod_python/apache.py", line
>185, in Dispatch
>    result = object(self.req)
>
>  File "/usr/local/lib/python2.0/site-packages/mod_python/cgihandler.py",
>line 76, in handler
>    imp.load_module(module_name, fd, path, desc)
>
>  File "/usr/local/www/cgi-bin/MoinMoin/wiki-moinmoin/moin.py", line 15, in
>?
>    cgimain.run()
>
>  File "/usr/local/www/cgi-bin/MoinMoin/cgimain.py", line 161, in run
>    cgi.print_exception()
>
>  File "/usr/local/lib/python2.0/cgi.py", line 870, in print_exception
>    print
>
>  File "/usr/local/lib/python2.0/site-packages/mod_python/apache.py", line
>484, in write
>    h, v = string.split(line, ":", 1)
>
>ValueError: unpack list of wrong size
>----------------------------------------------------------------------------
>-----------------------------
>
>
>
>
>
>What seems to be the problem?..
>Thanks in advance. :)
>p.s) If this news group is not proper to post this message, plz let me
>know. -- with suitable news group. :)
>
>
>



From sholden at holdenweb.com  Thu Mar 22 22:54:25 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 22 Mar 2001 22:54:25 -0500
Subject: extending the built in dictionary type
References: <umfKujAz2iu6EwHa@jessikat.demon.co.uk>
Message-ID: <6Rzu6.12951$_46.485491@e420r-atl2.usenetserver.com>

"Robin Becker" <robin at jessikat.fsnet.co.uk> wrote in message
news:umfKujAz2iu6EwHa at jessikat.demon.co.uk...
> I'm interested in extending the built in dictionary. It's fairly simple
> to wrap, but seems incredibly hard to chain the methods. Is there any
> obvious way to use most of the built in methods whilst modifying a few.
>
> Seems that although it exists I can't chain the original mapp_methods of
> dictobject as that is not visible outside pythonxx.dll. So is there an
> easier way to get at these internal types?

See FAQ 5.18: How do I define and create objects corresponding to
built-in/extension types?

regards
 Steve





From vlouie at telusplanet.net  Wed Mar  7 02:14:19 2001
From: vlouie at telusplanet.net (Victor Louie)
Date: Wed, 07 Mar 2001 07:14:19 GMT
Subject: sorry....never mind
References: <DZkp6.18811$pF4.3015652@news1.telusplanet.net>
Message-ID: <fdlp6.16982$go.2157953@news0.telusplanet.net>

found it...

The comparison operators <> and != are alternate spellings of the same
operator. != is the preferred spelling; <> is obsolescent.

...if anyone was wondering


"Victor Louie" <vlouie at telusplanet.net> wrote in message
news:DZkp6.18811$pF4.3015652 at news1.telusplanet.net...
> Hi,
>
> I was just wondering if any knew what <> does....I've got some sample code
> and can't figure out what's happening
>
> eg.
> x =10
> if x <> 10:
>  ....
>
> thanks
> Victor
>
>




From johnw at gnu.org  Sun Mar 11 14:47:18 2001
From: johnw at gnu.org (John Wiegley)
Date: 11 Mar 2001 12:47:18 -0700
Subject: Going ZODB
References: <mailman.984324988.25505.python-list@python.org>
Message-ID: <m3wv9w3x6h.fsf@gnu.org>

>>>>> On Sun Mar 11, Steven writes:

> I am curious as to how many people in this list actually use ZODB in
> production environment and what is the typical uses of it
> (especially ZODB with relational/berkeleydb storage). Aside from
> along with Zope, of course.

> Also, there are one issue I am considering before going ZODB. If
> later on the application grows beyond the scale that ZODB can
> provide at the moment, would it be awkwardly hard to convert the
> code to use other backend.  For example in the relational world, if
> I keep the SQL statements portable enough, I can just,
> theoretically, switch the backend with another DB.

I tried to use ZODB as the persistence engine for my MOO clone
(http://doop.sourceforge.net).  I found it's restrictions too onerous,
most especially the restriction that I not override __getattr__ or
__setattr__.  Lastly, I found that debugging problems was hideously
difficult, and that mostly I had to resort to a binary search using
print statements, just about all the time.  Maybe they've improved the
robustness of this.

Feature-wise, I would love to use it.  If I could ever find out how to
use ZODB *and* overload get/setattr, I think I'd switch.  What I'm
doing now (just checkpointing the whole database at time intervals) is
too basic.


From gerhard.nospam at bigfoot.de  Sun Mar 18 14:01:39 2001
From: gerhard.nospam at bigfoot.de (Gerhard =?iso-8859-1?Q?H=E4ring?=)
Date: Sun, 18 Mar 2001 20:01:39 +0100
Subject: installing 2.x over the top of 1.5.2
References: <mailman.984861502.27092.python-list@python.org>
Message-ID: <slrn9ba2cf.2mr.gerhard.nospam@lilith.hqd-internal>

On Sat, 17 Mar 2001 12:39:50 -0800, Timothy Grant wrote:
>This problem has been bothering me for a while.
>
>Python v1.5.2 is installed by default on RH6.2 boxes. 
>
>My problem is that upgrading an RH6.2 box by using the 2.x rpms
>now means that there are two versions of Python installed on my
>box, and any extra modules I have installed have to be copied
>to the new location for v2.x. I then need to uninstall the
>v1.5.2 rpms.

Have you looked if there are any Python 2.0 RPMs from Redhat themselves? For my
distribution (SuSE) they appeared shortly after the Python 2.0 release in the
contrib directory of the ftp server. I did it that way. Install Python 2.0 RPMs
from distributor. Fix .SPEC files and patches for Source RPMs. Recompile source
RPMs. For the most part, I simply could do a search for python1.5 and replace
it with python2.0 in the .SPEC files.

I always try to install things from RPM or create my own RPMs.

>I'm sure someone somewhere has come up with a clean way of
>upgrading Python. As with all thing Pythonic, there must be an
>easier way.

Yes, I think rpm --rebuild pyFooBang.SPEC is as clean as it can get.

Gerhard
-- 
mail:   gerhard <at> bigfoot <dot> de
web:    http://highqualdev.com


From eeskoe at students.wisc.edu  Mon Mar 19 23:46:55 2001
From: eeskoe at students.wisc.edu (Erika Skoe)
Date: Mon, 19 Mar 2001 22:46:55 -0600
Subject: setting Scrollbars in PMW ScrolledFrame
Message-ID: <3AB6E0BF.A5B9FCF0@students.wisc.edu>

Is there a way to set scrollbars programatically in a ScrolledFrame? and
if not, is it possible to bind scrollbars to a frame widget or do i need
to use a canvis?

thanks

e


From phd at phd.pp.ru  Mon Mar  5 09:46:15 2001
From: phd at phd.pp.ru (Oleg Broytmann)
Date: Mon, 5 Mar 2001 17:46:15 +0300 (MSK)
Subject: Little problem with the "+" operator
In-Reply-To: <983802310.155838463@news.t-online.de>
Message-ID: <Pine.LNX.4.30.0103051745110.13477-100000@fep132.fep.ru>

On Mon, 5 Mar 2001, Zamurai wrote:
> variableOne = 500
> myVariable = variableOne + "cool"
>
> But I get the following error: "unsupported operand types for +"

   Python is strict typed language. You cann add strings, or you can
integers, but you cannot add int to string. Use conversion functions -
convert int to string, e.g.

Oleg.
----
     Oleg Broytmann            http://phd.pp.ru/            phd at phd.pp.ru
           Programmers don't die, they just GOSUB without RETURN.




From ajm at enteract.com  Thu Mar 29 20:33:48 2001
From: ajm at enteract.com (Alan Miller)
Date: Thu, 29 Mar 2001 19:33:48 -0600
Subject: Text printing under Windows
References: <slrn9bk130.10l.ego@macondo.visconti.home>
Message-ID: <MPG.152d98a184ccb1fe98977c@news-enteract>

Gabriele Favalessa (ego at macondo.visconti.home) wrote:
>how do I print using a dot matrix printer in text mode under Windows
>98?  I have both a local and a remote printer.

For local printers, other people have pointed out that you just open the 
device and dump data directly to it.  For networked printers that you 
have a UNC name for, you can do much the same - the following code works 
just fine:
>>> myprinter = open(r"\\myserver\myprinter", "wb")
>>> myprinter.write("Here is my test string\n")
>>> myprinter.close()

There's also a way to open a locally-defined print queue (e.g. "HP 
LaserJet 5si in the back room") and dump data directly to it, but the 
required APIs aren't complete in Mark Hammond's win32print extensions.  
I've looked at extending them and mentioned it to Mark once, but haven't 
made the time to do so.  

The missing API calls are StartDocPrinter, WritePrinter and EndDocPrinter 
if anyone's interested; I also have a VB class that I put together to 
nicely encapsulate this if anyone wants a copy of it.  Sending raw data 
to a print queue doesn't actually seem to be all that urgent for most 
people - I did it because I needed to dump pregenerated PS and PCL files 
and was getting sick of running "copy x.ps \\servername\printername" in 
command windows.

ajm


From lluang at northwestern.edu  Tue Mar 20 20:49:21 2001
From: lluang at northwestern.edu (Louis Luangkesorn)
Date: Tue, 20 Mar 2001 19:49:21 -0600
Subject: stripping punctuation from tuples
References: <3AB57F0C.594CF3E9@northwestern.edu> <a1jt6.205998$df5.5255534@news1.crdva1.bc.home.com>
Message-ID: <3AB808A1.A6A9B7DD@northwestern.edu>

Yep, I finally got it to work.

Wayne Izatt wrote:

> Is this what you're looking for?
>
> >>> a = []
> >>> a.append(('Alan Tsang',))
> >>> a
> [('Alan Tsang',)]
> >>> a[0]
> ('Alan Tsang',)
> >>> b = a[0]
> >>> b
> ('Alan Tsang',)
> >>> (x, y) = string.split(str(b[0]))
> >>> x, y
> ('Alan', 'Tsang')
> >>> x
> 'Alan'
> >>> y
> 'Tsang'
> >>>
>
> cheerio

--
K Louis Luangkesorn
lluang at northwestern.edu  http://pubweb.nwu.edu/~kll560  PGP:0xF3E2D362
Whatsoever things are true, ... honest, ... just, ... pure, ... lovely, ... of
good report; if there be any virtue, and if there be any praise, think on
these things.- motto - Northwestern University




From hamish_lawson at btinternet.com  Fri Mar 16 20:43:50 2001
From: hamish_lawson at btinternet.com (Hamish Lawson)
Date: Sat, 17 Mar 2001 01:43:50 -0000
Subject: ZPT, a next-generation template technology
Message-ID: <98ufbu$kkh$1@uranium.btinternet.com>

The Zope Presentation Templates project is underway at Digital Creations
to develop a next-generation template technology for Zope* that is
XHTML-compliant and which makes it easier for designers to work with.
However it is hoped to design it in such a way that as much as possible
is not specific to Zope, in the belief that the technology is more
widely useful and could be adopted by other languages and tools.
Developers from other communities may want to consider contributing to
the design of ZPT to help keep it language-independent, and to consider
implementing the approach in their programming language of choice.

First of all some background on what is meant by "next-generation".

In the beginning programmers wrote program code with embedded HTML. Then
technologies like eperl, ASP, PHP and JSP allowed us to reverse this and
to embed program code in the HTML. However a school of thought wanted to
have greater separation of presentation and logic, and so various
template mechanisms were devised that provided just enough
programmability to control presentation, but which moved the business
logic out elsewhere. One example of this approach is Freemarker:

   <h1>$customername</h1>
    <table border="1">
        <tr>
            <td>Date</td>
            <td>Amount</td>
        </tr>
        <list deposits as deposit>
            <tr>
                <td>${deposit.date}</td>
                <td>${deposit.amount}</td>
            </tr>
        </list>
    </table>

Instead of custom tags some other template mechanisms use special
markers to designate control structures, but in general most template
mechanisms share the property that their templates are not compliant
with XHTML (though the rendered output may be). As such it can be
awkward trying to use such template mechanisms with authoring tools that
only understand XHTML.

The approach that the ZPT project is pursuing is to attach the control-
flow, formatting and placeholder directives as *attributes* of standard
XHTML tags. The example above thus becomes something like:

    <h1 zpt:insert="customername">Bob Smith</h1>
    <table border="1">
        <tr>
            <td>Date</td>
            <td>Amount</td>
        </tr>
        <tr zpt:repeat="deposit deposits">
            <td zpt:insert="deposit/date">01/02/2001</td>
            <td zpt:insert="deposit/amount">23.40</td>
        </tr>
    </table>

Because this doesn't make use of special tags, but employs attributes
instead, it should pass unscathed through most decent authoring tools.
Moreover (borrowing an idea from Enhydra's XMLC template scheme) the ZPT
template scheme allows typical sample content to be used as a proxy for
the real data (which is substituted by the template engine), so making
it easier to work with in an authoring tool.

More information on the rationale of attribute languages can be found

    http://dev.zope.org/Wikis/DevSite/Projects/ZPT/AttributeLanguage

An overview of ZPT can be found at

    http://www.zope.org/Wikis/DevSite/Projects/ZPT/VisionStatement


Hamish Lawson

* Zope is a leading open-source web application server and content-
management framework, written in Python by Digital Creations.






From jcollin at exis.net  Thu Mar 15 15:12:10 2001
From: jcollin at exis.net (Jay Collins)
Date: Thu, 15 Mar 2001 15:12:10 -0500
Subject: decss
Message-ID: <Pine.LNX.4.21.0103151511300.7874-100000@tarpon.exis.net>

Anyone every convert that perl script to python? I was just wondering how
it would look/perform.



From fuess at att.net  Tue Mar 13 08:37:46 2001
From: fuess at att.net (David Fuess)
Date: Tue, 13 Mar 2001 13:37:46 GMT
Subject: This math scares me
References: <3aad4062.1290021@News.CIS.DFN.DE> <mailman.984434913.16256.python-list@python.org>
Message-ID: <t98satkcr7gg4d8g155mtu93k88ndvteui@4ax.com>

On Mon, 12 Mar 2001 17:03:56 -0500, William Park <parkw at better.net>
wrote:

>On Mon, Mar 12, 2001 at 09:34:58PM +0000, costas at springmail.com wrote:
>> Ok, I can see maybe division having problems. But why does addition of
>> the two numbers below:
>> 
>> 5.01+5.54
>> 
>> give me this?
>> 
>> 10.550000000000001
>
>It's called 'Round-off error'.

Well, truncation error actually But who's counting? (The name derives
from truncating the infinite sequence to the finite number of bits
available for representing the floating point mantissa.)

Dave
>
>---William Park, Open Geometry Consulting, Linux/Python, 8 CPUs.



From park997 at aol.comnotospam  Tue Mar 13 11:59:14 2001
From: park997 at aol.comnotospam (Park997)
Date: 13 Mar 2001 16:59:14 GMT
Subject: main in mac?
References: <3AAE3978.3FCDC1BB@letterror.com>
Message-ID: <20010313115914.19243.00000247@ng-dh1.aol.com>

>No: if you look at the option popup menu -- which is very well hidden
>at the top right corner of your edit window, it looks like an arrow --
>there's an option called "run as __main__". This will set the __name__
>variable in the script to "__main__".
>
>Hope this helped,
>
>Just
>
Yes. That was the trick I needed. Thanks,

Wendell




From jafo at tummy.com  Mon Mar 19 00:55:31 2001
From: jafo at tummy.com (Sean Reifschneider)
Date: Sun, 18 Mar 2001 22:55:31 -0700
Subject: installing 2.x over the top of 1.5.2
In-Reply-To: <20010317123950.C24992@trufflehunter.avalongroup.net>; from tjg@exceptionalminds.com on Sat, Mar 17, 2001 at 12:39:50PM -0800
References: <20010317123950.C24992@trufflehunter.avalongroup.net>
Message-ID: <20010318225531.A21136@tummy.com>

On Sat, Mar 17, 2001 at 12:39:50PM -0800, Timothy Grant wrote:
>My problem is that upgrading an RH6.2 box by using the 2.x rpms
>now means that there are two versions of Python installed on my
>box, and any extra modules I have installed have to be copied

If you're using the SRPMs I put up, you may prefer to change the
.spec file to comment out the "%define binsuffix 2" line, so
that it builds an executable called "python" (instead of "python2"),
in which case it can't be installed along-side the 1.5.2 release.

So far I haven't heard any opinions on this.  I built them that
way so that I could install 2.0 and 2.1 and test them out, without
having to break any existing code.  I did make them easy to
build either way though.

Sean
-- 
 One person's data is another person's program. 
                 -- Guy L. Steele, Jr.
Sean Reifschneider, Inimitably Superfluous <jafo at tummy.com>
tummy.com - Linux Consulting since 1995. Qmail, KRUD, Firewalls, Python



From tino at wildenhain.de  Fri Mar 23 04:37:20 2001
From: tino at wildenhain.de (Tino Wildenhain)
Date: Fri, 23 Mar 2001 10:37:20 +0100
Subject: email address verification
References: <99d5qi$95l$1@solaria.cc.gatech.edu>
Message-ID: <3ABB1950.6AB582DB@wildenhain.de>

Hi,

Holland King wrote:
> 
> does python provide quick/simple way to verify that an email address exist?
> i looked at the smtp library and the verify method but that does not seem
> to be very standard anymore. is there anything else that is simalar?
> thank you.
 
you can go up to the point to send the "data" statement. If you get so 
far, you can assume, the address exists. However, you can only be sure
that it does not exists if you get an error beforehand. If you dont
get an error, the address may still be wrong.

Regards
Tino



From mhuening at zedat.fu-berlin.de  Sun Mar 11 07:43:53 2001
From: mhuening at zedat.fu-berlin.de (Matthias Huening)
Date: Sun, 11 Mar 2001 13:43:53 +0100
Subject: question , Does pytyon excist on a pda, like psion?
References: <xDJq6.28348$KE1.1127955@zwoll1.home.nl>
Message-ID: <98frub$1mhg0$1@fu-berlin.de>

> Is Python already ported to the psion or other pda, s?
>


http://www.cs.uu.nl/~otfried/Python/
http://dales.rmplc.co.uk:9080/Duncan/PyPsion.htm
(The second was unavailable when I tried it five minutes ago)

Matthias




From andreas at digicool.com  Wed Mar 14 07:54:12 2001
From: andreas at digicool.com (Andreas Jung)
Date: Wed, 14 Mar 2001 07:54:12 -0500
Subject: Help doing simple threading
Message-ID: <005101c0ac85$f8560a30$9865fea9@SUXLAP>

>I am trying to learn how to use threads.  I want multiple threads to be
>able to communicate.  I thought that threads started by a common process
>would share globals in that module.  I want thread to be able to store
>esults in a common array.  However the following returns all zeros, so
>clearly I am not passing anything back to the top level list 'z'.  Can
>someone tell me how I should be passing things back from threads?

Your code has several problems:

a) You don't synchronize your main thread with the threads you start.
    That means your main thread continues running BEFORE the
    started threads make any work.

b) Modifying shared resources within threads is dangerous. Think
   about using locks to get rid of this problem.

c) Try the threading module. It gives you a better control and programmer
    interface to work with thread in Python.


----
Andreas Jung (andreas at digicool.com)
Digital Creations - www.digicool.com, www.zope.org




From jepler at inetnebr.com  Mon Mar 12 09:28:31 2001
From: jepler at inetnebr.com (Jeff Epler)
Date: Mon, 12 Mar 2001 14:28:31 GMT
Subject: Extending Python: rewriting a single method in C
References: <g0wv9vxms0.fsf@scumbag.ecs.soton.ac.uk>
Message-ID: <slrn9apn8f.382.jepler@potty.housenet>

On 12 Mar 2001 11:13:51 +0000, Jacek Generowicz
 <jmg at ecs.soton.ac.uk> wrote:
>I have an object-oriented numerical code written in Python. It has
>come to the stage where the speed of the code is seriously impeding
>progress. Profiling reveals (surprise, surprise) that most of the time
>is spent in only 2 methods.
>
>Is there any way of re-writing just these methods in C ?

It's easy to write a function in C.  Other posters have mentioned references 
to documentation about doing this.

However, you cannot make a builtin function be a method of a class/instance.
Instead, you could write:

from cmodule import fast_a, fast_b

class fastKlass(Klass):
	def a(self, int1, int2):
		fast_a(self, int1, int2)
	def b(self, int1, int2):
		fast_b(self, int1, int2)

In fast_a and fast_b, you can get arguments by using the ParseTuple string
"(Oii)" and then use PyObject_SetAttrString and PyObject_GetAttrString
to store or retrieve attributes (including, if necessary, instance
methods) from the instance by their name.

Note that if there are *many* calls to the method, rather than each
invocation taking a large amount of time, then this approach is not
going to gain you much, since it actually *increases* the function-call
overhead.

Jeff


From emile at fenx.com  Sun Mar 18 21:25:48 2001
From: emile at fenx.com (Emile van Sebille)
Date: Sun, 18 Mar 2001 18:25:48 -0800
Subject: Subsetting a dictionary
References: <993dkm$ttr$1@news6.svr.pol.co.uk> <993qq3$4bajg$1@ID-11957.news.dfncis.de>
Message-ID: <993qsn$49989$1@ID-11957.news.dfncis.de>

"Emile van Sebille" <emile at fenx.com> wrote in message
news:993qq3$4bajg$1 at ID-11957.news.dfncis.de...
> Rather than copying and then deleting from the copy, try building a new
> dictionary while adding only those that qualify.
>
> Something like:  (untested)
>
> def subset(self, attr, value):
>                  # Copy semantics
>                  td = {}
without this line...Sorry :{
>                  td.data = self.data.copy()

>                  for ky, val in TrainingData.items():
>                           if ky[attr] == value:
>                                     td[ky] = val
>                  return td
>




From slhath at home.com  Tue Mar  6 15:51:01 2001
From: slhath at home.com (Scott Hathaway)
Date: Tue, 06 Mar 2001 20:51:01 GMT
Subject: wierd tkinter window issue
Message-ID: <V4cp6.142393$bb.1470686@news1.rdc1.tx.home.com>

I have a window that I call overridedirect(1) on to remove the titlebar on
Windows.  Then, I have a button that minimizes the app.  To minimize by
code, I have to call overridedirect(0) and then iconify() to minimize.

Then, what event is called when the window is maximized again, so that I can
remove the titlebar again with overridedirect(1)?

I have tried Activate, Enter, and FocusIn.

TIA,
Scott




From ullrich at math.okstate.edu  Sun Mar  4 14:13:36 2001
From: ullrich at math.okstate.edu (David C. Ullrich)
Date: Sun, 04 Mar 2001 19:13:36 GMT
Subject: Hitting two targets:  OO + group theory (pedagogy)
References: <u2k0atkgkiatdp9ukj3mpd1ej0b648ldsm@4ax.com> <3aa1077a.1110803@nntp.sprynet.com> <0932atgt9v8d8aboq90evoi5ipj5c0sk8s@4ax.com> <3aa11a91.5998558@nntp.sprynet.com> <p5vo6.10990$ML1.527923@e420r-atl2.usenetserver.com>
Message-ID: <3aa293a6.1642116@news.okstate.edu>

On Sun, 4 Mar 2001 12:35:47 -0500, "Steve Holden"
<sholden at holdenweb.com> wrote:

>"David C. Ullrich" <ullrich at math.okstate.edu> wrote in message
>news:3aa11a91.5998558 at nntp.sprynet.com...
>[ ... ]
>> Some people would say that the reference to Genesis
>> is out of place here... probably you know that some
>> people would say that. Might be regarded as just
>> generally inappropriate - some other people might
>> think that the bit about Abel is misleading, since
>> "Abelian" _does_ refer to a man named Abel, but
>> not the one in Genesis.
>>
>[ ... ]
>
>What's purple and commutes?
>

A purple commutative book of Genesis?

>
>A: an Abelian grape.

Right - I knew it was something like that.

>sorry
> Steve
>
>
>



From jcollin at exis.net  Tue Mar 13 10:30:10 2001
From: jcollin at exis.net (Jay Collins)
Date: Tue, 13 Mar 2001 10:30:10 -0500
Subject: string.find() ?
Message-ID: <Pine.LNX.4.21.0103131027570.13200-100000@tarpon.exis.net>

Hello,

 I'm using string.find to look for a string I get out of a popen() and it
doesn't appear to ever find what I'm looking for. Below is an example. It
always returns -1 no matter what I'm looking for. I was even looking for
root which should show up a bunch of times. Hopefully, someone can tell me
what I'm doing wrong.

-----------------------------<Cut Here>--------------------------------

import sys
import os
import string




output = os.popen('ps -auxww')
l = output.readlines()


size = len(l)
print "Size =",size
idx = 0



while idx < size:
        print string.find('snmpd',l[idx])
        idx = idx + 1




From bencr at bigpond.com  Thu Mar 29 02:59:16 2001
From: bencr at bigpond.com (Ben)
Date: Thu, 29 Mar 2001 17:59:16 +1000
Subject: Recommendations for Python Books
References: <3ac2582c$1@news.starhub.net.sg>
Message-ID: <xLBw6.5365$45.29190@newsfeeds.bigpond.com>

If you just starting out and want a good free (opensource) book to start
with try "How To Think Like A Computer Scientist In Python" ... you can get
the free download in a few different formats from
...http://www.andamooka.org ...




From aleaxit at yahoo.com  Tue Mar 13 17:13:43 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Tue, 13 Mar 2001 23:13:43 +0100
Subject: Extending Python: rewriting a single method in C
References: <g0wv9vxms0.fsf@scumbag.ecs.soton.ac.uk> <slrn9apn8f.382.jepler@potty.housenet> <g0ofv6xbj4.fsf@scumbag.ecs.soton.ac.uk> <98l1i901k4u@news1.newsguy.com> <g0lmq9xxz0.fsf@scumbag.ecs.soton.ac.uk>
Message-ID: <98m6g20fgv@news1.newsguy.com>

"Jacek Generowicz" <jmg at ecs.soton.ac.uk> wrote in message
news:g0lmq9xxz0.fsf at scumbag.ecs.soton.ac.uk...
    [snip]
> As I understand it, I have to provide a __getattr__ method in my
> python class which will get called when I use PyObject_GetAttr. (If
> this is so then I don't see why the function isn't automatically
> provided; any creative definition---ie differing from the blatantly
> obvious (and hence tedious)---surely only opens the door to
> horrors. Still, mine is not to reason why.)

You need only provide __getattr__ if you have to do something
special -- and when you do need it, it's nice to have the ability.
But attribute-getting works just fine for simple cases without
you writing anything special!


> In Beazley I see that the signature of the PyObject_GetAttr is
>
> PyObject * PyObject_GetAttr( PyObject *o, PyObject *attr_name )
>
> Nowhere can I find an explanation of the variables, so I'm guessing

The online Python/C API reference says of this function: '''Retrieve an
attribute named attr_name from object o. [...] equivalent of the Python
expression "o.attr_name". '''.  Seems clear enough to me -- although,
actually, a closer equivalent is "getattr(o,attr_name)" [a level of
indirectness issue wrt attr_name!-)].  This is in 6.1, "Object Protocol",
of said manual.


> ). But what on earth is attr_name ?  Is it a PyObject* pointing to a
> python string containing the name of the attribute I want ?
>
> How about PyObject_GetAttrString ?  Is it the same as PyObject_GetAttr
> except that I can use a C string to give the name of the attribute ?

Yes, and yes.  Again, the quoted subchapter documents that.


> Assuming that is so, and that these funcitons return a pointer to the
> attribute in question, how do I do anything with it ?  For example,
> assume that I know that PyObject * O points to a python integer, how
> do I convert it to a C integer? or is this The Wrong Thing To Do ?

No it isn't -- see e.g. 7.4.1, same manual: PyInt_Check lets you
check if the object is indeed an integer, PyInt_AS_LONG returns
the C int value for a PyObject* that passes PyInt_Check (or,
PyInt_AsLong coerces-to-integer if needed).


> (Would it be fair to say that the documentaion is, on the whole, less
> that satisfactory ?)

I find it pretty decent, myself.


> > > For example, I tried to write an extension type equivalent to
> > >
> > > class foo:
    [snip]
> > Getting equivalence to this is not going to be easy -- the
> > type/class separation hits hard,
>
> Could you elabore a bit on this, please ? (the bit after --).

Normally, the types you define in C are *not* equivalent to
Python-coded classes -- they can't be inherited from, etc.  To
get closer to equivalence, you need to use a "deep and dark
mystery" in the Python internals known as "the Beaudry hook" --
or use a framework that will do it on your behalf, such as the
Boost Python Library (www.boost.org -- but, be warned, it
needs C++, not just C).


Alex





From claird at starbase.neosoft.com  Thu Mar 15 09:12:55 2001
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 15 Mar 2001 08:12:55 -0600
Subject: Pattern talk (was: Who's minister of propaganda this week?)
References: <mailman.984595938.11127.python-list@python.org> <98otuq0nnf@news1.newsguy.com> <slrn9b0cc2.gsa.neelk@alum.mit.edu> <98q6ro2uoa@news2.newsguy.com>
Message-ID: <617676FE32F5EC92.40A2A304C29DF275.0015EAF449104233@lp.airnews.net>

In article <98q6ro2uoa at news2.newsguy.com>,
Alex Martelli <aleaxit at yahoo.com> wrote:
>"Neelakantan Krishnaswami" <neelk at alum.mit.edu> wrote in message
>news:slrn9b0cc2.gsa.neelk at alum.mit.edu...
>> On Thu, 15 Mar 2001 00:09:30 +0100, Alex Martelli <aleaxit at yahoo.com>
>wrote:
>> >"Steven D. Majewski" <sdm7g at virginia.edu> wrote in message
>> >news:mailman.984595938.11127.python-list at python.org...
>> >    [snip]
>> >> ( OK: not having multiple dispatch in Python means that some manual
>> >> dispatching is sometimes required, but usually this is burried deep
>> >> in some classes __coerce__ method. )
>> >
>> > "Usually" in arithmetic, perhaps.  For fancier stuff, lack of
>> > multimethods sometimes cramps (not that it cramps any less in C++,
>> > mind you), though some patterns (such as 'dynamic visitor') can help
>> > a bit in certain (reasonably frequent) special cases.
>>
>> What's a dynamic visitor? Is it a dictionary keyed on class? That's
>> what I usually do when emulating multimethods in Python.
>
>Sorry, the pattern-name is _Acyclic_ Visitor -- it's fully
>described by its author, Robert Martin, in the PDF at URL
>http://www.objectmentor.com/publications/acv.pdf for example.
			.
			.
			.
Coincidentally, the same Uncle Bob originated the quote which
launched this thread, and is available for open, public inter-
view through the end of business today, Thursday (USA time), at
<URL: http://forums.itworld.com/webx?230 at 96.WvrdaBXXgrN^0 at .ee6eed4 >
-- 

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


From konrad.koller at worldonline.de  Fri Mar 23 04:11:57 2001
From: konrad.koller at worldonline.de (Konrad Koller)
Date: Fri, 23 Mar 2001 09:11:57 GMT
Subject: Text printing under Windows
References: <slrn9bk130.10l.ego@macondo.visconti.home>
Message-ID: <3abb122e.5577975@news.worldonline.de>

On 22 Mar 2001 14:00:29 GMT, ego at macondo.visconti.home (Gabriele
Favalessa) wrote:

>Hi,
>
>how do I print using a dot matrix printer in text mode under Windows
>98?  I have both a local and a remote printer.
>
>I use python 2.0 and wxPython 2.2.5.  The latter seems to support
>printing only in bitmap mode.
>
>My knowledge of win32 APIs is poor, I'm looking for a simple
>open/write/close API.  Something similar to:
>
>   printer = os.popen ("lpr -Pprinter1")
>   printer.write ("Hello world!\n")
>   printer.close ()
>
>that works fine under unix.
>
>Thanks in advance.
>
>   Gabriele
Hello Gabriele,
I am using a laser printer instead of a matrix printer, but (under
Windows 98) I do it like this:

stdout=sys.stdout
sys.stdout=open("lpt1","w")
print "what you want to print\f"
sys.stdout=stdout

Yours Konrad


From aboufer at atlsci.com  Wed Mar 28 16:56:54 2001
From: aboufer at atlsci.com (aboufer at atlsci.com)
Date: Wed, 28 Mar 2001 16:56:54 -0500
Subject: Launching apps on Win32 from Python
Message-ID: <3AC25E26.9446DAA8@atlsci.com>

Greetings,

    I am writing an app in which I need to spawn a process on win32.
I use spawnv for this purpose. However, I need a way to determine
the full path of the task (say iexplore.exe) that is spawned.

Is there a way to do this ? If so, I'll appreciate a sample code.

Thanks



From sholden at holdenweb.com  Mon Mar  5 21:14:10 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Mon, 5 Mar 2001 21:14:10 -0500
Subject: I come to praise .join, not to bury it...
References: <mailman.983756766.27347.python-list@python.org> <97vnu3076q@news2.newsguy.com> <slrn9a89dk.2rm.hzhu@rocket.knowledgetrack.com>
Message-ID: <DNXo6.24266$1D5.975603@e420r-atl1.usenetserver.com>

"Huaiyu Zhu" <hzhu at users.sourceforge.net> wrote in message
news:slrn9a89dk.2rm.hzhu at rocket.knowledgetrack.com...
> It is all nice and well to say that sep.join(list) is good for
polymorphism,
> except that there are several practical annoyances:
>
> 1. The word join can be either transitive or nontransitive:
>    sep joins list.
>    list joins with sep.
>
It can also be a noun. This is no more help than your assertion.

>    Might sep.glue(join) be clearer?  Well, compatibility with str.join?
>
> 2. Can we really join all kinds of lists?
>
No. I was somewhat startled to see that a UserList of integers could not be
given as an argument to join(). Relief of a kind arrived when I realised
that the same was true of lists themselves.

>>> import UserList
>>> reallist = [1, 2, 3, 4, 5]
>>> userlist = UserList.UserList(reallist)
>>> ":".join(reallist)
Traceback (innermost last):
  File "<interactive input>", line 1, in ?
TypeError: sequence item 0: expected string, int found
>>> ":".join(userlist)
Traceback (innermost last):
  File "<interactive input>", line 1, in ?
TypeError: sequence item 0: expected string, int found
>>>

It would seem reasonable to expect the join() method to try and coerce
things to lists, but maybe I'm not a reliable guide to what's reasonable.

[ ... ]
>
> 3. What's the inverse of join?  Is it split?
>
> >>> b = [str(x) for x in a]
> >>> c = "".join(b)
> >>> c
> '0123456789'
> >>> d = "".split(c)
> >>> d
> ['']

Bzzt. This is along the lines of saying that since 0*x == o*y then x and y
must be equal. You really shouldn't argue from degenerate cases. The null
string is not a practical separator (for split()). Clearly it works for
join().

>
> Oops! The other way round.  Try again
> >>> d = c.split("")
> Traceback (most recent call last):
>   File "<stdin>", line 1, in ?
> ValueError: empty separator
>
Aha! So you''believe the interpreter, then?

> What?  Oh, in this particular case, the inverse is
> >>> d = list(c)
> >>> d
> ['0', '1', '2', '3', '4', '5', '6', '7', '8', '9']
>
> For other cases this is OK:
> >>> c = ".".join(b)
> >>> c
> '0.1.2.3.4.5.6.7.8.9'
> >>> d = c.split(".")
> >>> d
> ['0', '1', '2', '3', '4', '5', '6', '7', '8', '9']
>
Quite.

> 4. So can we just change the order of split?  No.  Consider
>
> file.readlines()
> file.read().splitlines()
> file.read().split("\n")
>
> More abstractly, when we split something, we have
>    whole - sep ==> components.
> But when we join things together, we do it like
>    glue + components ==> whole,
> rather then the "more natural"
>    components + glue => whole,
> where "more natural" is defined as "most likely to be guessed by
uninitiated
> minds".
>
You either have to understand the basis of a programming language's
operation, or copy other people's code in the hopes that it's well formed
and your use of it does not break it. "More natural" is an unprovable
assertion without usability testing, something which did feature in Python's
early development.

Nowadays, of course, everyone simply bitches like mad on c.l.py, and Guido
ignores our rantings and introduces execrescences such as "print >>" (don't
get me started :-)

> Of course, all these point do not detract from the fact that from
> implementer's point of view, sep.join(list) is the most natural.
>
There's the rub. We could all take the source and be our own implementer.
Few of us have the time and the stupidity^H^H^H^H^H^H^H^H^Hcourage.

> So what am I trying to say here?  I'd say that there's not much point in
> arguing, or even explaining which way is better.  What would help is to
> make things more symmetrical/smooth if possible.
>
> Huaiyu

Erm, use string.join() until it is taken away.

regards
 Steve





From matias.guijarro at utbm.fr  Thu Mar 29 06:04:44 2001
From: matias.guijarro at utbm.fr (Matias GUIJARRO)
Date: Thu, 29 Mar 2001 13:04:44 +0200
Subject: asyncore
Message-ID: <3AC316CB.597E85F3@utbm.fr>

Hello group :-)

I'm using the asyncore library in order to make a simple network card
game.

I'm relatively new to sockets programming, however I think the asyncore
library provides a very efficient way to what I'd like to obtain.

Here are my questions/problems :
- The server must be always listenning for new connections, so I call
the
asyncore.loop() method, but it seems to be "blocking" : the code after
isn't
executed. How can I make an "asyncronous" listenner ?
- Due to this behaviour, I can't develop my app under Windows because
every time I test it, the OS "hangs up", listenning for connections...
(I use IDLE and I directly execute the code with CTRL-F5).
- I don't really want to make a server very different from a client ;
before
the game one of the players says he's the server, then every client
connect
to him : that's what I expect.  What is the best solution for doing so ?

- Moreover, I build my GUI with Tkinter (I'm learning Python, so I
thought
Tkinter was more accessible, even if it's surely not the best), and
there is
also a loop for the Tkinter events (root.mainloop()), how can I have the
two
loops (asyncore.loop & this one) working as background processes ?
Using threads (thread module) in Windows isn't a good solution, it's too
slow
and sometimes the OS hangs up too !

If someone can help me about these questions, I will appreciate it a
lot.

Thanks in advance.



From w.t.stolker at getronics.com  Fri Mar 30 02:43:09 2001
From: w.t.stolker at getronics.com (Stolker, Wim)
Date: Fri, 30 Mar 2001 09:43:09 +0200
Subject: Converting an integer base 10 to a binary number
Message-ID: <F2E8F5623E44D41196150008C70D2F9601847F1D@getasdnt3.gen.getronics.nl>

I don't know of a built-in, or 'batteries included' function
but:

def tobase(b,n,result=''):
    if n == 0:
        if result:
            return result
        else:
            return '0'
    else:
        return tobase(b,n/b,str(n%b)+result)

or, more safe and complete:

def tobase(base,number):
    global tb
    def tb(b,n,result=''):
        if n == 0: return result
        else: return tb(b,n/b,str(n%b)+result)

    if type(base) != type(1):
        raise TypeError, 'invalid base for tobase()'
    if base <= 0:
        raise ValueError, 'invalid base for tobase(): %s' % base
    if type(number) != type(1) and type(number) != type(1L):
        raise TypeError, 'tobase() of non-integer'
    if number == 0:
        return '0'
    if number > 0:
        return tb(base, number)
    if number < 0:
        return '-' + tb(base, -1*number)

bin = lambda n: tobase(2,n)

does what you want.

Wim


-----Original Message-----
From: Peter St?hr [mailto:peter.stoehr at fh-hof.de]
Sent: Wednesday, March 28, 2001 3:35 PM
To: python-list at python.org
Subject: Converting an integer base 10 to a binary number


Hi out there,

I know the int(x[,radix]) function that converts a string or a number
(for a give radix) to an integer. Is there a built-in function that
provides a conversion the other way round?
Something like
    bin(32) = "10000"

Thanks in advance
    Peter
-- 
Fachhochschule Hof
Prof. Dr. Peter St?hr				Tel: 09281 - 409 485
Alfons-Goppel-Platz 1				Fax: 09281 - 409 477
95028 Hof / Germany



From BrianQ at ActiveState.com  Wed Mar 21 19:19:27 2001
From: BrianQ at ActiveState.com (Brian Quinlan)
Date: Wed, 21 Mar 2001 16:19:27 -0800
Subject: socket programming
In-Reply-To: <99bb0g0nce@enews3.newsguy.com>
Message-ID: <000b01c0b265$c28c9cd0$b503a8c0@activestate.ca>

Hi Kevin,

The bufsize parameter in recv is the maximim size of the buffer that recv
will return, not the minimum. recv will normally return any available data
and will only block if there is none.

I assume that the only way that you can determine if the message has ended
is by detecting that the socket has been closed? You could see if setsockopt
has something to help you or you could try something like this (as usual,
not tested in any way):

message = ''
try:
	message += mysocket.recv( 1024 )
except socket.error, e:
	# e.args could be a string too, might want to check for that
	if e.args[0] != errno.ECONNRESET
		raise e

--
Brian Quinlan
Python and XSLT Developer
BrianQ at ActiveState.com

Use the power of the dragon for your programming needs.
Download Komodo, our Mozilla-based, cross-platform IDE.
      http://www.ActiveState.com/Komodo


-----Original Message-----
From: python-list-admin at python.org
[mailto:python-list-admin at python.org]On Behalf Of Kevin X Lin
Sent: Wednesday, March 21, 2001 2:47 PM
To: python-list at python.org
Subject: socket programming


Hi there, I have the following problem with socket programming:

I'm trying to setup a tcp client to talk to a tcp server sitting at a
certain port. When I issue :

mysocket.setblocking(1)
mysocket.send('something')
mysocket.recv(1024)

I was expecting I can receive all the data sent by server. Unfortunately,
this is not necessary the case. Sometimes I can't get all of data, worse,
the application protocal doesn't define end tag of the data, which prevent
me from doing a loop until I receive all data.

But if I do this:

mysocket.setblocking(1)
mysocket.send('something')
sleep(0.1)
mysocket.recv(1024)

I can get all data, yet I believe this is not a good solution which might
casusing other problems in the future.

Any suggestion?

Thanks.

Kevin




--
http://mail.python.org/mailman/listinfo/python-list




From fuess at att.net  Sat Mar 31 10:31:28 2001
From: fuess at att.net (David Fuess)
Date: Sat, 31 Mar 2001 15:31:28 GMT
Subject: Problem stopping Tkinter apps
Message-ID: <dktbct8ul4jmbnlh4h6acplkauofdh5tup@4ax.com>

I am building Tkinter applications on W2K using ActivePython2.0 and
Tcl 8.3.2 with Pmw wigets. The problem is that if I try to exit the
program using a button handler I get an error. I have tried both
'sys.exit(0)' and 'root.quit()' in the button handler. In both cases I
often get the error:

pythonw.exe - Application Error, The instruction at "0x01744727"
referenced memory at "0x00000000". The memory could not be "read".

This never happens if I close the application using the X in the
window decorations.

Dave


From parkw at better.net  Wed Mar 21 17:48:44 2001
From: parkw at better.net (William Park)
Date: Wed, 21 Mar 2001 17:48:44 -0500
Subject: Simple regex that has me banging my head against a wall
In-Reply-To: <20010321143144.K24341@trufflehunter.avalongroup.net>; from tjg@exceptionalminds.com on Wed, Mar 21, 2001 at 02:31:44PM -0800
References: <20010321143144.K24341@trufflehunter.avalongroup.net>
Message-ID: <20010321174844.A2506@better.net>

On Wed, Mar 21, 2001 at 02:31:44PM -0800, Timothy Grant wrote:
> Hi again.
> 
> I actually thought I had this regex thing down pretty well, but
> I'm getting beat up by what I thought should be a simple one.
> 
> >>> x = '$1$asdfjke$lkjasdkjfsdasdfj'
> >>> re.sub(r'\$', r'\\\$', x)
> '\\\\$1\\\\$asdfjke\\\\$lkjasdkjfsdasdfj'
> 
> All I want to do is substitute \$ for $.

Solution:
    >>> re.sub('\$', r'\$', x)
    '\\$1\\$asdfjke\\$lkjasdkjfsdasdfj'

Note: 2nd argument is just string, not pattern.

:wq --William Park, Open Geometry Consulting, Linux/Python, 8 CPUs.



From donotreply at interbulletin.bogus  Thu Mar  8 11:54:34 2001
From: donotreply at interbulletin.bogus (Rob Earl)
Date: Thu, 08 Mar 2001 16:54:34 +0000
Subject: mxODBC on Windows 2000?
Message-ID: <3AA7B94A.23D5C849@interbulletin.com>

Has anyone been able to get the mxODBC module for Python to work properly with ActivePython on Windows 2000?  The reason I ask is because every time I try to execute a SQL command, the CPU usage goes up to 100% on my machine and the python shell is locked up.  Can anyone help?  Thanks.
_______________________________________________
Submitted via WebNewsReader of http://www.interbulletin.com



From tim.one at home.com  Sun Mar 11 23:22:03 2001
From: tim.one at home.com (Tim Peters)
Date: Sun, 11 Mar 2001 23:22:03 -0500
Subject: random number between -maxint and +maxint
In-Reply-To: <K2Sq6.360684$Pm2.5406628@news20.bellglobal.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCKEPIJEAA.tim.one@home.com>

[Dan Parisien]
> I can't figure out a good way to do it, does anyone know a better solution
> than
>
> from whrandom import whrandom
> rand = whrandom()
> max = 2**30 + (2**30-1) #one less than 2**31
> randomnumber = rand.randrange(0, max) - rand.randrange(0, max)

The last line is clearly biased; e.g., consider what happens if max is 2:

    LHS  RHS  result
      0    0       0
      0    1      -1
      1    0       1
      1    1       0

That is, not all results are equally likely.  Try instead:

    import sys, random, math
    nvalues = 2.0 * sys.maxint + 1
    raw = random.random() * nvalues - sys.maxint
    randomnumber = int(math.floor(raw))

If maxint is 2 there, then nvalues is 5., random.random()*nvalues is uniform
across [0., 5.); that minus maxint is uniform in [-2., 3.); so the floor is
uniform in [-2., -1., 0., 1., 2.] == -maxint thru maxint inclusive.

Somewhat simpler, if you're assuming you're on a 32-bit box, is:

    (random.randrange(65536) << 16) | random.randrange(65536)

If you're afraid of getting -sys.maxint-1, stick that in a loop and keep
going until you get something else.

even-simpler-hardwire-the-result-to-6<wink>-ly y'rs  - tim




From mark.limber at dvinc.com  Fri Mar 16 11:35:48 2001
From: mark.limber at dvinc.com (Mark Limber)
Date: Fri, 16 Mar 2001 09:35:48 -0700
Subject: redefining stdout
Message-ID: <3AB240E4.1EB6E144@dvinc.com>

Hi All,

    I appologize if this is terribly naive...

    I'm embedding Python in an application and would like a
way to redefine stdout so that it writes to my custom class.

    It seems that it would be as simple as:
-------------
File : t.py

import sys

class myPrint:
  def write(self,s):
    print >> sys.__stdout__, "Here it is: %s!" %s

  def writelines(self,lines):
    for i in lines:
      self.write(i)
------------

where, of course, the actual write method would call my
own write object.

However, when I try:

>>> import sys
>>> import t # short file names = less typing!
>>> sys.stdout = t.myPrint
>>> print "hey"
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
TypeError: unbound method must be called with class instance 1st
argument


Help?

Thanks,
mark



From tmhazel at mediaone.net  Sun Mar  4 19:31:13 2001
From: tmhazel at mediaone.net (Thomas Hazel)
Date: Mon, 05 Mar 2001 00:31:13 GMT
Subject: [ANNOUNCE] txObject ATK 1.3.5 Python released
Message-ID: <l7Bo6.9589$TL4.2459032@typhoon.ne.mediaone.net>

Announcing the release of txObject ATK Python Modules 1.3.5.
It provides a few more Python interfaces to the txObject Application Toolkit
and Libraries. In addition to previous releases containing only
Non-Preemptive
Python thread modules, this release now provides Preemptive modules as well.
The following are a list of some modules provided in this release.

Non-Preemptive:
   - txThread, txEvent, txLock, txEventQueue

 Preemptive:
   - txNativeThread, txNativeEvent, txNativeLock, txNativeEventQueue

What is txObject ATK? txObject ATK is a collection of platform-independent
C++ Class libraries (GPL) that provides a five-layer framework containing an
Object Library, IO and Timers, Object-Oriented Threads, Inter-Process
Communication, and Distributed Object Communication. txObject ATK can
be found at either of the following sites:

 Official Site
 http://www.txobject.org

 SourceForge (development & bug tracking)
 http://txobject.sourceforge.net

 Enjoy,

 --Thomas Hazel ( thazel at txobject.org )


 <P><A HREF="http://txobject.sourceforge.net">txObject ATK 1.3.5</A> -
Python interfaces to the txObject Application Toolkit and Libraries.
(04-Mar-01)





From insanc at cc.gatech.edu  Fri Mar 23 08:44:08 2001
From: insanc at cc.gatech.edu (Holland King)
Date: 23 Mar 2001 13:44:08 GMT
Subject: email address verification
References: <99d5qi$95l$1@solaria.cc.gatech.edu> <NFzu6.9043$G72.216162@e420r-atl1.usenetserver.com>
Message-ID: <99fjv8$bhf$1@solaria.cc.gatech.edu>

Steve Holden <sholden at holdenweb.com> wrote:
: The classic way would be to send it mail containing a link, which begins:
: "To remove yourself from this mailing list, all you have to do is click
: here". Naturally the URL clicked on adds the relevant email address to at
: least several hundred salable lists, including "gullible individuals who
: believe what they read in their inbox".

: do-i-seem-a-bit-cynical-today-or-what?-ly y'rs  - steve

despite the way it sounds i am not trying to set up a mass mailer program
or trying to spam people. i am trying to set my departments system so
that when an account expires it will parse the .forward file and use 
addresses in there to make an alias. however, we don't like the idea of 
sending mail to a list or to a non-existant address for obvious reason
(especially consider how large we are).


-- 
---
Joseph Holland King  | "God whispers to us in our pleasures, speaks in our
                     |  conscience, but shouts in our pains: it is His
                     |  megaphone to rouse a deaf world." C. S. Lewis


From danielk at aracnet.com  Sat Mar  3 13:18:33 2001
From: danielk at aracnet.com (Daniel Klein)
Date: Sat, 03 Mar 2001 10:18:33 -0800
Subject: IDLE: Refreshing Class Browser
Message-ID: <v5d2atc86rahttjsjhu8nvhdg3os07kslp@4ax.com>

This is not a big deal but I was wondering if there is there a shortcut key to
refresh the Class Browser? Presently you have to close and reopen to see the
updates.

Daniel Klein


From tjg at exceptionalminds.com  Wed Mar 21 23:56:40 2001
From: tjg at exceptionalminds.com (Timothy Grant)
Date: Wed, 21 Mar 2001 20:56:40 -0800
Subject: Please rate these Python books
In-Reply-To: <20010322.1445209@mis.configured.host>; from tm@kernelconsult.com on Thu, Mar 22, 2001 at 01:44:52AM +0000
References: <20010322.1445209@mis.configured.host>
Message-ID: <20010321205640.A29172@trufflehunter.avalongroup.net>

On Thu, Mar 22, 2001 at 01:44:52AM +0000, TM wrote:
> 3/ Python Essential Reference / David Beazley / New Riders Publishing 

The title says it all, this is *definitely* an Essential book.
Short, Informative, Concise. An alltogether excellent book.
-- 
Stand Fast,
    tjg.

Timothy Grant                         tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.         <><       (503) 246-3630
>>>>>>>>>>>>>Linux, because rebooting is *NOT* normal<<<<<<<<<
>>>>This machine was last rebooted:  64 days  9:10 hours ago<<



From root at [127.0.0.1]  Sat Mar 31 16:12:32 2001
From: root at [127.0.0.1] (deadmeat)
Date: Sat, 31 Mar 2001 21:12:32 GMT
Subject: list vs tuple
References: <hGlw6.782$p5.3210@news1.rivrw1.nsw.optushome.com.au><3AC20CD3.265A2D94@alcyone.com><ru69cto61jtmrqsf809e38hb3pbm2lhi7g@4ax.com><bC1x6.4319$4N4.905946@newsc.telia.net><71q9ct4le63anefmiajj5cbh3mg04qu1ga@4ax.com><mailman.985991202.5578.python-list@python.org><7Bfx6.989$p5.3791@news1.rivrw1.nsw.optushome.com.au><Ukhx6.4421$4N4.950603@newsc.telia.net><mailman.986049281.32365.python-list@python.org><uGqx6.1052$p5.3997@news1.rivrw1.nsw.optushome.com.au> <mailman.986072259.10548.python-list@python.org>
Message-ID: <4Lrx6.1055$p5.4170@news1.rivrw1.nsw.optushome.com.au>

> But that is not what is happening!

What is actually happening is not my point, the end result that the
programmer sees is, and that is not consistant with what is typed.




From benjamin.altman at noaa.gov  Wed Mar 21 13:02:29 2001
From: benjamin.altman at noaa.gov (Benjamin.Altman)
Date: Wed, 21 Mar 2001 13:02:29 -0500
Subject: learning python...
Message-ID: <3AB8ECB5.2197E940@noaa.gov>

Hello.  From the perspective of someone who already has programming
experience, would anyone know if O'reillys "Learning Python" book would
be better than say Prentice Hall's "Core Python Programming"?  Or would
it be better to go straight into something like "Programming Python"?

Thanks,
Ben
-------------- next part --------------
A non-text attachment was scrubbed...
Name: benjamin.altman.vcf
Type: text/x-vcard
Size: 413 bytes
Desc: Card for Benjamin.Altman
URL: <http://mail.python.org/pipermail/python-list/attachments/20010321/e6f519d1/attachment.vcf>

From mda at idatar.com  Sat Mar 31 16:56:32 2001
From: mda at idatar.com (David Allen)
Date: Sat, 31 Mar 2001 21:56:32 GMT
Subject: Problem stopping Tkinter apps
References: <dktbct8ul4jmbnlh4h6acplkauofdh5tup@4ax.com>
Message-ID: <kosx6.1123$NU3.1087875@typhoon2.ba-dsg.net>

In article <dktbct8ul4jmbnlh4h6acplkauofdh5tup at 4ax.com>, "David Fuess"
<fuess at att.net> wrote:

> I am building Tkinter applications on W2K using ActivePython2.0 and Tcl
> 8.3.2 with Pmw wigets. The problem is that if I try to exit the program
> using a button handler I get an error. I have tried both
> 'sys.exit(0)' and 'root.quit()' in the button handler. In both cases I
> often get the error:
> 
> pythonw.exe - Application Error, The instruction at "0x01744727" referenced
> memory at "0x00000000". The memory could not be "read".
> 
> This never happens if I close the application using the X in the window
> decorations.
> 
> Dave

Much easier is simply this:

toplevelwidget.destroy()

-- 
David Allen
http://opop.nols.com/
----------------------------------------
It's that time of year again, Defcon has descended upon us. Time to venture
to SYN City to see the lights, meet strange new people, and create new
blurs we later dub memories.
	-- Attrition.org


From bsb at winnegan.de  Wed Mar 21 12:26:26 2001
From: bsb at winnegan.de (Siggy Brentrup)
Date: 21 Mar 2001 18:26:26 +0100
Subject: Stupid String Question
In-Reply-To: <3AB8327C.493C70A5@direct.ca>
References: <3AB8327C.493C70A5@direct.ca>
Message-ID: <873dc79got.fsf@winnegan.de>

Brian & Colleen <greybria at direct.ca> writes:

> 1.  (*) text/plain          ( ) text/html           

Please don't send the same text twice in one message.

> I know I've seen this somewhere but dejanews has not yielded the
> answer with a variety of searches.

> What is the pretty way to generate a string of 30 asterisks. I'm
> sure it's not:

> def fill(c,n):
> ??? s=''
> ??? for i in range(n):
> ??????? s=s+c
> ??? return s
> 
> a=fill('*',30)

a = '*'*30

HIH
  Siggy
-- 
Siggy Brentrup - bsb at winnegan.de - http://www.winnegan.de/
****** ceterum censeo javascriptum esse restrictam *******




From dnew at san.rr.com  Fri Mar 23 14:47:02 2001
From: dnew at san.rr.com (Darren New)
Date: Fri, 23 Mar 2001 19:47:02 GMT
Subject: New PEP: The directive statement
References: <200103201042.LAA20248@pandora.informatik.hu-berlin.de> <mailman.985107140.4586.python-list@python.org> <9988ie02k7u@drn.newsguy.com> <ulmpzjrpb.fsf@cs.uu.nl> <mailman.985222709.15240.python-list@python.org> <wzr8zpwn4e.fsf@sunshine.cs.uu.nl> <3d1yrp1osi.fsf@ute.cnri.reston.va.us>
Message-ID: <3ABBA837.DEB8AB1B@san.rr.com>

Andrew Kuchling wrote:
> It's really a documentation issue; people who release software should
> say "I've tested this with Python 2.0", and people with 1.5.2 or 2.1
> will have to look out, and perhaps provide patches to fix problems.

What is the Python mechanism for ensuring a particular (or minimal) version
of Python is being used? Is there one?  I'm talking about something like
Tcl's [package require Tcl 8.3] or some such.

-- 
Darren New / Senior MTS & Free Radical / Invisible Worlds Inc.
San Diego, CA, USA (PST).  Cryptokeys on demand.
A million monkeys in a room with a million typewriters 
              will only yield half a million pregnant monkeys.


From aleaxit at yahoo.com  Wed Mar  7 02:56:28 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Wed, 7 Mar 2001 08:56:28 +0100
Subject: The in opperator -- How do I use it?
References: <983945437.96921@atlas.uniserve.ca> <dTkp6.13293$hn5.1934677@news1.rdc1.mb.home.com>
Message-ID: <984pjd0rsv@news1.newsguy.com>

"Brad Bollenbach" <bbollenbach at homenospam.com> wrote in message
news:dTkp6.13293$hn5.1934677 at news1.rdc1.mb.home.com...
> Think of the "in" operator as the list membership operator...because
that's
> what it is. :)

Slightly more general than that -- it's a membership operator
(not just in lists, but, rather, in all kinds of sequences,
user-defined instance objects that define a __contains__
method, and maybe one day in mappings as well, though I think
that addition was retracted for Python 2.1).


For example, module string contains an attribute, named
'letters', which is a string made up of all the letters.
You can test any single character for membership in that
string, using the in operator:

>>> import string
>>> if 'a' in string.letters: print 'a is a letter'
...
a is a letter
>>> if '3' not in string.letters: print '3 is not a letter'
...
3 is not a letter
>>>


Of course, that may not be optimal, since the .isalpha
string method does the same job, probably faster:

>>> 'a'.isalpha()
1
>>> '3'.isalpha()
0
>>>

...but it may come in handy for testing membership into
any short runtime-computed string -- you don't have to
make it into a list to be able to apply the in operator.


Alex





From bjennings at visionael.com  Mon Mar 19 17:58:08 2001
From: bjennings at visionael.com (Brad Jennings)
Date: Mon, 19 Mar 2001 16:58:08 -0600
Subject: (no subject)
Message-ID: <304D429618A8D41187D50008C7F430740B75CD@mail.visionael.com>

I want to execute an executable binary as thought through the Command-Line.
Python provides some powerful methods in the os package that use very quirky
syntax.  The two methods I've used are "startfile()" and "execv()". Neither
have performed the action I  need.  

I'm just trying to execute "ss.exe" and pass it some args but the method
syntax is laborious and prone to mistakes.  Do anyone know of  a better way
to call an executable from a python script without using a function to do
it?  

Brad



From mark at chaos.x-philes.com  Thu Mar 22 09:30:32 2001
From: mark at chaos.x-philes.com (Mark Carroll)
Date: Thu, 22 Mar 2001 14:30:32 GMT
Subject: Standalone executables
References: <998igr$4al$1@niagara.nonet>
Message-ID: <99d29u$mf$1@niagara.nonet>

Thanks very much for the helpful responses! I'll look into things some
more. (-:

-- Mark


From orlov at diasoft.ru  Thu Mar 22 09:41:09 2001
From: orlov at diasoft.ru (Oleg Orlov)
Date: Thu, 22 Mar 2001 17:41:09 +0300
Subject: bug in threading
Message-ID: <99d2oc$25um$1@gavrilo.mtu.ru>

My problem is: Condition.wait() is not protected against
KeyboardInterrupt. Fix is below.

Regards,
Oleg.

-----------------------------
#python 2.0, threading.py:

...
class _Condition(_Verbose):
...
    def wait(self, timeout=None):
...
        saved_state = self._release_save()
# line to be inserted
        try:
...
# line to be inserted
        finally:
            self._acquire_restore(saved_state)






From carifio.nospam at nospam.usys.com  Wed Mar 14 17:44:31 2001
From: carifio.nospam at nospam.usys.com (Mike Carifio)
Date: Wed, 14 Mar 2001 22:44:31 GMT
Subject: windows script host and python
References: <%tKr6.2570$pX.26174@news1.onlynews.com> <98oq3r$1e0$1@troll.powertech.no>
Message-ID: <jvSr6.34$cj1.4417@news1.onlynews.com>

> Just curious but what do plan to acheive by using python with wsh that you
> couldn't do using normal .py files?

I was testing Microsoft's assertion that "other" scripting engines worked.
Trust but verify.




From ben.held at staarinc.com  Mon Mar 19 08:55:14 2001
From: ben.held at staarinc.com (Ben Held)
Date: Mon, 19 Mar 2001 07:55:14 -0600
Subject: Executing a "dir" using PyRun_String??? URGENT!!!
Message-ID: <3ab61012$0$18892$272ea4a1@news.execpc.com>

I am trying to execute a dir command from C using PyRun_String where
pcObjectString is a module name imported using "import <module>".  The
problem is that PyRun_String always returns a NULL.  I have using code
similiar to this for calling other functions and they work, but I cannot get
this to work.  Any ideas?

Ben

  PyObject* pModule = PyImport_AddModule("__builtin__");
 PyObject* pDict = PyModule_GetDict(pModule);

  char pcCommand[1000];
  strcpy(pcCommand, "pSTADir = dir(\"");
  strcat(pcCommand, pcObjectString);
  strcat(pcCommand, "\")");
  PyObject* pyRunObject = 0;
  PyObject* pDir = 0;
  if (pyRunObject = PyRun_String(pcCommand, eval_input, pDict, pDict)) {
   pDir = PyDict_GetItemString(pDict, "pSTADir");
   if (!pDir) return false;
   int listsize = PyList_Size(pDir);
   PyObject* pItem = 0;
   CString strEntry;
   for (int i = 0; i < listsize; i++) {
    pItem = PyList_GetItem(pDir, i);
    strEntry = CString(PyString_AsString(pItem));
    strEntry.Remove('\r');
    if (strEntry[0] == '_') continue;
    dirEntries += strEntry + " ";
   }
   if (pDir) Py_DECREF(pDir);
  }
  if (pyRunObject) Py_DECREF(pyRunObject);





From stadel at excite.com  Sun Mar  4 20:55:41 2001
From: stadel at excite.com (stadel)
Date: Sun, 04 Mar 2001 18:55:41 -0700
Subject: Name of the Script
Message-ID: <20010304.185354.2114937732.8277@localhost.localdomain>

What is the method for returning the file name of the running script?

Thanks in advance.
Stadel


From emile at fenx.com  Tue Mar 27 09:32:46 2001
From: emile at fenx.com (Emile van Sebille)
Date: Tue, 27 Mar 2001 06:32:46 -0800
Subject: Combining Lists
References: <99q887$hl9$1@ctb-nnrp2.saix.net>
Message-ID: <99q8h8$299ne$1@ID-11957.news.dfncis.de>

list3 = list1 + list2

--

Emile van Sebille
emile at fenx.com

---------
"Mix" <mix77 at usa.net> wrote in message
news:99q887$hl9$1 at ctb-nnrp2.saix.net...
> How does one take two or more lists and combined them?
> For example, suppose one has the following 2 lists :
>  list1=['1','2']
>   and
>  list2=['3','9,'15']
>
> Now how  does one get the following list
>  list3=['1','2','3','9','15']
> from a combing the two lists (i.e list1 and list2)?
>
> Thanks.
> Mix
>
>




From bbollenbach at homenospam.com  Tue Mar  6 13:12:50 2001
From: bbollenbach at homenospam.com (Brad Bollenbach)
Date: Tue, 06 Mar 2001 18:12:50 GMT
Subject: Can Python replace Visual Basic? Should it?...and generalities :)
References: <5DZo6.11126$hn5.1578410@news1.rdc1.mb.home.com>
Message-ID: <CM9p6.12769$hn5.1803901@news1.rdc1.mb.home.com>

Unfortunately this thread is leading to exactly what I've seen in the past:
general answers, often by people that haven't even used much of VB or even
Windows. :)

I was really hoping someone would come along and say something like:

    "Yes, our company used to use VB for our main product, a universal
scheduler program with a clickable ledger. We saw the benefits of Python and
rewrote the whole thing in it, and our company have saved more time and
money than ever before, and our customers are thanking us for it."

Unfortunately, maybe that time won't come.


"Brad Bollenbach" <bbollenbach at homenospam.com> wrote in message
news:5DZo6.11126$hn5.1578410 at news1.rdc1.mb.home.com...
> Okay, sure this has been asked again and again, but I've never really seen
a
> clear answer to this question:
>
>     Can Python replace Visual Basic? Should it?
>
> As far as I can see, they both seem close to equal in their not-so-simple
> ways of distributing apps to end-users, which VB programmers have seem to
> overcome to some degree and so Python programmers should and probably do
> likewise. What I want is a programming language that can make Windows fun
> (like C or Perl does Unix), without having to shell out the big bucks for
> full installs of VB or VC++. Though I don't believe in using price as an
> important criteria in deciding what language I'll like (unless of course,
> it's Powerbuilder, which is over $4000 :), I still have to admit that I
like
> what Python is about, and the benefits of an Open Source language are
> obvious. As I work with Powerbuilder in my day job, I've also realized
that
> "languages" that are based on an IDE are not really my thing (I hate
having
> a 40,000 foot view of what I'm doing).
>
> But really, very specifically, were YOU a former Visual Basic (or
> Powerbuilder, or Delphi) programmer that is now using Python to write the
> complex GUI's VB used to handle for you? I'm an on again, off again,
student
> of Python, who really wants to give this a go, if there's a good chance
that
> I'll get the return on my investment: a programming tool that can take
over
> where VB, Powerbuilder, and Delphi used to play.
>
> Again, I'm much more interested in the real world experience of those who
> used to use the big visual development tools in Windows like VB, Delphi or
> Powerbuilder rather than theoretical answers.
>
> Thanks!
>
> Brad Bollenbach
>
>




From aahz at panix.com  Sun Mar 11 11:12:29 2001
From: aahz at panix.com (Aahz Maruch)
Date: 11 Mar 2001 08:12:29 -0800
Subject: any interest in type-scoped static constants?
References: <3AAB9578.4C8754E4@fxtech.com>
Message-ID: <98g85d$cih$1@panix6.panix.com>

In article <3AAB9578.4C8754E4 at fxtech.com>,
Paul Miller  <paul at fxtech.com> wrote:
>
>If I write an extension type that mirrors the functionality of a C++
>class, and the C++ has static constants within its scope, I can do this
>in C++:
>
>	Point p = Point::ZERO;
>
>Where ZERO is defined as:
>
>	class Point
>	{
>		static Point ZERO = Point(0, 0);
>	};
>
>If I create an extension type for Point, I can't do this:
>
>	p = Point.ZERO

Hmmmm....  The problem is that in Python, you've got a recursive
structure definition -- is that exact syntax really legal in C++?  In
Python itself, the way I'd do this is to create a module Point that
contains the Point class, and following the class definition create the
ZERO object.  Then you refer to it as Point.ZERO and create new points
with Point.Point().
-- 
                      --- Aahz (Copyright 2001 by aahz at pobox.com)

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

"Perhaps God rewards martyrs, but life seldom does..." --Ulrika O'Brien


From tjg at exceptionalminds.com  Wed Mar 21 15:05:03 2001
From: tjg at exceptionalminds.com (Timothy Grant)
Date: Wed, 21 Mar 2001 12:05:03 -0800
Subject: learning python...
In-Reply-To: <m77u6.56$Hp.3284@typhoon.aracnet.com>; from DanielK@jBASE.com on Wed, Mar 21, 2001 at 11:17:19AM -0800
References: <3AB8ECB5.2197E940@noaa.gov> <m77u6.56$Hp.3284@typhoon.aracnet.com>
Message-ID: <20010321120503.B24341@trufflehunter.avalongroup.net>

> additional advanced concepts. From there, I would recommend "Programming
> Python 2nd Ed." as well as the indispensible "Python Essential Reference"
> (altho I think a new edition is in the works so you might want to wait for
> that).

I'm am searching for a review of PP2nd. Is it a significantly
different book from PP1st?

Also. I'd love to see a new version of Beazley's book. I know
that Ivan isn't reading this list at the moment, but I found
the colophon of PER to be a fascinating and enjoyable bit of
reading. Hope he gets to do something in the update too.


-- 
Stand Fast,
    tjg.

Timothy Grant                         tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.         <><       (503) 246-3630
>>>>>>>>>>>>>Linux, because rebooting is *NOT* normal<<<<<<<<<
>>>>This machine was last rebooted:  64 days     1 hours ago<<



From emile at fenx.com  Sat Mar 31 21:01:52 2001
From: emile at fenx.com (Emile van Sebille)
Date: Sat, 31 Mar 2001 18:01:52 -0800
Subject: PythonLabs Team Moves to ActiveState
References: <mailman.986089142.10423.python-list@python.org>
Message-ID: <9a62cg$3it52$1@ID-11957.news.dfncis.de>

DOH!

Those-who-forget-the-past-etc-ly y'rs,

--

Emile van Sebille
emile at fenx.com

---------
"Carlos Alberto Reis Ribeiro" <cribeiro at mail.inet.com.br> wrote in message
news:mailman.986089142.10423.python-list at python.org...
> It's 2 hours before April Fool's Day here (?!?!?)
>







From qrczak at knm.org.pl  Fri Mar  9 15:49:41 2001
From: qrczak at knm.org.pl (Marcin 'Qrczak' Kowalczyk)
Date: 9 Mar 2001 20:49:41 GMT
Subject: "!=" is bad form. Re: sorry....never mind
References: <DZkp6.18811$pF4.3015652@news1.telusplanet.net> <fdlp6.16982$go.2157953@news0.telusplanet.net> <3AA6E308.5B63ECC6@cosc.canterbury.ac.nz> <3AA72253.FB021815@Lugoj.Com> <slrn9afcb5.lou.qrczak@qrnik.zagroda> <3aa93b8e.775214008@news.uswest.net>
Message-ID: <slrn9aigf5.sli.qrczak@qrnik.zagroda>

Fri, 09 Mar 2001 20:23:25 GMT, Jay O'Connor <joconnor at cybermesa.com> pisze:

> I prefer 
> 
> 	if not x == y:

This means a different thing. Should be:
    if x not == y:
consistent with 'not is'.

-- 
 __("<  Marcin Kowalczyk * qrczak at knm.org.pl http://qrczak.ids.net.pl/
 \__/
  ^^                      SYGNATURA ZAST?PCZA
QRCZAK


From pinard at iro.umontreal.ca  Fri Mar 30 15:29:14 2001
From: pinard at iro.umontreal.ca (=?iso-8859-1?q?Fran=E7ois?= Pinard)
Date: 30 Mar 2001 15:29:14 -0500
Subject: Should I close after popen??
In-Reply-To: <uofujnng0.fsf@ctwd0143.fitlinxx.com>
References: <m38zm6r8ih.fsf@master.athome>
	<fhOs6.510$Fb.777424@typhoon2.ba-dsg.net>
	<mailman.985919149.7252.python-list@python.org>
	<uofujnng0.fsf@ctwd0143.fitlinxx.com>
Message-ID: <oqelvfm25h.fsf@lin2.sram.qc.ca>

[David Bolen]

> pinard at iro.umontreal.ca (=?iso-8859-1?q?Fran=E7ois?= Pinard) writes:

> > I would completely agree with the above for someone who wants to write
> > Jython, or write Python that could be quickly turned into Jython.
> > But if you stick with Python, you can safely rely on the fact that a
> > file is automagically closed as soon as you do not refer to it anymore.

> I'd probably change Python in that sentence to CPython, since it's
> an implementation issue and not a language one (the language makes no
> such guarantees).

As it was pointed on this list quite a while ago, Python has no language
definition per itself.  It is rather defined by the documentation of its
main implementation.  It is true that the documentation of (C)Python does
not commit it on the particular issue above, and this is why I checked
with this list before starting to firmly rely on this feature, in real code.

About Jython being Python or not, or JPython being CPython (:-), the
context usually makes it clear what Python means.  I remember that I was
surprised to read that JPython, on its Web pages, had strong opinions about
what Python did rightly or wrongly, or what it should have done otherwise.
I got from there that there is definite distance.  I did not check these
pages recently, at least not since JPython has been renamed Jython.

In the meantime, I think we might continue to safely define Python as per
(C)Python, rather than by the intersection of Jython and Python.  I do know
that if I ever move from Python to Jython, I'll have to revise many things
in my code, but until then, I prefer to be as clean and legible as possible.

> > Code like:
> >     open(logfile, 'a').write('Done!\n')
> > or maybe:
> >     for line in open(logfile).readlines():
> >         process(line)
> > is very legible, and would rather be uselessly encumbered with explicit
> > closes and intermediate variables.

> I [...] would caution that [...] you have no guarantees that your data
> actually got to the logfile without doing the close and possibly handling
> any errors there.

The close *is* properly done.  This is guaranteed by the fact I do not
keep any reference to the file.  On the first example, the close is done
before proceeding with the next Python instruction.  On the second example,
the close is done even before the first "process(line)" is executed.

About errors, shouldn't we just rely on the fact that they raise exceptions,
automatically?  One of the beauties of Python, compared to C or Perl, is that
errors are always trapped and handled by the system, when not by the user.

So, we have all guarantees that the data got written to the logfile, given
the script does not fail with an I/O error.  Errors on implicit close will
also be properly detected.  So, there is no need to feel uncomfortable! :-)

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



From tim.one at home.com  Mon Mar 12 02:30:03 2001
From: tim.one at home.com (Tim Peters)
Date: Mon, 12 Mar 2001 02:30:03 -0500
Subject: Revive the types sig? 
In-Reply-To: <200103120711.AAA09711@localhost.localdomain>
Message-ID: <LNBBLJKPBEHFEDALKOLCEEACJFAA.tim.one@home.com>

Could we please prune followups on this to the Types-SIG now?  I don't really
need to see three copies of every msg, and everyone who has the slightest
interest in the topic should already be on the Types-SIG.

grumpily y'rs  - tim




From tim.one at home.com  Mon Mar 19 21:44:43 2001
From: tim.one at home.com (Tim Peters)
Date: Mon, 19 Mar 2001 21:44:43 -0500
Subject: This math scares me
In-Reply-To: <3AB6B91F.F201AA7C@cosc.canterbury.ac.nz>
Message-ID: <LNBBLJKPBEHFEDALKOLCOEMKJGAA.tim.one@home.com>

[Terry Reedy]
> BCD does not eliminate the problems of finite precision arithmetic.

[Greg Ewing]
> No, but it lets you see exactly what's going on
> in terms that most people are familiar with. It
> eliminates any *extra* surprises due to conversion
> between bases.

In addition, due to the lack of any sort of BCD support in C, any form of BCD
support in Python would most likely be coded via general multiprecision
algorithms.  Meaning you could conceivably boost the precision up to the
limits of memory.  This is very effective in the REXX language:  if your
results look fishy, rerun the program with (say) twice the precision, and see
whether the results agree up to the original precision.  If not, you're
suffering a numeric surprise *somewhere*, and where the two runs start to
diverge is a great clue as to where; or, if they do agree, your original
results were either OK or you're suffering a very unlucky fp surprise.

The original Cray Fortran compiler supported that gimmick "in reverse":  a
compile switch caused each (binary) fp result to get *truncated* to N bits (a
simply masking operation).  Non-experts loved it:  via plotting results
versus number of bits, unstable algorithms often stuck out like a broken
nose.

cleaning-up-the-blood-was-a-bit-unpleasant-though-ly y'rs  - tim




From wtanksle at dolphin.openprojects.net  Thu Mar  1 18:23:46 2001
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Thu, 01 Mar 2001 23:23:46 GMT
Subject: A counter-proposal to __future__ in PEP 236
References: <mailman.983479564.11788.python-list@python.org>
Message-ID: <slrn99tmg1.15u.wtanksle@dolphin.openprojects.net>

On Thu, 1 Mar 2001 15:45:17 -0500, Tim Peters wrote:

>from __future__ import nested_scopes
>assert nested_scopes[1] > sys.version_info, "don't need this anymore"

Is it really important that information such as this be stored as a tuple
rather than a dict (with symbolic names rather than numbers)?

assert nested_scopes.version_info > sys.version_info, \
                            "don't need this anymore"

I've always been taught that magic numbers are unpleasant.

-- 
-William "Billy" Tanksley


From none at none.com  Thu Mar 29 16:13:47 2001
From: none at none.com (Wong Da Qiang)
Date: Fri, 30 Mar 2001 05:13:47 +0800
Subject: Recommendations for Python Books
References: <3ac2582c$1@news.starhub.net.sg>
Message-ID: <3ac3a5eb@news.starhub.net.sg>

Thanks Guys, I will take note of the materials!
"Wong Da Qiang" <noen at none.com> wrote in message
news:3ac2582c$1 at news.starhub.net.sg...
> Just a newbie to programming and python, need some recommendations on
books
> to buy
>
>




From fredrik at pythonware.com  Fri Mar  2 03:04:19 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 02 Mar 2001 08:04:19 GMT
Subject: Performance when working directory is slow network drive
References: <983411179.898268@clam-ext> <mailman.983425921.14373.python-list@python.org> <983480535.813990@clam-ext>
Message-ID: <7uIn6.16286$Qb7.2659111@newsb.telia.net>

Mark Hadfield wrote:
> 1. Why having the current directory in the search path affects the import
> time for the os module much more than it does for other modules I tested
> (like string). Is this because os is a package?

try running the interpreter with -vv (very verbose) might
give you additional clues:

    python -vv -c "import os"

(look for "trying" messages)

Cheers /F




From aahz at panix.com  Wed Mar 14 10:00:22 2001
From: aahz at panix.com (Aahz Maruch)
Date: 14 Mar 2001 07:00:22 -0800
Subject: Atomic Event.{wait,clear}
References: <c22l89.2v1.ln@127.0.0.1>
Message-ID: <98o126$ia6$1@panix2.panix.com>

In article <c22l89.2v1.ln at 127.0.0.1>, David Given <dg at pearl.tao.co.uk> wrote:
>
>I'm using the threading module, with the Event class to handle
>synchronisation. It's all working nicely and is very easy to use.
>
>However, I can't seem to find a way of atomically waiting for an Event,
>and then clearing the flag, which is very strange. Currently I do
>Event.wait() and then Event.clear(). If another thread does a Event.set()
>in between the two, the event will be lost. For my current application,
>this isn't a problem, but what shold I do in the future?

I tell people to use Queue.Queue because it handles both synchronization
and data passing.  Is there some reason why that wouldn't work for you?
-- 
                      --- Aahz <*>  (Copyright 2001 by aahz at pobox.com)

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

Python:
    Programming the way
    Guido indented it


From py.list at mulga.com.au  Fri Mar 30 19:56:13 2001
From: py.list at mulga.com.au (Michael Hall)
Date: Sat, 31 Mar 2001 10:26:13 +0930 (CST)
Subject: newbie variables question
Message-ID: <Pine.LNX.4.10.10103311016540.31580-100000@localhost.localdomain>

I've been trying to do this:

SYSPRODIR = /home/admin/systemprofile
os.system('mv /tmp/install.log SYSPRODIR/install.log')

No matter how I play around with single and double quotes, this won't
work.

But this does work:

os.system('mv /tmp/install.log %s/install.log' %s SYSPRODIR)

Is this the normal/best way to do this in Python ... seems a little
convoluted me, like there must be a simpler way. Bash for example would be
happy with something like:

mv /tmp/install.log $SYSPRODIR/install.log

while PHP is happy with things like:

include ("$SYSPRODIR/install.log");

Thanks for any advice. I haven't found any clear info about this in
Learning Python yet.

Mick




From grante at visi.com  Thu Mar  8 13:32:43 2001
From: grante at visi.com (Grant Edwards)
Date: Thu, 08 Mar 2001 18:32:43 GMT
Subject: gah! I hate the new string syntax
References: <mailman.983615170.23313.python-list@python.org> <tN5o6.2746$eH3.14913@NewsReader> <Koco6.7518$1D5.244602@e420r-atl1.usenetserver.com> <mailman.983997252.5628.python-list@python.org> <m2vgpkyfjo.fsf@mycroft.actrix.gen.nz> <mailman.984063199.18779.python-list@python.org>
Message-ID: <ffQp6.9$Tg.2012@ruti.visi.com>

In article <mailman.984063199.18779.python-list at python.org>, D-Man wrote:
>On Thu, Mar 08, 2001 at 06:51:07PM +1300, Paul Foley wrote:
>| On Wed, 7 Mar 2001 15:33:30 -0500, D-Man  wrote:
>| 
>| > What about one who is both English and Scots, yet never set foot on
>| > that island ;-)?
>| 
>| AFAIK, nobody outside the US would admit the existence of such a
>| thing.  The American habit of calling themselves "Irish" or "Italian"
>| or whatever when they're not is *very* strange.  I've never heard
>| anything like it anywhere else.
>
>Heh.  Maybe Americans are very odd.  Wouldn't surprise me at all.  ;-)

I've always boggled at Americans' habit of claiming foriegn
nationality based on where one or two of their great, great,
grandparents were born. If you were born in the US, your
parents were born in the US, and your grandparents were born in
the US, you are _not_ Irish, Polish, German, Norwegian,
Chinese, or whatever.  You're an American.  Deal with it.

If your parents were born in France and you speak French at
home, I'll let you get buy with claiming to be "French" even
though you were born in the US and carry a US Passport.

-- 
Grant Edwards                   grante             Yow!  I hope I
                                  at               bought the right
                               visi.com            relish... zzzzzzzzz...


From claird at starbase.neosoft.com  Thu Mar 22 15:44:40 2001
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 22 Mar 2001 14:44:40 -0600
Subject: python, perl, lisp, tcl / got a decision to make, can you give me a view point please ? and a joke for python users :)
References: <3aba3076.14395287@news.vif.com> <99dkkd018e0@news1.newsguy.com>
Message-ID: <D5D3EFAC3F91249E.CFABFD6402779F64.3EEAB78CDA9D3126@lp.airnews.net>

In article <99dkkd018e0 at news1.newsguy.com>,
Alex Martelli <aleaxit at yahoo.com> wrote:
			.
			.
			.
>write all kinds of programs.  For your purpose [2], I suggest
>Haskell (it starts out easy, but when you get to monads they
>WILL give your mind a workout -- when, and if, you emerge
>alive and still sane, you will have moved to a higher level of
>consciousness) or javascript (chaotically and ramblingly hard
			.
			.
			.
<URL: http://www.jelovic.com/articles/why_people_arent_using_haskell.htm >
-- 

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


From gerhard.nospam at bigfoot.de  Tue Mar 13 12:20:25 2001
From: gerhard.nospam at bigfoot.de (Gerhard =?iso-8859-1?Q?H=E4ring?=)
Date: Tue, 13 Mar 2001 18:20:25 +0100
Subject: Compiling python under Windows
References: <QGnr6.17986$0N3.109906@news-server.bigpond.net.au>
Message-ID: <3AAE56D9.3E3E61BC@bigfoot.de>

Steve wrote:
> 
> Hi,
> 
> Is it possible to build python on win32 with either mingw (the win32 port of
> gcc) or the borland free compiler?  As I don't want to have to buy MS VC++
> just to try a changes that I plan to the core.

Python 2.1 beta 1 does. I have built it with a Cygwin compiler (you need
a recent one). IIRC Python 2.0 under GCC will not work out of the box.

Gerhard

PS: If you use the Cygwin compiler in Cygwin mode, be sure to compile
with --no-threads, as the Cygwin threads isn't quite there yet.
-- 
Sorry for the fake email, please use the real one below to reply.
contact: g e r h a r d @ b i g f o o t . d e
web:     http://highqualdev.com


From gReGlAnDrUm at earthlink.net  Tue Mar  6 18:59:33 2001
From: gReGlAnDrUm at earthlink.net (Greg Landrum)
Date: Tue, 6 Mar 2001 15:59:33 -0800
Subject: Open Source Organic Chemistry software project
References: <taaseuid59k047@corp.supernews.com>
Message-ID: <983tj3$fio$1@nntp9.atl.mindspring.net>

"mapMonkey" <jziniti at speakeasy.org> wrote in message
news:taaseuid59k047 at corp.supernews.com...
>
> The current code-base is geared to Organic Chemistry molecules
> and the IUPAC nomenclature system and it works quite well.  It
> was written entirely in Python, and entirely by me. My
> goals are to re-engineer this code for added flexibility, and to
> increase the scope of the model to include all (or most) of organic
> chemistry (reactions, submolecular thingies, etc.).
>

>From this description, I'm not sure what it is that you have developed, but
I'll express provisional interest anyway.  :-)

-greg




From DOUGS at oceanic.com  Sun Mar  4 15:02:18 2001
From: DOUGS at oceanic.com (Doug Stanfield)
Date: Sun, 4 Mar 2001 10:02:18 -1000 
Subject: IP Math anyone?
Message-ID: <8457258D741DD411BD3D0050DA62365907A698@huina.oceanic.com>

A cryptic suggestion was made before to follow this link:

http://www.vex.net/parnassus/apyllo.py?i=35559588

I'd like to reiterate that, with the information that it is a set of classes
that implements most of what you'll end up re-implementing in your "crude
hack".  As the discussion showed, its necessary to have both the IP and
netmask info in order to get the math right.  These classes give you that as
a start, and I'd guess adding the math functions would be easier than
starting over.

-Doug-

> -----Original Message-----
> From: Timothy Grant [mailto:tjg at exceptionalminds.com]
> Sent: Tuesday, February 27, 2001 9:09 AM
> To: python-list at python.org
> Subject: Re: IP Math anyone?
> 
> 
> Thanks to all of you who made this thread much more
> enlightening than expected! As I expected, it is a somewhat
> complex problem, for which I will probably throw together a
> class with a nice interface but internals of which are a crude
> hack, and then refine the internals as time allows.
> 
> 
> On Tue, Feb 27, 2001 at 07:38:56AM -0600, Chris Gonnerman wrote:
> > You neglected to tell us where to download your class from, 
> or to attach
> > the source.  I at least would like to see it.
> > 
> > ----- Original Message -----
> > From: "Tino Wildenhain" <tino at wildenhain.de>
> > Subject: Re: IP Math anyone?
> > > Hi Timothy,
> > >
> > > for teaching I made a class for this.
> > >
> > > you can instantiate it with a=ip('192.168.0.254/24') or
> > > a=ip('192.168.0.254','255.255.255.0') just using 
> a=ip('192.168.0.254')
> > > would assign a appropriate subnet mask for this class 
> (class B results in
> > > /16 or 255.255.0.0)
> > >
> > > Since IP-adresses are only 32bit non signed integers, 
> they are internally
> > > represented
> > > like that.
> > > If you make a class like that you can simply add the 
> support to add
> > > integers to
> > > it according to the subnet-mask you use.
> > >
> > > Regards
> > > Tino
> 
> -- 
> Stand Fast,
>     tjg.
> 
> Timothy Grant                         tjg at exceptionalminds.com
> Red Hat Certified Engineer            www.exceptionalminds.com
> Avalon Technology Group, Inc.         <><       (503) 246-3630
> >>>>>>>>>>>>>Linux, because rebooting is *NOT* normal<<<<<<<<<
> >>>>This machine was last rebooted:  41 days 23:20 hours ago<<
> 
> -- 
> http://mail.python.org/mailman/listinfo/python-list
> 



From claird at starbase.neosoft.com  Mon Mar 12 12:36:46 2001
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 12 Mar 2001 11:36:46 -0600
Subject: Well-edited material (was: Python In A Nutshell)
References: <nG5q6.54910$__6.11376601@typhoon.southeast.rr.com> <98c4ok$fhd$1@panix2.panix.com> <slrn9aj98s.ock.grante@isis.visi.com> <98e8vi22sqq@news1.newsguy.com>
Message-ID: <CFA23F735129AF9D.BFB342C6DC911114.A5F48EEE95E307DE@lp.airnews.net>

In article <98e8vi22sqq at news1.newsguy.com>,
Alex Martelli <aleaxit at yahoo.com> wrote:
>"Grant Edwards" <grante at visi.com> wrote in message
>news:slrn9aj98s.ock.grante at isis.visi.com...
>    [snip]
>> "Sorry it's so long, I didn't have time to write it short."
>>
>> No idea where that quote is from, but I noticed in grad school that
>
>Pascal (as in, Blaise, the mathematician and philosopher -- not the
>programming language later named after him!-), in one of his letters.
>
>
>Alex
>
>
>

In the US, this is commonly attributed to Mark Twain.
Goethe seems to believe he found it in Cicero <URL: 
http://lists.village.virginia.edu/lists_archive/Humanist/v05/0440.html >
and I've heard it ascribed to a pantheon of other
aphorists.  In any case, it does appear in Pascal's
*Lettres provinciales*, a book seminal both in theo-
logical apologetics and modern French prose style.
That's a rather different "one of his letters" than a
naive reader might understand.
-- 

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


From aschneid at mindspring.com  Sun Mar 18 20:56:43 2001
From: aschneid at mindspring.com (Boopy Bootles)
Date: Sun, 18 Mar 2001 20:56:43 -0500
Subject: Help Rewriting/Refactoring/Rethinking Parsing Algorythm
Message-ID: <3AB5675A.8363C9F2@mindspring.com>

I'm trying to write a simple piece of code to make programming by voice
recognition software easier to do.  I wrote a very simple function that
would, for example, convert "number customers equals 5" to
"numberCustomers = 5".  But once I started using it, I quickly
discovered several more cases I had to handle.  So far, this is the
list:

number customers            - >     numberCustomers
Abstract Class                 ->    AbstractClass
if error type == 5 then     ->    if errorType == 5:
whiskey x-ray python equals 6     ->     wxPython = 6
normal whiskey type = 'peachy'        ->    whiskeyType = 'peachy'
print "what's wrong with %s?" % first name     ->     print "what's
wrong with %s?" % firstName

(the whiskey x-ray stuff is the International Communications Alphabet,
which you use when spelling something out in Dragon NaturallySpeaking if
NaturallySpeaking is having trouble understanding you).

I've gotten all but the last case to work.  But once I started trying to
incorporate the last case--not messing up quoted strings--my already
overly messy code turned into a hideous snarl.  I know there's _got_ to
be a better way to parse this input, but I don't have a clue where to
start.  I'd like to avoid building a full-blown language parser, which
seems like overkill.

I've included the code below, which correctly translates all but the
last case.  Any thoughts would be greatly appreciated, esp. thoughts re:
a simple object-oriented approach (I'm pretty sure there is one, I just
don't have enough experience writing OO code to figure it out).

Thanks,
Anders Schneiderman

P.S. Once I've gotten this code in better shape, I'll post it
somewhere.  Even at this primitive stage, it makes a _huge_ difference
in writing code by voice (I use NaturallySpeaking plus Natlink, Joe
Gould's terrific system for writing NaturallySpeaking macros using
Python).

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

""" voicecode.py:  routines for translating and otherwise manipulating
voice input into code.
"""

from string import *

SpecialWords = {'equals': ' = ', '=': ' = ', '==': ' == ', '%': ' % ',
   'if': 'if ', 'then': ':', 'elsif': 'elif', 'dot': '.', '.':'.',
'open': ' = open(',
   'try': 'try:', 'init': '__init__ (self, ', 'define': 'def ',
'except':'except:',
   'finally':'finally:', 'tab':'   ', 'blank': ' '}
ICA = {'alpha': 1, 'bravo': 1, 'charlie':1, 'delta':1, 'echo':1,
       'foxtrot': 1, 'golf':1, 'hotel':1, 'india':1, 'juliet': 1,
       'kilo':1, 'lima': 1, 'mike':1, 'november':1, 'oscar':1,
       'papa':1, 'quebec':1, 'romeo':1, 'sierra':1, 'tango':1,
       'uniform':1, 'whiskey':1, 'x-ray':1, 'xray':1, 'yankee':1,
'zulu':1 }

quote = {"'":1, '"':1 }



def translate(words):
    """Given an array of words, translate into code.
The rules are:
 * In general, convert lists of words into wordWordWord
 * For certain words/symbols, aka "special words", convert 'em
 * For ICA words--alpha, bravo, etc.--convert to a single letter
 * If the word is "normal", use the next word exactly as is

NOTE:  Right now, this will NOT work on lines that are quoted:
    print 'this is a test' will translate to 'thisIsATest'.
    I need to find a cleaner way to solve this problem.
"""

    line = ''
    firstWord = 1
    normalWord = 0
    for word in words:
        if normalWord:
            # This word should be used exactly as is (prev word was
'normal')
            if firstWord:
                # Don't capitalize the first word of a variable or a
word in a quote
                line = line + word
                firstWord = 0
            else:
                line = line + capitalize(word)
            normalWord = 0
        elif lower(word) == 'normal':
            normalWord = 1            # next word should be used exactly
as is
        elif SpecialWords.has_key(word):
            line = line + SpecialWords[word]
            firstWord = 1
        elif ICA.has_key(lower(word)):
            # International Alphabet -- convert 'alpha' to 'a', etc.
            line = line + word[0]
            firstWord = 0
        elif firstWord:
            # Don't capitalize the first word of a variable or a word in
a quote
            line = line + word  # Don't capitalize the first word of a
variable
            firstWord = 0
        else:
            line = line + capitalize(word)
    return line






From hungjunglu at yahoo.com  Wed Mar 28 21:00:10 2001
From: hungjunglu at yahoo.com (hungjunglu at yahoo.com)
Date: Thu, 29 Mar 2001 02:00:10 -0000
Subject: Threading legality/morality
Message-ID: <99u4va+ej74@eGroups.com>

--- In python-list at y..., "Syver Enstad" <syver at N...> wrote:
> 
> "John J. Lee" <phrxy at c...> wrote in message > Perhaps you're
> right, but I've seen the same behaiviour on win95, without
> > multi-threading.  When you run command-line (do we mean the same 
thing by
> > console and command-line?) apps it freezes until a key gets 
pressed.  Not
> > using urllib either.
> >
> > It is possible (likely, even) that I was using cygwin, so maybe 
that's
> > where the blame lies.  I'll see if I can reproduce it with the 
standard
> > DOS shell.
> 
> I am using console program/command line whatever you like, and I am 
not
> using Cygwin. I've experienced this both on the BeOpen and 
ActiveState
> distros.

For whatever it's worth, I tried to build Python2.0 from source on 
cygwin, and it has problem with some pthread thing. I had to disable 
thread support when doing ./configure. If I have to guess, cygwin may 
have problems with thread support.

Hung Jung





From grante at visi.com  Thu Mar  8 11:18:44 2001
From: grante at visi.com (Grant Edwards)
Date: Thu, 08 Mar 2001 16:18:44 GMT
Subject: grabbing return codes from os.system() call
References: <%wFp6.9481$dL4.126404@vixen.cso.uiuc.edu>
Message-ID: <EhOp6.232$z6.36409@ruti.visi.com>

In article <%wFp6.9481$dL4.126404 at vixen.cso.uiuc.edu>, Damian Menscher wrote:

>How do I get the return code from an os.system call?  I would
>have expected I could do something like
>
>---returncode---
>#/bin/csh
>echo do stuff
>exit 3
>
>and then in my python program I could do
>
>print os.system('./returncode')
>
>But it prints out 768.  Not particularly useful, even after I
>recognize the trick of dividing by 256 (byte-swapping going on?
>No, because a return code of 768 reports as 0).

A return code of 0x0300 (768) means that the exit status was
0x03, and the signal number was 0x00.

>Given that my real return codes will be (possibly large)
>integers,

They won't be larger than 255.  At least not on any Unix system
I've ever heard of.

>this limitation will likely cause some serious problems down
>the line.
>
>Even better would be a way of returning a string (the script I
>run can be something other than csh, but it has to be a
>separate script).

Take a look at the "commands" module. The function
commands.getstatusoutput("foobar") will return a tuple of exit
status, plus stdout and stderr strings.  It's just a wrapper
around popen2 routines, so you could do the same thing
yourself. Last time I looked it was only available on Unix
systems.

-- 
Grant Edwards                   grante             Yow!  Bo Derek ruined
                                  at               my life!
                               visi.com            


From tim.one at home.com  Tue Mar 20 22:01:58 2001
From: tim.one at home.com (Tim Peters)
Date: Tue, 20 Mar 2001 22:01:58 -0500
Subject: This math scares me
In-Reply-To: <slrn9bfmua.v02.neelk@alum.mit.edu>
Message-ID: <LNBBLJKPBEHFEDALKOLCIEAMJHAA.tim.one@home.com>

[Tim]
> The original Cray Fortran compiler supported that gimmick "in
> reverse":  a compile switch caused each (binary) fp result to get
> *truncated* to N bits (a [simple] masking operation).  Non-experts
> loved it:  via plotting results versus number of bits, unstable
> algorithms often stuck out like a broken nose.

[Neelakantan Krishnaswami]
> Whoa. What an awesomely elegant hack. That goes straight into the box
> of tricks that will someday make me look extremely smart. Thanks!

Na, thank the departed Seymour Cray for designing floating-point units that
traded away accuracy for speed whenever they clashed.  The *original* use of
this compiler switch was to show people coming from 32-bit machines that
their results were wildly different on the 64-bit Cray boxes not because
Cray's arithmetic sucked, but because they had always been getting nonsense
results (when the compiler switch was set to truncate back to the precision
of the box they were coming from, and they got the results they were used to,
and saw that the results bounced around depending on # of bits but eventually
settled down as the # of bits increased beyond what they had before, it was
powerful evidence of numerical instability in their algorithms).

can't-argue-with-bits-the-bits-are-always-right-ly y'rs  - tim




From ckrill at qvlinc.com  Mon Mar 26 16:20:14 2001
From: ckrill at qvlinc.com (Coy Krill)
Date: Mon, 26 Mar 2001 13:20:14 -0800
Subject: Daemonizing a python programme.
References: <mailman.985376338.19672.python-list@python.org> <3ABBE09E.3287DEC5@qvlinc.com> <jtofuo1ucj.fsf@wazor.biostat.wisc.edu>
Message-ID: <3ABFB28E.A0A0A98D@qvlinc.com>

William Annis wrote:

> Coy Krill <ckrill at qvlinc.com> writes:
> >         # get our own session and fixup std[in,out,err]
> >         os.setsid()
>
>         I include these here:
>
>         os.chdir(some_sensible_root_dir_for_core_files)
>         os.umask(000)     # or whatever seems sensible
>

Thanks, I'll incorporate these ideas as well.

> >         sys.stdin.close()
> >         sys.stdout = NullDevice()
> >         sys.stderr = NullDevice()
>
>         This is curious.  Why not just close down STDOUT and STDERR,
> too?

The programmer is smart enough not to try to use any functions that read
from stdin, however, he might import a module that tries to write to
stdout or stderr which would cause the program to crash.  This allows
you to ignore any output with impunity.

Coy




From me at andy.cx  Tue Mar 13 10:30:30 2001
From: me at andy.cx (Lego Andy)
Date: Tue, 13 Mar 2001 10:30:30 -0500
Subject: Threads and subprocesses
Message-ID: <20010313103030.2c20f345.me@andy.cx>

Hello!

I am trying to make execute program from tkinter while still having control over user
interface.

So I tried with combination of threads and os.popen.
Well, it kind of works.

So what I did is this:

done=IntVar()
done.set(1)

def Display(fp, done):
   while not done.get():
	line = fp.read(1)
	if not line:
		break
	print line,

fp = os.popen("some program", 'r')
tr=threading.Thread(target=Display, args=(fp, done))
tr.start()

Now the question is:

Let say that this is done when you press Start. Then when you press Stop, it should
stop executing. Well, I tried by closing fp and I get segfault. I tried killing thread,
but I cannot find any way of doing that. Please help me.


					Andy

--
 /\  |  | |~\  \  /      ------------------------------------------------
/  \ |\ | |  |  \/      / e-mail: andy at x0.org  )\._.,--....,'``.
|--| | \| |  |  /      /  ICQ: 27889915       /,   _.. \   _\  (`._ ,.
|  | |  | |_/  /      /   http://andy.x0.org  `._.-(,_..'--(,_..'`-.;.'
----------------------                                  


From scarblac at pino.selwerd.nl  Sun Mar 11 14:24:13 2001
From: scarblac at pino.selwerd.nl (Remco Gerlich)
Date: 11 Mar 2001 19:24:13 GMT
Subject: Questions about no more nested imports
References: <98gi4l$n9n$0@216.39.170.247>
Message-ID: <slrn9ank32.irm.scarblac@pino.selwerd.nl>

Dave LeBlanc <whisper at oz.net> wrote in comp.lang.python:
> It appears that if this is enforced, one can no longer dynamically
> load extensions/whatever dynamically and all imports must be done at
> the top of a module prior to anticipated use - whether used or not. If
> this is true, then load times for python apps could go up drastically.

The *only* thing that will happen is that "from module import *" (only the
* form) in a scope other than the top module scope will be illegal. It is
already illegal, has always been illegal, and will be enforced by a future
version.

Nothing else. Normal imports are not affected.

-- 
Remco Gerlich



From claird at starbase.neosoft.com  Thu Mar 22 13:03:32 2001
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 22 Mar 2001 12:03:32 -0600
Subject: Jython is cool?
References: <20010322103954.A18539@sync.nyct.net> <mailman.985276416.27994.python-list@python.org>
Message-ID: <D8E322349EF749A7.FE591ED9F551D4F9.8F03B1CFC479FEF9@lp.airnews.net>

In article <mailman.985276416.27994.python-list at python.org>,
Neil Schemenauer  <nas at arctrix.com> wrote:
			.
			.
			.
>I'm hoping that the Mozilla guys catch the Python bug based on
>what Active State has been doing with Komoto and adds "<script
>language=python>" to Mozilla.
>
>  Neil
>

Idle chatter:  this doesn't match my image of "the Mozilla guys".
I recognize how close ActiveState is to Mozilla; the latter, though,
strikes me as eternally in crisis over a collection of challenges
each of which Mozilla collectively regards as more important than
  <script language = "Python">
I'd love to hear the view from a different perspective.
-- 

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


From jimhill at meldrick.swcp.com  Fri Mar 16 02:39:18 2001
From: jimhill at meldrick.swcp.com (Jim Hill)
Date: 16 Mar 2001 07:39:18 GMT
Subject: Announcement: Python Programming: Taming the Snake.
References: <mailman.984699860.10366.python-list@python.org> <kBis6.104$sk3.37174@newsb.telia.net>
Message-ID: <slrn9b3gh3.jnb.jimhill@meldrick.swcp.com>

In <kBis6.104$sk3.37174 at newsb.telia.net>,
Fredrik Lundh <fredrik at pythonware.com> wrote:
>> Open Docs is set to release "Python Programming: Taming the Snake" at the
>> end of June. It will cover 2.0 including features found in 2.1.
>
>a book written by anonymous authors?  that's interesting.

Think of it as "The Story of O" or "Primary Colors" ... only it's about
Python.  It's just crazy enough it might work!


Jim
-- 
jimhill at swcp.com                         http://www.swcp.com/~jimhill/

   "Let me preface my remarks by saying that I have absolutely no love
            for Barry Manilow in my heart."  --  Jason Lempka


From rickp at telocity.com  Wed Mar 14 09:52:43 2001
From: rickp at telocity.com (Rick Pasotto)
Date: Wed, 14 Mar 2001 14:52:43 GMT
Subject: gadfly problem
Message-ID: <slrn9av1dr.mng.rickp@tc.niof.net>

Can someone give me a step-by-step procedure for installing gadfly under
windows? After much fiddling it now imports but when it encounters
self.conn = gadfly.gadfly() it complains that there is no attribute
'gadfly' in module gadfly.

The script works fine under linux but I know very little about dealing
with windows.


-- 
"The state is that great fiction by which everyone tries to live at the
expense of everyone else."
		--Frederic Bastiat
		   Rick Pasotto email: rickp at telocity.com


From aleaxit at yahoo.com  Tue Mar  6 09:15:51 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Tue, 6 Mar 2001 15:15:51 +0100
Subject: I come to praise .join, not to bury it...
References: <mailman.983485923.1136.python-list@python.org> <97p7iq$kk2$1@nntp6.u.washington.edu> <97rka002k9@news1.newsguy.com> <kuu257kb1n.fsf@lasipalatsi.fi> <982a4f0su3@news1.newsguy.com> <kuofvfjb77.fsf@lasipalatsi.fi>
Message-ID: <982rep01hmu@news1.newsguy.com>

"Erno Kuusela" <erno-news at erno.iki.fi> wrote in message
news:kuofvfjb77.fsf at lasipalatsi.fi...
> In article <982a4f0su3 at news1.newsguy.com>, "Alex Martelli"
> <aleaxit at yahoo.com> writes:
>
> | "Erno Kuusela" <erno-news at erno.iki.fi> wrote in message
>
> || what's wrong with vague aesthetic?-) if lots of people share it,
> || it's not to be dismissed so lightly.
>
> | Lots of people share the 'vague aesthetic' of desiring
> | to use braces instead of whitespace to delimit blocks
> | of statements, for example.
>
> | Python dismisses that lightly, and that is definitely
> | a part of what makes it such a wonderful language.
>
> it could also be argued that python uses whitespace instead
> of braces because of vague aesthetic shared by lots of people. :)

If it's "shared by lots of people", how come Python, Haskell and
Occam -- languages which, between them, may account for _maybe_
10% of the world's programmers (much fewer, I suspect) -- are
the only ones to support such whitespace-for-grouping usage...?

Fortunately, Python's choices are not determined by majority
voting, but by our benevolent dictator -- whose good sense of
design, despite the lapse of print>>, is clearly of the highest
order.  I consider the choice of making .join a method of the
joiner object one more example of that excellent design-sense.


Alex





From duncan at rcp.co.uk  Sat Mar  3 11:42:51 2001
From: duncan at rcp.co.uk (Duncan Booth)
Date: Sat, 3 Mar 2001 16:42:51 +0000 (UTC)
Subject: Adding a method to an instance
References: <G2%n6.360714$JT5.11606234@news20.bellglobal.com> <6621ato2vknc667m3iuqvmm8vcj0h99qqk@4ax.com> <Sc0o6.361217$JT5.11613656@news20.bellglobal.com>
Message-ID: <Xns9059A3131B3EAduncanrcpcouk@127.0.0.1>

Bryan Mongeau <bryan at eevolved.com> wrote in <Sc0o6.361217$JT5.11613656
@news20.bellglobal.com>:

>Okay, based on these statements and referring to my previous example, I 
>should be able to do this:
>
>def newMethod(self):
>  print self.bar
>
>f.newMethod = newMethod
>
>Problem is:
>
>f.newMethod()
>TypeError: not enough arguments; expected 1, got 0 
>
>Understandably though, this works in this case:
>f.newMethod(f)

Try this:
    from new import instancemethod
    f = foo()
    f.newMethod = instancemethod(newMethod, f, foo)
    f.newMethod()
 


From root at [127.0.0.1]  Sat Mar 31 02:24:11 2001
From: root at [127.0.0.1] (deadmeat)
Date: Sat, 31 Mar 2001 07:24:11 GMT
Subject: list vs tuple
References: <hGlw6.782$p5.3210@news1.rivrw1.nsw.optushome.com.au> <3AC20CD3.265A2D94@alcyone.com> <ru69cto61jtmrqsf809e38hb3pbm2lhi7g@4ax.com> <bC1x6.4319$4N4.905946@newsc.telia.net> <71q9ct4le63anefmiajj5cbh3mg04qu1ga@4ax.com> <mailman.985991202.5578.python-list@python.org> <7Bfx6.989$p5.3791@news1.rivrw1.nsw.optushome.com.au>
Message-ID: <vCfx6.990$p5.4120@news1.rivrw1.nsw.optushome.com.au>

> >>> b

make that 'a'





From jarober at mail.com  Sat Mar  3 01:38:23 2001
From: jarober at mail.com (James A. Robertson)
Date: Sat, 03 Mar 2001 06:38:23 GMT
Subject: Collection interfaces
References: <3a90fa97.0@news1.mweb.co.za> <3A933830.7F98D118@ix.netcom.com> <CgJk6.515$v77.66568@newsread2.prod.itd.earthlink.net> <220220010813598042%cmh@bDistributed.com> <ud7cakvi8.fsf@roundpoint.com> <MPG.150312ceb68f1ada98969f@news.earthlink.net> <8Gem6.473353$U46.14105062@news1.sttls1.wa.home.com> <MPG.1503548620e632919896a4@news.earthlink.net> <MCjm6.474652$U46.14163288@news1.sttls1.wa.home.com> <MPG.150394195f7f04a19896a7@news.earthlink.net> <3A9A86B1.9A8C69A2@ix.netcom.com> <MPG.1504d14563dce6499896af@news.earthlink.net> <97fjla128ir@news2.newsguy.com> <MPG.1505a76831c34b749896b3@news.earthlink.net> <97h5s40vhc@news2.newsguy.com> <MPG.1506007591cf818b9896bc@news.earthlink.net> <97ifd60sen@news1.newsguy.com> <MPG.1506c16edd0cb6159896bf@news.earthlink.net> <SJ9n6.2187$Yw2.141138@news.uswest.net> <MPG.1506d244a87ac34b9896c3@news.earthlink.net> <97jfj3$9e2@gazette.lmig.com> <MPG.150744dc6bce99ee9896c4@news.earthlink.net>
Message-ID: <3AA0925C.6BEF996C@mail.com>

Topmind wrote:
> 


Nope.  Lack of focus by the primary vendor did the damage.

And it's not dead either.


> Do you know what killed Smalltalk's market? VB!
> I am not saying VB is wonderful, but to be killed by
> VB is not saying much for Smalltalk.
> 
> > "Topmind" <topmind at technologist.com> wrote in message
> > news:MPG.1506d244a87ac34b9896c3 at news.earthlink.net...
> > > > > Using procedural and table techniques, I have matched
> > > > > Smalltalk "can-you-do-this" challenges in the past,
> > > > > and usually with more K.I.S.S.
> > > >
> > ...> OO is far more Guru-oriented than anything before it. It is
> > > generating tons of lore, scriptures, and meditation materials
> > > without any objective proof or metrics that it is better...
> >
> 
> -tmind-

-- 
James A. Robertson
Product Manager (Smalltalk), Cincom
jarober at mail.com
<Talk Small and Carry a Big Class Library>


From slhath at home.com.nospam  Fri Mar  9 10:01:47 2001
From: slhath at home.com.nospam (Scott Hathaway)
Date: Fri, 09 Mar 2001 15:01:47 GMT
Subject: getting the field names from a resultset on Windows with mxODBC
Message-ID: <vf6q6.709$eH1.219552882@newssvr16.news.prodigy.com>

I am using mxODBC for Windows.  When I do a query, how can I get the names
of the fields that the resultset returns (without parsing the SQL myself or
having the user supply them)?

Thanks,
Scott




From tim.one at home.com  Thu Mar  1 16:32:35 2001
From: tim.one at home.com (Tim Peters)
Date: Thu, 1 Mar 2001 16:32:35 -0500
Subject: str behaviour
In-Reply-To: <xjvn1b55vs4.fsf@skepnuskapur.ifi.uio.no>
Message-ID: <LNBBLJKPBEHFEDALKOLCGECFJDAA.tim.one@home.com>

[Igor V. Rafienko]
> I was playing around with printing tuples, when I noticed how str
> behaves with "compound" types (such as tuples):
>
>
> >>> class A:
> ...     def __str__( self ):
> ...         return "I'm A"
> ...
> >>> a = A()
> >>> t = ( A, a )
> >>> str( t )
> '(<class __main__.A at 80fd498>, <__main__.A instance at 810f090>)'
> >>> str( a )
> "I'm A"
> >>>
>
>
> I expected str to be called recursively for the constituent types.
> But, alas, that did not happen. Am I misunderstanding something or is
> it the intended behaviour? If the latter is the case, could someone
> elaborate on why this behaviour was chosen?

You're on the edge of a very deep pit.  The builtin container types (lists,
tuples, dicts) "pass repr down" regardless of whether str() or repr() is
applied at the top level.  I've been whining about that for years myself
<wink>.

The primary reason is maddeningly simple:  it's for the benefit of contained
strings.  If str() "got passed down", then, e.g.,

    d = {"colon": ":", "comma": ",", "' greeting": "I'm A"}
    print d  # note that print implicitly applies str()

would print

   {' greeting: I'm A, comma: ,, colon: :}

instead of today's

   {"' greeting": "I'm A", 'comma': ',', 'colon': ':'}

so it would be an unreadable mess.

I call it "maddening" because, to date, nobody has thought of an alternative
that doesn't suck just as bad in one respect or another.  If DejaNews were
still useful, searching for ssctsoos ("str special casing the snot out of
strings") would yield more blather on the topic than you could tolerate.

Like most other issues involving repr-vs-str, it doesn't often bite so long
as you stick to the builtin types.  But once you write classes that implement
their own notions of __str__ and __repr__, it seems that more often than not
*neither* of them does what you want when instances end up in builtin
containers.

Good topic for a PEP!  Alas, that requires dreaming up a rational solution
first ...




From glevenson at mediaone.net  Tue Mar 20 18:39:29 2001
From: glevenson at mediaone.net (Gary)
Date: Tue, 20 Mar 2001 23:39:29 GMT
Subject: deleting files from a script
Message-ID: <RSRt6.2154$nd.2670928@typhoon.ne.mediaone.net>

Is the best way to do this to just issue a shell command using getstatus or
something similar? I have a file descriptor created with open. I'd rather
not deal with constructing the entire path string if I can avoid it.

Thanks for any advice,

-Gary




From mwh21 at cam.ac.uk  Sun Mar  4 05:49:09 2001
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 04 Mar 2001 10:49:09 +0000
Subject: A counter-proposal to __future__ in PEP 236
References: <j47l29lpo8.fsf@informatik.hu-berlin.de> <97ll2u02qmp@news2.newsguy.com> <3A9FBE5C.95615241@sewardweb.com> <mailman.983578147.29040.python-list@python.org>
Message-ID: <m3ae71dd22.fsf@atrus.jesus.cam.ac.uk>

Thomas Wouters <thomas at xs4all.net> writes:

> On Fri, Mar 02, 2001 at 09:38:04AM -0600, Lloyd Sommerer wrote:
> 
> > I should also point out that the "from  __future__ import..." is also
> > confusing to a newbie, but others have said that already.
> 
> And that's exactly as it should be: you shouldn't use the experimental
> feature(s) if you're a newbie :)

Bzzt.  If your new to Python, you should always use nested scopes, so
you don't have to unlearn the old behaviour when the new behaviour
becomes the default in 2.2.

Cheers,
M,

-- 
  C++ is a siren song.  It *looks* like a HLL in which you ought to
  be able to write an application, but it really isn't.
                                       -- Alain Picard, comp.lang.lisp


From ct at gocept.com  Wed Mar  7 18:43:44 2001
From: ct at gocept.com (Christian Theune)
Date: Thu, 08 Mar 2001 00:43:44 +0100
Subject: Copying objects
Message-ID: <986h3g$m3s$04$1@news.t-online.com>

How do I create a copy of an Instance?

I tried following:

a = [1, 2, 3, 4]
b = a

!will not work, because referencing

I found this tip:

b = a[:]

!works not, too.

How to do else?

Best regards.

Christian Theune


From whisper at oz.net  Thu Mar 29 15:06:43 2001
From: whisper at oz.net (Dave LeBlanc)
Date: 29 Mar 2001 20:06:43 GMT
Subject: NumPY: Reading/Writing Arrays To/From Files
References: <3aae3bd6$0$42882$272ea4a1@news.execpc.com> <3AC1F37D.E8ADA48@seebelow.org> <Pine.SOL.4.30.0103291907490.13267-100000@primrose.csv.warwick.ac.uk>
Message-ID: <9a04kj$kvh$0@216.39.170.247>

Not Found
The requested URL /~da/Travis/numpyio/numpyo-0.5.tgz was not found on
this server.


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

Apache/1.3.19 Server at starship.python.net Port 80

Dave LeBlanc

On Thu, 29 Mar 2001 19:18:32 +0100, "John J. Lee"
<phrxy at csv.warwick.ac.uk> wrote:

>On Wed, 28 Mar 2001, Grant Griffin wrote:
>
>> Ben Held wrote:
>> >
>> > I would like to be able to read/write NumPY array objects to/from ascii
>> > files.  I could not find support for this in the NumPY code.  Any ideas?
>> >
>> > Ben
>>
>> I forget exactly, but I think there's something out there called
>> "arrayio".  Look for Travis Oliphant's page: I think you can find it
>> there.
>[...]
>
>ArrayIO I think is for binary data, not ASCII.  As far as I can tell,
>Travis' pages haven't resurfaced since they disappeared a few months ago.
>There is a copy at
>
>http://starship.python.net/~da/Travis/numpyio/
>
>but this may be ancient for all I know.  You can do the same thing with
>the Python standard library too, IIRC.
>
>TableIO reads ASCII tables.
>
>http://php.iupui.edu/~mmiller3/python/
>
>There is a bug somewhere around line 40 that you'll get if you use Numeric
>arrays where it wants lists.  Just delete the line that checks for array
>type, and put in a try / except if you feel like it.
>
>This is faster than a python version of course, but the GNU C library
>implementation of strtok() has trouble with input like .3 or 3. (one or
>the other).
>
>
>John
>



From garabik-news at spam.melkor.dnp.fmph.uniba.sk  Tue Mar 27 09:31:07 2001
From: garabik-news at spam.melkor.dnp.fmph.uniba.sk (Radovan Garabik)
Date: 27 Mar 2001 14:31:07 GMT
Subject: unique items in lists
References: <_nZv6.2878$yU4.468236@news.infostrada.it>
Message-ID: <985703466.200356@cdwork.cvt.stuba.sk>

comp.lang.python <micampe at f2s.com> wrote:
 : Which is the best (read: fastest) way to remove duplicated items from a list
 : of strings?

import kjbuckets
l = kjbuckets.kjSet(l).items()


-- 
 -----------------------------------------------------------
| Radovan Garabik http://melkor.dnp.fmph.uniba.sk/~garabik/ |
| __..--^^^--..__    garabik @ melkor.dnp.fmph.uniba.sk     |
 -----------------------------------------------------------
Antivirus alert: file .signature infected by signature virus.
Hi! I'm a signature virus! Copy me into your signature file to help me spread!


From fgeiger at datec.at  Tue Mar 27 14:03:53 2001
From: fgeiger at datec.at (Franz GEIGER)
Date: Tue, 27 Mar 2001 21:03:53 +0200
Subject: dynamic web pages?
References: <99q701$jlr10@news-1.bank.dresdner.net>
Message-ID: <99qobm$l2m$1@newsreaderg1.core.theplanet.net>

Try Webware (currently 0.5.1): webware.sourceforge.net.

Reagrds
Franz GEIGER


"Dr. Peter Dintelmann" <Peter.Dintelmann at dresdner-bank.com> wrote in message
news:99q701$jlr10 at news-1.bank.dresdner.net...
> Hi,
>
> is there any module to create dynamic web pages
> using Python?
>
> I am looking for something like Perl's Mason (www.masonhq.com)
> (or PHP). The business logic of my web app should be implemented
> in Python classes which are accessed through a Python/HTML
> mixture.
>
> Thanks in advance,
>
>     Peter
>
>
>




From akuchlin at mems-exchange.org  Thu Mar 22 16:23:09 2001
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 22 Mar 2001 16:23:09 -0500
Subject: New PEP: The directive statement
References: <200103201042.LAA20248@pandora.informatik.hu-berlin.de> <mailman.985107140.4586.python-list@python.org> <9988ie02k7u@drn.newsguy.com> <ulmpzjrpb.fsf@cs.uu.nl> <mailman.985222709.15240.python-list@python.org> <wzr8zpwn4e.fsf@sunshine.cs.uu.nl>
Message-ID: <3d1yrp1osi.fsf@ute.cnri.reston.va.us>

Piet van Oostrum <piet at cs.uu.nl> writes:
> future_statement. But if you remove it after the feature has become
> standard your new modules that use the new division semantics become
> ambiguous. Suppose you have to use Python 1.5.2. or 2.0 because that's all
> your ISP provides. 

Backward incompatible changes have happened before; consider the
str(5L) -> '5' instead of '5L' change in 1.6, which made some 1.5.2
code produces incorrect results.  The __future__ stuff provides a way
to make the transition take longer, but more abrupt transitions have
been made in the past.

It's really a documentation issue; people who release software should
say "I've tested this with Python 2.0", and people with 1.5.2 or 2.1
will have to look out, and perhaps provide patches to fix problems.

> And you have this nifty module that someone wrote. How
> do you know what the semantics of division is in that module? It would be
> much better if the module indicated that it uses the new semantics, for the
> ages to come.

But then, 5 years from now, we'll be stuck with 'from __future__
import interfaces', plus 5 more years of cruft, in all our modules,
and after a few years, no one will consider the semantics novel.
Tools/scripts/classfix.py converts class definition syntax from the
pre-0.9.8 syntax to the current syntax; no one cares about this any
longer.

--amk


From neil.benn at cambridgeantibody.com  Wed Mar 28 14:26:58 2001
From: neil.benn at cambridgeantibody.com (Neil Benn)
Date: Wed, 28 Mar 2001 20:26:58 +0100
Subject: Printing mixtures in integers and strings
Message-ID: <3ac38cf6$0$12246$ed9e5944@reading.news.pipex.net>

Hello,

            I'm trying to use something like this:-

def move(row): #Moves the dispenser to the row on the plate
    print "Moved to Row - " + row

    Where row is expected to be a number.  However when I import this into
teh interpreter I get an error message - stating that __add__ and __radd__
have not been defined:-

Traceback (innermost last):
  File "<console>", line 1, in ?
  File "c:\projects\scripting\TrialScripting\AcmeBRD.py", li
TypeError: __add__ nor __radd__ defined for these operands

    However when I enter text e.g. "BOB" into the function it works fine!
Therefore I assume that I need to explicity convert the into to a string,
does anybody know how to do that - I can;t find the command in teh docs!!!

    Thanks, in advance, for your help.

Cheers,

Neil Benn
Automation Informatics Scientist
Cambridge Antibody Technology




From aleaxit at yahoo.com  Wed Mar 28 14:35:12 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Wed, 28 Mar 2001 21:35:12 +0200
Subject: Boost documentation.
References: <g0wv99x256.fsf@scumbag.ecs.soton.ac.uk>
Message-ID: <99tebh02ofs@news1.newsguy.com>

"Jacek Generowicz" <jmg at ecs.soton.ac.uk> wrote in message
news:g0wv99x256.fsf at scumbag.ecs.soton.ac.uk...
> Are there any sources of documentation for the Boost Python Library
> besides the source examples, getting_started<1-5>.cpp ?

http://www.boost.org/libs/python/doc/index.html

(but the same docs should also have been received when you
downloaded and unpacked the boost package).


> How would I write a C++ function which receives a python list and
> converts it into a std::vector<int> for its own use ?

The only example of mine that I have at hand happens to use a not very
elegant approach, I'm not sure why (maybe other attempts didn't
work back when I wrote it, and I was in a hurry...):

void Deal::set_deck(boost::python::list data) {
    if(data.size() != PACK_CARDS) {
        PyErr_SetString(PyExc_ValueError, "deck must have exactly 52
cards");
        throw boost::python::error_already_set();
    }
    int i;
    PyObject* ppl = boost::python::to_python(data);
    for(i=0; i<PACK_CARDS; ++i)
        deck[i] = boost::python::from_python(PyList_GET_ITEM(ppl, i),
            boost::python::type<int>());
}


Alex





From jkraska1 at san.rr.com  Sat Mar 24 22:40:16 2001
From: jkraska1 at san.rr.com (Courageous)
Date: Sun, 25 Mar 2001 03:40:16 GMT
Subject: time to invest ???
References: <3abd3a17.7103358@news.vif.com> <Yrcv6.2751$45.12829@newsfeeds.bigpond.com>
Message-ID: <vdqqbtk946qh6p7a0tfpf2kng2o0jkk7mm@4ax.com>

>Mastery of the language may take a while...

Mastery of the *language* shouldn't take more than a couple of weeks,
withstanding perhaps some of Python's esoterica. The various different
modules and so forth are another matter altogether.

If a new programmer couldn't be VERY effective in Python within a week
or two, I'd be seriously worried about that programmer. :)

C//



From aleaxit at yahoo.com  Fri Mar  9 10:57:47 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 9 Mar 2001 16:57:47 +0100
Subject: Can anyone recomend a good intoduction to C...
References: <mailman.983894288.19965.python-list@python.org> <ewwp6.1466$y6.241005@ruti.visi.com> <slrn9aeblb.cg1.scarblac@pino.selwerd.nl> <djagat4kk3cs9hg0c3k39cudk6goblhkg6@4ax.com> <989mbk$24nuv$1@fido.engr.sgi.com> <slrn9ah66m.k09.qrczak@qrnik.zagroda>
Message-ID: <98auhn0270j@news2.newsguy.com>

"Marcin 'Qrczak' Kowalczyk" <qrczak at knm.org.pl> wrote in message
news:slrn9ah66m.k09.qrczak at qrnik.zagroda...
    [snip]
> I dislike C too, but at least there is an area where it's a good
> choice: low-level programming, i.e. OS interface, OS itself, runtime
> systems for other languages, time-critical modules for other languages.
> C++ doesn't fit there, is unnecessarily too complicated for these
> tasks, and it isn't a high level language either.

C++ fits *perfectly well* for these tasks.  For example, in writing
Python extensions I'm _vastly_ more productive with the Boost Python
Library (BPL), the rest of Boost (www.boost.org), and the standard
C++ library, than in bare C -- and BPL (just as the rest of Boost and
the standard C++ libraries) just wouldn't be possible in C.

Similarly, high-performance COM objects are a *SNAP* to write in
C++ (here, thanks to the ATL framework) and a *HORROR* to write
in C.  Yes, I write from experience -- I do use C _when this is
absolutely necessary_, and, in particular, I used it for my Win32
tutorial, freely available online (in Italian, sorry), which
includes piece of simple C-coded COM _client_ code (_server_
code would be a lower-down-yet circle of Hell!-), so you can see
the kind of mess that results.  (Microsoft code examples in the
Platform SDK are also good - they do have some hot coders!-).

I haven't gotten into XPCOM yet, but I suspect I'll find a very
similar situation there -- components far easier to develop in
C++ (even if constrained to a limited subset), particularly with
some appropriate framework, and a LOT more work in C.


Alex





From msoulier at nortelnetworks.com  Tue Mar 20 16:19:51 2001
From: msoulier at nortelnetworks.com (Michael P. Soulier)
Date: 20 Mar 2001 21:19:51 GMT
Subject: re.match
References: <997skf$qoc$1@bmerhc5e.ca.nortel.com> <TrMt6.2490$4N4.318602@newsc.telia.net> <99874d$4tp$1@bmerhc5e.ca.nortel.com> <j4r8zsi3pw.fsf@informatik.hu-berlin.de>
Message-ID: <998hhn$cep$1@bmerhc5e.ca.nortel.com>

On 20 Mar 2001 21:29:31 +0100, Martin von Loewis
>
>You have to take two things into account:
>
>a) you were critizing the author of the regex code (in its sre incarnation),
>and 

    While a certain paranoid interpretation of my question could certainly
result in an assumption of criticism, it was in fact a question. My statement
of, "that's what the ^ is for" was perhaps too general, but if you ask someone
what the ^ is for in a regexp, they'll probably tell you, "it means to match
only if the pattern appears at the beginning of the line", no? It is within
this general context that I made that statement, and I'm astonished at the
amazing lack of tolerance I've experience here (and with Usenet in general
really). For some reason people seem to think it's ok to be impolite when you
have a computer to hide behind. I'm sick of it. The very fact that I was not
given the benefit of the doubt is, I believe, a symptom of how badly people
often treat each other on Usenet.
    Thank you for your polite and informative response btw. I'll work harder
to keep myself from being misinterpreted in the future, but it's quite
difficult without one's tone of voice and body language included, and the
first instinct of so many Usenet denizens is to immediatly flame. Be original,
and try tolerance, I say. 

>b) this is regular expressions, which seem be unfriendly by nature (which
>perhaps had an influence on comp.lang.perl.misc as well :-) >

    Naw, I think they just inherited it from comp.lang.c ;-)

c) Fredrik spent
>quite some time getting all the "start of line" and "start of string" cases
>right. It was difficult to write, so it should be difficult to understand :-)

    Now there's an argument against code documentation if I've ever heard one.
;-)

>I find match extremely useful, and I never ever use search. The reason is
>that I use regular expressions for tokenization of languages with a formal
>syntax. There, I don't care whether the point where I look for the next token
>happens to be the start of a line - but I absolutely must make sure that any
>match I get is at the current position (or else there is a syntax error).

    I suppose what I expected was that the ^ in the pattern would enable the
optimization of the search that match provides. My experience with regular
expressions is completely with the Unix tools that use them (ie. grep, awk,
sed, vi...) and Perl. This is the first time I've encountered two different
methods of invoking a regexp comparison. If that's because I've never used the
C API for them, fine. 
    Apparently match also provides additional functionality that these tools
did not previously, to my knowledge, give me access to. 

>Please have a look at the undocumented sre.Scanner class for an example of
>using match in a loop.

    Thanks.

>> As always, one asks questions not to simply annoy, but to learn. There was
>> a time when Usenet was a good place to do that.
>
>Well, maybe you should phrase your question in a way that encourages
>discussion. You've stated 

    I thought I did. See above.

    As the road to hell was paved with good intentions, I suppose what I meant
to imply doesn't count for much, but it did not warrant the initial response,
or the followup that I have chosen to ignore. I do not permit anyone to speak
to me like that "gentleman" already has. I really could care less who they
are. 

    Thanks again,

    Mike

-- 
Michael P. Soulier, TD12, SKY  Tel: 613-765-4699 (ESN: 39-54699)
Optical Networks, Nortel Networks, SDE Pegasus
"With sufficient thrust, pigs fly just fine. However, this is not necessarily a
good idea. It is hard to be sure where they are going to land, and it could be
dangerous sitting under them as they fly overhead." -- RFC 1925
Nortel Linux User's Group Ottawa: (internal) http://nlug.ca.nortel.com


From rol9999 at attglobal.net  Thu Mar 29 15:45:16 2001
From: rol9999 at attglobal.net (Roland Schlenker)
Date: Thu, 29 Mar 2001 15:45:16 -0500
Subject: Accessing variables in __main__ from modules....
References: <mailman.985891179.24578.python-list@python.org>
Message-ID: <3AC39EDC.8F5EC4EB@attglobal.net>

nanotech at europa.com wrote:
> 
> All:
> 
> When writing programs, I like to add print statements that are only
> displayed when the user provides a command line switch such as '--
> debug'.
> 
>   Ex:  %myprog spam.txt ham
>        Found the word 'ham' in 'spam.txt' 45 times
> 
>        %myprog --debug spam.txt ham
>        -I-: Reading 'spam.txt'....has 213 lines....
>        Found the word 'ham' in 'spam.txt' 45 times
> 
> Now, say most of the work is done in a module 'spam.py'. If I wrote
> code in this module like:
> 
>   lines=open(filename).readlines()
>   if (DEBUG):
>     print "-I-: Reading '%s'....has %s lines...."%(filename,len
> (lines))
> 
> DEBUG would be searched for at the module scope. But I want it to
> look for DEBUG back in __main__, like:
> 
>   if (sys.modules["__main__"].DEBUG):
>     print "-I-: Reading '%s'....has %s lines...."%len(filename,lines)
> 
> Is this the "best" way to access variable in the "__main__" scope?
> How else might I solve this (in the python way)??
> 
> Quentin Crain
> 
> Note: I might load many modules all wanting to know if DEBUG is
> turned on. I do not want to pass DEBUG to functions or classes and I
> do not want to set a module variable DEBUG to the same value
> (module1.DEBUG=DEBUG, module2.DEBUG=DEBUG, etc. in __main__). Unless
> these are "better" solutions! :)

How about a module named globals, which contains one line:

DEBUG = 0

Have all your modules import globals.

In __main__ when you want to perform debugging, set DEBUG = 1 in
globals.

Then write your conditional as:

if (globals.DEBUG):
  print "-I-: Reading '%s'....has %s lines...."%(filename,len(lines))

Roland Schlenker



From topmind at technologist.com  Tue Mar 20 01:39:31 2001
From: topmind at technologist.com (topmind)
Date: Tue, 20 Mar 2001 06:39:31 GMT
Subject: Collection interfaces
References: <3a90fa97.0@news1.mweb.co.za>   <MPG.150312ceb68f1ada98969f@news.earthlink.net> <8Gem6.473353$U46.14105062@news1.sttls1.wa.home.com> <MPG.1503548620e632919896a4@news.earthlink.net> <MCjm6.474652$U46.14163288@news1.sttls1.wa.home.com> <MPG.150394195f7f04a19896a7@news.earthlink.net> <3A9A86B1.9A8C69A2@ix.netcom.com> <MPG.1504d14563dce6499896af@news.earthlink.net> <97fjla128ir@news2.newsguy.com> <MPG.1505a76831c34b749896b3@news.earthlink.net> <97h5s40vhc@news2.newsguy.com> <MPG.1506007591cf818b9896bc@news.earthlink.net> <slrn99opop.gr.amk@mira.erols.com> <MPG.1506265ebf6d934d9896be@news.earthlink.net> <97ibmc0k47@news1.newsguy.com> <MPG.1506cf3293b094b49896c1@news.earthlink.net> <97m0c7080c@news2.newsguy.com> <MPG.15085517debf4b649896ca@news.earthlink.net> <zb%n6.31033$o%6.586026@news1.rdc1.bc.home.com> <MPG.150b383cea226d409896d0@news.earthlink.net> <kNko6.6986$eq3.707121@news.uswest.net> <MPG.150c46db9cec283e9896d4@news.earthlink.net> <L9Go6.2720$hZ4.875436@news.uswest.net>
Message-ID: <DWCt6.12229$Im6.1309770@newsread1.prod.itd.earthlink.net>

----- Original Message -----
From: "Randy A. Ynchausti" <mickeyz at qwest.net>
Newsgroups:
comp.object,comp.lang.smalltalk,comp.lang.python,comp.lang.java.advocacy
Sent: Sunday, March 04, 2001 10:20 PM
Subject: Re: Collection interfaces


> Topmind,
>
> > > Please implement code (from scratch) for the following:
> > >
> > > Floating Point Numerical Precision
> > >     default precision
> > >     radix
> > >     machine precision
> > >     negative machine precision
> > >     smallest number
> > >     largest number
> > > Solve Polynomials
> > > Solve Error Function
> > > Solve Gamma Function
> > > Solve Beta Function
> > > Bulirsch-Stoer interpolation algorithm
> > > Newton interpolation algorithm
> > > Spline interpolation algorithm
> > > Langrange interpolation algorithm
> > > Neville interpolation algorithm
> > >
> >
> >
> > Not exactly "typical custom business applications", the domain
> > in which I focus most of my whining.
>
> hum...  me thinks I have been had by vain rantings.
>
> These algorithms/functions are very applicable to reasonable business
> systems including forcasting involving manufacturing and production.  For
> example, if a company does most of its business during the six weeks prior
> to Christmas, an analysis of manufacturing products and inventory that can
> be sold during that period without spoiling is critical, etc.  These
> algothrims/functions and many others can be used to perform some of the
> analysis of sales and production data for managers who need to know
product
> demand, sales, etc., and provide intelligence on how to direct the
company's
> manufacturing and production schedules.
>

Okay.

For the sake of argument, let's say that OO shines if the app is very mathy.

But, what about the rest?

-tmind-







From mcherm_python at yahoo.com  Fri Mar  9 17:20:28 2001
From: mcherm_python at yahoo.com (Michael Chermside)
Date: Fri, 09 Mar 2001 16:20:28 -0600
Subject: Dictionary attributes and functions
References: <Pine.LNX.4.21.0103092014190.19640-100000@moroni.pp.asu.edu>
Message-ID: <3AA9572C.4090002@yahoo.com>

Well, let me take a shot at it (but bear with me... I'm fairly new at 
Python).

 >>> aList = [('a', 123), ('b', 456), ('a', 234)]
 >>> def readAList( aList ):
   """This function reads name-value pairs from a list
   and returns a dictionary whose keys are the names read.
   Each dictionary value is a list of all values which
   paired with that name in the original list."""
   dict = {}
   for (name, value) in aList:
       if dict.has_key(name):
           # it's an existing name
           dict[name].append(value)
       else:
           # it's a new name
           dict[name] = [value]
   return dict

 >>> readAList(aList)
{'b': [456], 'a': [123, 234]}

Of course, you won't be reading a python data structure, so you'll
probably set the variables name and values differently than I do, but
the if statement I've used shows how to add a value to the existing
list if the name is already in the dictionary and create a new, one-element
list if the name is NOT already in the dictionary.

Happy pythoning!

-- Michael Chermside


iddwb wrote:

> I'm looking to read a list of items where their may be duplicates.  I'd
> like to use a dictionary to do this but am way to new to python to know
> fully how to do it.  I'm hoping there is some set of fuctions or
> attributes that can be reference regarding dictionaries... specifically if
> key(a) is already in dictionsary x I'd like to know prior to
> adding.  rather than change the value of key(a) I'd like to add the value
> of key(`a) to the value of key(a) so I end up with
> 
> {keya: [123]}
> on insert of keya againg have something like
> {keya: [123, 234]}
> 
> Any pointers on what to look for in the docs?
> 
> btw, I've been going through Mark Lutz's book on "programming python" but
> I'd like something with a little more langugage reference in it.
> 
> David Bear
> College of Public Programs/ASU
> 




From cfelling at iae.nl  Thu Mar 15 19:29:01 2001
From: cfelling at iae.nl (Carel Fellinger)
Date: 16 Mar 2001 01:29:01 +0100
Subject: I come to praise .join, not to bury it...
References: <mailman.983485923.1136.python-list@python.org> <97p7iq$kk2$1@nntp6.u.washington.edu> <97rka002k9@news1.newsguy.com> <3AAF7E47.294617E1@sage.att.com> <98qeco02hn7@news1.newsguy.com>
Message-ID: <98rmod$c8p$1@animus.fel.iae.nl>

Alex Martelli <aleaxit at yahoo.com> wrote:
> "Garry Hodgson" <garry at sage.att.com> wrote in message
> news:3AAF7E47.294617E1 at sage.att.com...
...
>> don't be too quick to write off Aesthetics.  it's a significant
>> part of what attracted me to python (i'd seen enough curly braces
>> and semicolons to last me a lifetime).  avoiding scaring off newbies
>> until they've had time to appreciate other virtues is not a Bad Thing.

> And then again, the _appearance_ that too much emphasis
> was being placed on syntax-sugary issues kept me *AWAY*
> from Python for longer than it should have -- sure, sure,
> I could save some punctuation, but, *WHAT DID I CARE*?!

> Good thing (for me) that I eventually looked beyond the
> pretty face, of course:-).  But, appearing to care about
> looks more than about substance _is_ going to keep away

Just to get this straithened out:

   Python is not about Good Looks, it's about *Readability* 

But just like with Page Layout Design and Font Design the initiated
and experienced ones can tell just by the looks of something whether
it scores high on readability.  Unfortunately the masses when given
the opportunity --maybe having heard of the arguments of those wise
ones-- think they can do the same which results in horrors when
applied to Page Layout and Font Design as is shown by so many a
publishing nowedays.

So the fun thing is that Aestethics is a very bad guide in general,
but a highly precise although unformal one once one has gotten the
propper education (okee, and even then only for the gifted ones:)

And Aesthethics was about all they had some years ago in the above
mentioned fields, and still what they produced is highly respected.
To exagerate this, there was a time they even didn't know that it was
readability they were after, and yet their keen and seasoned sence of
aesthetics guided them to designs that even today stand as landmarks
of good design (with readability as a prime criterea).

> some people (just as it is going to attract others), so,
> it's not a clear "win" either way, even if you just frame
> it as a popularity-race.  Focus on *substance*, on the

but if you get too blinded by this substance thing, you might easily
overlook that in the end you have to weight things and there again the
seasoned sense of aesthetics of the gifted ones is what makes the
difference.

> other hand, is going to be a win for all who _are_ involved
> with Python -- not just those who happen to be on the
> "I like it" side vs the "I like it not" one (and I've
> never seen any argument about _aesthetics_ [as opposed to

I hope you'll see the argument in the above:)

> _usability_ -- human-factors studies DO have a substantial
> side too:-)] that doesn't boil down to such trifles]).


> Alex





-- 
groetjes, carel


From scarblac at pino.selwerd.nl  Mon Mar 19 03:59:19 2001
From: scarblac at pino.selwerd.nl (Remco Gerlich)
Date: 19 Mar 2001 08:59:19 GMT
Subject: Deleting dictionary items
References: <98ug6n$q9d$1@news6.svr.pol.co.uk> <98vs4p$k9e$1@slb7.atl.mindspring.net> <993570$71$1@newsg2.svr.pol.co.uk> <993tnu$9fh$1@news.udel.edu>
Message-ID: <slrn9bbifc.8sl.scarblac@pino.selwerd.nl>

Terry Reedy <tjreedy at udel.edu> wrote in comp.lang.python:
> 
> > This makes me think, would it be a sane way of doing things to use a list
> > comprehension to modify a dictionary?
> ...
> > filter(dict, lambda key: key[attr] == value)
> >
> > Is this possible? Desireable? Too stupid to contemplate?
> 
> Funny coincidence.  Just 3 miniute ago, before reading this, I tossed out a
> similar idea, in the thread Subsetting a Dictionary, of adding 'dict
> comprehensions' with syntax directly paralleling list comprehensions.
> 
> {k,v for k,v in dict if condition(k,v)}
> 
> Lists and dicts are subcategories of collection or set.  Subseting is, by
> axiom, a general set operation not limited to ordered sets (lists).  So
> this is hardly a stupid idea.

I'd like 

{ k:v for k,v in dict.items() if condition(k.v) }

better, as long as the "for k,v in dict:" loop isn't going to be in the
language. If it is, then the items() isn't necessary. And I like : better
than ,.

-- 
Remco Gerlich


From n8gray at caltech.edu.is.my.e-mail.address  Fri Mar 23 00:15:02 2001
From: n8gray at caltech.edu.is.my.e-mail.address (Nathaniel Gray)
Date: Thu, 22 Mar 2001 21:15:02 -0800
Subject: Good Editor? (What's the big deal with syntax colouring?)
References: <mailman.985291829.27633.python-list@python.org> <99dqqo$gpi$1@news.mathworks.com> <99e9sv$usmp$1@ID-11957.news.dfncis.de>
Message-ID: <99emg2$qif@gap.cco.caltech.edu>

Emile van Sebille wrote:

> No!  Black on pale yellow.  Routines not exceed one screen's worth.
> 
> Now-if-I-could-only-read-5-point-type-on-my-21"-monitor-ly y'rs
> 

My coworkers cringe when they see me coding.

White on dark olive-green, baby!  Comments, of course, are in italicized 
teal.

I think the reason I find it so appealing is that looks like the color of 
white chalk on a blackboard.  Very readable and low-strain.

-n8

-- 
_.~'^`~._.~'^`~._.~'^`~._.~'^`~._.~'^`~._
             Nathaniel Gray
   California Institute of Technology
     Computation and Neural Systems
     n8gray <at> caltech <dot> edu
_.~'^`~._.~'^`~._.~'^`~._.~'^`~._.~'^`~._



From bhk at dsl.co.uk  Wed Mar 14 11:39:44 2001
From: bhk at dsl.co.uk (Brian {Hamilton Kelly})
Date: Wed, 14 Mar 2001 16:39:44 GMT
Subject: Einstein's Riddle
References: <oNPq6.396644$ge4.138479566@news2.rdc2.tx.home.com> <UdSq6.2184$54.2166@www.newsranger.com> <984380185.768920@news.aei.ca> <uP%q6.2375$54.2207@www.newsranger.com> <4w6r6.193645$df5.5192565@news1.crdva1.bc.home.com> <3aae4114.63604234@nntp.interaccess.com>
Message-ID: <984587984snz@dsl.co.uk>

In article <3aae4114.63604234 at nntp.interaccess.com>
           olczyk at interaccess.com "Thaddeus L Olczyk" writes:

> I also figure the fish thing was simply a typo, and that this guy (
> who hasn't resdponded ) was probably given this as a homework
> assignement, and is simply looking for a cheap way to get it done.

It wasn't a typo; after all, one TROLLS for fish.

"No fair", starting a troll by cross-posting to EIGHT groups.  One's
supposed to limit the initial posting to no more than eight groups.

At least "Graham" hasn't cheated on the classic criteria for a troll by
responding.

-- 
Brian {Hamilton Kelly}                                          bhk at dsl.co.uk
    "We have gone from a world of concentrated knowledge and wisdom to one of
    distributed ignorance.  And we know and understand less while being incr-
    easingly capable."                          Prof. Peter Cochrane, BT Labs



From corour01 at motorola.com  Fri Mar 23 06:14:04 2001
From: corour01 at motorola.com (O'Rourke Clodagh-corour01)
Date: Fri, 23 Mar 2001 11:14:04 -0000
Subject: Tkinter - Widget Colours
Message-ID: <13EE655665F4D311AB4D0008C789498A01EE2C9D@zei02exm02.cork.cig.mot.com>

Sorry all - spellcheckers!! 
should have been bg(background)/fg(foreground)

-----Original Message-----
From: O'Rourke Clodagh-corour01
[mailto:Clodagh_O'Rourke-COROUR01 at email.mot.com]
Sent: 23 March 2001 11:10
To: 'python-list at python.org'
Subject: Tkinter - Widget Colours


Hi all,

I was wondering does anyone know how the bag/fag colours can be changed after the widget is packed.  

Thanks,
Clodagh

-- 
http://mail.python.org/mailman/listinfo/python-list



From lluang at northwestern.edu  Mon Mar 19 20:48:39 2001
From: lluang at northwestern.edu (Louis Luangkesorn)
Date: Mon, 19 Mar 2001 19:48:39 -0600
Subject: HELP - beginner to programing/python
References: <Vpxt6.416511$Pm2.6478704@news20.bellglobal.com>
Message-ID: <3AB6B6F7.5BF4BBD7@northwestern.edu>

Hi, this is practically a FAQ type question.

Look at this article on the O'Reilly network (O'Reilly is a publisher)  It is
aimed at people who are new to programming in general and Python in particular.

http://www.oreillynet.com/pub/a/python/2001/02/07/pythonnews.html

As proof of Stephen Figgins genuineness, even though he is writing for the
O'Reilly web site, he does not recommend any of the O'Reilly books :-)

Documents on the web are linked on http://www.python.org/doc/Intros.html  In
particular, there is a fairly new section titled "Introductions to Python
programming for non-programmers."  Most of these are described in detail in the
O'Reilly article.  If you are not able to process LaTeX, the one you should get
if you want to print something out is Non-Programmers Tutorial For Python, which
has a PDF version http://www.honors.montana.edu/~jjc/easytut/easytut.pdf The
others are HTML documents so you would have to download/print them chapter by
chapter.

The standard replies to this question are the tutorial that is on Python.org, as
well as O'Reilly's Learning Python.  But both of these (and all of O'Reilly's
Learning ... series) tend to be for people who already know how to program and
want to add Python to their repetoire.  (of course, this may just describe you.
Look at the Python Intro page for the section labled "Full Tutorials"

Louis
--
K Louis Luangkesorn
lluang at northwestern.edu  http://pubweb.nwu.edu/~kll560  PGP:0xF3E2D362

Whatsoever things are true, ... honest, ... just, ... pure, ... lovely, ... of
good report; if there be any virtue, and if there be any praise, think on these
things.
     - motto - Northwestern University (Phil 4:8)




From ralf.claus at teleatlas.com  Thu Mar 29 04:26:26 2001
From: ralf.claus at teleatlas.com (Ralf Claus)
Date: Thu, 29 Mar 2001 11:26:26 +0200
Subject: how to write an error message in a file....
Message-ID: <3AC2FFC2.A60F886E@teleatlas.com>

....befor the programm quit by an error ?

I would like to store the different errors in a file!?
Is there a possibility ?

cu

Ralf


From support at internetdiscovery.com  Sat Mar 17 16:34:59 2001
From: support at internetdiscovery.com (Mike Clarkson)
Date: Sat, 17 Mar 2001 21:34:59 GMT
Subject: How do I avoid the Tkinter 30 msec polling loop.
References: <98toci$ktjh$1@fido.engr.sgi.com> <QdOs6.506$Fb.774749@typhoon2.ba-dsg.net>
Message-ID: <3ab3d735.8139907@24.0.228.33>

On Sat, 17 Mar 2001 18:41:52 GMT, "David Allen" <mda at idatar.com>
wrote:
>
>Check out:
>http://www.pythonware.com/library/tkinter/introduction/x9129-alarm-handlers-and-other.htm
>
>It is a page on Tkinter non-event callbacks.  It has
>a set of methods described there for Tkinter where you
>can call a certain function once every X milliseconds
>where you specify X.  If you read that page, I think
>it will answer your questions.

No, his question is about a fundamental _tkinter.c inefficiency.

BTW, there is an incompleteness in the documentation on that page:
The documentation for after_cancel does not mention where `id' comes
from, and the documentation for after() does not note the most
important usage of after(), namely 
	id = after(ms,func)

>From the Tkinter.py source code:

    def after(self, ms, func=None, *args):
        """Call function once after given time.

        MS specifies the time in milliseconds. FUNC gives the
        function which shall be called. Additional parameters
        are given as parameters to the function call.  Return
        identifier to cancel scheduling with after_cancel."""

Mike.


From aleaxit at yahoo.com  Wed Mar  7 04:01:01 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Wed, 7 Mar 2001 10:01:01 +0100
Subject: I come to praise .join, not to bury it...
References: <mailman.983928669.28324.python-list@python.org>
Message-ID: <984tcd010qp@news1.newsguy.com>

"Mike C. Fletcher" <mcfletch at home.com> wrote in message
news:mailman.983928669.28324.python-list at python.org...
> Aesthetes are a different kind of professional, I'm a design theorist,
those
> are art wh#$3s ;) .  Sorry, I dislike the connotations of "aesthete" (too
> turn of the last century eclecticism for me), so I use aesthetic (which, I
> feel, has a much nicer resonance with "ascetic", which is closer to my
sense
> of beauty).

Funny enough, that is also very close to what _I_, a mere engineer,
perceive as 'beauty': lack of decoration, match between form and
function, spareness, simplicity.  As another 'engineer' put it,
perfection is reached, not when there's nothing more left to add,
but when there's nothing more left to be taken away (well, _almost_
an engineer -- he failed the examinations for his certificate --
although, as a pioneer of flight, he worked basically in the field
of engineering afterwards; maybe the epsilon of engineering he
lacked was the 'but practicality trumps purity' lithmus test...;-).


> I have no problem with the semantics of the string having methods.  I
asked
> for lists and tuples to have a consistent and complete set of methods 2 or
3
> years ago if I'm recalling correctly.  Healing the class/type split is
> something I would like to live to see (if the PSU doesn't get me first).

Indeed, such developments have contributed, and/or will/would
contribute, to the deep underlying beauty of the language in
the above-mentioned reference terms.


> I'll have to give that Python thing a try.  Can you recommend any good
books
> for someone who's been stuck in a backwater language for the last 6 years?

My favourite books for Python beginners are:
"Learning Python", Mark Lutz & David Ascher, O'Reilly, 1-56592-464-9
"The Quick Python Book", Daryl Harms & Kenneth McDonald , Manning,
1-88477-774-0

They're not up-to-date to Python 2.0, I believe, but either will give
a good grounding on the fundamentals, and learning the 2.0/2.1 changes
on top of that is no biggie.  Lutz & Ascher cover slowly, thoroughly,
and in depth, the language itself and the key parts of its libraries;
Harms & McDonald move faster, and cover (not quite as deeply) a wider
selection of topics.  Choice between them is a matter of taste.

> Isn't that Python thing slow though?  I heard they don't even have static
> type declarations and they don't support polymorphic function signature
> dispatches.

Python's polymorphism is by-method, not by 'function signature', but that's
hardly an issue of speed.  As for speed, I find Python perfectly adequate
for my needs, with the proviso that some low-level performance bottlenecks
may have to be coded as extension modules -- of course, lots of such high
performance 'extensions' already exist, so, depending on your needs, you
may need no non-Python coding (e.g., for numeric operation on large arrays
and matrices, the Numeric extensions may already provide what you need).


> "," . join( ) doesn't help much, as there's no graphical connection drawn
> between the two items, so you lose the correspondence of the . character
to
> "member of" (i.e. the glyph no longer represents the concept, but must be
> interpreted at a semantic level to make the construct clear).

Personally, I find ','.join(blahblah) quite adequate -- if I had to
use double rather than single quotes, I would indeed see an epsilon's
worth of "loss" in visual terms.


Alex





From larry at smith-house.org  Thu Mar 15 11:44:17 2001
From: larry at smith-house.org (Larry Smith)
Date: Thu, 15 Mar 2001 11:44:17 -0500
Subject: Cameron Laird Quote on www.unixinsider.com -- Probably old news to 
 y'all, but new to me.
References: <bJOr6.9481$TW.38047@tor-nn1.netcom.ca> <3AB00593.D1C4BBBF@fxtech.com>
Message-ID: <3AB0F161.90AB4D08@smith-house.org>

"Paul T. Miller" wrote (quoting Cameron Laird):

> I love Python. I especially love it as an embedded scripting language.
> But it is not going to replace graphics applications where speed is
> paramount.

Until, of course, we have speed to burn.  When we hit the
limit on single CPU systems, it will not be long before
the industry begins to seriously tackle multi-CPU systems.
In which case the above statement will eventually be some
permutation of:

I love Python.  I especially love it as a embedded scripting
language for raytracing and textures.  But it is not going
to replace genetic permutation and world simulation where
speed is paramount.

-- 
 .-.    .-. .---. .---. .-..-. | Do not mistake  my cynicism
 | |__ / | \| |-< | |-<  >  /  | for dispair for deep inside
 `----'`-^-'`-'`-'`-'`-' `-'   | me is an  optimist  forever
  http://www.smith-house.org/  | asking, "Why not?"


From tim_one at email.msn.com  Tue Mar 20 00:36:29 2001
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 20 Mar 2001 00:36:29 -0500
Subject: Why "from __future__" stinks; a counter-offer
In-Reply-To: <3AB6E557.6D9DFF05@cosc.canterbury.ac.nz>
Message-ID: <LNBBLJKPBEHFEDALKOLCAENGJGAA.tim_one@email.msn.com>

[Greg Ewing]
>   from __future__ import no_future_statement
> 
> :-)

I expect it will be more like:

    from __future__ import directive_is_a_keyword
    directive future no_future_statement

Although, clearly, they will have to be given in that order <wink>.

if-you-ask-me-they're-both-ugly-ly y'rs  - tim




From mcfletch at home.com  Sun Mar 11 15:46:49 2001
From: mcfletch at home.com (Mike C. Fletcher)
Date: Sun, 11 Mar 2001 15:46:49 -0500
Subject: Pythonwin default directory init
In-Reply-To: <3aab04bc.2359756@News.CIS.DFN.DE>
Message-ID: <000001c0aa94$ba7f23b0$a9807018@cr706570a>

Edit the shortcut (right-click, choose properties) so that "start in" (in
shortcut properties, second tab in the dialogue) points to your desired
directory.

HTH,
Mike

-----Original Message-----
From: costas at meezon.com [mailto:costas at meezon.com]
Sent: Saturday, March 10, 2001 11:58 PM
To: python-list at python.org
Subject: Pythonwin default directory init


I use Pythonwin and I would like it if the File/Open or File/Save
would default to a directory of my choice when I start it up.

I have tried making a startup module to set the paths of sys but that
doesn't seem to work. I have looked thru the .py files trying to
figure out where I could fix this and can't find it.

Also setting it up as a shortcut on the Desktop defaults to the
Desktop.

Does anyone know how to do this?

TIA

Costas
--
http://mail.python.org/mailman/listinfo/python-list




From grey at despair.rpglink.com  Fri Mar 23 22:29:46 2001
From: grey at despair.rpglink.com (Steve Lamb)
Date: Sat, 24 Mar 2001 03:29:46 -0000
Subject: Good Editor? (What's the big deal with syntax colouring?)
References: <mailman.985291829.27633.python-list@python.org> <99dqqo$gpi$1@news.mathworks.com> <99e9sv$usmp$1@ID-11957.news.dfncis.de> <99emg2$qif@gap.cco.caltech.edu> <ha3obts0kp05lonh3o25fsesr38bouo839@4ax.com>
Message-ID: <slrn9bo559.48h.grey@teleute.dmiyu.org>

On Sat, 24 Mar 2001 02:52:48 GMT, Courageous <jkraska1 at san.rr.com> wrote:
>*shrug*. I like black on parchment. Looks like print on paper. Very
>readable and low-strain. :)

    Don't do much coding in the dark, do you?

    I never understood why people insist on it "looking like paper" when there
is one major difference.  Paper is reflective, a monitor is projective.  I've
never been blinded by a 19" piece of paper 3' from my face but I sure have
been blinded by a 19" monitor 3' from my face because of this "looking like
paper" bit.  :/

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


From aleaxit at yahoo.com  Fri Mar  9 18:12:04 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sat, 10 Mar 2001 00:12:04 +0100
Subject: function access to caller's symbols?
References: <mailman.984165207.5080.python-list@python.org>
Message-ID: <98dpog22ckn@news1.newsguy.com>

"Lance E Sloan" <lsloan at yahoo.com> wrote in message
news:mailman.984165207.5080.python-list at python.org...
> I'm a Python newbie, so if I'm approaching this problem
> the wrong way, let me know.
    [snip]
> So, how can I make my function access its caller's
> symbol table?

This is indeed an unnecessarily complicated approach in
most cases (though you can access the caller's global
symbols by raising and catching an exception, and using
the resulting traceback object to walk the stack).  I would
strongly suggest NOT using such black-magic tricks, by
making the directory arg not optional -- the caller may
very well pass globals() explicitly if it wants.


Alex





From m.hadfield at niwa.cri.nz  Wed Mar  7 14:41:36 2001
From: m.hadfield at niwa.cri.nz (Mark Hadfield)
Date: Thu, 8 Mar 2001 08:41:36 +1300
Subject: ScriptControl (was Re: Can Python replace Visual Basic? Should it?...and generalities : ))
References: <mailman.983905329.22236.python-list@python.org> <2Ybp6.12845$hn5.1818206@news1.rdc1.mb.home.com> <9852ui0166a@news1.newsguy.com>
Message-ID: <983994152.56521@clam-ext>

"Alex Martelli" <aleaxit at yahoo.com> wrote in message
news:9852ui0166a at news1.newsguy.com...
> "Brad Bollenbach" <bbollenbach at homenospam.com> wrote in message
> news:2Ybp6.12845$hn5.1818206 at news1.rdc1.mb.home.com...
> >
> > >   A Python "OCX" control needs to be written.  This control
> > >   would simply encapsulate Python for us Visual Basic idiots.
> > >   This way we could 'pop in the control in our VB project and
> > >   start using Python for parts of the program.
> >
> > That's a neat idea. I hope someone smarter than me decides to do it. :)
>
> Microsoft has done it!-)

With a little help from Mark Hammond :-)

---
Mark Hadfield
m.hadfield at niwa.cri.nz  http://katipo.niwa.cri.nz/~hadfield
National Institute for Water and Atmospheric Research




From dan at eevolved.com  Thu Mar  1 17:25:19 2001
From: dan at eevolved.com (Dan Parisien)
Date: Thu, 01 Mar 2001 22:25:19 GMT
Subject: catching a del
Message-ID: <j%zn6.322252$f36.11513719@news20.bellglobal.com>

class obj:
        def __init__(self, name, parent):
                parent.children[name] = self
                self.parent = parent
                self.name = name

        def onDelete(self): #?!
                del self.parent[self.name]

I want to be able to catch a del of an instance of the above class so I can 
remove it from its parent dictionary. Is there a way? (note it cannot be in 
a destructor because there will always be a reference inside the parent 
dict)

Thanks,
Dan


From paulp at ActiveState.com  Sun Mar 11 17:31:15 2001
From: paulp at ActiveState.com (Paul Prescod)
Date: Sun, 11 Mar 2001 14:31:15 -0800
Subject: any interest in type-scoped static constants?
References: <3AAB9578.4C8754E4@fxtech.com> <mailman.984338774.14395.python-list@python.org> <3AABF174.3C986EB4@fxtech.com>
Message-ID: <3AABFCB3.4F13837@ActiveState.com>

Paul Miller wrote:
> 
> ....
> 
> Correct me if I am wrong, but do you not need an *object* of the type
> before you can call getattr on it?
> 
> In other words, I would have to do this:
> 
>         constant_value = Point().CONSTANT_VALUE

Python type objects are not, in general, constructors. 

>>> types.FloatType()
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
TypeError: call of non-function (type type)

Typically modules expose a function (unrelated to the type object) and
the function is the constructor. If you want, though, you could make an
object that was callable and also had a getattr.

It does seem sometimes that the constructor and the type object should
be formally related somehow (as they are for classes) but usually they
are not. I would be interested in a PEP that added *both*
getattr/setattr behavior AND callable-behavior to type objects so that
they were more like class objects.

Fixing this sort of stuff typically falls under the heading "mending the
type/class dichotomy."

http://www.google.com/search?q=class+type+dichotomy

-- 
Python:
    Programming the way
    Guido
    indented it.
       - (originated with Skip Montanaro?)



From mysticsguy at yahoo.ca  Thu Mar 22 12:12:18 2001
From: mysticsguy at yahoo.ca (Yvon Boulianne)
Date: Thu, 22 Mar 2001 17:12:18 GMT
Subject: python, perl, lisp, tcl / got a decision to make, can you give me a view point please ? and a joke for python users :)
Message-ID: <3aba3076.14395287@news.vif.com>

Hello,  i'm a bit confused with so many language ;-)
theres the story, i like to learn a programming language for 2 reasons
1- i want to be able to make some script on my web site (i heard about
zope, don't know what it have to do with python)

2- i want to learn to keep my mind trained, i'm a mystic and i want to
make this mental grow in ability ;-)

so i read about these language, i know a bit of perl and tcl, (i find
perl too much cryptics, difficult to understand other people scripts)
and for tcl i find it too much simple at my taste,  i also find tcl
unclear to read, it's like reading a bunch of chinese sometimes ;-)

so do you think that python can be a good choice or maybe another one
? i am presently downloading python to try some example, maybe it will
fix my mind but i doubt it,  i haven't enough experiences to compare
subject as complicated as computer languages.

oh, theres a joke i found on a website, maybe it will make some of you
laugh ;-)

EXTERIOR: DAGOBAH -- DAY      With Yoda strapped to his back, Luke
climbs up one of the   many thick vines that grow in the swamp until
he reaches the   Dagobah statistics lab. Panting heavily, he continues
his   exercises -- grepping, installing new packages, logging in as
root, and writing replacements for two-year-old shell scripts   in
Python.

YODA: Code!  Yes.  A programmer's strength flows from code
maintainability. But beware of Perl.  Terse syntax... more than one
way to do it... default variables.  The dark side of code
maintainability are they. Easily they flow, quick to join you when
code you write.  If once you start down the dark path, forever will it
dominate your destiny, consume you it will.

LUKE: Is Perl better than Python?

YODA: No... no... no.  Quicker, easier, more seductive.

LUKE: But how will I know why Python is better than Perl?

YODA: You will know.  When your code you try to read six months from
now.
 

Thanks verry much for the time you took to read me.
Yvon.
Yvon Boulianne ; Saphi
mysticsguy at yahoo.ca
www.monunivers.net
Humour,Paradox & Changements


From cce at clarkevans.com  Tue Mar 27 10:15:14 2001
From: cce at clarkevans.com (Clark C. Evans)
Date: Tue, 27 Mar 2001 10:15:14 -0500 (EST)
Subject: who is my caller?
In-Reply-To: <99q8nv$2aibn$1@ID-11957.news.dfncis.de>
Message-ID: <Pine.LNX.4.21.0103271010460.24308-100000@clarkevans.com>

On Tue, 27 Mar 2001, Emile van Sebille wrote:
> If I'm not mistaken, with 2.0 you can directly import traceback without
> needing to generate an exception first.

I was actually looking to do something like..
    
    caller = traceback.call_stack.top()
    try:
      if '0h3kfd' = caller.check('key'):
          # ok... you passed security
    except:
      pass

A string print-out is useless... I'm after
the function objects.... 

This brings up another question, if I
have a function can I get the module
handle; or if I have a bound-function can
I fetch the object that it is part of?

Thanks!

Clark

       




From paul.moore at uk.origin-it.com  Fri Mar  9 09:02:03 2001
From: paul.moore at uk.origin-it.com (Paul Moore)
Date: Fri, 9 Mar 2001 15:02:03 +0100 
Subject: win32com:MakePy -- Early Binding without the size jitters?
References: <mailman.984043211.10915.python-list@python.org> <98ae5r01hn0@news2.newsguy.com>
Message-ID: <guKoOvyz2tkgZ9zg6THcL9zanB2o@4ax.com>

On Fri, 9 Mar 2001 12:08:10 +0100 , "Alex Martelli" <aleaxit at yahoo.com> wrote:

>"Clark C. Evans" <cce at clarkevans.com> wrote in message
>news:mailman.984043211.10915.python-list at python.org...
>    [snip]
>> each library.  I was wondering... is it safe to delete
>> methods/interfaces which I'm not using?  I've deleted
>
>Yes.
>
>> a few, and it still works for my purposes, so I was
>> wondering if I'm shooting myself in the foot -- or if
>> there is a methodological approach to this pruning?
>
>Yes: study the sources and look at the bForDemand
>flag (introduced reasonably recently).  That flag
>ensures that only the wrappers you're actually using
>are generated (it uses packages rather than modules
>to house the wrappers).

More specifically, run makepy -d to generate "on demand" support. Support for
the basic object is generated, but then sub-object support is built as
necessary.

This is quite nice, and I didn't know it was available. Thanks for the
pointer, Alex!

Paul



From lsommerer at sewardweb.com  Fri Mar  2 10:38:04 2001
From: lsommerer at sewardweb.com (Lloyd Sommerer)
Date: Fri, 02 Mar 2001 09:38:04 -0600
Subject: A counter-proposal to __future__ in PEP 236
References: <j47l29lpo8.fsf@informatik.hu-berlin.de> <97ll2u02qmp@news2.newsguy.com>
Message-ID: <3A9FBE5C.95615241@sewardweb.com>

Speaking as someone who is very new to python (3 weeks) and has only
written two programs, using a colon in this manner would have been
confusing.  Right now, my understanding is that the colon exists (apart
from slicing) only to allow statements to follow immediately after a
looping or control structure (on the same line).

This is obviously very distinct from its use with slices.  But it seems
reasonable to me that new python programmers would wonder if this:

directive: nested_scopes

might be equivalent to:

directive:
     nested_scope

I should also point out that the "from  __future__ import..." is also
confusing to a newbie, but others have said that already.

Lloyd Sommerer

Alex Martelli wrote:

> "Martin von Loewis" <loewis at informatik.hu-berlin.de> wrote in message
> news:j47l29lpo8.fsf at informatik.hu-berlin.de...
>     [snip]
> > In this implementation, directive is only considered as a keyword if
> > it appears at the beginning of the module. Therefore, code that
> > happens to use directive as an identifier will not break (unless it
> > assigns to directive as the first thing in the module).
> >
> > I'd appreciate any comments as to whether this directive statement is
> > better or worse for the purpose of indicating that nested scopes are
> > used in a module.
>
> It's indisputably much more explicit and direct.  Recycling
> 'from/import' to mean something very different than other uses
> of from/import is subtle and cute.  This implies a huge plus
> for the 'directive' patch.
>
> Yes, 'directive' does theoretically imply possible code
> breakage -- although, if it were possible to further
> special-case things so that the directive statement
> always took TWO or more directive-words, as in, e.g.
>
> directive nested scopes
>
> or
>
> directive simple scopes
>
> and/or took a colon, e.g.
>
> directive: scopes nest
>
> or
>
> directive: no nesting
>
> it might perhaps be possible to avoid breaking ANY code
> that currently might be working by any stretch of the
> imagination (I have not checked how hard it might be to
> look-forward in the lexer so that only directive-colon
> triggered the new keyword-token -- just reasoning in
> the abstract).  [It seems to me the colon would not
> damage readability at all here -- on the contrary].
>
> Even if the theoretical breakage of some hypothetical
> old module starting with
>     directive = 23
> or
>     directive, floopres = 23,42
> was to be the price of avoiding the cuteness of
> 'from/import' "overloading", I would still vote for
> directive (or, rather, against the overloading thing)
> if I did have a vote in the matter.  Such special cases
> and cuteness are overdone in other languages (C++ and
> its different uses of 'static' come to mind...), and
> they're a bear to teach to newbies.  The _semantics_
> of PEP 236 are a great idea; I do earnestly hope there
> is some way, ANY way, to avoid its syntax, and this
> 'directive' idea gives me a glimmer of hope...!
>
> Alex



From erno-news at erno.iki.fi  Sat Mar 10 09:20:57 2001
From: erno-news at erno.iki.fi (Erno Kuusela)
Date: 10 Mar 2001 16:20:57 +0200
Subject: Urgent Question about Python licensing
References: <98cpfe$5oj$0@216.39.170.247>
Message-ID: <kupufpiu2e.fsf@lasipalatsi.fi>

In article <98cpfe$5oj$0 at 216.39.170.247>, whisper at oz.net (Dave
LeBlanc) writes:

| Hi;
| While reading up on the recent Python conference on O'Reilly's site
| (http://python.oreilly.com/news/pythonday1_0301.html - actually on the
| 'day 2' page), I cam across the following:
| ----------------------------------------------------
[...]
| This difficulty means that Python
| can't be distributed with software licensed under the GPL.

that is false. the issue is not distributiong python with gpl software,
it is mixing the python interpreter code with gpl'd code. (ie where
the other code and the python interpreter would combine to produce
what the gpl refers to as a "derivate work").

  -- erno


From aahz at panix.com  Thu Mar 22 11:29:27 2001
From: aahz at panix.com (Aahz Maruch)
Date: 22 Mar 2001 08:29:27 -0800
Subject: bug in threading
References: <99d2oc$25um$1@gavrilo.mtu.ru>
Message-ID: <99d997$mrj$1@panix6.panix.com>

[posted and e-mailed]

Please send this as a bug report and patch to
http://python.sourceforge.net

In article <99d2oc$25um$1 at gavrilo.mtu.ru>, Oleg Orlov <orlov at diasoft.ru> wrote:
>My problem is: Condition.wait() is not protected against
>KeyboardInterrupt. Fix is below.
>
>Regards,
>Oleg.
>
>-----------------------------
>#python 2.0, threading.py:
>
>...
>class _Condition(_Verbose):
>...
>    def wait(self, timeout=None):
>...
>        saved_state = self._release_save()
># line to be inserted
>        try:
>...
># line to be inserted
>        finally:
>            self._acquire_restore(saved_state)
>
>
>
>


-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

"I won't accept a model of the universe in which free will, omniscient
gods, and atheism are simultaneously true."  -- M


From tim.one at home.com  Sat Mar 17 19:46:37 2001
From: tim.one at home.com (Tim Peters)
Date: Sat, 17 Mar 2001 19:46:37 -0500
Subject: Interactive startup file
In-Reply-To: <3dTs6.2144$ge.55105@news3.mia>
Message-ID: <LNBBLJKPBEHFEDALKOLCEEGBJGAA.tim.one@home.com>

[Brendhan Horne]
> ...
> Can some one please explain this [PYTHONSTARTUP] to me in novice
> english and for a guy who is using windows98 with his python.

Here's a screen dump from a Win98 DOS box:

C:\Python20>echo print "hi" > somefile.py

C:\Python20>type somefile.py
print "hi"

C:\Python20>set PYTHONSTARTUP=c:\python20\somefile.py

C:\Python20>python
Python 2.0 (#8, Oct 16 2000, 17:27:58) [MSC 32 bit (Intel)] on win32
Type "copyright", "credits" or "license" for more information.
hi
>>>

Note that somefile.py got executed (it printed "hi") when I started Python
from the cmdline.

Note that very little of this has to do with Python:  "echo" and "type" and
"set" are all Windows DOS-box commands.  If you want to make the value of the
environment variable PYTHONSTARTUP persist across sessions, you're going to
have to plumb the mysterious depths of autoexec.bat in Win98.  Or you could
decide that this isn't something you actually need <wink>.

more-than-one-way-to-get-out-of-a-corner-ly y'rs  - tim




From bedge at troikanetworks.com  Fri Mar  2 14:32:55 2001
From: bedge at troikanetworks.com (Bruce Edge)
Date: Fri, 02 Mar 2001 11:32:55 -0800
Subject: create new statement?
References: <3A9FE7C2.1F5525D2@troikanetworks.com> <003101c0a349$d5c76ad0$a325fea9@his>
Message-ID: <3A9FF567.920D1027@troikanetworks.com>

Ken Seehof wrote:
> 
> Sorry, not without rebuilding python.
> 
> What is the actual problem that you are trying to solve?  Maybe there's
> another approach.

I'm creating a command line interface to a product.
I want to have access to python as a scripting tool, and be able to
access my own commands.
Ideally, this could be used by the end user. So, I'd like to not require
the my_func(args) syntax,
as that may look a bit kludgey/unfinished from a users standpoint. I'd
like to be able to use "my_command args" from the python cmd line.
Everything else I need is already there, readline line editing,
customizable tab-completer, python language support. 

My expertise is in Tcl, where this is trivial. I'm trying out python as
there are others here that already use it, and it's well, somewhat
syntactically cleaner.


-Bruce.


> 
> ----- Original Message -----
> From: "Bruce Edge" <bedge at troikanetworks.com>
> Newsgroups: comp.lang.python
> To: <python-list at python.org>
> Sent: Friday, March 02, 2001 10:34 AM
> Subject: create new statement?
> 
> > Can you add a new statement to python?
> > AFAICT the only type of commands that don't require () or . after them
> > are statements.
> >
> > I want to add my own command to the interp, but not have to use object
> > calling syntax
> > eg:
> >     create obj opt
> > not
> >     create(obj, opt)
> >
> > How do I define new statements, or, a way of calling functions using the
> > former syntax?
> >
> > Thanks, Bruce.
> > --
> > http://mail.python.org/mailman/listinfo/python-list
> >


From aahz at panix.com  Mon Mar 19 18:55:29 2001
From: aahz at panix.com (Aahz Maruch)
Date: 19 Mar 2001 15:55:29 -0800
Subject: [Very Long (11K)] Numeric PEPs, first public posts
References: <slrn9bcp3h.ns2.qrczak@qrnik.zagroda> <mailman.984748941.31248.python-list@python.org> <Q5Ns6.432845$ge4.152574719@news2.rdc2.tx.home.com> <mailman.985035504.26190.python-list@python.org>
Message-ID: <99669h$so4$1@panix6.panix.com>

In article <mailman.985035504.26190.python-list at python.org>,
Moshe Zadka  <moshez at zadka.site.co.il> wrote:
>
>I have to wonder how come people have not noticed my request to
>*discuss it in the python-numerics sig*, which has a parallel discussion
>to this one.

Because a lot of people don't have time to subscribe to every mailing
list that comes along.  If I have the time, I'm going to be posting some
comments about your proposal, and they're going to be here on c.l.py.
Note carefully this quote from PEP 1:

    A PEP that has not been discussed on python-list and python-dev will
    not be accepted for review.  However, wherever possible, long
    open-ended discussions on public mailing lists should be avoided.
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

There's a difference between a person who gets shit zie doesn't deserve
and a person who gets more shit than zie deserves.  --Aahz


From paul.moore at uk.origin-it.com  Mon Mar  5 10:40:08 2001
From: paul.moore at uk.origin-it.com (Paul Moore)
Date: Mon, 5 Mar 2001 16:40:08 +0100 
Subject: Name of the Script
References: <20010304.185354.2114937732.8277@localhost.localdomain> <NJCo6.6720$qv3.2834377@nnrp5-w.sbc.net> <FpmjOjvwe5bm0Bg09LDTMPhh5vji@4ax.com> <Pine.LNX.4.21.0103051715150.18263-100000@kev.ioc.ee>
Message-ID: <cbKjOqPRwNza00jOm7A1HBDkRYj7@4ax.com>

On Mon, 5 Mar 2001 16:17:38 +0100 , Pearu Peterson <pearu at cens.ioc.ee>
wrote:
>--- a.py ---
>import os,sys
>print os.path.abspath(sys.argv[0])
>---
>might be what you want.

Close, but not quite...

----
C:\Data>python a.py
C:\Data\a.py
C:\a.py

C:\Data>type a.py
import os,sys
print os.path.abspath(sys.argv[0])
os.chdir("C:/")
print os.path.abspath(sys.argv[0])
----

OK, so you could say "don't do that". But if I'm looking at some code
deep in a big program, can I be *sure* that no-one has changed
directory on me?

Interestingly, although the __name__ symbol has the value __main__ in
a script being run as a "main program", the __file__ symbol doesn't
exist. That's a pity, as it would probably be what I wanted...

Paul.



From stephen_purcell at yahoo.com  Fri Mar  2 06:04:19 2001
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Fri, 2 Mar 2001 12:04:19 +0100
Subject: multipart/form-data submission with httplib + MimeWriter?
Message-ID: <20010302120419.A3850@freedom.puma-ag.com>

Wise ones,

Can anyone point me towards some example code for performing
'multipart/form-data' submissions of form values and file contents
using httplib and MimeWriter?

It looks like the topic is covered in 'Internet Programming with Python', but
I don't have a copy to hand, and nothing shows up on google.

Thanks,

-Steve

-- 
Steve Purcell, Pythangelist
Get testing at http://pyunit.sourceforge.net/
Get servlets at http://pyserv.sourceforge.net/
"Even snakes are afraid of snakes." -- Steven Wright



From chris at collegewafer.com  Sun Mar 11 23:35:29 2001
From: chris at collegewafer.com (Chris Baker)
Date: Sun, 11 Mar 2001 23:35:29 -0500
Subject: We have your Wafers Ge, InP, GaAs, SOI, ZnO, ZnS, ZnSe, Sapphire & More!
Message-ID: <E14cK4d-0003ke-00@smtp01.mrf.mail.rcn.net>

Visit http://www.collegewafer.com or call 
800-713-9375, Fax 888-832-0340 or email chris at collegewafer.com

Just arrived! New stocks of 2"6" GaAs, GaSb, Cap Germanium, InP, InAs, InSb, 
Sapphire and more! They're ready to ship!

Get your FREE quote today!

We also have:
Cleaning, Lapping, Etching, Edge round,           graphic printouts (2D & 3D), 
Certificate of      Conformance, Certificate of Analysis,     Polishing, Reclaim 
Services, Laser-cutting and           Oxide Services 

Ask about In-House Oxide Services: 

Thermally grown oxide for 2"-8" wafers ranging in thickness from 500-100,000 
Angstroms.

To Remove yourself, please click the link
http://www.collegewafer.com/contact_us/Remove/remove.html


mtlll








From dwig at advancedmp.net  Thu Mar 15 12:48:36 2001
From: dwig at advancedmp.net (Don Dwiggins)
Date: 15 Mar 2001 09:48:36 -0800
Subject: Who's minister of propaganda this week?
In-Reply-To: "Alex Martelli"'s message of "Thu, 15 Mar 2001 12:37:18 +0100"
References: <B98041329020E007.7DA84CAD9D0562AB.817BBC572F480869@lp.airnews.net>
	<98nu1m$12e@dispatch.concentric.net>
	<660ACCA1E4816F7E.C359CDF4168EF3FB.3B6B05538D619770@lp.airnews.net>
	<3AAFA13D.76E32C3F@divalsim.it> <98oa4o01i4v@news2.newsguy.com>
	<mailman.984619039.7856.python-list@python.org>
	<98q9h60122v@news2.newsguy.com>
Message-ID: <uy9u69b4b.fsf@advancedmp.net>

Alex Martelli writes:
> Another way to express it: an interface is not just, not even
> _mostly_, about existence and signature of methods -- it's mostly
> about prereq's, post-conditions, and invariants; and nobody can
> check those at compile-time in enough cases to make a difference
> to your software's reliability.

I agree whole-heartedly with the first clause, but I wonder: if you made the
same statement over on comp.lang.eiffel, what kind of response would you
get from folks who use the language extensively?  Are Eiffel's prereq's,
post-conditions, and invariants mostly ignored in practice, or are they
commonly and effectively used?

-- 
Don Dwiggins                    "Solvitur Ambulando"
Advanced MP Technology
dwig at advancedmp.net





From aleaxit at yahoo.com  Tue Mar  6 04:02:08 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Tue, 6 Mar 2001 10:02:08 +0100
Subject: I come to praise .join, not to bury it...
References: <mailman.983485923.1136.python-list@python.org> <97p7iq$kk2$1@nntp6.u.washington.edu> <97rka002k9@news1.newsguy.com> <3AA2F4B3.A7496E76@cosc.canterbury.ac.nz> <97vmoh061p@news2.newsguy.com> <3AA46134.D385208B@cosc.canterbury.ac.nz>
Message-ID: <98292h0s0g@news1.newsguy.com>

"Greg Ewing" <greg at cosc.canterbury.ac.nz> wrote in message
news:3AA46134.D385208B at cosc.canterbury.ac.nz...
    [snip]
> > indeed, the string module supplies ready-made versions
> > of such calls
>
> For the time being, yes, but the intention has been
> hinted at on python-dev to deprecate and eventually
> remove the string module. That worries me.

It doesn't worry _me_, but I see where you're coming
from.  Still, I would pass on the issue.  The advantage
of having "one obvious way to do it" IS big, but so is
the problem of hurting the feelings of joiner.join-haters.
I'm glad I have no say on that particular decision!


> > if the string.join function had to typeswitch on the
> > joiner-object to distinguish Unicode from single-byte --
>
> I don't get this business about avoiding type switches
> by dispatching on the joiner, because you still have to
> typeswitch on all the OTHER strings that you're joining.

No!  string_join (in stringobject.c) currently just _checks_
PyString_Check on the items -- if any item fails that, it
delegates to PyUnicode_Join (if PyUnicode_Check OK's that).

The latter is (IMHO) better-coded -- it uses the 'access
through standard interface' paradigm, i.e., it calls
PyUnicode_FromObject to let the items being joined have
a chance to make themselves into Unicode.

Unfortunately, there is (yet) no standard __unicode__
(or whatever) magic method, so the "make yourself into
Unicode" has to go through __str__.

Still, despite these signs of 'early design stages, still
needs to mature', the Unicode-joiner behavior is quite a
bit better already -- consider a typical case:

But absolutely *NOT*!  You access the OTHER stuff through
standard interfaces.  Take a typical case:

class stringable:
    def __init__(self, x):
        self.x = x
    def __str__(self):
        return str(self.x)

class singlebyte:
    def __init__(self, N):
        self.N = N
    def __len__(self):
        return self.N
    def __getitem__(self, index):
        if index>=self.N:
            raise IndexError, index
        return stringable(index)

Now, with something such as:

seq = singlebyte(7)

print '+'.join(seq)

print u'/'.join(seq)

the first print will raise an exception (as the joiner wants
the items to BE string-object already -- it does NOT use the
'access through standard interface' paradigm to give them a
chance to MAKE themselves into strings); the second one will
work just fine (as, IMHO, should the first one).

But anyway, the call to PyUnicode_FromObject is NOT a
typeswitch -- it's an "access through standard interface",
CONCEPTUALLY.  There should be a slot in the type's method
table that gets directly probed to see if the object knows
how to make a Unicode object out of itself -- a fallback
to the _existing_ slot that may indicate a function whereby
the object knows how to make a singlebyte-string object out
of itself should also be included of course.

The implementation may not (yet) EXPLOIT the design's strengths,
but that is no basis by which to judge the DESIGN itself.


> So you save one (very fast C-level) typeswitch out of
> quite a lot. Nowhere near a big enough gain to justify
> anything, to my mind.

You gain a clean approach, which the implementation could
(and should) exploit to access the items in the sequence
through the _appropriate_ standard interface (also, there
should be no mandate on the sequence to define 'length',
I think, just as there isn't for a for-loop -- but I do
understand there are tradeoffs here).  Some gain today,
and an excellent architectural/design basis for further
gains tomorrow when the implementation catches up with
the design's potential for excellence.

You further gain, as I have already shown, the advantage
of polymorphism when you want a joiner object to do
something special -- at NO COST, as usual for uses of
polymorphism -- even code you wrote months ago, with no
idea that 'special joining' would be needed, will get
this extra flexibility if it accepts the joiner object
to use as an argument and just calls its join method.

There are NO technical costs associated with these
advantages.  Your purely-aesthetical, technically
unmotivated dislike for joiner.join, on a strictly
syntax-sugar level, is clearly just about balanced
by the just-as-aesthetical, just-as-arbitrary liking
it by others.  Only technical advantages and costs
can be fairly put on the balance -- and, here, we
have ZERO technical costs to weigh against clear
technical advantages; it does not matter, therefore,
if one judges those technical advantages large (as
I do) or small (as others might) -- they're >0, and
so the joiner.join approach is a 100% winner on
technical grounds (the only rationally debatable ones).

If a single approach should be allowed, then it would
therefore HAVE to be the one presenting technical
pluses, rather than the one presenting NO pluses
on that level.

Note that I've seen NO arguments AT ALL for having
.join be a method on SEQUENCE object (rather than
on the joiner), despite this being often mentioned
on an irrational-purely-aesthetical basis.  Any
taker...?


> (And if ordinary strings and Unicode strings are ever
> unified, this argument will go away completely.)

If it should ever happen that NO polymorphism can
ever appear on the joiner object, then there would
be no technical advantage in allowing polymorphic
behavior on it -- a tautology.  In fact, special kinds
of quasi-string objects might still emerge in the same
kind of far-away-hypothetical-future frame as the
one in which single-byte/Unicode distinctions can
disappear, and then, again, the polymorphic aspects
of joiner.join will show to advantage (as they always
will for any special-joining needs).


Alex





From gradha at iname.com  Sun Mar 11 19:01:31 2001
From: gradha at iname.com (gradha at iname.com)
Date: Mon, 12 Mar 2001 01:01:31 +0100
Subject: Problems writting a wrapper around linux' ssh command
References: <31rd89.lki.ln@127.0.0.1> <Fdwq6.236804$o91.30298800@typhoon.nyroc.rr.com>
Message-ID: <rk3h89.gkk.ln@127.0.0.1>

Dan Maas <dmaas at dcine.com> wrote:
> Right, ssh tries to bypass stdin and read the password from its controlling
> terminal. You can avoid this by creating a virtual terminal and attaching to
> it before executing ssh. The standard 'pty' modules does it for you:

Thank you very much for the info. Now I finally have a working version of
my program and can avoid the tedious typing.

-- 

 Grzegorz Adam Hankiewicz   gradha at iname.com - http://gradha.infierno.org
 Other web pages:  http://glub.ehu.es/  -  http://welcome.to/gogosoftware/


From jhauser at ifm.uni-kiel.de  Tue Mar 13 11:11:25 2001
From: jhauser at ifm.uni-kiel.de (Janko Hauser)
Date: 13 Mar 2001 17:11:25 +0100
Subject: NumPY: Reading/Writing Arrays To/From Files
References: <3aae3bd6$0$42882$272ea4a1@news.execpc.com>
Message-ID: <87snkhhcnm.fsf@lisboa.ifm.uni-kiel.de>

Look for TableIO at

http://php.iupui.edu/~mmiller3/python/

HTH,
__Janko
-- 
  Institut fuer Meereskunde             phone: 49-431-597 3989
  Dept. Theoretical Oceanography        fax  : 49-431-565876
  Duesternbrooker Weg 20                email: jhauser at ifm.uni-kiel.de
  24105 Kiel, Germany


From aahz at panix.com  Wed Mar 21 20:29:01 2001
From: aahz at panix.com (Aahz Maruch)
Date: 21 Mar 2001 17:29:01 -0800
Subject: Simple regex that has me banging my head against a wall
References: <20010321143144.K24341@trufflehunter.avalongroup.net> <20010321235241.D807@apone.network.loc> <mailman.985215628.28804.python-list@python.org>
Message-ID: <99bkgt$i5r$1@panix6.panix.com>

In article <mailman.985215628.28804.python-list at python.org>,
Timothy Grant  <tjg at exceptionalminds.com> wrote:
>
>>>> re.sub(r'\$', '\$', x)
>'\\$1\\$asdfjke\\$lkjasdkjfsdasdfj'


>>>'\\$1\\$asdfjke\\$lkjasdkjfsdasdfj' == r'\$1\$asdfjke\$lkjasdkjfsdasdfj'
1
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

"I won't accept a model of the universe in which free will, omniscient
gods, and atheism are simultaneously true."  -- M


From kens at sightreader.com  Wed Mar 21 11:45:17 2001
From: kens at sightreader.com (Ken Seehof)
Date: Wed, 21 Mar 2001 08:45:17 -0800
Subject: Cookies
References: <mailman.985050745.11337.python-list@python.org> <210320010046053593%timo@alum.mit.edu>
Message-ID: <012b01c0b226$7b1a33b0$a325fea9@his>

Wow, that is absolutementally brilliantissimo!

Also, it might even work :-)

And it solves the problem of providing cheap UI capability to
the python executable via the browser!

Thanks,
- Ken Seehof

Timothy O'Malley <timo at alum.mit.edu>'s brain emits:
> Here's a wacky idea.
> 
> You could generate the id on the server or client side.  If the
> client generates it in JavaScript, then he sends it to the server.  If
> the server generates it, then he already has it.  In either case, the
> the server who responds with an web page that includes an HTTP link
> that points to the following URL
>  
>          http://localhost:8080/id=1234
> 
> On this page, you instruct the user to
>     1) start up your Python application
>     2) click on the link
> 
> The Python application should include a small (and stupid) HTTP sever
> that listens on port 8080 of the localhost.  When it detects a
> connection, it reads the ID out of the requested URL.  (In the above
> example, the ID is the string "1234".)
> 
> When the user clicks on the link, the client browser makes a request
> to the Python application running on his machine.  By this mechanism
> is the unique key passed to the application.  (To be nice, the Python
> application should respond with a nice web page.)





From aleaxit at yahoo.com  Mon Mar 26 05:04:45 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Mon, 26 Mar 2001 12:04:45 +0200
Subject: file locking
References: <5zCv6.583$p5.2667@news1.rivrw1.nsw.optushome.com.au> <mailman.985596999.26319.python-list@python.org>
Message-ID: <99n47u02fk@news2.newsguy.com>

"Steve Purcell" <stephen_purcell at yahoo.com> wrote in message
news:mailman.985596999.26319.python-list at python.org...
> deadmeat wrote:
> > How do I lock a file on opening it?  the bultin open() only allows me to
set
> > the read/write mode, not any deny modes.
>
> On Unix, look at the 'flock' function in the module 'fcntl'.
>
> You can also try the 'posixfile' module -- posixfile objects support
locking.

On Win32, look at module win32file; its CreateFile function has
a shareMode parameter.  Getting a Python fileobject from the
HANDLE returned by CreateFile is a tad tedious -- I _think_ you
have to use win32file._open_osfhandle to get an (integer) C
level handle from the opaque HANDLE, then os.fdopen to get from
said int-handle to a Python fileobject.


Alex





From costas at springmail.com  Mon Mar 12 17:07:30 2001
From: costas at springmail.com (costas at springmail.com)
Date: Mon, 12 Mar 2001 22:07:30 GMT
Subject: This math scares me
References: <3aad4062.1290021@News.CIS.DFN.DE> <Hwbr6.1139$Tg.127571@ruti.visi.com>
Message-ID: <3aad4485.2348499@News.CIS.DFN.DE>

On Mon, 12 Mar 2001 21:50:31 GMT, grante at visi.com (Grant Edwards)
wrote:

>In article <3aad4062.1290021 at News.CIS.DFN.DE>, costas at springmail.com wrote:
>
>>Ok, I can see maybe division having problems. But why does addition of
>>the two numbers below:
>>
>>5.01+5.54
>>
>>give me this?
>>
>>10.550000000000001
>
>Sorry. That's how floating point numbers on computers work.
>
>Many numbers (including 10.01) can't be represented exactly
>in the binary representation used by your computer.
>
>Therefore floating-point results are never going to be exact.  
>
>>I hope its some setting that can take care of this. I am using Python
>>under Windows 98.
>
>Use integer or fixed-point BCD math if you need exact results.

I will also tell my customers that this is the new math. Maybe I will
tell them to use a slide ruler too .

Joking aside, this should be fixed to display the number correctly. If
I use 'print 5.01+5.54" then it gives the result correctly. It seems
to happen only if you type it in the command line without a print
command.

Costas




From dsh8290 at rit.edu  Thu Mar  8 22:15:19 2001
From: dsh8290 at rit.edu (D-Man)
Date: Thu, 8 Mar 2001 22:15:19 -0500
Subject: learning C
In-Reply-To: <3AA82422.53026D3D@cosc.canterbury.ac.nz>; from greg@cosc.canterbury.ac.nz on Fri, Mar 09, 2001 at 01:30:26PM +1300
References: <mailman.983972829.23481.python-list@python.org> <983975751.18833@newsmaster-04.atnet.at> <3AA82422.53026D3D@cosc.canterbury.ac.nz>
Message-ID: <20010308221519.A4367@harmony.cs.rit.edu>

On Fri, Mar 09, 2001 at 01:30:26PM +1300, Greg Ewing wrote:
| Werner Schiendl wrote:
| > 
| > The fine thing is: C++ still HAS the scalpel, so what?
| 
| C++ is a Swiss army knife with both scissors and scalpel,
| and a nasty tendency for the scalpel to shoot out and
| stick you in the palm when all you were trying to do
| was use the scissors...
| 

Nice.  I like this description. 

yet-not-joining-the-war-ly y'rs  -D




From aleaxit at yahoo.com  Thu Mar 29 07:02:29 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 29 Mar 2001 14:02:29 +0200
Subject: Converting an integer base 10 to a binary number
References: <3AC1E898.D5929C44@fh-hof.de>
Message-ID: <99v88q027o3@news2.newsguy.com>

"Peter St?hr" <peter.stoehr at fh-hof.de> wrote in message
news:3AC1E898.D5929C44 at fh-hof.de...
"""
I know the int(x[,radix]) function that converts a string or a number
(for a give radix) to an integer. Is there a built-in function that
provides a conversion the other way round?
Something like
    bin(32) = "10000"
"""

There is no 'bin' built-in function -- you have to roll your own, e.g:

trits = ('000','001','010','011','100','101','110','111')
from string import octdigits as o

def nol0(s):
    for i in range(0, len(s)):
        if s[i]!='0': return s[i:]
    return '0'

def bin(n):
    return nol0(''.join([trits[o.find(x)] for x in oct(n)]))


Alex





From moshez at zadka.site.co.il  Fri Mar 23 14:16:14 2001
From: moshez at zadka.site.co.il (Moshe Zadka)
Date: Fri, 23 Mar 2001 21:16:14 +0200
Subject: newbie question
In-Reply-To: <73125560B5E4EF16.CC3FC0DB3FD693E3.4AF0ACCBD2EFF29B@lp.airnews.net>
References: <73125560B5E4EF16.CC3FC0DB3FD693E3.4AF0ACCBD2EFF29B@lp.airnews.net>, <575F0D65CC45F12D.7D165177C452F6D5.01903532892CC93D@lp.airnews.net> <99fll8$996$1@rex.ip-plus.net> <mailman.985367791.3511.python-list@python.org>
Message-ID: <E14gX2k-0001P8-00@darjeeling>

On 23 Mar 2001, claird at starbase.neosoft.com (Cameron Laird) wrote:

<snipped a lot of wise things about CGI and Apache>
> That's a lot more exciting
> for them than hearing that, with a mere week or two of
> study, they can install Zope on their own.

That's pure FUD. Zope takes <depending on your internet connection> time
to install, about 5 minutes to compile and 30 seconds to startup. And
using it is not necessarily harder then learning CGI. I'm only mentioning
it because "Zope is hard" seems to be a commonly uttered phrase, one that
I believed until I actually had to get a Zope server up and running.

Hooking up Zope to Apache (say, with reverse proxying) is very easy, though
it does use some Apache configuration knowledge. This can all be learned
in a day, and anyone seriously using Apache needs to know about this
stuff.
-- 
"I'll be ex-DPL soon anyway so I'm        |LUKE: Is Perl better than Python?
looking for someplace else to grab power."|YODA: No...no... no. Quicker,
   -- Wichert Akkerman (on debian-private)|      easier, more seductive.
For public key, finger moshez at debian.org  |http://www.{python,debian,gnu}.org



From robin at jessikat.fsnet.co.uk  Sun Mar  4 20:00:07 2001
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Mon, 5 Mar 2001 01:00:07 +0000
Subject: the demise of 'from foo import * and its implications?
References: <3kVBAfATrDo6EwEM@jessikat.fsnet.co.uk>
 <mailman.983582584.8198.python-list@python.org>
 <z8dAGJAyqLo6EwlX@jessikat.fsnet.co.uk>
 <mailman.983746566.27513.python-list@python.org>
Message-ID: <GqQAdIAXUuo6EwyY@jessikat.fsnet.co.uk>

In article <mailman.983746566.27513.python-list at python.org>, Jeremy
Hylton <jeremy at alum.mit.edu> writes
>>>>>> "RB" == Robin Becker <robin at jessikat.fsnet.co.uk> writes:
>
>  RB> my preference would be that statements legal in one context
>  RB> should be legal in another where they make sense. I cannot see
>  RB> how using import * or eval etc in nested scopes is somehow
>  RB> obviously bad or different from the same usage in non-nested
>  RB> scopes. The fact that the prohibition comes from making things
>  RB> easier for the compiler adds to my confusion.
>
>Using import * in function's is asking for trouble, because it makes
>it impossible for someone reading the code to know what a variable
>refers to.  Consider this hypothetical bit of code:
>
>    from string import *
>
>    def f(x):
>        from amodule import *
>        return strip(x)
>
....

doesn't from foo import * make things hard to locate at the global level
as well? exec/eval/import can always do this surely.

This is not an argument against this particular usage alone.

>
>One other thing you're confused about is that the prohibition comes
>from making things easier for the compiler.  That's bunk.  The
>prohibition comes because there is no reasonable and intuitive
>semantics for these features in the presence of nested scopes.  When
>the issue first came up on python-dev, two seasoned Python programmers
>explained what the obvious semantics where.  They chose completely
>opposite semantics.  There's just no sensible meaning to assign to
>this kind of code in most cases.
>
>Jeremy
>

-- 
Robin Becker


From wayne.izatt at _nospam_myself.com  Fri Mar 23 00:50:33 2001
From: wayne.izatt at _nospam_myself.com (Wayne Izatt)
Date: Fri, 23 Mar 2001 05:50:33 GMT
Subject: python, perl, lisp, tcl / got a decision to make, can you give me a view point please ? and a joke for python users :)
References: <3aba3076.14395287@news.vif.com>
Message-ID: <JuBu6.1270$NE.129508@news1.crdva1.bc.home.com>

I'll bite...

Without a doubt, Python is the better of the three. If only I'd
known about it before I leaned the other two.

cheers

"Yvon Boulianne" <mysticsguy at yahoo.ca> wrote in message
news:3aba3076.14395287 at news.vif.com...
> Hello,  i'm a bit confused with so many language ;-)
> theres the story, i like to learn a programming language for 2 reasons
> 1- i want to be able to make some script on my web site (i heard about
> zope, don't know what it have to do with python)
>
> 2- i want to learn to keep my mind trained, i'm a mystic and i want to
> make this mental grow in ability ;-)
>
> so i read about these language, i know a bit of perl and tcl, (i find
> perl too much cryptics, difficult to understand other people scripts)
> and for tcl i find it too much simple at my taste,  i also find tcl
> unclear to read, it's like reading a bunch of chinese sometimes ;-)
>
> so do you think that python can be a good choice or maybe another one
> ? i am presently downloading python to try some example, maybe it will
> fix my mind but i doubt it,  i haven't enough experiences to compare
> subject as complicated as computer languages.
>
> oh, theres a joke i found on a website, maybe it will make some of you
> laugh ;-)
>
> EXTERIOR: DAGOBAH -- DAY      With Yoda strapped to his back, Luke
> climbs up one of the   many thick vines that grow in the swamp until
> he reaches the   Dagobah statistics lab. Panting heavily, he continues
> his   exercises -- grepping, installing new packages, logging in as
> root, and writing replacements for two-year-old shell scripts   in
> Python.
>
> YODA: Code!  Yes.  A programmer's strength flows from code
> maintainability. But beware of Perl.  Terse syntax... more than one
> way to do it... default variables.  The dark side of code
> maintainability are they. Easily they flow, quick to join you when
> code you write.  If once you start down the dark path, forever will it
> dominate your destiny, consume you it will.
>
> LUKE: Is Perl better than Python?
>
> YODA: No... no... no.  Quicker, easier, more seductive.
>
> LUKE: But how will I know why Python is better than Perl?
>
> YODA: You will know.  When your code you try to read six months from
> now.
>
>
> Thanks verry much for the time you took to read me.
> Yvon.
> Yvon Boulianne ; Saphi
> mysticsguy at yahoo.ca
> www.monunivers.net
> Humour,Paradox & Changements




From not.this at seebelow.org  Sun Mar 25 23:54:44 2001
From: not.this at seebelow.org (Grant Griffin)
Date: Sun, 25 Mar 2001 22:54:44 -0600
Subject: time to invest ???
References: <3abd3a17.7103358@news.vif.com>
Message-ID: <3ABECB94.901ADE54@seebelow.org>

Yvon Boulianne wrote:
> 
> Hello, i start to learn Python some day ago, with lot of knowledge as
> a system admin but little as programmer and i wonder what is the
> average time/hours someone have to invest to be able to use a language
> like that easily.  i talk as well about devlopping the way of thinking
> of a programmer and the knowledge of the language. please don't tell
> me theres no average, at least take the idea of a normal guy, not a
> genius and not stupid  ;-)

Guilty as charged.  Perhaps your experience will be like mine.

I "learned" Python in a day by going through the tutorial that came with
it.  Then, having "learned" it, I was able to write very simple
programs.  (My personal favorite exercise at this stage is writing a
Fibonacci function.)

I got nominally "good" at it in few months.  The main thing here was
learning the important library functions, and how to use them.  I
suggest you start with sys, os, os.path, string, re (and probably a
couple of others.)

My background helped a lot: I had been doing Perl and C++ for several
years.  If you aren't versed in scripting languages and object-oriented
languages, it might take you longer.

After six months, I think I was thinking somehwhat Pythonically, that
is, I began to really understand how things fit together.

Now, after a year, I say I'm "fairly good" at Python.  There was a lot
of study along the way.  I read two books ("Learning Python" and "Python
Essential Reference"), and I read a lot of web and newsgroup stuff.  I
also studied Numerical Python's manual.  (Not everybody needs that,
though.)

In hope to truly "master" Python in the next few years.  (That is,
assuming I don't find something better first. <wink>)

Python is simple, but still complex.  It has a lot of power, and can do
a lot of things.  The library and all the third-party modules are
extensive.  There's no getting around the fact that it takes time to
learn.  That's the bad news.  The good news is that it's "approachable":
you can learn it a little at a time, in layers, and you can begin to do
useful things with it within a day.  IMHO, it's very much worth the time
and effort, both in thems of what you can _do_ with it and also in terms
of what you will _learn_ from it.

> Thanks verry much and sorry for my english, my first language is
> french.

Tres bon! (Or whatever it is you people say. <wink>)

your-english-is-better-than-my-dutch,-yet-we-both-can-learn-to
   -think-pythonic-ly y'rs,

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation            http://www.iowegian.com


From chrisl at gnuchina.org  Thu Mar  1 21:02:59 2001
From: chrisl at gnuchina.org (Christopher Li)
Date: Fri, 2 Mar 2001 10:02:59 +0800 (CST)
Subject: define a new func on the fly?
In-Reply-To: <3A9EF31B.4105F05C@troikanetworks.com>
Message-ID: <Pine.LNX.4.33.0103020957430.3393-100000@chris.cn.mvd>


On Thu, 1 Mar 2001, Bruce Edge wrote:

> >>> class xx:
> ...   def __init__(self):
> ...     exec( "def yyy():  print \"yyyyyyyyy\"" )
> ...
> >>> x=xx()
> >>>
> >>> x.__dict__
> {}
>
> Where is yyy defined?
In local namespace of __init__.
>
> Ideally, I'd like to be able to create it in the global namespace.
Do yon mean in class xx?
How about

>>> class xx:
...     def __init__(self):
...         exec( "def tmp(self): print 'yyyyyy'")
...         self.yyy = tmp
...
>>> x = xx()
>>> x.__dict__
{'yyy': <function tmp at 0x81df40c>}

Chris




From js.graham at home.com  Mon Mar 19 07:04:07 2001
From: js.graham at home.com (Steve Graham)
Date: Mon, 19 Mar 2001 12:04:07 GMT
Subject: Einstein's Riddle
References: <oNPq6.396644$ge4.138479566@news2.rdc2.tx.home.com>
Message-ID: <XAmt6.447643$ge4.157329333@news2.rdc2.tx.home.com>

A variety of answers has been posted at
http://members.home.net/js.graham/einstein/index.html.

Steve Graham

===

"Steve Graham" <js.graham at home.com> wrote in message
news:oNPq6.396644$ge4.138479566 at news2.rdc2.tx.home.com...
> Who would be interested in using his/her brain (and his computer) to solve
> the following?
>
>
> Steve Graham
>
> ===
>
> Einstein's Riddle
>
> Albert Einstein wrote this riddle this century [ed. 20th century].  He
said
> 98% of the world could not solve it.
>
> There are 5 houses in 5 different colors.  In each house lives a person
with
> a different nationality.  The 5 owners drink a certain type of beverage,
> smoke a certain brand of cigar, and keep a certain pet.  No owners
> have the same pet, smoke the same brand of cigar or drink the same
beverage.
>
> The question is: "Who owns the fish?"
>
> Hints:
>
> The Brit lives in the red house.
>
> The Sweed keeps dogs as pets.
>
> The Dane drinks tea.
>
> The green house is on the left of the white house.
>
> The green house's owner drinks coffee.
>
> The person who smokes Pall Mall rears birds.
>
> The owner of the yellow house smokes Dunhill.
>
> The man living in the center house drinks milk.
>
> The Norwegian lives in the first house.
>
> The man who smokes Blends lives next to the one who keeps cats.
>
> The man who keeps the horse lives next to the man who smokes Dunhill.
>
> The owner who smokes Bluemasters drinks beer.
>
> The German smokes Prince.
>
> The Norwegian lives next to the blue house.
>
> The man who smokes Blends has a neighbor who drinks water.
>
>
>
>
>
>
>
>
>




From kens at sightreader.com  Tue Mar 20 14:38:45 2001
From: kens at sightreader.com (Ken Seehof)
Date: Tue, 20 Mar 2001 11:38:45 -0800
Subject: Sequence-spreading
References: <3AB7167F.2816AA21@student.liu.se> <99790u01c73@news2.newsguy.com> <99851p$ba6$2@neptun.beotel.net>
Message-ID: <008101c0b175$60dd9c00$04090a0a@upcast.com>

Yikes! No!

>>> a = [[]]*3
>>> a
[[], [], []]
>>> a[2].append(4)
>>> a
[[4], [4], [4]]

Be careful, python does lists by reference, so you get multiple
instantiations of the same list.

offsets = [ [] for i in range(pm) ]  # is much better!

----- Original Message -----
From: "gzeljko" <gzeljko at sezampro.yu>
Newsgroups: comp.lang.python
To: <python-list at python.org>
Sent: Tuesday, March 20, 2001 9:44 AM
Subject: Re: Sequence-spreading


>
> Alex Martelli <aleaxit at yahoo.com> wrote in message
> > offsets = [ [] for i in range(pm) ]
> > for i in range(len(arg)):
> >     offsets[i%pm].append(i)
> >
> >
> > Alex
> >
>
> Old syntax yet works:
>
> offsets = [[]]*pm
>
> ly-y'rs-gzeljko
>
>
> --
> http://mail.python.org/mailman/listinfo/python-list
>




From othello at javanet.com  Mon Mar 12 17:48:46 2001
From: othello at javanet.com (Raymond Hettinger)
Date: Mon, 12 Mar 2001 17:48:46 -0500
Subject: Help with Regular Expressions
Message-ID: <3AAD524E.746338DE@javanet.com>

Is there an idiom for how to use regular expressions for lexing?

My attempt below is unsatisfactory because it has to filter the
entire match group dictionary to find-out which token caused
the match. This approach isn't scalable because every token
match will require a loop over all possible token types.

I've fiddled with this one for hours and can't seem to find a
direct way get a group dictionary that contains only matches.

Many Thanks,

Raymond

from re import *

def tokenize(pattern, string):
    'Create an items() style list of token types and token values'
    pos = 0
    ans = []
    while 1:
        m = pattern.search(string, pos)
        if m == None: break
        pos = m.end()
        # the following gobblygook finds the first matched group name and value
        ans.append( filter( lambda (k,v): v ,m.groupdict().items())[0] )
    return ans

pgm = ' (a1->b & b->c) imp (a1->c)'
lexpat = compile(
r'(?P<op>(->)|(imp)|(&))|(?P<lpar>\()|(?P<rpar>\))|(?P<id>[A-Za-z]\w*)' )
print tokenize(lexpat, pgm)

# result is:
[('lpar', '('), ('id', 'a1'), ('op', '->'), ('id', 'b'), ('op', '&'), ('id',
'b'),
('op', '->'), ('id', 'c'), ('rpar', ')'), ('op', 'imp'), ('lpar', '('), ('id',
'a1'),
('op', '->'), ('id', 'c'), ('rpar', ')')]



From fjh at cs.mu.oz.au  Mon Mar 26 20:44:16 2001
From: fjh at cs.mu.oz.au (Fergus Henderson)
Date: 27 Mar 2001 01:44:16 GMT
Subject: CfP: Babel01 Workshop on Multi-Language Infrastructure and Interoperability
References: <3abf4ae1@msrnews.research.microsoft.com> <99o2n2$9on$1@216.39.170.247>
Message-ID: <99or9g$6tg$1@mulga.cs.mu.OZ.AU>

whisper at oz.net (Dave LeBlanc) writes:

>Gotta wonder why Microsoft is asking for papers to be submitted in a
>format it doesn't support.

"Microsoft" is not asking anything.  This is an academic workshop, not
a Microsoft workshop.  Two of the programme committee members, in
particular the organizer and the chair, are researchers who happen to
work at Microsoft Research in Cambridge.  But none of the other
programme committee members are Microsoft employees.  The "Call for
Papers" was posted on behalf of the whole programme committee.

>I bet this ends up all C-sharpish.

The intent of the workshop is to focus on language interoperability,
not on any single language or platform.  Papers will be judged on their
academic merit.

The content of the workshop will depend on what papers get submitted.
But we strongly encourage submissions on *all* areas that relate to
multi-language infrastructure and/or interoperability.  If you have
done something interesting with regard to that in the context of
a non-Microsoft-related language and/or platform, e.g. Linux, the GNU
compiler collection, Java, ANDF, etc., then please submit it!

In summary, the program committee is very strongly committed to
adhering to the highest academic standards, particularly with
regard to integrity and fairness in the peer review process.
We will do all that we can to avoid bias or the perception of bias.
The programme committee includes many reputable academics who
have considerable experience in organizing conferences and workshops.
BABEL'01 is the first in what we hope will be an ongoing series of
workshops/conferences devoted to language interoperability.
We will do our best to ensure that BABEL'01 upholds the highest
academic standards and forms the foundation of a reputable series
of refereed academic workshops/conferences.

-- 
Fergus Henderson <fjh at cs.mu.oz.au>  |  "I have always known that the pursuit
                                    |  of excellence is a lethal habit"
WWW: <http://www.cs.mu.oz.au/~fjh>  |     -- the last words of T. S. Garp.

P.S.  To ensure full transparency, I must point out that the research
group that I work in at the University of Melbourne has received
substantial funding from Microsoft.  However, this has not affected
my loyalties (at least in so far as I am conscious of them).
I remain a vocal advocate of the things I believe in, such as the
importance of academic integrity, and I am not afraid to publically
criticize those who fund us if criticism is due.

>On Mon, 26 Mar 2001 14:58:07 +0100, "Andrew Kennedy"
><akenn at microsoft.com> wrote:
>
>>			       BABEL'01
>>			   CALL FOR PAPERS
>>
>>		   First Workshop on Multi-Language
>>		 Infrastructure and Interoperability
>>
>>	   http://research.microsoft.com/~nick/babel01.htm
>>
>>			 Satellite to PLI'01
>>
>>		  Firenze, Italy, 8th September 2001
>>
>>
>>AIMS AND SCOPE
>>
>>Recent years have seen a resurgence of interest in multi-language
>>tools and intermediate languages, and in interoperability between
>>programs and components written in different programming
>>languages. Shared infrastructure such as code generators, analysis
>>tools and garbage collectors can greatly ease the task of producing a
>>high-quality implementation of a new programming language, whilst
>>being able to interoperate easily with code written in existing
>>languages is essential for such an implementation to be useful in
>>practice.  This workshop aims to bring together researchers and
>>developers working on multi-language integration. Contributions are
>>welcome on topics including, but not limited to:
>>
>>
>>* Compilation of high level languages to common executable formats
>>  such as Sun's Java Virtual Machine, the DRA's ANDF or Microsoft's
>>  .NET Common Language Runtime.
>>
>>* Defining and using bindings for component middleware such as OMG's
>>  CORBA or Microsoft's COM.
>>
>>* Language constructs to support interoperability between different
>>  languages, particularly from different paradigms (e.g. OO/functional).
>>
>>* Multi-language backends for compilation and/or analysis
>>  (e.g. MLRISC, FLINT, C--, TAL, SUIF).
>>
>>* Type theoretic and semantic foundations for multi-language work.
>>
>>* Multi-language development environments and tools (e.g. debuggers,
>>  profilers).
>>
>>Submissions may address any programming paradigm.  Experience papers
>>which describe the use of multiple languages in application
>>development are also welcome. Authors unsure of the appropriateness of
>>a potential submission should email the PC chair (nick at microsoft.com).
>>
>>
>>PROGRAMME COMMITTEE
>>
>>Nick Benton (chair)          Microsoft Research
>>Fergus Henderson             University of Melbourne
>>Andrew Kennedy (organiser)   Microsoft Research
>>Greg Morrisett               Cornell University
>>Martin Odersky               Ecole Polytechnique Federale de Lausanne
>>John Reppy                   Bell Labs
>>Andrew Tolmach               Portland State University
>>David Wakeling               University of Exeter
>>
>>
>>INVITED SPEAKERS
>>
>>Zhong Shao                   Yale University
>>
>>
>>IMPORTANT DATES
>>
>>Deadline for submission      1st June 2001
>>Notification of acceptance   9th July 2001
>>Final version due            10th August 2001
>>
>>
>>SUBMISSION DETAILS
>>
>>Papers should be at most 14 pages and should be submitted in
>>Ghostscript-compatible PostScript or PDF format and be printable on
>>both US letter and A4 paper.  Authors are strongly encouraged to use
>>ENTCS style files (see http://math.tulane.edu/~entcs/).  Papers should
>>be submitted electronically from the workshop website given above.
>>Submissions should not overlap significantly with work which has
>>already been published or submitted to any other conference or
>>journal.  An informal proceedings will be published as a technical
>>report and distributed at the workshop.  It is intended that a final
>>proceedings will be published in a volume of ENTCS.


From fredrik at pythonware.com  Tue Mar 20 02:20:31 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Tue, 20 Mar 2001 07:20:31 GMT
Subject: Why "from __future__" stinks; a counter-offer
References: <mailman.985051704.15160.python-list@python.org>
Message-ID: <3xDt6.2394$4N4.290991@newsc.telia.net>

Tim Peters wrote:
> Fine, so I'm an idiot

Me too!

Cheers /F




From xyzmats at laplaza.org  Fri Mar  2 08:43:20 2001
From: xyzmats at laplaza.org (Mats Wichmann)
Date: Fri, 02 Mar 2001 13:43:20 GMT
Subject: extended slices and buffer objects ????
Message-ID: <3a9f9581.5089898@news.laplaza.org>

Perhaps someone can shed light on my confusion about these two Python
features?

(1) Extended slicing.  I've seen claims that it's useful for
"multidimensional data structures such as matrices and arrays", but my
brain just isn't making sense of this. Can anyone give me a practical
example or two?

[2] Buffer objects.  What are buffer objects for, anyway? I find this
description from the Reference Manual somewhat less than enlightening:

Buffers are not directly supported by Python syntax, but can be
created by calling the builtin function buffer(). 


Thanx,

Mats Wichmann



From root at rainerdeyke.com  Sat Mar  3 13:14:37 2001
From: root at rainerdeyke.com (Rainer Deyke)
Date: Sat, 03 Mar 2001 18:14:37 GMT
Subject: Adding a method to an instance
References: <G2%n6.360714$JT5.11606234@news20.bellglobal.com>
Message-ID: <hwao6.335854$ge4.119442611@news2.rdc2.tx.home.com>

"Bryan Mongeau" <bryan at eevolved.com> wrote in message
news:G2%n6.360714$JT5.11606234 at news20.bellglobal.com...
> Quick one for you guys:
>
> I would like to add a method to an instantiated class.
>
> class foo:
>   def __init__(self):
>     self.bar = 123
>
> f = foo()
>
> Now to add a method to the instance f of class foo:
>
> def newMethod(self):
>   print self.bar

In this particular case:

f = foo()

def newMethod(self=f):
  print self.bar

f.newMethod = newMethod


In general:

def add_method(instance, method, name):
  # We want nested scopes!  We want nested scopes!  We want nested scopes
  # In the absence of nested scopes, there is no easy way to pass on
  # non-keyword args to 'method'.
  def call_method(instance=instance, method=method, **kwargs):
    method(instance, **kwargs)
  setattr(instance, name, call_method)

add_method(f, newMethod, 'newMethod')


--
Rainer Deyke (root at rainerdeyke.com)
Shareware computer games           -           http://rainerdeyke.com
"In ihren Reihen zu stehen heisst unter Feinden zu kaempfen" - Abigor




From aleaxit at yahoo.com  Mon Mar 26 11:22:51 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Mon, 26 Mar 2001 18:22:51 +0200
Subject: portable string localization and resource bundles...
References: <mailman.985605333.14558.python-list@python.org>
Message-ID: <99nqcl0mg@news2.newsguy.com>

"Clark C. Evans" <cce at clarkevans.com> wrote in message
news:mailman.985605333.14558.python-list at python.org...
> I was wondering if there is a solution to the
> common problem of string localization
> (menu and message box titles, for instance).

The gettext module in the Python library helps.


Alex





From ^NOSPAM^ifi at ifi.it  Wed Mar  7 05:45:50 2001
From: ^NOSPAM^ifi at ifi.it (accattatevillo)
Date: Wed, 07 Mar 2001 11:45:50 +0100
Subject: Using a jar with jython
Message-ID: <3AA6115E.7B2ACDEE@ifi.it>

Hi.

I'm trying to convert a java application to jython.

I use the IBM java toolbox JT400.jar to access data on a AS/400 bat I'm
not able to do the same thing with jython.

With java I added jt400.jar to the CLASSPATH but in jython???
Anyone could help me?

TIA
Stefano


From stephen_purcell at yahoo.com  Sun Mar  4 10:20:08 2001
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Sun, 4 Mar 2001 16:20:08 +0100
Subject: Java guy interested in Python
In-Reply-To: <003801c0a459$a29c97f0$63da1542@wisdom>; from kknbk2@bellsouth.net on Sat, Mar 03, 2001 at 09:17:24PM -0600
References: <003801c0a459$a29c97f0$63da1542@wisdom>
Message-ID: <20010304162008.C2779@freedom.puma-ag.com>

Kyle Wayne Kelly wrote:
> Can anyone tell me some of the things that Python offers which are not
> available in Java?  I like Python's syntax, but what can Python do that
> Java can not do?  Any comments are helpful.

Just try it and see.

The 2 hours that you might spend trying to solicit and evaluate responses
from those on this list would already be enough to get a feel for Python!

You can whiz through the tutorial, then write a toy program or browse through
some existing code. (e.g. standard library modules)

(Hint: don't expect to choose between Python and other languages based on
'features' -- you won't find legibility, sanity and brevity mentioned
on any Java feature list.)

Have fun,

-Steve

-- 
Steve Purcell, Pythangelist
Get testing at http://pyunit.sourceforge.net/
Get servlets at http://pyserv.sourceforge.net/
"Even snakes are afraid of snakes." -- Steven Wright



From dwig at advancedmp.net  Wed Mar 14 21:46:06 2001
From: dwig at advancedmp.net (Don Dwiggins)
Date: 14 Mar 2001 18:46:06 -0800
Subject: site.py on Windows
In-Reply-To: "Syver Enstad"'s message of "Wed, 14 Mar 2001 23:05:39 +0100"
References: <%tKr6.2570$pX.26174@news1.onlynews.com>
	<98oq3r$1e0$1@troll.powertech.no>
Message-ID: <u3dcfagwh.fsf_-_@advancedmp.net>

I'm trying to use the site-packages facility, as documented for site.py, in
my ActiveState distribution (2.03, I think).  After some investigation, I
found that the site.py code essentially punts if the path separator
character isn't '/'.  After seeing that, I also noticed in the fine print of
the documentation: "Note that for some non-Unix systems, sys.prefix and
sys.exec_prefix are empty, and the path manipulations are skipped; however
the import of sitecustomize is still attempted."  Doh!

OK, has anyone fixed site.py to do the right stuff under Windows?  (I'd
rather not resort to writing a sitecustomize.)  If not, I'm willing to have
a go at it, and submit what I come up with.  It doesn't look particularly
difficult, but if there are subtle issues I'm not seeing, please let me
know.

-- 
Don Dwiggins                    "Solvitur Ambulando"
Advanced MP Technology
dwig at advancedmp.net




From qrczak at knm.org.pl  Fri Mar  9 03:50:38 2001
From: qrczak at knm.org.pl (Marcin 'Qrczak' Kowalczyk)
Date: 9 Mar 2001 08:50:38 GMT
Subject: from Tkinter import *
References: <mailman.983920449.5230.python-list@python.org> <slrn9ab0g7.s6.matt@happy-hour.mondoinfo.com> <989dot$t6t$1@panix3.panix.com> <uae6vk3up.fsf@ctwd0143.fitlinxx.com>
Message-ID: <slrn9ah6au.k09.qrczak@qrnik.zagroda>

08 Mar 2001 22:39:42 -0500, David Bolen <db3l at fitlinxx.com> pisze:

> the one place I still see a use for "import *" even in production
> code is large wrapped C libraries that already name everything
> according to a flat namespace (for example, staying in the GUI
> vein, wxPython).

I would prefer to remove these prefixes while porting to a language
with a module system.

-- 
 __("<  Marcin Kowalczyk * qrczak at knm.org.pl http://qrczak.ids.net.pl/
 \__/
  ^^                      SYGNATURA ZAST?PCZA
QRCZAK


From parz at home.com  Thu Mar  1 01:46:38 2001
From: parz at home.com (Parzival Herzog)
Date: Thu, 01 Mar 2001 06:46:38 GMT
Subject: "import X" and "from X import y" produce different results, Why?
Message-ID: <ifmn6.6635$hn5.857862@news1.rdc1.mb.home.com>

In the Essential Python Reference, it states (p. 72):

"The from statement is identical to import, e3xcept that instead of creating
a
name referring to the newly created module namespace, references to one or
more of the objects defined in the module are placed into the current
namespace."

In the following example, apperently the abovve claim is not true:
----------
# xx.py
X = 35
----------
# yy.py
from xx import X
import xx

print "BEFORE, X =", X
print "BEFORE, xx.X =", xx.X
X = 99
print "AFTER, X =", X
print "AFTER, xx.X =", xx.X
----------

Running yy.py produces the result:

BEFORE, X = 35
BEFORE, xx.X = 35
AFTER, X = 99
AFTER, xx.X = 35

So, apparently the imported X is a copy of xx.X, not the same object by
another name.
Can anyone explain this result? Is this result justified by a Python
language specifcation?

- Parzival





From lac at cd.chalmers.se  Tue Mar  6 10:57:31 2001
From: lac at cd.chalmers.se (Laura Creighton)
Date: Tue, 6 Mar 2001 16:57:31 +0100 (MET)
Subject: Can anyone recomend a good intoduction to C...
Message-ID: <200103061557.QAA26974@boris.cd.chalmers.se>

Werner Schiendl <ws-news at gmx.at> writes:
>If you have a good reason to choose C over C++, take the K&R book the others
>recommend.
>
>Otherwise you should consider going straight to C++.
>This will save you hours of headache with C memory stuff, string handling,
>and so on.

This leaves the novice with the idea that C programmers, by and large
all became happy C++ programmers and are pleased as punch with the
new language.  The reason to learn C is to eventually learn C++.
This is misleading.

There are a very large number of people who think that C
is a terrific language; extremely good at what it does, and well worth
studying, in part to understand what it is that a language designer
does when he *gets* *it* *right*, while also believing that studying C++ is 
of particular use when you want to see what happens when a language 
designer *gets* *it* *wrong*.  Or when forced to do so at gunpoint.  

Laura Creighton



From kens at sightreader.com  Fri Mar 16 15:31:09 2001
From: kens at sightreader.com (Ken Seehof)
Date: Fri, 16 Mar 2001 14:31:09 -0600
Subject: Cookies
References: <20010316220554.18816.qmail@web9307.mail.yahoo.com>
Message-ID: <009401c0ae58$095e8b80$04090a0a@upcast.com>

That's true, if I am willing to make the user do stuff for me, my problem
is solved.  The whole idea is to minimize user interaction, but apparently
there is no way to leave the user out of the loop.  I think I'll just
display
a serial number for the user to type in when the python program asks for
it.  The users can use copy/paste if they want to.

Thanks,
- Ken

----- Original Message -----
From: "Greg Jorgensen" <gregjor at yahoo.com>
To: "Ken Seehof" <kens at sightreader.com>
Cc: "python-list" <python-list at python.org>
Sent: Friday, March 16, 2001 4:05 PM
Subject: Re: Cookies


> What about copy/paste? Or tell the user to save the file on their disk,
then
> let your Python script dig the ID out of it later. The cookie thing could
be
> made to work but you are right about the OS and browser incompatibilities.
>
> Good luck,
>
> Greg Jorgensen
> gregj at pobox.com
>
>
> --- Ken Seehof <kens at sightreader.com> wrote:
> > Yeah, that's the closest I've come up with too, but the difficulty is
making
> > the system reliably support all browsers.  Unfortunately from the point
of
> > view of a standalone application, there is no standard for cookies.
Cookies
> > are standardized from the point of view of servers and javascript, but
that
> > doesn't help me with a standalone app.
> >
> > The best approach so far seems to be to have the server send back a page
> > with a 10 digit unique identifier in a large font.  The python app would
> > then ask the user to type in the number manually.  Seems kinda dumb that
I
> > need the users fingers as part of my protocol. :-)
> >
> > - Ken
> >
> > ----- Original Message -----
> > From: "Greg Jorgensen" <gregj at pobox.com>
> > Newsgroups: comp.lang.python
> > To: <python-list at python.org>
> > Sent: Thursday, March 15, 2001 11:35 PM
> > Subject: Re: Cookies
> >
> >
> > > A cookie is just a text file stored in a special place where the
browser
> > can
> > > find it. Where exactly the cookies go depends on your OS and browser.
> > >
> > > Using a text file to store the key would seem the easiest solution,
but
> > > client-side Javascript can't create or write a file.
> > >
> > > I think you should generate unique IDs on the server, not on the
client.
> > > Here's one way to handle it:
> > >
> > > 1. Client browser sends form data to server.
> > > 2. Server processes form, generates a unique ID, stores the form data
> > keyed
> > > by the unique ID.
> > > 3. Server replies with a "thank you" or "database updated" page that
also
> > > writes a cookie.
> > > 4. Python program starts up and looks for the cookie. Depending on
your OS
> > > and browser the location and name of the cookie will vary, but with a
> > little
> > > experimenting you can easily find the most likely locations.
> > > 5. Python program reads the cookie. Open your cookie with a text
editor to
> > > see what's in there--it's just name: value pairs.
> > > 6. Python program sends HTTP request to server with unique ID it got
from
> > > the cookie.
> > >
> > > Obviously this won't work if the client has disabled cookies. With a
> > little
> > > more trouble you can detect this and perhaps put the unique ID on the
> > > response page and tell the user to save it or something. Or embed the
> > unique
> > > ID in the reply page and tell the user to save that page and give it a
> > > specific name, then have your Python program open that text file and
find
> > > the unique ID in there.
> > >
> > > Depending on your requirements and security concerns, you can either
have
> > > your database program generate the unique ID (my preference), use the
> > > Windows GUID functions (which are available to ASP on the server), or
> > > implement one of the GUID functions you can easily find on the web.
Many
> > > programmers have solved this in one way or another; send email to me
if
> > you
> > > want specific advice on generating unique IDs from your database
engine.
> > >
> > > Good luck.
> > >
> > > <<<
> > > "Ken Seehof" <kens at sightreader.com> wrote in message
> > > news:mailman.984707839.22839.python-list at python.org...
> > > Short version:
> > >
> > > Is it possible for a stand-alone python application to read a cookie
> > without
> > > the help of a browser?"
> > >
> > > Long version:
> > >
> > > I have a web page W and a client side python script P.  I want W to
> > generate
> > > a GUID (perhaps using javascript) which is then sent to a server S,
where
> > it
> > > is to be used as a key in a database.  Now I run P (on the client, and
not
> > > in a browser) and I want P to somehow find the value of the GUID, so
that
> > it
> > > can query the database on S.
> > >
> > > Alternate question:
> > >
> > > Is there a technique common to both javascript and python that will
obtain
> > a
> > > unique identifier for a computer (e.g. IP address).  It must not be
> > browser
> > > specific.
> > >
> > > Motivation:
> > >
> > > I need to write a record of data from an html form, keyed by some kind
of
> > > unique client id, to the server and then be able to query that record
from
> > a
> > > stand alone python script running on the client.  The missing link is
> > > getting the html form and the python script to agree on a value for
the
> > id;
> > > the rest is easy.
> > > >>>
> > >
> > > --
> > > Greg Jorgensen
> > > PDXperts
> > > Portland, Oregon, USA
> > > gregj at pobox.com
> > >
> > >
> > > --
> > > http://mail.python.org/mailman/listinfo/python-list
> > >
> >
> >
>
>
> __________________________________________________
> Do You Yahoo!?
> Get email at your own domain with Yahoo! Mail.
> http://personal.mail.yahoo.com/
>




From phd at phd.pp.ru  Fri Mar  9 07:06:36 2001
From: phd at phd.pp.ru (Oleg Broytmann)
Date: Fri, 9 Mar 2001 15:06:36 +0300 (MSK)
Subject: virtual hosting in Zope
In-Reply-To: <987lbo$119t$1@infocom.km.ua>
Message-ID: <Pine.LNX.4.30.0103091502530.23299-100000@fep132.fep.ru>

On Thu, 8 Mar 2001, Anton A. Zotov wrote:
> Urgent: I'd appreciate a person who would tell me how to implement virtual
> hosting under Zope/Medusa environment

   Russian-speaking Python and Zope User Group is at your service. Please
visit us at zope.net.ru and subscribe to our mailing lists:
   http://zope.net.ru/About/mail-list/
The very zope.net.ru had implemented and reimplemented virtual hosting in
many different way. Just ask.
   (I'll be available 11 March, but other people may answer your quetsions
in the list)

Oleg.
----
     Oleg Broytmann            http://phd.pp.ru/            phd at phd.pp.ru
           Programmers don't die, they just GOSUB without RETURN.





From max at alcyone.com  Sun Mar 11 05:17:00 2001
From: max at alcyone.com (Erik Max Francis)
Date: Sun, 11 Mar 2001 02:17:00 -0800
Subject: Can anyone recomend a good intoduction to C...
References: <mailman.983894288.19965.python-list@python.org> <ewwp6.1466$y6.241005@ruti.visi.com> <slrn9aeblb.cg1.scarblac@pino.selwerd.nl> <djagat4kk3cs9hg0c3k39cudk6goblhkg6@4ax.com> <989mbk$24nuv$1@fido.engr.sgi.com> <slrn9ah66m.k09.qrczak@qrnik.zagroda> <98auhn0270j@news2.newsguy.com> <slrn9aiaq7.rmu.qrczak@qrnik.zagroda> <98dpoe12ckn@news1.newsguy.com> <3AAA85C1.81FB90AA@shore.net> <98eb1f02uqq@news1.newsguy.com> <3AAAE0BB.CBCC99EC@alcyone.com> <slrn9amfea.u2j.qrczak@qrnik.zagroda>
Message-ID: <3AAB509C.5FF0F19F@alcyone.com>

Marcin 'Qrczak' Kowalczyk wrote:

> What about the 'export' keyword?

Sigh.  Almost all.  Nothing's perfect, obviously.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ Success and failure are equally disastrous.
\__/ Tennessee Williams
    REALpolitik / http://www.realpolitik.com/
 Get your own customized newsfeed online in realtime ... for free!


From sholden at holdenweb.com  Mon Mar 26 17:15:12 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Mon, 26 Mar 2001 17:15:12 -0500
Subject: Why Python/Jython?
References: <99cl9e$b2b$1@Merlin.et.bocholt.fh-gelsenkirchen.de> <mailman.985260089.28444.python-list@python.org> <srzu6.12868$_46.475705@e420r-atl2.usenetserver.com> <3abc8d36.83155745@24.0.228.33>
Message-ID: <3fPv6.1460$jh3.126995@e420r-atl2.usenetserver.com>

<support @ internetdiscovery.com (Mike Clarkson)> wrote in message
news:3abc8d36.83155745 at 24.0.228.33...
> On Thu, 22 Mar 2001 22:26:59 -0500, "Steve Holden"
> <sholden at holdenweb.com> wrote:
>
> >"Steve Purcell" <stephen_purcell at yahoo.com> wrote in message
> >> Programmer performance is usually gauged by coding time and
> >> number of bugs, both of which are vaguely proportional to the number of
> >> lines of code in most sane languages*.
> >>
> >> * Perl is a noteworthy exception
> >>
> >I saw no inconsistency here. Why would Perl be an exception to your rule?
>
> It's not a sane language :-)
>
> It's like vicegrips: the wrong tool for every job.
>
Interesting. I thought Perl wasn't an exception precisely because your
original wording clearly didn't apply to it (because it's not a sane
language).

not-wishing-to-get-tied-up-in-semantics-ly y'rs  - steve





From grante at visi.com  Fri Mar 23 10:25:55 2001
From: grante at visi.com (Grant Edwards)
Date: Fri, 23 Mar 2001 15:25:55 GMT
Subject: newbe   how to read in a string?
References: <3ABB4E70.54885C9F@noos.fr> <99fo39$gsq$2@news.mathworks.com>
Message-ID: <7WJu6.1993$SB2.256465@ruti.visi.com>

In article <99fo39$gsq$2 at news.mathworks.com>, Joshua Marshall wrote:

>> I whant to extract the 87.1. In C i would code by a
>> scanf("%s,%f,%f,%f\n"),...
>
>> How can i do this in python?
>
>I don't think there's any generic way.  It would be nice to see
>something similar to scanf in Python.

While sscanf() is handy, scanf() is considered harmful...

-- 
Grant Edwards                   grante             Yow!
                                  at               TAILFINS!!... click...
                               visi.com            


From rcameszREMOVETHIS at dds.removethistoo.nl  Sat Mar 31 12:10:30 2001
From: rcameszREMOVETHIS at dds.removethistoo.nl (Robert Amesz)
Date: Sat, 31 Mar 2001 17:10:30 GMT
Subject: HTML Tidy and Python wrapper
References: <985945106.28359.0.nnrp-02.c1c3e154@news.demon.co.uk>
Message-ID: <Xns9075C2ED17AE0rcamesz@127.0.0.1>

Paul Brian wrote:


> I was sure that I recently read that there was a python wrapper
> around the HTML Tidy (http://www.w3.org/People/Raggett/tidy/).
> Unfortunately I cannot find the reference. Does any one know if
> such a thing exists, as simple searches through deja have come up
> nought. 

I don't know about any python wrapper, but I *do* know there's a COM 
version, named - you guessed it - TidyCOM. It's at:

    http://perso.wanadoo.fr/ablavier/TidyCOM/index.html


If you properly registered the COM-component with your system, the 
following snippet should work ('content' should contain the HTML-page 
in full, the cleaned-up version will be stored in 'newcontent'):

--------------------------------------------------------
import win32com.client

tidyobj = win32com.client.Dispatch("TidyCOM.TidyObject")

# Tidy COM call
newcontent =  tidyobj.TidyMemToMem(content)
--------------------------------------------------------


This is only very basic, of course, you really should read the 
documention to learn about things like how TidyCOM reports errors.


Robert Amesz


From dwig at advancedmp.net  Thu Mar 22 13:44:52 2001
From: dwig at advancedmp.net (Don Dwiggins)
Date: 22 Mar 2001 10:44:52 -0800
Subject: Python and emacs
In-Reply-To: John McMonagle's message of "Thu, 22 Mar 2001 11:13:31 +1000"
Message-ID: <upuf9k5i3.fsf@advancedmp.net>

A semi-idle question: has anyone looked into embedding Python as an
extension language in Emacs, alongside Elisp?  I'm toying with the idea of
doing some Emacs enhancements/modifications, and given the choice, I'd much
rather do them in Python.  (One such enhancement would be to add
Pythonwin-style "intellisense" popups to python-mode.)

-- 
Don Dwiggins                    "Solvitur Ambulando"
Advanced MP Technology
dwig at advancedmp.net




From cribeiro at mail.inet.com.br  Fri Mar 30 06:30:23 2001
From: cribeiro at mail.inet.com.br (Carlos Alberto Reis Ribeiro)
Date: Fri, 30 Mar 2001 08:30:23 -0300
Subject: NumPy loop efficiency
In-Reply-To: <m3itkr7drt.fsf@chinon.cnrs-orleans.fr>
References: <3ac44e27$0$16692$73bec57d@news.be.uu.net>
Message-ID: <5.0.2.1.0.20010330082234.0269a0e0@mail.inet.com.br>

>It does create a temporary array. Getting rid of this would be a lot
>of work, array expressions would have to use lazy evaluation.

(out of the top of my mind; I dont have Numeric installed on my home 
computer, and have no way to test it right now)

The intermediate array is not really needed, but the python expresion 
evaluator have no way to know it. However, something like the ufuncs could 
be used. For instance, something like this could be done (the names are 
just examples):

   z = arrayadd(a, b, c)

or the more generic solution:

   z = arraymap(add, a, b, c)

The first one could be optimized for the most common cases (I'm sure adding 
arrays is one of these operations).


Carlos Ribeiro





From mal at lemburg.com  Mon Mar 19 12:35:21 2001
From: mal at lemburg.com (M.-A. Lemburg)
Date: Mon, 19 Mar 2001 18:35:21 +0100
Subject: [ANNOUNCE] More Python tamers (Europeen Python Meeting in Bordeaux)
References: <mailman.985020805.14821.python-list@python.org>
Message-ID: <3AB64359.E5B099@lemburg.com>

Marc Poinot wrote:
> 
> Hi you all,
> 
> please heck again http://lsm.abul.org
> where we have now at least three more Python
> tamers for our European Python Meeting:
> 
> Andy Robinson (UK) will talk about ReportLab
> Christian Tismer, Dinu Gherman (DE) will talk
> about stackless python.
> And maybe two other Python tamers soon...

I think you forgot to mention the most interesting URL for this
target group ;-)

    http://lsm.abul.org/program/topic19/topic19.php3

    "Europeen Python Meeting, Bordeaux, 2001-07-04 - 07-09"

-- 
Marc-Andre Lemburg
______________________________________________________________________
Company & Consulting:                           http://www.egenix.com/
Python Pages:                           http://www.lemburg.com/python/



From jkraska1 at san.rr.com  Sat Mar 24 00:12:15 2001
From: jkraska1 at san.rr.com (Courageous)
Date: Sat, 24 Mar 2001 05:12:15 GMT
Subject: Good Editor? (What's the big deal with syntax colouring?)
References: <mailman.985291829.27633.python-list@python.org> <99dqqo$gpi$1@news.mathworks.com> <99e9sv$usmp$1@ID-11957.news.dfncis.de> <99emg2$qif@gap.cco.caltech.edu> <ha3obts0kp05lonh3o25fsesr38bouo839@4ax.com> <slrn9bo559.48h.grey@teleute.dmiyu.org>
Message-ID: <0bbobt0c50nnftjf6amjnhhmfbah69lorq@4ax.com>

>    Don't do much coding in the dark, do you?

Actually, no. I always turn the lights on. I hate dark rooms. They
depress me.

> Paper is reflective, a monitor is projective.

While indeed what you say is true, it's not all that relevant. For example,
outdoors, in the sun, paper can be quite blinding. I'd say that it's the
intensity of the light coming from the surface without regard to its source
that really matters.

Personally, my monitor doesn't blind me. It's very sharp (a ViewSonic 17
original, analogous to the professional series), and is quite a bit better
than reading a naked piece of paper under direct sunlight, let me assure
you.

:)

C//



From loewis at informatik.hu-berlin.de  Mon Mar 26 03:41:28 2001
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 26 Mar 2001 10:41:28 +0200
Subject: PEP 245: Python interfaces
References: <Pine.LNX.4.32.0103222253510.26302-100000@localhost.localdomain> <j4hf0jz36a.fsf@informatik.hu-berlin.de> <Pine.LNX.4.32.0103241227160.26302-100000@localhost.localdomain>
Message-ID: <j44rwgc47b.fsf@informatik.hu-berlin.de>

Michel Pelletier <michel at digicool.com> writes:

> > Even though this is quite a long document, it is incomplete.
> 
> Yes, *sigh*.  I know I need more formalism, but should this be at the
> expense of the section on goals?  I'm not sure.  I was raised in an XP
> environment, the goals are the most important part to me.  This is
> something I've been debating myself over.  I obviously want the PEP to be
> accessable and not boring, but I also want it to be comprehensive.

I agree with Tim's comments on this issue. I personally would prefer
it to be complete but boring. There should be a clear rationale of
what this wants to achieve, and a specification of how it achieves
that. Examples why it is useful to have that feature, and a tutorial
on using it are both optional. Readers are then expected to evaluate
whether they like the rationale, and whether they think the
specification fulfills the goals in the best possible way.

> > - packaging. I think the hierarchy is too deep, everything should be
> >   contained in the interfaces module (there is not an awful lot of
> >   stuff in it, is there?).
> 
> How do you mean?  I don't think the PEP proposes any depth beyond the
> Interface module.

It mentions an Interface.Exceptions module. The reference
implementation also has Interface.InterfaceBase, Interface.Method,
Interface.Standard and other modules.

> Ah yes, this is an interesting point of debate.  The idea here is
> that the interface of an interface is *not* the interface that it
> describes.  I was not around when the types-sig (and Jim) came up
> with this; I cannot give you the exact motivation, but it feels
> right to me.  If you'd like, I can add a section to 'Dissenting
> Opinions' section on this point.

I agree that an interface of an interface is not the interface that it
describes. That does not rule out using attribute access syntax to
navigate through the interface description. Suppose you have

interface Foo:
  def bar(self):
    "Hello"

class Foo_impl implements Foo:
  def bar(self):
    return "Hello"

foo = Foo_impl()

then foo.bar() is usage of the interface being described; foo.bar is a
bound method. I'm not requesting that Foo.bar is a method (bound or
unbound); I'm requesting that it returns a method description.

Regards,
Martin


From wayne.izatt at _nospam_myself.com  Fri Mar 23 00:46:07 2001
From: wayne.izatt at _nospam_myself.com (Wayne Izatt)
Date: Fri, 23 Mar 2001 05:46:07 GMT
Subject: Sockets in Python
References: <99dcpm$am1$1@wanadoo.fr>
Message-ID: <zqBu6.1268$NE.131211@news1.crdva1.bc.home.com>

> here is my problem : I'd like to create a simple network card game with
> Python,

sound like fun

> but I've got some troubles with the sockets...

ok

> In fact I'd like to find some kind of tutorial about this subject, because
> although
> my first-time code quite works, I don't think it is really good.
>
> When I try all the samples I've found on the 'net about sockets, the
"main"
> loop for communicating, generally implemented with a While 1: ..., makes
my
> computer frozen up (under Windows, not under Linux).
> That's why I thought it would be better to use threads, for the listening
> function
> for example, but it doesn't work well under Windows (sometimes it hangs
up,
> I can't figure out why !).

nor can I, since I can't see your code

> I'm sure it's possible creating a robust and nice app which uses the
> sockets encapsulation provided with Python under Windows, but how ?

Python sockets are the easiest I've used. However, I think it can be useful
to
know a little bit more about socket programming than what one normally gets
from a Python or Perl or whatever tutorial. Even reading through some of the
C-based tutorials will help (if not, at least it'll make you grateful you
get to use the
Python socket API rather than the raw C API). I'm not saying you have to
digest Stephens completely, but I've seen aome pretty good intros for linux
programmers. If you can't find one, let me know and I'll dig a little.

> What are the "good" things to do ?

Whatever gets the job done. Socket programming is easy and difficult - easy
to get something up and running, difficult to get working perfectly.

> What is the best module to use between socket or asyncore ?

asyncore allows you to avoid threads. I don't know why anyone would want
to avoid threads, especially Python threads.

> Well, if someone can help me with the sockets in general and the sockets
> with Python, it would be greatly appreciated...
> (I've already read the "classic" tutorials about sockets, I followed the
> links
> on the Python website, etc..)

As I said, I think the C tuts are pretty good (just so you know what's going
on
under the hood). The socket programming chapters in the 2nd edition of
Programming Python (Lutz) are particularly excellent - I wish I'd had them
years ago. Look in the Google directory for good, oatsy socket stuff. Post
the
problem sections of your code, or send them and I'll take a look.

cheers

>
> Thanks in advance.
>
>
>
>
>
>
>




From sholden at holdenweb.com  Mon Mar  5 20:42:37 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Mon, 5 Mar 2001 20:42:37 -0500
Subject: Suffering For Your Art
References: <n2rl6.1686$v5.5388@newsfeeds.bigpond.com> <5iRo6.44561$o%6.860997@news1.rdc1.bc.home.com>
Message-ID: <MjXo6.19373$ML1.964362@e420r-atl2.usenetserver.com>

"Dethe Elza" <delza at antarcti.ca> wrote in message
news:5iRo6.44561$o%6.860997 at news1.rdc1.bc.home.com...
> But just think, if computer science was taught with python, the 4-year
> curriculum would only take one semester, then all the students would drop
> out to start writing code and making money.  C++ was designed to make
things
> hard to learn to keep comp sci instructors gainfully employed and off the
> streets.
>
> --Dethe
>
Oh well, in that case I'm (suddenly) all in favor of C++. Never did like the
streets that much.

Python-training-not-being-a-major-revenue-earner-here-ly y'rs  - steve





From fredrik at pythonware.com  Wed Mar 28 10:37:33 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 28 Mar 2001 15:37:33 GMT
Subject: Speed problems with Python vs. Perl
References: <99sv33$8hk$1@news.cineca.it>
Message-ID: <1znw6.1969$sk3.665386@newsb.telia.net>

Georg Umgiesser wrote:

> ----------------------------------------------------------
>
> #!/usr/bin/python
>
> import sys
> import re
>
> whitespace = re.compile("\s+")
>
> def main():
>
>     icount = 0
>     for line in sys.stdin.readlines():
>         icount = icount + 1
>         f = whitespace.split(line)
>     print "Total lines read: " + `icount`
>
> if __name__ == '__main__':
>     main()
>
> --------------------------------------------------------

assuming you're using Python 2.1, the following version is
about 16 times faster on my box:

import sys

def main():

    icount = 0
    for line in sys.stdin.xreadlines():
        icount += 1
        f = line.split()
    print "Total lines read", icount

if __name__ == '__main__':
    main()

for a backwards-compatible version of xreadlines, use
the double-loop pattern from:

http://effbot.org/guides/readline-performance.htm

Cheers /F




From dick.streefland at tasking.com  Fri Mar 16 10:49:05 2001
From: dick.streefland at tasking.com (Dick Streefland)
Date: 16 Mar 2001 15:49:05 GMT
Subject: Python style questions
References: <98tb91$9t9$1@saltmine.radix.net>
Message-ID: <98tclh$ap6$2@lahti.tasking.nl>

cobrien at Radix.Net (Cary O'Brien) wrote:
| 4. I need to change things like "88aa99bb" (hex string) into integers.  Should I
| 
|    a. use expr("0x"+s) 
|    b. not use such things
|    c. ???

c. use string.atoi(s, 16), or for Python >= 2.0, the builtin int(s, 16).

-- 
Dick Streefland                      ////            TASKING Software BV
dick.streefland at tasking.com         (@ @)         http://www.tasking.com
--------------------------------oOO--(_)--OOo---------------------------


From jeremy at alum.mit.edu  Thu Mar  1 11:43:18 2001
From: jeremy at alum.mit.edu (Jeremy Hylton)
Date: Thu, 1 Mar 2001 11:43:18 -0500 (EST)
Subject: d = {}; d[0:1] = 1; d[0:1] = 2; print d[:]
In-Reply-To: <97kotr$su9$0@216.39.151.169>
References: <mailman.983419142.24027.python-list@python.org>
	<97kotr$su9$0@216.39.151.169>
Message-ID: <15006.31782.195212.326426@w221.z064000254.bwi-md.dsl.cnc.net>

Looks good to me.

Jeremy



From jfontain at free.fr  Sun Mar 18 14:58:34 2001
From: jfontain at free.fr (Jean-Luc Fontaine)
Date: Sun, 18 Mar 2001 19:58:34 GMT
Subject: ANNOUNCE: moodss-14.1
Message-ID: <3AB5134E.1B1A84DC@free.fr>

Hi everybody: here is a new version of moodss.


Check it out! (or at least the screenshots at
http://jfontain.free.fr/moodss3.gif or
http://jfontain.free.fr/moodss4.gif :).


Notes:
- moodss modules can now be written in Python! Sample randpy and
minipy modules are provided. Documentation (see randpy.py file)
needs improving. Feedback is highly welcomed.
(modules can also be written in Perl, and Tcl of course)
- please let me know if you are interested in writing modules in
Ruby (I'm on a roll :-)

### CHANGES ###

--- version 14.1 ---
allow multiple concurrent Python modules (tclpython or tclpython2
  version 2.0.1 required)
implemented dynamic loading of Python modules
module programmers: module data display was updated even if updates
  counter was not increased, whereas now that condition is actually
  checked
cpustats module could fail with a division by zero error on repetitive
  updates
MySQL myvars module updated to support 3.23.34 and 3.23.35 servers

### README ###

This is moodss (Modular Object Oriented Dynamic SpreadSheet) version
14.1.

Moodss won in the Best System Admin Technology category (Tcl Tips and
Tricks, Valuable Real World Programming Examples) at the O'Reilly
Tcl/Tk 1999 Conference.
Linux Magazine calls it a "lifesaver".
Tucows gives it 5 stars (cows or penguins :-).

Moodss is a modular application. It displays data described and
updated in one or more modules, which can be specified in the command
line or dynamically loaded or unloaded while the application is
running. Data is originally displayed in tables. Graphical views
(graph, bar, 3D pie charts, ...), summary tables (with current,
average, minimum and maximum values) and free text viewers can be
created from any number of table cells, originating from any of the
displayed viewers. Thresholds can be set on any number of cells.

Specific modules can easily be developed in the Tcl, Perl and Python
scripting languages or in C.

A thorough and intuitive drag'n'drop scheme is used for most viewer
editing tasks: creation, modification, type mutation, destruction,
... and thresholds creation. Table rows can be sorted in increasing or
decreasing order by clicking on column titles. The current
configuration (modules, tables and viewers geometry, ...) can be saved
in a file at any time, and later reused through a command line switch,
thus achieving a dashboard functionality.

The module code is the link between the moodss core and the data to be
displayed. All the specific code is kept in the module package. Since
module data access is entirely customizable (through C code, Tcl,
Perl, Python, HTTP, ...) and since several modules can be loaded at
once, applications for moodss become limitless.
For example, thoroughly monitor a dynamic web server on a single
dashboard with graphs, using the Apache, MySQL, cpustats, memstats,
...  modules. If you have replicated servers, dynamically add them to
your view, even load the snmp module on the fly and let your
imagination take over...

Along with a core trace module, random, ps, cpustats, memstats,
diskstats, mounts, route, arp, kernmods, netdev, pci, system, MySQL
(myquery, mystatus, myprocs, myvars) modules for Linux, ping, snmp and
snmptrap for UNIX, apache and apachex modules are included (running
"wish moodss ps cpustats memstats" mimics the "top" application with a
graphic edge and remote monitoring capability).

Thorough help is provided through menus, widget tips, a message area,
a module help window and a global help window with a complete HTML
documentation.

Moodss is multi-langual thanks to Tcl internationalization
capabilities. So far only English and partially French are
supported. Help with other languages will be very warmly welcomed.

Development of moodss is continuing and as more features are added in
future versions, backward module code compatibility will be maintained.

I cannot thank the authors of the tkTable, BLT, MIME/SMTP and the HTML
libraries enough for their great work.

In order to run moodss, you need to install the following packages
(unless you can use the rpm utility, see below):
Tcl/Tk 8.3.1 or above, at (or at a mirror near you)
  http://dev.scriptics.com/ or ftp://ftp.scriptics.com/
the latest tkTable widget library at:
  http://www.hobbs.wservice.com/tcl/main.html
the latest BLT library at:
  ftp://tcltk.sourceforge.net/pub/tcltk/blt/
eventually the latest tclperl library for writing modules in Perl, or
the latest tclpython library for writing modules in Python at:
  http://jfontain.free.fr/
(see the INSTALL file for complete instructions, for UNIX and also
Windows platforms).

You also have the option of using the moodss rpm file (also in my
homepage), if you are using a Redhat Linux system (6.0 or above).
You can find the required tcl, tk, tktable, blt, tcpperl and other
rpms at: http://jfontain.free.fr/

Whether you like it (or hate it), please let me know. I would like to
hear about bugs and improvements you would like to see. I will correct
the bugs quickly, especially if you send me a test script (module code
with a data trace would be best).

###

you may find it now at my homepage:

http://jfontain.free.fr/moodss-14.1.tar.gz
http://jfontain.free.fr/moodss-14.1-1.i386.rpm
http://jfontain.free.fr/moodss-14.1-1.spec

and a bit later at:

ftp://contrib.redhat.com/ in libc6 sub-directory.

Enjoy and please let me know what you think.


--
Jean-Luc Fontaine  mailto:jfontain at free.fr  http://jfontain.free.fr/





From jeremy at alum.mit.edu  Mon Mar 12 19:09:52 2001
From: jeremy at alum.mit.edu (Jeremy Hylton)
Date: Mon, 12 Mar 2001 19:09:52 -0500 (EST)
Subject: (Port-induced?) memory leak in 2.0
In-Reply-To: <3AAD6201.8F549AD7@packeteer.com>
References: <3AAD6201.8F549AD7@packeteer.com>
Message-ID: <15021.25936.221795.719840@w221.z064000254.bwi-md.dsl.cnc.net>

It looks like the leftovers are all interned strings.  These are not
freed when you call Py_Finalize() because they must persist across
multiple interpreter creations.  The interned strings live as long as
the process.

Jeremy



From fredrik at pythonware.com  Wed Mar 21 12:53:10 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 21 Mar 2001 17:53:10 GMT
Subject: re.match
References: <997skf$qoc$1@bmerhc5e.ca.nortel.com> <3AB8E6DF.A2C32AD8@mentor.com>
Message-ID: <aU5u6.914$sk3.297681@newsb.telia.net>

Bob Cannard wrote:
> Another possibility would have been to combine compiled regular
> expressions, but there doesn't seem to be a way to do this. Does
> anyone know of a function or piece of magic that I've missed?

not really.  you can make it look like you're reusing the old
pattern, but it doesn't save you any time:

>>> import re
>>> p1 = re.compile("foo")
>>> p1.pattern
'foo'
>>> p2 = re.compile(p1.pattern + "$")

>>> print p1.match("foo")
<SRE_Match object at 008CCB00>
>>> print p1.match("foobar")
<SRE_Match object at 008CCD90>
>>> print p2.match("foo")
<SRE_Match object at 008CCB00>
>>> print p2.match("foobar")
None

Cheers /F




From tony at lsl.co.uk  Thu Mar  1 08:30:09 2001
From: tony at lsl.co.uk (Tony J Ibbs (Tibs))
Date: Thu, 1 Mar 2001 13:30:09 -0000
Subject: [Distutils] Re: CPAN functionality for python - requirements
In-Reply-To: <032e01c0a24b$13d30fa0$e000a8c0@thomasnotebook>
Message-ID: <005d01c0a253$bc513dc0$f05aa8c0@lslp7o.int.lsl.co.uk>

John Lee grumbled about a lack of documentation for distutils, and
Thomas Heller replied:
> I wrote some docs. They are online already in Fred Drake's development
> version:
> http://python.sourceforge.net/devel-docs/

Erm - all I can see there that sound at all relevant are "Distributing
Python Modules" and "Installing Python Modules", both credited to Greg
Ward, and both in that <rant>infernally infuriating "one smidgeon of
text per HTML page" format (which is worse than useless, so far as I'm
concerned!)[1]</rant>.

Could you give a clearer reference, please?

..[1] or am I missing some (undocumented) trick of the trade which will
lead me to whole documents?

Tibs

--
Tony J Ibbs (Tibs)      http://www.tibsnjoan.co.uk/
Give a pedant an inch and they'll take 25.4mm
(once they've established you're talking a post-1959 inch, of course)
My views! Mine! Mine! (Unless Laser-Scan ask nicely to borrow them.)




From jwbnews at scandaroon.com  Mon Mar 19 23:45:49 2001
From: jwbnews at scandaroon.com (John W. Baxter)
Date: Mon, 19 Mar 2001 20:45:49 -0800
Subject: Why "from __future__" stinks; a counter-offer
References: <mailman.985029808.7088.python-list@python.org> <memo.20010319213204.8771C@digitig.compulink.co.uk> <9965t7$s57$1@panix6.panix.com> <9969ta$2is$1@newsg3.svr.pol.co.uk>
Message-ID: <jwbnews-5EED43.20454819032001@corp.supernews.com>

In article <9969ta$2is$1 at newsg3.svr.pol.co.uk>,
 Will Newton <will at nospam.misconception.org.uk> wrote:

> Aahz Maruch wrote:
> 
> > So far, no other person has actually come up with a scheme that addreses
> > the problem at hand, as opposed to simply whining, suggesting minor and
> > worthless variations, or putting forth even *more* complicated solutions.
> > One of the nice things about this solution is that if it is decided to be
> > an ugly wart, it can be easily dropped for Python 2.2.
> 
> How often is this "from __future__" expected to be used?
> 
> Just experimental code written for bleeding edge versions of Python or a 
> significant portion of code (that will eventually become legacy code)?
> 

It seems to be for gentle introduction of changes which are considered 
good (a category in which I would file the nested scopes as it is in 2.1 
and will be), but which will cause code breakage.  If something doesn't 
break code, it can go in without the mechanism.  

Note that no one ever has to write such an import (although folks will 
have to read them)...one can instead elect to wait until the future 
feature becomes *the way things are* before using it.  (But dropping the 
import into existing code long enough to observe breakage in 2.1 would 
seem more prudent than waiting...even without making use of the nested 
scopes.)

And the from __future__ import xxx has at least the advantage that it 
seems to work.

I would expect there will not be scores of those; more like a handful or 
two, since the BDFL has made it clear he would rather not break code.

Meanwhile, the
   from __future__ import xxx
is presently in use for exactly one feature, with no more on the horizon 
for 2.1.

That would seem to leave time for someone(s) who prefer some 
alternative(s) to write (a) PEP(s) proposing same...convincing enough 
and implementable enough, and such input could make
   from __future__ import nested_scopes
the only instance of from __future__.  

But I don't think the proponents of something else should wait very 
long, and I think they should spend their energy on the (er) future, not 
on this instance in this release.  Once from __future__ has been used in 
a couple of releases, changing it would be harder than it would be now 
for 2.2 et seq.  I think Tim's description of the from __future__ PEP as 
darn near retroactive might enhance the chances of an alternative PEP 
displacing it (but not for the nested scopes thing).

  --John

-- 
John W. Baxter   Port Ludlow, WA USA  jwbnews at scandaroon.com


From objectway at divalsim.it  Tue Mar 13 12:02:49 2001
From: objectway at divalsim.it (Nicola Musatti)
Date: Tue, 13 Mar 2001 18:02:49 +0100
Subject: This math scares me
References: <3aad4062.1290021@News.CIS.DFN.DE> <Hwbr6.1139$Tg.127571@ruti.visi.com> <3aad4485.2348499@News.CIS.DFN.DE> <jwbnews-7CC03F.19050312032001@corp.supernews.com> <yu993dcimjt8.fsf@europa.research.att.com> <slrn9asbvc.dfq.neelk@alum.mit.edu>
Message-ID: <3AAE52B9.3CE00110@divalsim.it>

Hey, you already *stole* Alex Martelli from us, now are you trying to
steal Andrew Koenig away too? <wink>

it.comp.lang.c++-is-nothing-like-it-used-to-be-ly y'rs
Nick

In case you wonder: http://www.research.att.com/info/ark


From thomas.heller at ion-tof.com  Fri Mar  2 12:39:06 2001
From: thomas.heller at ion-tof.com (Thomas Heller)
Date: Fri, 2 Mar 2001 18:39:06 +0100
Subject: Unresolved symbols in MSVCRT.DLL when running a py2exe packaged program.
References: <mailman.983539023.17143.python-list@python.org> <4rQn6.14308$UZ4.4134344@news4.rdc1.on.home.com>
Message-ID: <97olrr$qalbl$1@ID-59885.news.dfncis.de>

> "Rolander, Dan" <Dan.Rolander at marriott.com> wrote in message
> news:mailman.983539023.17143.python-list at python.org...
> > I agree with Robin and Gordon's assessment. It's risky at best to
> distribute
> > system dll's with your application, and impossible to upgrade system dll's
> > on Win2K because of system file protection.
> >
> > Dan

"Tom" <NoSpam at NoSpam.com> wrote in message news:4rQn6.14308$UZ4.4134344 at news4.rdc1.on.home.com...
> I disagree.
>
> You can distribute system dll's.  You are expected to, and sometimes you
> have to.  You must insure that the version you load is at least equal to the
> version that you linked against.
>

These are different resposibilities:

py2exe should simply create an executable, and should
copy dlls needed by the executable into the dist directory.
Unfortunately is does this also for SYSTEM dll's.

py2exe shouldn't, it should maybe only do this for user
dlls, and print a list of system dll's needed.
Maybe it shouldn't even print this list...

Ensuring that the system dll's are installed/upgraded
on the target system should be left to an installer
like WISE or InnoSetup: They can replace in-use dlls,
they can do proper version checking, update the ref-count
of the dlls, and so on.

Regards,

Thomas

--------------------------------------------------------
py2exe - http://starship.python.net/crew/theller/py2exe/





From aahz at panix.com  Sun Mar 18 10:31:04 2001
From: aahz at panix.com (Aahz Maruch)
Date: 18 Mar 2001 07:31:04 -0800
Subject: PEP 42 and Py2.1
References: <3AB45C34.C8D1E82D@javanet.com>
Message-ID: <992kbo$blb$1@panix2.panix.com>

In article <3AB45C34.C8D1E82D at javanet.com>,
Raymond Hettinger  <othello at javanet.com> wrote:
>
>I've checked out the Py2.1 beta and my most desired small feature didn't
>make it.  To quote PEP 42:
>    Generalize eval() to accept any mapping objects for locals and globals.
>
>Is it too late to get this one in?

Yes.
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

Three sins: BJ, B&J, B&J


From TCHUR at doh.health.nsw.gov.au  Thu Mar  8 01:48:24 2001
From: TCHUR at doh.health.nsw.gov.au (Tim CHURCHES)
Date: Thu, 08 Mar 2001 17:48:24 +1100
Subject: Who am I: can a class instance determine its own name?
Message-ID: <saa7c5ef.027@dohprimo.ccf.health.nsw.gov.au>

This is probably an elementary question and the answer is probably writ large in multiple places in the Python documentation, but...

...can an instance of a class determine the name of the variable to which it is assigned? For example:

###########################
class Foo:
    def whoami(self):
         return "You are a Foo() but I do not know your name"

FooBar = Foo()

print FooBar.whoami()
###########################

How does one define the method whoami() so that it returns "FooBar"? This sort of navel gazing is formally called introspection, I think (therefore I am)?

Tim Churches
Sydney, Australia
(where, due to the Coriolis effect, the Python prompt does indeed look like this: <<< - or maybe its because we are upside-down)







From fredrik at pythonware.com  Tue Mar 13 14:38:55 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Tue, 13 Mar 2001 19:38:55 GMT
Subject: Tkinter with PIL crashing on large images - help
References: <3aacc208.1842603@news.ne.mediaone.net>
Message-ID: <jHur6.113$4d3.22203@newsb.telia.net>

Joel Gould wrote:
> Is this expected behavior?  If I load a 3.3 megapixel JPG file into a
> PIL (Python Image Library) PhotoImage object and then draw that on a
> Tkinter.Canvas object, Python crashes with an access violation (under
> WIndows 98).

fwiw, it works fine my Win2K box.

don't have time to dig up a Win98 box right now, but I suspect Tk's
running out of GDI resources in this case (under win9X).

Cheers /F




From jlowe at mentos.urbana.css.mot.com  Mon Mar 12 19:42:57 2001
From: jlowe at mentos.urbana.css.mot.com (Jason Lowe)
Date: 13 Mar 2001 00:42:57 GMT
Subject: Python, threads, and signals (oh my!)
Message-ID: <98jqeh$6ut$1@newshost.mot.com>

I've been playing around with Python and threads, and I've noticed some
odd and often unstable behavior.  In particular, on my Solaris 8 box I
can get Python 1.5.2, 1.6, or 2.0 to core dump every time with the
following sequence.  I've also seen this happen on Solaris 6.


1. Enter the following code into the interactive interpreter:
--
import threading

def loopingfunc():
  while 1: pass

threading.Thread(target=loopingfunc).start()
--

2. Send a SIGINT signal (usually Ctrl-C, your terminal settings may
   vary).  "Keyboard Interrupt" is displayed and so far everything looks 
   fine.

3. Now simply press the <Enter> key to enter a blank line in the
   interpreter.  For my Solaris 8 box with the GNU readline 2.2 module
   present, this always ends up in a core dump.  It may take a while,
   since at this point the readline signal handler is being re-entered
   recursively until the stack overflows.


Looking more into this, it appears that on Solaris, more than one
thread is processing the signal handler installed by the readline
module (according to truss output of the process).  Unfortunately, the
readline signal handlers don't support being re-entered, as they have
global data that's not protected.  

Now granted, signals and threads are a dangerous business.  However, it
would be nice if the user sending SIGINT to a script wouldn't cause
instability.  Looking at the Python module sources, I noticed that the
signal module is somewhat "thread aware" -- it allows signal
handlers to be installed only by the main thread.  However, I found it a
little odd that the thread support, both in the core interpreter and in
the thread module, had no support code for signals -- even in the
specific thread cases like POSIX (pthreads).

According to various pthread documentation, it appears the "right" way
to handle the often volatile signal/thread mix is to mask all signals
except in one thread (usually the main thread) so that only one thread
will receive the signal.  Unfortunately, I couldn't find any Python
module that would allow one to change the signal mask of a thread
(pthread_sigmask) or even a process (sigmask, sigprocmask, etc.).  I'm
assuming the lack of this interface has to do with portability across
platforms.

I was able to solve the problem by modifying Python/thread_pthread.h's
PyThread_start_new_thread() to block all signals before creating the new 
thread and then restoring the signal mask after the new thread was off
and running.  Therefore, all threads created by Python except the
initial thread will have all signals masked with this change.  Masking
all signals in new Python threads makes sense to me, given that the
signal module doesn't like other threads installing handlers anyway.

As a side note, I think part of the problem with sending SIGINT at the
interactive prompt while threading is aggravated by the Python readline
module's signal handler.  It setjmp()'s and longjmp()'s to do its dirty
work, and many thread implementations require the longjmp() be performed 
by the thread that did the setjmp().  If signals aren't masked in all
threads except the one doing the readline() call, then this isn't
guaranteed.

So what's everyone else's take on this?  Has anyone else experienced
problems in Python when using threads and receiving signals (like
SIGINT)?  Should the thread code be at least a little more "signal
aware", so that new Python threads have all signals masked on platforms
that support this?

Thanks in advance.

Jason Lowe
--
Jason Lowe                                    Urbana Design Center
Motorola Personal Communications Sector       1800 South Oak Street
jlowe at urbana.css.mot.com                      Champaign, IL  61820-6947
                                              (217) 384-8513


From parkw at better.net  Thu Mar 22 16:26:14 2001
From: parkw at better.net (William Park)
Date: Thu, 22 Mar 2001 16:26:14 -0500
Subject: Good Editor? (What's the big deal with syntax colouring?)
In-Reply-To: <20010322131648.V24341@trufflehunter.avalongroup.net>; from tjg@exceptionalminds.com on Thu, Mar 22, 2001 at 01:16:48PM -0800
References: <008F0A63472BD311AF9800104BCD102561CD3E@minirex.vmlabs.com> <20010322131648.V24341@trufflehunter.avalongroup.net>
Message-ID: <20010322162614.A1133@better.net>

On Thu, Mar 22, 2001 at 01:16:48PM -0800, Timothy Grant wrote:
> On Thu, Mar 22, 2001 at 12:10:59PM -0800, John Schmitt wrote:
> > Why do people want syntax colouring?  I can't stand it.
>
> It's pretty simple. Forget to close a ("|')--or open one for
> that matter, and you automatically know somethings wrong
> because the colours are wrong, misspell(sp?) a keyword and you
> know it because it doesn't hi-light in the correct colour.
>
> It is a huge benefit to getting syntax correct before you save
> your file.

Yeah, same here.  Colour is good for quick scan, but not for real work.
So, usually I type with colour off, and turn it on before I exit.

:wq --William Park, Open Geometry Consulting, Linux/Python, 8 CPUs.



From sholden at holdenweb.com  Sat Mar  3 14:45:27 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Sat, 3 Mar 2001 14:45:27 -0500
Subject: Thinking Pythonically (was Re: gah! I hate the new string syntax)
References: <mailman.983485923.1136.python-list@python.org> <97opsb$nhu$1@nntp6.u.washington.edu> <97oupq$qijrh$1@ID-77331.news.dfncis.de> <97pcu2$n6n$1@animus.fel.iae.nl> <97ptag$81s65$1@fido.engr.sgi.com>
Message-ID: <YUbo6.4854$ML1.258365@e420r-atl2.usenetserver.com>

"Paul Jackson" <pj at sgi.com> wrote in message
news:97ptag$81s65$1 at fido.engr.sgi.com...
>
> Sorry - couldn't resist the following ...
>
[YASOL]
>
> ... I write this trusting that everyone else on this thread has
> more sense than to follow this up with yet more ways to write
> shell one-liners ... clearly I lack such sense ...
>
Pythonic thinking at its best ;-)

"""From personal experience I have to say that Python users are a great
bunch of people. Flames are noticeably less frequent on comp.lang.python
than other language newsgroups, despite recent increases in volume. Whether
this comes from using the language, or the group is effective in socializing
its members, or whether Python just naturally attracts a particular type of
personality is difficult to say. The list is also exceptionally welcoming to
newcomers, and patiently revisits topics that could be answered by a search
of the FAQ."""

[An anthropological note on Python users: forthcoming]

Do you mind if I quote you, with appropriate attribution?

regards
 Steve





From tuttledon at hotmail.com  Thu Mar  1 23:29:59 2001
From: tuttledon at hotmail.com (Don Tuttle)
Date: Fri, 02 Mar 2001 04:29:59 GMT
Subject: PythonJournal.com becomes Interactive, new issue today
References: <mailman.983500743.14313.python-list@python.org> <hzEn6.1869$GD.875814@typhoon.snet.net>
Message-ID: <blFn6.13318$__6.3143457@typhoon.southeast.rr.com>

"Paul Magwene"
> Ouch, to be perfectly honest, if the above website was my first exposure
to
> Python's web application capabilities I'd run for the hills.  The layout
of
> the site is AWFUL.

Gotta agree.  Especially if we have to wait 2 -3 years for Issue III.

Don




From geek+ at cmu.edu  Fri Mar 16 12:45:12 2001
From: geek+ at cmu.edu (Brian E Gallew)
Date: Fri, 16 Mar 2001 12:45:12 -0500
Subject: X properties
Message-ID: <200103161745.MAA07707@smtp2.andrew.cmu.edu>

In any of the GUI toolkits, is there a way to add random window
manager properties?  After the window has been created?

-- 
=====================================================================
| JAVA must have been developed in the wilds of West Virginia.      |
| After all, why else would it support only single inheritance??    |
=====================================================================
| Finger geek at cmu.edu for my public key.                            |
=====================================================================
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 267 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/20010316/de3a2600/attachment.sig>

From cce at clarkevans.com  Mon Mar 26 05:04:06 2001
From: cce at clarkevans.com (Clark C. Evans)
Date: Mon, 26 Mar 2001 05:04:06 -0500 (EST)
Subject: win32com:MakePy -- Early Binding without the size jitters?
In-Reply-To: <3AAAE3BB.5090506@ActiveState.com>
References: <mailman.984043211.10915.python-list@python.org> <3AAAE3BB.5090506@ActiveState.com>
Message-ID: <Pine.LNX.4.21.0103260458020.19297-100000@clarkevans.com>

On Sun, 11 Mar 2001, Mark Hammond wrote:
> If you are using the gencache functions to create makepy files, try 
> passing "bForDemand=1" to the functions.  This will only generate what 
> is actually used by the program, as it is used.

Mark,  

  I spent time trimming the wrapper file for MSXML,
  microsoft's XML/XSLT implementation down to the
  very few methods which I require.  I'm now creating
  DOMDocument via the following call:

      doc = msxml.DOMDocument.default_interface(
                pythoncom.CoCreateInstance( 
                    "MSXML2.DOMDocument",None, 
                     pythoncom.CLSCTX_INPROC,
                     pythoncom.IID_IDispatch))

  It seems to be working as expected, and my wrapper
  file is very very small (good deal).  So.  I'm wondering:

     Am I shooting myself in the foot?

  If not, I had another question... could something
  like the above by default be put in the wrapper file
  automagically...  I noticed that DOMDocument has a
  comment nearby saying "MSXML2.DOMDocument".

  How about making the CoClasses default to somethign
  like the above in their constructor when an argument
  (The IDispatch) is not passed. Then it would be very
  easy to use..

   import msxml3
   doc = msxml3.DOMDocument()
 

Thank you for your thoughts!

Clark





From jp at NOSPAMdemonseed.net  Mon Mar 12 14:37:31 2001
From: jp at NOSPAMdemonseed.net (jason petrone)
Date: Mon, 12 Mar 2001 19:37:31 +0000
Subject: Tkinter canvas double buffering
Message-ID: <rh8j89.pu7.ln@demonseed.net>

Is there an easy way to do double buffering with the canvas?  The only way
I see is rasterizing it offscreen myself into a BitmapImage, then drawing that.

I've done this plenty of times in Xlib and GTK, but I would prefer to use
Tk for this.  Could someone post some example code?

thanks

-jason



From root at rainerdeyke.com  Wed Mar 28 21:48:52 2001
From: root at rainerdeyke.com (Rainer Deyke)
Date: Thu, 29 Mar 2001 02:48:52 GMT
Subject: Is inheritance broken?
References: <15042.11167.848530.564353@beluga.mojam.com><NDBBLAKACLGJJBDNGGCMOEJBCCAA.jcc.ugm@ix.netcom.com> <mailman.985809879.27201.python-list@python.org>
Message-ID: <ooxw6.513708$ge4.182488142@news2.rdc2.tx.home.com>

"Skip Montanaro" <skip at pobox.com> wrote in message
news:mailman.985809879.27201.python-list at python.org...
>
>     Jonathan> Ugly and error prone, no?
>
> Well, perhaps there is a better way...  ;-)
>
> How about creating mixins, then composing a set of classes from them:

An alternate approach would be to explicitly pull methods into a class:

class Square(Rectangle, Rhombus):
  side = Rhombus.side


--
Rainer Deyke (root at rainerdeyke.com)
Shareware computer games           -           http://rainerdeyke.com
"In ihren Reihen zu stehen heisst unter Feinden zu kaempfen" - Abigor




From gregj at pobox.com  Fri Mar 16 00:35:56 2001
From: gregj at pobox.com (Greg Jorgensen)
Date: Fri, 16 Mar 2001 05:35:56 GMT
Subject: Cookies
References: <mailman.984707839.22839.python-list@python.org>
Message-ID: <0Dhs6.576315$U46.17170213@news1.sttls1.wa.home.com>

A cookie is just a text file stored in a special place where the browser can
find it. Where exactly the cookies go depends on your OS and browser.

Using a text file to store the key would seem the easiest solution, but
client-side Javascript can't create or write a file.

I think you should generate unique IDs on the server, not on the client.
Here's one way to handle it:

1. Client browser sends form data to server.
2. Server processes form, generates a unique ID, stores the form data keyed
by the unique ID.
3. Server replies with a "thank you" or "database updated" page that also
writes a cookie.
4. Python program starts up and looks for the cookie. Depending on your OS
and browser the location and name of the cookie will vary, but with a little
experimenting you can easily find the most likely locations.
5. Python program reads the cookie. Open your cookie with a text editor to
see what's in there--it's just name: value pairs.
6. Python program sends HTTP request to server with unique ID it got from
the cookie.

Obviously this won't work if the client has disabled cookies. With a little
more trouble you can detect this and perhaps put the unique ID on the
response page and tell the user to save it or something. Or embed the unique
ID in the reply page and tell the user to save that page and give it a
specific name, then have your Python program open that text file and find
the unique ID in there.

Depending on your requirements and security concerns, you can either have
your database program generate the unique ID (my preference), use the
Windows GUID functions (which are available to ASP on the server), or
implement one of the GUID functions you can easily find on the web. Many
programmers have solved this in one way or another; send email to me if you
want specific advice on generating unique IDs from your database engine.

Good luck.

<<<
"Ken Seehof" <kens at sightreader.com> wrote in message
news:mailman.984707839.22839.python-list at python.org...
Short version:

Is it possible for a stand-alone python application to read a cookie without
the help of a browser?"

Long version:

I have a web page W and a client side python script P.  I want W to generate
a GUID (perhaps using javascript) which is then sent to a server S, where it
is to be used as a key in a database.  Now I run P (on the client, and not
in a browser) and I want P to somehow find the value of the GUID, so that it
can query the database on S.

Alternate question:

Is there a technique common to both javascript and python that will obtain a
unique identifier for a computer (e.g. IP address).  It must not be browser
specific.

Motivation:

I need to write a record of data from an html form, keyed by some kind of
unique client id, to the server and then be able to query that record from a
stand alone python script running on the client.  The missing link is
getting the html form and the python script to agree on a value for the id;
the rest is easy.
>>>

--
Greg Jorgensen
PDXperts
Portland, Oregon, USA
gregj at pobox.com




From rozen at rgelpc254.rgv.hp.com  Wed Mar  7 12:09:13 2001
From: rozen at rgelpc254.rgv.hp.com (Don Rozenberg)
Date: 07 Mar 2001 09:09:13 -0800
Subject: ANNOUNCE PAGE 0.1.1 - Drag and Drop GUI Generator.
Message-ID: <80ld7bt1p6u.fsf@rgelpc254.rgv.hp.com>

This version of PAGE, 0.1.1, contains my setup.py file which
simplifies installation under Linux of the required python files.

PAGE is a tool which helps to create GUI for Python programs.  It uses
Tk and Tix widgets.  PAGE is not an end-all, be-all tool, but rather
one that attempts to ease the burden on the Python programmer. It is
aimed at the user who will put up with a less-than-general GUI
capability in order to get an easily generated GUI. It does not build
an entire application but rather is aimed at building a single GUI
window.  

PAGE is based on Stewart Allen's Visual Tcl program. In fact, it is
vTcl, with an additional module that generates Python code and some
bug fixes.  PAGE is built on version 1.2.2 of Visual Tcl, version
8.1.1 of Tix, and Tcl/Tk version 8.3.2. It works with Python 2.0.
This is an early version which run on Linux. It supports many Tk
widgets and a subset of the Tix widget set.  The user will have to
acquire and install Tix8.1.1 and tcl8.3.2 to use this program.


The PAGE web site is 

	http://www.geocities.com/page_python

-- 
Don Rozenberg
rozen at rgv.hp.com
408-343-6266


From -$Paul$- at verence.demon.co.uk  Thu Mar  8 19:01:56 2001
From: -$Paul$- at verence.demon.co.uk (Paul Wright)
Date: 9 Mar 2001 00:01:56 -0000
Subject: learning C
References: <mailman.983972829.23481.python-list@python.org> <983975751.18833@newsmaster-04.atnet.at> <slrn9adklr.bv9.scarblac@pino.selwerd.nl> <984047775.592522@newsmaster-04.atnet.at>
Message-ID: <9896hk$ig8$1@verence.demon.co.uk>

In article <984047775.592522 at newsmaster-04.atnet.at>,
Werner Schiendl <ws-news at gmx.at> wrote:
>
>Remco Gerlich <scarblac at pino.selwerd.nl> wrote in message
>news:slrn9adklr.bv9.scarblac at pino.selwerd.nl...
>> A baseless statement. There are many types of project for which C is still
>> the best choice. For one thing, think embedded systems.
>
>Would say, depends.
>
>If you have some really simple thing to do (and those are much rarer breed
>than some believe...) you are probably better of not using C++ extended
>features, libraries and syntax. Especially if the code has to be inspected
>externally for safety reasons.
>
>But if you have a need for string handling, containers, dynamic object
>creation and deletion, etc. you can of course do that in C, but doing it in
>C++ will get you up and running much faster.

I think you have a different definition of "embedded" from me
(admittedly it's a broad term). C is still in wide use in the industry I
work in (which produces small, hard real time systems). This is unlikely
to change in the near future: C++ is perceived as being slow and unsafe.
There's the feeling that you never quite know what it's going to do.  By
contrast, C compilers are pretty efficient and what they'll end up
producing is more or less in line with what you'd expect from looking at
the code. C itself can be unsafe, of course, but by using restricted
subsets of the language you can help with this somewhat (banning dynamic
allocation is common, for example).

I don't think these systems are simple. If they were, writing them would
be a lot easier.

>I personally use embedded platforms and C++ is readily available there
>(VxWorks, to be precise).

ObFourYorkshiremen: You had an OS? You were lucky! In my day we had some
interrupts and timers, and we liked it! (Actually, my current project
does have an OS: such luxury :-)

Not-speaking-for-my-employer-ly yours

Paul

-- 
----- Paul Wright ------| Nobody will ever win the battle of the sexes. There's
-paul.wright at pobox.com--| too much fraternizing with the enemy.
http://pobox.com/~pw201 | 


From moshez at zadka.site.co.il  Mon Mar 26 05:54:43 2001
From: moshez at zadka.site.co.il (Moshe Zadka)
Date: Mon, 26 Mar 2001 12:54:43 +0200
Subject: Python ready to replace perl (was Re: Install Python on Apache)
In-Reply-To: <86snk1qafw.fsf_-_@jfh00.fernuni-hagen.de>
References: <86snk1qafw.fsf_-_@jfh00.fernuni-hagen.de>, <99fph5$a2n$1@rex.ip-plus.net> <wzelvnz4l5.fsf@sunshine.cs.uu.nl>
Message-ID: <E14hUe3-00040h-00@darjeeling>

On 26 Mar 2001, Fritz Heinrichmeyer <fritz.heinrichmeyer at fernuni-hagen.de> wrote:
> So i consider using python in the future.

I am fluent in both, and I think you're making the right decision.

> For this python has to have
> 
> 1. sql-interface like dbi, at least for postgres (i think this exists).

db-api, with two postgresql implementations available.

> 2. interface to openldap 

Exists. I can't remember offhand where to download this...

> 3. mod_perl beeing able to program authentification procedures using
>    1. and 2. for apache and beeing able to program arbitrary URL-redirect
>    procedures.

Err....mod_perl doesn't support Python last I checked.
You can do this in mod_snake, AFAIK.

> 4. a template system beeing able to manage multiple language versions
>    (to replace the perlish WML-System)?.
> 
> Are there solutions apart from using zope?

These days you can't throw a rock without hitting a Python templating
framework. Note you can rip out DTML from Zope and use it in your
Python scripts, or you can use PyML (http://www.pyml.org) or
Quixote or WebWare and couple more I'm not aware about.
Note that in Python, the solution "write your own" is not as idiotic
as it sounds (see all the available alternatives!). With stuff
like sgmllib it's easy to embed Python in HTML, and with stuff
like the compiler package it's possible to embed HTML in Python.
You can also go in a more xmlish direction with stuff like ZPT,
but I'm not sure how finished it is.

 
-- 
"I'll be ex-DPL soon anyway so I'm        |LUKE: Is Perl better than Python?
looking for someplace else to grab power."|YODA: No...no... no. Quicker,
   -- Wichert Akkerman (on debian-private)|      easier, more seductive.
For public key, finger moshez at debian.org  |http://www.{python,debian,gnu}.org



From sebastien.libert at comexis.com  Mon Mar 12 10:29:42 2001
From: sebastien.libert at comexis.com (Sébastien Libert)
Date: Mon, 12 Mar 2001 16:29:42 +0100
Subject: file - codecs - unicode ??? 
Message-ID: <3aace9da$0$382$446d705c@news.skynet.be>

Hi,

here is a sample of what i do :

>>> h=open("c:\\myfile.log")
>>> line=h.readline()
>>> line
'\377\3760\000.\0000\0000\000\011\000i\000n\000f\000o\000\011\000L\000o\000g
\000_\000S\000t\000a\000n\000d\000a\000r\000d\000\011\000n\000g\000L\000o\00
0g\000\015\000\012'

What can i do with this kind of thing ?????
I want to convert thid in ascii.... how can i do thid in python ?

I use python 2.0 on win32 !!!

thx to answerd ;-)




From grante at visi.com  Thu Mar 22 12:44:15 2001
From: grante at visi.com (Grant Edwards)
Date: Thu, 22 Mar 2001 17:44:15 GMT
Subject: Cross-platform serial port access
References: <99d274$ogv$1@news.netpower.no> <PRou6.1592$SB2.181656@ruti.visi.com> <99d7g3$qed$1@news.netpower.no>
Message-ID: <PRqu6.1603$SB2.186660@ruti.visi.com>

In article <99d7g3$qed$1 at news.netpower.no>, David Brown wrote:

>>Currently, serial port stuff in Python is not portable between
>>Mac/MS/Unix.
>
>I suppose it should be easy enough to make a small wrapper around the
>different platforms' code.

Yup -- somebody should do that someday...

>>>Does anyone know if Linux supports RTS toggle handshaking
>>>(i.e., the RTS line is high while transmitting, for convenient
>>>connection to RS-485 convertors),
>>
>>Not with the standard PC serial port driver.  I hacked the
>>16x50 driver to do RTS toggle several years ago, but it never
>>worked reliably.  It's very difficult to do it reliably in
>>software at high baud rates.

Back when I tried to do it with the Linux driver, I was trying
to use extra transmit data bytes (didn't work well) and later
the shift-register empty bit to shut off RTS.  I should have
done it using loop-back: enable loopback in the UART and shut
off RTS in the receive handling code when the end of the tx
message is seen.  That only works if loopback in the UART
doesn't disable the transmit data pin...

>>Your best bet is to buy a board that supports RTS toggle in
>>hardware or use a 232-485 converter that handles it
>>automatically by detecting the presence of data.

>This is definitely a shame.  The RTS toggle works well under NT
>(no idea about Win9x) - I have written a Delphi program that
>uses it to run a RS-485 convertor at 57600.  I don't know for
>sure how fast RTS is turned off after the last byte - the other
>nodes on the bus wait 10 ms before answering anyway.

That's the trick.  10ms is long enough that it can be done in
software.  I had to get it shut off within 1 byte-time of the
last stop bit.  At 57.6K that's 17us.

>I guess under Linux I would have to toggle RTS manually.
>Turning it on before sending is easy enough, it is turning it
>off at the right time that is more of a hassle.

Yup - it's extremely tricky.  I finally gave up and bought a
board that did it in hardware.  If you've got a converter that
loops back transmit data, you can use that: do a write(), then
do a read() -- when you've seen the right number of bytes, shut
off RTS.

Some of the middle and upper level serial driver layers in
Linux have a 10ms granularity -- so you might not be able do
anything from user-space faster than 10ms.

>So-called "intelligent" RS-485 convertors that read the
>incomming data and change direction appropriately have their
>own problems.  We have delivered systems that communicated at
>300 baud, and "intelligent" convertors frequently do not work
>at such slow rates.  Additionally, they generally need
>configured with DIP-switches and the like, which is an extra
>hassle if you need to change the baud rate.

Yup.  The ones without dip-switches (fixed tx data timeout)
will only work well at certain baud rates.

>Simple RTS-controlled convertors are much easier in that
>respect.

I agree.

>Although it is not really Python-specific, do you know if there
>are serial drivers for Linux that simulate 9-bit UARTs (i.e.,
>swapping the parity bit from space to mark as appropriate)?

I don't think so.

-- 
Grant Edwards                   grante             Yow!  .. Do you like
                                  at               "TENDER VITTLES?"?
                               visi.com            


From fredrik at pythonware.com  Thu Mar  1 02:01:00 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Thu, 01 Mar 2001 07:01:00 GMT
Subject: "import X" and "from X import y" produce different results, Why?
References: <ifmn6.6635$hn5.857862@news1.rdc1.mb.home.com>
Message-ID: <Msmn6.16082$Qb7.2629029@newsb.telia.net>

Parzival Herzog wrote:
> So, apparently the imported X is a copy of xx.X, not the same object
> by another name.

it's another *reference* to the same object, not a copy.

I suggest reading up on Python's object semantics:

    http://effbot.org/guides/python-objects.htm

and maybe also:

    http://effbot.org/guides/import-confusion.htm

Cheers /F

<!-- pythonworks 1.2 coming soon:
http://www.pythonware.com/products/works/index.htm
-->




From aleaxit at yahoo.com  Fri Mar  9 06:08:10 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 9 Mar 2001 12:08:10 +0100
Subject: win32com:MakePy -- Early Binding without the size jitters?
References: <mailman.984043211.10915.python-list@python.org>
Message-ID: <98ae5r01hn0@news2.newsguy.com>

"Clark C. Evans" <cce at clarkevans.com> wrote in message
news:mailman.984043211.10915.python-list at python.org...
    [snip]
> each library.  I was wondering... is it safe to delete
> methods/interfaces which I'm not using?  I've deleted

Yes.

> a few, and it still works for my purposes, so I was
> wondering if I'm shooting myself in the foot -- or if
> there is a methodological approach to this pruning?

Yes: study the sources and look at the bForDemand
flag (introduced reasonably recently).  That flag
ensures that only the wrappers you're actually using
are generated (it uses packages rather than modules
to house the wrappers).


Alex





From ahto at jussila.as  Tue Mar 27 09:49:43 2001
From: ahto at jussila.as (Ahto)
Date: Tue, 27 Mar 2001 17:49:43 +0300
Subject: C++ and Python
Message-ID: <99q99l$g21$1@news.clinet.fi>

I'm very new to Python and haven't quite figured out how to use c++ and
python together on windows...if someone could send me an example program of
tutorial webpage where this is explained?!?

Thanks




From mvanier at endor.bbb.caltech.edu  Thu Mar  1 02:59:36 2001
From: mvanier at endor.bbb.caltech.edu (Michael Vanier)
Date: 28 Feb 2001 23:59:36 -0800
Subject: writing a C function to manipulate python lists
Message-ID: <jllmqpnclz.fsf@endor.bbb.caltech.edu>

Hi,

I've been programming in python for many years but have never had to solve
this particular problem.  I would like to write a short C function that will
be part of a python-callable module.  The function will receive two python
lists as arguments.  Currently I do this in python, but my code uses this
function very heavily and it's the major speed bottleneck in the code.  The
hope is that the C function will be much faster than iterating through the
lists in python (which may not be true, but I'd like to see for myself).  

If anyone has some code lying around that does something like this, could
they pass it on to me?  I haven't seen this kind of thing in the official
python docs.  I'd also prefer to use the C representation of python lists
directly for speed, because using the C API would probably slow things down
considerably.

Mike


--------------------------------------------------------------
Mike Vanier	mvanier at bbb.caltech.edu
Department of Computation and Neural Systems, Caltech 216-76



From robin at jessikat.fsnet.co.uk  Sat Mar  3 04:34:42 2001
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Sat, 3 Mar 2001 09:34:42 +0000
Subject: the demise of 'from foo import * and its implications?
References: <3kVBAfATrDo6EwEM@jessikat.fsnet.co.uk>
 <mailman.983582584.8198.python-list@python.org>
Message-ID: <z8dAGJAyqLo6EwlX@jessikat.fsnet.co.uk>

In article <mailman.983582584.8198.python-list at python.org>, Tim Peters
<tim.one at home.com> writes
>[Robin Becker]
>> it's an ambitious act to suppose that the new complicated
>> __twilight_zone__ rules are better than the old simpler rules because of
>> some optimisation.
>
>The official rule was that import* at other than module scope has undefined
>behavior.  Still is.  The complication comes from spelling out the accidents
>of the 2.1 implementation that intersect with the accidents of the 2.0
>implementation, for the benefit of whatever users may have been violating the
>rules without realizing it.
>
>Would you prefer that we let you dope that out for yourself?  I'd be
>delighted to stick to the official rule, not catering at *all* to code that
>ignored the Ref Man.  Then we could replace the entire explanation with a
>one-line quote from the manual:  "you cheated; you got caught; tough luck".
>
>couldn't-be-simpler-than-that!-ly y'rs  - tim
>
>
my preference would be that statements legal in one context should be
legal in another where they make sense. I cannot see how using import *
or eval etc in nested scopes is somehow obviously bad or different from
the same usage in non-nested scopes. The fact that the prohibition comes
from making things easier for the compiler adds to my confusion.

I have no particular objection to being able to access variables in
intermediate scopes.

If we really want to improve visibility we could try to make

>>> def bingo(a):
...     if a<=1: return 0
...     return bingo(a-1)*a
...

work naively, but I'm fairly sure this will be rejected as unpythonic or
somesuch.
-- 
Robin Becker


From gustav at morpheus.demon.co.uk  Sun Mar  4 08:44:20 2001
From: gustav at morpheus.demon.co.uk (Paul Moore)
Date: Sun, 04 Mar 2001 13:44:20 +0000
Subject: urllib.urlopen and proxies
Message-ID: <vse4atgcgcfafth7c5tq3po7srd3l32f3h@4ax.com>

I've just looked at urllib for the first time. I was pleased to find
that it automatically picks up my proxy configuration from the system
(this is ActiveState's Python 2.0 on Windows 2000). However, I have a
problem - my proxy requires a user name and password, and there is no
place that I can see to enter them. So all I get back from urlopen()
is the polite "you need authorisation" message that our proxy gives
:-(

Can anyone tell me how to pass a proxy userid and password to
urllib.urlopen()?

Thanks,
Paul Moore


From donn at u.washington.edu  Thu Mar  8 17:42:21 2001
From: donn at u.washington.edu (Donn Cave)
Date: 8 Mar 2001 22:42:21 GMT
Subject: Get terminal size in perl and python
References: <20010308194028.A1616@bork.demon.nl> <mailman.984079092.10641.python-list@python.org> <DORp6.25$Tg.3395@ruti.visi.com>
Message-ID: <9891sd$hng$1@nntp6.u.washington.edu>

Quoth grante at visi.com (Grant Edwards):
[ ... vs. COLUMNS and LINES ]
| In most Unixes you'd be better off using the ioctl() call (the
| details of which I can't remember at the moment -- I thought it
| was in the termios struct, but I can't find it).

It has its own struct, and its own ioctls:

>>> x = fcntl.ioctl(0, 0x40087468, '12345678')
>>> struct.unpack('hhhh', x)
(24, 80, 240, 800)

The symbol would be TIOCGWINSZ, if you have it, but I don't think
that many do.  0x40087468 is good for NetBSD 1.5.  I don't see any
value for LINES or COLUMNS, and I'm not surprised - I think of that
as more of a SysV 3.2 thing than Berkeley.  But it really depends
on how you got there - ssh/telnet/console login, bash/ksh/csh, etc.

	Donn Cave, donn at u.washington.edu


From robin at jessikat.fsnet.co.uk  Tue Mar 27 11:15:45 2001
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Tue, 27 Mar 2001 17:15:45 +0100
Subject: warrior geeks
References: <3ABF72F1.98CB78F@sage.att.com>
 <znTv6.762$cgz.2752875@news.randori.com>
Message-ID: <nvBYMUAxyLw6Ewm$@jessikat.fsnet.co.uk>

In article <znTv6.762$cgz.2752875 at news.randori.com>, Robin Dunn
<robin at stop.spam.alldunn.com> writes
>
>I don't know about Martial Arts, but almost every other person named Robin
>that I know are in the Python community.  I've always thought that there is
>some sort of cosmic thing going on there...
>
>
>--
>Robin Dunn
...
well I share a name with the celebrated Harvard lesbian poetess. I
wonder what she knows about the pythonic arts.
-- 
Robin Becker


From max at alcyone.com  Tue Mar  6 12:38:19 2001
From: max at alcyone.com (Erik Max Francis)
Date: Tue, 06 Mar 2001 09:38:19 -0800
Subject: Pickling arbitrary character strings on Windows
References: <3AA51D2D.3251921E@lbl.gov>
Message-ID: <3AA5208B.A15ADC33@alcyone.com>

"Nicholas K. Sauter" wrote:

> It would be good to be able to pickle strings containing arbitrary
> ASCII
> characters, for example to encode data at the C level.  But on Windows
> it is impossible to do a binary unpickle of a string that has an
> embedded ASCII 26 (octal 032).

Under Windows, when you see anything like this, or start to see bizarre
games with end-of-line sequences, it means you have the files open in
the wrong mode.  You want the files open (for both reading and writing)
in binary mode, not text.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ I am become death, the destroyer of worlds.
\__/ J. Robert Oppenheimer (quoting Hindu legend)
    Kepler's laws / http://www.alcyone.com/max/physics/kepler/
 A proof of Kepler's laws.


From sholden at holdenweb.com  Mon Mar  5 01:05:50 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Mon, 5 Mar 2001 01:05:50 -0500
Subject: Subclassing builtin types
References: <i6u4atslg4euv732pvaujk6u036kemcidf@4ax.com> <obvo6.14838$1D5.510512@e420r-atl1.usenetserver.com> <ia15at4undkk6lpv7kn8l2nl9s3bgt30ln@4ax.com>
Message-ID: <A4Go6.14941$ML1.734525@e420r-atl2.usenetserver.com>

"Daniel Klein" <danielk at aracnet.com> wrote in message
news:ia15at4undkk6lpv7kn8l2nl9s3bgt30ln at 4ax.com...
> On Sun, 4 Mar 2001 12:41:55 -0500, "Steve Holden" <sholden at holdenweb.com>
> wrote:
>
> >> 1) Are wrapper classes for the builtin types a feasible (and pythonic)
> >thing to
> >> do?
> >>
> >Yes and yes: so feasible the library includes a UserList class which
> >provides esxactly what you seem to want: it wraps a list inside a class
so
> >you can inherit from it!
>
> Thanks Steve, you have just saved me a ton of work, something Python has a
> habit of doing consistently. :-)
>
Yes. Problem for me used to be doing some of the owrk before finding out
about the library module concerned. THe longer I use Python the less of a
problem this becomes, however.

> > Delegation is also a useful trick.
>
> I'm not sure what you mean by this. Are there any simple examples I can
look
> at?
>
> Dan

Well, there's the ftpStream mosule to which I referred in my last post. The
idea behind delegation is to create an instance of some other object,
storing it as an instance variable. Then you implement a __getattr__()
method which, when your object is asked for a non-existent method or
attribute attempts to return a reference to that method or attribute of the
saved instance object.

In ftpStream it works like this:

class ftpStream:
    "Specialised FTP with file-like interface."

    def __init__(self, host='', name='', passwd='', acct=''):
        "Create FTP object and set stream socket inactive."
        self.ftp = ftplib.FTP(host, name, passwd, acct)
        self.strsock = None

    def __getattr__(self, attrname):
        "Delegate unrecognised methods/attributes to the FTP object."
        return getattr(self.ftp, attrname)

In this case, calling ftpStream.cwd(path) causes the underlying FTP object's
cwd method to be used (because ftpStream has not cwd() method of its own).

The advantage of this is in avoiding naming conflicts: ftpStream has its own
close method, but it can easily call the close() method of its FTP object.

regards
 Steve





From sholden at holdenweb.com  Fri Mar  9 18:17:41 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Fri, 9 Mar 2001 18:17:41 -0500
Subject: Setting an attribute on a Python object from C
References: <mailman.984151691.9715.python-list@python.org>
Message-ID: <8Adq6.46418$1D5.2158622@e420r-atl1.usenetserver.com>

"G. Sumner Hayes" <SumnerH at fool.com> wrote in message
news:mailman.984151691.9715.python-list at python.org...
> Hi,
>
> I'm working on a Python extension in C and I've reached a point I can't
> get past.  How do I create an attribute on an object from C?
> PyObject_SetAttrString (or _SetAttr) doesn't seem to work for this.
> Here's what I'm doing (implementing a Python DB 2.0 driver for Microsoft
> SQL server):
>
> self=PyObject_NEW(mssql_cursor, &mssql_cursor_type);
> v=PyInt_FromLong(i);
> rv=PyObject_SetAttrString((PyObject*) self, "rowcount", Py_None);
>
> But rv comes back as -1.  Do I need to declare room in the mssql_cursor
> struct and somehow declare the rowcount attribute to point at that?
>
Are you aware of mxODBC? Without wishing to steal your thunder, it's a
pretty good ODBC for (SQL Server, Access, ...)

regards
 Steve





From mcalla at home.com  Wed Mar 28 08:43:38 2001
From: mcalla at home.com (Mike Callahan)
Date: Wed, 28 Mar 2001 13:43:38 GMT
Subject: Argument to Tkinter mainloop
Message-ID: <eUlw6.50034$Q47.13083876@news1.rdc1.tn.home.com>

What is the optional argument to mainloop used for? For example:
root.mainloop() vs root.mainloop(1)?

Mike Callahan





From aleaxit at yahoo.com  Thu Mar  1 10:46:36 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 1 Mar 2001 16:46:36 +0100
Subject: PEP 227 (was Re: Nested scopes resolution -- you can breathe again!)
References: <LNBBLJKPBEHFEDALKOLCMEEPJBAA.tim.one@home.com><200102230259.VAA19238@cj20424-a.reston1.va.home.com> <mailman.983384008.21434.python-list@python.org>
Message-ID: <97lqsr0311d@news2.newsguy.com>

"Jeremy Hylton" <jeremy at alum.mit.edu> wrote in message
news:mailman.983384008.21434.python-list at python.org...
    [snip]
>   Rebinding names in enclosing scopes
    [snip]
>       opposed to it.  It is difficult to support, because it would
>       require a new mechanism that would allow the programmer to specify
>       that an assignment in a block is supposed to rebind the name in an
>       enclosing block; presumably a keyword or special syntax (x := 3)
>       would make this possible.

If this were to be allowed (and I'm not saying it should be!), the
most natural way to allow it would appear to be to be a new clause
on the global statement -- most natural, because "allow rebinding
in an outer scope" is exactly global's job (of course, up to Python
2, the only 'outer scope' was the module-one:-).

Maybe something like...:

    def spam():
        def inner():
            global in spam, x
            x += 1
        x = 23
        inner()


Alex






From kens at sightreader.com  Sun Mar  4 01:54:16 2001
From: kens at sightreader.com (Ken Seehof)
Date: Sat, 3 Mar 2001 22:54:16 -0800
Subject: Java guy interested in Python
References: <003801c0a459$a29c97f0$63da1542@wisdom>
Message-ID: <00d301c0a477$ef8d6b20$a325fea9@his>

The article http://www.python.org/~rmasse/papers/java-python96/, which you may have seen because it is linked to in a previous post, gives a good overview of the differences.  However you should be aware that it was written in 1996 and python has made several advances since then (I'm sure java has as well, but naturally I haven't kept up on java after I learned python).

In particular:

- Python support for web development has grown substantially (e.g. Zope (www.zope.org)).

- Jython http://www.jython.org/ has arrived.  This lets you run python on a Java Virtual Machine (JVM) and derive python classes from java classes.

- Look in http://www.vex.net/parnassus/ for lots of other stuff.

And if you go outside the scope of web programming, IMHO, you find that python seems to have much broader scope.

In terms of the languages themselves, I think the biggest difference is that Java has declarations and is statically typed, while Python is dynamically typed and doesn't have declarations (except the rarely used 'global' keyword).  Many programmers are confused in that they think the purpose of type declarations is to protect the programmer from errors by providing enforcement of "programming by contract".  IMHO, type declarations are a sacrifice necessary to give compilers enough information to produce efficient code.  Type declarations make sense in C and C++, but they are (IMHO) a design flaw in high level languages.

Of course, as has been mentioned many times, no single language is best at everything, so it's good to learn a few.  On the other hand I haven't encountered any problems lately that can't be solved efficiently in python, combined with extensions written in C++.

Oops, I seem to have gotten carried away and forgot I'm a gentle python programmer.  I hope the "IMHO"s prevent a flame war :-).

  ----- Original Message ----- 
  From: Kyle Wayne Kelly 
  To: python-list at python.org 
  Sent: Saturday, March 03, 2001 7:17 PM
  Subject: Java guy interested in Python


  Can anyone tell me some of the things that Python offers which are not available in Java?  I like Python's syntax, but what can Python do that Java can not do?  Any comments are helpful.

  Kyle Wayne Kelly
  Computer Science Student
  University of New Orleans
  http://www.cs.uno.edu/~kkelly
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20010303/2cd8e3cd/attachment.html>

From owen at astrono.junkwashington.emu  Mon Mar 19 17:17:36 2001
From: owen at astrono.junkwashington.emu (Russell E. Owen)
Date: Mon, 19 Mar 2001 14:17:36 -0800
Subject: Wrong reasons to enhancing the language
References: <mailman.984971990.29857.python-list@python.org>
Message-ID: <9960i1$tp0$1@nntp6.u.washington.edu>

In article <mailman.984971990.29857.python-list at python.org>,
 Carlos Ribeiro <carribeiro at yahoo.com> wrote:

>Le me try to summarize what I've seen on the past ten days or so on this 
>list, regarding the awful lot of "enhancements" proposed to the Python 
>language.
>
>1) Proposing "syntactic sugar" changes to make some particular code 
>constructions look better;

Somebody recently posted a well-written message supporting the following 
idea: a very good predictor of how easily one can write bug-free code in 
a language is the expressiveness of the language. This matches my own 
experience.

This suggests that clean syntax to express an idea can be a very good 
thing, if done carefully.

Of course changes can be bad. They may be inconsistent with the 
language, add too much complexity, or add pitfalls to the language. But 
if a modification is well designed and allows one to express an idea 
more naturally or simply (especially a common idea) then it may pay rich 
dividends.

I believe many of the improvements in Python 2.0 fall into the category 
of helpful improvements. I am more doubtful about some of the stuff 
coming up (including nested scopes, // and rational numbers), but we'll 
see how it works out.

-- Russell


From andreas3004 at hotmail.com  Mon Mar 19 11:34:25 2001
From: andreas3004 at hotmail.com (Andreas Kremer)
Date: Mon, 19 Mar 2001 17:34:25 +0100
Subject: text file to array and back
References: <990fgo$cdm$04$1@news.t-online.com> <3AB58B8A.E1EAFA44@crosswinds.net>
Message-ID: <995cb5$e6t$07$1@news.t-online.com>

Hi,

thanks for your help. It looks much nicer now. Additionally I found a module
in the scientific python package, which does exactly what I want, i.e.
reading an array or writing it back to disk. Thanks a lot, Andreas.





From grante at visi.com  Sat Mar 24 23:36:19 2001
From: grante at visi.com (Grant Edwards)
Date: Sun, 25 Mar 2001 04:36:19 GMT
Subject: time to invest ???
References: <3abd3a17.7103358@news.vif.com> <Yrcv6.2751$45.12829@newsfeeds.bigpond.com> <vdqqbtk946qh6p7a0tfpf2kng2o0jkk7mm@4ax.com>
Message-ID: <slrn9br4j8.tf.grante@tuxtop.visi.com>

On Sun, 25 Mar 2001 03:40:16 GMT, Courageous <jkraska1 at san.rr.com> wrote:

>>Mastery of the language may take a while...
>
>Mastery of the *language* shouldn't take more than a couple of weeks,
>withstanding perhaps some of Python's esoterica. The various different
>modules and so forth are another matter altogether.
>
>If a new programmer couldn't be VERY effective in Python within a week
>or two, I'd be seriously worried about that programmer. :)

I would think most people familiar with a couple other
languages could master the commonly used language features in a
day or two.

-- 
Grant Edwards                   grante             Yow!  HUGH BEAUMONT died
                                  at               in 1982!!
                               visi.com            


From moshez at zadka.site.co.il  Mon Mar 26 17:26:44 2001
From: moshez at zadka.site.co.il (Moshe Zadka)
Date: Tue, 27 Mar 2001 00:26:44 +0200
Subject: Release 2.0.1: Heads Up
Message-ID: <E14hfRk-00051d-00@darjeeling>

Greetings, earthlings!

As Guido said in the last conference, there is going to be a bugfix release
of Python 2.0, Python 2.0.1. Originally meant to be only a license bugfix
release, comments in the Python community have indicated a need for a real
bugfix release. PEP 6[1] has been written by Aahz, which outlines a procedure
for such releases. With Guido's blessing, I have volunteered to be the
Patch Czar (see the PEP!) for the 2.0.1 release. In this job, I intend
to be feared and hated throughout the Python community -- men will 
tremble to hear the sounds of my footsteps...err...sorry, got sidetracked.

This is the first Python pure bugfix release, and I feel a lot of weight
rests on my shoulders as to whether this experiment is successful. Since
this is the first bugfix release, I intend to be ultra-super-conservative.
I can live with a release that does not fix all the bug, I am very afraid
of a release that breaks a single person's code. Such a thing will give
Python bugfix releases a very bad reputation. So, I am going to be a very
strict Czar.

I will try to follow consistent rules about which patches to integrate,
but I am only human. I will make all my decisions in the public, so they
will be up for review of the community.

There are a few rules I intend to go by

1. No fixes which you have to change your code to enjoy. (E.g., adding a new
   function because the previous API was idiotic)
2. No fixes which have not been applied to the main branch, unless they
   are not relevant to the main branch at all. I much prefer to get a pointer
   to an applied patch or cvs checkin message then a fresh patch. Of course,
   there are cases where this is impossible, so this isn't strict.
3. No fixes which have "stricter checking". Stricter checking is a good
   thing, but not in bug fix releases.
4. No fixes which have a reasonable chance to break someone's code. That
   means that if there's a bug people have a good change of counting on,
   it won't be fix.
5. No "improved documentation/error message" patches. This is stuff that
   gets in people's eyeballs -- I want bugfix upgrade to be as smooth
   as possible.
6. No "internal code was cleaned up". That's a good thing in the development
   branch, but not in bug fix releases.

Note that these rules will *not* be made more lenient, but they might
get stricter, if it seems such strictness is needed in order to make
sure bug fix releases are smooth enough.

However, please remember that this is intended to help you -- the Python
using community. So please, let me know of bugfixes that you need or want
in Python 2.0. I promise that I will consider every request.
Note also, that the Patch Czar is given very few responsibilities ---
all my decisions are subject to Guido's approval. That means that he
gets the final word about each patch.

I intend to post a list of patches I intend to integrate soon -- at the
latest, this Friday, hopefully sooner. I expect to have 2.0.1a1 a week
after that, and further schedule requirements will follow from the
quality of that release. Because it has the dual purpose of also being
a license bugfix release, schedule might be influenced by non-technical
issues. As always, Guido will be the final arbitrator.

-- 
"I'll be ex-DPL soon anyway so I'm        |LUKE: Is Perl better than Python?
looking for someplace else to grab power."|YODA: No...no... no. Quicker,
   -- Wichert Akkerman (on debian-private)|      easier, more seductive.
For public key, finger moshez at debian.org  |http://www.{python,debian,gnu}.org



From digitig at cix.co.uk  Tue Mar 20 13:40:00 2001
From: digitig at cix.co.uk (Tim Rowe)
Date: Tue, 20 Mar 2001 18:40 +0000 (GMT Standard Time)
Subject: What IDE are you'all using for Python 2.1 for Windows?
References: <mailman.985033225.14770.python-list@python.org>
Message-ID: <memo.20010320184015.54039E@digitig.compulink.co.uk>

In article <mailman.985033225.14770.python-list at python.org>, 
Jaeho.Lee at brooks.com (Lee, Jaeho) wrote:

> emacs.
> 
> You can get automatic indentation (not so dramatic compared with other
> languages, though) and highlights syntax.
> 
> Optionally you can use IDLE or Pythonwin along with emacs. But usually 
> emacs
> is enough for me.

Ditto, though I do use Pythonwin too. If only emacs understood 
triple-quoted strings (one of the things I consider unpythonic that seems 
to have been in Python from the start, FWIW!)


From danielk at aracnet.com  Mon Mar 19 09:00:40 2001
From: danielk at aracnet.com (Daniel Klein)
Date: Mon, 19 Mar 2001 06:00:40 -0800
Subject: The order of classes
Message-ID: <9r3cbt04tg3c3li151416n9b2a0gg3eccb@4ax.com>

Is it a requirement to place classes in a particular order if one class uses
another in its __init__ method?

To illustrate this in its simplest form...

class bar:
    def __init__(self, f = foo()):
        pass

class foo:
    def __init__(self):
        pass

...then at the interactive prompt...

>>> import foobar
Traceback (innermost last):
  File "<pyshell#0>", line 1, in ?
    import foobar
  File "C:\DanielK\python\foobar.py", line 1, in ?
    class bar:
  File "C:\DanielK\python\foobar.py", line 2, in bar
    def __init__(self, x = foo()):
NameError: There is no variable named 'foo'

...however, it works if I reverse the order of the classes and put 'foo' before
'bar'.

I would like to keep my classes in a particular order but it does not appear
that I can do this in all cases.

Thanks,
Daniel Klein


From sheila at spamcop.net  Sun Mar 18 00:43:39 2001
From: sheila at spamcop.net (Sheila King)
Date: Sun, 18 Mar 2001 05:43:39 GMT
Subject: reading stdinput
References: <mailman.984885323.10683.python-list@python.org>
Message-ID: <o4i8bt0b32s56h21c64259a33gcn1hgvut@4ax.com>

On Sun, 18 Mar 2001 11:11:00 +0930 (CST), Michael Hall <admin at mulga.com.au>
wrote in comp.lang.python in article
<mailman.984885323.10683.python-list at python.org>:

:How do I read standard input into a Python script? I'm trying to do the
:same as 'read' in a shell script to create a variable. I've played about
:with various combinations of 'sys.stdin.read()' and such but no go.

Here are two lines from a script I use for filtering my incoming e-mail:

	origheaders=rfc822.Message(sys.stdin, 0)
	raw_body = sys.stdin.read()

You have to import the sys module to use that.

sys.stdin is a file object. It behaves like a file pointer (fp).

Later in my script, I have the following line:

	body = string.join(raw_body, "")

This takes the raw_body, and joins it into one single string, instead of a
list of strings.

:If there is a better list for newbie questions like this, please let me
:know.

So far as I can tell (I'm a newbie, too), this list/newsgroup is fine. But,
you could also try the Python Tutor mailing list, as well. Info is here:

http://mail.python.org/mailman/listinfo/tutor

--
Sheila King
http://www.thinkspot.net/sheila/
http://www.k12groups.org/



From db3l at fitlinxx.com  Fri Mar  9 21:09:08 2001
From: db3l at fitlinxx.com (David Bolen)
Date: 09 Mar 2001 21:09:08 -0500
Subject: Divorcing Command Line and GUI
References: <mailman.984160633.14724.python-list@python.org> <xqxae6ulqev.fsf@cola.ca.boeing.com>
Message-ID: <uy9ue74u3.fsf@ctwd0143.fitlinxx.com>

Harry George <hgg9140 at cola.ca.boeing.com> writes:

> A couple of options:
> 
> 1. The Model-View-Controller pattern, where the model is your
> underlying application engine and you have two views: commandline and
> gui.
> 
> You will probably need as abstract "user interface" class which
> defines the available callbacks, and then subclass the comline and gui
> off that.  In the abstract case, you need ways to let the engine get
> the data it wants (modal dialogs), and for the ui to trigger actions
> in the engine (modeless actions).  

Or if you don't want to go the whole MVC setup, Python is still
extremely friendly to having a shared module between command line and
GUI parent modules, and even in general to embed the command line code
right with the main module.

All of the functions within a module can be treated essentially as
methods of that module object, when that module is imported by
another.  So if, for example, you've got a module that has a small
entry point (within the if __name__ == "__main__" block) that just
handles the command line and then executes via other functions/classes
in the module, your GUI can just import that top script and then
directly call the methods as appropriate - bypassing the block used
from the command line.

It can be pretty simple to design a module this way such that it runs
fine from the command line, but under a GUI, it's imported and its
functions used directly as dictated by the GUI.  You don't need to
worry as much about callbacks and what not this way since the module
is just providing support code used by the GUI.  Although depending on
the code involved, you may still separate out much of the action from
your main GUI thread for performance reasons.

The really neat thing is that while you can make this really effective
if you've designed with this in mind from the beginning, it can be
amazingly effective even with random Python scripts that you never
originally thought would be used as modules by later scripts.  About the
only place you can hurt yourself is by placing too much true script
application logic in the if __name__ block, or if you place command
line logic outside of such a block.

> 2. Keep the comline app as is.  Run it from the gui (e.g., via pipes).

And this case doesn't necessarily have to imply that the user even
knows that the raw command line tool is running beneath the scenes.
You can trap all of the output from the command line application and
parse/massage it into some other format for inclusion in the GUI.  If
not normally that friendly for parsing, you can always add a hidden
option to the command line tool to produce its output in a
parsing-friendly way.

Of course, you can also just re-route all of the output from the tool
into a common GUI element like a text or list box just for display
purposes.

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


From kens at sightreader.com  Mon Mar 19 18:16:49 2001
From: kens at sightreader.com (Ken Seehof)
Date: Mon, 19 Mar 2001 17:16:49 -0600
Subject: Cookies
Message-ID: <000e01c0b0ca$ad230650$04090a0a@upcast.com>

Thanks for your help and persistence.  I have come to an adequate
compromise solution. . . .

"David Bolen" <db3l at fitlinxx.com> writes:
> "Ken Seehof" <kens at sightreader.com> writes:
>
> > > Or, if it isn't going to connect to the server, how about just letting
> > > the Python application open up it's own web page internally to run
> > > appropriate code on your web server to generate the identifier (which
> > > could run as a server script), and then Python could parse it right
> > > from the resulting HTML information.
> >
> > Again, it's the same catch-22.  I need the server to generate the same
> > id that the javascipt generated earlier, but I need each client to have
> > a unique id.
>
> I'm still somewhat confused as to the requirement that your javascript
> needs the identifier - is there something else it gets used for as
> part of the web page other than to generate it on behalf of the
> application?

Yes, the page where the user downloads the application also contains
a form (to be filled out while the user is waiting for the download).
The data in the form is written to a database on the server keyed by
the newly generated identifier.  When the application is downloaded,
it uses the same identifier to query the server database for the form
data.  Of course since there are multiple clients, the server needs to
know which client is making the request.

The solution I've ended up with is for the server to generate the id,
the web page to display it, and the application to prompt the user to
manually type it in.

It would be nice to skip the step where the application prompts the
user (hence the search for a cookie oriented solution), but now that
I see it in action it's not really a big deal (i.e. problem solved, unless
there's a really clever idea that doesn't involve any other compromise).

> If you support generating the id the first time via a script running
> from a normal browser, couldn't you do the same thing as I suggested
> above by letting your application itself be that browser - you could
> just generate the id server side and parse the results right within
> your application, so no need to transfer the id at all.
>
> In other words, why not just use the web page to download the app, and
> that's it.  Don't depend on the web page and browser access for
> anything else.  Let your application connect to your server (since you
> mention it'll do it anyway) and during its first execution get the
> server to generate the id and then use it from there.

I want the browser to be the user interface for the form because the
the application must be very lightweight (i.e. not be a GUI).  There are
other reasons as well to do it this way.  The executable comes from a
possibly overburdened central server, but the data entry form (which
may be customized) is on any of several custom servers (which don't
have enough security to offer an executable).

> If you need to have the first generation take place in the context of
> an HTTP session, that can still be encapsulated by your Python
> application rather than depending on a separate Javascript execution
> in a different browser.

Thanks again everyone,
- Ken



-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20010319/49d94153/attachment.html>

From bsb at winnegan.de  Mon Mar  5 10:07:42 2001
From: bsb at winnegan.de (Siggy Brentrup)
Date: 05 Mar 2001 16:07:42 +0100
Subject: urllib.urlopen and proxies
In-Reply-To: <008301c0a57d$1331cde0$a100000a@local>
References: <vse4atgcgcfafth7c5tq3po7srd3l32f3h@4ax.com>
	<008301c0a57d$1331cde0$a100000a@local>
Message-ID: <871ysc8da9.fsf@winnegan.de>

"Chris Gonnerman" <chris.gonnerman at usa.net> writes:

> It *appears* (from reading the source) that username and password
> information for
> the proxy server must be passed as part of the URL:
> 
>     http://username:password at my.host.com/index.html
> 

I suggest using urllib2, changing the proxy setup described in the
module's docstring to

  proxy_support = urllib2.ProxyHandler({"http" :
                                        "http://user:pw at proxy:port"})

Untested too, since my proxy doesn't require authorization.

HTWorks
  Siggy




From aleaxit at yahoo.com  Tue Mar 13 11:07:38 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Tue, 13 Mar 2001 17:07:38 +0100
Subject: This math scares me
References: <3aad4062.1290021@News.CIS.DFN.DE> <Hwbr6.1139$Tg.127571@ruti.visi.com> <3aad4485.2348499@News.CIS.DFN.DE> <jwbnews-7CC03F.19050312032001@corp.supernews.com> <yu993dcimjt8.fsf@europa.research.att.com> <slrn9asbvc.dfq.neelk@alum.mit.edu>
Message-ID: <98lgkc02dso@news1.newsguy.com>

"Neelakantan Krishnaswami" <neelk at alum.mit.edu> wrote in message
news:slrn9asbvc.dfq.neelk at alum.mit.edu...
    [snip]
> Point Numbers Accurately_ and Byrger and Dybvig's _Printing Floating
> Point Numbers Quickly and Accurately_ and submit it as a patch I don't

It's Burger, not Byrger (just in case someone should be searching
or quoting this -- the analogy with Dybvig is hard to resist!-),
and I think it's in the proceedings of the 1996 SIGPLAN conference.


Alex





From hinsen at cnrs-orleans.fr  Thu Mar 29 05:40:30 2001
From: hinsen at cnrs-orleans.fr (Konrad Hinsen)
Date: 29 Mar 2001 12:40:30 +0200
Subject: Numeric question, putting one matrix into another
References: <3AC30D1C.620E197B@psi.ch>
Message-ID: <m3u24crh7l.fsf@chinon.cnrs-orleans.fr>

Jorn Verwey <jorn.verwey at psi.ch> writes:
> Using Numeric-arrays I have made a (3x3) rotation-matrix. Now I would like to put this rotation
> matrix in a (4x4) transformation matrix.

That's easy:

   rotation = Numeric.zeros((3,3), Numeric.Float)
   transformation = Numeric.zeros((4,4), Numeric.Float)

   transformation[:3, :3] = rotation

BTW, you might be interested in the module
Scientific.Geometry.Transformation, which is part of Scientific Python
(http://starship.python.net/crew/hinsen/scientific.html). It handles
linear transformations with some added luxury.
-- 
-------------------------------------------------------------------------------
Konrad Hinsen                            | E-Mail: hinsen at cnrs-orleans.fr
Centre de Biophysique Moleculaire (CNRS) | Tel.: +33-2.38.25.56.24
Rue Charles Sadron                       | Fax:  +33-2.38.63.15.17
45071 Orleans Cedex 2                    | Deutsch/Esperanto/English/
France                                   | Nederlands/Francais
-------------------------------------------------------------------------------


From robin at jessikat.fsnet.co.uk  Sat Mar  3 14:23:11 2001
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Sat, 3 Mar 2001 19:23:11 +0000
Subject: the demise of 'from foo import * and its implications?
References: <3kVBAfATrDo6EwEM@jessikat.fsnet.co.uk>
 <mailman.983582584.8198.python-list@python.org>
 <z8dAGJAyqLo6EwlX@jessikat.fsnet.co.uk>
 <2V8o6.1383$8q3.219969@news1.rdc1.az.home.com>
Message-ID: <AQyhzIAfSUo6EwQ9@jessikat.fsnet.co.uk>

In article <2V8o6.1383$8q3.219969 at news1.rdc1.az.home.com>, Tim Hochberg
<tim.hochberg at ieee.org> writes
>
>"Robin Becker" <robin at jessikat.fsnet.co.uk> wrote in message
>
>[SNIP]
>
>> If we really want to improve visibility we could try to make
>>
>> >>> def bingo(a):
>> ...     if a<=1: return 0
>> ...     return bingo(a-1)*a
>> ...
>>
>> work naively, but I'm fairly sure this will be rejected as unpythonic or
>> somesuch.
>
>
>?????
>
>In what sense doesn't this work now? Under 2.0 or 1.5.2 or even, I think,
>under 0.9 this 'works' in the sense that it iterates for a while then
>returns 0. If one replaces "return 0" with "return 1" it even returns the
>factorial as I assume it should.
>
>What doesn't work is the following:
>
>def spam(n):
>   def fact(a):
>      if a <= 1: return 1
>      else: return fact(a-1)*a
>   return fact(n) + 42
>
>With the current three level scope rule this fails, so a function that works
>fine when defined at the top level fails miserably when defined at an
>intermediate level. Given your earlier statement that:
.....

you're right of course, this is what I meant. I haven't played with the
new statically scoped stuff, but I probably wrongly assumed that bingo
would remain invisible inside bingo.

>>my preference would be that statements legal in one context should be
>>legal in another where they make sense.
>
>This seems like it would be of concern to you as well.
>
>My understanding is we can have two of the following three options:
>
>1. Decent performance (~current performance)
>2. "Real" lexical scoping
>3. Unrestricted use of  "from X import *" at other than module scope.
>
>Given those choices I'd choose 1 and 2. Which is fortunate for me, since
>that's what were going to get anyway.
...

I don't really disagree. The optimisation is currently discardable in
favour of external language simplicity. The dynamic nature of python
bindings is somehow being more compromised with the new scope rule. I
hadn't really appreciated before now that python wishes to enforce a
fixed scope for references (for optimisation purposes) and that itself
seems to me to be wrong as it's more difficult as a concept than fixed
type. I get told about reference semantics and then find that these
pointers are somehow scope const.  Oh well :(

>Going off on a tangent: In listening to the various debates here, and when I
....
>
>If you've put up with my rambling to this point I congratulate you.
>
>Ramblin on,
>
>-tim
>
>
>
>
>
>
>
>
>.
>
>

-- 
Robin Becker


From fgeiger at datec.at  Tue Mar 27 14:21:22 2001
From: fgeiger at datec.at (Franz GEIGER)
Date: Tue, 27 Mar 2001 21:21:22 +0200
Subject: wxPython: How to get a child by name?
Message-ID: <99qpce$ljk$1@newsreaderg1.core.theplanet.net>

I currently try to polish some of my apps with a GUI and decided to go for
wxPython.

I use wxDesigner for the layout. As it generates code after every change, I
named all the text controls and checkboxes etc. of the dialog hosting all
those controls by applying SetName() to the controls. The Dialog is small,
so I do not use validators yet.

To get a text control of the dialog I thought I could write
theValue = self.FindWindowByName("tcSurname").GetValue().
This raises an exception, because the return value is of type wxWindowPtr
which does not have such a method.

After a while I found a method which worked, at least for text controls:
GetLabel().

But for checkboxes GetLabel() returns the button's text and not the boolean
value checked/unchecked. This were returned by GetValue(), but GetValue()
... (see above).

For a "quick" hack in a MSVC project I would downcast the returned pointer
of GetDlgCtrl() to be able to call the subclassed object's methods. But how
do I proceed here?

Any hints greatly welcome.

Best regards
Franz GEIGER








From objectway at divalsim.it  Wed Mar 14 11:50:05 2001
From: objectway at divalsim.it (Nicola Musatti)
Date: Wed, 14 Mar 2001 17:50:05 +0100
Subject: Who's minister of propaganda this week?
References: <B98041329020E007.7DA84CAD9D0562AB.817BBC572F480869@lp.airnews.net> <98nu1m$12e@dispatch.concentric.net> <660ACCA1E4816F7E.C359CDF4168EF3FB.3B6B05538D619770@lp.airnews.net>
Message-ID: <3AAFA13D.76E32C3F@divalsim.it>

Hi,
I'm a Python curious not yet sold to the advantages of typeless
languages; I tend to think that typelessness is suited to a set of tasks
that has only a limited overlap with the tasks for which static typing
is preferable. I have a question and a provocation for you.

Cameron Laird wrote:
> 
> In article <98nu1m$12e at dispatch.concentric.net>,
> Phlip  <phlip_cpp at my-deja.com> wrote:
>                         .
> >Robert C. Martin meant two things. A> If you write wall-to-wall unit tests
> >when you code, if you use a typeless language you will add bullet-proof
> >features faster than if you use a statically typed language, even though
> >the former has less built-in error checking at type conversion time. The
> >latter supports refactoring easier.

But those things that the compiler used to check for you may well have
become new tests for you to write: if the trade-off between the decrease
in actual code size and the increase in test number is negative, even
the better support for refactoring turns out to be an illusion. 

[...]
>   Expressiveness ... is a far better aid to
>   correctness ...

Here I'm puzzled, but maybe it's only a "language" problem. What do you
mean by "Expressiveness"? I ask because until now I thought
expressiveness was an advantage of strongly typed languages (namely,
C++) over typeless ones.

Best regards,
Nicola Musatti


From sholden at holdenweb.com  Mon Mar  5 12:08:23 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Mon, 5 Mar 2001 12:08:23 -0500
Subject: the demise of 'from foo import * and its implications?
References: <3kVBAfATrDo6EwEM@jessikat.fsnet.co.uk><mailman.983582584.8198.python-list@python.org><z8dAGJAyqLo6EwlX@jessikat.fsnet.co.uk> <mailman.983746566.27513.python-list@python.org> <suPo6.16824$ML1.841861@e420r-atl2.usenetserver.com> <3AA3C82F.E2F6CA7F@alcyone.com>
Message-ID: <YNPo6.21752$1D5.847097@e420r-atl1.usenetserver.com>

"Erik Max Francis" <max at alcyone.com> wrote in message
news:3AA3C82F.E2F6CA7F at alcyone.com...
> Steve Holden wrote:
>
> > I did begin to wonder whether it might be possible to specify a name
> > prefix,
> > along the lines of
> >
> >     from modulewithaverylongname import * prefix sh_
> >
> > to allow
> >
> >     sh_i
> >
> > as a reference to the module's "i", but I'm not sure that the gain is
> > worth
> > the pain. Even if such a proposal were to be PEP-ified and accepted it
> > could
> > not be mandated because that would break humongous amounts of existing
> > code.
> >
> > Thoughts?
>
> I don't see how changing a . to an _ really helps anything, and all it
> does is make the language more complex, and add a new keyword which is
> _very_ likely to be used in existing code.
>
I'll take this as a "me, too", then! Of course if the prefix didn't end in
an inderscore it would be easier to type, but your arguments still apply.

so-that's-another-PEP-I-needn't-write-ly y'rs  - steve





From akuchlin at mems-exchange.org  Sat Mar 17 12:08:36 2001
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 17 Mar 2001 12:08:36 -0500
Subject: ZPT, a next-generation template technology
References: <98ufbu$kkh$1@uranium.btinternet.com>
Message-ID: <3ditl8migb.fsf@ute.cnri.reston.va.us>

"Hamish Lawson" <hamish_lawson at btinternet.com> writes:
> The approach that the ZPT project is pursuing is to attach the control-
> flow, formatting and placeholder directives as *attributes* of standard
> XHTML tags. The example above thus becomes something like:

After a glance at the initial specs, I like it; it looks like a
simpler processing language than XSLT, and one more suited to
interfacing with some other imperative language.  Given the recent
squabbling over the scripting features in XSL, there's probably a
significant niche for something like this.

You've notified various generic scripting communities, but you should
also really let the XML community know about this work; they're more
likely to take an interest.  Send a note to mailing lists such as
xml-dev, or to news sites such as xmlhack.com, xml.com, and
xml.coverpages.org.  comp.text.xml exists, but discussion is
relatively low-level and few serious developers hang out there.

--amk



From syver at NOSPAMcyberwatcher.com  Thu Mar 15 15:36:40 2001
From: syver at NOSPAMcyberwatcher.com (Syver Enstad)
Date: Thu, 15 Mar 2001 21:36:40 +0100
Subject: windows script host and python
References: <%tKr6.2570$pX.26174@news1.onlynews.com> <98oq3r$1e0$1@troll.powertech.no> <jvSr6.34$cj1.4417@news1.onlynews.com>
Message-ID: <98r991$t4f$1@troll.powertech.no>

Ok, if you're using SQL server it might interest you to know that you can
use Python (ActiveScripting) in the SqlAgent, a kind of scheduler that is
part of the SQL server package. This is a nice way to setup running of
scripts that are related to clean up of database files, but you aren't
restricted in any way to database related things.

"Mike Carifio" <carifio.nospam at nospam.usys.com> wrote in message
news:jvSr6.34$cj1.4417 at news1.onlynews.com...
> > Just curious but what do plan to acheive by using python with wsh that
you
> > couldn't do using normal .py files?
>
> I was testing Microsoft's assertion that "other" scripting engines worked.
> Trust but verify.
>
>




From tdelaney at avaya.com  Sun Mar  4 20:45:06 2001
From: tdelaney at avaya.com (Delaney, Timothy)
Date: Mon, 5 Mar 2001 12:45:06 +1100 
Subject: I come to praise .join, not to bury it...
Message-ID: <B43D149A9AB2D411971300B0D03D7E8B0FBF11@natasha.auslabs.avaya.com>

> I have not heard ANY _technical_ arguments opposed to
> this last time the discussion went around -- nothing but
> vague aesthetic, "it should be the other way", "I find it
> ugly" kind of complaints.  Unless and until further notice,
> then, I class these together with the complaints of people
> who dislike 0-based indexing, on similar vague bases -- as
> 0-based indexing _works_ better than 1-based, then, for
> me, it is a superior choice.  Aesthetics is in the eye of the

I truly have no complaints with the technical merits of ''.join().

However, I have *severe* complaints with the aesthetics and consistency.
Quite simply, the way we look at things tends to be

"result" = "thing to be acted on to get the result" modified by "thing which
does the action"

This is consistently followed in the standard python libraries in almost all
cases ... but ''.join() is a glaring inconsistency. As a result, it seems
unintuitive and ugly.

This is why I prefer string.join(sequence, string). It is now syntactic
sugar for the above, and documented as such. However, it is *important*
syntactic sugar. It is known that the method is called polymorphically on
the second argument (a standard interface) but it feels that it is the first
argument which is the basis for the expression. This remains consistent with
the rest of the standard libraries.

OTOH, I suggest an alternative syntax ...

(sequence)join.''

There. Everyone should be happy with that ;)

Tim Delaney
Avaya Australia
+61 2 9352 9079

All recipient(s) of this email have permission to forward or reply to this
email, quoting this email in full or in part.



From emile at fenx.com  Sat Mar 24 17:24:51 2001
From: emile at fenx.com (Emile van Sebille)
Date: Sat, 24 Mar 2001 14:24:51 -0800
Subject: [Python-Dev] Anyone using weakrefs?
References: <mailman.985469191.25050.python-list@python.org>
Message-ID: <99j71s$1dmph$1@ID-11957.news.dfncis.de>

Thanks!  That's cool.  I'd wondered if there was a way to re-swizzle in
objects if they still existed from the ID alone, and it looks like this can
do it.  I've got a process that takes up about 700Mb of my system's 512Mb
memory, and for testing runs about 5 1/2 hours.  It runs about 10 minutes on
the client's 2Gb system, even though their processor is 1/3 the speed.  I
never realized that thrashing could make such a difference.

At this point, it's in place and they've paid the price, but I'll start
playing with weak references to see if it improves my local test
performance.  I'm practicing one-man-shop XP ;-) and when the both of us get
into it, being able to do more than a single iteration per day would really
help.

--

Emile van Sebille
emile at fenx.com

---------
"Tim Peters" <tim.one at home.com> wrote in message
news:mailman.985469191.25050.python-list at python.org...
> [Fred Drake]
> > Is anyone out there playing with the weak references support yet?
> > I'd *really* appreciate receiving a short snippet of non-contrived
> > code that makes use of weak references to use in the documentation.
>
> Alas, it's kinda like asking for a "short snippet" illustrating
coroutines:
> the need isn't apparent in tiny examples.
>
> Still, I've been happily playing w/ this little module:
>
> import weakref
> id2obj_dict = weakref.mapping()
>
> def remember(obj):
>     id2obj_dict[id(obj)] = obj
>
> def id2obj(id):
>     return id2obj_dict.get(id, None)
>
> In various introspective tools (like Cyclops.py), there's a need to keep
> track of "other peoples'" class instances.  Before weakrefs, it was
> impossible to do this without keeping the instances artificially alive.
Now
> I can pass arbitrary instances to remember() above, then just use their
ids
> internally.  This doesn't keep the objects alive, but id2obj() above can
> still get the objects back so long as they're still alive.
>
> I suppose I could work with weakrefs directly instead of indirecting
through
> id()s, but most code of this nature I have is already mucking w/ ids
anyway
> (ints are small, fast, cheap and make great dict keys themselves).
>
>




From spankme at ucla.edu  Tue Mar 20 16:02:47 2001
From: spankme at ucla.edu (Greg Kobele)
Date: Tue, 20 Mar 2001 13:02:47 -0800
Subject: Einstein's Riddle
References: <oNPq6.396644$ge4.138479566@news2.rdc2.tx.home.com>
Message-ID: <998gru$du1$1@persian.noc.ucla.edu>

You actually don't need to have this tidbit of info:
>
> The man who smokes Blends has a neighbor who drinks water.
>
The problem then becomes a little more fun to do if you don't.

Greg.


Steve Graham <js.graham at home.com> wrote in message
news:oNPq6.396644$ge4.138479566 at news2.rdc2.tx.home.com...
> Who would be interested in using his/her brain (and his computer) to solve
> the following?
>
>
> Steve Graham
>
> ===
>
> Einstein's Riddle
>
> Albert Einstein wrote this riddle this century [ed. 20th century].  He
said
> 98% of the world could not solve it.
>
> There are 5 houses in 5 different colors.  In each house lives a person
with
> a different nationality.  The 5 owners drink a certain type of beverage,
> smoke a certain brand of cigar, and keep a certain pet.  No owners
> have the same pet, smoke the same brand of cigar or drink the same
beverage.
>
> The question is: "Who owns the fish?"
>
> Hints:
>
> The Brit lives in the red house.
>
> The Sweed keeps dogs as pets.
>
> The Dane drinks tea.
>
> The green house is on the left of the white house.
>
> The green house's owner drinks coffee.
>
> The person who smokes Pall Mall rears birds.
>
> The owner of the yellow house smokes Dunhill.
>
> The man living in the center house drinks milk.
>
> The Norwegian lives in the first house.
>
> The man who smokes Blends lives next to the one who keeps cats.
>
> The man who keeps the horse lives next to the man who smokes Dunhill.
>
> The owner who smokes Bluemasters drinks beer.
>
> The German smokes Prince.
>
> The Norwegian lives next to the blue house.
>
> The man who smokes Blends has a neighbor who drinks water.
>
>
>
>
>
>
>
>
>




From colinmeeks at home.com  Thu Mar  1 08:05:47 2001
From: colinmeeks at home.com (Colin Meeks)
Date: Thu, 01 Mar 2001 13:05:47 GMT
Subject: ANN: GuestBook 1.0.0
Message-ID: <LOrn6.265427$KP3.67810961@news3.rdc1.on.home.com>

Continuing on from our release of MailList (the mailing list management
script http://www.programmedintegration.com/pmaillist.htm) comes our new
script for
putting a guestbook on your website.

GuestBook 1.0.0 (http://www.programmedintegration.com/pguestbook.htm) was
written
to provide a simple guestbook for our family website (http://www.meeks.ca).
A full sample
is included. GuestBook was written and tested under Python 1.52 and isn't
dependent on
any non standard Python modules.

All our Python software is written on an as needed basis and we are pretty
much learning
Python as we go along, so the code may not be the most efficient.  Please
get back to us
if you have any ideas for fixes or improvements and we will make sure they
are incorporated
into future releases. We would also like to hear from you if you incorporate
GuestBook on your
website.

GuestBook is provided with full source code and is released into the public
domain.

Colin Meeks
colin at programmedintegration.com




From thomas at cintra.no  Mon Mar 19 08:07:07 2001
From: thomas at cintra.no (Thomas Weholt)
Date: Mon, 19 Mar 2001 14:07:07 +0100
Subject: Looking for web indexer in python
References: <roy-AD5737.12020617032001@news.panix.com>
Message-ID: <1wnt6.2190$mh4.123235@news3.oke.nextra.no>

I got a indexer-project going, though still in the testing/alpha stage. It's
a general purpose indexer, but it's trivial to build it into a full-text
web-indexer with a cgi-frontend.

Look for the SnipThis project at sourceforge or download the latest version
from http://www.weholt.no/files/hs/hs_0_1_5_2.zip

Best regards,
Thomas Weholt

"Roy Smith" <roy at panix.com> wrote in message
news:roy-AD5737.12020617032001 at news.panix.com...
> The software development group I work in has a small internal web site (a
> few hundred documents) which we don't currently have any way of searching.
> I'd like to put an automated indexer/search system in place.  If possible,
> I'd like to use python, since I think it would be a nifty way to introduce
> the group to the language.  There's no funding for this, so it has to be
> free software.
>
> A cursory look around didn't turn up anything that would fit the bill, but
> I have to assume somebody's already done a free python web indexer.  Any
> suggestions?




From montana at stats.warwick.ac.uk  Thu Mar 29 22:12:44 2001
From: montana at stats.warwick.ac.uk (Giovanni Montana)
Date: Fri, 30 Mar 2001 04:12:44 +0100
Subject: Python and BLT Graph - win98
Message-ID: <9a0tku$8qr$1@wisteria.csv.warwick.ac.uk>

Hi all,

I installed the BLT Graph library written in Tcl for Python 2.0, which I use
together with Pmw megawidgets.

My problem is the following: I am able to load the module "Pmw.py" and run
some of the BLT.PMW examples with no problem ... but only once! I cannot
interrupt a running application, and the second time I wish to run the
same - or a different - script, the BLT doesn't seem to work anymore ....

Any suggestions or idea PLEASE?

Thanks a lot!
Giovanni





From mwh21 at cam.ac.uk  Thu Mar 29 17:57:49 2001
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 29 Mar 2001 23:57:49 +0100
Subject: diff lists
References: <3AC1DD4E.4BA7D421@aon.at>
Message-ID: <m3zoe4b2tu.fsf@atrus.jesus.cam.ac.uk>

Oliver Vecernik <vecernik at aon.at> writes:

> Hi,
> 
> I've got following two lists:
> 
> ['a', 'b', 'c', 'd', 'e', 'f']
> ['e', 'c', 'f']
> 
> I'd like to have the result:
> 
> ['a', 'b', 'd']
> 
> The list need not to be ordered. ['d', 'a', 'b'] will also be ok. What
> is the most effective way to achive that result?

Tangentially, does anyone know of any good algorithms for "edit
distance" between two sequences?  E.g. if I have 

"abcdef"

and want to get to

"abQUACKcde"

I want to get the answer back "insert 'QUACK' at position 3 and delete
a character at position 11".

"Good" means "pretty quick", here.

Cheers,
M.

-- 
  If your telephone company installs a system in the woods with no
  one around to see them, do they still get it wrong?
                                 -- Robert Moir, alt.sysadmin.recovery


From kalle at gnupung.net  Tue Mar 20 06:07:18 2001
From: kalle at gnupung.net (Kalle Svensson)
Date: Tue, 20 Mar 2001 12:07:18 +0100
Subject: The most important modules
In-Reply-To: <Pine.GSO.4.21.0103191732210.22234-100000@laplace>; from mrq@for.mat.bham.ac.uk on Mon, Mar 19, 2001 at 05:39:10PM +0000
References: <Pine.GSO.4.21.0103191732210.22234-100000@laplace>
Message-ID: <20010320120718.A10259@father>

Sez Martyn Quick:
> I've seen a few comments along the lines of "When you are a newbie it will
> take a while to realize which the important modules are."  As someone who
> is in the middle of wading through Chapter 3 of the Library Reference, it
> would be very helpful to know which ones are particularly worth
> noting.  (Presumably  sys, (c)Pickle and shelve, from the ones I've so far
> in the chapter - but what other treats are out there for me?)

That very much depends on what you're trying to do.  I tend to use "os" a
lot, because most of my scripts need it.  "traceback" can be useful.
If you're a Unixer, check out "rlcompleter".  "getopt" is good.
But most importantly (IMHO):
http://www.python.org/doc/current/lib/builtin.html

Peace,
  Kalle
-- 
Email: kalle at gnupung.net     | You can tune a filesystem, but you
Web: http://www.gnupung.net/ | can't tune a fish. -- man tunefs(8)
PGP fingerprint: 0C56 B171 8159 327F 1824 F5DE 74D7 80D7 BF3B B1DD
 [ Not signed due to lossage.  Blame Microsoft Outlook Express. ]



From mwh21 at cam.ac.uk  Sun Mar 11 18:06:53 2001
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 11 Mar 2001 23:06:53 +0000
Subject: Static typing (was Re: Java guy interested in Python)
References: <mailman.984334635.27345.python-list@python.org> <98glhc02sga@news2.newsguy.com> <m3hf1080or.fsf@atrus.jesus.cam.ac.uk> <98gt3f064l@news2.newsguy.com>
Message-ID: <m3elw39a7m.fsf@atrus.jesus.cam.ac.uk>

"Alex Martelli" <aleaxit at yahoo.com> writes:

> "Michael Hudson" <mwh21 at cam.ac.uk> wrote in message
> news:m3hf1080or.fsf at atrus.jesus.cam.ac.uk...
[snip]
> > > int is no "reserved word", after all.
> >
> > Yes it is.  "identifier" and "keyword" are disjoint classes of token
> > in C++, at least by my reading of the 1997-12 working draft I've just
> > dredged up with google.
> >
> > Unless I've missed some subtle joke...
> 
> Sorry, too subtle -- the point was meant to be that a #define int
> foo _would_ make it legal (and incomprehensible to anybody, if the
> #define is well-hidden enough) -- sort of trying to see who'd fail
> to spot this.  But on re-reading it falls pretty flat -- sorry!

Actually, I thought

#define int foo

was pedantically illegal C++ that was almost certain to work, but on
checking I see that it's probably legal.  I also hadn't taken this
possibility into account when replying above; catching people out with
the preprocessor isn't really fair...

Cheers,
M.

-- 
  Those who have deviant punctuation desires should take care of their
  own perverted needs.                  -- Erik Naggum, comp.lang.lisp


From aleaxit at yahoo.com  Thu Mar 29 03:56:51 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 29 Mar 2001 10:56:51 +0200
Subject: i need help on builtin function math
References: <mailman.985822299.17200.python-list@python.org>
Message-ID: <99utcp01tiu@news2.newsguy.com>

"Marcus Konermann" <m_konermann at gmx.de> wrote in message
news:mailman.985822299.17200.python-list at python.org...
> hello !
>
> i want to use the sqrt() function of the builtin funtion math in that
> kind:
>
> import math
> ...
> a=0
> a=sqrt(whatever)
>
> and than a NameError: sqrt occured and i don?t know why ? Is it
> depending on the OS (i use Windows98 and the pywin
> editor). Other functions, like pow(), from the math module are working.

There is a *built-in* function pow, which is not the same
as math.pow, and is no doubt what IS being executed here:

>>> pow
<built-in function pow>
>>> import math
>>> math.pow
<built-in function pow>
>>> pow is math.pow
0
>>> math.pow(2, 35000)
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
OverflowError: math range error
>>> pow(2, 35000)
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
OverflowError: integer exponentiation
>>>


> Can anyone help me with my problem ?

Either:
    import math
    a=math.sqrt(whatever)
or:
    from math import sqrt
    a=sqrt(whatever)


Alex





From bernhard at abnoba.intevation.de  Tue Mar  6 08:55:49 2001
From: bernhard at abnoba.intevation.de (Bernhard Reiter)
Date: 6 Mar 2001 13:55:49 GMT
Subject: Can Python replace Visual Basic? Should it?
References: <5DZo6.11126$hn5.1578410@news1.rdc1.mb.home.com>
Message-ID: <982q95$3gs$1@newsserver.rrzn.uni-hannover.de>

In article <5DZo6.11126$hn5.1578410 at news1.rdc1.mb.home.com>,
	"Brad Bollenbach" <bbollenbach at homenospam.com> writes:

>     Can Python replace Visual Basic? Should it?

Yes. Yes.

> As far as I can see, they both seem close to equal in their not-so-simple
> ways of distributing apps to end-users, which VB programmers have seem to
> overcome to some degree and so Python programmers should and probably do
> likewise. 

Did you see the nice windows installers you can make with innosetup?
Checkout out cowTipper for an example: 
	http://bluesine.com/archives/software/cowTipper/

The main advantage for python over VB is that python is free software.
	
> What I want is a programming language that can make Windows fun
> (like C or Perl does Unix), without having to shell out the big bucks for
> full installs of VB or VC++. 

Well python worked for me.. :)

> But really, very specifically, were YOU a former Visual Basic (or
> Powerbuilder, or Delphi) programmer that is now using Python to write the
> complex GUI's VB used to handle for you? 

No. I gave up on MS products a long time ago.
But I had to control SPSS under Windows and I had a look at
documentation I could get and it was fun to do it with python in the end.

For GUIs you can use wxpython which gives you a native look and feel
under MS windows.

	Bernhard
-- 
Professional Service around Free Software                (intevation.net)  
The FreeGIS Project                                         (freegis.org)
Association for a Free Informational Infrastructure            (ffii.org)
FSF Europe                                            	  (fsfeurope.org)


From sholden at holdenweb.com  Sat Mar  3 12:22:54 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Sat, 3 Mar 2001 12:22:54 -0500
Subject: A counter-proposal to __future__ in PEP 236
References: <mailman.983610546.9172.python-list@python.org>
Message-ID: <AP9o6.6393$1D5.198210@e420r-atl1.usenetserver.com>

"Tim Peters" <tim.one at home.com> wrote in message
news:mailman.983610546.9172.python-list at python.org...
[ ... ]
>
> can-make-it-foolproof-but-not-damned-fool-proof<wink>-ly y'rs  - tim
>
Besides which, every time we invent a new and foolproof syste the world has
a way of producing a complementary new and improved fool who finds another
hole.

there's-no-end-to-improving-ease-of-use-ly y'rs  - steve





From tim.one at home.com  Fri Mar  2 13:10:25 2001
From: tim.one at home.com (Tim Peters)
Date: Fri, 2 Mar 2001 13:10:25 -0500
Subject: Unresolved symbols in MSVCRT.DLL when running a py2exe packaged program.
In-Reply-To: <4rQn6.14308$UZ4.4134344@news4.rdc1.on.home.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCAEHDJDAA.tim.one@home.com>

[Tom]
> I disagree.
>
> You can distribute system dll's.  You are expected to, and sometimes you
> have to.  You must insure that the version you load is at least
> equal to the version that you linked against.
>
> I have done this (distribute system DLL's) on all windows except WinME and
> the 16bit ones.  Unfortunately I can't say how I did it precisely, because
> I just used InstallShield.

And Python just uses the Wise installer.  They aren't that fancy under the
covers, though -- they just "follow the rules", checking version numbers,
bumping refcounts on refcounted DLLs, and arranging for a reboot if they need
to replace something.  But that stuff is ineffective on the newer systems
with SFP.

> But I believe that, where SFP exists, you must use specific and new
> API's to replace DLL's, otherwise the system will back out your changes.

My understanding is that SFP isn't so easily fooled; e.g.,

    http://support.microsoft.com/support/kb/articles/Q253/5/71.ASP

    http://www.microsoft.com/hwdev/sfp/wfp.htm

SFP won't allow replacement of protected files via just using API tricks; it
also sucks in a digital signature scheme, and only MS can supply signatures
that work.

> One solution is to specify the changes in an .INF file and let the system
> worry about these details (but this has its limitations).
>
> I assume that Python itself updates the MSVCRT.dll.

The PythonLabs Windows installer tries to, yes.

> Might look at how it does it on Win2K.

My understanding is that it can't (but Win2K ships with a "good enough" set
of DLLs, so the version check shouldn't trigger there -- yet).

BTW, I'm *glad* MS is doing SFP!  Whatever Windows Life may be in theory,
DLLs in practice cannot be mixed and matched -- you need a self-consistent
set else Very Strange Things Happen.  And with about 1,000 DLLs under a
typical Windows system directory, that's just impossible for app developers
to keep straight.  May well be impossible for MS to keep straight too, but at
least it's no longer impossible for them to *try* to keep 'em straight now.

something-to-ponder-each-time-your-win9x-crashes-ly y'rs  - tim




From aleaxit at yahoo.com  Thu Mar 15 18:42:46 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 16 Mar 2001 00:42:46 +0100
Subject: Python and ADO
References: <98r9qn$tbi$1@troll.powertech.no> <mailman.984690501.20811.python-list@python.org>
Message-ID: <98rk2k04jt@news2.newsguy.com>

"Geoff Talvola" <gtalvola at nameconnector.com> wrote in message
news:mailman.984690501.20811.python-list at python.org...
> Syver Enstad wrote:
>
> > Has anyone considered making a DB API compliant wrapper around MS ADO?
Or
> > isn't this considered worthwhile?

I _considered_ it often, but never got a round tuit... yet!


> > As ADO is the database api that I use, I could have a look at making a
> > wrapper for it if there was any interest in it and if making such a
thing is
> > not too involved.
>
> Michel Orengo already wrote a preliminary version of a DB-API wrapper for
ADO but I don't think he ever finished it.  I have a copy of it, but I have
never actually tried it.  If you can't find it in the newsgroup archives,
drop me an email and I'll send you a copy.

I wouldn't mind getting a copy -- I can't find it in the NG archives (maybe
I'm still unclear on how to use google for those...).


Alex





From andrew_dot_henshaw_at_earthling_dot_net  Wed Mar 21 23:05:25 2001
From: andrew_dot_henshaw_at_earthling_dot_net (Andrew Henshaw)
Date: Wed, 21 Mar 2001 23:05:25 -0500
Subject: Accessing Oracle from Python remotely ?
References: <Xns906AB9C901807PaCmAnRDLM@161.48.76.20> <eE3u6.776$9%.203510@newsr1.maine.rr.com> <99bj2o$u91$1@nntp9.atl.mindspring.net>
Message-ID: <tbiuhdp9eno4e9@corp.supernews.com>

Does this support  the DBI API?  If not, should it?

"Andreas Jung" <andreas at andreas-jung.com> wrote in message
news:99bj2o$u91$1 at nntp9.atl.mindspring.net...
> There are currently two different DCOracle versions available. DCOracle
> 1.3.x
> supports is mostly build on top of Oracle 7 but I works although with
Oracle
> 8i
> except features like BLOBs and CLOBs. When you need complete support
> for Oracle 8i take a look at Matt Kromers new version of DCOracle 2. It is
> still alpha so be warned :-) You can find DCO2 on www.zope.org  - search
for
> DCO2.
>
> Andreas Jung
> Digital Creations
>





From thomas at xs4all.net  Mon Mar  5 10:30:41 2001
From: thomas at xs4all.net (Thomas Wouters)
Date: Mon, 5 Mar 2001 16:30:41 +0100
Subject: the demise of 'from foo import * and its implications?
In-Reply-To: <GqQAdIAXUuo6EwyY@jessikat.fsnet.co.uk>; from robin@jessikat.fsnet.co.uk on Mon, Mar 05, 2001 at 01:00:07AM +0000
References: <3kVBAfATrDo6EwEM@jessikat.fsnet.co.uk> <mailman.983582584.8198.python-list@python.org> <z8dAGJAyqLo6EwlX@jessikat.fsnet.co.uk> <mailman.983746566.27513.python-list@python.org> <GqQAdIAXUuo6EwyY@jessikat.fsnet.co.uk>
Message-ID: <20010305163041.Z9678@xs4all.nl>

On Mon, Mar 05, 2001 at 01:00:07AM +0000, Robin Becker wrote:
> In article <mailman.983746566.27513.python-list at python.org>, Jeremy
> Hylton <jeremy at alum.mit.edu> writes
> >>>>>> "RB" == Robin Becker <robin at jessikat.fsnet.co.uk> writes:
> >
> >  RB> my preference would be that statements legal in one context
> >  RB> should be legal in another where they make sense. I cannot see
> >  RB> how using import * or eval etc in nested scopes is somehow
> >  RB> obviously bad or different from the same usage in non-nested
> >  RB> scopes. The fact that the prohibition comes from making things
> >  RB> easier for the compiler adds to my confusion.
> >
> >Using import * in function's is asking for trouble, because it makes
> >it impossible for someone reading the code to know what a variable
> >refers to.  Consider this hypothetical bit of code:
> >
> >    from string import *
> >
> >    def f(x):
> >        from amodule import *
> >        return strip(x)
> >
> ....
> 
> doesn't from foo import * make things hard to locate at the global level
> as well? exec/eval/import can always do this surely.

The global namespace is a 'last call' namespace. The compiler doesn't (and
can't, since you can modify global namespaces from outside the namespace)
keep track of global names because it isn't necessary. Neither is true for
function namespaces.

-- 
Thomas Wouters <thomas at xs4all.net>

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



From py.list at localhost.localdomain  Sun Mar 25 02:07:53 2001
From: py.list at localhost.localdomain (py.list at localhost.localdomain)
Date: Sun, 25 Mar 2001 16:37:53 +0930 (CST)
Subject: time to invest ???
In-Reply-To: <vdqqbtk946qh6p7a0tfpf2kng2o0jkk7mm@4ax.com>
Message-ID: <Pine.LNX.4.10.10103251637190.23294-100000@localhost.localdomain>


> Mastery of the *language* shouldn't take more than a couple of weeks,
> withstanding perhaps some of Python's esoterica. The various different
> modules and so forth are another matter altogether.
> 
> If a new programmer couldn't be VERY effective in Python within a week
> or two, I'd be seriously worried about that programmer. :)




From fredrik at pythonware.com  Mon Mar 12 16:50:15 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Mon, 12 Mar 2001 21:50:15 GMT
Subject: Tkinter canvas double buffering
References: <rh8j89.pu7.ln@demonseed.net>
Message-ID: <rwbr6.17912$Qb7.3115491@newsb.telia.net>

jason petrone wrote:
> Is there an easy way to do double buffering with the canvas?  The only way
> I see is rasterizing it offscreen myself into a BitmapImage, then drawing that.
>
> I've done this plenty of times in Xlib and GTK, but I would prefer to use
> Tk for this.  Could someone post some example code?

the canvas is always double-buffered.  just create or modify
canvas items as usual.  when Tk returns to the event loop,
the canvas is redrawn as soon as possible.

to force redraw, use w.update_idletasks()

to get an unbuffered canvas, use the wck (tkinter.effbot.org)

Cheers /F

<!-- pythonworks 1.2 coming soon:
http://www.pythonware.com/products/works/index.htm
-->




From sholden at holdenweb.com  Tue Mar 13 02:31:01 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Tue, 13 Mar 2001 02:31:01 -0500
Subject: gah! I hate the new string syntax
References: <mailman.983485923.1136.python-list@python.org> <97sudi01i26@news1.newsguy.com> <98elrk$li8$1@nntp.Stanford.EDU> <98fc070thh@news1.newsguy.com> <98iqo2$56o$1@nntp.Stanford.EDU>
Message-ID: <j4kr6.68739$ML1.3282814@e420r-atl2.usenetserver.com>

"Amit Patel" <amitp at Xenon.Stanford.EDU> wrote in message
news:98iqo2$56o$1 at nntp.Stanford.EDU...
> Alex Martelli <aleaxit at yahoo.com> wrote:
> | "Amit Patel" <amitp at Xenon.Stanford.EDU> wrote in message
> | news:98elrk$li8$1 at nntp.Stanford.EDU...
> |     [snip]
> | > What I mean to say is that I cannot write new methods on strings.  So
> |
> | Perfectly true, just as you cannot on dictionaries, lists, tuples,
> | file objects, and so on.  Only classes and extension-classes
> | are extensible (by inheritance and overriding), not types.
>
> Even if strings were a class, and I could extend it, it is NOT the
> thing to do to add new functionality like capitalize().  Suppose
> strings do not have a capitalize method, and I want to add one.  And
> suppose strings are classes and can be extended.  If I wrote:
>
>   class MyString(string):
>      def capitalize(self):
>         ...
>
>
> then I can only capitalize strings I created!  I can't take
> ' '.join(['lister', 'is', 'a', 'smeghead'] and run .capitalize() on
> it -- because it's an ordinary string and not a MyString.
>
> The advantage of method functions like string.capitalize is that they
> do not require me to change something I can't change (the string
> type).  I can write capitalize(x) when x is a string.  I can't write
> x.capitalize() when x is a string.
>
Well, being pedantic, of course, the capitalize() method is already
provided:

>>> x = "that seems wrong"
>>> x.capitalize()
'That seems wrong'

However, your point is valid for a general transmogrify() method.

> Defining methods is like running things as root.  Yes, you sometimes
> need to do it, but it's best to limit yourself to only those things
> that need superuser access.  Run as much as you can as non-root.  And
> IMO that goes for methods and functions too -- string type methods are
> "root" (where "root" = "Guido & pals") and have access to the
> internals of a string, whereas functions on strings are user-level and
> go through the abstractions.
>
But surely the idea of inheritance is precisely that if you want
self-capitalizing strings you should be prepared to subclass the string
class. Or, in Python (since strings are a type) to wrap them then subclass
the wrapper.

The type/class dichotomy can lead to awkwardness, but I've seen no proposal
to get rid of it before Python 3, probably way in the future.

regards
 Steve





From mwh21 at cam.ac.uk  Mon Mar  5 15:21:45 2001
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 05 Mar 2001 20:21:45 +0000
Subject: Coroutines, and memory leaks
References: <8oh089.7it.ln@127.0.0.1>
Message-ID: <m3r90carvq.fsf@atrus.jesus.cam.ac.uk>

dg at pearl.tao.co.uk (David Given) writes:

> 2. I have this massive memory leak. How can I find it?
> 
> The more I use the program, the more memory it uses. I've seen it use
> >100MB before I killed it. I guess that I've got cyclic memory structures
> in the message cache, so it's never freeing any of the mail I look at. Are
> there any suggestions for how I can find these and kill them?

(a) switch to Python 2.
(b) try using Tim Peters' cyclops.py:
    http://www.python.org/ftp/python/contrib-09-Dec-1999/System/Cyclops.py

Cheers,
M.

-- 
  This song is for anyone ... fuck it.  Shut up and listen.
                         -- Eminem, "The Way I Am"


From aleaxit at yahoo.com  Fri Mar 23 18:42:14 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sat, 24 Mar 2001 00:42:14 +0100
Subject: PEP 245: Python interfaces
References: <mailman.985380515.28565.python-list@python.org>
Message-ID: <99gn1f060d@news2.newsguy.com>

"Clark C. Evans" <cce at clarkevans.com> wrote in message
news:mailman.985380515.28565.python-list at python.org...
    [snip]
> you may have an interface with four functions.  The implementation
> of three functions may be completely optional or trivial.  With
> the class approach, you could label the one critical method as
> "pure" and then provide a reasonable, default implementation
> for the other methods.
>
> > Classes could in some way, be extened to have an inline "interface"
> > but that is mixing your interface with your implementation.
>
> I'm not convinced that they are entirely seperate ideas.

I used to suspect this was just a C++ distortion, but after
seeing the way Haskell does typeclasses I'm now convinced
otherwise.  A Haskell typeclass need not have any 'pure'
method -- they may ALL be implemented in terms of each
other (unbounded recursion!), a type that instances that
typeclass just needs to override a minimal set of methods
to ensure no unbounded recursion occurs.  And meanwhile,
as well as providing convenience, the mutually recursive
definitions in the typeclass specify the semantics!

Example: consider a typeclass modeling a subset of what
fileobject does today, specifically write and writelines.
Each may be defined in terms of the other (pythonic-ish
syntax for the Haskell typeclass idea!-):

typeclass writeablefile:
    def write(self, astring):
        self.writelines([astring])
    def writelines(self, stringslist):
        for astring in stringslist:
            self.write(astring)

See?  Neither needs to be defined as 'primary' -- they
are two methods of equal dignity, EITHER may be overridden
by a type instancing writeablefile and the other will be
picked up through its semantic definition in the
typeclass.  Now is this cool, or what?!-)

(Of course, BOTH may be overridden, e.g. for efficiency,
but at least the definitions in the typeclass stand as
good documentation of the intended semantics!-).


> > How can you reuse that?
>
> Just like you use classes today... look at the PyXML stuff,
> specifically the Sax interfaces.  Providing a default, or
> trivial implemenation for optional methods is a common
> technique in C++ and very valueable.
>
> In Java you are forced to have two constructs, an interface,
> and then a "base handler".  Having the interfae and the
> base implementation in two places just makes more places
> where errors can pop-up.

I disagree on this -- I like the interface/basehandler
separation so much that I tend to write C++ that way,
too (via mixins, generally, and templates too -- Java
doesn't have those, so more manual boilerplate is
needed -- but that's a SMOP).


Alex





From SBrunning at trisystems.co.uk  Wed Mar 28 09:20:03 2001
From: SBrunning at trisystems.co.uk (Simon Brunning)
Date: Wed, 28 Mar 2001 15:20:03 +0100
Subject: list vs tuple
Message-ID: <31575A892FF6D1118F5800600846864D78BA44@intrepid>

> From:	deadmeat [SMTP:root@[127.0.0.1]]
> Whats the practical difference between a list[] and a tuple() ?
> 
> A tuple seems to be a very basic list, so unless it's faster, what's the
> point of it?
 
Off the top of my head - A tuple is immutable, and therefore hashable, and
therefore can be used as a key to a dictionary. I'm sure that there are
other things...

Cheers,
Simon Brunning
TriSystems Ltd.
sbrunning at trisystems.co.uk





-----------------------------------------------------------------------
The information in this email is confidential and may be legally privileged.
It is intended solely for the addressee. Access to this email by anyone else
is unauthorised. If you are not the intended recipient, any disclosure,
copying, distribution, or any action taken or omitted to be taken in
reliance on it, is prohibited and may be unlawful. TriSystems Ltd. cannot
accept liability for statements made which are clearly the senders own.



From Simon.Budig at unix-ag.org  Fri Mar 16 20:56:58 2001
From: Simon.Budig at unix-ag.org (Simon Budig)
Date: 17 Mar 2001 02:56:58 +0100
Subject: list.sort(cmpfunc) question
References: <mailman.984788006.18156.python-list@python.org>
Message-ID: <3ab2c46a@si-nic.hrz.uni-siegen.de>

Tim Peters <tim.one at home.com> wrote:
> [Simon Budig]
>> ...
>> Is the sort-function guaranteed to be stable? Does it keep the order,
>> when cmpfunc returns 0 for a pair of items?
> 
> No, Python's list.sort() is not stable.  Stability can be "faked", though, by
> sorting a derived list y, where each element y[i] == (x[i], i).  Then no two
> list elements are equal.  Because tuples are compared lexicographically,
> cmp((x, i), (x, j)) has the same sense as cmp(i, j), thus preserving the
> original order of equal elements in x.

Thanks, that was helpful.

Bye,
        Simon
-- 
      Simon.Budig at unix-ag.org       http://www.home.unix-ag.org/simon/


From aahz at panix.com  Tue Mar 20 00:49:50 2001
From: aahz at panix.com (Aahz Maruch)
Date: 19 Mar 2001 21:49:50 -0800
Subject: relational/boolean operator overloading
References: <98tjq6$qre$1@news.flashnet.it> <98tk4b$sbv$1@panix3.panix.com> <994h71$ohq$1@news.flashnet.it>
Message-ID: <996r1u$67f$1@panix6.panix.com>

In article <994h71$ohq$1 at news.flashnet.it>,
Renzo Tomaselli <renzo.tomaselli at tecnotp.it> wrote:
>Aahz Maruch wrote in message <98tk4b$sbv$1 at panix3.panix.com>...
>>
>>Not sure what advice to give you with current Python, but Python 2.1
>>introduces rich comparisons, which may give you what you want.  See
>>PEP 207, http://python.sourceforge.net/peps/pep-0207.html
>
>Pep 207 looks great for my posted topic. However its solves half of the
>problem. The missing part is about overloading 'and'/'or' operators, also
>because this PEP allows for comparisons returning any object. How to chain
>such results with booleans, beside a trivial truth test ?
>Say we have: "(sa > 10) and (sb < 100)". I would expect PEP overloading to
>carry out relational filtering for the two sets and to return two result
>sets, while some "and" overloading would connect them to return a final
>result set.

There's some stuff being done on the numeric side of Python that may be
relevant (in terms of set operators and such).  I don't know enough
about it to give you a good search term, though.
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

There's a difference between a person who gets shit zie doesn't deserve
and a person who gets more shit than zie deserves.  --Aahz


From jafo at tummy.com  Sat Mar 17 04:39:07 2001
From: jafo at tummy.com (Sean Reifschneider)
Date: Sat, 17 Mar 2001 02:39:07 -0700
Subject: Pep 238 Why new // operator (Non-integer Division)
In-Reply-To: <98tc2s$qir$1@news2.ottawa.cyberus.ca>; from g_will@cyberus.ca on Fri, Mar 16, 2001 at 10:39:10AM -0500
References: <98tc2s$qir$1@news2.ottawa.cyberus.ca>
Message-ID: <20010317023907.I29139@tummy.com>

On Fri, Mar 16, 2001 at 10:39:10AM -0500, Gordon Williams wrote:
>All these changes make my head spin.  There is more to life than learning
>all the fine details in a computer language.  The fewer number of ways to do
>something the better.

Just because the changes are being made doesn't mean you have to USE them.
You wouldn't be the only one sticking to 1.5.2.  I like the more rapid
pace of development, but I'm mostly running 1.5.2 right now because I
don't want to start writing code that's going to break on a few dozen
machines.  ;-)

Sean
-- 
 I think you are blind to the fact that the hand you hold
 is the hand that holds you down.  -- Everclear
Sean Reifschneider, Inimitably Superfluous <jafo at tummy.com>
tummy.com - Linux Consulting since 1995. Qmail, KRUD, Firewalls, Python



From tjg at exceptionalminds.com  Tue Mar  6 18:16:31 2001
From: tjg at exceptionalminds.com (Timothy Grant)
Date: Tue, 6 Mar 2001 15:16:31 -0800
Subject: from Tkinter import *
Message-ID: <20010306151631.M3579@trufflehunter.avalongroup.net>

Hi all,

With all the discussion around here recently about the

from foo import *

idiom, I thought I'd ask about Tkinter. All of Fredrik's
examples in his guide on Tkinter use 

from Tkinter import *

Now, I'm aware that the effbot is never wrong about
anything<wink>, but is it safe to assume that I should not be
using the above idiom, but the import Tkinter idiom instead?


-- 
Stand Fast,
    tjg.

Timothy Grant                         tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.         <><       (503) 246-3630
>>>>>>>>>>>>>Linux, because rebooting is *NOT* normal<<<<<<<<<
>>>>This machine was last rebooted:  49 days  3:27 hours ago<<



From tim.one at home.com  Thu Mar  1 15:48:21 2001
From: tim.one at home.com (Tim Peters)
Date: Thu, 1 Mar 2001 15:48:21 -0500
Subject: A counter-proposal to __future__ in PEP 236
In-Reply-To: <97m2p5$3au$2@news.mathworks.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCIECCJDAA.tim.one@home.com>

[Joshua Marshall]
> ...
> Note that it doesn't need to be implemented this way.  Currently,
>
>   directive foo
>
> is a syntax error.  Meaning can be assigned to this syntax without
> breaking any existing code--it is unambiguous.

You should study Python's parser before getting too optimistic about that
<wink>.




From asle at spam.com  Thu Mar 15 03:50:41 2001
From: asle at spam.com (asle at spam.com)
Date: 15 Mar 2001 08:50:41 GMT
Subject: HTMLparsing abnormal html pages
Message-ID: <98pvp1$15t$1@news.netmar.com>

Considering the small program below. Running it will show that the
HTMLparser
is truncating urls in the HTML page. Now, most of you will probably say that
the page and in particular the URL's of this page are not valid according to
the RFC1738 protocol --bad luck. But there must be a work-around for this?

import htmllib
import urllib
import formatter

url='http://di.se/Scripts/Sections/allarticles.asp'
parser=htmllib.HTMLParser(formatter.NullFormatter())
parser.feed(urllib.urlopen(url).read())
parser.close()
urls=parser.anchorlist
print urls

One solution is of course to preprosess the whole HTML page and
replacing invalid URL's whith valid URL's (using regex??), however I have
also tried to look into HTMLparser and the formatter to see what can be done
there to correct the problem, but with no sucess on the latter.

Any comments on what to do?

/Asle



 -----  Posted via NewsOne.Net: Free (anonymous) Usenet News via the Web  -----
  http://newsone.net/ -- Free reading and anonymous posting to 60,000+ groups
   NewsOne.Net prohibits users from posting spam.  If this or other posts
made through NewsOne.Net violate posting guidelines, email abuse at newsone.net


From fig at monitor.net  Thu Mar 22 15:49:09 2001
From: fig at monitor.net (Stephen R. Figgins)
Date: Thu, 22 Mar 2001 12:49:09 -0800
Subject: Puzzling input problem
References: <3AB9449C.A1D14738@monitor.net> <mailman.985233449.28064.python-list@python.org> <3AB97E58.7B01D485@monitor.net> <99cbni0149a@news2.newsguy.com>
Message-ID: <3ABA6545.F2959F81@monitor.net>

Thanks, Alex.  That is crystal clear now.  

Stephen


From spahievi at vega.bg  Wed Mar 28 15:05:54 2001
From: spahievi at vega.bg (Niki Spahiev)
Date: Wed, 28 Mar 2001 23:05:54 +0300
Subject: daemon app in win32
In-Reply-To: <99r4hq$sq3$1@troll.powertech.no>
References: <99nk49$a2q$02$1@news.t-online.com>
 <mailman.985619314.10299.python-list@python.org>
 <99nsn3$q39$06$1@news.t-online.com> <99r4hq$sq3$1@troll.powertech.no>
Message-ID: <21795650.20010328230554@vega.bg>

28.3.2001, 01:32:18, Syver Enstad wrote:

>> Hmm, seems like this is only available under NT. Is there a more general
>> way to accomplish this task independently of the underlying
SE> Windows-version?

SE> I remember reading some time about rudimentary supprort for services under
SE> Win9x so maybe you can use one version after all. Check out the platform
SE> sdk, or msdn for details.

WinGate is HTTP proxy that runs as service under win9x.

-- 
Best regards,
 Niki Spahiev





From chrislamb at btinternet.com  Sat Mar  3 07:52:35 2001
From: chrislamb at btinternet.com (chris)
Date: Sat, 3 Mar 2001 12:52:35 +0000
Subject: gah! I hate the new string syntax
References: <mailman.983615170.23313.python-list@python.org>
Message-ID: <tN5o6.2746$eH3.14913@NewsReader>

> at latest since the 19th century; in America, it appears this is normally
> introduced in the 6th or 7th grades (ages 11-12, for those of you in
> inferior countries <wink>

In Scotland that kind of stuff is pre-School! <wink-back>

Chris


From skip at pobox.com  Wed Mar 28 16:22:21 2001
From: skip at pobox.com (Skip Montanaro)
Date: Wed, 28 Mar 2001 15:22:21 -0600 (CST)
Subject: Is inheritance broken?
In-Reply-To: <NDBBLAKACLGJJBDNGGCMIEJDCCAA.jcc.ugm@ix.netcom.com>
References: <15042.17325.125458.471763@beluga.mojam.com>
	<NDBBLAKACLGJJBDNGGCMIEJDCCAA.jcc.ugm@ix.netcom.com>
Message-ID: <15042.22029.707094.352865@beluga.mojam.com>

    >> How about creating mixins, then composing a set of classes from them:

    Jonathan> Interesting. Are there conventions for mixin classes which
    Jonathan> distinguish them from normal classes? For example, are mixins
    Jonathan> not suppposed to use inheritance (or at least not use mulitple
    Jonathan> inheritance)? Thanks for the help,

I'm not sure what all conventions are typical that distinguish mixins from
normal classes.  I'm sure other people can answer from a lot more
experience.  Typically, mixins can't stand alone, however.  Also, there are
often multiple variants of a class of mixins.  For example, if you
implemented a networking class you might compose your concrete classes from
an abstract base class that implements the desired semantics and appropriate
mixins that know how to setup and shutdown various types of connections
(TCP, UDP, Unix sockets, etc).  Also, in some environment mixins seem to
often contain "Mixin" as a suffix of the class name so that it's clear what
their purpose is.

You might want to check recent c.l.py archives (within the last week or
three).  I seem to recall someone posting a URL about mixins.

-- 
Skip Montanaro (skip at pobox.com)
(847)971-7098



From slhath at home.com  Fri Mar  9 22:06:15 2001
From: slhath at home.com (Scott Hathaway)
Date: Sat, 10 Mar 2001 03:06:15 GMT
Subject: getting the field names from a resultset on Windows with mxODBC
References: <vf6q6.709$eH1.219552882@newssvr16.news.prodigy.com> <u1ys68k0n.fsf@ctwd0143.fitlinxx.com>
Message-ID: <HSgq6.152230$bb.1522991@news1.rdc1.tx.home.com>

Thanks!

Scott

"David Bolen" <db3l at fitlinxx.com> wrote in message
news:u1ys68k0n.fsf at ctwd0143.fitlinxx.com...
> "Scott Hathaway" <slhath at home.com.nospam> writes:
>
> > I am using mxODBC for Windows.  When I do a query, how can I get the
names
> > of the fields that the resultset returns (without parsing the SQL myself
or
> > having the user supply them)?
>
> Use <cursor>.description - it's a list of tuples corresponding to your
> fields that map to each element of the result set you are getting
> back.  It includes name and type information and some other stuff.
> This is part of the standard Python DB-API, and is covered in both the
> DB-API and mxODBC-specific documentation included with mxODBC.
>
> --
> -- David
> --
> /-----------------------------------------------------------------------\
>  \               David Bolen            \   E-mail: db3l at fitlinxx.com  /
>   |             FitLinxx, Inc.            \  Phone: (203) 708-5192    |
>  /  860 Canal Street, Stamford, CT  06902   \  Fax: (203) 316-5150     \
> \-----------------------------------------------------------------------/




From dav-nospam-id at westco-nospam-ntrol.com  Thu Mar 22 09:28:49 2001
From: dav-nospam-id at westco-nospam-ntrol.com (David Brown)
Date: Thu, 22 Mar 2001 15:28:49 +0100
Subject: Cross-platform serial port access
Message-ID: <99d274$ogv$1@news.netpower.no>

I am relatively new to Python, but am planning to use it for a number of
small programs that should be able to run both on Windows (for initial
development and testing) and Linux (once we get a server running properly).
Some of these programs will need to communicate with external devices using
a serial port - is this going to work in a portable fashion using Python?
Does anyone know if Linux supports RTS toggle handshaking (i.e., the RTS
line is high while transmitting, for convenient connection to RS-485
convertors), and is it easy to control from Python?  Are there any Python
modules designed to work with modems?

Thanks for any help in getting me started.

David Brown
Norway





From rwklee at home.com  Mon Mar 12 22:48:11 2001
From: rwklee at home.com (Rick Lee)
Date: Tue, 13 Mar 2001 03:48:11 GMT
Subject: Python, threads, and signals (oh my!)
References: <98jqeh$6ut$1@newshost.mot.com>
Message-ID: <3AAD987A.B4A026C2@home.com>

On a related note about threads.  Does anyone have any experience on which
unix system is the fastest with threads, by supporting native thread
switching?  It seems that Linux does not have native thread switching, and
in an environment where the processor is fairly heavily loaded running
several threads, it is much slower compared to WinNT on the same hardware.

- Rick Lee

Jason Lowe wrote:

> I've been playing around with Python and threads, and I've noticed some
> odd and often unstable behavior.  In particular, on my Solaris 8 box I
> can get Python 1.5.2, 1.6, or 2.0 to core dump every time with the
> following sequence.  I've also seen this happen on Solaris 6.
>
> 1. Enter the following code into the interactive interpreter:
> --
> import threading
>
> def loopingfunc():
>   while 1: pass
>
> threading.Thread(target=loopingfunc).start()
> --
>
> 2. Send a SIGINT signal (usually Ctrl-C, your terminal settings may
>    vary).  "Keyboard Interrupt" is displayed and so far everything looks
>    fine.
>
> 3. Now simply press the <Enter> key to enter a blank line in the
>    interpreter.  For my Solaris 8 box with the GNU readline 2.2 module
>    present, this always ends up in a core dump.  It may take a while,
>    since at this point the readline signal handler is being re-entered
>    recursively until the stack overflows.
>
> Looking more into this, it appears that on Solaris, more than one
> thread is processing the signal handler installed by the readline
> module (according to truss output of the process).  Unfortunately, the
> readline signal handlers don't support being re-entered, as they have
> global data that's not protected.
>
> Now granted, signals and threads are a dangerous business.  However, it
> would be nice if the user sending SIGINT to a script wouldn't cause
> instability.  Looking at the Python module sources, I noticed that the
> signal module is somewhat "thread aware" -- it allows signal
> handlers to be installed only by the main thread.  However, I found it a
> little odd that the thread support, both in the core interpreter and in
> the thread module, had no support code for signals -- even in the
> specific thread cases like POSIX (pthreads).
>
> According to various pthread documentation, it appears the "right" way
> to handle the often volatile signal/thread mix is to mask all signals
> except in one thread (usually the main thread) so that only one thread
> will receive the signal.  Unfortunately, I couldn't find any Python
> module that would allow one to change the signal mask of a thread
> (pthread_sigmask) or even a process (sigmask, sigprocmask, etc.).  I'm
> assuming the lack of this interface has to do with portability across
> platforms.
>
> I was able to solve the problem by modifying Python/thread_pthread.h's
> PyThread_start_new_thread() to block all signals before creating the new
> thread and then restoring the signal mask after the new thread was off
> and running.  Therefore, all threads created by Python except the
> initial thread will have all signals masked with this change.  Masking
> all signals in new Python threads makes sense to me, given that the
> signal module doesn't like other threads installing handlers anyway.
>
> As a side note, I think part of the problem with sending SIGINT at the
> interactive prompt while threading is aggravated by the Python readline
> module's signal handler.  It setjmp()'s and longjmp()'s to do its dirty
> work, and many thread implementations require the longjmp() be performed
> by the thread that did the setjmp().  If signals aren't masked in all
> threads except the one doing the readline() call, then this isn't
> guaranteed.
>
> So what's everyone else's take on this?  Has anyone else experienced
> problems in Python when using threads and receiving signals (like
> SIGINT)?  Should the thread code be at least a little more "signal
> aware", so that new Python threads have all signals masked on platforms
> that support this?
>
> Thanks in advance.
>
> Jason Lowe
> --
> Jason Lowe                                    Urbana Design Center
> Motorola Personal Communications Sector       1800 South Oak Street
> jlowe at urbana.css.mot.com                      Champaign, IL  61820-6947
>                                               (217) 384-8513



From zeitlin at seth.lpthe.jussieu.fr  Sat Mar 10 21:32:45 2001
From: zeitlin at seth.lpthe.jussieu.fr (Vadim Zeitlin)
Date: 11 Mar 2001 02:32:45 GMT
Subject: loading Python20.dll during run-time under Windows
Message-ID: <slrn9alouc.sor.zeitlin@seth.lpthe.jussieu.fr>

 Hello,

 I have an application which can use Python for scripting but it is not
essential to its work, i.e. it works mostly fine without Python. So I'd
like to be able to install it even on systems lacking Python installation
but for this I need to [try to] load Python.dll during run-time instead
of linking to it directly.

 Has anyone done this before and is there any (C) code I could reuse?
Normally, there is no problem with doing what I want (especially
after realizing that I need to have USE_DL_EXPORT defined...) but
the perspective of having to manually write all the code needed to load
the functions from the DLL is not very, hmm, encouraging just because there
are quite a few of functions and so I'm looking for a slightly less painful
way to do it.

 Thanks in advance for any help!
VZ

-- 
GCS/GM d? H+ s++:-- p2 au--- a- w+ v C+++ UBLS+++ P- L++ N++ E--- W++++ M? V--
-po+ R++ G`` !tv b+++ D--- e++++ u++ h--- f+ r++ n- y?


From fuess at att.net  Sat Mar 10 10:06:38 2001
From: fuess at att.net (David Fuess)
Date: Sat, 10 Mar 2001 15:06:38 GMT
Subject: import headaches/gripes
References: <3aaa3f94.15940553@News.CIS.DFN.DE>
Message-ID: <tegkatsj8gdmmoctjqivgfdjq7gckfg41h@4ax.com>

I think the problem is that PythonWin sets up the Python environment
in the beginning and everything runs under that environment. It gets
dirty as you develop so you eventually have to clean house. Your
import statements are ignored because the module already exists in the
environment. Changing that would affect multiple instances of import
on a module in an application. Kind of a Catch-22, eh?

You should try programming Tkinter apps under PythonWin. The
environment gets ditry after one or two iterations and sometimes has
to be killed from outside!!

Dave

On Sat, 10 Mar 2001 14:52:08 GMT, costas at meezon.com (Costas Menico)
wrote:

>I am using Pythonwin to develop apps.  I usually have a few modules
>open. My main module imports all the necessary files at the start.
>
>However when I change a module, I have to remember to reload() it
>manually or quit Pythonwin. 
>
>Why can't the import statement be smart enough to check if a file was
>changed to import it anyway? All it has to do is compare .py to the
>last import date/time.
>
>Given my gripe is there some way I can force imports? It seems the
>only way to do this is somehow to use reload(module name). Maybe some
>logic like this would work:
>
>if module_exists(utils):
>	reload(utils)
>else:
>	import utils
>
>
>where module_exists() is some imaginary function that can check if the
>module is loaded.
>
>Any ideas would be appreciated.
>
>Costas



From carmstro at dynup.net  Sun Mar  4 08:35:30 2001
From: carmstro at dynup.net (Chris Armstrong)
Date: Sun, 04 Mar 2001 18:35:30 +0500
Subject: Static class methods
References: <rdzo6.154$gc4.80445@news.pacbell.net>
Message-ID: <ta5ka1ks0buc3b@corp.supernews.com>

I have trouble believing that something like this is really needed.
Could you give a real-world example?

In article <rdzo6.154$gc4.80445 at news.pacbell.net>, "Alex Shindich"
<alex at shindich.com> wrote:

> I was
> thinking that a new keyword "unbound" could be used to identify
> functions that belong to a class but do not expect "self" as their first
> attribute. From outside the class "unbound" methods would be invoked by
> specifying the class name -- Foo.my_static_method (). From within the
> class they would work similar to local functions that are defined at the
> class level. The __ trick should work for the "unbound" methods as well
> to allow for creation of private static methods.

-- 
Chris Armstrong   http://twistedmatrix.com/~carmstro     carmstro at dynup.net
There is a 90% chance that this message was written when the author's been
awake longer than he should have. Please disregard any senseless drivel.


From steven at haryan.to  Wed Mar 21 02:50:13 2001
From: steven at haryan.to (Steven Haryanto)
Date: Wed, 21 Mar 2001 14:50:13 +0700
Subject: slightly OOT: packaging Python 2.1b1 in RPM
In-Reply-To: <4.3.2.7.2.20010320130831.00bb0dc0@202.53.255.51>
Message-ID: <4.3.2.7.2.20010321144928.067f5280@bdg.centrin.net.id>

Sorry to bother everyone. This has been solved. rpm did
implicitly compress manpages after %install
(using /usr/lib/rpm/brp-compress script), so I compress
them manually at %install.

Steve

At 3/20/2001 01:15 PM, Steven Haryanto wrote:
>I am packaging Python 2.1b1 on RedHat 6.2, rpm 3.0.5, using
>spec file modified from Jeremy's beopen-python.spec for
>BeOpen-Python 2.0.
>
>In the very latest step of %install, I generate an automatic
>file list using find:
>
>find $RPM_BUILD_ROOT/ -type f | grep -v 'file-list\.txt' | \
>     sed -e "s|^$RPM_BUILD_ROOT||g" > file-list.txt
>find $RPM_BUILD_ROOT/usr/lib/python2.1/ \
>     $RPM_BUILD_ROOT/usr/include/python2.1/ -type d | \
>     sed -e "s|^$RPM_BUILD_ROOT|%dir |g" >> file-list.txt
>
>and feed it to %files:
>
>%files -f file-list.txt
>
>so I can be pretty sure that the package will not be missing
>any file.
>
>However, when rpm is scanning for files, /usr/man/man1/python.1
>is not found because it has been compressed and becomes
>/usr/man/man1/python.1.gz. This must happen after the %install,
>since 'find' still lists the uncompressed one.
>
>Any hint why/what program did this?
>
>Thanks,
>Steve




From duncan at rcp.co.uk  Wed Mar 14 05:42:09 2001
From: duncan at rcp.co.uk (Duncan Booth)
Date: Wed, 14 Mar 2001 10:42:09 +0000 (UTC)
Subject: Extending Python: rewriting a single method in C
References: <g0wv9vxms0.fsf@scumbag.ecs.soton.ac.uk> <slrn9apn8f.382.jepler@potty.housenet> <g0ofv6xbj4.fsf@scumbag.ecs.soton.ac.uk> <98l1i901k4u@news1.newsguy.com> <g0lmq9xxz0.fsf@scumbag.ecs.soton.ac.uk> <98m6g20fgv@news1.newsguy.com> <g0g0ggy7ct.fsf@scumbag.ecs.soton.ac.uk>
Message-ID: <Xns90646C7A9204Dduncanrcpcouk@127.0.0.1>

Jacek Generowicz <jmg at ecs.soton.ac.uk> wrote in 
<g0g0ggy7ct.fsf at scumbag.ecs.soton.ac.uk>:

>>>> print a.__getattr__("bar")
>Traceback (most recent call last):
>  File "<stdin>", line 1, in ?
>AttributeError: 'foo' instance has no attribute '__getattr__'
>
>
What you want is:
>>> print getattr(a, "bar")
4



From micampe at f2s.com  Tue Mar 27 04:50:50 2001
From: micampe at f2s.com (comp.lang.python)
Date: Tue, 27 Mar 2001 09:50:50 GMT
Subject: unique items in lists
Message-ID: <_nZv6.2878$yU4.468236@news.infostrada.it>

Which is the best (read: fastest) way to remove duplicated items from a list
of strings?

10x, Michele




From ws-news at gmx.at  Fri Mar  9 05:55:27 2001
From: ws-news at gmx.at (Werner Schiendl)
Date: Fri, 9 Mar 2001 11:55:27 +0100
Subject: pipe problems
References: <3AA6D886.4DD8D3A4@iems.nwu.edu> <984045555.82149@newsmaster-04.atnet.at> <3AA7DD16.93C0730D@iems.nwu.edu>
Message-ID: <984135609.599401@newsmaster-04.atnet.at>

Maybe to much decoration in the last post...

readline() is a blocking function for pipes, ttys and so on.
If you readline() at the end of a file, it immediately returns, because the
file will (normally) not grow later.

But a pipe (like a tty) is filled from one side (a tty is 'filled' by a user
typing on the keyboard) and emptied from your application.
If there is currently nothing that can be read in the buffer, readline waits
for such data to become available

-> it blocks

This behaviour is to ease the use on one hand (you do not need to poll in a
loop, but just do one call and have the result - or know that there will not
be one) and to make multitasking more efficient on the other hand (it is of
no use to have your programm polling for data that the IO system knows much
better when it arrives).

I guess the programm you call does not print (a whole) line of text after
you output your command to it and the whole thing stops.

Does this clearify the situation?

regards
werner
Leonardo B Lopes <leo at iems.nwu.edu> wrote in message
news:3AA7DD16.93C0730D at iems.nwu.edu...
> Werner Schiendl wrote:
> >
> > If there is no data to read, readline it will block for such to become
> > available.
> > I don't think that cout.flush() is of any use, btw.
>
> When I tried the program with 'cat', it only worked if I used
> cout.flush(). But maybe it doesn't work in general. My guess is that
> this call is the crux of the problem.
>
> > Maybe I'm wrong, but I think data can only be flushed out by the sender
of a
> > pipe...
> > Further, what's if the program prints out something in between, or more
than
> > 1 line?
>
> That is something I'll deal with at a later point, when I get the basic
> stuff working. I have the impression there is something about pipes that
> I don't understand correctly.
>
> Thanks,
> Leo.
>
> >
> > Maybe you can do the reading in a seperate thread.
> > Alternatively you can try to find out if there _is_ sth you can read.
> >
> > hth
> > werner
> >
> > Leonardo B Lopes <leo at iems.nwu.edu> wrote in message
> > news:3AA6D886.4DD8D3A4 at iems.nwu.edu...
> > > I use a program that doesn't have readline compiled in. So I figured,
it
> > > would be easy to add a python wrapper around it. So I wrote the
wrapper,
> > > using 'cat' as a test command. With 'cat' it works fine. But when I
try
> > > using another program, such as 'python', the program just hangs at
> > > readline(). Anybody know how to accomplish this? Here is the code:
> > >
> > > #!/usr/local/bin/python
> > >
> > > import sys,os
> > > import cmd
> > >
> > > class Amplcmd(cmd.Cmd):
> > >
> > >         def __init__(self):
> > >                 cmd.Cmd.__init__(self)
> > >                 self.cin,self.cout =
os.popen2('/usr/local/bin/python')
> > >
> > >         def default(self,line):
> > >                 self.cin.write(line+'\n')
> > >                 self.cin.flush()
> > >                 self.cout.flush()
> > >                 print self.cout.readline(),
> > >
> > >         def do_EOF(self,line):
> > >                 sys.exit()
> > >
> > > mycmd = Amplcmd()
> > > mycmd.cmdloop()
> > >
> > >
> > >
> > > --
> > >
=======================================================================
> > > Leonardo B. Lopes
leo at iems.nwu.edu
> > > Ph.D. Student
(847)491-8470
> > > IEMS - Northwestern University
http://www.iems.nwu.edu/~leo
>
> --
> =======================================================================
> Leonardo B. Lopes                                      leo at iems.nwu.edu
> Ph.D. Student                                             (847)491-8470
> IEMS - Northwestern University             http://www.iems.nwu.edu/~leo




From jack at oratrix.nl  Fri Mar  2 04:15:45 2001
From: jack at oratrix.nl (Jack Jansen)
Date: Fri, 02 Mar 2001 10:15:45 +0100
Subject: [Distutils] Re: CPAN functionality for python - requirements 
In-Reply-To: Message by Sean Reifschneider <jafo@tummy.com> ,
	     Wed, 28 Feb 2001 16:54:49 -0700 , <20010228165449.A1781@tummy.com> 
Message-ID: <20010302091545.DD722373C95@snelboot.oratrix.nl>

> On Wed, Feb 28, 2001 at 10:13:51AM +0100, Jack Jansen wrote:
> >Well, one of the things I'd like to do is integrate all this in MacPython IDE, 
> >and the same probably holds for PythonWin and Idle. Then the "what do you 
> >have" is pretty useful for building menus and such.
> 
> I anticipate having a web-based interface to the database as well.  I'm
> sure that we can get the CGI to let the user do that sort of search [...]

Right, but an IDE can do better. For instance, if an import fails it can offer 
to do a lookup on the packages site to see whether such a package is 
available. On the other hand: it could also do so by firing up your browser 
with a carefully constructed URL. Ah well...
--
Jack Jansen             | ++++ stop the execution of Mumia Abu-Jamal ++++
Jack.Jansen at oratrix.com | ++++ if you agree copy these lines to your sig ++++
www.oratrix.nl/~jack    | see http://www.xs4all.nl/~tank/spg-l/sigaction.htm 





From jwbnews at scandaroon.com  Fri Mar  9 21:13:20 2001
From: jwbnews at scandaroon.com (John W. Baxter)
Date: Fri, 09 Mar 2001 18:13:20 -0800
Subject: gah! I hate the new string syntax
References: <mailman.983615170.23313.python-list@python.org> <mailman.984014353.5058.python-list@python.org> <BcCp6.37817$1D5.1582482@e420r-atl1.usenetserver.com> <98amlj01sp6@news2.newsguy.com> <98asrq$645$1@panix6.panix.com> <G5eq6.47941$ML1.2142034@e420r-atl2.usenetserver.com>
Message-ID: <jwbnews-8E2915.18132009032001@corp.supernews.com>

In article <G5eq6.47941$ML1.2142034 at e420r-atl2.usenetserver.com>,
 "Steve Holden" <sholden at holdenweb.com> wrote:

> "Aahz Maruch" <aahz at panix.com> wrote in message
> news:98asrq$645$1 at panix6.panix.com...
> > In article <98amlj01sp6 at news2.newsguy.com>,
> > Alex Martelli <aleaxit at yahoo.com> wrote:
> > >"Steve Holden" <sholden at holdenweb.com> wrote in message
> > >news:BcCp6.37817$1D5.1582482 at e420r-atl1.usenetserver.com...
> > >    [snip]
> > >> > | > What about one who is both English and Scots, yet never set foot
> on
> > >> > | > that island ;-)?
> > >    [snip]
> > >> Either you were born in England, or you were born in Scotland, or you
> are
> > >> neither English nor a Scot. Part-either-or-both, maybe, but not fully
> > >
> > >Wrong!  You're talking about people whose seafaring skills and
> > >traditions are (deservedly) legendary.  It's perfectly acceptable
> > >for either an English or a Scot to be born aboard a ship...!!!
> >
> > Since when are *women* allowed aboard ships?
> 
> Women were allowed aboard British ships in the eighteenth and nineteenth
> centuries, I believe. The phrase "shake a leg" (meaning to stop lazing about
> and start work) is held to have been used by ships' mates to get the seamen
> out of their bunks. A woman would show her leg to let the mate know she
> should not be rousted out to work.
> 
And at least one of the US Clipper ships finished a New York to San 
Fransicso voyage under the command of the Captain's widow.  The captain 
died en route.  The wife was aboard as navigator (being better at the 
task than the mate), and the crew made it clear that things would go 
much better with her in command than with the mate in command.  Sorry, I 
don't have a reference handy (and the right Google search for this one 
could be a bit tricky to create).

   --John

-- 
John W. Baxter   Port Ludlow, WA USA  jwbnews at scandaroon.com


From chris.gonnerman at usa.net  Thu Mar  1 00:22:31 2001
From: chris.gonnerman at usa.net (Chris Gonnerman)
Date: Wed, 28 Feb 2001 23:22:31 -0600
Subject: shadow.h
References: <97ggvg$mqf$1@solaria.cc.gatech.edu> <97jlpi$6os$1@solaria.cc.gatech.edu>
Message-ID: <00de01c0a20f$a735bd60$a100000a@local>

Attached is C source to a shadow module which contains the functions
lckpwdf() and ulckpwdf().  Semantics are exactly as given in your man 
page for shadow.h.

Find your Makefile.pre.in and copy it into a directory with the supplied
files and run

    make -f Makefile.pre.in boot
    make

Copy shadowmodule.so to your site-packages directory (or wherever) and 
there you go.

Hope this helps...

----- Original Message ----- 
From: "Joseph Holland King" <insanc at cc.gatech.edu>
Newsgroups: comp.lang.python
To: <python-list at python.org>
Sent: Wednesday, February 28, 2001 2:07 PM
Subject: Re: shadow.h


> Joseph Holland King <insanc at cc.gatech.edu> wrote:
> 
> i have searched the python web site including the starship page and the 
> vaults of parnassus page, to no avail. i have not been able to find a 
> port or any other information that would help me to use the c function
> lckpwdf. i am really close to just writing it myself. does anyone know 
> of anywhere else to look or any way that is easier to convert the c 
> function to a python module? if not do you have any warnings or 
> comments that i should heed before attempting to write (if you can't 
> tell i am not all that enthusiastic about this idea.) thank you.

-------------- next part --------------
A non-text attachment was scrubbed...
Name: Setup
Type: application/octet-stream
Size: 31 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/20010228/29c53003/attachment.obj>
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: shadowmodule.c
URL: <http://mail.python.org/pipermail/python-list/attachments/20010228/29c53003/attachment.c>

From uche at ogbuji.net  Sun Mar 11 20:19:07 2001
From: uche at ogbuji.net (Uche Ogbuji)
Date: Mon, 12 Mar 2001 01:19:07 GMT
Subject: sql query results to xml
References: <ljuo6.134776$bb.1432827@news1.rdc1.tx.home.com>
Message-ID: <3AAC4C17.490C0936@ogbuji.net>

Scott Hathaway wrote:
> 
> I want to create a tool that will take a valid SQL query and return the
> results as valid xml.  Before I reinvent the wheel, is there a python class
> or code snippet that someone has that does this already?

Not that I know of.  Are you planning to use the syntax from Oracle's
XSQL, or DB2's XML Extender, or any of the other hundreds of formats
available (see

http://www.linuxworld.com/lw-2001-02/lw-02-xml3databases.html

)

But all that is just to point out that you probably don't have to
reinvent a resultant XML format.  A Python implementation for one of the
existing formats would be great.  Please make sure to mention it to the
Python XML SIG if you do it.


-- 
Uche Ogbuji
Personal:   uche at ogbuji.net		http://uche.ogbuji.net
Work:       uche.ogbuji at fourthought.com	http://Fourthought.com


From mda at idatar.com  Thu Mar 29 13:58:18 2001
From: mda at idatar.com (David Allen)
Date: Thu, 29 Mar 2001 18:58:18 GMT
Subject: Evaluating *EVERYTHING*
Message-ID: <eBLw6.3059$j_3.1383318@typhoon2.ba-dsg.net>

Quick question:

How do I get the interpreter to 'validate' code by
trying to evaluate *everything* so I can catch bugs?

For example, recently while hacking I discovered that
in a file where I was using FoobarException, I wrote
this line:

raise FoobarExceptoin

The typo would have caused a runtime error.  I never
found this though, because that condition never came
up in testing.  (It's pretty obscure and time related,
I don't think I could have manufactured it if I wanted 
to)

Is there a way to get python to look through a whole
file and say "Whoa there bucky - I've never heard
of FoobarExceptoin"

I know that there are other more subtle problems,
like:

typo = 5
tyop = typo * 5

and that the compiler can't probably catch those.  But
surely it should be able to catch the programmer
trying to raise an exception that doesn't exist?

-- 
David Allen
http://opop.nols.com/
----------------------------------------
When everything's coming your way, you're in the wrong lane.


From kno at jtan.com  Thu Mar 15 05:09:24 2001
From: kno at jtan.com (Matt Dunford)
Date: Thu, 15 Mar 2001 10:09:24 GMT
Subject: Regular Expressions and Python 1.5.2
References: <98q0k8$2b77$1@norfair.nerim.net>
Message-ID: <slrn9b156k.lgn.kno@io.jtan.com>

* Alexandre Fayolle <alf at leo.logilab.fr>:
> Is this a known bug in the module re of python 1.5.2, or am I missing 
> something obvious?
> 
> >>> import re
> >>> t = "m&n"
> >>> re.sub('[^/_ -]','*',t)
> '***'
> >>> re.sub('[^_ -/]','*',t)
> '*&*'

Your regex's look fine to me.  Remember that an '-' indicates a range
of characters.  '[ -/]' will match exactly that: <space> through
<forward slash>.  An '&' is between those two.  Check out this ascii
chart.

http://www.jimprice.com/ascii-0-127.gif

-- 


From phlip_cpp at my-deja.com  Tue Mar  6 13:50:21 2001
From: phlip_cpp at my-deja.com (Phlip)
Date: Tue, 6 Mar 2001 10:50:21 -0800
Subject: Long Shifty Bits
Message-ID: <kmap6.15$8Q.28605@news.pacbell.net>

Not hyp:

I want this expression to make sense:

        x = 1L << 200

If someone tells me "it will roll over at the 32nd bit", or mayby even "it 
will roll over when the CPU damn well feels like it", that's okay. I'l just 
construct a Bit Vector of some ilk and drop my bits into it. But the above 
notation is closest to our problem space, so it would be nice if it worked 
as expected.

-- 
  Phlip                          phlip_cpp at my-deja.com
============== http://phlip.webjump.com ==============
  --  "In my experience, the customer doesn't know what he wants
       until you don't give it to him." --David Brady  --


From hgg9140 at cola.ca.boeing.com  Thu Mar 22 09:33:49 2001
From: hgg9140 at cola.ca.boeing.com (Harry George)
Date: Thu, 22 Mar 2001 14:33:49 GMT
Subject: Is there python code highlighting in emacs ?
References: <mailman.985223430.22832.python-list@python.org> <3AB978A9.353A0E4B@yahoo.com> <3AB97F1F.70FF97DD@velseis.com.au> <mailman.985235612.31492.python-list@python.org>
Message-ID: <xqxelvp50vm.fsf@cola.ca.boeing.com>

You solution looks a bit complex.  I'm using:
  (add-hook 'python-mode-hook 'turn-on-font-lock)
as in:

(load "python-mode")
(setq auto-mode-alist
      (cons '("\\.py$" . python-mode) auto-mode-alist))
(setq interpreter-mode-alist
      (cons '("python" . python-mode)
            interpreter-mode-alist))
(autoload 'python-mode "python-mode" "Python editing mode." t)
(add-hook 'python-mode-hook 'turn-on-font-lock)
(setq python-mode-hook 'python-initialise)
(defun python-initialise ()
  (interactive)
  (setq default-tab-width 4)
  (setq indent-tabs-mode nil))


John McMonagle <jmcmonagle at velseis.com.au> writes:

> > 
> > > >
> > > > The hilit19 package in emacs provides keyword highlighting when editing
> > > > code.  For example, if you are writing in C, comments are italicised in
> > > > red, strings are in grey, import statements are in lavender, functions
> > > > are in bold-blue, etc.
> > > >
> > > > Does anyone know if there is python support in the hilit19 emacs lisp
> > > > package ?
> > >
> > > I don't know about hilit19, but I use python-mode which highlights
> > > using font-lock-mode. Comes with GNU Emacs 20.whatever. Try M-x
> > > python-mode and see if anything happens.
> > >
> > > python-mode does many other useful things, too. I can't imagine
> > > working without it at this point.
> > >
> > 
> > I already use python-mode which works fine for indenting, etc.  How do I
> > configure this font-lock-mode for python  ? What do I add to my .emacs
> > file ?
> > 
> 
> Worked it out.
> 
> Add the following to your .emacs file:
> 
> (add-hook 'python-mode-hook '(lambda() (font-lock-mode 1)))
> 
> 

-- 
Harry George                E-mail:  harry.g.george at boeing.com
The Boeing Company          Renton:  (425) 237-6915
P. O. Box 3707  02-CA       Everett: (425) 266-3868
Seattle, WA 98124-2207      Page:    (425) 631-8803  


From tjg at exceptionalminds.com  Wed Mar 21 17:31:44 2001
From: tjg at exceptionalminds.com (Timothy Grant)
Date: Wed, 21 Mar 2001 14:31:44 -0800
Subject: Simple regex that has me banging my head against a wall
Message-ID: <20010321143144.K24341@trufflehunter.avalongroup.net>

Hi again.

I actually thought I had this regex thing down pretty well, but
I'm getting beat up by what I thought should be a simple one.

>>> x = '$1$asdfjke$lkjasdkjfsdasdfj'
>>> re.sub(r'\$', r'\\\$', x)
'\\\\$1\\\\$asdfjke\\\\$lkjasdkjfsdasdfj'

All I want to do is substitute \$ for $.

Thanks.

-- 
Stand Fast,
    tjg.

Timothy Grant                         tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.         <><       (503) 246-3630
>>>>>>>>>>>>>Linux, because rebooting is *NOT* normal<<<<<<<<<
>>>>This machine was last rebooted:  64 days  2:43 hours ago<<



From syver at NOSPAMcyberwatcher.com  Tue Mar 27 17:49:45 2001
From: syver at NOSPAMcyberwatcher.com (Syver Enstad)
Date: Wed, 28 Mar 2001 00:49:45 +0200
Subject: How to increase PythonWin v2.0: COM browser limits?
References: <994psh$so5$1@news1.xs4all.nl> <995sgi$7u0$04$1@news.t-online.com> <996flh$e9m$1@news1.xs4all.nl>
Message-ID: <99r5ig$tbg$1@troll.powertech.no>

Try using OleView instead, it think it's freely downloadable with the
platform sdk if you don't have visual studio.

"Twan van der Schoot" <twanvds at xs4all.nl> wrote in message
news:996flh$e9m$1 at news1.xs4all.nl...
> thanks,
>
>
> But I think it is strange that nobody else seems to have the problem,
wonder
> if anybody uses the COM facilities in Pythonwin.
>
> gr
>
> Twan
>
>
> "Eduard Hiti" <KatMouse at gmx.de> wrote in message
> news:995sgi$7u0$04$1 at news.t-online.com...
> > I just tried the COM browser here on my machine, and it gives me the
same
> > error. I'm suspecting an error in Pythonwin...
> >
> > Eduard
> >
> > "Twan van der Schoot" <twanvds at xs4all.nl> wrote in message
> > news:994psh$so5$1 at news1.xs4all.nl...
> > | I'm trying to use the COM Browser in Pythonwin version 2.0.  But I run
> > into
> > | limitations of some kind.  As a Python newby I dutifully scanned FAQs
> and
> > | newsgroups, but alas!
> > |
> > | To reproduce the problem: On a heavily used NT 4.0 system with a lot
of
> > | registered COM stuff, start Pythonwin version 2.0 go to the menu
entry:
> > COM
> > | Browser to start the COM browser and expand the "Registered
Categories"
> > | branch in the tree view.
> > |
> > | Pythonwin shows in the interactive window a backtrace on an exception.
> > The
> > | culprit raising this exception was the response:
> > |
> > | File "D:\Python20\win32com\client\util.py", line 53, in __GetIndex
> > |     result = self._oleobj_.Next(1)
> > | pywintypes.com_error: (-2147024882, 'Not enough storage is available
to
> > | complete this operation.', None, None)
> > | win32ui: Exception in OnNotify() handler
> > |
> > | Something suggests that the number of registered objects is too large
> for
> > | the COM browser.  As I've -of course- the sources available, I was
> wonder
> > if
> > | there is some parameter which I can increase to solve the problem.  Or
> am
> > I
> > | running into limitations of the treeview control of MS Windows?
> > |
> > | TIA
> > |
> > | Twan van der Schoot
> > | twanvds at NOSPAM.xs4all.nl
> > |
> > |
> >
> >
>
>




From cfelling at iae.nl  Thu Mar 29 06:54:45 2001
From: cfelling at iae.nl (Carel Fellinger)
Date: 29 Mar 2001 13:54:45 +0200
Subject: Getting all the *files* from a directory -- A better way??
References: <mailman.985816117.6859.python-list@python.org> <99uu6401u9c@news2.newsguy.com> <PjEw6.4142$4N4.841410@newsc.telia.net>
Message-ID: <99v7q5$30i$1@animus.fel.iae.nl>

Fredrik Lundh <fredrik at pythonware.com> wrote:
> Alex Martelli wrote:
>> So wrap it up...:
>>
>> def joined_listdir(dir):
>>     return [os.path.join(dir,f) for f in os.listdir(dir)]

> or: glob.glob(os.path.join(dir, "*"))

this won't show `hidden' files on unix/linux.


-- 
groetjes, carel


From loewis at informatik.hu-berlin.de  Wed Mar  7 04:48:32 2001
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 07 Mar 2001 10:48:32 +0100
Subject: supporting asyncronous messages
References: <mailman.983793067.17928.python-list@python.org>
Message-ID: <j466hloqof.fsf@informatik.hu-berlin.de>

Joshua Spoerri <josh at pobox.com> writes:

> for example:
> 	a = wfmessage()
> 	b = wfmessage()
> 	a.send()
> 	b.send()
> 	if a.getResponse() || b.getResponse():
> 	whatever
> 
> If b.getResponse() is already available and true,
> but a.getReponse() is blocking, the program shouldn't
> wait for it. (Yes, this does funny things to
> shortcircuiting order.)
> 
> Is there a sane way to support this behaviour without
> hacking up the compiler? Is it bad news for other
> reasons?

Sure. The "design pattern" being used is the same as in select
(reactor):

ready = reactor.pick_one([a,b], timeout)

Then you have delegated the blocking business into pick_one, where you
have a number of options: busy-waiting, multi-threading, the operating
system's select/WaitForMultipleObjects mechanism, and perhaps others -
it much depends on how getResponse does its blocking.

Regards,
Martin


From fredrik at pythonware.com  Wed Mar 14 16:54:09 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 14 Mar 2001 21:54:09 GMT
Subject: reading binary data
References: <3AAFE69D.FF01FEC8@yahoo.fr>
Message-ID: <5MRr6.326$4d3.74189@newsb.telia.net>

Francois Forest wrote:
> I found that I can read a binary file and get the result in a string,
> Is there a way to convert a string slice into an int16, int32, or
> floatting point ?

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

> More generally, how can I browse through standard libraries ?

http://www.pydoc.org

Cheers /F

<!-- (the eff-bot guide to) the standard python library:
http://www.pythonware.com/people/fredrik/librarybook.htm
-->




From gregj at pobox.com  Mon Mar 12 14:50:49 2001
From: gregj at pobox.com (Gregory Jorgensen)
Date: Mon, 12 Mar 2001 19:50:49 GMT
Subject: Einstein's Riddle
References: <mailman.984369618.12493.python-list@python.org> <3AACBB41.5854666A@yahoo.com>
Message-ID: <tM9r6.2585$54.2700@www.newsranger.com>

In article <3AACBB41.5854666A at yahoo.com>, Steve says...
>
>Making assumptions based on vague statements is not a feature of
>"advanced" IQ tests.  Only one IQ test, the Stanford-Binet form L-M, has
>been shown to reliably discern among IQ's higher than about 150, and if
>this question were on that test, the subject would probably be awarded
>extra points if he or she pointed out that this puzzle is unsolvable
>without making an assumption.

I thought making assumption based on vague statements was a feature of mental
illness. I took that Stanford-Binet test once, and I used a tinfoil hat to make
sure they couldn't detect my IQ.


Greg Jorgensen
Deschooling Society
Portland, Oregon, USA
gregj at pobox.com


From wware at world.std.com  Mon Mar  5 17:00:42 2001
From: wware at world.std.com (Will Ware)
Date: Mon, 5 Mar 2001 22:00:42 GMT
Subject: MySQL-python build griefs
Message-ID: <G9qvt6.L0I@world.std.com>

I'm trying to build MySQL-python on a Red Hat 6.2 box. In order
not to confuse the Red Hat installation, which depends heavily
on Python, I've left Python at 1.5.2. I have Distutils 1.0.1
installed. The version of MySQL-python is 0.3.0. To get things
to work at all, I needed to build my own copy of Python-1.5.2/*
to populate the /usr/include/python1.5 tree, running "configure"
to produce a suitable "config.h". Now I'm getting an error on
"python setup.py build" that looks like this:

running build
running build_py
not copying MySQLdb.py (output up-to-date)
not copying CompatMysqldb.py (output up-to-date)
running build_ext
Traceback (innermost last):
  File "setup.py", line 90, in ?
    extra_objects=extra_objects,
  File "/usr/lib/python1.5/site-packages/distutils/core.py", line 138, in setup
    dist.run_commands()
  File "/usr/lib/python1.5/site-packages/distutils/dist.py", line 829, in run_commands
    self.run_command(cmd)
  File "/usr/lib/python1.5/site-packages/distutils/dist.py", line 849, in run_command
    cmd_obj.run()
  File "/usr/lib/python1.5/site-packages/distutils/command/build.py", line 106, in run
    self.run_command(cmd_name)
  File "/usr/lib/python1.5/site-packages/distutils/cmd.py", line 328, in run_command
    self.distribution.run_command(command)
  File "/usr/lib/python1.5/site-packages/distutils/dist.py", line 849, in run_command
    cmd_obj.run()
  File "/usr/lib/python1.5/site-packages/distutils/command/build_ext.py", line 200, in run
    customize_compiler(self.compiler)
  File "/usr/lib/python1.5/site-packages/distutils/sysconfig.py", line 106, in customize_compiler
    (cc, opt, ccshared, ldshared, so_ext) = \
  File "/usr/lib/python1.5/site-packages/distutils/sysconfig.py", line 368, in get_config_vars
    func()
  File "/usr/lib/python1.5/site-packages/distutils/sysconfig.py", line 274, in _init_posix
    parse_makefile(filename, g)
  File "/usr/lib/python1.5/site-packages/distutils/sysconfig.py", line 207, in parse_makefile
    value = value[:m.start()] + done[n] + after
TypeError: illegal argument type for built-in operation

Am I using the wrong color duct tape here? Or just not enough
bubblegum?

-- 
-----------------------------------+---------------------
 22nd century: Esperanto, geodesic | Will Ware
 domes, hovercrafts, metric system | wware at world.std.com


From ws-news at gmx.at  Wed Mar  7 09:31:11 2001
From: ws-news at gmx.at (Werner Schiendl)
Date: Wed, 7 Mar 2001 15:31:11 +0100
Subject: learning C
References: <mailman.983972829.23481.python-list@python.org>
Message-ID: <983975751.18833@newsmaster-04.atnet.at>

Hi,

Laura Creighton <lac at cd.chalmers.se> wrote in message
news:mailman.983972829.23481.python-list at python.org...
> Greetings from Sweden:
>
> I think that objecting to using C because pointers and strings are hard
> is akin to objecting to using a scalpel because it is sharp and you
> can cut yourself.

Really? Is it that obviously off sense?

Do you give your children a scalpel to cut paper? Probably not.
But in surgery there is a need for it.

The fine thing is: C++ still HAS the scalpel, so what?

>  Now if the original poster had said, `Look, I have
> never written anything but Python before, but I have to write some
> application in C or C++, and I don't have any time or inclination
> to actually learn the language much, what should I do?' then I would
> be in utter agreement with you.  Don't learn C unless you really
> want to.  It is hard to be a good C programmer.  It requires discipline
> which the language will not give you, which you must bring to the
> experience yourself.  It helps if you know a really good C programmer
> who will say things like ``no, you cannot put that nauseous piece of slime
> into our nice clean v7 kernel'' which will give you real incentive to
learn
> how to have that discipline.  The world is full of *bad* C programmers,
> though, so finding such a person is hard, especially if you are new
> enough that you do not know the difference between good and bad C code.

Maybe I interpreted the original post somewhat too much.
I actually thought the person was intending to use C to add some
functionality to Python.
Which in fact means to me he does not want to know about every hole in the
road but get it running quickly.

If this was wrong, sorry for my ignorance.

> It was easier in the days when we all just wanted to code like
> Ken Thompson. The person who suggested reading v7 source actually had
> good advice, if only you could get a v7 kernel and a pdp-11 to run it on.

If one is interested in archeology, this is probably really interesting
stuff.
I admit thsat sometime I start some old-system simulator on my own.

But to get productive, I try to use up to date technology.

Maybe you appreciate your silent, quick page printer more if you had a 9
needle piece for years.
But the outcome will stay the same, nice printouts.

> I assumed the original poster wanted to become a good C code writer, or at
> least a good C code reader.  You tell him to learn C++ instead, as if
learning
> C is obselete.

If it is not obsolete, it should be at least considered deprecated for new
projects.

The thing is, you get C for free a much better programming habits I think if
you start out with objects rather than a bunch of unrelated functions (as
long as syntax is concerned, in your mind they will of course have some
relation, but mind tends to fade after a - usually very short - period of
doing sth else) and data pieces.

I compare that to learning to drive. You do not need how your injection
system works _exactly_ and how the wires are mounted and things like that.
You only need to know that if you are really getting a freak in cars... Or
if you make repairing cars your business.
In former days, you had your car engineer with you because it was necessary
to have so. Like one had to use Assembly language and C in former days.
Today you visit your garage every some month to check the thing up - and you
only need Assembly and plain C seldom.

>
> But then you also say that there are no good and bad languages.  Are you
sure
> that you are on the correct mailing list?  The on-going debate about the
new
> string syntax is all about whether or not it is ``good''.  Does it make
> Python *good*? or *bad*?   Important stuff.
>

Well I think that one can learn quite a lot by monitoring what problems
other ones encounter and how they are solved.
This is the main reason why I'm in this list right now.

Hope I can contribute more helpful posts in some time.

If I would not consider Python an interesting language I would not be here.
But I would not choose it to write a windows kernel driver right now.
(I would take C++)


just my opinions

regards
werner




From cce at clarkevans.com  Mon Mar 26 02:31:31 2001
From: cce at clarkevans.com (Clark C. Evans)
Date: Mon, 26 Mar 2001 02:31:31 -0500 (EST)
Subject: Yet Another PEP:  Interface Adapter Mechanism __adapt__
In-Reply-To: <Pine.LNX.4.21.0103252141090.9359-100000@server01.shindich.com>
Message-ID: <Pine.LNX.4.21.0103260212510.15705-100000@clarkevans.com>

On Sun, 25 Mar 2001, Alex Shindich wrote:
> Let's see... The abstract section of the PEP says: "This proposal allows 
> a standard and extensible method to ask two types of questions:  Is a
> particular object compliant with a given protocol?" The following
> describes the strategy pattern: "The Strategy pattern provides a means of
> parameterizing a component to accept an algorithm.  This keeps the
> component simpler, and lets clients pick and choose an algorithm
> specifically tailored to their needs." See any similarities?

The strategy pattern and the adapter pattern do use the concept of
abstraction, but this is as far as the similarity goes.

The strategy is used when there are two or more *algorithems* 
for accomplishing something, and rather than hard-code the 
technique used, this becomes a parameter.  The example used in 
the GOF text is line-breaking algorithems.

The adapter pattern is used to: "convert the interface of a 
class into another interface the clients expect".  See the
pluggable adapter example discussed a TreeDisplay, p142.  

Think of this proposal (and it isn't a PEP yet) as an
bi-directional adapter factory so that users of classes 
don't have to explore to find the appropriate adapter.
That's all, it is rather simple; however, it is a singleton
and standardization would help interoperability.  I'm 
putting it up as a sourceforge project and we will see who
uses it... if it is used and becomes popular, then it
can be submitted as a PEP.  I was too pre-mature and perhaps
out-of-line to suggest otherwise.

> 1. You are making an assumption that each class will provide adaptors to
> some set of protocols, thus entangling the class and available set of
> protocol conversions.

The technique ALLOWS a class to provide an adapter for objects
that it knows about, to convert the interface so that users
expecting the class's behavior will not be suprised.  Nothing
says a class must do this...

> Why can't protocol conversion live outside of the class?

It can.  It can live in an object, whatever.

> 2. Is adaption symmetric? If class A provides an adaptor for class B,
> mustn't class B provide an adaptor for class A?

Adaptation is often not symmetric.  

> 3. Versioning is not mentioned anywhere in your PEP. Once a class defines
> an adaptor for a protocol, the protocol better not change. Otherwise every
> adaptor needs to change to match the new definition of the
> protocol. 

This is not an interface proposal... it is not defining 
what a protocol means nor what it doesn't mean.  Any 
valid Python object can be a protocol.  The impl just
happens to have default behavior for Type and Class
protocols.

> That's all I have got so far... I'll send you more once 
> I read more

Thank you!

Clark




From costas at meezon.com  Fri Mar  9 17:59:17 2001
From: costas at meezon.com (Costas Menico)
Date: Fri, 09 Mar 2001 22:59:17 GMT
Subject: Pythonwin Initialization
Message-ID: <3aa95fb8.350774771@News.CIS.DFN.DE>

I would like to execute a .py file that initializes my Pythonwin or
Python environment. Is there some recommended way of doing this
automatcially when Python starts up?

Thanks

Costas 


From pj at sgi.com  Sun Mar  4 21:46:24 2001
From: pj at sgi.com (Paul Jackson)
Date: 5 Mar 2001 02:46:24 GMT
Subject: Thinking Pythonically (was Re: gah! I hate the new string syntax)
References: <mailman.983485923.1136.python-list@python.org> <97pcu2$n6n$1@animus.fel.iae.nl> <97ptag$81s65$1@fido.engr.sgi.com> <YUbo6.4854$ML1.258365@e420r-atl2.usenetserver.com>
Message-ID: <97uum0$h0dk$1@fido.engr.sgi.com>

Steve asks:
|> Do you mind if I quote you, with appropriate attribution?

Sure, that's fine.  Thanks for the smile you
just put on my face.
-- 
                          I won't rest till it's the best ...
                          Manager, Linux System Software
                          Paul Jackson <pj at sgi.com> 1.650.933.1373


From johnson2 at hiwaay.net  Sun Mar 25 22:54:15 2001
From: johnson2 at hiwaay.net (Lyle Johnson)
Date: Sun, 25 Mar 2001 21:54:15 -0600
Subject: Good GUI for Python
References: <mailman.985314033.5120.python-list@python.org> <AADu6.436656$f36.14223632@news20.bellglobal.com> <3ABBFB52.1FB59ABA@NoNospammytheworld.com>
Message-ID: <tbtettlsjttk47@corp.supernews.com>

> What are the licensing issues referred to below on pyQT?

Here is a link to one of TrollTech's FAQ pages concerning the licensing of
Qt:

    http://www.trolltech.com/developer/faq/simple.html

To summarize, there is a free edition of Qt for Unix/X11 only, which can be
used for developing free software. Because this free edition is covered by
the GPL it cannot be used to developed closed-source or commercial software.
Note that if you're just an end-user of the software (including, for
example, the KDE) you can use Qt free edition with no worries.

There is no free edition of Qt for Windows at this time. Furthermore, if you
want to develop closed-source or commercial software (regardless of the
target platform) you must purchase developer licenses for either the
Professional or Enterprise editions of Qt; it appears that these developer
licenses start out at around $1500 per developer, with discounts if you buy
multiple licenses. There are no runtime royalties; you just pay for the
developer licenses.

So this puts Qt at a bit of a disadvantage (with regards to other, "freer",
GUIs like GTK, wxWindows and FOX) if you want to develop free software for
both Unix and Windows.

I do not speak for Trolltech and have not had any business dealings with
them; I'm just trying to summarize what I think I'm reading on their web
site (http://www.trolltech.com). Apologies in advance if I have
misrepresented anything there.

Hope this helps,

Lyle




From aleaxit at yahoo.com  Tue Mar  6 04:25:28 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Tue, 6 Mar 2001 10:25:28 +0100
Subject: I come to praise .join, not to bury it...
References: <mailman.983756766.27347.python-list@python.org> <97vnu3076q@news2.newsguy.com> <slrn9a89dk.2rm.hzhu@rocket.knowledgetrack.com>
Message-ID: <982ae80t56@news1.newsguy.com>

"Huaiyu Zhu" <hzhu at users.sourceforge.net> wrote in message
news:slrn9a89dk.2rm.hzhu at rocket.knowledgetrack.com...
> It is all nice and well to say that sep.join(list) is good for
polymorphism,
> except that there are several practical annoyances:
>
> 1. The word join can be either transitive or nontransitive:
>    sep joins list.
>    list joins with sep.

The first way one tries to parse a method-name is as the
imperative form of some verb -- transitive if it takes
arguments, intransitive if it's without arguments.  I am
not a native speaker of English, but I strongly suspect
this generalizes to those who are.

> 2. Can we really join all kinds of lists?

We can join any kind of sequence which defines a length,
with some irregularities and bugs (regarding, not the
sequence itself, but the types of its _items_) depending
on the kind of joiner object: when the joiner object is
a single-byte string, the items must also be (unicode
object items are supposed to be accepted too, but there
is a bug in Python 2.0 which interferes sometimes); when
the joiner object is a Unicode string, the items may be
any object that defines a way to make it into Unicode
(in Python 2, e.g., an instance object with a __str__
method).

I hope some of these irregularities go away in 2.1...:-).


Alex





From aahz at panix.com  Fri Mar 16 19:14:11 2001
From: aahz at panix.com (Aahz Maruch)
Date: 16 Mar 2001 16:14:11 -0800
Subject: HTMLparsing abnormal html pages
References: <98pvp1$15t$1@news.netmar.com>
Message-ID: <98ua8j$a6h$1@panix2.panix.com>

In article <98pvp1$15t$1 at news.netmar.com>,  <asle at spam.com> wrote:
>
>Considering the small program below. Running it will show that the
>HTMLparser
>is truncating urls in the HTML page. Now, most of you will probably say that
>the page and in particular the URL's of this page are not valid according to
>the RFC1738 protocol --bad luck. But there must be a work-around for this?

For this specific case, Mark's solution may well work (haven't tested it
myself).  But you cannot easily find a generic solution because of all
the different ways to mangle HTML.
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

Three sins: BJ, B&J, B&J


From loredo at spacenet.tn.cornell.edu  Thu Mar 29 19:53:06 2001
From: loredo at spacenet.tn.cornell.edu (Tom Loredo)
Date: Thu, 29 Mar 2001 19:53:06 -0500
Subject: NumPY: Reading/Writing Arrays To/From Files
References: <3aae3bd6$0$42882$272ea4a1@news.execpc.com> <3AC1F37D.E8ADA48@seebelow.org> <Pine.SOL.4.30.0103291907490.13267-100000@primrose.csv.warwick.ac.uk>
Message-ID: <3AC3D8F2.1BC4E930@spacenet.tn.cornell.edu>

"John J. Lee" wrote:
> 
> ArrayIO I think is for binary data, not ASCII.  As far as I can tell,
> Travis' pages haven't resurfaced since they disappeared a few months ago.

The new URL was posted here a while ago:

http://pylab.sourceforge.net/

This stuff is important enough that the link should be in some
prominent place, like at the NumPy project web page.  Oh... it is!

http://pfdubois.com/numpy/

8-)

-Tom Loredo


From aleaxit at yahoo.com  Wed Mar  7 03:29:54 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Wed, 7 Mar 2001 09:29:54 +0100
Subject: sorting on IP addresses
References: <3A7F26CB.BE85C96F@esec.com.au> <mailman.983939410.22287.python-list@python.org>
Message-ID: <984ri10uem@news1.newsguy.com>

"Justin Sheehy" <justin at iago.org> wrote in message
news:mailman.983939410.22287.python-list at python.org...
> Sam Wun <swun at esec.com.au> writes:
>
> > Does anyone know what is the quickly way to sort a list of IP addresses?
> >
> > ie. 203.21.254.89 should be larger than 203.21.254.9
>
> Convert to longs and sort on those values?
>
> Not all that quick, I suppose, but I don't know of a faster correct way.
>
> I suppose you could do it one octet at a time, which might be faster
> for cases like your example where the addresses are fairly close.
>
> The actual code for either of those approaches is short and
> straightforward.  Just write a little function to compare two
> addresses using the method of your choice, and pass it to your list's
> sort method.

...or, if the list is longish, use the decorate/sort/undecorate
design-pattern, which is faster for all but very short lists.

def sort_list_of_ips(list_of_ips):
    import socket
    temp = map(socket.inet_aton, list_of_ips)
    temp.sort()
    return map(socket.inet_ntoa, temp)

The full-fledged d/s/u pattern would append each item to its
sort-key, but we don't need it here since we can recover the
item's full information-content from the key itself.  Thus,
we can use the faster map for decoration and undecoration,
rather than list-comprehensions as are generally needed.


Alex





From alex at shindich.com  Sun Mar  4 20:47:46 2001
From: alex at shindich.com (Alex Shindich)
Date: Sun, 4 Mar 2001 17:47:46 -0800
Subject: Static class methods
References: <rdzo6.154$gc4.80445@news.pacbell.net> <ta5ka1ks0buc3b@corp.supernews.com> <97ul9k$n2l$02$1@news.t-online.com>
Message-ID: <aeCo6.166$VX5.157081@news.pacbell.net>

----- Original Message -----
From: "Sven Peters" <svpeters at t-online.de>
Newsgroups: comp.lang.python
Sent: Sunday, March 04, 2001 4:06 PM
Subject: Re: Static class methods


> Chris Armstrong wrote:
>
> > I have trouble believing that something like this is really needed.
> > Could you give a real-world example?
> Factoty methods, for getting objects created earlier and no longer needed,
> yet expensive to create e.g. database connection pooling.
> Sure, you can place this elsewhere(module level or other objects).
> Interresting:
> You CAN have class methods:
> def bert:
>         print "Bert"
> class.method = bert
> class.method()
> "Bert"
>
> ... but no language support to create them as easy as instance methods
>
> --
> Sven Peters
Amen! Most factories are implemented as module-level functions today. This
decouples classes from factories. Sometimes this behavior is desired.
Sometimes a factory method pattern is more appropriate. As of today, Python
doesn't allow for factory methods.
Another need for static methods comes from a very basic OOP principle --
encapsulation. Sometimes a private class-level utility function is needed.
The only way to have a function like that today is to have a private
instance method. This is damn ugly!

Regards,

Alex Shindich
visit http://www.shindich.com/sources/
P.S. The code snippet above won't work, because an operation of adding a
function attribute to any instance creates an attribute of type instance
method:
>>> def bert ():
 print "Bert"


>>> class Foo: pass

>>> Foo.method = bert
>>> Foo.method ()
Traceback (most recent call last):
  File "<pyshell#6>", line 1, in ?
    Foo.method ()
TypeError: unbound method bert() must be called with instance as first
argument
>>>




From lac at cd.chalmers.se  Wed Mar  7 08:45:44 2001
From: lac at cd.chalmers.se (Laura Creighton)
Date: Wed, 7 Mar 2001 14:45:44 +0100 (MET)
Subject: learning C
Message-ID: <200103071345.OAA06502@boris.cd.chalmers.se>

Greetings from Sweden:

I think that objecting to using C because pointers and strings are hard
is akin to objecting to using a scalpel because it is sharp and you
can cut yourself.  Now if the original poster had said, `Look, I have
never written anything but Python before, but I have to write some 
application in C or C++, and I don't have any time or inclination
to actually learn the language much, what should I do?' then I would
be in utter agreement with you.  Don't learn C unless you really
want to.  It is hard to be a good C programmer.  It requires discipline
which the language will not give you, which you must bring to the 
experience yourself.  It helps if you know a really good C programmer
who will say things like ``no, you cannot put that nauseous piece of slime 
into our nice clean v7 kernel'' which will give you real incentive to learn
how to have that discipline.  The world is full of *bad* C programmers,
though, so finding such a person is hard, especially if you are new
enough that you do not know the difference between good and bad C code. 
It was easier in the days when we all just wanted to code like 
Ken Thompson. The person who suggested reading v7 source actually had 
good advice, if only you could get a v7 kernel and a pdp-11 to run it on.
I assumed the original poster wanted to become a good C code writer, or at
least a good C code reader.  You tell him to learn C++ instead, as if learning
C is obselete.

But then you also say that there are no good and bad languages.  Are you sure 
that you are on the correct mailing list?  The on-going debate about the new 
string syntax is all about whether or not it is ``good''.  Does it make
Python *good*? or *bad*?   Important stuff.

Laura Creighton



From eddthompson at ssi.nospamhere.parlorcity.com  Thu Mar 29 23:25:08 2001
From: eddthompson at ssi.nospamhere.parlorcity.com (EuphoriaDj)
Date: Thu, 29 Mar 2001 23:25:08 -0500
Subject: list vs tuple
References: <mailman.985789357.19622.python-list@python.org>
Message-ID: <3ac40acd$0$10576@wodc7nh0.news.uu.net>

"Simon Brunning" <SBrunning at trisystems.co.uk> wrote in message
news:mailman.985789357.19622.python-list at python.org...
> > From: deadmeat [SMTP:root@[127.0.0.1]]
> > Whats the practical difference between a list[] and a tuple() ?
> >
> > A tuple seems to be a very basic list, so unless it's faster, what's the
> > point of it?
>
> Off the top of my head - A tuple is immutable, and therefore hashable, and
> therefore can be used as a key to a dictionary. I'm sure that there are
> other things...
>
what do you mean by hashable?




From MarkH at ActiveState.com  Fri Mar  2 17:27:27 2001
From: MarkH at ActiveState.com (Mark Hammond)
Date: Fri, 02 Mar 2001 22:27:27 GMT
Subject: Communicating with serial port
References: <ndNn6.5161$t21.152593@news3.oke.nextra.no>
Message-ID: <3AA01D88.4020503@ActiveState.com>

Peter Blom wrote:

> I shall write a python program for downloading data from a memory card in a
> measuring instrument, via a serial port.
> For this purpose, I have downloaded SioModule.zip - a serial comm extension
> for windows 95, upgraded to Python 1.5.1.
> This does not seem to work - when importing the module 'serial.py', my
> system hangs (Windows 98, Activestate Python 2.0).
> Could anybody give a hint?
> 
> Peter (a newbie)

I dont know where an SioModule that works with Python 2.0 can be found. 
  It sounds like you found a 1.5.1 version, but are trying to use it 
with Python 2.0.

Check out win32\demos\win32comport_Demo.py for an alternative way of 
hitting the serial port on Windows.

Mark.



From syver at NOSPAMcyberwatcher.com  Tue Mar 13 08:52:12 2001
From: syver at NOSPAMcyberwatcher.com (Syver Enstad)
Date: Tue, 13 Mar 2001 14:52:12 +0100
Subject: win32all buld 137 win32gui fails
Message-ID: <98l8qk$oeo$1@troll.powertech.no>

I've just installed the win32all build 137 that contains the support for
Vtable interfaces in win32com. I've suddenly found that import win32gui
fails with the message:

ImportError: DLL load failed: The specified module could not be found.

This is mighty strange as the pyd file is lying there among the other .pyd
files like win32help and so on. Maybe there are some other modules that
win32gui depends on that it didn't depend on before?

Anybody have a suggestion to fix this?

For the record I am running Win2000 Pro with no service packs, VC++ 6 to
service pack 3, IE 5.





From duncan at rcp.co.uk  Wed Mar 21 09:53:30 2001
From: duncan at rcp.co.uk (Duncan Booth)
Date: Wed, 21 Mar 2001 14:53:30 +0000 (UTC)
Subject: [Newbie] line separators in file-output
References: <3AB8AEDA.2D2BD105@bg-sys.de>
Message-ID: <Xns906B97111E9AEduncanrcpcouk@127.0.0.1>

Daniel Orlowski <daniel.orlowski at bg-sys.de> wrote in <3AB8AEDA.2D2BD105 at bg-
sys.de>:

>If I in Python uder Windows do smth like this:
>
>>>> File=open("test","w")
>>>> File.write("a line"+"\n")
>>>> File.close()
>
>I have data in file test:
>'a line\015\012'
>
>How can I prevent output of  '\015'?
>
Open the file in binary mode:
>>> File=open("test", "wb")



From grante at visi.com  Tue Mar 13 17:54:45 2001
From: grante at visi.com (Grant Edwards)
Date: Tue, 13 Mar 2001 22:54:45 GMT
Subject: This math scares me
References: <3aad4062.1290021@News.CIS.DFN.DE> <20010312170356.B3513@better.net> <mailman.984480856.18097.python-list@python.org> <984486395.407379@newsmaster-04.atnet.at> <slrn9asf1d.ja3.qrczak@qrnik.zagroda> <984518199.127140@newsmaster-04.atnet.at> <jwbnews-09D75F.13313313032001@corp.supernews.com> <984521930.27710@newsmaster-04.atnet.at>
Message-ID: <Vyxr6.1238$Tg.168802@ruti.visi.com>

In article <984521930.27710 at newsmaster-04.atnet.at>, Werner Schiendl wrote:
>> Bill Gates got this right in the early days, delivering two Basic
>> interpreters:  binary floating point for speed, and decimal floating
>> point for money and such.  (I never looked at how well his interpreters
>> did...I just used them.)
>
>Does this make a difference, as soon as arithmetics takes place?

Yes.  All FP calculations took place in BCD.  You got exact
results as long as your values didn't exceed the
representations limits.

-- 
Grant Edwards                   grante             Yow!  ... The waitress's
                                  at               UNIFORM sheds TARTAR SAUCE
                               visi.com            like an 8" by 10" GLOSSY...


From korpo at 01019freenet.de  Thu Mar 29 13:12:17 2001
From: korpo at 01019freenet.de (Oliver Korpilla)
Date: Thu, 29 Mar 2001 20:12:17 +0200
Subject: Recommendations for Python Books
References: <3ac2582c$1@news.starhub.net.sg>
Message-ID: <3AC37B01.F009E937@01019freenet.de>

Wong Da Qiang wrote:

> Just a newbie to programming and python, need some recommendations on books
> to buy

There's a new book on the market "Core Python" from Prentice Hall, you may get
it for example at amazon. I have O'Reilly books "Learning Python" and
"Programming Python", the 1st one being really a good introduction to Python,
the 2nd one just a rather lousy collection of examples . If "Core Python" will
be as good as "Core Java2", you may use it as a very good reference and as a
learning book as well. Get may hands on it this weekend / next week, but I
guess it will be very recommnedable.

Ciao, Korpo



From smnordby at yahoo.com  Wed Mar 14 07:25:54 2001
From: smnordby at yahoo.com (Steve)
Date: Wed, 14 Mar 2001 04:25:54 -0800
Subject: Einstein's Riddle
References: <mailman.984369618.12493.python-list@python.org> <3AACBB41.5854666A@yahoo.com> <3AAF1885.FF28CC21@yahoo.com>
Message-ID: <3AAF6352.726046A2@yahoo.com>

Reliability is an indication of whether a measurement is repeatable:
Does it give consistent results under the same conditions?  Subjects in
both the lowest and highest IQ ranges (more 3 standard deviations from
the mean) score more consistently on repeated administrations of the SM
L-M (not the more common SB IV) than on any other IQ test available. 
Your question is along the lines of external validity: Is the SM L-M a
valid IQ test?  The answer to that comes from consistency with other
tests in the range for which those other tests are reliable.  The SB L-M
does well here also.

-SteveN-
Glossary of GIfted Education
http://members.aol.com/svennord/ed/GiftedGlossary.htm

Paul Winkler wrote:
> 
> Steve wrote:
> (snip)
> > Only one IQ test, the Stanford-Binet form L-M, has
> > been shown to reliably discern among IQ's higher than about 150,
> 
> If there are no other reliable measures of IQ > 150, how can the S-B
> test be proven reliable?
> 
> --
> .................    paul winkler    ..................
> slinkP arts:   music, sound, illustration, design, etc.
>            web page:  http://www.slinkp.com
>       A member of ARMS:   http://www.reacharms.com


From joconnor at cybermesa.com  Mon Mar  5 12:59:27 2001
From: joconnor at cybermesa.com (Jay O'Connor)
Date: Mon, 05 Mar 2001 17:59:27 GMT
Subject: Little problem with the "+" operator
References: <983802310.155838463@news.t-online.de> <mailman.983804468.16428.python-list@python.org> <980i6q$crs$1@news.mathworks.com> <mt1yscm7yu.fsf@astron.berkeley.edu>
Message-ID: <3aa3d3a7.420988249@news.uswest.net>

On 05 Mar 2001 09:38:49 -0800, Johann Hibschman
<johann at physics.berkeley.edu> wrote:

>Joshua Marshall writes:
>
>>> This is infinitely more sane than:
>
>>> % perl -e 'print 500 + "cool"'
>>> 500
>
>> Comparing Python to Perl here doesn't seem very fair.  They're very
>> different languages.  In general, I'm surprised by the urgency with
>> which people often jump to compare them.
>
>Well, Perl's the only language that I know of that lets you add 500 to
>"cool" without an error.  Well, I suppose tcl might.  But it's still a
>fairly safe inference that the original poster was coming from Perl.


Actually, JavaScript will allow you to add numbers to strings, with
weird results

'100' + 200 = '100200'
100 + '200' = 300

I *think* that's how it works, I just know it's not very safe	





From jtorborg at nd.edu  Thu Mar  8 12:23:58 2001
From: jtorborg at nd.edu (Julie Torborg)
Date: Thu, 08 Mar 2001 11:23:58 -0600
Subject: Some simply python/Tkinter problems
Message-ID: <3AA7C02E.9A542F34@nd.edu>

I'm a relatively new user of Python, trying to make some GUIs using
Tkinter.  Postings on this newsgroup have been quite helpful, but there
are still some things I can't find information on:

1.  I'd like to be able to put a working "print" command into my GUI, so
that I can print what is displayed on-screen.  I'm wondering if this is
even possible using Tkinter, since there seems to be no info on it at
all.  (p.s. Versatility is nice, but all I really need is to print to a
postscript printer.)

2. This GUI pops up multiple windows, and I'd like to be able to close
them individually.  All the Python books put a "close" option in their
menu, but none of them have a working callback other than
"sys.exit(0)"--but I don't want to exit the whole system, just the
Toplevel.  I've tried using destroy(), but my process is iterative, so
some windows have the same name and only the last called can be
destroyed (I've tried indexing the window names and passing the window
name as an argument, but these don't seem to work).

3. I'd like to organize my modules into different folders (I'm running
on Windows 98), but my scripts can only "see" modules in the same folder
as Python.exe.

If anybody has any solutions, or even where to look for them, yell.



From thomas at xs4all.net  Thu Mar  1 09:22:13 2001
From: thomas at xs4all.net (Thomas Wouters)
Date: Thu, 1 Mar 2001 15:22:13 +0100
Subject: Allow anything in identifiers?
In-Reply-To: <3A9E3ACE.60F8C639@home.se>; from sverker.is@home.se on Thu, Mar 01, 2001 at 01:04:30PM +0100
References: <3A9DC13A.1647EE51@home.se> <mailman.983420883.30039.python-list@python.org> <3A9E3ACE.60F8C639@home.se>
Message-ID: <20010301152213.K9678@xs4all.nl>

On Thu, Mar 01, 2001 at 01:04:30PM +0100, Sverker Nilsson wrote:
> David Porter wrote:

> > > alu.id'+' = lambda x, y...

> > It would break any code using the name 'id', likely a lot of code.

> How? Not id <string>, should likely be on the same line, may allow
> space in between. That gives a syntax error currently, except when id
> is some of a few special ones like r, u, etc. 

If you want for id"string" as a string literal, the example above will not
work. String literals create strings, not identifiers. Creating a special
kidn of string that really creates an identifier would be... Weird.

>   operator.__lt__ (though I am just guessing about
>   [that] one and I  am sure it's wrong - that's a point of the
>   argument :-)

The latter doesn't exist. But neither does operator.__cmp__, so it's not
that big a suprise.

> * Same holds for the corresponding class methods __add__ etc.

Actually, you guessed exactly right with __lt__. That's the name for the
less-than operator 'magic method'. But even with special syntax to declare
operator'+', you can't throw away all magic names. What to do with __cmp__,
__init__, __del__, __setitem/attr__ ?

> * Command interpreters could dispatch to functions that had obvious
>   names corresponding to the actual commands.

> * Variables corresponding to files with fixed name, could have
>   the obvious name.

>>> class X:
...     pass
... 
>>> x = X()
>>> setattr(x, '%$@(*#@$+', 2)
>>> getattr(x, '%$@(*#@$+')   
2

I don't see a lot of people using that hack. I also don't see how using that
kind of variable name can be any clearer. I especially dont believe in
'fixed named files': you should name your variables according to what
function they perform, not what they contain -- what they contain is already
obvious ;) I might call a template file
'/usr/local/lib/<application group>/<application>.tmpl', but the variable
would simply be called 'tmpl' or 'TEMPLATE' or some such.

> One drawback I can think of myself right now is a kind of uglyness
> by using quotes but at least they make the construct stand out so it
> should not be so confusing to ordinary id's.

The main drawback I see is your 'strong point': I don't believe allowing
more characters in identifiers is going to make programs easier to read or
follow, or going to make constructs more selfexplanatory. I just see added
complexity with no obvious benifit.

-- 
Thomas Wouters <thomas at xs4all.net>

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



From aleaxit at yahoo.com  Thu Mar 22 07:00:01 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 22 Mar 2001 13:00:01 +0100
Subject: Functionalism, aesthetics Was:(RE: I come to praise .join,   not
References: <mailman.985052844.17993.python-list@python.org> <mailman.985145848.16117.python-list@python.org>
Message-ID: <99cpg101iij@news2.newsguy.com>

"Carlos Ribeiro" <carribeiro at yahoo.com> wrote in message
news:mailman.985145848.16117.python-list at python.org...
> At 08:11 20/03/01 +0100, Alex Martelli wrote:
> >Back when "Concorde" and "Jumbo" first appeared on the scene, for
> >example, non-experts found the former quite beautiful, the latter quite
> >plain.  Sure the sleek, fast-looking lines of Concorde were superior to
> >the goofy-looking fatness of Jumbo?
>
> Are you sure that this is a good example? One thing is to be impressed by
> the look, and the other is to actually use the product. Opinions may (and

Oh yes, it's an EXCELLENT parallel, EXACTLY because of this distinction.

People who are attacking the joiner.join(sequence) syntax [some arguing
for sequence.join(joiner), others for string.join(sequence, joiner)] are
NOT doing so on the basis of having actually used the various possible
syntaxes and having found functional/operational defects in the one our
BDFL has blessed.  Rather, they handwave about this operation 'clearly'
belonging to the sequence-object rather than to the joiner-object, or
'clearly' to neither one, or (more often) the resulting code 'looking
bad'.  The first two lines of reasoning are based on considerations of
functionality; are, thus, rationally debatable; and I think I have
shown why (in a language, such as Python, where general polymorphism
applies to the object to the left of the '.') the various functional
characteristics of the architecture chosen by the BDFL are Just Right.

The line of, well, 'argument' (loosely speaking...) based on "it looks
bad", on the other hand, is basically a waste of good bits.  If the
functionality IS right (let's settle *that*, first!), and the chosen
form follows that (right) functionality, then any perception that
the chosen form is 'bad' must be the consequence of an optical
illusion -- and, since it IS well known that humans (even quite
experienced and well-trained ones) ARE rather subject to optical
illusions and other misperceptions, this hypothesis is anything
but far-fetched.  The only item of interest remaining is how best
to present the rationale for the joiner.join(sequence) syntax, so
as to help as many users as possible overcome this particular
misperception (ideally by understanding enough of its root causes
that correcting this identified misperception may also help with
as-yet-unidentified further ones).  "Arguing" (loosely speaking...)
on "pretty" vs "ugly", on the other hand, is not productive.


> >issues.  Vox populi is NOT necessarily vox Dei -- not when
> >the underlying technology is too new (a few centuries) and/or
> >not reflected in immediately apperceivable traits.
>
> I'm not talking about completely untrained users here. Most doctors would
> appreciate the explanation given by the pharmaceutical industry. And while

Definitely -- they *ARE* well trained in the rudiments of chemistry
and pharmacology, after all!

> I agree that Vox Populi != Vox Dei, this has *nothing* to do with the
> technology maturity as measured by time alone. Our innate reaction against
> change play a big role here too.

There is a dynamic balance between neophily and neophoby -- if our
innate reaction was ONLY against change, then you would hardly see
the "new! improved!" labels on detergents boxes on supermarket
shelves, would you?  Nor would the clothing fashion industry thrive.

Craftily triggering and exploiting the -phily or -phoby part is
an important marketing technique, but not very germane here.  What
IS germane is that there are some fields (where we're biologically
predisposed and/or collectively trained by centuries or millennia
of shared cultural experiences) where "innate" reactions of
not-specially-trained users must carry enormous weight; in other
fields, ones that are newer and/or farther from our biological
specializations (characteristics that often go together), the
issues are very different (training is needed to give us enough
appropriate memes to overcome the lack, or counterproductivity,
of our genes' influence on untrained/instinctive judgments).


> > > To summarize it: a  *technically* optimal solution for a problem is
not
> > > always the *best* solution. If it does not look intuitively good for
most
> > > users, its probably a bad solution.
> >
> >So, you're arguing that the amount of artificial coloring should
> >have been kept high (who cares about giving users cancer, right?),
> >since that bright color was definitely what "looked intuitively
> >good for most users" as proved by the endless carping...?
>
> Ok, ok. ' Users' is misleading. However, for Python, 'users' usually means
> programmers, people with sufficient knowledge to make good questions about
> proposed changes.

I think it's a _minority_ of programmers who fully understand
and "intuitively" grasp issues of polymorphism (and others that
relate to the depths of O-O: inheritance, delegation, multiple
inheritance, singletons, holding vs referring-to, etc, etc) in
languages that are richly multi-paradigm (and Python is very
much so, at least on a par with C++).  Ingrained habits, and
principles not critically examined but subconsciously distilled
from experience that may well not be relevant to the current
situation, stand in the way, far too much so.

If you heard such assertions from somebody whose business is
selling training or advice, you might reasonably be suspicious
that his self-interest may be prompting them.  As for me, while
I do like teaching and advising, I'd still rather be _DOING_
coding and design -- yet, I *must* invest a goodly proportion
of my working days to consulting, design-reviews, code-reviews,
preparing and teaching seminars, etc, because the firm I work
for has determined that this is a very high-leverage use of
my time (and, I will admit, they ARE right about it).  To be
clear about it, it's my _colleagues_ I'm training and advising
(and I have been doing so, for a growing fraction of my work
days, ever since I joined this company over 12 years ago);
_brilliant_ people, one and all, programmers of varying but
generally considerable experience, guys and gals I'm PROUD to
be working with; yet, with all that, the consulting and the
teaching are _still_ absolutely necessary to make them as
productive as can be in their design and coding activities.

I think this gives me considerable perspective and experience
on what conceptual and perceptual errors brilliant (but
typically overworked) practising programmers are prone to.
(I've also taught programming newbies, e.g. back when I was
moonlighting as a university professor, so I also know
something of the rather different set of misperceptions
that _they_ are most at risk of -- but my experience of
that is nowhere close to what I have in dealing with people
who have been programming professionally for years).

And I don't find "discussion" (loosely speaking) based on
'aesthetics' of some system-architecture choice is very
productive at all in this context.  Far too often, the
(brilliant, mind you!) practising programmer who can't
articulate his/her architectural objections beyond the
stage of "it _looks_ bad" is subconsciously 'deducing'
said objections from other, not-applicable-here contexts
he or she is more familiar with.  _Refusing_ to 'debate'
(what debate?!) on the plane of 'aesthetics', and firmly
insisting that very specific and concrete (functional!)
grounds for the objections be discovered, is by far the
most productive stance in such cases, in my experience.


> > > Its unfortunate that in many cases the non-expert user keeps quiet,
> > > assuming that whatever opinion the "master" haves, its going to be
better
> > > than theirs - after all they are the so-called experts.
> >
> >Ask the experts in any field whether untrained users keep too quiet,
> >or not enough, in their field, and you'll get a different opinion (of
> >course, self-serving conscious or unconscious biases are at work
> >as well!).
>
> The fact that many people complain without reason is not a valid excuse
for
> ignoring all complaints.

But refusing to accept complainants who are unable or
unwilling to work to put their complaints in specific
and concrete terms can be by far the best approach.


Alex





From grisha at verio.net  Fri Mar  9 13:39:34 2001
From: grisha at verio.net (Gregory (Grisha) Trubetskoy)
Date: Fri, 9 Mar 2001 13:39:34 -0500 (EST)
Subject: [mod_python] help needed with forms' parameters handling
In-Reply-To: <86FC34EA3C4F074AA6BBFFE52ED83FED27BE3E@METAEXCH.sfo.metamarkets.com>
Message-ID: <Pine.BSF.4.32.0103091338010.53766-100000@localhost>

This is probably because FieldStorage indeed doesn't have "items()" - just
like it says ;-)

May be in the future I'll add it. Until then - you'll have to copy it to
real dictionary, or you can subclass FieldStorage and add your own
items().

Grisha

On Fri, 9 Mar 2001, Petar Karafezov wrote:

> Hello all -
>
> I am running Apache/1.3.14 (Unix) mod_python/2.7.2 Python/2.0. At this point
> I am not even sure that this is a mod_python issue, as I have seen this
> same behavior before I ported my cgi to mod_python. I am not sure what to do
> at this point..
>
> Here's the issue:
>
> When you do a POST to a python script you obtain the parameters using the
> util.FieldStorage, when using cgi it's cgi.FieldStorage. The returned result
> is a dictionary { key, value }. Everything is OK if you just what to work
> with this dictionary. However, if you try to pass this dictionary object to
> urllib.urlencode, we raise an exception:
>
> Mod_python error: "PythonHandler parse2"
>
> Traceback (most recent call last):
>
>   File "/usr/local/lib/python2.0/site-packages/mod_python/apache.py", line
> 189, in Dispatch
>     result = object(self.req)
>
>   File "/usr/local/apache/htdocs/test/parse2.py", line 147, in handler
>     params=urllib.urlencode(form)
>
>   File "/usr/local/lib/python2.0/urllib.py", line 1086, in urlencode
>     for k, v in dict.items():
>
> AttributeError: 'FieldStorage' instance has no attribute 'items'
>
>
> Now, the funny part is that if I get this object 'form' - which was defined
> as form=util.FieldStorage(req) and just get the keys and values and form a
> new dictionary, I can pass that new one to urllib.urlencode with no
> problems.
>
> ...
>     newparams={}
>     for key in form.keys():
>           newparams[key]=form[key]
>     params=urllib.urlencode(newparams)
> ...
>
> Again, I run into the same issue with both util.FieldStorage when working
> with mod_python and with cgi.FieldStorage when using my cgi version.
>
> I am quite happy with this workaround until I bumped into a different issue,
> which I am not exactly sure if it's related or not. When there're multiple
> checkboxes on a form:
>
> <input type="checkbox"  name="frequencies" value="DEM" priority="6" checked>
> <input type="checkbox"  name="frequencies" value="HLF" priority="6">
> <input type="checkbox"  name="frequencies" value="DAY" priority="6">
> <input type="checkbox"  name="frequencies" value="WEK" priority="6">
>
> and we have selected more than 1, then the form parameter value gets passed
> to the server as an array. For example, if I'd select DEM and DAY from the
> checkboxed above, my params would be:
>
> 'frequences' : ['DEM','DAY']
>
> This is all great, but when I pass the parameters to the server - my python
> script is playing the role of a proxy, then I am passing as value for
> 'frequences' the STRING "['DEM','DAY']" and not the ARRAY ['DEM','DAY'].
>
> So, somewhere in what I do/use the array gets converted to a string. The way
> I see it is:
>
> 1) It could be the python's FieldStorage when I originally get the keys and
> values
> 2) It could be that funky dictionary rewriting that I am doing before I pass
> the form parameters to urllib.urlencode.
> 3) I could be something in what urlencode does.
>
>
> I am not sure where to look at this point.
>
> Any ideas, anyone ?? Any help would be greatly appreciated!
>
> Thanks for your help!
>
> regards,
>
> Petar Karafezov
>
> MetaMarkets.com
> 415-575-3015
> -------------------------------------------
> Investing Out Loud at
> http://www.metamarkets.com
> -------------------------------------------
>
>
> _______________________________________________
> Mod_python mailing list
> Mod_python at modpython.org
> http://www.modpython.org/mailman/listinfo/mod_python
>




From takto at alum.mit.edu.-  Tue Mar 20 10:15:50 2001
From: takto at alum.mit.edu.- (Tak To)
Date: Tue, 20 Mar 2001 10:15:50 -0500
Subject: Collection interfaces
References: <3a90fa97.0@news1.mweb.co.za>   <MPG.1503548620e632919896a4@news.earthlink.net> <MCjm6.474652$U46.14163288@news1.sttls1.wa.home.com> <MPG.150394195f7f04a19896a7@news.earthlink.net> <3A9A86B1.9A8C69A2@ix.netcom.com> <MPG.1504d14563dce6499896af@news.earthlink.net> <97fjla128ir@news2.newsguy.com> <MPG.1505a76831c34b749896b3@news.earthlink.net> <97h5s40vhc@news2.newsguy.com> <MPG.1506007591cf818b9896bc@news.earthlink.net> <slrn99opop.gr.amk@mira.erols.com> <MPG.1506265ebf6d934d9896be@news.earthlink.net> <97ibmc0k47@news1.newsguy.com> <MPG.1506cf3293b094b49896c1@news.earthlink.net> <97m0c7080c@news2.newsguy.com> <MPG.15085517debf4b649896ca@news.earthlink.net> <zb%n6.31033$o%6.586026@news1.rdc1.bc.home.com> <MPG.150b383cea226d409896d0@news.earthlink.net> <kNko6.6986$eq3.707121@news.uswest.net> <MPG.150c46db9cec283e9896d4@news.earthlink.net> <L9Go6.2720$hZ4.875436@news.uswest.net> <DWCt6.12229$Im6.1309770@newsread1.prod.itd.earthlink.net> <jWEt6.7867$3_.998377@news.uswest.net>
Message-ID: <3AB77426.671D92F9@alum.mit.edu.->

"Randy A. Ynchausti" wrote:

RY> It use to be that people said you should never implement
RY> a compiler using OO because for all of the "obvious" reasons.

Who said that?  What were the stated reasons?  References?

Pardon me, but this sounds like a strawman to me.

Tak
-----------------------------------------------------------+----------
Tak To                                            takto at alum.mit.edu.-
--------------------------------------------------------------------^^
 [taode takto ~{LU5B~}]      NB: trim the .- to get my real email addr


From sholden at holdenweb.com  Tue Mar  6 19:08:53 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Tue, 6 Mar 2001 19:08:53 -0500
Subject: Can Python replace Visual Basic? Should it?
References: <5DZo6.11126$hn5.1578410@news1.rdc1.mb.home.com> <98266k0p5h@news1.newsguy.com> <3q8p6.12338$hn5.1798569@news1.rdc1.mb.home.com>
Message-ID: <T1fp6.23934$ML1.1225519@e420r-atl2.usenetserver.com>

"Brad Bollenbach" <bbollenbach at homenospam.com> wrote in message
news:3q8p6.12338$hn5.1798569 at news1.rdc1.mb.home.com...
>
> "Alex Martelli" <aleaxit at yahoo.com> wrote in message
> news:98266k0p5h at news1.newsguy.com...
> > "Brad Bollenbach" <bbollenbach at homenospam.com> wrote in message
> > news:5DZo6.11126$hn5.1578410 at news1.rdc1.mb.home.com...
> > > Again, I'm much more interested in the real world experience of those
> who
> > > used to use the big visual development tools in Windows like VB,
Delphi
> or
> > > Powerbuilder rather than theoretical answers.
> >
> > Why do you focus on 'used to use'?  One may perfectly well keep using
> > some VB (&c) for tasks it's particularly well-suited to, while using
> > Python wherever feasible (and, indeed, using both in the same project
>
> Certainly. And for all intents and purposes "used to use" isn't necessary,
> but the question happened to be directed (as its subject suggests) at
those
> who _have_ replaced VB with Python. :)
>
> But realistically, I just wanted to emphasize my interest in hearing
answers
> from those who have or currently are using VB or one of the other Win32
> visual development tools. Personally, I'm a former VB programmer now
working
> in Powerbuilder and am unhappy with both, again based on my dislike of
IDE's
> that obscure the programmer's view of what's going on.
>
> I could explain _why_ I don't like the 40,000 foot views of IDE's, but
> that's another thread.
>
> Brad
>
Well, I have both used and taught PowerBuilder.  I built a contact manager
system using it. It's good for creating GUIs, and embedded SQL is
beautifully easy. However, PowerScript is an abortion, with a half-assed
implementation of object orientation and decrepit implementations of many
functions. I hate not knowing whether a function is returning zero because
that's the result, or because an error was caught and translated into a
sentinel value.

Having said all this, I haven't yet found a tool as handy as VB for
generating GUIs, so I regards PowerBuilder as a lost cause. Not doing much
VB, come to that.

Python GUI construction is becoming easier by the day, tools like BlackAdder
and WingIDE will be commonplace within a year. So for now I'm not expending
too much effort ont he GUI side, as I'm not a Tkinter afficionado (I can use
it, but I'm sure Tk3k will be much better when it's finished, and would like
to see a GUI builder for that toolkit).

Don't know whether any of thishelps, just my 2c.

regards
 Steve




From mwh21 at cam.ac.uk  Tue Mar 20 09:30:40 2001
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 20 Mar 2001 14:30:40 +0000
Subject: Python style questions
References: <98tb91$9t9$1@saltmine.radix.net> <m34rwpilxg.fsf@atrus.jesus.cam.ac.uk> <995o0u$itc$1@news.mathworks.com> <0oodbtgdgc3tv452iiqs29n2rd3534dc2k@4ax.com>
Message-ID: <m37l1kh5rj.fsf@atrus.jesus.cam.ac.uk>

Tim Roberts <timr at probo.com> writes:

> Joshua Marshall <jmarshal at mathworks.com> wrote:
> >
> >A switch statement may just be syntactic sugar, but not necessarily in
> >a simple way.  A good compiler will compile switch statements into
> >jump-tables or, at worst, binary searches.  Cluttery to code by hand.
> >
> >Sure this is "just" an efficiency thing, but switches can get quite big.
> 
> But Python, at least today, is not compiled.

Yes it is (at least in this context).

> Switch statements in an interpreted language are not as big of a
> "win".

There's no reason that the bytecode compiler couldn't do binary search
jiggery-pokery, although if I was doing it, I'd probably use a
dictionary.

Cheers,
M.

-- 
  MGM will not get your whites whiter or your colors brighter.
  It will, however, sit there and look spiffy while sucking down
  a major honking wad of RAM.              -- http://www.xiph.org/mgm/


From woetzel at gmd.de  Fri Mar 23 11:08:30 2001
From: woetzel at gmd.de (Gerd Woetzel)
Date: 23 Mar 2001 16:08:30 GMT
Subject: Is setup.py a "good thing"? (long)
References: <3aba76e2.167083508@24.0.228.33> <3dvgp1zasw.fsf@ute.cnri.reston.va.us>
Message-ID: <99fsdu$p2u$1@hop.gmd.de>

Andrew Kuchling <akuchlin at mems-exchange.org> writes:

>If a module is enabled in Modules/Setup, then setup.py won't try to
>build it, so you could just enable Tkinter there with whatever options
>are required.

Hmm, maybe that's true for Tkinter. With socket an zlib I had some
Problems on SunOS 5.8.

I need to load libssl.so and libz.so from a special place.
Hence I used Modules/Setup. Make went well, all tests passed.

However after installing, python2.1 did not load these libs anymore :(.

What happened?

Modules/Setup has buit zlibmodule.so and socketmodule.so which worked
fine during the test phase, but setup.py has (secretly?) created
zlib.so (socket.so) too! Finally "make install" has installed all of them.

Of course, python selcts between xxx.so and xxxmodule.so according to
murphey's law <oops-wink>. 

Yes, it was easy to remove the wrong libs from lib/python21 in the end.
However, I was quite confused for some time. 
--
Gerd Woetzel            | email: gerd.woetzel at gmd.de
GMD FIT.CSCW            | phone: ++49 2241 142648 (fax: 142084)
D-53754 Sankt Augustin  | www:   http://orgwis.gmd.de/~woetzel/


From ws-news at gmx.at  Tue Mar  6 08:58:21 2001
From: ws-news at gmx.at (Werner Schiendl)
Date: Tue, 6 Mar 2001 14:58:21 +0100
Subject: Can anyone recomend a good intoduction to C...
References: <3aa366c6_2@nnrp1.news.uk.psi.net>
Message-ID: <983887378.363030@newsmaster-04.atnet.at>

If you have a good reason to choose C over C++, take the K&R book the others
recommend.

Otherwise you should consider going straight to C++.
This will save you hours of headache with C memory stuff, string handling,
and so on.

To get an good introduction to C++ for an programmer, I would recommend to
have a look at
http://www.bruceeckel.com/

and look up his free electronic book "Thinking in C++"
the printed edition is accompanied with a CD rom that teaches C (the subset
used with C++) from ground up
but there is also a chapter in the online book that does a good job if you
are well familiar with Python

hth
Werner

p. s. forget anything about that tells you C++ is too difficult to get in
one step... its wrong ;-)


Simon Brunning <sbrunning at trisystems.co.uk> wrote in message
news:3aa366c6_2 at nnrp1.news.uk.psi.net...
> ... for a Python programmer?
>
> I'm going to have to bite the bullet and learn a bit of C. I'd at
> least like to learn the *right* way to do it.
>
> --
> Cheers,
> Simon Brunning
> sbrunning at bigfoot.com
>
>
>
>




From aleaxit at yahoo.com  Thu Mar  1 04:05:48 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 1 Mar 2001 10:05:48 +0100
Subject: "import X" and "from X import y" produce different results, Why?
References: <ifmn6.6635$hn5.857862@news1.rdc1.mb.home.com>
Message-ID: <97l3e8027ho@news2.newsguy.com>

"Parzival Herzog" <parz at home.com> wrote in message
news:ifmn6.6635$hn5.857862 at news1.rdc1.mb.home.com...
> In the Essential Python Reference, it states (p. 72):
>
> "The from statement is identical to import, except that instead of
creating
> a name referring to the newly created module namespace, references to one
or
> more of the objects defined in the module are placed into the current
> namespace."

Right.  I.e., it's identical, except that it isn't:-).

> # yy.py
> from xx import X
> import xx

Note that this *PAIR* of statements IS identical in effect to the following:

import xx
X = xx.X

Is this semantical identity clear to you...?  We're doing two bindings in
the current module's namespace: we bind a name 'xx' to a module object
(the same as sys.modules['xx'], though we'd need an import sys to check);
also, we bind a name 'X' (again, in the current module's namespace) to
whatever object name 'X' was bound to in the xx module's namespace.

All right so far?

> print "BEFORE, X =", X
> print "BEFORE, xx.X =", xx.X

And you might also say
    print "Before, are they the same...?", X is xx.X
and get a nice 1 result from the 'is' operator (which does not prove
much when the object being referred to is immutable -- specifically,
a small integer -- but you could try with more 'elaborate' objects and
still pass the 'is' test; X and xx.X *MUST* refer to the SAME object
here, whatever that object might be).

Still OK...?

> X = 99

This *RE-BINDS* name X (in the current namespace) to something else.
This, *as usual*, does *NOT* affect any other binding that might be
in effect to the object that this name was bound to before the
re-binding.

Side-note with explanatory purposes, which you may want to skip
except for the first 6 lines (the rest are for completeness)...:

Binding affects names; specifically, it (only!) changes what object
that name refers to (until further notice, e.g. another rebinding of
the same name).  It does *NOT* affect the object being bound to; in
fact, the object need not even be aware of how it's being referred to,
in one or more ways, from whatever number of places -- bindings,
rebindings, unbindings, names, slots in lists, entries in dictionaries
(which is what names basically boil down to), whatever  -- _except that _:
  if the _number_ of such bindings to the object falls to zero, the object
  may be destroyed (it's immediately destroyed when that happens, in the
  current version of CPython, but that's an implementation artefact --
  Jython, and possibly future CPython's [and other implementations yet,
  such as Python.NET], are more 'relaxed' about such time-of-destruction
  issues; also, it's not *necessary* for the number of bindings to fall
  to zero in order for object-destruction to be triggered -- Jython and
  .NET, and under some conditions CPython 2.0 and higher, are also able
  to destroy objects caught in circular-reference loops, whose numbers of
  bindings never falls to zero but which still become 'unreachable').

'import' has very little to do with what is happening here.  Through
*whatever* mechanism a name was bound to a value, re-binding the name
does not affect the value, nor does it affect other bindings to the
same value.


Alex





From mrq at for.mat.bham.ac.uk  Mon Mar 19 12:39:10 2001
From: mrq at for.mat.bham.ac.uk (Martyn Quick)
Date: Mon, 19 Mar 2001 17:39:10 +0000
Subject: The most important modules
Message-ID: <Pine.GSO.4.21.0103191732210.22234-100000@laplace>

I've seen a few comments along the lines of "When you are a newbie it will
take a while to realize which the important modules are."  As someone who
is in the middle of wading through Chapter 3 of the Library Reference, it
would be very helpful to know which ones are particularly worth
noting.  (Presumably  sys, (c)Pickle and shelve, from the ones I've so far
in the chapter - but what other treats are out there for me?)

As a general comment along these lines, I notice that as I become aware of
more of Python's capabilities, I find that previous bits of code can be
greatly simplified.  Presumably this is a common experience?

Cheers,
Martyn
(becoming less of a newbie day-by-day)

P.S. Thanks to everyone (particularly Cameron Laird) who helped me out
earlier.  I unexpectedly came into enough money to buy a couple of books
on Python - Lutz & Ascher's "Learning Python" and Grayson's "Python and
Tkinter Programming".  Both are proving to be very useful - I especially
found that having been tinkering with Python for a few weeks before I
started reading, Lutz & Ascher managed to tell me almost exactly the
things I wanted to know.  Whenever a question jumped into my head, they
seemed to answer it in the next paragraph.

--------------------------------------------------------
Dr. Martyn Quick  (Research Fellow in Pure Mathematics)
University of Birmingham, Edgbaston, Birmingham, UK.
http://www.mat.bham.ac.uk/M.R.Quick



From aaron.ginn at motorola.com  Mon Mar 19 12:48:13 2001
From: aaron.ginn at motorola.com (Aaron Ginn)
Date: 19 Mar 2001 10:48:13 -0700
Subject: Evaluating the contents of a string
Message-ID: <snitl5r6oy.fsf@motorola.com>

Is there a simple way to evaluate the contents of a string assuming
that string contains only nunbers and mathematical operators?

For example, I'm parsing a file that contains expressions contained in 
parantheses such the following:

12 net64 a vss vss n w=1.1u l=0.5u ad=(1.1 * 1.15) as=(1.1 * 1.15)
                                      ^^^^^^^^^^^^    ^^^^^^^^^^^^

I'm using a stack to determine when an expression can be evaluated
(i.e., when a closing paranthesis is found), and I'm storing the
contents of the parantheses in a string.  When the first closing
parenthesis is found in the example above, the string to be evaluated
will be '1.1 * 1.15'.  Is there a module that will mathematically
evaluate this string?

Thanks,
Aaron


-- 
Aaron J. Ginn                    Phone: 480-814-4463
Motorola SemiCustom Solutions    Pager: 877-586-2318
1300 N. Alma School Rd.          Fax  : 480-814-4463
Chandler, AZ 85226 M/D CH260     mailto:aaron.ginn at motorola.com


From Eugene.Leitl at lrz.uni-muenchen.de  Mon Mar 26 16:46:51 2001
From: Eugene.Leitl at lrz.uni-muenchen.de (Eugene.Leitl at lrz.uni-muenchen.de)
Date: Mon, 26 Mar 2001 23:46:51 +0200
Subject: Python MPI
References: <mailman.985354471.4871.python-list@python.org> <3u0g99.vn3.ln@dollerup.csn.tu-chemnitz.de>
Message-ID: <3ABFB8CB.28705FD3@lrz.uni-muenchen.de>

Karsten Petersen wrote:
>
> Well, that only describes how to use Python with C-modules using MPI.

But that covers 95% of what I need, so it's plenty.

> I found it even more interesting to call MPI functions directly from
> Python - thats possible with e.g. the MPI-module contained in

I'm going to use this for realtime visualization, so there's no 
way I can call a bunch of Python code if I'm to expect ~10 fps
refresh, where each frame takes many MBytes of memory to stream
through.

> ScientificPython 2.1.
>         http://starship.python.net/crew/hinsen/scientific.html
> The above link also describes, that you have to patch your Python
> executable.  But my tests with lam-mpi and mpich had shown that it is
> not really needed - they finish themself cleanly even if MPI_Finalize is
> not called.

Good to know.
 
> > Any MPI Python users here? I'll be interested in hearing your experiences.
> I used Python with the MPI-Module from Scientific-Python 2.1 during the
> last months to do my MPI-coding homework.  Currently I am trying to
> write a parallizing fractal-generator in Python using MPI, NumPy and
> PyGTK...

I'm going to write a cellular automata MPI package around Rolf Rabenseifner's
heat transfer MPI demo, and the VolPack voxel rendering library, to run on
my 8-node K7 Beowulf (I guess DDR memory, since this is about memory stream
voxels).



From michael at stroeder.com  Sun Mar  4 11:29:18 2001
From: michael at stroeder.com (Michael =?iso-8859-1?Q?Str=F6der?=)
Date: Sun, 04 Mar 2001 17:29:18 +0100
Subject: Concurrency and shelve
Message-ID: <3AA26D5E.2FD28B40@stroeder.com>

HI!

Can anybody summarize issues with shelve (based on anydbm) and
concurrent access? I experienced with bsddb that changes to the DB
within one instance are only visible after calling close() to a
second instance of the same database.

Ciao, Michael.


From matt at mondoinfo.com  Sat Mar 10 17:18:42 2001
From: matt at mondoinfo.com (Matthew Dixon Cowles)
Date: Sat, 10 Mar 2001 22:18:42 GMT
Subject: Tkinter focus_get()
References: <ziwq6.18374$tV.3430599@newsrump.sjc.telocity.net>
Message-ID: <slrn9ala21.mp.matt@happy-hour.mondoinfo.com>

On Sat, 10 Mar 2001 14:39:57 -0600, dsavitsk <dsavitsk at e-coli.net>
wrote:

>I have a grid (actually a dictionary of dictionaries) of entry
>widgets that are dynamically created.  Each has <KeyPress> bound to
>it.  I am trying to determine which one has focus, so i used

>self.root.focus_get()

>which returns something that looks like

>.8854796.8898804.18282492.18282532.18283228.18248316.17887444.17854708

>is there a way to interpret this? or a better way to get focus?

That's the actual Tk widget, it just prints as a string. You can
compare it with the widgets you've created:

def reportUp(self,event):
  w=self.tkRoot.focus_get()
  print w
  print type(w)
  if w is self.entry1:
    print "It's e1"
  elif w is self.entry2:
    print "It's e2"

But you probably don't need the focus_get() call because the widget
attribute of the event object ought to be the same:

def reportUp(self,event):
  if event.widget is self.e1:
    print "It's e1"
  elif event.widget is self.e2:
    print "It's e2"
    
An exhaustive search like this simple and very practical if the number
of widgets is small. If it grows large, one practical thing to do is
to save a reference to the widget in a callable object at the time the
widget is created:

class reportUp:
  def __init__(self,widget):
    self.w=widget
    return None

  def __call__(self,eventIgnored):
    self.w.insert(0,"Me") # Or something more useful
    return None

And when widgets are being created:

  self.e1=Entry(self.tkRoot,text="one")
  self.e1.pack(side=TOP)
  self.e1.bind('<KeyPress-Up>',reportUp(self.e1))

Regards,
Matt


From claird at starbase.neosoft.com  Fri Mar 16 08:39:05 2001
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 16 Mar 2001 07:39:05 -0600
Subject: Basic information requested
References: <MHks6.144$lC5.52172@news2-win.server.ntlworld.com> <98snjh$hsu$1@norfair.nerim.net> <1Mls6.207$lC5.83390@news2-win.server.ntlworld.com>
Message-ID: <8577DF580FF89E5F.C22924F4413747C2.9E9B3A61B5DE0343@lp.airnews.net>

In article <1Mls6.207$lC5.83390 at news2-win.server.ntlworld.com>,
Jamie MacIsaac <jamie at cctech.co.uk> wrote:
>
>Hi,
>
>Thanks for the suggestion. I've been looking for that title in the bookshops
>around here (Glasgow, Scotland), but I can't find it. I'm somewhat loathe to
>order it online without actually knowing that it will cover what I'm after -
>I've done that too often.
>
>Do you have 'Python Programming on Win32'? Can you (or anyone else) verify
>that it deals with mapping and disconnecting remote shares?
It does.

I'm as wary of spending money, particularly on
things unseen, as any Glaswegian (notice my name).
I can assure you that this is an opportunity you'll
be glad you exercised.  *Python Programming on
Win32* is a superb resource for folks in your situ-
ation.

As are the archives of this newsgroup, and other
collections mentioned by other posters.
			.
			.
			.
-- 

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


From kapblp at bellsouth.nOt  Sat Mar 31 14:33:06 2001
From: kapblp at bellsouth.nOt (KEVIN)
Date: Sat, 31 Mar 2001 14:33:06 -0500
Subject: I want to learn PYTHON!
Message-ID: <1iqx6.3744$aD4.234302@news2.atl>

Where do I start.

I am not a programmer but have an intermediate level of computer knowledge
in Windows only.

--
KEVIN

"Si hoc legere scis, nimis eruditionis habes."
-----------------------------------------------------------------
Please repond in the newsgroup.
My email address is altered for anti-spam
Just change .nOt to .net




From smigowsky at dspace.de  Tue Mar 13 03:24:40 2001
From: smigowsky at dspace.de (Stefan Migowsky)
Date: Tue, 13 Mar 2001 09:24:40 +0100
Subject: PyEval_CallObject and threads
Message-ID: <84257D042AA7D411B3F700D0B7DB9B7C13A99A@newsintern.dspace.de>


> [Phlip]
>
> Tim Peters wrote:
> 
> > [Phlip]
> >> My colleague came over and asked me to write this:
> >>
> >> "We got multiple C++ pthreads calling PyEval_CallObject. 
> How threadsafe
> >> is the Python API? A C++ module using Boost BPL creates multiple
threads,
> >> which then call back into Python. When two threads simultaneously call
> >> PyEval_CallObject we go boom."
> >>
> >> I will now go examine his serialization primitives. But if anyone has
any
> >> better ideas...
> > 
> > With very few exceptions, all calls to Python C API functions must be
made
> > while holding the global interpreter lock.  The exceptions have mostly
to
> > do with a handful of functions related to initialization and shutdown.
See
> > the section "Thread State and the Global Interpreter Lock" in the
Python/C
> > API Reference Manual for details.
> 
> Further inspection reveals our architect(s) think we need to call Python 
> script asynchronously from multiple threads. By 'lock' we assume both you 
> and the section you cite mean 'block'. Isn't this what Stackless Python is

> for?

We had a similar problem having both : threads started with Python commands
and threads started and controlled by the C-runtime. Since we couldn't be
sure if a threadstate is always set( a NULL Threadstate would end up in a 
PyFatalError if calling PyEvalSaveThread) we emulated the creation of a
Python
thread in the C-Runtime thread. For this you have to store the
Interpreterstate
and the use some code like this: 

// This code is running is a thread spawned by the C-program.
....
if(NULL != g_ppyinstInterp)  // Check global interpreterstate
{
    return;
}
if(NULL != g_ppyoPyThreadFunc ) // Check the function to be called
{
    PyThreadState *ppythstState  = NULL;
    PyObject      *ppyobRes      = NULL;
    PyObject      *ppyobArgs     = NULL;
    PyObject      *ppyobFirstArg = NULL;

    // 1. Set up the args
    ppyobFirstArg = GetFirstArg();
    ppyobArgs     = PyTuple_New(1);
    if(NULL == ppyobArgs )
    {
        Py_XDECREF(ppyobFirstArg);
        return;
    }
    PyTuple_SetItem(ppyobArgs, 0, ppyobFirstArg);
    // 2. Create a new threadstate for the current thread. This thread 
    // state is needed to insert the current thread into the Python 
    // interpreter.
    ppythstState  = PyThreadState_New(g_ppyinstInterp);
 
    // 3. Acquire Thread
    PyEval_AcquireThread(ppythstState);

    // 4. Call the function
    ppyobRes = PyEval_CallObjectWithKeywords(g_ppyoPyThreadFunc, ppyobArgs,
NULL);

    Py_DECREF(ppyobArgs);

    // 5. check result of callback function
    if (res == NULL) 
    {
        if (PyErr_ExceptionMatches(PyExc_SystemExit))
        {
	     PyErr_Clear();
        }
    	  else 
        {
	    	fprintf(stderr, "Unhandled exception in thread:\n");
	      PyErr_PrintEx(0);
	  }
    }
    else
    {
        Py_DECREF(res);
    }

    // 6. Cleanup
    PyThreadState_Clear(ppythstState);
    PyEval_ReleaseThread(ppythstState);
    PyThreadState_Delete(ppythstState);
}

   Stefan



From tim.one at home.com  Mon Mar 12 18:37:14 2001
From: tim.one at home.com (Tim Peters)
Date: Mon, 12 Mar 2001 18:37:14 -0500
Subject: Help with Regular Expressions
In-Reply-To: <3AAD524E.746338DE@javanet.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCAEDDJFAA.tim.one@home.com>

[Raymond Hettinger]
> Is there an idiom for how to use regular expressions for lexing?
>
> My attempt below is unsatisfactory because it has to filter the
> entire match group dictionary to find-out which token caused
> the match. This approach isn't scalable because every token
> match will require a loop over all possible token types.
>
> I've fiddled with this one for hours and can't seem to find a
> direct way get a group dictionary that contains only matches.

That's because there isn't a direct way; best you can do now is seek to order
your alternatives most-likely first (which is a good idea anyway, given the
way the engine works).

If you peek inside sre.py (2.0 or later), you'll find an undocumented class
Scanner that uses the undocumented .lastindex attribute of match objects.
Someday I hope this will be the basis for solving exactly the problem you're
facing.  There's also an undocumented .lastgroup attribute:

Python 2.1b1 (#11, Mar  2 2001, 11:23:29) [MSC 32 bit (Intel)] on win32
Type "copyright", "credits" or "license" for more information.
IDLE 0.6 -- press F1 for help
>>> import re
>>> pat = re.compile(r"(?P<a>aa)|(?P<b>bb)")
>>> m = pat.search("baab")
>>> m.lastindex  # numeral of group that matched
1
>>> m.lastgroup  # name of group that matched
'a'
>>> m = pat.search("ababba")
>>> m.lastindex
2
>>> m.lastgroup
'b'
>>>

They're not documented yet because we're not yet sure whether we want to make
them permanent parts of the language.  So feel free to play, but don't count
on them staying around forever.  If you like them, drop a note to the effbot
saying so.

for-more-docs-read-the-source-code-ly y'rs  - tim




From spike at oztourer.nl  Wed Mar 21 04:45:08 2001
From: spike at oztourer.nl (Steve Pike)
Date: Wed, 21 Mar 2001 09:45:08 GMT
Subject: Problem posting an app to Vaults of Parnassus
Message-ID: <3ab87668.4593002@news.cablewanadoo.nl>

I've tried several times to add a Python application to the Vaults of
Parnassus at http://www.vex.net/parnassus/, and each time it seems to
complete successfully, but the app never actually appears on the site.
I've sent a query via the site's feedback page, but got no response
there either.

Does anyone have contact information for the site's webmaster, or any
ideas what the problem might be?

Thanks,
Steve


From paul.robinson at businesscollaborator.com  Thu Mar 22 12:55:52 2001
From: paul.robinson at businesscollaborator.com (Paul Robinson)
Date: Thu, 22 Mar 2001 17:55:52 +0000
Subject: Help with creating Python COM
References: <Dwqu6.79766$Ch.10654381@typhoon.we.rr.com>
Message-ID: <3ABA3CA8.A53F1623@businesscollaborator.com>

Charlotte Williams wrote:
> 
> I am following a tutorial for creating a Python COM server that will
> interact with a Visual Basic client. 

> ImportError: No module named ni
> 
> All the tutorials I have encountered tells me that these are the modules
> that must be imported:
> 
>  import ni
>  from win32com.server.exception import Exception
>  import win32com.server.util

I think the first line of the tutorial sums it up nicely:
"Last Updated 4 June 1997 9:00 a.m."

The ni module has not been used for a while now. I beleive it's use was
required pre 1.5 to enable package importing....or something like that.

I would reccommend find a more upto date tutorial on Python and COM.
e.g. the docs that come with Pythonwin, Mark Hammonds book - "Python
Programming on win32" - ISBN: 1565926218, Mark's starship page also has
links to sample code tutorials etc:
http://starship.python.net/crew/mhammond/

Hope that helps.

	Paul.

Paul Robinson
Business Collaborator Development Manager
Enviros Software Solutions
WWW: http://www.businesscollaborator.com



From root at rainerdeyke.com  Wed Mar  7 21:07:05 2001
From: root at rainerdeyke.com (Rainer Deyke)
Date: Thu, 08 Mar 2001 02:07:05 GMT
Subject: Automagic type conversions, was: Re: I come to praise .join, not to bury it...
References: <mailman.983485923.1136.python-list@python.org> <97p7iq$kk2$1@nntp6.u.washington.edu> <97rka002k9@news1.newsguy.com> <3AA2F4B3.A7496E76@cosc.canterbury.ac.nz> <97vmoh061p@news2.newsguy.com> <3AA46134.D385208B@cosc.canterbury.ac.nz> <98292h0s0g@news1.newsguy.com> <3AA58BD0.3324C709@cosc.canterbury.ac.nz> <9850ge013ns@news1.newsguy.com> <gRup6.364652$ge4.128679630@news2.rdc2.tx.home.com> <qZvp6.29197$ML1.1439945@e420r-atl2.usenetserver.com>
Message-ID: <dPBp6.367558$ge4.129410307@news2.rdc2.tx.home.com>

"Steve Holden" <sholden at holdenweb.com> wrote in message
news:qZvp6.29197$ML1.1439945 at e420r-atl2.usenetserver.com...
> "Rainer Deyke" <root at rainerdeyke.com> wrote in message
> > I'm curious: how do you feel about the automagic conversions of
functions
> > (and only true function, not function-like objects) into unbound
methods,
> > and from unbound methods into bound methods?  I'm thinking that the
former
> > could be eliminated entirely, and the latter replaced by a __bind__
magic
> > method that allows the mechanism to apply to function-like objects.
> >
> >
> Surely no type conversion is involved: the interpreter just converts
>
>     instance.method(*args)
>
> into
>
>     class.method(instance, *args)
>

>>> class X: pass
...
>>> def f(self): pass
...
>>> type(f)
<type 'function'>
>>> X.f = f
>>> type(X.f)
<type 'instance method'>
>>> x = X()
>>> x.f is X.f
0

As of 2.0, 'x.f' and 'X.f' have the same type but are different objects.
This is misleading, and I prefer the 1.5.2 behavior where the types of 'x.f'
and 'X.f' were different.  In either version, two automagic conversions took
place, both based on hardwired type.  Consider:


>>> class Y:
...   class Z:
...     def __init__(self, outerself):
...        print outerself
...
>>> y = Y()
>>> Y.Z(y)
<__main__.Y instance at 0077221C>
<__main__.Z instance at 00773BCC>
>>> y.Z()
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
TypeError: not enough arguments; expected 2, got 1


--
Rainer Deyke (root at rainerdeyke.com)
Shareware computer games           -           http://rainerdeyke.com
"In ihren Reihen zu stehen heisst unter Feinden zu kaempfen" - Abigor




From rdsteph at earthlink.net  Sun Mar 25 11:59:46 2001
From: rdsteph at earthlink.net (Ron Stephens)
Date: Sun, 25 Mar 2001 16:59:46 GMT
Subject: Newbie needs help to find bug in his first Perl program
Message-ID: <3ABCC757.6C6F17CA@earthlink.net>

Ok, I am a total newbie. I took some programming course in college, did
some relatively simple programs in Fortran, Basic, PL/1 and even one in
assembler for an old IBM "mainframe" using punched cards. But that was
along time ago.

OK, I have just fallen head over heels in love with Python. Wrote my
first little "program", about 200 lines of a thing to do simple
"Decision Analysis" in a general sort of way, to take input from an
"expert" in any given (simple) field, and set up criteria based on the
experts experience, then calculate a weighted average to decide which
option to choose.

I find the debugging to be extraordinarily good, but I have one bug I
cant seem to figure out yet. Been up all night its noon of the next day
;-))) Need some sleep

The offending line of code is

sixtwo = raw_input ("where does %s rank for %y ?" % (option6,
criteria2))

Can anyone see what is wrong with this? I have several lines like this
and they all give me error messages.

Sorry for the trivial newbie nature of my question and first post ;-)))

But I do luv Python....



From loewis at informatik.hu-berlin.de  Fri Mar  2 07:27:37 2001
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 02 Mar 2001 13:27:37 +0100
Subject: A counter-proposal to __future__ in PEP 236
References: <j47l29lpo8.fsf@informatik.hu-berlin.de> <97m2p5$3au$2@news.mathworks.com>
Message-ID: <j41ysgl5ja.fsf@informatik.hu-berlin.de>

Joshua Marshall <jmarshal at mathworks.com> writes:

> > In this implementation, directive is only considered as a keyword if
> > it appears at the beginning of the module. Therefore, code that
> > happens to use directive as an identifier will not break (unless it
> > assigns to directive as the first thing in the module).
> 
> Note that it doesn't need to be implemented this way.  Currently,
> 
>   directive foo
> 
> is a syntax error.  Meaning can be assigned to this syntax without
> breaking any existing code--it is unambiguous.

Yes, but that is harder to implement - it requires a look-ahead for
the next token. Traditionally, Python has the notion of 'reserved
words' (aka keywords). 'directive' in above statement would be a
keyword - if it was a name, it won't parse. 

Allowing a 'name name atom' production probably introduces a
grammatical ambiguity for the simple recursive-descent parser used in
Python. Classifying directive as a keyword only if it is followed by a
name is more than the classify function can do. If it is considered
necessary to break absolutely no code, I could look into implementing
it that way.

Regards,
Martin


From brendhanhorne at bellsouth.net  Sat Mar 17 20:13:35 2001
From: brendhanhorne at bellsouth.net (Brendhan Horne)
Date: Sat, 17 Mar 2001 20:13:35 -0500
Subject: Interactive startup file
References: <3dTs6.2144$ge.55105@news3.mia> <mailman.984876862.21693.python-list@python.org>
Message-ID: <cZTs6.2148$ge.55348@news3.mia>

> Well, it's exactly what it says.
>
> If you set environment variable

Go slow here, What is an environment variable?


>     PYTHONSTARTUP='.pythonrc'
> then commands in file '.pythonrc' will be executed everytime you run
> python in interactive mode (ie. when you get '>>>' prompt).  For
> example, mine contains
>     import rlcompleter
>     rlcompleter.readline.parse_and_bind('tab: complete')
It sounds kinda like a program that runs everytime you start the command
interpretter>>>>  Am I correct?
>
> If you want to run those commands from a script, then you have to call
> the file explicitly, using
>     execfile('.pythonrc')
If I am at the command interpretter>>>> I can open a file from there by
typing:
>>> execfile('.filenamehere')    and it would open the file?

--
Thanks,
Brendhan




From rkiendl at gmx.net  Tue Mar 27 10:31:46 2001
From: rkiendl at gmx.net (Robert Kiendl)
Date: Tue, 27 Mar 2001 17:31:46 +0200
Subject: dynamic web pages?
References: <99q701$jlr10@news-1.bank.dresdner.net>
Message-ID: <3AC0B262.6F0E30DE@gmx.net>

we have the BIG GUN here: Zope (www.zope.org) dynamic app.server in
native python logic including an OODBMS!

or use cgi programming with apache 

/R

"Dr. Peter Dintelmann" wrote:
> 
> Hi,
> 
> is there any module to create dynamic web pages
> using Python?
> 
> I am looking for something like Perl's Mason (www.masonhq.com)
> (or PHP). The business logic of my web app should be implemented
> in Python classes which are accessed through a Python/HTML
> mixture.
> 
> Thanks in advance,
> 
>     Peter


From ws-news at gmx.at  Tue Mar 13 16:05:27 2001
From: ws-news at gmx.at (Werner Schiendl)
Date: Tue, 13 Mar 2001 22:05:27 +0100
Subject: learning C
References: <mailman.983972829.23481.python-list@python.org> <983975751.18833@newsmaster-04.atnet.at> <slrn9adklr.bv9.scarblac@pino.selwerd.nl> <984047775.592522@newsmaster-04.atnet.at> <9896hk$ig8$1@verence.demon.co.uk> <984134612.781779@newsmaster-04.atnet.at> <3AACE22D.3A9EEB12@bt.com> <j6kr6.68748$ML1.3282888@e420r-atl2.usenetserver.com> <DYsr6.1213$Tg.160308@ruti.visi.com>
Message-ID: <984517813.15780@newsmaster-04.atnet.at>

> Yes, but not much.  C is little more than a portable assembly
> language.

So I think even plain C is a large improvement over assembler.
Of course you know what is going on, but doing multilevel pointer and array
dereferencing in asm is pretty awkward and errorprone in assembler.
Structured programming is also not supported (but of course possible) by
assembler.

regards
werner




From whisper at oz.net  Sun Mar 11 18:27:00 2001
From: whisper at oz.net (Dave LeBlanc)
Date: 11 Mar 2001 23:27:00 GMT
Subject: Questions about no more nested imports
References: <98gi4l$n9n$0@216.39.170.247> <m3n1as850y.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <98h1k4$ai4$0@216.39.170.247>

Ok, read the PEP (more or less again). It's not all that clear, but
does answer my original questions.

Could someone please clarify the scoping wrt to classes? I would
expect that the following code would refer to the variable at the
class scope, but it's not clear to me if it would do so:

class foo
	bar = 1
	def fee
		print bar

Dave LeBlanc

On 11 Mar 2001 19:44:13 +0000, Michael Hudson <mwh21 at cam.ac.uk> wrote:


>Read the PEP:
>
>  http://python.sourceforge.net/peps/pep-0227.html
>
>(specifically the section "Backwards compatibility").
>
>Cheers,
>M.
>



From seandc at att.net  Wed Mar 14 21:00:52 2001
From: seandc at att.net (Sean)
Date: Thu, 15 Mar 2001 02:00:52 GMT
Subject: Spreadsheet-like array in Tkinter??
Message-ID: <3AB021B0.3F3DCA1F@att.net>

I'm writing a database browser which I'd
like to look like the FoxPro "browse" 
window, or like the body of a spreadsheet.
Each row would display a record (tuple).

Has anyone seen sample code in Tkinter
or Python Megawidgets which might offer 
a place to start?

Thanks.

Sean
seandc at att.net


From _gragra at nomore.spam.com  Sun Mar 25 21:55:49 2001
From: _gragra at nomore.spam.com (GraGra)
Date: Mon, 26 Mar 2001 02:55:49 GMT
Subject: Python script to create HTML gallery, thumbnails
References: <3ABCAC94.B095CB9A@ozemail.com.au>
Message-ID: <3abeae5c.18990177@news.itmel.bhp.com.au>

Thumbs Plus is shareware from http://www.cerious.com/
and does a great job managing images and creating thumbnails and
linked html pages.

GraGra




From shaleh at valinux.com  Thu Mar 15 17:27:04 2001
From: shaleh at valinux.com (Sean 'Shaleh' Perry)
Date: Thu, 15 Mar 2001 14:27:04 -0800 (PST)
Subject: Should I close after popen??
In-Reply-To: <m38zm6r8ih.fsf@master.athome>
Message-ID: <XFMail.20010315142704.shaleh@valinux.com>

On 15-Mar-2001 Andrew Markebo wrote:
> I am hacking away a small background program, using popen to call
> commands, should I close the file in some way when done or it is taken
> care of automagically??
> 
> I do something like this once every 10 mins..
> 
>         fd1=os.popen("command")
>         fd2=os.popen("command")
>         os.wait(fd1)
>         os.wait(fd2)
>         values=fd1.read()+fd2.read()
> 
> And.. anything to worry about, nesting popen/wait like this??
> (Linux/Solaris)
> 

if that is in a loop, it should be fine:

  while 1:
     f = open(file) # every time the loop repeats, f gets closed and opened
                    # again
     # do things to f



From cobrien at Radix.Net  Tue Mar 20 15:50:23 2001
From: cobrien at Radix.Net (Cary O'Brien)
Date: 20 Mar 2001 15:50:23 -0500
Subject: python suitability for large critical run forever apps
References: <mailman.985105416.28251.python-list@python.org>
Message-ID: <998fqf$6h$1@saltmine.radix.net>

In article <mailman.985105416.28251.python-list at python.org>,
Oleg Broytmann  <phd at mail2.phd.pp.ru> wrote:
>Hi!
>
>   I recommend you to look briefly into some projects, written in Python.
>Zope (www.zope.org) is well-know web application server. It is
>multithreaded server, that can run for months without any problem. Mailman

I can vouch for this also.  We have a (lightly loaded) zope server
that handles bug tracking with a PostgreSQL database.  As far as I can
tell it will run forever, at least it has never failed in the year and
a half we've had it up. The only thing keeping us from 3-figure
uptimes (100s of days) is the power.  I guess we should buy a ups, eh?

The major thing you need to watch for is memory leaks.  Run the thing
with load for a while and watch the thing with top or ps or something
to make sure the memory usage doesn't grow.

-- cary


From skip at mojam.com  Mon Mar  5 19:41:50 2001
From: skip at mojam.com (Skip Montanaro)
Date: Mon, 5 Mar 2001 18:41:50 -0600 (CST)
Subject: [Python-Dev] Re: Bug fix releases
In-Reply-To: <LNBBLJKPBEHFEDALKOLCEELHJDAA.tim.one@home.com>
References: <15009.17989.88203.844343@cj42289-a.reston1.va.home.com>
	<LNBBLJKPBEHFEDALKOLCEELHJDAA.tim.one@home.com>
Message-ID: <15012.12878.853762.563753@beluga.mojam.com>

    Tim> Note there was never a bugfix release for 1.5.2, despite that 1.5.2
    Tim> had some serious bugs, and that 1.5.2 was current for an
    Tim> unprecedentedly long time.  Guido put out a call for volunteers to
    Tim> produce a 1.5.2 bugfix release, but nobody responded.  Past is
    Tim> prelude ...

Yes, but 1.5.2 source was managed differently.  It was released while the
source was still "captive" to CNRI and the conversion to Sourceforge was
relatively speaking right before the 2.0 release and had the added
complication that it more-or-less coincided with the formation of
PythonLabs.  With the source tree where someone can easily branch it, I
think it's now feasible to create a bug fix branch and have someone
volunteer to manage additions to it (that is, be the filter that decides if
a code change is a bug fix or a new feature).

Skip



From tim at vegeta.ath.cx  Fri Mar 23 21:34:26 2001
From: tim at vegeta.ath.cx (Tim Hammerquist)
Date: Sat, 24 Mar 2001 02:34:26 GMT
Subject: python, perl, lisp, tcl / got a decision to make, can you give me a view point please ? and a joke for python users :)
References: <3aba3076.14395287@news.vif.com>
Message-ID: <slrn9bo25q.omf.tim@vegeta.ath.cx>

Yvon Boulianne <mysticsguy at yahoo.ca> wrote:
> i know a bit of perl and tcl, (i find
> perl too much cryptics, difficult to understand other people scripts)

Damn!  Not even a mystic can make sense of Perl.  =)

-- 
-Tim Hammerquist <timmy at cpan.org>
Disclaimer: If anyone disagrees with anything I
say, I am quite prepared to not only retract it,
but also to deny under oath I ever said it.
	-- T. Lehrer


From topmind at technologist.com  Thu Mar 22 01:53:15 2001
From: topmind at technologist.com (topmind)
Date: Thu, 22 Mar 2001 06:53:15 GMT
Subject: Collection interfaces
References: <3a90fa97.0@news1.mweb.co.za>   <MPG.150394195f7f04a19896a7@news.earthlink.net> <3A9A86B1.9A8C69A2@ix.netcom.com> <MPG.1504d14563dce6499896af@news.earthlink.net> <97fjla128ir@news2.newsguy.com> <MPG.1505a76831c34b749896b3@news.earthlink.net> <97h5s40vhc@news2.newsguy.com> <MPG.1506007591cf818b9896bc@news.earthlink.net> <slrn99opop.gr.amk@mira.erols.com> <MPG.1506265ebf6d934d9896be@news.earthlink.net> <97ibmc0k47@news1.newsguy.com> <MPG.1506cf3293b094b49896c1@news.earthlink.net> <97m0c7080c@news2.newsguy.com> <MPG.15085517debf4b649896ca@news.earthlink.net> <zb%n6.31033$o%6.586026@news1.rdc1.bc.home.com> <MPG.150b383cea226d409896d0@news.earthlink.net> <kNko6.6986$eq3.707121@news.uswest.net> <MPG.150c46db9cec283e9896d4@news.earthlink.net> <3AA2A898.450B659E@mail.com> <gRCt6.12224$Im6.1308560@newsread1.prod.itd.earthlink.net> <3AB755F8.67E84126@mail.com> <%a%t6.17245$227.1601569@newsread2.prod.itd.earthlink.net> <3AB8A730.4BE13D22@mail.com>
Message-ID: <vjhu6.18486$Im6.1943577@newsread1.prod.itd.earthlink.net>

"James A. Robertson" <jarober at mail.com> wrote in message
news:3AB8A730.4BE13D22 at mail.com...
> topmind wrote:
> >
>
> > > http://www.cincom.com/smalltalk
> > >
> >
> > I only see a front page, not a parsing example.
> >
>
> Follow the tutorial link - it's right there on that page.
>

I will try again on the weekend.

>
> > Besides, what are you comparing it to so say
> > it is better than a procedural/relational approach?
>
> It's vastly easier than similar code done in C or basic, for example
>

Hardly the best of bread you are comparing. (Oh oh, I keep offending
C fans even though they sometimes side with me. What to do.)

> >
> > > >
> > > > > [snip]
> > > >
> > > > -tmind-
> > > > oop.ismad.com
> > >
> > > --
> > > James A. Robertson
> > > Product Manager (Smalltalk), Cincom
> > > jarober at mail.com
> > > <Talk Small and Carry a Big Class Library>
> > >
>
> --
> James A. Robertson
> Product Manager (Smalltalk), Cincom
> jarober at mail.com
> <Talk Small and Carry a Big Class Library>




From ^NOSPAM^ifi at ifi.it  Wed Mar  7 08:33:13 2001
From: ^NOSPAM^ifi at ifi.it (accattatevillo)
Date: Wed, 07 Mar 2001 14:33:13 +0100
Subject: Using a jar with jython
References: <3AA6115E.7B2ACDEE@ifi.it> <3aa62b59_1@nnrp1.news.uk.psi.net>
Message-ID: <3AA63899.90BA964E@ifi.it>


Simon Brunning wrote:
> 
> accattatevillo <^NOSPAM^ifi at ifi.it> wrote in message
> <3AA6115E.7B2ACDEE at ifi.it>...
> >I'm trying to convert a java application to jython.
> >
> >I use the IBM java toolbox JT400.jar to access data on a AS/400 bat
> I'm
> >not able to do the same thing with jython.
> >
> >With java I added jt400.jar to the CLASSPATH but in jython???
> >Anyone could help me?
> 
> Works for me:
> 
> Jython 2.0 on java1.1.8 (JIT: symcjit)
> Type "copyright", "credits" or "license" for more information.
> >>> import com.ibm.as400.access
> >>> my400 = com.ibm.as400.access.AS400()
> >>> my400.getVersion()
> 4
> 
> What are the problems that you are having? Error messages?
> 

C:\jdk1.3>jython
Jython 2.0 on java1.3.0 (JIT: null)
Type "copyright", "credits" or "license" for more information.
>>> import com.ibm.as400.access
Traceback (innermost last):
  File "<console>", line 1, in ?
ImportError: No module named access
>>> import com.ibm.as400
>>> dir()
['__doc__', '__name__', 'com']
>>> dir(com.ibm.as400)
['__name__', 'ui']
>>>

Tnx for the crosspost: I didn't know that there was that group.

TIA
Stefano


From qrczak at knm.org.pl  Fri Mar  9 03:24:36 2001
From: qrczak at knm.org.pl (Marcin 'Qrczak' Kowalczyk)
Date: 9 Mar 2001 08:24:36 GMT
Subject: "!=" is bad form. Re: sorry....never mind
References: <DZkp6.18811$pF4.3015652@news1.telusplanet.net> <fdlp6.16982$go.2157953@news0.telusplanet.net> <3AA6E308.5B63ECC6@cosc.canterbury.ac.nz> <3AA72253.FB021815@Lugoj.Com> <slrn9afcb5.lou.qrczak@qrnik.zagroda> <3AA822EE.F77D956F@cosc.canterbury.ac.nz>
Message-ID: <slrn9ah4q4.k09.qrczak@qrnik.zagroda>

Fri, 09 Mar 2001 13:25:18 +1300, Greg Ewing <greg at cosc.canterbury.ac.nz> pisze:

> > Icon uses "~=" for numbers and "~==" for strings. 
> 
> And I seem to recall it also uses ~=== for something else as well
> (lists? dictionaries?).

Identity comparison ('not is' in Python). I think value comparison
for lists must be coded by hand.

-- 
 __("<  Marcin Kowalczyk * qrczak at knm.org.pl http://qrczak.ids.net.pl/
 \__/
  ^^                      SYGNATURA ZAST?PCZA
QRCZAK


From andreas at andreas-jung.com  Sat Mar 17 09:23:18 2001
From: andreas at andreas-jung.com (Andreas Jung)
Date: Sat, 17 Mar 2001 09:23:18 -0500
Subject: Deleting dictionary items
References: <98ug6n$q9d$1@news6.svr.pol.co.uk>
Message-ID: <98vs4p$k9e$1@slb7.atl.mindspring.net>

"Will Newton" <will at nospam.misconception.org.uk> schrieb im Newsbeitrag
news:98ug6n$q9d$1 at news6.svr.pol.co.uk...


>                 for item in td.data.keys():

#                What td.data ?  I assume td.data is your dictionary ?
#                 item is now a tuple (x1,x2,x3...)

>                         if item[attr] != value:

#                  you can access the tuples elements by their position
using the [] operator
#                  e.g. item[0] gives x1, item[1] gives x2 ...
#
>                                 del td.data[item]

#                  This should work

Andreas






From aleaxit at yahoo.com  Fri Mar  9 18:07:36 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sat, 10 Mar 2001 00:07:36 +0100
Subject: Can anyone recomend a good intoduction to C...
References: <mailman.983894288.19965.python-list@python.org> <ewwp6.1466$y6.241005@ruti.visi.com> <slrn9aeblb.cg1.scarblac@pino.selwerd.nl> <djagat4kk3cs9hg0c3k39cudk6goblhkg6@4ax.com> <989mbk$24nuv$1@fido.engr.sgi.com> <slrn9ah66m.k09.qrczak@qrnik.zagroda> <98auhn0270j@news2.newsguy.com> <slrn9aiaq7.rmu.qrczak@qrnik.zagroda>
Message-ID: <98dpoe12ckn@news1.newsguy.com>

"Marcin 'Qrczak' Kowalczyk" <qrczak at knm.org.pl> wrote in message
news:slrn9aiaq7.rmu.qrczak at qrnik.zagroda...
    [snip]
> > > systems for other languages, time-critical modules for other
languages.
> > > C++ doesn't fit there, is unnecessarily too complicated for these
> > > tasks, and it isn't a high level language either.
> >
> > C++ fits *perfectly well* for these tasks.
>
> Maybe under Windows, but it's rarely used at all under Linux.

The Boost Python Library works perfectly well under Linux,
and so does Mozilla and the attendant XPCOM component
model.  They're C++ babies.  Writing such stuff, that fits
perfectly well with your model of tasks for which C++ is (you
claim) "unnecessarily too complicated", is made a LOT more
productive by using C++ rather than C.

> Kernel is in C, system interface is in C, system-level libraries are
> almost exclusively in C. C++ is used in some projects (KDE, nsgmls)
> but lower level libraries are generally expected to be available in C.

"Expected" is correct; it's a social issue, NOT a technical one --
technically, C++ is a better choice for most Python extensions &c
on a Linux platform.  But, if one wants acceptance, one writes in
C (e.g., that's what I did for GMPY, because I'd like people to use
it -- but it would be already out of alpha if had dared use C++, and
it would be just as fast &c).

Your opinion is widespread, and self-perpetuating, but it has no
current technical basis for such tasks as Python extensions on Linux.


Alex





From grzegorz at ugrad.cs.ualberta.ca  Sun Mar 11 01:33:54 2001
From: grzegorz at ugrad.cs.ualberta.ca (Grzegorz Dostatni)
Date: Sat, 10 Mar 2001 23:33:54 -0700
Subject: Buffer Objects...
Message-ID: <Pine.GSO.4.21.0103102331430.18879-100000@csu032.cs.ualberta.ca>

Hello, it's me again... I think I can solve my problem.  does anyone know
how to use Buffer objects from C?  A couple of examples would be nice.  I
can convert the PyString into Buffer Object, but how do I read from
it?  (getting the char array would be nice).

Can anyone help me??

Thanx in advance...
Grzegorz Dostatni


########################################################################
The world is full of people who have never, since childhood, met an
open doorway with an open mind.
                -- E.B. White

Chemistry is physics without thought.
Mathematics is physics without purpose.

"If you want to know what God thinks about money, just look at the people
He gives it to." 
-- Old Irish Saying 




From boswell at ugcs.caltech.edu  Thu Mar 22 04:02:33 2001
From: boswell at ugcs.caltech.edu (dustybos)
Date: Thu, 22 Mar 2001 01:02:33 -0800
Subject: automatic cookie handling with urllib?
Message-ID: <3AB9BFA9.56301B22@ugcs.caltech.edu>

I am doing a project that essentially involves rewriting a browser.  The
server
I am dealing with uses cookies and lots of redirects, and I was hoping
to use
urllib.py seamlessly.  But since urllib doesn't handle cookies, I feel
like I have
no choice but hack away at urllib.py and httplib.py to make a version
that
handles it for me.  It's not *that* big of a deal, but I would like to
use a module
that has already implemented the RFC (path, expiration, domain, etc...)
for me.

Has anyone run into a similar problem or know of a good
library/module/whatever
to make cookies automatically handled like redirects?

Thanks a lot.
Dustin Boswell.



From andy at reportlab.com  Thu Mar 15 14:27:15 2001
From: andy at reportlab.com (Andy Robinson)
Date: Thu, 15 Mar 2001 19:27:15 +0000
Subject: ReportLab commercial toolkit enters beta testing
Message-ID: <pj52bt87k6il1v961thjbt85u5aq9ak4s2@4ax.com>

ReportLab Inc. is introducing a new commercial product range which
builds on our existing Open Source Python PDF generation toolkit.

RML2PDF and PageCatcher are entering beta testing today.  These are
the first elements of a tightly integrated suite of packages for
building world-class reporting solutions for the Web and back office.

Report Markup Language and RML2PDF
----------------------------------
RML2PDF is a powerful tool which allows you to generate reports in PDF
with simplicity, speed and flexibility, and without the level of
programming knowledge you need to use our Open Source ReportLab
toolkit. The primary market is web or application servers, though it
can also be embedded in client applications.

RML2PDF uses the Report Markup Language. RML is an XML dialect which
is simple, flexible and powerful. It allows you to produce PDF
documents that include drawings, photographs, multicolumn page layouts
- all with the ease of writing basic HTML rather than raw PDF. It?s
even easier than Python.

We offer an attractive package to resellers and consultants who
already deliver solutions to their customers and who can use this as
one more piece of their tool suite; and will consider licensing the
technology for use within third party applications.

See http://www.reportlab.com/rml/

PageCatcher
-----------
This tool allows you to import graphical elements or whole pages from
existing PDF files and use them in RMl2PDF or with our Open Source
reporting tools. It can be used to
- overprint PDF files with extra information
- grab specific pages, reorder and assemble documents
- save trees by doing 4-up versions of documents
- use existing documents as 'forms' and draw over them

It gives your graphic designers the freedom to use any tools they wish
to design artwork, which can then be included in reports.  It also
opens up many possibilites for forms processing.

See http://www.reportlab.com/pageCatcher/

What about the existing ReportLab toolkit?
------------------------------------------
The ReportLab toolkit is free, and will always be free. If you want to
use Python to output PDF, then the ReportLab toolkit will always be
available to you. RML2PDF is aimed at a different market: those
organizations who want a "straight out of the box" solution, or those
who aren?t Python-savvy. RML and any other commercial products we
release will _always_ be in addition to the ReportLab toolkit and
never replace it.

If you?re interested, send an email to mailto:info at reportlab.com. Or
have a look at the products section on our web site:
    http://www.reportlab.com/rml/


Andy Robinson
CEO and Chief Architect, ReportLab Inc.


From kens at sightreader.com  Thu Mar 15 16:27:08 2001
From: kens at sightreader.com (Ken Seehof)
Date: Thu, 15 Mar 2001 15:27:08 -0600
Subject: Magnitude of the wx* market (was: Python In A Nutshell - suggestions)
References: <3AB26DAC@MailAndNews.com> <98revc$5rn@spool.cs.wisc.edu>
Message-ID: <004f01c0ad96$b0e60c60$04090a0a@upcast.com>

I'll definitely buy a wxPython book when it comes out.  As a
C++ programmer, I don't mind reading C++ documentation
(i.e. wxWindows), but people who don't know C++ tend to
get very confused and annoyed by it.  There seems to be
enough volume on this mailing list to make a wxPython book
worthwhile.

What we really need is an AI enhanced version of swig to
automatically convert wxWindows documentation to wxPython
documentation. :-)

----- Original Message -----
From: "Manoj Plakal" <plakal-NO at SPAM-cs.wisc.edu>
Newsgroups: comp.lang.python
To: <python-list at python.org>
Sent: Thursday, March 15, 2001 4:16 PM
Subject: RE: Magnitude of the wx* market (was: Python In A Nutshell -
suggestions)


> FYI, Robin Dunn (wxPython czar) mentioned (on the wxpython-users mailing
> list a few days back) that a wxPython book was in the works. But it's
still
> a while away.
>
> I think the wxWindows web site does mention plans for a "wxBook".
>
> Manoj
>
>
> On Wednesday 14 March 2001 19:11, Laura Lewin wrote:
> > Sure, we'll take this suggestion into consideration (on expanding
wxPython
> > coverage) when we publish the next edition of Python Programming on
Win32.
> >
> >>===== Original Message From tim at digitig.cix.co.uk =====
> >>In article
> <A909AD29CBD32AFC.4F1641C135749F87.7149617C7C166896 at lp.airnews.net>,
> >>claird at starbase.neosoft.com (Cameron Laird) wrote:
> >>> In article <3AAA4FE3.B471B689 at northwestern.edu>,
> >>> Louis Luangkesorn  <lluang at northwestern.edu> wrote:
> >>> .
> >>> >And while I'm at it, one useful book to have would be on
> >>> wxWindows/wxPython.  I see references to it in several places, and a
> >>> few intros on the web, but
>
> --
> http://mail.python.org/mailman/listinfo/python-list
>




From joconnor at cybermesa.com  Sat Mar  3 10:48:58 2001
From: joconnor at cybermesa.com (Jay O'Connor)
Date: Sat, 03 Mar 2001 15:48:58 GMT
Subject: Adding a method to an instance
References: <G2%n6.360714$JT5.11606234@news20.bellglobal.com>
Message-ID: <3aa1110e.93290162@news.cybermesa.com>

On Sat, 03 Mar 2001 05:12:06 GMT, Bryan Mongeau <bryan at eevolved.com>
wrote:

>Quick one for you guys:
>
>I would like to add a method to an instantiated class. 
>
>class foo:
>  def __init__(self):
>    self.bar = 123
>
>f = foo()
>
>Now to add a method to the instance f of class foo:
>
>def newMethod(self):
>  print self.bar
>
>f.__class__.newMethod = newMethod
>
>This works, except that it modifies the definition of class foo!
>
>>>> dir(foo)
>['__doc__', '__init__', '__module__']
>>>> f.__class__.newMethod = newMethod
>>>> f.newMethod()
>123
>>>> dir(foo)
>['__doc__', '__init__', '__module__', 'newMethod']
>
>Is there a way to add the method only to the instance?  Oh, and please 
>don't ask me *why* I want to do this. Just trust me :)
>


You can use the module 'new' to create a new instance method for an
object

>>> import new
>>>class foo:
	def __init__(self):
		self.bar = 123

>>>f = foo()

>>> def newMethod (self):
		print self.bar

>>>f.newMethod = new.instancemethod (newMethod, f, foo)
>>>f.newMethod()

123

>>> g = foo()
>>> g.newMethod()
Traceback (innermost last):
  File "<pyshell#16>", line 1, in ?
    g.newMethod()
AttributeError: 'foo' instance has no attribute 'newMethod'

Notice that f has the newMethod, but it only applies to f, g does not
get the new method


Jay O'Connor
joconnor at cybermesa.com
http://www.cybermesa.com/~joconnor

Python Language Discussion Forum - http://pub1.ezboard.com/fobjectorienteddevelopmentpython

"God himself plays on the bass strings first, when he tunes the soul"


From chocolate at toblerone.yum  Wed Mar 21 14:35:02 2001
From: chocolate at toblerone.yum (Bubba Brains)
Date: Wed, 21 Mar 2001 14:35:02 -0500
Subject: learning python...
References: <3AB8F4E1.184DD9C@noaa.gov> <Z97u6.16733$Im6.1761313@newsread1.prod.itd.earthlink.net>
Message-ID: <3ab90559_6@goliath.newsfeeds.com>

Forget Learning Python?  I think it's the best reference/tutorial around.  I
use python once every couple months, and sometimes forget the simple python
syntax.  The Learning Python book is great for getting up to speed fast, and
to look up common stuff.  The other OReilly book (Programming Python) is
absolutely annoying to use.  It's not even a good reference book (good luck
finding the info you want quickly).  Don't know about the Prentice book.

M
http://www.mp3.com/maddogskullcap

"Simon Kesenci" <tomega at earthlink.net> wrote in message
news:Z97u6.16733$Im6.1761313 at newsread1.prod.itd.earthlink.net...
> If you know programming, forget _Learning_ _Python_.  I tried
_Programming_
> _Python_ last year, but it covered version 1.3.  A good book if they
> updated it since then.  As a programmer, you should find the online
> tutorial easy and helpful <http://python.org/doc/current/tut/tut.html>.
>
> --S.
>
> Benjamin.Altman wrote:
>
> > Hello.  From the perspective of someone who already has programming
> > experience, would anyone know if O'reillys "Learning Python" book would
> > be better than say Prentice Hall's "Core Python Programming"?  Or would
> > it be better to go straight into something like "Programming Python"?
> >
> > Thanks,
> > Ben
> >
> >
>




-----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
-----==  Over 80,000 Newsgroups - 16 Different Servers! =-----


From aleaxit at yahoo.com  Wed Mar 28 14:49:25 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Wed, 28 Mar 2001 21:49:25 +0200
Subject: diff lists
References: <3AC1DD4E.4BA7D421@aon.at>
Message-ID: <99tf9m02ph7@news1.newsguy.com>

"Oliver Vecernik" <vecernik at aon.at> wrote in message
news:3AC1DD4E.4BA7D421 at aon.at...
> Hi,
>
> I've got following two lists:
>
> ['a', 'b', 'c', 'd', 'e', 'f']
> ['e', 'c', 'f']
>
> I'd like to have the result:
>
> ['a', 'b', 'd']
>
> The list need not to be ordered. ['d', 'a', 'b'] will also be ok. What
> is the most effective way to achive that result?

No doubt, it's a loop testing with the 'in' operator if one list is
pretty small, one using an auxiliary dictionary if they're more
substantial:

def smallExcluded(big, small):
    return [ x for x in big if x not in small ]

def anyExcluded(big, any):
    auxdict = {}
    for x in any: auxdict[x] = None
    return [ x for x in big if not auxdict.has_key(x) ]

There are also interesting approaches when the lists are known
to be sorted, but I guess those would be of purely theoretical
interest to you:-).


Alex







From Markus_Reitz at yahoo.com  Wed Mar 21 06:47:22 2001
From: Markus_Reitz at yahoo.com (Markus Reitz)
Date: Wed, 21 Mar 2001 12:47:22 +0100
Subject: Checking the type
Message-ID: <99a493$bhl$1@sun.rhrk.uni-kl.de>

Hi,

is there a possibilty to check if a variable refers to an object of type x?
Does a typeof-command exist in Python:

  if a typeof list:
     #Commands only useable with list

I read the Python Tutorial, but have not found a hint to this topic.

Thanks

Markus




From ajm at enteract.com  Thu Mar 29 20:33:46 2001
From: ajm at enteract.com (Alan Miller)
Date: Thu, 29 Mar 2001 19:33:46 -0600
Subject: Updated edition of Programming Python out
References: <pj4iat8kevh2fp52gkua87ig16h76oh3m0@4ax.com>
Message-ID: <MPG.152d8de631a0ecbf98977b@news-enteract>

Simone Paddock (simone at oreilly.com) wrote:
>The updated, 1256 pages strong tome is finally here:
>"Programming Python, 2nd Edition"
>http://www.oreilly.com/catalog/python2/

Having picked up the one copy* that one of the local Borders had ordered 
(I had them dig it out of the back room before it even hit the shelves), 
I have one question:
  Is there such a thing as a second edition of a computer (or other 
  technical) book that is _thinner_ than the original?

ajm

* Another had six coming in within the next week or so but I didn't feel 
like waiting; none of the B&N had it in stock or on order - 
disappointing.


From bbhaydon at bigpond.com  Sat Mar 10 10:21:24 2001
From: bbhaydon at bigpond.com (Brett Haydon)
Date: Sun, 11 Mar 2001 02:21:24 +1100
Subject: mxODBC on Windows 2000?
References: <3AA7B94A.23D5C849@interbulletin.com> <3AA803B0.7E08E836@interbulletin.com>
Message-ID: <xqrq6.12046$v5.51961@newsfeeds.bigpond.com>

I've been using  mxODBC 1.1.1 no problems using W2k and Active Py.

The most common setup problem is changing the name of the existing
odbc.py that comes with Active Py.

I have been using it for 6 months with both MySQL and MS Access.

Exactly where are the failures

regards,

bbhaydon
"Bill Tolbert" <donotreply at interbulletin.bogus> wrote in message
news:3AA803B0.7E08E836 at interbulletin.com...
> It's not limited to ActivePython: I get the same (non) result under IDLE
> and the command line. I'm seeing it fail under Win2000 and Win98.
>
> Bill
>
> Rob Earl <donotreply at interbulletin.bogus> wrote in article
> <3AA7B94A.23D5C849 at interbulletin.com> :
> >
> >Has anyone been able to get the mxODBC module for Python to work properly
> with ActivePython on Windows 2000?
> The reason I ask is because every time I try to execute a SQL command,
> the CPU usage goes up to 100% on my machine and the python shell is
> locked up.  Can anyone help?  Thanks.
> _______________________________________________
> Submitted via WebNewsReader of http://www.interbulletin.com
>




From nickson at centropolisfx.com  Thu Mar 22 18:22:01 2001
From: nickson at centropolisfx.com (Nickson Fong)
Date: Thu, 22 Mar 2001 15:22:01 -0800
Subject: Multiple Processes on Unix with Python
Message-ID: <3ABA8919.835D8BED@centropolisfx.com>

Can anyone show me a quick example on how to fork multiple processes on
a multiple cpu Irix machine. I know I need os.fork() and os.execvp() but
can't figure out how they work together.

-- 
thanks


        (:~~~ (:~~~ (:~~~    NO MORE EGGS   ~~~:) ~~~:) ~~~:)
			        JUST 
                               BADFROG

	
	      
                     Tel : (310) 204-7300 Ext 120
                         Pager : 626-582-5459
                      Office Fax : 310-204 7301

                       nickson at centropolisfx.com
	                  ubadfrog at yahoo.com


From barry at digicool.com  Mon Mar 12 10:12:22 2001
From: barry at digicool.com (Barry A. Warsaw)
Date: Mon, 12 Mar 2001 10:12:22 -0500
Subject: "!=" is bad form. Re: sorry....never mind
References: <15018.30583.852611.836468@anthem.wooz.org>
	<LNBBLJKPBEHFEDALKOLCCELJJEAA.tim.one@home.com>
Message-ID: <15020.59222.128369.486729@anthem.wooz.org>

>>>>> "TP" == Tim Peters <tim.one at home.com> writes:

    TP> If you look at the CVS history of Grammar/Grammar, you'll
    TP> discover that In The Beginning, Python allowed only the <>
    TP> spelling.  != wasn't added until revision 1.5 (of Grammar, not
    TP> Python).  So this is a curious case where Guido disowned his
    TP> first impression (from only <>, to <> is obsolescent); but
    TP> since this is the same process by which your print>>file got
    TP> into the language, you had better not protest too loudly
    TP> <wink>.

Ah, so I wonder if Guido would like my "print <>" patch.  This one
prints to a second file object everything it reads in from a first
file object, e.g.:

    print <> sys.stdin, sys.stdout

I can't change that to "print !=" because I've got a second patch that
gives this the semantics such that it prints an expression to a file
object, but only if they are really equal.  Here "print !=" is the
same as "print =!" but the former is obsolescently deprecated.

my-calendar-must-be-running-ahead-about-20-days-ly y'rs,
-Barry



From ppe at albert.com  Thu Mar  8 11:02:47 2001
From: ppe at albert.com (Pablo Pernot)
Date: Thu, 8 Mar 2001 17:02:47 +0100
Subject: Ldap / Python help
Message-ID: <20010308170247.6aa8e6a4.ppe@albert.com>

Hello all,

I'm new in python

I found an ldap module.

Where can I find documentation about ldap/python

my modify instruction won't work => modify_s is waiting for a string (dn)
and a list (mylist) formed by tuple (mytuple) and this tuple is formed for 3 strings.

So I can't understand the error message about integer.

mytuple=("MOD_REPLACE","sn","test")
mylist=[mytuple]

>>> l.modify_s(dn,mylist)
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
TypeError: an integer is required


l is my ldap bind & connection which works.

Thanks for the help

Pablo

-- 
Pablo Pernot --Albert-Inc-- [http://www.albert.com] 
email : ppe at albert.com   | phone  : +33 499 130 900            
fax   : +33 499 130 901  | mobile : +33 620 805 810     



From ullrich at math.okstate.edu  Sat Mar  3 10:15:51 2001
From: ullrich at math.okstate.edu (David C. Ullrich)
Date: Sat, 03 Mar 2001 15:15:51 GMT
Subject: Hitting two targets:  OO + group theory (pedagogy)
References: <u2k0atkgkiatdp9ukj3mpd1ej0b648ldsm@4ax.com>
Message-ID: <3aa1077a.1110803@nntp.sprynet.com>

On Fri, 02 Mar 2001 18:07:16 -0800, Kirby Urner
<urner at alumni.princeton.edu> wrote:

>
>============== OREGON CURRICULUM NETWORK =====================

Does the Oregon Curriculum Network have any official status, ie
any actual connection with the state of Oregon?

[...]
>
>OO + GROUP THEORY 
[...]
>Let's define our sandbox to be groups of relative primes less
>than a particular prime modulus n.  These are groups under 
>multiplication, where a*b is defined as (a*b) mod n. 

One comment would be that if we're talking about pedagogy and
curricula and things we probably want to get the terminology
straight. There's no such thing as a "relative prime" - you
meant "the positive integers less than n which are
relatively prime to n."

And more important there's no reason why the modulus n
should be prime! This gizmo is a group under multiplication
whether n is prime or not. (That's the point to specifying
we're talking about the integers relatively prime to n...)

Seems more interesting to me to start with a class
Group that includes stuff common to all groups, so
that a subclass can represent any group at all. If we're
going to implement what's above as part of a
"curriculum" then at least we should make it clear
that our sandbox is not "groups", it's a tiny part of
the category of all groups.


From piet at cs.uu.nl  Sat Mar 24 08:20:06 2001
From: piet at cs.uu.nl (Piet van Oostrum)
Date: 24 Mar 2001 14:20:06 +0100
Subject: Install Python on Apache
References: <99fph5$a2n$1@rex.ip-plus.net>
Message-ID: <wzelvnz4l5.fsf@sunshine.cs.uu.nl>

>>>>> "Sharp" <s.sharp at fh-aargau.ch> (S) writes:

S> Hi Everybody
S> I sucked down Python from python.org and started to install the program
S> on a win2000 apache system. that has worked perfectly. but then I tried
S> to set up the apache httpd.conf file with the following options

S> AddType application/x-httpd-py .py
S> Action application/x-httpd-py /cgi-bin/Python/python.exe

For a start you should never put python.exe (or any other interpreter) in
cgi-bin. This is a large security hole!
-- 
Piet van Oostrum <piet at cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: P.van.Oostrum at hccnet.nl


From loewis at informatik.hu-berlin.de  Fri Mar  2 10:52:04 2001
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 02 Mar 2001 16:52:04 +0100
Subject: A counter-proposal to __future__ in PEP 236
References: <mailman.983487602.7750.python-list@python.org>
Message-ID: <j4lmqojhi3.fsf@informatik.hu-berlin.de>

"Tim Peters" <tim.one at home.com> writes:

> It's not possible-- in Python's parser --to make the identifier
> "directive" a keyword in some contexts but not others

Sure it is; my patch does that. It is even possible to implement the
syntax that Joshua wants to have, it is just not easy.

> You also have to consider the parsers in tools, from the Emacs
> python-mode thru whatever the heck PythonWorks uses.

The Emacs Python mode deals just fine with the proposed syntax as it
is now; i.e. it does not highlite directive (as it might do in some
future release), however, its indentation behaves just fine even in
presence of directive. Once it knows that directive is a keyword
(i.e. by adding it to the kw1 list), it will highlite it in all
contexts (even if used as an identifier) - just as it does for, say,
print.

I don't know whatever the heck PythonWorks uses, so I don't know
whether it would break. My guess is that it would not.

> Doesn't mean it can't be hacked, but it's ugly, and you have to get
> a lot of different people to sign up to do all that hacking.

That's only for the transitional period for introducing the new
keyword; I firmly believe that most tools will continue to work as
much as reasonable, instead of outright breaking, even if they are not
touched at all.

Regards,
Martin



From jschmitt at vmlabs.com  Thu Mar 22 15:10:59 2001
From: jschmitt at vmlabs.com (John Schmitt)
Date: Thu, 22 Mar 2001 12:10:59 -0800
Subject: Good Editor? (What's the big deal with syntax colouring?)
Message-ID: <008F0A63472BD311AF9800104BCD102561CD3E@minirex.vmlabs.com>

In the dog days of DOS, I was a brief guy.  Now I use CodeWright.  It's the
best editor I've used (and I've used vi, although not too extensively).

Why do people want syntax colouring?  I can't stand it.  One of my minor
beefs with CodeWright is that there isn't a one button way to tell it to
change _all_ of my colours all at once.  Am I missing something by disabling
syntax colouring?  It always reminds me of the artwork my 6-year-old nephew
brings home from school to show his mommy.

The one-and-only colour combination that works best for source code is
bright green on black.  Easy on the eyes and a 9 point, sans-serif,
mono-spaced font looks good at 1600x1200 85Hz on a garden variety 17"
monitor.

John



From chris.gonnerman at usa.net  Wed Mar 28 23:01:58 2001
From: chris.gonnerman at usa.net (Chris Gonnerman)
Date: Wed, 28 Mar 2001 22:01:58 -0600
Subject: OO Windows Registry API
References: <99t4mu42c57@news1.newsguy.com>
Message-ID: <00f401c0b805$0e988e00$a100000a@local>

Yes, please post it!  It sounds most useful.

----- Original Message -----
From: <york at ficom.net>
Newsgroups: comp.lang.python
To: <python-list at python.org>
Sent: Wednesday, March 28, 2001 10:49 AM
Subject: OO Windows Registry API


>
> After using the _winreg module I was pining for the fjords -- oh, sorry --
I was pining for a more Pythonesc API.  I noticed that the Python
documentation mentions that a better API may be written in the future, but
none has been forthcoming.  So -- while up at 2AM with my 6-week-old son
last night -- I wrote a wrapper that exposes the registry as a
dictionary-like object.  It seems to work really well, and more importantly
it feels more like a Python module should.
>
> I decided to search Python-Dev for mentions of a higher-level API and
found a squabble over various proposals.  Did the idea of an OO interface
die last August, or is there any interest in what I've done?
>
> (Untested) Example code:
>
> from winreg import Registry, NewKey, HKEY_CURRENT_USER
> reg = Registry()
>
> # Create a new "My App" key under HKEY_CURRENT_USER/Software.
> reg[HKEY_CURRENT_USER]['Software']['My App'] = NewKey()
>
> # Add some values to the newly created key.
> # Note: the module only supports strings right now, but that's
> #  easy to change.
> reg[HKEY_CURRENT_USER]['Software']['My App']['foo'] = 'value 1'
> reg[HKEY_CURRENT_USER]['Software']['My App']['bar'] = 'value 2'
>
> # Create a new subkey.
> reg[HKEY_CURRENT_USER]['Software']['My App']['subkey'] = NewKey()
>
> # Delete a value.
> del reg[HKEY_CURRENT_USER]['Software']['My App']['foo']
>
> # Delete a key.
> del reg[HKEY_CURRENT_USER]['Software']['My App']
>
>
> Any comments or suggestions?  If anyone is interested in the code, I'll
post it on the web.
>
> --
> Benji York
> york at ficom.net
>
>
> ==================================
> Posted via http://nodevice.com
> Linux Programmer's Site
> --
> http://mail.python.org/mailman/listinfo/python-list
>




From tjreedy at udel.edu  Mon Mar 12 13:49:29 2001
From: tjreedy at udel.edu (Terry Reedy)
Date: Mon, 12 Mar 2001 13:49:29 -0500
Subject: Extending Python: rewriting a single method in C
References: <g0wv9vxms0.fsf@scumbag.ecs.soton.ac.uk> <98idtn013v3@news2.newsguy.com>
Message-ID: <98j5d4$nqh$1@news.udel.edu>

> "Jacek Generowicz" <jmg at ecs.soton.ac.uk> wrote in message
> news:g0wv9vxms0.fsf at scumbag.ecs.soton.ac.uk...
> > I have an object-oriented numerical code written in Python. It has
> > come to the stage where the speed of the code is seriously impeding
> > progress. Profiling reveals (surprise, surprise) that most of the time
> > is spent in only 2 methods.
> >
> > Is there any way of re-writing just these methods in C ?

Are you already using the Numerical Extensions?  They include many
functions rewrittin in C for speed.




From nomad*** at ***freemail.absa.co.za  Mon Mar 26 08:15:13 2001
From: nomad*** at ***freemail.absa.co.za (nomad)
Date: Mon, 26 Mar 2001 13:15:13 GMT
Subject: Help how do I read `word' variables from a binary file?
References: <3abf3b0b.1487830@news.is.co.za> <aafubtkrem3o3fu5j5sbf9vnh2e3dnrea1@4ax.com>
Message-ID: <3abf4091.2901634@news.is.co.za>

On Mon, 26 Mar 2001 13:01:29 GMT, David Fuess <fuess at att.net> wrote:

>Read the data in recordsize groups like you're doing now, then use
>struct.py do decompose the records into Python variables.
>
>Dave
>
>On Mon, 26 Mar 2001 12:57:31 GMT, nomad***@***freemail.absa.co.za
>(nomad) wrote:

><SNIP>

>>Any help would be greatfully recieved.
>>
>>TIA.
>
Man that was Quick!!!  Thanks for the tip, I'll try it right away.

Cheers.

Nomad


From fluxent at yahoo.com  Fri Mar 16 12:55:37 2001
From: fluxent at yahoo.com (fluxent at yahoo.com)
Date: Fri, 16 Mar 2001 17:55:37 -0000
Subject: KeyboardInterrupt in IIS?
Message-ID: <98tk2p+qshm@eGroups.com>

I'm running Python via ASP pages in IIS (win2k).

Occasionally I start getting errors when I ask for any page that look 
like
HTTP 500.100 - Internal Server Error - ASP error
Internet Information Services

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

Technical Information (for support personnel)

Error Type:
Python ActiveX Scripting Engine (0x80020009)
Traceback (innermost last): KeyboardInterrupt: 
/test/biztrack/admin/timechargesReview.asp

It eventually seems to go away on its own, or else I reboot (gee, 
thought I didn't have to do that anymore...)






From stowasser.h at idowa.de  Thu Mar 15 02:53:02 2001
From: stowasser.h at idowa.de (Stowasser Harald)
Date: Thu, 15 Mar 2001 08:53:02 +0100
Subject: Better string.translate?
References: <3aaf758f$1_2@news.zet.net> <mailman.984590438.506.python-list@python.org>
Message-ID: <3ab0749f$1_3@news.zet.net>

"Steve Purcell" <stephen_purcell at yahoo.com> schrieb :
> Use the 'sub' function in the 're' module:

Thank you, but i don't want do delete the characters. I need to
translate them into Spaces.

Anybody knows how?





From alex at shindich.com  Sat Mar 10 04:30:26 2001
From: alex at shindich.com (Alex Shindich)
Date: 10 Mar 2001 10:30:26 +0100
Subject: Static class methods
References: <3AA6E554.DFB04DBF@cosc.canterbury.ac.nz>
Message-ID: <3AA9F409.9957A813@shindich.com>

>  [Previous]   [Next]   [Reply]   [Index]   [Home]   [Find]
>
>    * Newsgroups: comp.lang.python
>    * From: Greg Ewing <greg at cosc.canterbury.ac.nz>
>    * Subject: Re: Static class methods
>    * Date: Thu, 08 Mar 2001 14:50:12 +1300
>    * Organization: Computer Science Dept, University of Canterbury
>
>   ------------------------------------------------------------------------
> The class scope is already special with respect to
> nested scopes anyway, since it doesn't count as a
> scope when you're looking out from inside a method.
> So even if there were class methods, you wouldn't
> be able to call them from within a method without
> qualification of some sort.
>
> However, class scopes *do* behave like other scopes
> in that the code in the body of the following function:

I would argue that this is inconsistent. Since nested scopes are supported now, classes should be subjected to the same rules functions are.

--
Alex Shindich
mailto:alex at shindich.com
Visit http://www.shindich.com/


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20010310/a618c1fe/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: alex.vcf
Type: text/x-vcard
Size: 194 bytes
Desc: Card for Alex Shindich
URL: <http://mail.python.org/pipermail/python-list/attachments/20010310/a618c1fe/attachment.vcf>

From fredrik at pythonware.com  Fri Mar 23 02:01:29 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 23 Mar 2001 07:01:29 GMT
Subject: Digest Number 1946
References: <mailman.985313369.3107.python-list@python.org>
Message-ID: <dxCu6.3134$4N4.460882@newsc.telia.net>

Bill Bell wrote:
> I'm new to Python, have been playing with someone's whois client,
> which contains the following lines:
>
>         s = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
>         s.setblocking(1)
>         s.connect((self.whoisserver, 43))
>
> The code works fine (as long as the requested whois server is
> available) if the argument to 'setblocking' is 1. However, if I change
> this to 0 I receive the diagnostic,
>
> "The socket operation could not complete without blocking."
>
> I cannot find the code that contains this literal.

did you look in the sources for your operating system? ;-)

> Would someone please suggest what might be going wrong.

nothing's going wrong here -- the system's reminding you that
when you're using non-blocking mode, the fact that connect
returns doesn't mean that it has succeeded yet.

you're supposed to catch that exception, and check for com-
pletion using select or poll.  see your favourite socket pro-
gramming guide for more info.

Cheers /F




From alex at shindich.com  Sun Mar  4 17:22:10 2001
From: alex at shindich.com (Alex Shindich)
Date: Sun, 4 Mar 2001 14:22:10 -0800
Subject: Static class methods
Message-ID: <rdzo6.154$gc4.80445@news.pacbell.net>

Just out of my curiosity, have any of you ladies and gentlemen experienced a
need for an equivalent of C++/Java static method?
Because I have, but it is impossible add them to Python classes -- all class
methods are treated as instance methods.
This is especially unfortunate given that local functions now work properly
thanks to PEP 227.
I was thinking that a new keyword "unbound" could be used to identify
functions that belong to a class but do not expect "self" as their first
attribute.
>From outside the class "unbound" methods would be invoked by specifying the
class name -- Foo.my_static_method (). From within the class they would work
similar to local functions that are defined at the class level. The __ trick
should work for the "unbound" methods as well to allow for creation of
private static methods.

Examples:
>>>class Foo:
    """This is a sample class..."""

    __myPrivateStaticVariable  = None

    unbound getMyPrivateStaticVariable ():
        """getMyPrivateStaticVariable (obj) -> obj

        This method returns the value stored in private static variable...
        """
        return Foo.__myPrivateStaticVariable


    unbound modifyMyPrivateStaticVariable (obj):
        """modifyMyPrivateStaticVariable (obj) -> void

        This method assigns obj to the private static variable...
        """
        Foo.__myPrivateStaticVariable = obj

>>> type (Foo.modifyMyPrivateStaticVariable)
<type 'unbound method'>
>>> type (Foo.getMyPrivateStaticVariable ())
<type 'None'>
>>> Foo.modifyMyPrivateStaticVariable ("abc")
>>> Foo.getMyPrivateStaticVariable ()
"abc"
>>>





From dwig at advancedmp.net  Tue Mar 27 12:42:27 2001
From: dwig at advancedmp.net (Don Dwiggins)
Date: 27 Mar 2001 09:42:27 -0800
Subject: WinNT problem with Python and Cygwin Bash
In-Reply-To: David Bolen's message of "26 Mar 2001 21:55:01 -0500"
References: <mailman.985632095.6515.python-list@python.org>
	<uwv9b53ay.fsf@ctwd0143.fitlinxx.com>
Message-ID: <uy9tr9kho.fsf@advancedmp.net>

David Bolen writes:
> However, am I correct in guessing that "/e" isn't actually a top level
> directory on your current drive, but rather a Cygwin mount point for
> your E: drive?  If that's the case, then you'll definitely have a
> problem because there's no way for Python (unless you've built a
> Cygwin version of it) to handle that mount translation, as such
> translation is built into the Cygwin DLL.

You're guessing right.

> In either case, you can use the cygpath utility to convert any Cygwin
> path into a Windows path (or vice versa, or other conversions), and it
> will handle replacing any mount references with appropriate Windows
> native paths and adding drive references as necessary.

> So if you modify the created shell script slightly to take whatever
> pathname it would normally send into Python and run it through
> "cygpath -w" first, you should be ok.  Or, just create a small
> front-end shell script to run instead of Python which can do the
> expansion and then call Python, and tell the Happydoc created script
> to run that instead of Python.

Thanks; that's pretty much what I'd surmised (but I'd hoped I was missing
something simpler).  I'll have to change the Happydoc-generated script to
run Bash instead of Python, and call Python inside the script.

-- 
Don Dwiggins                    "Solvitur Ambulando"
Advanced MP Technology
dwig at advancedmp.net




From neotaku at free.fr  Tue Mar  6 08:09:05 2001
From: neotaku at free.fr (Neotaku)
Date: Tue, 06 Mar 2001 13:09:05 GMT
Subject: Access to sybase API
References: <obOo6.1158$f07.1696858@nnrp6.proxad.net> <3AA3B0D0.90876BA7@schlund.de> <m3zoezxjly.fsf@vole.object-craft.com.au>
Message-ID: <Rj5p6.1003$XO1.4290677@nnrp1.proxad.net>

    Yes of course thanks a lot i will try you new module with a friend

see ya

"Dave Cole" <djc at object-craft.com.au> a ?crit dans le message news:
m3zoezxjly.fsf at vole.object-craft.com.au...
> >>>>> "Carsten" == Carsten Gaebler <cg at schlund.de> writes:
>
> Carsten> Neotaku wrote:
> >> I search a code source for python to access Sybase thru API
> >> connexion the Sybase DB-Library API or Sybase CT-Library (aka the
> >> Client Library) API
>
> Carsten> Have a look at Dave Cole's Sybase module:
> Carsten> http://www.object-craft.com.au/projects/sybase/
>
> I have been slowly rewriting this module to make it as close to 100%
> DB-API 2.0 compliance as possible.  My new version has split the
> module into a Python part and a C extension module which is a fairly
> thin wrapper on top of the Sybase CT API.  I am getting fairly close,
> but the pressures of earning money keep forcing me back onto real
> work.
>
> For some people, the new version of the module might be good enough
> already to replace the current 0.12 module.  I have not placed a link
> to the new module on the homepage because I want to wait until the new
> module can completely replace the old module.
>
> If you are interested in giving the new module a spin, or you are
> interested in how the CT library works, you can get the work in
> progress from:
>
>         http://www.object-craft.com.au/projects/sybase/Sybase-0.21.tar.gz
>
> If you were not happy with the speed of cursors in the old module then
> you should try this one.  Cursors are now about five times faster
> because I am no longer using ct_cursor().
>
> In some ways, the new module is much closer to DB-API compliance than
> the old one.
>
> - Dave
>
> P.S.  If you want to hekp out with the implementation, I will
>       gratefully accept patches.
>
> --
> http://www.object-craft.com.au
>



From dsh8290 at rit.edu  Wed Mar  7 15:33:30 2001
From: dsh8290 at rit.edu (D-Man)
Date: Wed, 7 Mar 2001 15:33:30 -0500
Subject: gah! I hate the new string syntax
In-Reply-To: <Koco6.7518$1D5.244602@e420r-atl1.usenetserver.com>; from sholden@holdenweb.com on Sat, Mar 03, 2001 at 03:19:05PM -0500
References: <mailman.983615170.23313.python-list@python.org> <tN5o6.2746$eH3.14913@NewsReader> <Koco6.7518$1D5.244602@e420r-atl1.usenetserver.com>
Message-ID: <20010307153329.C29445@harmony.cs.rit.edu>

On Sat, Mar 03, 2001 at 03:19:05PM -0500, Steve Holden wrote:
| "chris" <chrislamb at btinternet.com> wrote in message
| news:tN5o6.2746$eH3.14913 at NewsReader...
| > > at latest since the 19th century; in America, it appears this is
| normally
| > > introduced in the 6th or 7th grades (ages 11-12, for those of you in
| > > inferior countries <wink>
| >
| > In Scotland that kind of stuff is pre-School! <wink-back>
| >
| > Chris
| 
| Ha! As an Englishman married to a Scot I'm not going to let you get away
| with that one: this is classic Scottish braggadocio [I expect Italian
| education form tne martellibot about that]. Brass neck, as we simple
| Yorkshire folk would term it.
| 

What about one who is both English and Scots, yet never set foot on
that island ;-)?

-D




From fredrik at pythonware.com  Wed Mar 14 11:08:07 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 14 Mar 2001 16:08:07 GMT
Subject: Newbie: Is there a more pythonic approach?
References: <3AAE87E8.FE11B4F9@sewardweb.com>
Message-ID: <HHMr6.253$4d3.54012@newsb.telia.net>

loyd Sommerer wrote:
> Thanks, the program follows below.

looks fine to me.  minor nits below.

> =========================================
> error_page = '/404.html'
> base_url = 'http://churchsite.gatheringspot.com'
> base_directory = '/home/gatheri/public_html/churchsite/'
> extension_list = ['.html','.shtml','.htm','.txt','.php','.asp']
>
> import string
> import os
>
> lowercase_url = string.lower(os.environ['REDIRECT_URL'])
> url_path,url_extension = os.path.splitext(lowercase_url)
> new_url = base_url+error_page

to make the program a bit more robust, you might wish to
use os.path.join and urlparse.urljoin instead of "+" when
building filenames/urls.

> #
> # Check to see if simply converting the URL to lowercase corrects the
> problem.
> # If it doesn't, check to see if the extension is incorrect and fix it
> if it is.
> #
> if os.path.exists(base_directory+lowercase_url):
>     new_url = base_url+lowercase_url

if you like source code optimizations, you could prepend an empty
extension to the extensions list and get rid of the if clause:

      for extension in [''] + extension_list:

> else:
>     for extension in extension_list:
>         if os.path.exists(base_directory+url_path+extension):
>             new_url = base_url+url_path+extension

you should probably add a break here, to get out of the loop once
you've found an existing file.

for extra python-style points, you can use the for-else construct
to set new_url to the error page URL only if you don't find a file:

    filename = os.path.join(base_directory, url_path)
    for extension in [''] + extension_list:
            if os.path.exists(filename + extension):
                new_url = urlparse.urljoin(base_url, url_path + extension)
                break # this skips the else clause
    else:
        new_url = urlparse.urljoin(base_url, error_page)

> #
> # Send HTTP header information to the browser so that it can request the
> "correct" document.
> #
> print "Status: 301 Permanently moved"
> print "Location: " + new_url

or:

    print "Location:", new_url

> print

Cheers /F

<!-- (the eff-bot guide to) the standard python library:
http://www.pythonware.com/people/fredrik/librarybook.htm
-->




From stephan at proceryon.at  Tue Mar 27 06:53:37 2001
From: stephan at proceryon.at (Stephan Herschel)
Date: Tue, 27 Mar 2001 13:53:37 +0200
Subject: Implicit Type Conversion buffer -> string
Message-ID: <3AC07F41.766F0787@proceryon.at>

Hi,

Have a look at this example about use of buffers. Can someone explain?

stephan at tilos:~ > python
Python 1.5.2 (#1, Jul 29 2000, 14:28:37)  [GCC 2.95.2 19991024
(release)] on linux2
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> a='abcdefghi'
>>> aB=buffer(a)
>>> b='ABCDEFGHI'
>>> bB=buffer(b)
>>> print aB
abcdefghi
>>> type(aB)
<type 'buffer'>
>>> print aB[1:-1]
bcdefgh
>>> type(aB[1:-1])
<type 'string'>
>>> print aB[0:len(aB)]
abcdefghi
>>> type(aB[0:len(aB)])
<type 'buffer'>
###############################################
this is interesting: when slicing a buffer from beginning to end no type
conversion is happening
###############################################
>>> print a+b
abcdefghiABCDEFGHI
>>> print aB+bB
abcdefghiABCDEFGHI
>>> type(aB+bB)
<type 'string'>
>>> print a+bB
Traceback (innermost last):
  File "<stdin>", line 1, in ?
TypeError: illegal argument type for built-in operation
###############################################
this is strange: type conversion is happening when concatenating 2
buffers,
but why is it leading to a type error when concatenating a string with a
buffer??
###############################################
   
regards,
stephan


From aleaxit at yahoo.com  Thu Mar 15 07:30:10 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 15 Mar 2001 13:30:10 +0100
Subject: Who am I: can a class instance determine its own name?
References: <mailman.984607702.5236.python-list@python.org>
Message-ID: <98qckb02eva@news1.newsguy.com>

"Tim Peters" <tim.one at home.com> wrote in message
news:mailman.984607702.5236.python-list at python.org...
> [Emile van Sebille]
> > Has locals changed so that it may be written to reliably?  My
> > documentation still warns:
> >
> > locals ()
> > Return a dictionary representing the current local symbol table.
Warning:
> > The contents of this dictionary should not be modified; changes may not
> > affect the values of local variables used by the interpreter.
>
> No, this has not changed.  As the docs say, writing to locals() is not
> allowed.
>
> and-more-likely-than-not-will-be-enforced-someday-too-ly y'rs  - tim

It *WOULD* be nice if all dictionaries had a "read-only" switch one
could throw to ensure no further modifications were accepted -- this
would make it trivial to enforce the semantics of 'locals' AND be
of similar use in similar application-level scenarios too.

(My dream is actually of a collection of bits in an attribute...:
    bit 1    no further modification/deletion of entries corresponding
                to already-existing keys
    bit 2    no addition of entries for keys that are not already present
    bit 4    no more changes to this attribute itself
and if all three were set, a dictionary might then become hashable,
and the world would suddenly be wonderful...:-).


Alex





From aleaxit at yahoo.com  Tue Mar  6 04:16:55 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Tue, 6 Mar 2001 10:16:55 +0100
Subject: I come to praise .join, not to bury it...
References: <mailman.983756766.27347.python-list@python.org> <97vnu3076q@news2.newsguy.com> <slrn9a89dk.2rm.hzhu@rocket.knowledgetrack.com> <DNXo6.24266$1D5.975603@e420r-atl1.usenetserver.com>
Message-ID: <9829u70spn@news1.newsguy.com>

"Steve Holden" <sholden at holdenweb.com> wrote in message
news:DNXo6.24266$1D5.975603 at e420r-atl1.usenetserver.com...
    [snip]
> > 2. Can we really join all kinds of lists?
> >
> No. I was somewhat startled to see that a UserList of integers could not
be
> given as an argument to join(). Relief of a kind arrived when I realised
> that the same was true of lists themselves.
    [snip]
> It would seem reasonable to expect the join() method to try and coerce
> things to lists, but maybe I'm not a reliable guide to what's reasonable.

It's not an issue of lists vs other stuff -- .join accepts as
its argument any sequence which defines a length, which is
reasonable (although it might be nice to remove that need for
the length being defined -- the sequence-length is only used
to control a for-loop, after all, so the test-for-IndexError
might suffice; but I guess unbounded-sequences might prove to
be a problem here).

The key issue, anyway, is with the _items_ in the sequence,
rather than with the sequence itself.  The join method on
a single-byte string wants each item to *BE* a single-byte
string -- peculiar behavior follows if it isn't, consider:

class Sequence:
    def __init__(self, N, value):
        self.N = N
        self.value = value
    def __len__(self):
        return self.N
    def __getitem__(self, index):
        if index>=self.N:
            raise IndexError, index
        return self.value

>>> sb=Sequence(3,'aa')
>>> 'x'.join(sb)
'aaxaaxaa'
>>> u'y'.join(sb)
u'aayaayaa'
>>> uc=Sequence(3,u'bbb')
>>> u'z'.join(uc)
u'bbbzbbbzbbb'
>>> 't'.join(uc)
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
TypeError: coercing to Unicode: need string or buffer, tuple found

and a Control-Z now crashes/hangs the Python interpreter (2.0, on
Win32).  OK, some kind of ugly bug in 2.0 (gotta get 2.1b1 and
check if the bug is still there...!!!).


Alex





From danielk at aracnet.com  Sun Mar 25 12:23:57 2001
From: danielk at aracnet.com (Daniel Klein)
Date: Sun, 25 Mar 2001 09:23:57 -0800
Subject: Coding choices
Message-ID: <2basbt4oe4tkf6hasrlmtmnr5jvbtoq2mm@4ax.com>

I've seen Python coded in both of these manners. Is there some advantage of one
over the other? Or is it just preference?

if condition:
    return something
raise SomeException

if condition:
    return something
else:
    raise SomeException

Thanks,
Daniel Klein

"Earth is full. Go home!"


From jdgorman at att.net  Sat Mar 24 22:41:00 2001
From: jdgorman at att.net (John D. Gorman)
Date: Sun, 25 Mar 2001 03:41:00 GMT
Subject: Python script to create HTML gallery, thumbnails
References: <3ABCAC94.B095CB9A@ozemail.com.au>
Message-ID: <3ABD68C6.AC7A844D@att.net>

For what it's worth, Jeff Kunce wrote a similar utility for image
resizing that included a simple gui. You might find some other useful
thngs in his "image squeezer" at:
http://starship.python.net/crew/jjkunce/ . This also uses PIL.

--John Gorman

ccang wrote:

> Hi there all,
> I recently got a digital camera (Kodak DC4800) and wanted to
> put all the crappy photos I shoot on the web.
>
> I looked for software to create thumbnails and corresponding
> HTML files but couldnt find any so I wrote my own using Python
> and the PIL.
>
> It creates a consolidated thumbnail and the corresponding HTML files.
>
> No GUI yet so u'll have to edit the script to customize
> your settings eg page headers, comments.
>
> # Does only JPGs for now.
> # Usage : python mkgallery.py [ filenames ]
> #   Will produce 1 HTML file for the thumbnail index.
> #                1 HTML file per JPG image.
> #                1 consolidated thumbnail JPG.
> #
> # There will be GUI for this pretty soon.
> #
> # The HTML file for the thumbnail is generated in the same
> # directory as where the script is run from.
> # However, it assumes that it will reside
> # in a subdirectory above where the other files (HTML, JPGS) reside.
> # ie u'll have to move it.
> #
> # eg
> #   myGallery/
> #   myGallery/topic.html
> #   myGallery/topic/topic_index.jpg
> #   myGallery/topic/zpic1.jpg
> #   myGallery/topic/zpic1.html
> #       ...
> #       ...
> #   myGallery/topic/zpicN.jpg
> #   myGallery/topic/zpicN.html
> #
>
> http://www.ozemail.com.au/~calcium/scripts.html
>
> Sample output
> http://www.ozemail.com.au/~calcium/rogaine.html
>
> Feedback welcome,
> Chai
> --
> | mailto:calcium at ozemail.com.au   | +61 401 688 408
> |  http://www.ozemail.com.au/~calcium
> |  I dont use the voice of Bart when I'm making love
> |  to my husband, but Marge's voice turns him on a
> |  little - Nancy Cartwright.



From zeitlin at seth.lpthe.jussieu.fr  Wed Mar 14 12:53:40 2001
From: zeitlin at seth.lpthe.jussieu.fr (Vadim Zeitlin)
Date: 14 Mar 2001 17:53:40 GMT
Subject: wxPython UI refresh
References: <98jku0$ifi$1@newstest.laserlink.net>
Message-ID: <slrn9avc13.p8n.zeitlin@seth.lpthe.jussieu.fr>

On Mon, 12 Mar 2001 17:17:49 -0600, Wiwih Gunadi <wcanopy at covad.net> wrote:
>I've tried Refresh(), Validate, wxYield() on the OnSize event handler to no
>avail.

 Try calling Win32 ::UpdateWindow() directly on the embedded window (assuming
you can find its HWND somehow) and/or sending it WM_PAINT explicitly (again,
you need the HWND).

>BTW, Excel is not an ActiveX control is it? i.e. it's impossible to "wrap"
>Excel like IE with the activexwrapper, right?

 Wrong, you can embed it, of course. Try putting an Excel table inside a Word
document and editing it.

 Regards,
VZ

-- 
GCS/GM d? H+ s++:-- p2 au--- a- w+ v C+++ UBLS+++ P- L++ N++ E--- W++++ M? V--
-po+ R++ G`` !tv b+++ D--- e++++ u++ h--- f+ r++ n- y?


From echuck at mindspring.com  Thu Mar  1 21:03:13 2001
From: echuck at mindspring.com (Chuck Esterbrook)
Date: Thu, 01 Mar 2001 21:03:13 -0500
Subject: [ANNOUNCE] Webware for Python 0.5
Message-ID: <5.0.2.1.0.20010301205419.047bfd90@mail.mindspring.com>

"Don Tuttle" <tuttledon at hotmail.com> wrote in message 
news:<fFCn6.13480$Z8.1902263 at typhoon.southeast.rr.com>...
>"Chuck Esterbrook"
> >    * Python Server Pages (PSP, similar to ASP, PHP and JSP)
>I 'm trying to get a handle on why I would want to use PSP on WinNt with
>IIS.  What's the difference/advantage to using PSP compared to using ASP
>with embedded Python script?
>Don

[1] Webware's PSP offers more syntax options than ASP. If I remember 
correctly, using Python in ASP involves some funkiness because of the clash 
between Pythonic indentation and HTML. I don't remember the details, but 
I'm positive they are chronicled in the archives of c.l.p.

PSP also allows you to inherit one or more base classes which can be good 
for reusing code. I don't know if ASP lets you do that or not.

In any case, if you check the docs for PSP, you can see what various 
options you have compared to ASP:

http://webware.sourceforge.net/Webware/PSP/Docs/


[2] Webware's PSP sits on top of WebKit, which means that when you use it, 
you can also create plain old Python servlets. You can also make your own 
"servlet factories". For example, you can write a handler that serves up 
*.sql, *.mytemplate, etc. There are also other components you can use with 
WebKit such as UserKit, MiddleKit, etc.  Of course, that doesn't matter if 
you're not interested in these options, although using Webware puts you in 
a position to use them any time later.


[3] A colleague of mine has had "caching" problems in ASP regarding out of 
date code residing in memory despite file modifications and reload()s. I 
can't say I know the details, and I won't say that you are guaranteed to 
suffer this if you use ASP+Python.


Hope that helps. More details might be available from people who have 
extensively used both.

-Chuck




From malte.forkel at berlin.de  Fri Mar 30 01:52:09 2001
From: malte.forkel at berlin.de (Malte Forkel)
Date: Fri, 30 Mar 2001 08:52:09 +0200
Subject: Python 2.0 vs 68k Mac
Message-ID: <9a1agl$o55$2@news.cs.tu-berlin.de>

I'm looking for Python 2.0 for 68k Macs. Has anyone build such a beast?
Please cc: your reply to me (malte.forkel at berlin.de).

Thanks, Malte

----------------------------------------------------------------------------
Malte Forkel                                             Tel +49 30 21018926
Kurfuerstenstrasse 127                                   Fax +49 30 21018927
10785 Berlin
Germany                                        mailto:malte.forkel at berlin.de




From anthony at interlink.com.au  Mon Mar 26 09:49:29 2001
From: anthony at interlink.com.au (Anthony Baxter)
Date: Mon, 26 Mar 2001 14:49:29 +0000
Subject: SNMPY now working against net-snmp 4.2, python 2.0
Message-ID: <200103261449.AAA04047@mbuna.arbhome.com.au>

I've mostly completed the port of SNMPY to use 
net-snmp 4.2. It also builds with Python 2.0, 
and distutils.

Code is currently available via sourceforge CVS, 
from http://sourceforge.net/cvs/?group_id=5313

Once I'm a bit more comfortable with it's stability, 
I'll make a new release available.

Testing and feedback welcome.

--
Anthony Baxter     <anthony at interlink.com.au>
It's never too late to have a happy childhood.



From michele at diee.unica.it  Sun Mar 11 08:40:46 2001
From: michele at diee.unica.it (Michele Marchesi)
Date: Sun, 11 Mar 2001 14:40:46 +0100
Subject: eXtreme Programming Conference XP2001
Message-ID: <3AAB805E.A6134FD5@diee.unica.it>

                  XP2001 - http://www.xp2001.org/

         2nd Int.l Conference on XP and Flexible Processes 
                     in Software Engineering

          May 20-23, 2001 - Villasimius, Sardinia, Italy

     *** THE PRELIMINARY PROGRAM OF XP2001 IS NOW ONLINE!!! ***
         *** THE REGISTRATION PROCEDURE IS NOW ACTIVE!! ***

The Conference schedule has been extended to accomodate all 
WORKSHOPS, Tutorial, Activity Sessions, Panels, Presentations!

XP2001 will give an unprecedented insight on Extreme Programming
and other lightweight techniques for software development!!

We remember that the Calls for Practitioner's Reports and
Demonstrations ARE STILL OPEN! (Deadline is April, 1, 2001.) See:
   http://www.xp2001.org/schedule.html#schedule

XP2001 WORKSHOPS are open for enrollment:
- Extreme Programming, revisiting the LEGO brick road.
- Agile Modeling (AM): A Discussion of Principles and Practices.
- Customer Involvement.
- Experience Exchange.
- XP and Requirements Negotiation.

For more information: http://www.xp2001.org/program.html
----------------------------------------------------------------
Apologize if you receive this announcement more than once.


From aahz at panix.com  Fri Mar 23 14:53:19 2001
From: aahz at panix.com (Aahz Maruch)
Date: 23 Mar 2001 11:53:19 -0800
Subject: TimeoutSocket exceptions handling
References: <mailman.985371211.10506.python-list@python.org>
Message-ID: <99g9jf$qp8$1@panix6.panix.com>

In article <mailman.985371211.10506.python-list at python.org>,
Bill Bell <bill-bell at bill-bell.hamilton.on.ca> wrote:
>
>However, not so: Python says "NameError: There is no variable 
>named 'Timeout'" for my 'except' statement.

Try TimeoutSocket.Timeout
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

"I won't accept a model of the universe in which free will, omniscient
gods, and atheism are simultaneously true."  -- M


From wcuddy at lserv.ja10629.home  Wed Mar 28 19:24:39 2001
From: wcuddy at lserv.ja10629.home (Wayne Cuddy)
Date: Thu, 29 Mar 2001 00:24:39 GMT
Subject: setvbuf() ??
Message-ID: <slrn9c50mp.me7.wcuddy@lserv.ja10629.home>

I need to print a period on an xterm every few seconds to provide status to the
user, I don't want to force a newline.  How do I turn off line buffering for
standard out?



From paul.moore at uk.origin-it.com  Wed Mar 14 11:39:18 2001
From: paul.moore at uk.origin-it.com (Paul Moore)
Date: Wed, 14 Mar 2001 17:39:18 +0100
Subject: windows script host and python
References: <%tKr6.2570$pX.26174@news1.onlynews.com>
Message-ID: <=Z6vOm2BLX=EI19YpBrnMFBLqw5l@4ax.com>

On Wed, 14 Mar 2001 14:36:59 +0100, "Mike Carifio"
<carifio.nospam at nospam.usys.com> wrote:

>I'm looking for some resources on using python with wsh generally.
>
>Specifically, I'm trying to use .wsf files with python. For example,
>given a "one liner" in (say) hello.py:
>
>    WScript.Echo("hello")
>
>then wscript //E:python hello.py works. But hello.wsf below doesn't:
>
><?xml version="1.0" ?>
><?job error="true" debug="true" ?>
>
><package id="package1">
>   <job id="job1"> <!-- run by default -->
>     <comment>
>       blah blah
>     </comment>
>     <script language="python">
>     <![CDATA[
>         WScript.Echo("hello")

  ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

Don't indent this line. Remember whitespace is significant in Python. Put it
at the left margin, and it works fine (at least it does for me!)

Paul.



From tonetheman at hotmail.com  Thu Mar 22 00:21:32 2001
From: tonetheman at hotmail.com (Tonetheman)
Date: Thu, 22 Mar 2001 05:21:32 GMT
Subject: Dynamically reloading extension modules
References: <007f01c0b252$852bb370$04090a0a@upcast.com> <20010321145901.M24341@trufflehunter.avalongroup.net> <00e201c0b25f$b95d4a70$04090a0a@upcast.com> <20010321154600.Q24341@trufflehunter.avalongroup.net> <mailman.985218751.8244.python-list@python.org>
Message-ID: <wZfu6.103264$lk6.8537193@typhoon.midsouth.rr.com>

It looks like you should be able to do this.

You should change the file dynload_win.c to look more like the
dynload_beos.c Whoever wrote that module saved the HMODULE (or whatever it
is called in Beos) in a dictionary. Then you can FreeLibrary when you get
unloaded...

Just an idea...


"Ken Seehof" <kens at sightreader.com> wrote in message
news:mailman.985218751.8244.python-list at python.org...
>
> > On Wed, Mar 21, 2001 at 03:36:15PM -0800, Ken Seehof wrote:
> > > Not exactly.  I didn't state the problem very clearly.  The
> > > problem is that while an extension module is loaded, the
> > > module file (pyd on windows) is read-only.
> > >
> > > At least this is true on windows.  On unix, I think maybe you
> > > can rename the module file, rebuild it, and do a reload(module).
> > > Is this true?  Windows doesn't allow you to rename or rewrite
> > > a dll while it is in use, so something like FreeLibrary would be
> > > necessary.
> >
> > Sorry, I can't answer Windows questions. I haven't had to run
> > Windows in years.
> >
> > But I would still think that reload() should work as long as
> > you are using python modules, *not* c/c++ modules.
>
> I -am- talking about c/c++ modules (which I have been referring
> to as "extension" modules).  Hence there's a bit more to it than
> reload().
>
> I want to recompile and reload a c/c++ extension module from
> within a continuously running python program.
>
> > Beazley has no disclaimers about reload working only on
> > specific platforms.
> >
> > There is a caveat, objects created with the old module won't be
> > updated.
>
>
>




From seandc at att.net  Thu Mar 15 16:15:30 2001
From: seandc at att.net (Sean)
Date: Thu, 15 Mar 2001 21:15:30 GMT
Subject: wxPython/Tkinter/pyGTK
References: <3AB11488.3010006@Antefacto.com>
Message-ID: <3AB1304D.B73714D3@att.net>

Take a look at Python Megawigets....they take a lot
of the pain out of using Tkinter:
    http://pmw.sourceforge.net/

Both Tkinter and Megawidgets are *really*well* described
in John Grayson's book Python and Tkinter Programming.

Sean
====
Padraig Brady wrote:
> 
> Hi,
> I'm just starting to use python, and my first
> app will have the following widgets:
> 
> Tabbed interface, with tabs in horizontal position.
> toolbar.
> treeview.
> listview (with embedded checkboxes)
> the odd button.
> 
> like:
> 
> +--------------------------+
> | MyApp                    |
> +--------------------------+
> |       toolbar            |
> |--------------------------|
> | tab1 +-----------------+ |
> | tab2 |  treeview &/or  | |
> | ..   |  listview       | |
> | tabn +-----------------+ |
> +--------------------------+
> 
> So
> 
> 1. I'm wondering what to use?
>     wxPython, Tkinter, PMW+Tkinter or pyGTK
> 
> 2. Is there any documents comparing
>     these options.
> 
> 3. Are there other options I haven't considered.
> 
> 4. What option do people consider as having
>     the least dependencies (easiest to install),
>     as this app is GPL'd and must be easy to install.
> 
> Thanks,
> Padraig.


From robin at stop.spam.alldunn.com  Fri Mar 16 17:26:34 2001
From: robin at stop.spam.alldunn.com (Robin Dunn)
Date: Fri, 16 Mar 2001 14:26:34 -0800
Subject: socket.ssl for win32
References: <mailman.984773016.25714.python-list@python.org>
Message-ID: <Jqws6.22$quX.1114625@news.randori.com>

"Robin Dunn" <robin at alldunn.com> wrote in message
news:mailman.984773016.25714.python-list at python.org...
> Does anybody have a pre-built _socket.pyd (and a _socket_d.pyd too if
> possible) for Python 2.0 that includes the OpenSSL support?
>

I got impatient and just did it myself.  (Okay, I only waited about 5
minutes...)  I've put the binaries at http://alldunn.com/python/ in case
anyone else wants them.


--
Robin Dunn
Software Craftsman
robin at AllDunn.com       Java give you jitters?
http://wxPython.org      Relax with wxPython!





From lux_mundt at t-online.de  Thu Mar  8 17:13:04 2001
From: lux_mundt at t-online.de (Thomas Mundt)
Date: Thu, 8 Mar 2001 23:13:04 +0100
Subject: Database with Python ??
Message-ID: <988vmn$t77$05$1@news.t-online.com>

Has Python a own database language and do someone know a link to a Tutorial
in database-working with python?

Thanks
Thomas




From tismer at tismer.com  Wed Mar 28 06:29:12 2001
From: tismer at tismer.com (Christian Tismer)
Date: Wed, 28 Mar 2001 13:29:12 +0200
Subject: Name of a method of an instance
References: <3AC1BE92.ACE82741@ua.ac.be>
Message-ID: <3AC1CB08.CA169C59@tismer.com>


Richard Philips wrote:
> 
> A question on introspection:
> 
> class Log:
> 
>     def __init__(self):
>         self.previous = None
> 
>     def methodA(self):
>         self.previous = "methodA"
> 
>     def methodB(self):
>         self.previous = "methodB"
> 
> myinstance = Log()
> myinstance.methodA()
> myinstance.methodB()
> 
> How can I calculate the name of the method from within the method itself

<module name="funcname.py">
# obtaining the name of a function/method
# Christian Tismer
# March 2001

# the following function might be implemented in the
# sys module, when generators are introduced.
# For now, let's emulate it. The way is quite inefficient
# since it uses an exception, but it works.

import sys

def _getframe(level=0):
    try:
        1/0
    except:
        import sys
        tb = sys.exc_info()[-1]
    frame = tb.tb_frame
    while level >= 0:
        frame = frame.f_back
        level = level - 1
    return frame

sys._getframe = _getframe
del _getframe

# we now assume that we have sys._getframe

def funcname():
    return sys._getframe(1).f_code.co_name

class Log:

    def __init__(self):
        self.previous = None

    def methodA(self):
        self.previous = funcname()

    def methodB(self):
        self.previous = funcname()

myinstance = Log()
myinstance.methodA()
print myinstance.previous
myinstance.methodB()
print myinstance.previous

printout="""
methodA
methodB
"""
# that's all folks
</module>

ciao - chris

-- 
Christian Tismer             :^)   <mailto:tismer at tismer.com>
Mission Impossible 5oftware  :     Have a break! Take a ride on Python's
Kaunstr. 26                  :    *Starship* http://starship.python.net/
14163 Berlin                 :     PGP key -> http://wwwkeys.pgp.net/
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
     where do you want to jump today?   http://www.stackless.com/



From bill_tolbert at bigfoot.com  Wed Mar  7 14:36:53 2001
From: bill_tolbert at bigfoot.com (bill_tolbert at bigfoot.com)
Date: Wed, 07 Mar 2001 19:36:53 -0000
Subject: cursor.execute fails under mxODBC
Message-ID: <9862kl+69cq@eGroups.com>

I just upgraded to the latest version of mxODBC. Calling execute on a 
cursor object is hosing python. This code worked perfectly under 1.5.1 
and version 1.1.1 of mxODBC.

Having trouble on both Win98 and Win2000.

I can stay on windows, so I'm gonna give ADO a try, but I'd still like 
to know what's going on.

Thanks,

Bill




From mcfarlan at neca.com  Sat Mar  3 16:37:30 2001
From: mcfarlan at neca.com (D. Michael McFarland)
Date: Sat, 03 Mar 2001 21:37:30 GMT
Subject: gah! I hate the new string syntax
References: <mailman.983615170.23313.python-list@python.org> <tN5o6.2746$eH3.14913@NewsReader> <Koco6.7518$1D5.244602@e420r-atl1.usenetserver.com>
Message-ID: <m3pufyedhg.fsf@katrine.mcfhome>

"Steve Holden" <sholden at holdenweb.com> writes:

> Lest people might think this is intended as criticism rather than humo(u)r,
> I hasten to add that I lived in Scotland for three years before moving to
> the USA, and found the Scots to be extraordinarily warm and generous people.

Well, sure, after my ancestors left at sword-point.  We go back for a
visit from time to time, though, so be careful.

> So, for the record, they don't get their mathmatics degrees until fourth
> grade in Scotland.

Ah, maybe that's why they left---impatience with the educational
system.  I've never believed great-great-granddad would have backed
down from a mere pointy stick.

This is the most fun I've had with Python strings since the Great
Objectification Debates began.

Cheerfully yours,
Michael

-- 
    D. Michael McFarland   <mcfarlan at caber-eng.com>
    Caber Engineering, Inc., Storrs, Connecticut


From stephen_purcell at yahoo.com  Wed Mar 28 05:00:51 2001
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Wed, 28 Mar 2001 12:00:51 +0200
Subject: Jython and JPython Questions...How similar to Python, Tkinter?
In-Reply-To: <20010328114505.E8325@freedom.puma-ag.com>; from stephen_purcell@yahoo.com on Wed, Mar 28, 2001 at 11:45:05AM +0200
References: <3ABFF14A.30D0096D@earthlink.net> <20010328114505.E8325@freedom.puma-ag.com>
Message-ID: <20010328120051.I8325@freedom.puma-ag.com>

Steve Purcell wrote:
> Now if someone ever comes up with a library that gives Swing/AWT a Python
> interface, that would be another story...

That should read 'a Tkinter' interface, obviously. Need more coffee...

-Steve

-- 
Steve Purcell, Pythangelist
Get testing at http://pyunit.sourceforge.net/
Any opinions expressed herein are my own and not necessarily those of Yahoo



From aleaxit at yahoo.com  Tue Mar 20 01:45:07 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Tue, 20 Mar 2001 07:45:07 +0100
Subject: How to increase PythonWin v2.0: COM browser limits?
References: <994psh$so5$1@news1.xs4all.nl> <995sgi$7u0$04$1@news.t-online.com> <996flh$e9m$1@news1.xs4all.nl>
Message-ID: <996ucp012ha@news2.newsguy.com>

"Twan van der Schoot" <twanvds at xs4all.nl> wrote in message
news:996flh$e9m$1 at news1.xs4all.nl...
> thanks,
>
> But I think it is strange that nobody else seems to have the problem,
wonder
> if anybody uses the COM facilities in Pythonwin.

I do, but I've switched to using makepy directly more often than the 'COM
browser'.  Having a zillion COM servers installed tends to give problems to
all utility programs that try to grep throughout the registry for registered
COM servers -- e.g., Microsoft's excellent "OleView" (comes with VC++ and
can also be downloaded for free in the platform SDK) regularly gives me
runtime errors when trying to list "all" servers of certain kinds.  Maybe I
should procure and regularly run registry-cleaning programs...!-)


Alex





From flognat at flognat.myip.org  Thu Mar 15 17:09:01 2001
From: flognat at flognat.myip.org (Andrew Markebo)
Date: Thu, 15 Mar 2001 22:09:01 GMT
Subject: Should I close after popen??
Message-ID: <m38zm6r8ih.fsf@master.athome>

I am hacking away a small background program, using popen to call
commands, should I close the file in some way when done or it is taken
care of automagically??

I do something like this once every 10 mins..

        fd1=os.popen("command")
        fd2=os.popen("command")
        os.wait(fd1)
        os.wait(fd2)
        values=fd1.read()+fd2.read()

And.. anything to worry about, nesting popen/wait like this??
(Linux/Solaris)

        /Andy


From aleaxit at yahoo.com  Tue Mar  6 05:02:12 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Tue, 6 Mar 2001 11:02:12 +0100
Subject: polymorphjsm &c (was Re: I come to praise .join, not to bury it...)
References: <mailman.983485923.1136.python-list@python.org> <97p7iq$kk2$1@nntp6.u.washington.edu> <97rka002k9@news1.newsguy.com> <9811a8$k5k$1@nntp6.u.washington.edu>
Message-ID: <982cja0v5c@news1.newsguy.com>

"Russell E. Owen" <owen at astrono.junkwashington.emu> wrote in message
news:9811a8$k5k$1 at nntp6.u.washington.edu...
    [snip]
> > (long discussion about polymorphism omitted).
>
> I'm afraid I don't understand where the polymorphism comes in, so I
> cannot comment on that.

If you don't understand polymorphism in general, then I don't
see how you can comment on any OO design.

If you're instead saying that you don't see where polymorphism
can help for this _specific_ case, what's so mysterious about
the specific example I gave?  Here it is again, more or less,
with a bit more fluff around it:

good_things = (
    'money', 'bridge', 'champagne',
    'wine', 'women', 'song'
    )
def joined_niceties(things, joiner=', '):
    nice_things = [thing for thing in things if thing in good_things]
    return joiner.join(nice_things)

>>> words = 'programming money bridge spam champagne'.split()
>>> print joined_niceties(words)
money, bridge, champagne
>>> print joined_niceties(words, ' - ')
money - bridge - champagne

class MixedJoiner:
    def __init__(self, joiner1, joiner2):
        self.joiner1 = joiner1
        self.joiner2 = joiner2
    def join(self, sequence):
        return joiner2.join((
            joiner1.join(sequence[:-1]),
            sequence[-1]))

>>> print joined_niceties(words, MixedJoiner(', ', ' and '))
money, bridge and champagne


We have some code (here, function joined_niceties) that somehow
prepares or obtains a sequence (here, nice_things, prepared by
selection from an argument to the function via membership in a
global-data list) and must return the string obtained by joining
said sequence with an arbitrary joiner (here, argument joiner,
defaulting to ", " -- comma-then-blank, often a decent way to
join words into a string).

Thanks to the fully general polymorphism that joiner.join affords,
we can then use a *special* joiner-object to do fancy joining.

This imposes NO technical cost whatsoever on the code that does
the joining (here, function joined_niceties) -- indeed, such code
can have been written months ago, before we even had an inkling
that the joiner-object would ever be other than a string.  This is
the common magic of polymorphism -- code can be closed to changes
yet open for modification (here, this applies to joined_niceties --
we need not change it to obtain fancy-joining, it all comes from
free thanks to the polymorphism on the joiner-object!).

So, we code a suitable joiner-object -- here, one that conjoins
two other joiners, using one for all but the last join, the other
for the last join only.  This use, of course, easily generalizes:

>>> commas_then_and = MixedJoiner(", ", " and ")
>>> commas_and_even = MixedJoiner(commas_then_and, " or even ")
>>> words = 'women money work bridge beans wine champagne spam'.split()
>>> print joined_niceties(words, commas_and_even)
women, money, bridge and wine or even champagne

without any further complexity, all through the magic of general
polymorphism.  Polymorphism is what's _truly_ great about object
oriented programming (and can also be obtained via some other
paradigms, such as generic-programming, with different costs and
advantages in various programming languages) -- getting the RIGHT
behaviour for a given specific case, from GENERAL client-code that
just appropriately asks one or more objects to "do the right
thing for purpose X" (by calling the appropriate methods with
the right arguments).


> >Say that I'm implementing a sequence-like object.  Would
> >having .join as a part of the set of methods I must write
> >be an _advantage_ to me?  As things stand now, to "be a
    [snip]
> I'm afraid we disagree here. Collections can all inherit from a parent
> collection class, a perfectly reasonable requirement. You then define
> the few methods that are required to make your own kind of collection.
> The parent class should define things like join (which is, of course, a
> very simple method).

So, there is NO advantage in having .join as a part of my
set of methods -- there IS the cost that I have to inherit
from some base-class, which may be 'perfectly reasonable'
(or not) but still IS a cost I have to pay, compared with
the present situation where no inheritance is REQUIRED.

For example, if I want to implement a sequence-like object
in a C-coded extension to Python (hardly an unreasonable
need!), then .join, and any other convenience methods that
may be required to "be a sequence", are PURE OVERHEAD.  It
makes my programming task harder; and, *where are the
benefits* corresponding to this cost?!  I have seen no
technical benefits claimed yet for making .join a method
of the sequence-like object -- in what use cases would the
resulting polymorphism be a functional _benefit_, to have
the sequence behave differently when it is to be joined
rather than (say) when it's to be written to a file-like
object (via the .writelines method of the f-l obj) or
otherwise iterated upon?

There's also some issue here with 'Collection' (a more
general concept than 'Sequence' -- why should we constrain
ALL collections to be sequenceable?!), but let's let that
pass for the moment.

Mixin base-classes are a good way to add convenience
methods to my own objects through the design pattern
'template' (as named by the Gof4 -- no relation to such
things as C++ templates, etc).  We have no disagreement
on THAT -- *IF* a method (that is generally suitable for
implementation via 'template' dp) is appropriate, then
mixin inheritance is a good implementation approach.
We DO have disagreement that .join is at all appropriate
as a method of sequences -- let alone more general
collections!!!

Furthermore: PLEASE show us how you would implement
the 'very simple method' that you claim the base
class .join would be.  Be sure, at the very least, to
handle both joiners that are single-byte strings, and
ones that are unicode-strings, assuming of course they
have no 'join' methods themselves.  It is also very
important that your .join's performance characteristics
be *reasonable* -- O(N) in time and space, where N
is the length of the resulting string; O(N*N) would
be a major disaster, as I hope you'll agree (a design
that forces an O(N) algorithm to become O(N*N) IS an
intrinsic disaster -- no aesthetic justification can
make up for that...:-).


Alex





From gunturu_neelima at rediffmail.com  Tue Mar  6 02:29:59 2001
From: gunturu_neelima at rediffmail.com (neelima  g)
Date: 6 Mar 2001 07:29:59 -0000
Subject: (no subject)
Message-ID: <20010306072959.28300.qmail@mailFA9.rediffmail.com>

after creating the empty database,how to enter the values into it  which is having morethan two columns.
                       confirm 881077


_____________________________________________________
Chat with your friends as soon as they come online. Get Rediff Bol at
http://bol.rediff.com







From scarblac at pino.selwerd.nl  Mon Mar 19 20:00:50 2001
From: scarblac at pino.selwerd.nl (Remco Gerlich)
Date: 20 Mar 2001 01:00:50 GMT
Subject: Why "from __future__" stinks; a counter-offer
References: <mailman.984949523.1896.python-list@python.org> <4Aft6.2171$pP2.372173@paloalto-snr1.gtei.net> <mailman.985029625.6313.python-list@python.org> <995pl4$jta$1@news.mathworks.com>
Message-ID: <slrn9bdap3.ac9.scarblac@pino.selwerd.nl>

Joshua Marshall <jmarshal at mathworks.com> wrote in comp.lang.python:
> As responses to suggestions, there's a lot of "write up a PEP or it
> won't get implemented" going around.  I can understand this, but it's
> also important that good ideas don't go to waste just because the
> proposer _didn't_ write a PEP.

If it's such a great idea, *someone* will care enough to write a PEP for it.
If noone can be bothered to do even that, then maybe it's not that important
after all. It's not as if writing a PEP is weeks of work...

> I guess I just hope that the PEP-mechanism doesn't make it so that
> good proposals get overlooked in favor of proposals with enthusiastic
> proponents.

Maybe the PEP process is good at weeding out the endless Usenet advocacy
from the important stuff...


(disclaimer: when I give my opinion in most of these discussions it's
definitely in the Usenet advocacy section... But maybe one day I will think
something is really important and then I will write a PEP)

-- 
Remco Gerlich


From nospam at mega-nerd.com  Mon Mar 19 02:28:19 2001
From: nospam at mega-nerd.com (Erik de Castro Lopo)
Date: Mon, 19 Mar 2001 07:28:19 GMT
Subject: Tkinter : resizing Text widget with main window
Message-ID: <3AB5B50B.52257D09@mega-nerd.com>

Hi all,

I'm reposting this as the original question seemed to have been
lost (it didn't find its way to google).

How does one force a Tkinter.Text widget to be resized whenever 
the main window is resized?

TIA,
Erik
-- 
-----------------------------------------------------------------
Erik de Castro Lopo     nospam at mega-nerd.com      (Yes its valid)
-----------------------------------------------------------------
"If POSIX threads are a good thing, perhaps I don't want to know 
what they're better than." -- Rob Pike


From jsoons at juilliard.edu  Sat Mar 24 09:16:37 2001
From: jsoons at juilliard.edu (Jonathan Soons)
Date: Sat, 24 Mar 2001 09:16:37 -0500
Subject: Please rate these Python books
Message-ID: <B7C74E1A6FBAD411BFC60002B309DDF4E439@mailbox.juilliard.edu>

 
"Snap!" is not understood outside former Commonwealth countries.
It's use causes culture shock.


-----Original Message-----
From: Brett Haydon
To: python-list at python.org
Sent: 3/24/01 1:00 AM
Subject: Re: Please rate these Python books

Snap!
> Prime collection is currently "Learning Python" (despite its age),
"Win32
> Programming in Python" and "Python Essential Reference". The rest have
their
> place, but it's in the bookshelf, not the briefcase.

though haven't really used Python Essential Ref much..
Looking foward to the new OReilly Python book, though I fear fitting it
in the briefcase may be out of the question.

Brett


-- 
http://mail.python.org/mailman/listinfo/python-list



From aleaxit at yahoo.com  Fri Mar  9 18:17:01 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sat, 10 Mar 2001 00:17:01 +0100
Subject: Divorcing Command Line and GUI
References: <mailman.984160633.14724.python-list@python.org>
Message-ID: <98dpoi32ckn@news1.newsguy.com>

"Timothy Grant" <tjg at exceptionalminds.com> wrote in message
news:mailman.984160633.14724.python-list at python.org...
> Hello Everyone.
>
> This is something of a strange question, but please bear with
> me.
>
> I have a series of tools that I have created that work from the
> command line. I have now been tasked with creating GUIs for
> those tools. The problem I'm faced with is that the command
> line tools and the GUI tools will probably be licensed
> differently (e.g, command line will be Open Source, GUI will be
> something undetermined as of yet.) This means that I can't put
> my GUI code in with my CL code.
>
> My first cut at this demarcation sort of worked, but was also
> troublesome. So while I have some new ideas, I'm looking to see
> if anyone else in the python community has ever had to do
> something like this, and if there are any easily maintainable
> solutions to this problem.

Piece of cake -- just make sure your CL tools take a switch
(-u or --unbuffered) to work without any input buffers on
stdin/stdout, and similarly-unbuffered Python GUI front-ends
will be easily able to drive them (after an os.popen2).


Alex





From aleaxit at yahoo.com  Tue Mar 13 03:52:31 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Tue, 13 Mar 2001 09:52:31 +0100
Subject: learning C
References: <mailman.983972829.23481.python-list@python.org> <983975751.18833@newsmaster-04.atnet.at> <slrn9adklr.bv9.scarblac@pino.selwerd.nl> <984047775.592522@newsmaster-04.atnet.at> <9896hk$ig8$1@verence.demon.co.uk> <984134612.781779@newsmaster-04.atnet.at> <3AACE22D.3A9EEB12@bt.com> <j6kr6.68748$ML1.3282888@e420r-atl2.usenetserver.com>
Message-ID: <98kn4b0nkr@news2.newsguy.com>

"Steve Holden" <sholden at holdenweb.com> wrote in message
news:j6kr6.68748$ML1.3282888 at e420r-atl2.usenetserver.com...
    [snip]
> > > This is nothing more difficult than in C.
> >
> > Sorry but it is. There's just so much less lurking under
> > the hood in C.
> >
> But by this argument, assembly code is even better!  It's always a
question
> of selecting language features because they are appropriate to your
problem
> domain.

Like with political parties, you can't really pick and choose
features with programming languages -- they come as a package.

C++ is a very big package, which IS the essential problem with
it -- and you can't really subset it effectively.  Its designer
chose very challenging design-goals, and managed to meet them,
but simplicity and linearity of the resulting language was not
compatible with the large set of goals.

Modern-day assembly language also has a lot "lurking", for very
different detailed reasons (although the need to get 100% of the
potential performance is a "common-mode factor" to the complexity
of both today's machine-code and C++:-).  But, mostly, the issue
there is one of portability -- the finely tuned code you write
today will be badly mistuned (if it even works at all) tomorrow,
or even this afternoon if you need to run it on another box.  But
this is really a red-herring, and I think you know it -- maybe
mentioning assembly in a discussion of programming languages should
be considered to apply a variant of Godwin's rule...!-)

"A simple and easily predictable semantic model"; "potential for
100% machine-speed exploitation"; "convenient high-level usage";
pick any two.  C gives you 1 and 2, but definitely not 3; C++
gives you 2 and 3, but definitely not 1; Python gives you 1 and
3, but definitely not 2.  The choice among them will depend on
the relative importance of the three desirable goals.

In _most_ cases, programmer productivity is paramount, speed of
the resulting code isn't -- which is why Python dominates (or,
rather, _should_ dominate:-), as the two nice things it offers
are both highly conducive to such productivity.  If speed IS a
big issue (e.g., in low-level components, meant to be scripted
by Python of course:-), then one needs to evaluate whether the
convenience of C++, OR the simplicity of C, serves one's goals
best.  My opinion is that, *in security-critical code*, a
language that is too complex (does stuff behind your back,
is prone to buggy compilers, etc) may be too big a minus; that
10% of 10% that is crucial for both speed and security should
thus probably be best coded in C.  When security is not that
big an issue (you're worried about errors rather than deliberate
exploits), then the _ability_ of C++ to be used in a safer way
(exceptions, std::vector & std::auto_ptr rather than any use
of explicit memory allocation, etc) makes it a better choice.

But you still can't pick-and-choose feature.  And C++ has so
MANY foot-level tripwires that it almost seems that must have
been one of its design goals... although in fact it's clear
that compatibility with C is what caused most of them.  E.g.,

struct foop {
    foop();
    void flip(int i);
    // some other stuff here
};

void flipthefoops(foop fooparray[], int nFoops)
{
    for(int i=0; i<nFoops; ++i)
        fooparray[i].flip(i*23-42);
}

struct biggerfoop: public foop {
    int addafield;
};

void flip17biggerfoops()
{
    biggerfoop severalfoops[17];
    flipthefoops(severalfoops, 17);
}

and "bang you're dead" -- nor would recoding flipthefoops() in
a style closer to the standard library (with a begin and an end
parameter, rather than begin and number-of-elements) help in
the least.  So, OK, this is just "typical tripwire number 83 in
a cast of thousands" -- "inheritance (extension) and arrays do
NOT play together well", and in this toy case this is quite easy
to see.  But, IS it easy when it happens in a less overt way and
as one tiny part of a complicated software system?  Nope.

I'm still a C++ "paladin" (well... I _do_ consider it the least
of evils in most speed-needed circumstances -- guess that's not
much, as far as language advocacy goes:-) mainly because of the
splendid power that frameworks can give the C++ developer -- the
standard C++ library to start with, ATL if you're into COM, the
Boost Python Library (and all of the rest of Boost)... the lack
of templates (and, secondarily, OO) in C just doesn't let any
C-specific "frameworks" be ANYWHERE close to what is available
for C++.  But don't kid yourself -- it IS going to trip you up
(have LOTS and LOTS and LOTS of unit-tests...!!!-).


Alex





From tim.hochberg at ieee.org  Wed Mar  7 15:41:13 2001
From: tim.hochberg at ieee.org (Tim Hochberg)
Date: Wed, 07 Mar 2001 20:41:13 GMT
Subject: Name of the Script
References: <20010304.185354.2114937732.8277@localhost.localdomain> <NJCo6.6720$qv3.2834377@nnrp5-w.sbc.net> <FpmjOjvwe5bm0Bg09LDTMPhh5vji@4ax.com> <Pine.LNX.4.21.0103051715150.18263-100000@kev.ioc.ee> <cbKjOqPRwNza00jOm7A1HBDkRYj7@4ax.com> <upuft9z0s.fsf@ctwd0143.fitlinxx.com>
Message-ID: <J1xp6.6661$Xt3.1009487@news1.rdc1.az.home.com>

"David Bolen" <db3l at fitlinxx.com> wrote in message
news:upuft9z0s.fsf at ctwd0143.fitlinxx.com...
> Paul Moore <paul.moore at uk.origin-it.com> writes:
[SNIP]
> > Interestingly, although the __name__ symbol has the value __main__ in
> > a script being run as a "main program", the __file__ symbol doesn't
> > exist. That's a pity, as it would probably be what I wanted...
>
> Right - if you're being imported you can reference __file__, but a top
> level script really only has argv[0], I believe.
>
> But argv[0] always references the path used to start the script which,
> at the point of startup, should always be a path that you can use to
> reference the same location.

Another approach, if __file__ is what you want is to have a module import
itself and grab the __file__ from the imported copy. Here's a (contrived)
example:

# file test.py
if __name__ == "__main__":
    import test
    print `test.__file__`

Unfortunately, this whole __file__ approach fails when using files from an
archive (such as produced with Gordon McMillan's installer), so this isn't
always useful.

> I'd just suggest that if you need this sort of thing later that you
> identify your location at startup, and then make that information
> available in a standard way to the rest of the application independent
> of sys.argv.  And if you know you might be moving directories, store
> the os.path.abspath() of sys.argv[0] at startup rather than just
> sys.argv[0].

Good advice I think.

-tim




From fredrik at pythonware.com  Fri Mar 23 11:24:23 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 23 Mar 2001 16:24:23 GMT
Subject: tempfile problem on win98
References: <g3nu6.369$hb.2141@news1.belrs1.nsw.optushome.com.au> <99d0dg$os92$1@ID-11957.news.dfncis.de> <viJu6.473$hb.2279@news1.belrs1.nsw.optushome.com.au>
Message-ID: <XMKu6.3196$4N4.475465@newsc.telia.net>

Steven Adams wrote:
> is there a way to have a suffix included in the calculation of the
> filename instead of the prefix?

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

    mktemp([suffix])

    Return a unique temporary filename. This is an absolute
    pathname of a file that does not exist at the time the call
    is made. No two calls will return the same filename.

    suffix, if provided, is used as the last part of the generated
    file name. This can be used to provide a filename extension
    or other identifying information that may be useful on some
    platforms.

worth trying...

Cheers /F




From brian at sweetapp.com  Mon Mar 19 01:36:13 2001
From: brian at sweetapp.com (Brian Quinlan)
Date: Sun, 18 Mar 2001 22:36:13 -0800
Subject: stripping punctuation from tuples
In-Reply-To: <3AB57F0C.594CF3E9@northwestern.edu>
Message-ID: <HPEJKHHCMJHPMNLDAPLMGEHPCEAA.brian@sweetapp.com>

Tuples are sequence types, like lists, and you can use the same notation to
reference their elements.

Using your example, you might want to try the following:

>>> a[10][0]
'Alan Tsang'
>>> name, = a[10] # Another alternative
>>> name
'Alan Tsang'

-----Original Message-----
From: python-list-admin at python.org
[mailto:python-list-admin at python.org]On Behalf Of Louis Luangkesorn
Sent: Sunday, March 18, 2001 7:38 PM
To: python-list at python.org
Subject: stripping punctuation from tuples


How do you strip the punctuation from tuples?  In particular, say I have
a tuple a[10] (which happened to come from a not very well designed
database)

>>> a[10]
('Alan Tsang',)

How do I get something to say 'Alan Tsang'  with out the parens or the
','

Or, what I'm really trying to do is make this so
>>>b = breakname(a[10])
>>>b[0], b[1]

'Alan','Tsang'

I've tried to do:
>>> d = split(str(a[0]))
>>> d
["('Aaron", "Fan',)"]
>>> d[0]
"('Aaron"
>>> d[1]
"Fan',)"

But that is not quite what I want.
Many thanks.
Louis
--
K Louis Luangkesorn
lluang at northwestern.edu  http://pubweb.nwu.edu/~kll560  PGP:0xF3E2D362

Whatsoever things are true, ... honest, ... just, ... pure, ... lovely,
... of good report; if there be any virtue, and if there be any praise,
think on these things.
     - motto - Northwestern University (Phil 4:8)
--
http://mail.python.org/mailman/listinfo/python-list




From neelk at alum.mit.edu  Tue Mar 20 16:29:02 2001
From: neelk at alum.mit.edu (Neelakantan Krishnaswami)
Date: 20 Mar 2001 21:29:02 GMT
Subject: This math scares me
References: <mailman.985056328.725.python-list@python.org>
Message-ID: <slrn9bfmua.v02.neelk@alum.mit.edu>

On Mon, 19 Mar 2001 21:44:43 -0500, Tim Peters <tim.one at home.com> wrote:
>
> The original Cray Fortran compiler supported that gimmick "in
> reverse": a compile switch caused each (binary) fp result to get
> *truncated* to N bits (a simply masking operation).  Non-experts
> loved it: via plotting results versus number of bits, unstable
> algorithms often stuck out like a broken nose.

Whoa. What an awesomely elegant hack. That goes straight into the box
of tricks that will someday make me look extremely smart. Thanks!


Neel



From mda at idatar.com  Thu Mar 29 14:01:06 2001
From: mda at idatar.com (David Allen)
Date: Thu, 29 Mar 2001 19:01:06 GMT
Subject: how query function args
References: <3AC375A5.507C4D73@abaqus-sn.com>
Message-ID: <SDLw6.3062$j_3.1384803@typhoon2.ba-dsg.net>

In article <3AC375A5.507C4D73 at abaqus-sn.com>, "david winkler"
<winkler at abaqus-sn.com> wrote:

> I need to query the names and values of a function's keyword arguments. I
> suspect that there is an easy way to do this but cannot find it.  Would
> someone care to make a suggestion?

This may help - you can pass all arguments as a 
dictionary:

def some_function(**dict):
  for key in dict.keys():
    print "%s %s" % (key, dict[key])


some_function(foo='bar', baz='quux', spam='eggs')

this will print:

foo bar
baz quux
spam eggs

If you do argument passing in this way, you can queyr
the arguments all you want and even use keynames as
strings of course.  Is this what you mean?

-- 
David Allen
http://opop.nols.com/
----------------------------------------
When everything's coming your way, you're in the wrong lane.


From gerhard at lilith.hqd-internal  Fri Mar 16 08:42:13 2001
From: gerhard at lilith.hqd-internal (Gerhard HƤring)
Date: Fri, 16 Mar 2001 14:42:13 +0100 (CET)
Subject: Python and ADO
In-Reply-To: <mailman.984690501.20811.python-list@python.org>
References: <98r9qn$tbi$1@troll.powertech.no> <mailman.984690501.20811.python-list@python.org>
Message-ID: <20010316134213.022172CC1A@lilith.hqd-internal>

In comp.lang.python, you wrote:
>Syver Enstad wrote:
>
>> Has anyone considered making a DB API compliant wrapper around MS ADO? Or
>> isn't this considered worthwhile?
>>
>> As ADO is the database api that I use, I could have a look at making a
>> wrapper for it if there was any interest in it and if making such a thing is
>> not too involved.
>
>Michel Orengo already wrote a preliminary version of a DB-API wrapper for ADO
>  but I don't think he ever finished it.  I have a copy of it, but I have never
>  actually tried it.  If you can't find it in the newsgroup archives, drop me an
>  email and I'll send you a copy.

Could you please send me a copy, too? I had downloaded it when it was announced
on c.l.p., but I can no longer find it.

TIA

Gerhard
-- 
mail:   gerhard <at> bigfoot <dot> de
web:    http://highqualdev.com



From rapto at arrakis.es  Thu Mar 15 12:21:58 2001
From: rapto at arrakis.es (Marcos Sįnchez Provencio)
Date: Thu, 15 Mar 2001 18:21:58 +0100
Subject: Instant Hacking and Instant Python
References: <98oanu$4da$1@tyfon.itea.ntnu.no>
Message-ID: <newscache$vh19ag$ayh$1@proxy.bitmailer.com>

I'll review that (spanish translation)




From calcium at ozemail.com.au  Mon Mar 26 07:51:52 2001
From: calcium at ozemail.com.au (C.C. Ang)
Date: Mon, 26 Mar 2001 22:51:52 +1000
Subject: Python script to create HTML gallery, thumbnails
References: <3ABCAC94.B095CB9A@ozemail.com.au> <3abeae5c.18990177@news.itmel.bhp.com.au>
Message-ID: <3ABF3B68.D3049188@ozemail.com.au>

GraGra wrote:
> 
> Thumbs Plus is shareware from http://www.cerious.com/
> and does a great job managing images and creating thumbnails and
> linked html pages.

Hiya,
Thanks for the info. Looks quite good.
It looks like windows only.
Pity as I run Linux about 50% of the time.

Chai

-- 
| mailto:calcium at ozemail.com.au   | +61 401 688 408
|  http://www.ozemail.com.au/~calcium
|  The face of a child can say it all, especially the
|  mouth part of the face.


From paul.moore at uk.origin-it.com  Fri Mar  9 05:44:43 2001
From: paul.moore at uk.origin-it.com (Paul Moore)
Date: Fri, 9 Mar 2001 11:44:43 +0100 
Subject: Name of the Script
References: <20010304.185354.2114937732.8277@localhost.localdomain> <NJCo6.6720$qv3.2834377@nnrp5-w.sbc.net> <FpmjOjvwe5bm0Bg09LDTMPhh5vji@4ax.com> <Pine.LNX.4.21.0103051715150.18263-100000@kev.ioc.ee> <cbKjOqPRwNza00jOm7A1HBDkRYj7@4ax.com> <upuft9z0s.fsf@ctwd0143.fitlinxx.com> <uFCnOk=ag9iJ0BBH3CPw9xrMiTRB@4ax.com> <ug0gnk548.fsf@ctwd0143.fitlinxx.com>
Message-ID: <p6uoOgwTJk=zxrdMpt4ziF4CK8Rq@4ax.com>

[The first bit of this posting is related to my logging example, and not too
generally relevant - mainly because I agree withy most of the points David
makes, and concede tghat my example wasn't a particularly good one. But the
last part contains a much better statement of what I'm really after.]

On Fri, 9 Mar 2001 04:12:23 +0100 , David Bolen <db3l at fitlinxx.com> wrote:

>Paul Moore <paul.moore at uk.origin-it.com> writes:
>
>> The obvious, and most user-friendly, way of doing this is using the
>> script name to generate a logfile name. This means looking at
>> sys.argv[0] from within the module.
>
>By "user-friendly" do you mean the user of the script or the
>developer?

I meant the writer of the script which uses the logging module. But some of
your points below make me wonder whether I am right here...

>If you mean the end user of
>the script, I have to admit that I rare place log files where my
>scripts are - scripts are executable code and logs are data, and I
>like to keep those things separate.

Agreed, and that is a good point. But I spend a lot of time managing a wide
range of customer machines, where directory structures and the like vary
wildly. In that situation, having scripts which depend as little as possible
on hard-coded file locations is important. I have found from experience that
locating all files related to a script in directories relative to the script's
location, is a great advantage from a portability point of view.

And no, I don't keep data and code in the same directory (well, not often :-)
- I usually put logs in <script dir>\..\logs or something similar.

>In terms of implementation, I'd probably (and I do) encapsulate my
>logging functionality in a class

If you have a class, then an initialisation argument isn't a great hardship.
Thinking about this, I can see a number of approaches. But the details are
more logging interface design issues, which wasn't really my point. So I'll
just say that I probably agree with you, although we may differ on details.

>> (Which in turn requires sys and os imports which the main script may
>> otherwise not use).
>
>True, but I wouldn't avoid it for that reason - or rather, needing
>standard library modules shouldn't be a negative.

OK, I apologise. That comment was FUD. I shouldn't have made it.

>> This is a real-life case - I had a (VB Script, sorry!) program which
>> did exactly this. And the initialisation step was a definit pain...
>
>Can you clarify the pain part further?  I see this as normal
>initialization that just makes sense and clarity.

Thinking about it, it's probaby not worth explaining further - it's a VB
clumsiness, rather than a general issue.

[...
 comments on the implementation of sys.argv[0] as an absolute path skipped
...]

These are good points, and I agree with most of them.

But my key point remains: There is no 100% reliable way in Python to refer to
a filename which is relative to the location of the main script. The reason
for needing such a file varies (data files, output files, whatever) but that
is a side issue. I suspect that the area of difficulty is that if the (vague)
examples I have of why such a thing might be needed were looked at on a
case-by-case basis, it would be possible to find alternative approaches for
each one in turn. But that's a change in mindset which I haven't achieved yet.

Actually, thinking some more, one way of doing this would be to have the main
script call a module to do all the work. Locate the script and module in the
same directory, and the module can then locate relative filenames via
__file__.

Example:

--- driver.py ---
import app
app.main()

--- app.py ---

def main():
    # whatever

import os
import os.path

def relative(name):
    """Generate a filename relative to the location of this
       module. Absolute filenames are left unchanged, whereas
       relative filenames are interpreted as if the current
       directory was the module's location"""
    if (os.path.isabs(name)): return name
    return os.path.join(os.path.dirname(__file__), name)

--------

Gaah. I tested this and it doesn't work, because __file__ isn't an absolute
pathname!

How _do_ applications locate their data files without hard-coding directory
names, or making assumptions of the current directory? (Actually, that's a
better statement of my issue...)

Paul.



From ejr at cs.berkeley.edu  Sat Mar 17 00:17:35 2001
From: ejr at cs.berkeley.edu (Edward Jason Riedy)
Date: 17 Mar 2001 05:17:35 GMT
Subject: More accurate summations [was Re: This math scares me]
References: <mailman.984726142.2994.python-list@python.org> <98umht$7e0$1@agate.berkeley.edu>
Message-ID: <98us1f$896$1@agate.berkeley.edu>

And I mistakenly blather:
 - 
 - The accumulated error in s from floating-point calculations is at 
 - most one unit in the last place of each summand x, [...]

Geez, I can't get anything right.  It's not one ulp, but
rather the distance from 1.0 to the next higher floating-
point number (either epsilon or 2*epsilon, depending on
whose definition you use).  But self.c still gives a bound
on the total rounding error.

Jason
-- 


From f8dy at diveintopython.org  Wed Mar  7 01:57:35 2001
From: f8dy at diveintopython.org (Mark Pilgrim)
Date: Wed, 07 Mar 2001 01:57:35 -0500
Subject: sorting on IP addresses
In-Reply-To: <mailman.983939410.22287.python-list@python.org>
References: <3A7F26CB.BE85C96F@esec.com.au> <mailman.983939410.22287.python-list@python.org>
Message-ID: <B6CB460F.4AEB%f8dy@diveintopython.org>

in article mailman.983939410.22287.python-list at python.org, Justin Sheehy at
justin at iago.org wrote on 3/6/01 11:29 PM:

> Sam Wun <swun at esec.com.au> writes:
> 
>> Does anyone know what is the quickly way to sort a list of IP addresses?
>> 
>> ie. 203.21.254.89 should be larger than 203.21.254.9
> 
> Convert to longs and sort on those values?
> 
> Not all that quick, I suppose, but I don't know of a faster correct way.

I could have sworn this was discussed about a month ago, but I can't find it
on groups.google.com (damn you Googja!).  Several methods were bandied
about, including custom sort functions passed to .sort().  After some timing
tests, the fastest appears to be munging the data with the magical
socket.inet_aton() function so you can do the sort natively, then munging it
back:

>>> import socket
>>> ips = ['127.0.0.1', '198.162.0.1', '10.1.2.5', '10.1.2.4']
>>> tempips = [socket.inet_aton(ip) for ip in ips]
>>> tempips
['\177\000\000\001', '\306\242\000\001', '\012\001\002\005',
'\012\001\002\004']
>>> tempips.sort()
>>> tempips
['\012\001\002\004', '\012\001\002\005', '\177\000\000\001',
'\306\242\000\001']
>>> [socket.inet_ntoa(ip) for ip in tempips]
['10.1.2.4', '10.1.2.5', '127.0.0.1', '198.162.0.1']

hang-around-long-enough-and-you-can-look-smart-by-recycling-other-people's-c
ode-ly,
-M
You're smart; why haven't you learned Python yet?
http://diveintopython.org/





From wjdandreta at worldnet.att.net  Tue Mar 13 14:37:14 2001
From: wjdandreta at worldnet.att.net (William Dandreta)
Date: Tue, 13 Mar 2001 19:37:14 GMT
Subject: Python interacting with Java Script webpage?
Message-ID: <KFur6.8651$Vg3.600799@bgtnsc06-news.ops.worldnet.att.net>

I would like to write a Python program that will be able to connect to a
webpage that runs a Java Script, post data to the form displayed, and save
the data retrieved to a file.

I have been looking for a book, tutorial, or any docs that will take me from
not having a clue as to how to do it to being able to write the program.

I read some of the info in the Python refs about the http built in classes
but I did not fully understand it and there really was not enough info there
to get me started.

Any suggestions would be appreciated.

Bill




From grante at visi.com  Fri Mar  2 18:24:16 2001
From: grante at visi.com (Grant Edwards)
Date: Fri, 02 Mar 2001 23:24:16 GMT
Subject: cataloging words in text file
References: <3AA0124E.74A39783@comm.mot.com> <v9Vn6.396$z6.33544@ruti.visi.com> <eFVn6.400$z6.34905@ruti.visi.com>
Message-ID: <AYVn6.401$z6.35605@ruti.visi.com>

In article <eFVn6.400$z6.34905 at ruti.visi.com>, Grant Edwards wrote:
>In article <v9Vn6.396$z6.33544 at ruti.visi.com>, Grant Edwards wrote:
>
>>>I remember this homework assignment for my data structures (c++)
>>>class: read in a large file, and create a data structure containing
>>>every word in the file and the number of times it appears.

[...]

>>You may want something a little more sophisticated
[...]
>A call to translate() with an appropriate translation table
>before the split() solves those issues.

----------------------------------------------------------------------
#!/usr/local/bin/python2.1
import sys
t = "".join([(" ",chr(x))[chr(x).isalnum()] for x in range(128)]) + " "*128
d={}
for w in sys.stdin.read().translate(t).split():
    if d.has_key(w):
        d[w] += 1
    else:
        d[w] = 1
print d
----------------------------------------------------------------------

I'm particularly proud of the obfuscated construction of the
translation table.  Looks like time spent reading those
threads on ?: wasn't wasted after all. ;)

Surprisingly, isalnum() returns true for any value larger than
chr(128).  I assume that's a characteristic of isalnum() in the
underlying libc.  Generally one uses isascii() to determine if
the call to isalnum() will return a meaningful value, but
Python doesn't expose isascii().

-- 
Grant Edwards                   grante             Yow!  Life is selling
                                  at               REVOLUTIONARY HAIR
                               visi.com            PRODUCTS!


From raj at grserv.med.jhmi.edu  Sat Mar  3 04:19:57 2001
From: raj at grserv.med.jhmi.edu (RajGopal Srinivasan)
Date: Sat, 03 Mar 2001 09:19:57 GMT
Subject: Adding a method to an instance
References: <G2%n6.360714$JT5.11606234@news20.bellglobal.com> <6621ato2vknc667m3iuqvmm8vcj0h99qqk@4ax.com> <Sc0o6.361217$JT5.11613656@news20.bellglobal.com> <ej41atcongof4vpi5i2vkrilalbdbc06m4@4ax.com> <pd61atcceals21f7kj157cfrq75bp71aq4@4ax.com>
Message-ID: <3AA0B719.FF0BE731@grserv.med.jhmi.edu>

Here is an attempt.  I am completely unsure of what the problems with
this
approach are, but here it goes

>>> class A:
...     def __init__(self):
...             self.value = 2
... 
>>> a = A()
>>> 
>>> def func():
...     self.value = self.value + 2
... 
>>> # the ugly part
... 
>>> func.func_globals['self'] = a
>>> 
>>> a.func = func
>>> 
>>> a.func()
>>> print a.value
4
>>> dir(A)
['__doc__', '__init__', '__module__']
>>> dir(a)
['func', 'value']

raj


From aotto at t-online.de  Mon Mar 26 04:46:03 2001
From: aotto at t-online.de (Andreas Otto)
Date: Mon, 26 Mar 2001 11:46:03 +0200
Subject: "Compiler" 1.2
Message-ID: <3ABF0FDB.42020B0F@t-online.de>

Hi,

  "Compiler" has a new home

      http://www.compiler-factory.com

  Release  1.2  is out


  what's new :

    * new WEB interface
    * new user interface ( master - slave )
    * new licence key's
    * new examples


  i look for :

    * tcl community place(s) to put a banner on
    * sales representative for different country's
    * please email  mailto:aotto at t-online.de   


mfg

  aotto :)


-- 
================================================================
Andreas Otto              Phone: ++49-(0)8152-399540
IPN Ingenieurbuero fuer   mailto:aotto at t-online.de
    Praezisionsnumerik    Web: http://home.t-online.de/home/aotto
Ulmenstrasse 3            Skills: Unix,Sql,Tcl,Shell,Sybase...
D-34289 Zierenberg        Area: Bank, FX, MM, shares, options
=================================================================


From grante at visi.com  Mon Mar  5 09:47:51 2001
From: grante at visi.com (Grant Edwards)
Date: Mon, 05 Mar 2001 14:47:51 GMT
Subject: Little problem with the "+" operator
References: <983802310.155838463@news.t-online.de>
Message-ID: <rGNo6.442$y6.129204@ruti.visi.com>

>I want to make the following:
>
>variableOne = 500
>myVariable = variableOne + "cool"
>
>But I get the following error: "unsupported operand types for +"
>
>Does someone know how to solve the problem?

You'll have to define the problem better.  What value do you
want myVariable to have?

-- 
Grant Edwards                   grante             Yow!  .. I'm IMAGINING a
                                  at               sensuous GIRAFFE, CAVORTING
                               visi.com            in the BACK ROOM of a
                                                   KOSHER DELI --


From admin at mulga.com.au  Fri Mar 16 19:50:01 2001
From: admin at mulga.com.au (Michael Hall)
Date: Sat, 17 Mar 2001 10:20:01 +0930 (CST)
Subject: image handling
Message-ID: <Pine.LNX.4.10.10103171013150.10770-100000@localhost.localdomain>

G'day people:

I'm brand new on this list and am looking for resources about image
processing in a web environment in Python. I'm wondering if there exists a
module that does things like the PHP image handling functions
(creating, copying, resizing, etc). 

I've had a look at the Python Functions Library but there doesn't appear
to be much in there. I've started to use Python a bit in a web environment
and like what I've seen so far, and from what I've heard I'm sure there
must be some pretty flash image handling capabilities hanging around
somewhere.

Thanks
Mick

***********************************************
Michael Hall - mick at mulga.com.au - mulga.com.au
***********************************************





From SCook at pobox.com  Mon Mar 12 07:33:33 2001
From: SCook at pobox.com (Steve)
Date: Mon, 12 Mar 2001 12:33:33 GMT
Subject: Building Python on Win32
Message-ID: <xm3r6.15418$0N3.94318@news-server.bigpond.net.au>

Is it possible to build the python source on Windows with either mingw (the
win32 port of gcc) or the Borland free compiler?




From ws-news at gmx.at  Mon Mar 12 17:14:06 2001
From: ws-news at gmx.at (Werner Schiendl)
Date: Mon, 12 Mar 2001 23:14:06 +0100
Subject: binary comparison of file object
References: <mailman.984429074.25277.python-list@python.org>
Message-ID: <984435530.746122@newsmaster-04.atnet.at>

If the files are small enough, you can read both into a string (with
read() ) and compare those.

hth
werner

Lee, Jaeho <Jaeho.Lee at brooks.com> wrote in message
news:mailman.984429074.25277.python-list at python.org...
> Hello,
>
> I am trying to compare the contents of binary file. Checking if the files
> are same or not is enough.
> Does anybody has idea? Comparing file object in python always return
FALSE.
>
> Thanks,
> /Jaeho
>




From timr at probo.com  Fri Mar 30 02:52:11 2001
From: timr at probo.com (Tim Roberts)
Date: Thu, 29 Mar 2001 23:52:11 -0800
Subject: printing columns question
References: <PXbv6.52892$Lm2.6031361@news0.telusplanet.net>
Message-ID: <rie8cts4m0nbrc1hrlae7lv8udoqmfb39b@4ax.com>

"Victor Louie" <vlouie at telusplanet.net> wrote:
>
>...i was just wondering if anybody could explain to me how the print
>worked...
>    ie. print '\t%-15s%-15s%4d' % (prev, i, citymap[prev][i][0]),
>
>I understand that \t is a tab....but i don't get anything after that

%s says to suck a string out of the parameter tuple and insert it here.
%15s says to right-justify the string in a field 15 columns wide.
%-15s says to left-justify the string in a field 15 columns wide.

  for s in ('x','xx','xxx','xxxx'):
    print "%15s***%-15s***" % (s,s)

              x***x              ***
             xx***xx             ***
            xxx***xxx            ***
           xxxx***xxxx           ***

With %15s or %-15s, the ENTIRE string is inserted, even if it is longer
than 15 characters.  If you want the string limited to at MOST 15
characters, use "%-15.15s".
--
- Tim Roberts, timr at probo.com
  Providenza & Boekelheide, Inc.


From fredrik at pythonware.com  Tue Mar 20 14:42:57 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Tue, 20 Mar 2001 19:42:57 GMT
Subject: Can Python execute a compiled C program
References: <3AB7A784.6C2E2BDA@lucent.com>
Message-ID: <5pOt6.2529$4N4.329020@newsc.telia.net>

Carl Moser wrote:
> Can Python execute a compiled C-program?

    os.system("program")

or

    output = os.popen("program").read()

for the full story, and more options (including os.exec, os.spawn),
see the os module docs in the library reference.

Cheers /F

<!-- (the eff-bot guide to) the standard python library:
http://www.pythonware.com/people/fredrik/librarybook.htm
-->




From paulp at ActiveState.com  Sun Mar 25 15:37:43 2001
From: paulp at ActiveState.com (Paul Prescod)
Date: Sun, 25 Mar 2001 12:37:43 -0800
Subject: static methods
References: <Pine.LNX.4.21.0103231640070.5553-100000@clarkevans.com> <mailman.985386272.7736.python-list@python.org> <slrn9bpcb0.bm7.qrczak@qrnik.zagroda>
Message-ID: <3ABE5717.22AE2EF4@ActiveState.com>

Marcin 'Qrczak' Kowalczyk wrote:
> 
> Fri, 23 Mar 2001 14:27:42 -0800, Paul Prescod <paulp at ActiveState.com> pisze:
> 
> > Probably because they are simply not neccessary.
> >
> > def _static():
> >       XXX.foo=bar
> >       return XXX.foo
> >
> > class Class:
> >       staticmethod = _static
> >       ...
> 
> It doesn't work. Try to call it.
> 
> The fact that you are not the first person who thinks it works
> suggests that the current Python's solution is not ideal :-)

Doh! Okay, it will work if you add it to the instances in the __init__:

class Class:
	def __init__(self):
		self.staticmethod = _static

But the deeper point is that static methods are just a syntactic
short-cut and there are a variety of ways to emulate them...

I can't, off the top of my head, think of any problems with having
Class.methodname return a function object rather than an unbound method.

-- 
Take a recipe. Leave a recipe.  
Python Cookbook!  http://www.activestate.com/pythoncookbook



From phlip_cpp at my-deja.com  Mon Mar 12 14:12:53 2001
From: phlip_cpp at my-deja.com (Phlip)
Date: Mon, 12 Mar 2001 11:12:53 -0800
Subject: PyEval_CallObject and threads
References: <H58r6.9$697.25923@news.pacbell.net>
Message-ID: <rf9r6.16$P26.44570@news.pacbell.net>

Phlip wrote:

> Thonpy:
> 
> My colleague came over and asked me to write this:
> 
> "We got multiple C++ pthreads calling PyEval_CallObject. How threadsafe is
> the Python API? A C++ module using Boost BPL creates multiple threads,
> which then call back into Python. When two threads simultaneously call
> PyEval_CallObject we go boom."
> 
> I will now go examine his serialization primitives. But if anyone has any
> better ideas...

Forgot to mention: We are using Stackless Python.

-- 
  Phlip                          phlip_cpp at my-deja.com
============== http://phlip.webjump.com ==============
  --  DARE to resist drug-war profiteering  --


From gabriel_ambuehl at buz.ch  Fri Mar 30 14:11:41 2001
From: gabriel_ambuehl at buz.ch (Gabriel Ambuehl)
Date: Fri, 30 Mar 2001 21:11:41 +0200
Subject: Speedup hint for users of viewcvs package
In-Reply-To: <m14j45Y-000CnEC@artcom0.artcom-gmbh.de>
References: <m14j45Y-000CnEC@artcom0.artcom-gmbh.de>
Message-ID: <190551356909.20010330211141@buz.ch>

-----BEGIN PGP SIGNED MESSAGE-----

Hello Peter,

Friday, March 30, 2001, 8:57:35 PM, you wrote:
> Explanation: viewcvs.cgi is a ~ 68 kB source file.  After applying
the
> procedure outlined above, this is compiled into a byte code file.
> So there is no longer recompilation needed on each cgi request.

This can, of course, be applied to almost every Python CGI script.
OTOH,
one might think about going straight with mod_python anyway...


Best regards,
 Gabriel

-----BEGIN PGP SIGNATURE-----
Version: PGP 6.0.2i

iQEVAwUBOsTMXsZa2WpymlDxAQHkHgf/YG6S+hK6+gyUxbajlSF7uHLpPGXX5fuT
3frRagE44403hCsZNPj1c/RxjEtdsP/6kXwqJCDCP4JKPMRMQ47xgfuAgGhlTPHV
rb4rrPmT8lha3pWx9ScQZ07JVWu14nYjgqIRa+Pj3+S5UoTelrLJZ4qsaU07aX+D
D0V9qYh68oDc+p/BtE85kZbEyamMstA87ybIDNJFOXYUSmsajyhQSB2PCCiIc4HU
wZ1+VGwlUgeD4Hb+VkGOhqWUqMxESi3tLFHpOSGNSrYlUL2klZJ4W2oywVKOeUR6
BLJI62yS6qAZDiTE1sWDPDJ1HXKtCQSp63SYj1Jyn/cOYbvn/Pwikg==
=hbU0
-----END PGP SIGNATURE-----





From dav-nospam-id at westco-nospam-ntrol.com  Fri Mar 23 04:04:48 2001
From: dav-nospam-id at westco-nospam-ntrol.com (David Brown)
Date: Fri, 23 Mar 2001 10:04:48 +0100
Subject: Python 2.1 and MySQL 3.23.33 on Windows
References: <99d1te$ogo$1@news.netpower.no> <3ABA2F78.87E037CA@bigfoot.de>
Message-ID: <99f3j7$l9g$1@news.netpower.no>

The new version seems to be working fine now.  Many thanks, Gerhard.


Gerhard H?ring wrote in message <3ABA2F78.87E037CA at bigfoot.de>...
>David Brown wrote:
>>
>> I have MySQL 3.23.33 and Python 2.1 installed on my Windows machine.
Both
>> work fine (I use MySQL along with PHP and Apache), and I have used Python
>> for a few short scripts.  But I cannot get Python to work with MySQL.
The
>> mySQLdb windows installation is built for Python 2.0, not 2.1, and I do
not
>> have MSVC to recompile it myself.  Does anyone have any hints or pointers
to
>> get me started?  Thanks.
>
>A new version of the MySQLdb Windows binary distribution is available at
>my homepage. This one contains binaries for Python 1.5.2, 2.0 and 2.1
>Beta 1.
>
>Gerhard
>--
>Sorry for the fake email, please use the real one below to reply.
>contact: g e r h a r d @ b i g f o o t . d e
>web:     http://highqualdev.com




From erno-news at erno.iki.fi  Tue Mar  6 08:09:32 2001
From: erno-news at erno.iki.fi (Erno Kuusela)
Date: 06 Mar 2001 15:09:32 +0200
Subject: I come to praise .join, not to bury it...
References: <mailman.983485923.1136.python-list@python.org> <97p7iq$kk2$1@nntp6.u.washington.edu> <97rka002k9@news1.newsguy.com> <kuu257kb1n.fsf@lasipalatsi.fi> <982a4f0su3@news1.newsguy.com>
Message-ID: <kuofvfjb77.fsf@lasipalatsi.fi>

In article <982a4f0su3 at news1.newsguy.com>, "Alex Martelli"
<aleaxit at yahoo.com> writes:

| "Erno Kuusela" <erno-news at erno.iki.fi> wrote in message

|| what's wrong with vague aesthetic?-) if lots of people share it,
|| it's not to be dismissed so lightly.

| Lots of people share the 'vague aesthetic' of desiring
| to use braces instead of whitespace to delimit blocks
| of statements, for example.

| Python dismisses that lightly, and that is definitely
| a part of what makes it such a wonderful language.

it could also be argued that python uses whitespace instead
of braces because of vague aesthetic shared by lots of people. :)

  -- erno





From wsryu at fas.harvard.edu  Tue Mar 13 11:01:56 2001
From: wsryu at fas.harvard.edu (William Ryu)
Date: Tue, 13 Mar 2001 11:01:56 -0500
Subject: Digest Number 1883
In-Reply-To: <984498113.566.89109.l8@yahoogroups.com>
Message-ID: <4.3.2.7.2.20010313105956.02f834d0@pop.fas.harvard.edu>

An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20010313/a6b7a50a/attachment.html>

From bbollenbach at homenospam.com  Tue Mar  6 13:05:09 2001
From: bbollenbach at homenospam.com (Brad Bollenbach)
Date: Tue, 06 Mar 2001 18:05:09 GMT
Subject: Can Python replace Visual Basic? Should it?
References: <5DZo6.11126$hn5.1578410@news1.rdc1.mb.home.com> <98266k0p5h@news1.newsguy.com> <3q8p6.12338$hn5.1798569@news1.rdc1.mb.home.com> <983898657.328387@newsmaster-04.atnet.at>
Message-ID: <pF9p6.12685$hn5.1803263@news1.rdc1.mb.home.com>

"Werner Schiendl" <ws-news at gmx.at> wrote in message
news:983898657.328387 at newsmaster-04.atnet.at...

[snip]
> In general I agree that one should know how tha mechanics of the program
> works. But I think it is not necessary to know how Visual Basic registers
> Active X Controls in detail. It works and I know how to invoke it, what's
> the problem? I do not know in detail how the .EXE file is made up or how
my
> harddisk handles CRCs. And still the programs work fine...
[snip]

Okay...I said this should be a separate thread, but you asked for it... :)
Here are just a few of the things I don't like about IDE's, with my point
being that they obscure the programmer's view of what's going on, and
therefore can often do more harm than good (and quite honestly, make one a
worse programmer than when you started).

I'll refer to Powerbuilder 6.5.1 on Win32 systems, which is what my day job
pays me to use right now. For those not familiar with Powerbuilder, it's
very similar to VB in its GUI-building capacity, and is arguably best suited
to developing client-server heavily database oriented programs. It's also
object-oriented, or not, depending on who you ask (so don't flame that
statement please).

Now...what don't I like about this IDE?

First of all, if you create a basic visual user object (akin to a "class"),
say a regular button to which you add some custom user events then you want
to place this on your form. Now you want to put this on one of your forms,
so you click on the User Object icon and click the "Inherit" button, then
find this new button you've inherited from PB's base button and modified,
and put it onto your form. The property sheet's title bar for this button on
your form will now say "cb_whatever inherited from u_cb_baseclass", when in
fact this isn't even inherited because cb_whatever cannot be further
inherited. In other words, what you've really done is instantiated it.

This throws a wrench in the works for those of us who think we know OO
terminology, already instilling a misunderstanding of how our objects exist
and interact within the system.

Another hindrance...Powerbuilder keys objects in the system by the .pbl (a
Powerbuilder library of objects) they are located in, _rather than the name
of the object_. I've found this wastes huge amounts of time when trying to
locate an object when I can't remember the .pbl it's in (and it certainly
doesn't take any kind of genius to figure out that this ordering scheme is
ridiculous).

Yet another annoyance, again related solely to the IDE: in Powerbuilder you
_cannot_ have a base class and one of its direct descendants open at the
same time (however two levels down is okay). Do I need to explain how
annoying this is when I'm trying to debug code by reading it through, and a
function that is defined in the super class can't be accessed because of
this limitation without closing down the object whose code I was currently
reading? Ugh...this is another limitation that I encounter at least a few
times a week.

Finally, briefly, I hate the fact that Powerbuilder only allows you to see
(to my knowledge) the code of a single method of each object at a time, when
sometimes I want to be able to let my eyes jump from method to method as I
scroll so I can follow what's going on smoothly, without always narrowing my
view to one method at a time. Combining this point with the previous
paragraph, and the idea of trying to debug your code in your head before
jumping to the debugger is an absolute nightmare.

I'll stop there, to keep the size of this already lengthy post to a minimum.
But the points I mentioned cover only about half of the things that _really_
annoy me about that IDE, not to mention the scores of other little bugs that
have occassionally caused me to lose a whole method's worth of code now and
then. As for VB, it can be said simply that I love playing with OO (yes, I
call it "playing" even though it's what's paying the bills :) and of course,
VB doesn't offer that yet.


Brad




From sholden at holdenweb.com  Thu Mar 22 22:46:46 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 22 Mar 2001 22:46:46 -0500
Subject: extending the built in dictionary type
References: <umfKujAz2iu6EwHa@jessikat.demon.co.uk>
Message-ID: <YJzu6.12928$_46.482847@e420r-atl2.usenetserver.com>

Look in the library and inherit from UserDict. A search on UserDict revelas
that this probably isn't quite as FAQtually recorded as it might be.

regards
 Steve

"Robin Becker" <robin at jessikat.fsnet.co.uk> wrote in message
news:umfKujAz2iu6EwHa at jessikat.demon.co.uk...
> I'm interested in extending the built in dictionary. It's fairly simple
> to wrap, but seems incredibly hard to chain the methods. Is there any
> obvious way to use most of the built in methods whilst modifying a few.
>
> Seems that although it exists I can't chain the original mapp_methods of
> dictobject as that is not visible outside pythonxx.dll. So is there an
> easier way to get at these internal types?
> --
> Robin Becker




From mkx at excite.com  Fri Mar 23 11:18:17 2001
From: mkx at excite.com (mkx at excite.com)
Date: Fri, 23 Mar 2001 11:18:17 -0500
Subject: Ruby will overtake Python within four years.
References: <mailman.985324111.22754.python-list@python.org>
Message-ID: <egtmbt0ccsq5msg9tf509gga8b1h2d00fh@4ax.com>

Funny how folks like to make predictions based on imagined statistics.

Given the way that this field changes, I doubt anyone could accurately
make such a prediction based on established languages.

Personally, I have lost of reasons not to use either Perl of VB, but
they are _very_ popular in my market. The positive apsects of a
language rarely can overcome a mindshare based on history and
investment.


On 23 Mar 2001 04:58:09 -0000, "prashant  patel"
<cognoscente at rediffmail.com> wrote:

>No I am not a ruby programmer not even using any scripting language other once used vbscript in ASP.
>my major programming language has been java. But with the experience in java, I want learn a scripting language which increases my productivity, java is simply to redundant.
>for example : doing an IO required to use so many classes.
>for the past one month i have been find a alternative, 
>and found three scripting language python,perl,and ruby.
>for now python is my choice, perl is surelty a no for its syntax.but how about python and ruby, as i would be 
>allocating a lot of time learing language the particular language, could some one give out facts about ruby.
>I would be unjustified actually to ask python lover's the good and bad of ruby compared to python, but these discusions will help us also evaluate python.
>And yes the subject line and be taken from. 
>http://www.amazon.com/exec/obidos/tg/feature/-/99126/102-2233024-4095364
>But my inclination are towards python.




From carmstro at dynup.net  Sun Mar  4 08:01:21 2001
From: carmstro at dynup.net (Chris Armstrong)
Date: Sun, 04 Mar 2001 18:01:21 +0500
Subject: python cgi syntax highlighter
Message-ID: <ta5i9s84d78ke5@corp.supernews.com>

You can use the 'htmlize' command in Xemacs (and probably Emacs as well)
to do this.

In article <Pine.BSF.4.32.0103032234260.88338-100000 at localhost>,
grisha at ispol.com wrote:

> 
> Hello -
> 
> I remember coming across a script that displayed python code in HTML
> with color syntax highlighting - does anyone know where I can find one?
> My searches don't turn up anything.
> 
> Thanks,
> 
> Grisha
> 


-- 
Chris Armstrong   http://twistedmatrix.com/~carmstro     carmstro at dynup.net
There is a 90% chance that this message was written when the author's been
awake longer than he should have. Please disregard any senseless drivel.


From emile at fenx.com  Tue Mar 27 09:00:41 2001
From: emile at fenx.com (Emile van Sebille)
Date: Tue, 27 Mar 2001 06:00:41 -0800
Subject: who is my caller?
References: <mailman.985700854.31956.python-list@python.org>
Message-ID: <99q6kv$25vk6$1@ID-11957.news.dfncis.de>

Take a look at trace.extract_stack

--

Emile van Sebille
emile at fenx.com

---------
"Clark C. Evans" <cce at clarkevans.com> wrote in message
news:mailman.985700854.31956.python-list at python.org...
> In many mud languages, you can query the
> call-stack to find out who called you.
> Is there a stack of function objects
> that can be obtained somehow?  Is this
> non-portable black magic?
>
> I'm asking beacuse this technique is
> often useful for implementing security
> (ok, not very good security)
>
> Thanks!
>
> Clark
>
>




From kwoeltje at mail.mcg.edu  Tue Mar 27 08:09:52 2001
From: kwoeltje at mail.mcg.edu (Keith F. Woeltje)
Date: Tue, 27 Mar 2001 13:09:52 GMT
Subject: Hot breakfast foods (was Oatmeal (was warrior geeks))
References: <3ABF72F1.98CB78F@sage.att.com> <3ABFA209.CF27BC20@optushome.com.au> <mailman.985651834.18706.python-list@python.org>
Message-ID: <3AC09120.90CB5A5D@mail.mcg.edu>

Oatmeal is OK, but I think there can be no possible disagreement that
grits are far superior, followed closely by farina (Cream of Wheat).
Anyone who thinks otherwise is clearly the inbred descendent of genetic
defectives, etc., etc. etc.
>K

Ken Seehof wrote:
> 
> I like oatmeal (porridge) too.
>


From emile at fenx.com  Sat Mar 17 09:32:54 2001
From: emile at fenx.com (Emile van Sebille)
Date: Sat, 17 Mar 2001 06:32:54 -0800
Subject: Deleting dictionary items
References: <98ug6n$q9d$1@news6.svr.pol.co.uk>
Message-ID: <98vsoe$3ro7i$1@ID-11957.news.dfncis.de>

I'm not clear on what you're asking, but I think you want to delete keys
from a dictionary where a certain value is one of the elements of the tuple
that is the key.  Does this do it:

# delete keys from mvDict where value in key
for key in myDict.keys():
    if value in key:
        del myDict(key)

Emile van Sebille
emile at fenx.com

"Will Newton" <will at nospam.misconception.org.uk> wrote in message
news:98ug6n$q9d$1 at news6.svr.pol.co.uk...
>
> What's the best way to delete dictionary items?
>
> I have a dictionary with tuples as keys, and I would like to delete all
> keys with a certain element. First I wrote:
>
>                 for item in td.data.keys():
>                         if item[attr] != value:
>                                 del td.data[item]
>
>
> But will this work?




From padiolea at merlin.irisa.fr  Tue Mar 13 15:28:25 2001
From: padiolea at merlin.irisa.fr (Yoann Padioleau)
Date: 13 Mar 2001 21:28:25 +0100
Subject: Einstein's Riddle
References: <oNPq6.396644$ge4.138479566@news2.rdc2.tx.home.com>
Message-ID: <wi6elw1pg5y.fsf@merlin.irisa.fr>

"Steve Graham" <js.graham at home.com> writes:

> Who would be interested in using his/her brain (and his computer) to solve
> the following?
here it is:

%----------------------------------------------------------------
% a sicstus program
:- use_module(library(clpfd)).                                                                                            

relation(Unknown) :-
	Unknown =
	[[Color1, Nationality1, Drink1, Smoke1, Pet1]
	,[Color2, Nationality2, Drink2, Smoke2, Pet2]
	,[Color3, Nationality3, Drink3, Smoke3, Pet3]
	,[Color4, Nationality4, Drink4, Smoke4, Pet4]
	,[Color5, Nationality5, Drink5, Smoke5, Pet5]
	]
	, Colors =        [Color1, Color2, Color3, Color4, Color5]
	, Nationalities = [Nationality1, Nationality2, Nationality3, Nationality4, Nationality5]
	, Drinks =        [Drink1, Drink2, Drink3, Drink4, Drink5]
	, Smokes =        [Smoke1, Smoke2, Smoke3, Smoke4, Smoke5]
	, Pets   =        [Pet1, Pet2, Pet3, Pet4, Pet5],
			
	Red = 1, Green = 2, Yellow = 3, Blue = 4, White = 5,
	Brit = 1, Sweed = 2, Dane = 3, Norvegian = 4, German = 5,
	Tea = 1, Coffee = 2, Milk = 3, Beer = 4, Water = 5,
	Pall_mall = 1, Dunhill = 2, Blends = 3, Bluemaster = 4, Prince = 5,
	Dogs = 1, Birds = 2, Cats = 3, Fish = 4, Horse = 5,

	domain([Color1, Nationality1, Drink1, Smoke1, Pet1
	       ,Color2, Nationality2, Drink2, Smoke2, Pet2
	       ,Color3, Nationality3, Drink3, Smoke3, Pet3
	       ,Color4, Nationality4, Drink4, Smoke4, Pet4
	       ,Color5, Nationality5, Drink5, Smoke5, Pet5], 1, 5)
	, all_different(Colors)
	, all_different(Nationalities)
	, all_different(Drinks)
	, all_different(Smokes)
	, all_different(Pets)
	,

% cant do the member, cos member need instantiated things!! and if enum and then test member
%  we gain nothing cos always big search space, the labeling must be at the end !!

	 A1 in 1..5, element(A1, Nationalities, Brit),  element(A1,Colors,Red),
	 A2 in 1..5, element(A2, Nationalities, Sweed), element(A2,Pets, Dogs),
	 A3 in 1..5, element(A3, Nationalities, Dane),  element(A3,Drinks, Tea),
	 A4 in 1..5, element(A4, Colors,        Green),
	 A5 in 1..5, A5 #= A4+1, element(A5, Colors, White),
	 A6 in 1..5, element(A6, Colors, Green),        element(A6, Drinks, Coffee),
	 A7 in 1..5, element(A7, Smokes, Pall_mall),    element(A7, Pets,  Birds),
	 A8 in 1..5, element(A8, Colors, Yellow),       element(A8, Smokes, Dunhill),
	 element(3, Drinks, Milk),
	 element(1, Nationalities, Norvegian),
	 A9 in 1..5, element(A9, Smokes, Blends),
	 A10 in 1..5, ((A10 #= A9-1) #\/ (A10 #= A9+1)), element(A10, Pets, Cats),
	 A11 in 1..5, element(A11, Pets, Horse),
	 A12 in 1..5 , ((A12 #= A11-1) #\/ (A12 #= A11+1)), element(A12, Smokes, Dunhill),
	 A13 in 1..5, element(A13, Smokes, Bluemaster),  element(A13, Drinks, Beer),
	 A14 in 1..5, element(A14, Nationalities, German), element(A14, Smokes, Prince),
	 A15 in 1..5, element(A15, Nationalities, Norvegian),
	 A16 in 1..5 , ((A16 #= A15-1) #\/ (A16 #= A15+1)), element(A16, Colors, Blue),
	 A17 in 1..5, element(A17, Smokes, Blends),
	 A18 in 1..5 , ((A18 #= A17-1) #\/ (A18 #= A17+1)), element(A18, Drinks, Water),
	
	labeling([],
		 [Color1, Nationality1, Drink1, Smoke1, Pet1
		 ,Color2, Nationality2, Drink2, Smoke2, Pet2
		 ,Color3, Nationality3, Drink3, Smoke3, Pet3
		 ,Color4, Nationality4, Drink4, Smoke4, Pet4
		 ,Color5, Nationality5, Drink5, Smoke5, Pet5
		 , A1, A2, A3, A4, A5, A6, A7, A8, A9, A10, A11, A12, A13, A14, A15, A16, A17, A18
		 ]).
		
%relation(X).
%  X = [[3,4,5,2,3],[4,3,1,3,5],[1,1,3,1,2],[2,5,2,5,4],[5,2,4,4,1]] ? ; corresponding to
%  [Yellow, Norvegian, Water, Dunhill, Cats]
%  [Blue,   Dane,       Tea,  Blends, Horse]
%  [Red,    Brit,      Milk,  Pall_mall, Birds]
%  [Green,  German,  Coffee,  Prince,    Fish]
%  [White,  Sweed,   Beer,  Bluemaster,  Dogs]

%----------------------------------------------------------------


> 
> 
> Steve Graham
> 
> ===
> 
> Einstein's Riddle
> 
> Albert Einstein wrote this riddle this century [ed. 20th century].  He said
> 98% of the world could not solve it.
> 
> There are 5 houses in 5 different colors.  In each house lives a person with
> a different nationality.  The 5 owners drink a certain type of beverage,
> smoke a certain brand of cigar, and keep a certain pet.  No owners
> have the same pet, smoke the same brand of cigar or drink the same beverage.
> 
> The question is: "Who owns the fish?"
> 
> Hints:
> 
> The Brit lives in the red house.
> 
> The Sweed keeps dogs as pets.
> 
> The Dane drinks tea.
> 
> The green house is on the left of the white house.
> 
> The green house's owner drinks coffee.
> 
> The person who smokes Pall Mall rears birds.
> 
> The owner of the yellow house smokes Dunhill.
> 
> The man living in the center house drinks milk.
> 
> The Norwegian lives in the first house.
> 
> The man who smokes Blends lives next to the one who keeps cats.
> 
> The man who keeps the horse lives next to the man who smokes Dunhill.
> 
> The owner who smokes Bluemasters drinks beer.
> 
> The German smokes Prince.
> 
> The Norwegian lives next to the blue house.
> 
> The man who smokes Blends has a neighbor who drinks water.
> 
> 
> 
> 
> 
> 
> 
> 
> 

-- 
Yoann  Padioleau,  INSA de Rennes, France,   http://www.irisa.fr/prive/padiolea
Opinions expressed here are only mine. Je n'?cris qu'? titre personnel.
**____   Get Free. Be Smart.  Simply use Linux and Free Software.   ____**


From tuttledon at hotmail.com  Tue Mar 13 17:52:06 2001
From: tuttledon at hotmail.com (Don Tuttle)
Date: Tue, 13 Mar 2001 22:52:06 GMT
Subject: win32all buld 137 win32gui fails
References: <98l8qk$oeo$1@troll.powertech.no> <DYqr6.83020$__6.16207072@typhoon.southeast.rr.com>
Message-ID: <qwxr6.83805$__6.16506760@typhoon.southeast.rr.com>

"Don Tuttle"
> Don't know.  It imports fine for me.  Perhaps unistall 137, then
reinstall.

Oops.  I was sitting at a AP"203" machine.  It fails for me too on build
137.

Don




From sholden at holdenweb.com  Wed Mar  7 14:03:11 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Wed, 7 Mar 2001 14:03:11 -0500
Subject: Can Python 1.5.2 code run on Python 2.x?
References: <985c9m$3gg$1@hobbes2.crc.com>
Message-ID: <FFvp6.35244$1D5.1473084@e420r-atl1.usenetserver.com>

Most code will run unchanged: only a few obscure bugs in library routines
led to incompatibilities, when the routines were fixed.

Try it, you'll be pleasantly surprised.

regards
 Steve

"Glen Mettler" <gem at hsv.crc.com> wrote in message
news:985c9m$3gg$1 at hobbes2.crc.com...
> Can code written in Python 1.5.2 run without modification on Python 2.x?
> If not, what modification in code is required?
>
> Glen
>
>
>




From aleaxit at yahoo.com  Sun Mar 11 17:08:48 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sun, 11 Mar 2001 23:08:48 +0100
Subject: Static typing (was Re: Java guy interested in Python)
References: <mailman.984334635.27345.python-list@python.org> <98glhc02sga@news2.newsguy.com> <m3hf1080or.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <98gt3f064l@news2.newsguy.com>

"Michael Hudson" <mwh21 at cam.ac.uk> wrote in message
news:m3hf1080or.fsf at atrus.jesus.cam.ac.uk...
    [snip]
> > > Since that's all "int i;" *can* mean in C++, seems tautological
<wink>.
> >
> > Surely you're joking, Mister Peters?
> >
> >     class int {
    [snip]
> > Isn't this perfectly legal C++
>
> No.
>
> > , meaning something very different from "i behaves like an integer
> > variable"...?  int is no "reserved word", after all.
>
> Yes it is.  "identifier" and "keyword" are disjoint classes of token
> in C++, at least by my reading of the 1997-12 working draft I've just
> dredged up with google.
>
> Unless I've missed some subtle joke...

Sorry, too subtle -- the point was meant to be that a #define int foo
_would_ make it legal (and incomprehensible to anybody, if the
#define is well-hidden enough) -- sort of trying to see who'd fail
to spot this.  But on re-reading it falls pretty flat -- sorry!


Alex





From borealis3 at home.com  Sat Mar 31 16:14:27 2001
From: borealis3 at home.com (Gary Walker)
Date: Sat, 31 Mar 2001 21:14:27 GMT
Subject: Newbie Dictionary Question
References: <jfmx6.7861$iU.1496746@news1.rdc1.md.home.com> <slrn9cc7i9.9n3.quinn@riyal.ugcs.caltech.edu>
Message-ID: <TMrx6.8831$iU.1687713@news1.rdc1.md.home.com>

Thanks!! I'd put sample code but there's no need. You've more than answered
the question, and brilliantly too, I might add. I'm actually a Delphi/Kylix
programmer (that is to say, I get paid to do that), and I'm learning Python
to amuse myself. Your answer made total sense, in a way that the online
tutorials and O'Reilly books have not.

I really appreciate the nitpicking part, too! That was the detail that I was
missing. In fact, within two minutes of reading your message, I had my
little program working!!

Thanks again!

Gary


Quinn Dunkan wrote in message ...
>On Sat, 31 Mar 2001 14:57:19 GMT, Gary Walker <borealis3 at home.com> wrote:
>>I *think* I want to do the following:
>>
>>I want to create a list (this I'm able to do)
>>Then I want to dynamically create a dictionary for each item in the list,
>>add some keys (and values), and then put the dictionary in the list.
>>
>>(This part is frustrating me, because my newbie python code is referencing
>>the SAME dictionary for each of the list indexes, this NOT what I want. I
>>want a new instance of a dictionary for each pass thru the list as I loop)
>
>list = []
>for elt in thingsies:
>    list.append({'key': elt})
>
>>I want to do this because this way I can have an ordered list of key/value
>>pairs... at least that was my plan...
>
>Well, then a dictionary isn't going to help you, it's not ordered.  You
might
>want a list of pairs:
>
>a = [
>    ('foo', 'a'),
>    ('bar', 'q'),
>    ('baz', 'j'),
>    ('faz', 't'),
>]
>
>Or, if you mean 'sorted' when you say 'ordered', you could sort a
dictionary's
>keys and iterate over that:
>
>d = { ... }
>ks = d.keys()
>ks.sort()
>for k in ks:
>    print d[k]
>
>>In other words:
>>How do I dynamically create a dictionary whose name is not known until
>>runtime?
>
>Use braces:
>
>{'foo': bar}
>
>Nitpick: dictionaries don't have names.  No objects have names.
Dictionaries
>might be bound to one, two, three, a million, or zero names, but they
>themselves don't have names.  So "{'foo': bar}" creates a dictionary
without a
>name.  "d = {'foo': bar}" creates a dictionary without a name *and* a name
>in the current environment (global or local) that refers to that
dictionary.
>"def f(x): ..." creates a function without a name and a name in the current
>environment that refers to that function.  "a = b = {}" creates a function
>without a name, and two names in the current environment that refer to it.
>"a[0] = {}" creates a dictionary without a name and asks 'a' to store a
>reference to it in its 0th element.
>
>>Actually, they don't have to be ordered, so I suppose (thinking aloud
here)
>>that I could use a dictionary of dictionaries, but again - it seems to me
>>that I'd need to dynamically create a dictionary whose name isn't known
>>until runtime...
>>
>>Or am I missing the obvious? This is the most likely scenario :)
>
>You should post some example code.  I'm not sure I know what you're talking
>about.




From phd at phd.pp.ru  Thu Mar  1 12:36:53 2001
From: phd at phd.pp.ru (Oleg Broytmann)
Date: Thu, 1 Mar 2001 20:36:53 +0300 (MSK)
Subject: Newbie createing .PYC for import
In-Reply-To: <3a9e83fa$1@news1.dbtech.net>
Message-ID: <Pine.LNX.4.30.0103012036130.31158-100000@fep132.fep.ru>

On Thu, 1 Mar 2001, Larry Bates wrote:
> I have spent the last couple of hours (unsuccessfully) attempting to figure
> out how to generate a .PYC file for a module that I've defined.  Everything
> appears to imply that it's "automatic", but I'm not getting anywhere.  I
> have a module and I wish to generate a .PYC file so that I can import it
> into future projects.  Can anyone shed any light on how to accomplish this?

import sys
from py_compile import compile

if len(sys.argv) <= 1:
   sys.exit(1)

for file in sys.argv[1:]:
   compile(file)

Oleg.
----
     Oleg Broytmann            http://phd.pp.ru/            phd at phd.pp.ru
           Programmers don't die, they just GOSUB without RETURN.




From lmhamric at aol.com  Fri Mar 30 02:13:22 2001
From: lmhamric at aol.com (LMHAMRIC)
Date: 30 Mar 2001 07:13:22 GMT
Subject: Where are the examples/samples?
Message-ID: <20010330021322.08468.00001688@ng-fe1.aol.com>

After looking at every Python book in the local bookstore and at every
Python-related site I could find on the Internet I have to admit that I'm
perplexed. I've seen a number of posts that have asked for examples of various
things and when I follow the links - where are the examples? Lots of theory,
lots computer-science indulgence, but no examples. Most recently tried to get
started on using BaseHTMLProcessor.py (or HTMLLIB or SGMLIB or whatever) and an
example called dialect.py was given to illustrate its use. The latter
translates things into Pig Latin, among other things! I can't tell what part of
it involves HTML and what part is just fooling around with something else. Does
anyone know of examples anywhere that illustrate usage at a bare bones level
where you can simply see how the interfaces work? I suspect that learning
Python would go a lot better if one could skip the theoretical stuff and simply
get on wiht working examples. You can come back and pick up the theory later.
Anyhow, if anyone else sees this as a problem and has found an answer then I'd
appreciate hearing what it is.  


From aleaxit at yahoo.com  Thu Mar  1 12:24:57 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 1 Mar 2001 18:24:57 +0100
Subject: PEP 227 (was Re: Nested scopes resolution -- you can breathe again!)
References: <LNBBLJKPBEHFEDALKOLCMEEPJBAA.tim.one@home.com><200102230259.VAA19238@cj20424-a.reston1.va.home.com><mailman.983384008.21434.python-list@python.org><97lqsr0311d@news2.newsguy.com> <15006.33055.663211.136994@w221.z064000254.bwi-md.dsl.cnc.net>
Message-ID: <97m0lb08bi@news2.newsguy.com>

"Jeremy Hylton" <jeremy at alum.mit.edu> wrote in message
news:15006.33055.663211.136994 at w221.z064000254.bwi-md.dsl.cnc.net...
> Or: global NAME in FUNCTION
>
> def eggs():
>     z = 2
>     def spam():
>         def inner():
>             global x in spam, z in eggs

Yep, looks better.  And it should be such a rare
use that the occasional need for repetition,
    global x in spam, y in spam
is no big deterrent.


Alex





From dkuhlman at rexx.com  Tue Mar  6 15:04:38 2001
From: dkuhlman at rexx.com (Dave Kuhlman)
Date: Tue, 06 Mar 2001 20:04:38 -0000
Subject: Proposal: allow '?' and '!' in identifiers
References: <96sh22$jq5$1@news.mathworks.com> <Vklk6.6049$pS4.2369984@typhoon2.ba-dsg.net> <96sp4s$ni7$1@news.mathworks.com> <xqyk6.533$S7.49256@ruti.visi.com> <3A9308FF.D07FB37F@javanet.com> <sMcm6.959$aw5.1833@www.newsranger.com>
Message-ID: <983909085.374175@rexx.com>

Gregory Jorgensen <gregj at pobox.com> wrote:
> 
> While I'm ranting... features of programming languages aren't intuitive. Some
> people may deduce the syntax easier than others, but that is experience, not
> intuition. My intuition tells me that any non-alphanumeric character in a symbol
> name may mean something special to the compiler; why would I think ? relates to
> its English usage when _ doesn't? In the context of reading and learning Python
> neither ? nor _ say anything to my intuition.

Actually, there are no English words or Spanish words that contain
a question mark.  A question mark is a punctuation symbol that
comes at the end of a sentence.  So, your English and Spanish
intuition will be fooled too.  In the following sentences "Here is
a shoe." and "Where is the shoe?" the last word is spelled the same.


  - Dave

> 
> 
> 
> Greg Jorgensen
> Deschooling Society
> gregj at pobox.com

-- 
Dave Kuhlman
dkuhlman at rexx.com


From fredrik at pythonware.com  Tue Mar 13 14:29:40 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Tue, 13 Mar 2001 19:29:40 GMT
Subject: This math scares me
References: <3aad4062.1290021@News.CIS.DFN.DE> <slrn9aqkth.aki.scarblac@pino.selwerd.nl> <3aad6ffe.3692710@News.CIS.DFN.DE>
Message-ID: <Eyur6.111$4d3.21486@newsb.telia.net>

Costas Menico wrote:
> Fortunately I discovered I can use "print 5.01 + 5.54" and it does the
> job. It has the extra intelligence built into it to autoformat

your definition of "intelligence" seems to be a bit unusual...




From josh at pobox.com  Mon Mar  5 06:46:37 2001
From: josh at pobox.com (Joshua Spoerri)
Date: Mon, 5 Mar 2001 06:46:37 -0500
Subject: supporting asyncronous messages
Message-ID: <20010305064637.A40204@acis.mc.yu.edu>

I'm designing a workflow server,
and want to use python for process descriptions,
using just a few additional messaging objects.

I want to block on receiving messages, but only
on those messages that will affect the program execution.

for example:
	a = wfmessage()
	b = wfmessage()
	a.send()
	b.send()
	if a.getResponse() || b.getResponse():
	whatever

If b.getResponse() is already available and true,
but a.getReponse() is blocking, the program shouldn't
wait for it. (Yes, this does funny things to
shortcircuiting order.)

Is there a sane way to support this behaviour without
hacking up the compiler? Is it bad news for other
reasons?

(Please cc me, since i'm not subscribed.)

Thanks much,
Joshua Spoerri



From jafo at tummy.com  Sat Mar 24 01:50:08 2001
From: jafo at tummy.com (Sean Reifschneider)
Date: Fri, 23 Mar 2001 23:50:08 -0700
Subject: Python 2.1b2 SRPM (was: Re: Python 2.1b2 released)
In-Reply-To: <200103232229.RAA19777@cj20424-a.reston1.va.home.com>; from guido@digicool.com on Fri, Mar 23, 2001 at 05:29:02PM -0500
References: <200103232229.RAA19777@cj20424-a.reston1.va.home.com>
Message-ID: <20010323235008.A30668@tummy.com>

Shy of RPMs because of library or other dependancy problems with most of
the RPMs you pick up?  The cure, in my experience is to pick up an SRPM.
All you need to do to build a binary package tailored to your system is run
"rpm --rebuild <packagename>.src.rpm".

I've just put up an SRPM of the 2.1b2 release at:

   ftp://ftp.tummy.com/pub/tummy/RPMS/SRPMS/

Again, this one builds the executable as "python2.1", and can be installed
along-side your normal Python on the system.  Want to check out a great new
feature?  Type "python2.1 /usr/bin/pydoc string".

Download the SRPM from above, and most users can install a binary built
against exactly the set of packages on their system by doing:

   rpm --rebuild python-2.1b2-1tummy.src.rpm
   rpm -i /usr/src/redhat/RPMS/i386/python*2.1b1-1tummy.i386.rpm

Note that this release enables "--with-pymalloc".  If you experience
problems with modules you use, please report the module and how it can be
reproduced so that these issues can be taken care of.

Enjoy,
Sean
-- 
 Total strangers need love, too; and I'm stranger than most.
Sean Reifschneider, Inimitably Superfluous <jafo at tummy.com>
tummy.com - Linux Consulting since 1995. Qmail, KRUD, Firewalls, Python



From MarkH at ActiveState.com  Sat Mar 10 19:58:04 2001
From: MarkH at ActiveState.com (Mark Hammond)
Date: Sun, 11 Mar 2001 00:58:04 GMT
Subject: Crashed PythonWin. How can I fix it?
References: <3aa04cf5.34464140@News.CIS.DFN.DE>
Message-ID: <3AAACCDE.80703@ActiveState.com>

Costas Menico wrote:

>     import glob
>   File "glob.py", line 11, in ?
>     glob2.what2()
>   File "glob2.py", line 3, in what2
>     print x
> exceptions.NameError: There is no variable named 'x'

It looks alot like a "glob.py" is being found that isnt the glob.py in 
the standard library.

Mark.



From jsoons at juilliard.edu  Tue Mar 27 09:49:39 2001
From: jsoons at juilliard.edu (Jonathan Soons)
Date: Tue, 27 Mar 2001 09:49:39 -0500
Subject: Changing every other instance of <B> in a file
Message-ID: <B7C74E1A6FBAD411BFC60002B309DDF4E43E@mailbox.juilliard.edu>

This seems to fail for tags that do not have a closing tag, which occurs
frequently in non-automated HTML.


-----Original Message-----
From: Lars Kl?boe [mailto:larskl at klassekampen.no]
Sent: Tuesday, March 27, 2001 9:28 AM
To: python-list at python.org
Subject: Re: Changing every other instance of <B> in a file


On Tue, 27 Mar 2001 05:56:26 -0800, "Emile van Sebille" <emile at fenx.com>
wrote:
>Here's one way (although I'm interested to see how re might do this):

[--- interresting code ---]

This is another (fancier ;)) implementation of the solution I've
been working on, but I still refuse to believe that the script has
to be this elaborate/big. *twisting my tired gray matter*

I guess re would do the trick (at least if combined with a simple
loop, I'll look into it when I'm really awake.

Thanks for your input !

-- 
Lars Kl?boe
-- 
http://mail.python.org/mailman/listinfo/python-list



From ransen_spam_me_not at nemo.it  Sat Mar  3 06:12:28 2001
From: ransen_spam_me_not at nemo.it (Owen F. Ransen)
Date: Sat, 03 Mar 2001 11:12:28 GMT
Subject: Python 2.1b1 released - debug versions?
References: <mailman.983573701.16010.clpa-moderators@python.org>
Message-ID: <3aa1cb88.4366424@news.newsguy.com>

>Find out all about 2.1b1, including docs and downloads (Windows
>installer and source tarball), at the 2.1 release page:
>    http://www.python.org/2.1/

I was hoping to try out Python 2.1 in order to see if
it fixed the Py_Finalize() bug and the PyRun_SimpleFile()
crash, but when I downloaded the windows installer
Python-2.1b1.exe I can't find the debug library and
dll, which I suppose would be called:

  python21_d.lib
  python21_d.dll

Any pointers anyone? I probably do not yet know my way
around Python releases, please bare with me!



--
Owen F. Ransen
http://www.ransen.com/
Home of Gliftic & Repligator Image Generators


From Karl.Bellve at umassmed.edu  Thu Mar 22 14:27:54 2001
From: Karl.Bellve at umassmed.edu (Karl Bellve)
Date: Thu, 22 Mar 2001 14:27:54 -0500
Subject: Debug and NonDebug versions of Python.DLL loading...
Message-ID: <3ABA523A.95D362FE@umassmed.edu>



I am linking in Python into my application. But, when I run the
application in DEBUG mode, it tries to load Python.dll and Python_d.dll
and then does an "Automatic DLL Relocation." I am not quite sure why the
python.dll is being loaded. This is with Python 2.1b1.

MSVC Studio.
Windows NT 4.0SP6

-- 
Cheers,



Karl Bellve



From mikael at isy.liu.se  Fri Mar  2 09:58:07 2001
From: mikael at isy.liu.se (Mikael Olofsson)
Date: Fri, 02 Mar 2001 15:58:07 +0100 (MET)
Subject: Splitting comp.lang.python
In-Reply-To: <3A9FA38B.1E5E03E2@olsen.ch>
Message-ID: <XFMail.20010302155807.mikael@isy.liu.se>

 >  Warren Postma wrote:
 > > Not again!?  We had this discussion ten times over. [snip]

On 02-Mar-2001 Nick Jacobs wrote:
 >  But comp.lang.python gets more traffic than it used to. The case for
 >  splitting it is stronger than it was.  IMHO it is almost inevitable that
 >  at some point in the not-very-distant future, it will be split. The
 >  meaningful debate is about when and how, not about whether.

The last time I remember discussing this was about a year ago. At that
particular time there was approximately 1000 postings a week, exacly as
it is today. Then the rate has gone down and up again. I'm not totally
against a split, but we did not split the group then, and will probably 
not do that now either. I say: Come back when the rate has really gone 
up. Then maybe you have a case.

/Mikael

-----------------------------------------------------------------------
E-Mail:  Mikael Olofsson <mikael at isy.liu.se>
WWW:     http://www.dtr.isy.liu.se/dtr/staff/mikael               
Phone:   +46 - (0)13 - 28 1343
Telefax: +46 - (0)13 - 28 1339
Date:    02-Mar-2001
Time:    15:51:31

         /"\
         \ /     ASCII Ribbon Campaign
          X      Against HTML Mail
         / \

This message was sent by XF-Mail.
-----------------------------------------------------------------------


From bencr at bigpond.com  Wed Mar 28 00:14:00 2001
From: bencr at bigpond.com (Ben)
Date: Wed, 28 Mar 2001 15:14:00 +1000
Subject: Controling the execution of external programs on Win2K
Message-ID: <n6dw6.4720$45.24451@newsfeeds.bigpond.com>

Hi,

I placed a post last night looking for a way to run exe's from a Python
programme, where the Python programme waits for each instance of the exe to
finish/return and exit code before proceding with it's next command.

As suggested I have looked at os.system( ) but on Win2K it seems every
external program I execute spawns an external command shell and there
doesn't seem a way to prevent this behaviour ... which is not good (atleast
from what I am trying to do) ...  after a bit more investigation I believe
it is possible to achieve this using the win32process module. I am just
unsure of what args to pass to make it behave in this way.

If anyone could provide further suggestions I would be most grateful :)

# Here is the piece of test code I have been toying with

import win32process



def CreateProcess(commandline):
                              sui = win32process.STARTUPINFO()
                              dummyvalue = win32process.CreateProcess(

None,

commandline,

None,

None,

0,

win32process.NORMAL_PRIORITY_CLASS,

None,

None,

sui)

# Spawns first executable ... python programme pauses until program1.exe
finishes running and closes

 CreateProcess("Programme1.exe /args")

# Once Programme1.exe returns an exit code ... the Python programme resumes
and Programme2 starts

 CreateProcess("Programme2.exe /args") #Once the instance of




From apardon at trout.vub.ac.be  Tue Mar 27 02:59:46 2001
From: apardon at trout.vub.ac.be (apardon at trout.vub.ac.be)
Date: 27 Mar 2001 07:59:46 GMT
Subject: Is this an inconsistency?
References: <slrn9bud88.2le.apardon@trout.vub.ac.be> <slrn9budgl.mbh.scarblac@pino.selwerd.nl>
Message-ID: <slrn9c0ieo.2le.apardon@trout.vub.ac.be>

On 26 Mar 2001 12:32:05 GMT,
Remco Gerlich <scarblac at pino.selwerd.nl> wrote:
>apardon at trout.vub.ac.be <apardon at trout.vub.ac.be> wrote in comp.lang.python:
>>   map(None , list)
>> 
>> Doesn't turn into a list of singletons but
>> just returns the list.
>> 
>> This seems a bit odd to me because if you
>> use apply with a single parameter function
>> the second argument has to be either a singleton
>> or a list with one element.
>> 
>> 
>> My question now: Is this behaviour changed in
>> the following versions? If not, wouldn't it be
>> more consistent if it was?
> 
>It's inconsistent if you see it this way, yes. On the other hand, if you
>think of different functions to call map with that take one argument, it's
>pretty nice that map(None, l) just returns l unchanged. It would be weird if
>the 'None' function made tuples.

It  wouldn't be the None function that made tuples. It is the map
function that prepares tuples so that they can be applied to  the
function.  So  if  the map function prepares tuples, if more than
one list is given it should IMO prepare singletons if only a sin-
gle list is given.

One  should also consider should one for the sake of the exercise
try to implement map onseself, with the help of the  apply  func-
tion, that apply expects a singleton if used with a single param-
eter, so that it is logical that the map function would deliver a
singleton for processing, which the None function would leave un-
changed.

-- 
Antoon Pardon


From aleaxit at yahoo.com  Wed Mar 21 10:34:37 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Wed, 21 Mar 2001 16:34:37 +0100
Subject: Checking the type
References: <Pine.LNX.4.30.0103211504200.15322-100000@fep132.fep.ru> <mailman.985177829.12938.python-list@python.org>
Message-ID: <99ahma01smr@news2.newsguy.com>

"Moshe Zadka" <moshez at zadka.site.co.il> wrote in message
news:mailman.985177829.12938.python-list at python.org...
> On Wed, 21 Mar 2001 15:05:10 +0300 (MSK), Oleg Broytmann <phd at phd.pp.ru>
wrote:
>
> >    Yes, builtin fuction type(x):
> >
> > if type(x) <> type([]):
> >    print "not list!"
>
> Or even better, isinstance
>
> if not isinstance(x, types.ListType):
> print "not list!"

Much more elegant (it will have identical semantics
in current Python, where types can't be inherited
from, but keeps a door open to future extension...).


Alex





From cribeiro at mail.inet.com.br  Tue Mar 27 12:19:15 2001
From: cribeiro at mail.inet.com.br (Carlos Alberto Reis Ribeiro)
Date: Tue, 27 Mar 2001 14:19:15 -0300
Subject: Yet Another PEP:  Query Protocol Interface or __query__
In-Reply-To: <Pine.LNX.4.21.0103270930100.24308-100000@clarkevans.com>
References: <Pine.LNX.4.21.0103270747160.17716-100000@clarkevans.com>
Message-ID: <5.0.2.1.0.20010327120719.02185e30@mail.inet.com.br>

Warning: this is a long message, but I think it's worth the time :-)

It's good to know that my Adapter/Proxy implementation was useful. I'm 
quite surprised also, because I'm relatively new to Python, but I found 
this technique to be extremely flexible while keeping code simple and 
readable ("pythonically correct" :-) I dreamed of similar things in my 
Object Pascal/C++ days...

As for your example: the proposal seems to be moving toward to a more 
intelligent protocol adapter. I think it is better to have 'dumb' adapters 
(in fact as dumb as possible), and helper methods to do the adaptation. In 
fact i remember some thoughts by JvR and Alex Martelli about a related 
topic - what to put inside constructors, and why it's good practice to use 
class factories. So I'm advocating the following definitions, with a 
thorough explanation following:

1) the "protocol specification" (which may be given as a class,
    interface, or simply as a list of methods);

2) the "protocol instance", which may be any Python object that
    conforms to the protocol specification;

3) the "adapter", or "proxy", which is a special Python object
    that implements *only* the protocol specification, and
    nothing more. The adapter hides all details; the adapted
    object is "opaque".

4) the "protocol identifier" is a unique string that can be used
    whenever a protocol specification is desired.

In the end we talk a little bit about some efficiency issues I'm worried about.

-----
1) What is the protocol specification

You see that I'm using the word "protocol" in a very broad meaning here. A 
protocol is an "abstract" beast, and it may be specified by a number of ways.

1.1) We can use a class to specify the protocol. For instance, the UserDict 
class can be used as the specification of the generic dictionary protocol.

1.2) We can use a interface to specify the protocol. In fact I dont see 
much difference between an "abstract class" defined with stub methods and a 
interface declaration. There are a few points that PEP245 is already 
tackling, such as different behavior when inheriting class and implementing 
interfaces. Also note that interfaces are a good point to check for method 
signatures (parameter types and so on).

1.3) A list of methods is a very simple and effective way to describe a 
interface. In fact, while we don't have any synctatic support for 
interfaces, I think that a simple list of strings is the best approach for 
protocol specification. It allows us to focus on the adaptation layer, 
while leaving the rest of the discussion for the types-sig and PEP245 guys 
<wink>.


It is relatively easy to convert between these three formats. I'm thinking 
about ways to support all the three approaches in my adapter/proxy module. 
I have done some work on it already; I'll post it soon.

-----
2) What is the protocol instance

2.1) Any Python object that implements a protocol is a protocol instance. 
In this sense, any dictionary - built in or inherited from UserDict - is an 
instance of the dictionary protocol.

2.2) Using this definition, adapt() job is to *return a protocol instance*. 
It may be the object itself (returning self in the __adapt__ module); it 
may be another generic object that handles the adaptation between the 
desired interface and the one implemented in the object; or it may be a 
special adapter built on the fly.

2.3) The code that I have running today is equivalent to the last case as 
outlined above - it is the special "adapter/proxy" object.

-----
3) What is an "adapter/proxy" object

It is possible to build a new object in Python, on the fly, to act as a 
proxy to another object.

3.1) This proxy object does not have methods of it's own. Instead of having 
methods bound to itself, all the methods it contains are bound to other 
objects. So the object ends up acting as a transparent proxy for other objects.

3.2) The transparent proxy is as efficient as possible, because there are 
no intermediate calls. All the calls go directly to the target object.

3.3) The resulting proxy object is opaque. It hides implementations details 
from the user. This is considered to be a good practive in object-oriented 
programming.

3.4) A generic adapter factory function can be used to build a new adapter 
from the "protocol specification", given in any of the three ways described 
in (1.1), (1.2) or (1.3). In fact, this is what my helper funciont 
currently does.

3.5) I would call "proxy" objects of this type that dont implement any 
behavior of their own. "Adapters" may use similar techniques, but they can 
implement supplementary methods. In some cases the "adapter" behavior may 
be needed, but in *most* cases, I think that the proxy approach is sufficient.


I believe that proxy objects built this way are the best answer for "adapt" 
calls, also for security reasons. I'll give an example:

Suppose that you have a class to handle user account profiles. You have two 
interfaces - one is the "public" one, and the other is the "administrator" 
interface. Your class could return an interface for public code to give 
visibility only to the public functions. To get to administrative 
interface, you must "log in" in some meaningful way. However, if you just 
return "self" in any case, it is fairly easy for "public authorized" code 
to figure out how to access the administrative functions, because they're 
all visible by anyone. Worse - this can happen by mistake.

What is the advantage of using interfaces to do that, instead of using, let 
us say, two different classes? The main advantage that I see is that we can 
keep the data syncronized between the two "protocol instances", because 
they both are pointing to the same target object.

Note that using Python, it is possible to follow the pointers through the 
method (using im_self) to get to the target object. However, this is not 
immediatelly obvious, and it at least avoid that someone end up using wrong 
methods by mistake. If *absolutely* needed, we could devise a way to deny 
access to the im_* attributes of the method through a proxy object, but 
this is beyond the scope of this discussion.

-----
4) What is a protocol identifier

The first draft if this PEP used strings as a parameter to specify the 
protocol in the __adapt__ call. Somewhere along it's development, this has 
been changed to a class parameter. However, it is not clear that this is 
sufficiently flexible.

My proposal is to define a unique protocol identifier as a string with a 
very-well defined format:

4.1) The protocol identifier is a unique string that identifies every protocol

4.2) The protocol identifier should be structured in a way that makes 
parsing easier for automated tools. It also must be usable as means of 
documentation inside the program. In the other hand, it must be kept simple 
to avoi clutter, keeping the code clear and easy to understand. There are 
two main proposals:

4.2.1) Using a DNS-like syntax. For example, org.python.UserDict;

4.2.2) Using a URL-like syntax. For example, python.org/protocol/userdict.xml.

The second approach has one advantage. It makes it possible to use the URL 
returned to actually retrieve some meaningful description of the object. Of 
course, this can't be mandatory, because not all applications are 
web-enabled; but anyway it makes fairly easy for the developer to locate 
and retrieve the documentation.

4.3) When presented by means of classes or interfaces (cases 1.1 and 1.2 
above), we should define an attribute called __protocol__ to retrieve the 
protocol identifier.

4.4) A 'protocol factory' could be implemented as a standard module. This 
module would contain:

4.4.1) One dictionary of well-known "protocol specifications", indexed by 
the "protocol identifier";

4.4.2) One function to automatically retrieve the XML representation of the 
protocol and use it to generate a valid "protocol specification", that 
could in turn be supplied to the adapter factory function.

4.4.3) One function to generate the XML representation of any give protocol 
specification (ok, this is not exactly a factory function, but is related 
to the job).

-----
Efficiency issues

I'm a little bit afraid of handling so much information and processing 
inside __adapt___ methods. These methods can be very useful, and there is a 
possibility to start using them a lot as part of a 
'programming-by-contract' paradigm. There are some ways to solve this problem:

1) Building helper structures on the __init__ method to make for a faster 
lookup inside __adapt__. For now I think that this is the best solution.

2) *If, and when* we get interfaces as part of the language, it is possible 
to dream about some optimizations. Namely, interface method tables could be 
implemented as direct access vectors instead of dictionaries. This can be 
done if we think of interfaces as the "immutable" counterpart to "classes".



Carlos Ribeiro





From claird at starbase.neosoft.com  Tue Mar  6 13:27:43 2001
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 6 Mar 2001 12:27:43 -0600
Subject: Questions from a beginner
References: <1103_983792888@breakdown> <7D924B83F563035A.78ADE854BCA8F2C6.DA35B8F7D112D810@lp.airnews.net>
Message-ID: <BAC83A82CDF10F60.77025EDCDF82B08B.EF81409F68BD8FE6@lp.airnews.net>

In article <7D924B83F563035A.78ADE854BCA8F2C6.DA35B8F7D112D810 at lp.airnews.net>,
I abbreviated:
Cameron Laird <claird at starbase.neosoft.com> wrote:
>In article <1103_983792888 at breakdown>,
>Alison Myers  <squirmy at freeuk.com> wrote:
>>Hi. I'm a beginning programmer at a fairly late age and have been looking for a language to learn.
>			.
>			.
>			.
>Most of the answers are, "Yes."  I've argued before that
>Python is the best first language
><URL: http://www.oreillynet.com/pub/a/network/2000/06/02/magazine/python_first_language.html >.
			.
			.
			.
A correspondent, with a return address I couldn't get to
work, wrote me privately (?) to wonder why I didn't mention
LISP, which is the *real* language-done-right for
Gabrielites <URL: http://www.naggum.no/worse-is-better.html >.
Well, Scheme's my second choice for a first language <URL:
http://www.sunworld.com/sunworldonline/swol-10-2000/swol-1013-regex.html >
Why only second?  For me, it's 'cause Python remains easier
to embed-and-extend, and has a wider range of what I regard
as significant GUI toolkit bindings, and Win* conveniences.
Also, I still have mixed feelings about Scheme in large-team
projects.  I can see Scheme making great advances on all
these fronts, though ...

I like to think Scheme and Python complement each other in
making the pie bigger, rather than squabbling over how to
cut the pieces.
-- 

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


From kmoffat at drizzle.com  Sat Mar 10 10:48:29 2001
From: kmoffat at drizzle.com (Ken Moffat)
Date: Sat, 10 Mar 2001 07:48:29 -0800
Subject: is it possible to use Tkinter on linux
References: <98714n$3p8$1@news2-2.kornet.net> <Xj_p6.189678$df5.5158755@news1.crdva1.bc.home.com> <GJaq6.266$Tg.42959@ruti.visi.com>
Message-ID: <3AAA4CCD.5B4AA238@drizzle.com>

Grant Edwards wrote:

> Tkinter wasn't enabled by default in the build process, but
> don't most Linux's come with Tkinter already installed? 

Caldera comes with tkinter enabled. (eDesktop2.4, at least.)
-- 
Ken Moffat
kmoffat at drizzle.com


From robin at stop.spam.alldunn.com  Wed Mar 28 17:33:17 2001
From: robin at stop.spam.alldunn.com (Robin Dunn)
Date: Wed, 28 Mar 2001 14:33:17 -0800
Subject: How to update status bar in wxPython?
References: <5Tpw6.2172$K%.667185@typhoon1.ba-dsg.net>
Message-ID: <OEtw6.238$nyH.5111956@news.randori.com>

"Steven Sartorius" <ssartor at bellatlantic.net> wrote in message
news:5Tpw6.2172$K%.667185 at typhoon1.ba-dsg.net...
> I'm a newbie to GUI programming in general so apologies if this is a
FAQ....
>
> I've got a Python script that does some heavy duty number crunching and
file
> I/O.  To track the progress of the script I print a series of messages to
the
> console/x-term ("Working on file #....", etc,etc).  As a project, I
decided to
> guify the script with wxPython and Boa-constructor.  I've successfully
built a
> window with a single button, a couple of text input fields and a status
bar;
> clicking the button fires off the script.  What I'm trying to do (so far
> unsuccessfully) is redirect the progress messages from the console to the
> status bar on the window.  Because I'm retrofitting an existing script,
I've
> just replaced the original 'print' statements with
> 'statusBar.SetStatusText("Working on file....") and (for good measure)
> 'statusBar.Refresh()'.  This approach kind of works....the status bar
shows
> the final progress message ("All data processed")  but only after I force
the
> window to redraw (moving it or covering and then uncovering it).  I'm sure
I'm
> missing something basic here so any help would be much appreciated.
>

When you click the button and "fire off the script" you are effectivly
blocking the GUI until the script is complete.  You need to allow the GUI's
main loop to run in order for new events to be dispatched and other
housework types of things to happen.  When you call statusBar.Refresh() you
are simply requesting that a paint event be sent ot the statusBar, but as
you've seen if the event is never dispatched to the window then it is never
actually refreshed.

One way to allow the main loop to run from within the middle of some long
running task is to call wxYield, which will return when there are no more
events waiting to be proccessed.  Unfortunatly there are potential
reentrancy issues you need to deal with when using wxYield, for example what
if the user clicks on your launcher button while a script is already
running, and then the code calls wxYield?

A better way is to refactor your long running task so that it can be
executed a chunk (or iteration) at a time from an EVT_IDLE event handler.
Each time the handler is called do the next chunk of the task and if there
is more to be done call event.RequestMore() before returning.  This lets the
code return to the main loop frequently so the GUI continues receiving
events and can be updated, but all extra time is spent running your task.

If the task is not easily broken into chunks then you can run it in a
background thread, but then there are a few things you need to do specially
to allow your thread to communicate back to the GUI thread.  See the demo
and also the wxpython-users archives for details.

--
Robin Dunn
Software Craftsman
robin at AllDunn.com       Java give you jitters?
http://wxPython.org      Relax with wxPython!





From kuntsi at cc.helsinki.fi.spam.no  Fri Mar 23 04:30:46 2001
From: kuntsi at cc.helsinki.fi.spam.no (Antti Kuntsi)
Date: 23 Mar 2001 09:30:46 GMT
Subject: close(), exceptions and problems
Message-ID: <slrn9bm5u5.klj.kuntsi@myntti.helsinki.fi>

Hi,
I have run into the following problem with python and close()s. When the
system is under a heavy load the close() system call can be interrupted,
but the Python never detects this. This leads into a lot of problems,
including at least the following:

socket object: the socket is not closed, and the FD reserved for the
connection is left to the process, but the FD-number is lost.

file object: the file is not closed, FD is left unreferenced.

the pipes from popen2.popen3(): The pipe(s) are left open, eg a sub-process
waiting for EOF waits forever.

So far I have checked the implementation of the socket object .close() -method
and it ignores the return value from close().

As a result Python becomes extremely unstable under if the system is under
a heavy load. Circumventing this at python-level does not work, as the
following code can go thru without executing the except-branch in a
multithreaded application regardless of if the close fails or succeeds.
Here's a quick piece of code demonstrating how not to avoid the problem,
even with stricter except:s you cannot guarantee that the socket has been
closed.

import os
...
def badClose(socket_object):
	fd=socket_object.fileno() # we loose the FD number in close()
	stats=os.fstat(fd)
	# Today we don't know if this succeeds or not
	socket_object.close() 
	try:
		if os.fstat(fd) == stats:
			raise Exception, "The socket was not closed"
# or so you'd think. the fstat() results can be identical, at least on
# Solaris 7. The socket could have been closed and a new socket with the
# same ctime and inode might have already been given out between the
# close() and the seconds fstat()
	except: # fstat() throws an exception if the FD is bad
		print "Socket was closed"

This seems to affect at least Python 1.5.2 and Python 2.0.


Antti Kuntsi

-- 
         /"\                           |    iki.
         \ /     ASCII Ribbon Campaign |    fi/
          X      Against HTML Mail     |    mickut
         / \


From donn at oz.net  Fri Mar  2 00:55:58 2001
From: donn at oz.net (Donn Cave)
Date: 2 Mar 2001 05:55:58 GMT
Subject: [Python-Dev] Re: d = {}; d[0:1] = 1; d[0:1] = 2; print d[:]
References: <15006.46879.607192.367739@w221.z064000254.bwi-md.dsl.cnc.net> <200103012101.QAA18516@cj20424-a.reston1.va.home.com> <mailman.983485324.29650.python-list@python.org>
Message-ID: <97ncle$tef$1@216.39.151.169>

Quoth Thomas Wouters <thomas at xs4all.net>:
|
| [ Dicts accept slice objects as keys in assignment, but not in retrieval ]
|
...
| I think making slices hashable objects makes the most sense. They can just
| be treated as a three-tuple of the values in the slice, or some such.
| Falling back to just sq_item/__getitem__ and not mp_subscript might make
| some sense, but it seems a bit of an artificial split, since classes that
| pretend to be mappings would be treated differently than types that pretend
| to be mappings.

The problem for me is that when we write ob[i:j], we clearly
must have expected that ob is a sliceable sequence.  If it is
instead a mapping, then the expression has no sensible meaning,
so it's an error and ought to raise an exception.

I'm not all that devastated to hear that my ceval.c hack wasn't
the right thing, it was a pretty hasty experiment.  But I do
think that ob[i:j] should raise an exception for dict objects.
I don't care so much if ob[i:j:1] does, but I would if that
notation started working for lists and tuples.

I think it's OK to apply this logic to mapping types, and defer
it for mapping classes.  It's up to the class implementor to
decide whether a slice is meaningful there.

	Donn Cave, donn at oz.net


From shonn at shonn.com  Wed Mar  7 00:54:24 2001
From: shonn at shonn.com (Shonn Gilson)
Date: Tue, 6 Mar 2001 21:54:24 -0800
Subject: Python, ASP, ADO, and Image string/binary problems
References: <ta0amgebq0sn0b@corp.supernews.com>
Message-ID: <tabjei1p4a9481@corp.supernews.com>

It looks like I've figured out what part of the problem is but I don't have
a solution (well I do, but it's really ugly). Oh, and I'm new to python
coming from a VB background.
Okay, Python is processing the strings right (the tutorial didn't specify
that '\377' is octal not decimal.). The problem seems to be that when Python
passes strings through COM that they get translated to Unicode. This happens
for things like Response.BinaryWrite along with things like recordsets
(objRS = Server.CreateObject("ADODB.Recordset")).  The result is that the
beginning of my jpg file looks like
'\377\000\330\000\377\000\340\000\000\000\020\000J\000F\000I\000F' instead
of '\377\330\377\340\000\020JFIF'.

Does anyone know how to handle this?

Shonn

"Shonn Gilson" <shonn at shonn.com> wrote in message
news:ta0amgebq0sn0b at corp.supernews.com...
> Okay, here it goes.
> I have a site in ASP where I need to upload an image (already working) and
> then make a thumbnail of it. The image is stored in a MS SQL Server
> database, I'm using IIS5 (Win2k), and the python is Active State's with
all
> the COM functionallity added.
> Here's what I'm trying to do:
>
> <script runat="server" language="Python">
> import os, sys
> import PIL.Image
> import StringIO
>
> strSQL = "select * from [FILE] where FILE_Guid =
> '406D58D8C4EE4FA28DF1CFC988A49C11'"
> objConn = Server.CreateObject("ADODB.Connection")
> objRS = Server.CreateObject("ADODB.Recordset")
> objConn.Open (connection_string)
> objRS.Open (strSQL,objConn,3,1)
> bstrIn = objRS("FILE_File").value # returns a buffer, not a string
> im = PIL.Image.open(StringIO.StringIO(bstrIn))
> im.thumbnail((128, 128))
> bstrOut = StringIO.StringIO()
> im.save(bstrOut, "JPEG",optimize=1,progressive=1)
> Response.ContentType = 'text/html' #'image/jpeg'
> Response.Write(" ")
> Response.BinaryWrite(bstrOut.getvalue())
> #Response.Write ('size')
> #Response.Write (`im.size`)
> objRS.Close()
> objConn.Close()
> </script>
>
>
> If I do a Response.BinaryWrite(bstrIn) I get the image in the browser but
if
> I do Response.BinaryWrite(bstrOut.getvalue()) I get spaces between every
> character (I've seen this with unicode problems). It's not just an
> incompatibility with Response.BinaryWrite because I've written the results
> back to the database with the same results. It seems like it's a problem
> with how python deals with strings.
>
> Anyone have any ideas?
>
> Thanks in advance
>
> Shonn
>
>




From fredrik at pythonware.com  Wed Mar 21 02:40:49 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 21 Mar 2001 07:40:49 GMT
Subject: Base 2 to long integers and back
References: <mailman.985157126.16257.python-list@python.org>
Message-ID: <5WYt6.846$sk3.282417@newsb.telia.net>

Tim Churches wrote:
> but there is no built-in or library function which does the
> reverse, it would seem. However your code does the trick
> nicely and quite quickly

for larger numbers, something like this is 8-10 times
faster:

binmap = []
for i in range(256):
    binmap.append(
        "%d%d%d%d%d%d%d%d" % (
            (i&0x80)!=0, (i&0x40)!=0, (i&0x20)!=0, (i&0x10)!=0,
            (i&0x08)!=0, (i&0x04)!=0, (i&0x02)!=0, (i&0x01)!=0
            )
        )

def f2(val, joiner="".join, binmap=binmap):
    out = []
    append = out.append
    while 1:
        append(binmap[val & 255])
        val >>= 8
        if not val:
            break
    out.reverse()
    return joiner(out)

Cheers /F

<!-- (the eff-bot guide to) the standard python library:
http://www.pythonware.com/people/fredrik/librarybook.htm
-->




From root at [127.0.0.1]  Wed Mar 28 08:28:45 2001
From: root at [127.0.0.1] (deadmeat)
Date: Wed, 28 Mar 2001 13:28:45 GMT
Subject: list vs tuple
Message-ID: <hGlw6.782$p5.3210@news1.rivrw1.nsw.optushome.com.au>

Whats the practical difference between a list[] and a tuple() ?

A tuple seems to be a very basic list, so unless it's faster, what's the
point of it?





From xyzmats at laplaza.org  Sat Mar 10 12:54:58 2001
From: xyzmats at laplaza.org (Mats Wichmann)
Date: Sat, 10 Mar 2001 17:54:58 GMT
Subject: Urgent Question about Python licensing
References: <98cpfe$5oj$0@216.39.170.247> <kupufpiu2e.fsf@lasipalatsi.fi>
Message-ID: <3aaa6973.5847127@news.laplaza.org>

On 10 Mar 2001 16:20:57 +0200, Erno Kuusela <erno-news at erno.iki.fi>
wrote:

>In article <98cpfe$5oj$0 at 216.39.170.247>, whisper at oz.net (Dave
>LeBlanc) writes:
>
>| Hi;
>| While reading up on the recent Python conference on O'Reilly's site
>| (http://python.oreilly.com/news/pythonday1_0301.html - actually on the
>| 'day 2' page), I cam across the following:
>| ----------------------------------------------------
>[...]
>| This difficulty means that Python
>| can't be distributed with software licensed under the GPL.
>
>that is false. the issue is not distributiong python with gpl software,
>it is mixing the python interpreter code with gpl'd code. (ie where
>the other code and the python interpreter would combine to produce
>what the gpl refers to as a "derivate work").

That's right.

This all /ought/ to be straightforward: a GPL'd piece of code is
'free' and through being combined with some other code, it may not
lose (any of) that freedom.  Putting GPL'd code into the Python
interpeter would cause it to come under the terms of the Python
license, thus changing the terms under which that piece of GPL'd code
was available, and that's not allowed unless the licenses are
considered "compatible", which has been one of the hopes for Python.

The issue seems to be more complicated than this though, because there
are people still worried about including new Python versions on Linux
distributions because of the CNRI license, and I don't fully get why
this is yet (not sure I want to, either).  Like other folks have said,
the subtleties will make your brain explode (always pick up a new one
at Curry's, though).



From gzeljko at sezampro.yu  Tue Mar 20 15:04:48 2001
From: gzeljko at sezampro.yu (gzeljko)
Date: Tue, 20 Mar 2001 21:04:48 +0100
Subject: Sequence-spreading
References: <3AB7167F.2816AA21@student.liu.se> <99790u01c73@news2.newsguy.com> <99851p$ba6$2@neptun.beotel.net> <998ba7$5j7$1@news.mathworks.com>
Message-ID: <998d5u$cqj$1@neptun.beotel.net>

Joshua Marshall <jmarshal at mathworks.com> wrote in message
news:998ba7$5j7$1 at news.mathworks.com...
> gzeljko <gzeljko at sezampro.yu> wrote:
>
> > Alex Martelli <aleaxit at yahoo.com> wrote in message
> >> offsets = [ [] for i in range(pm) ]
> >> for i in range(len(arg)):
> >>     offsets[i%pm].append(i)
> >>
> >>
> >> Alex
> >>
>
> > Old syntax yet works:
>
> > offsets = [[]]*pm
>
> Careful here:
>
>  >>> x = [[]]*2
>  >>> y = [[] for i in range(2)]
>  >>> x
>  [[], []]
>  >>> y
>  [[], []]
>  >>> x[0] is x[1]
>  1
>  >>> y[0] is y[1]
>  0
>
> The first way gives you the same value at each element ("[]" is
> evaluated once), while the second way gives you distinct values ("[]"
> is evaluated for each element).
> --

[Careful]*3

Thanx.

ly-y'rs-gzeljko








From michel at digicool.com  Mon Mar 26 17:50:29 2001
From: michel at digicool.com (Michel Pelletier)
Date: Mon, 26 Mar 2001 14:50:29 -0800
Subject: PEP 245: Python interfaces
In-Reply-To: <99o9g4$nh3$0@216.39.170.247>
References: <Pine.LNX.4.32.0103222253510.26302-100000@localhost.localdomain> <99o9g4$nh3$0@216.39.170.247>
Message-ID: <Pine.LNX.4.21.0103261430360.25260-100000@localhost.localdomain>

On 26 Mar 2001, Dave LeBlanc wrote:

> Having read the PEP briefly, I have one question:
> 
> Is there a good reason to add the "implements" keyword?
> 
> Current syntax is (from Language Reference 7.6):
> classdef:       "class" classname [inheritance] ":" suite
> inheritance:    "(" [expression_list] ")"
> classname:      identifier
> (i'm adding this based on the text of 7.6:)
> expression_list: Expression(s) which evaluate to a class object.
> 
> Any reason why expression_list couldn't include an interface? 

Because that expression list defines inheritance, and interfaces are not
inherited.  They do not define any of the same behaviors, like shared
implementation, as inheritance.  Readers of your code would (wrongly)
assume that you were mixing in implementation into your class instead of
asserting the implementation of a certain interface.  

> Seems to
> me it's easier to just extend the semantics of expression_list to
> include interface_name(s) instead of adding a variation like '"class"
> classname [inheritance] ["implements"   interfaces] ":" suite'.
> (Besides, two optional elements in a row makes for messy parsing :-).)

You mean human visual parsing or the actual Python parser?  I don't know
about the former, but the syntax is no problem for the latter.

> The end of a class definition might be slightly more complex to check
> that all "pure" methods have implementations, but (almost certianly)
> not hopelessly so... and you get to detect the "not implemented" error
> early with a "missing definition" error rather then later with a
> "can't instantiate" error.

You need to re-read PEP 245; no such compile time checking or error
conditions are proposed, and I think that requireing that checking is a
bad idea (in addition to being impossible for anything but the most naive
cases, since you can change a class at run time).

These are not like Java interfaces.  Interface enforcement can certainly
be done optionally, but it is not a requirment and it never should be
because a) it is expensive, b) the object should be implicitly trusted
unless you feel otherwise (and then you use optional enforcement) and c)
the problem is unbounded: at first you want to check just attributes, and
then parameters, and then assertions, and then types, whoa!  Everybody's
requirement is different (go read the types-sig and you'll see what I
mean), by making all of these enforcements optional, everybody gets to be
happy with their own.

Java and other static languages get away with this because all interfaces
do is assert compile time enforcement, this would not work for Python
which is much more dynamic.

> Surely a python interpreter could be smart enough to realize that some
> name in the list of inherited objects is actually an interface?

The interpreter? Surely.  The guy reading your code? No way.

Thanks for you comments, if you'd like, I can add your suggestion to the
PEP under the different opinions section.

-Michel



From db3l at fitlinxx.com  Wed Mar 28 13:29:34 2001
From: db3l at fitlinxx.com (David Bolen)
Date: 28 Mar 2001 13:29:34 -0500
Subject: list vs tuple
References: <mailman.985801119.10859.python-list@python.org>
Message-ID: <u66gtzr01.fsf@ctwd0143.fitlinxx.com>

"Sean 'Shaleh' Perry" <shaleh at valinux.com> writes:

> dir_list is declared as a tuple, which is as close to a const one can find in
> python.

I'm not sure I'd agree - I'd say that dir_list is defined to reference
a tuple, but not that dir_list itself is defined as a tuple.  Python
doesn't associate types with the reference labels but with the objects
themselves.  I don't think that I'd read into that definition anything
more than that dir_list was being initialized to point to an immutable
object - not that dir_list itself might never point to any other object.

It may be as close to a const definition as you can get (at least
textually - an object that was used to retrieve the directory list and
thus could exert control over any ability to change that list would be
symantically closer), but if that was the intent, I think some
comments along with the code would be advisable, albeit only
informational.

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


From mix77 at usa.net  Tue Mar  6 06:02:52 2001
From: mix77 at usa.net (Mix)
Date: Tue, 06 Mar 2001 13:02:52 +0200
Subject: Python and Printing
Message-ID: <3AA4C3DC.95EF7249@usa.net>

I have pictures (JPEG) which I get from a database and dump in an
arbitrary directory.
To print a picture I do the following:

      import os
      os.system("lpr lena.jpg")


My problem is that page orientation is always"landscape" instead
"portatrait".  What
switch is there to specify orientation for "lpr" if any?

P.S. I'm working under linux (redhat 7.0 distribution ).




From cce at clarkevans.com  Tue Mar 27 09:40:27 2001
From: cce at clarkevans.com (Clark C. Evans)
Date: Tue, 27 Mar 2001 09:40:27 -0500 (EST)
Subject: Yet Another PEP:  Query Protocol Interface or __query__
In-Reply-To: <Pine.LNX.4.21.0103270747160.17716-100000@clarkevans.com>
Message-ID: <Pine.LNX.4.21.0103270930100.24308-100000@clarkevans.com>

I've started to write an example usage of this PEP.  It is
far from perfect, however it should give you an idea of
the usage.  I could use some help with examples and tightening
this up... 

Thanks!  Clark

...


Example Usage (written but not checked)

    Suppose a user has a function that requires a file-like 
    object.  Here is how adapt can be used to ensure that
    an argument does indeed follow the file protocol:

        import types
        from adapt import adapt
        def first_line(file):
            file = adapt(file,types.FileType)
            print file.readline()

        >>> first_line(open("c:/autoexec.bat"))
        SET WXWIN=F:\PROGRA~1\wx2
        >>> first_line('bad')
        ...
        TypeError: test cannot be adapted to <type 'file'>

    Let us assume that the code above becomes the norm, and
    now the user needs to re-use this code with a BLOB
    stored in a database.  Here is how that would be done.

        class PseudoFile:
            def readline(self):             # read line from BLOB
               return "This is a new line"  # real code goes here
            # (other file functions go here)
            def __conform__(self,protocol):
                if protocol is types.FileType: return self

        >>> first_line(PseudoFile())
        This is a new line

    Let us say that this works for a while, but sooner or later
    the concept of a forward-access read-only file is needed.
    So, perhaps the following "standard interface" emerges:

    class ForwardFile:
        def readline(self): pass
        def read(self,size=-1): pass
        class AdaptRead(ForwardFile):
            def readline(self):  # implement using read
        class AdaptReadline(ForwardFile):
            def read(self,size=-1):  # implement using readline
        class Adapt:
            def __call__(self, obj):
                # objects adaptable to FileType is are compliant
                retval = adapt(obj,types.FileType)
                if retval: return retval
                # objects that have both readline and read are ok
                if getattr(obj,'readline',None) and \
                    getattr(obj,'read',None): return obj
                # wrap objects exposing a readline
                if getattr(obj,'readline',None):
                    retval = AdaptReadline()
                    retval.readline = obj.readline
                    return retval
                # wrap objects exposing a read
                if getattr(obj,'read',None):
                    retval = AdaptRead()
                    retval.read = obj.read
                    return retval
        __adapt__= Adapt()

    This would allow our first-line program to be re-written to
    use the new interface:

        def first_line(file):
            file = adapt(file,ForwardFile)
            print file.readline()

    Note that anything which inherits from ForwardFile is a
    adaptable as well as anything which considers itself a FileType
    More so, do to the __adapt__ class method above, anything
    that has read or readline can be adapted to work!

    Let's say that time passes on, and a record-set class is
    created.  And after the above interface is well in use
    (and now immutable due to wide distribution), the creator
    of the record-set class wishes to make this new class
    substitutable for any ForwardFile.

        class RecordSet:
            def readline(): pass  # some intelligent impl
            def read():  pass     # more intelligent impl
            # other record set specific stuff

    There are a few choices.  First, the RecordSet class
    could inherit from ForwardFile.  Or, the following
    __adapt__ method could be added!

            def __conform__(self,protocol):
                if protocol is ForwardFile: return self

    This has a slight problem in that a dependency is now
    created (ForwardFile must be on the user's system).  This
    can be mitigated with something like the following:

            def __conform__(self,protocol):
                if protocol.__name__ = 'ForwardFile': return self

    Note that readline and read could be put in a wrapper class
    instead (so that they don't clutter the RecordSet class!  
    Regardless, in every case where a ForwardFile is accepted, 
    the adapter mechanism will also pass off for objects of 
    this class.


Questions:  

  1. Is there a way to get a fully-qualified name
     (including the package?)

  2.  What kind of "wrapper helpers" could be made?
      (I like Carlos's examples!)

  3.  Perhaps a __conform__ function would also
      like to know who the caller is... I wonder
      if this would be useful.  Can this be
      obtained automatically?

  3.  Would a __protocol__ item also help?  

          class ForwardFile:
               ...
              __protocol__ = "com.zoober.protocols.ForwardFile"

      This would allow the __conform__ above to be re-written:

           def __conform__(self,protocol):
               if protocol.__protocol__ =
                  'com.zoober.protcols.ForwardFile': 
                      return self
          
      In this way, support for multiple protocols could
      be put in place without requiring said classes/interface
      declarations on the user's box.

  4.  Is a protcol registry needed, or does the
      above suffice?  (I think it's good enough)
      A proposal would be like:

        register_adapter(relevant_protocol, the_handler)

      And then, after exhausting the object and the
      protcol, it could check in the registry...


Thank you all for your help!

Clark




From stephen_purcell at yahoo.com  Tue Mar 27 12:25:20 2001
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Tue, 27 Mar 2001 19:25:20 +0200
Subject: large arbitrary binary data in python
In-Reply-To: <slrn9c1hf7.m5q.wcuddy@lserv.ja10629.home>; from wcuddy@lserv.ja10629.home on Tue, Mar 27, 2001 at 04:46:19PM +0000
References: <slrn9c1hf7.m5q.wcuddy@lserv.ja10629.home>
Message-ID: <20010327192520.F1260@freedom.puma-ag.com>

Wayne Cuddy wrote:
>     1) simply read the data into a python binary string

Yes, see

    http://www.python.org/doc/lib/module-StringIO.html
    http://www.python.org/doc/lib/module-cStringIO.html


>     2) pipe the string to md5sum to verify integrity

Yes, but don't pipe it to md5sum if you can avoid it. Instead, use the
'md5' standard module:

    http://www.python.org/doc/lib/module-md5.html


>     3) pipe the data to a shell command like this "gunzip -dc | dd of=device"


Yep, see os.popen() and the popen2 module:

    http://www.python.org/doc/lib/os-newstreams.html
    http://www.python.org/doc/lib/module-popen2.html


> Is there a limit to the size of a binary string in python?

Yes. The limit is the amount of available virtual memory.

-Steve


-- 
Steve Purcell, Pythangelist
Get testing at http://pyunit.sourceforge.net/
Any opinions expressed herein are my own and not necessarily those of Yahoo



From tim.one at home.com  Sat Mar  3 16:59:01 2001
From: tim.one at home.com (Tim Peters)
Date: Sat, 3 Mar 2001 16:59:01 -0500
Subject: A counter-counter-proposal for PEP 236: #pragma( nested_scopes)
In-Reply-To: <j48zmnxjga.fsf@informatik.hu-berlin.de>
Message-ID: <LNBBLJKPBEHFEDALKOLCCELNJDAA.tim.one@home.com>

[Tim continues nagging Martin to write a PEP; Martin continues resisting]

[Martin von Loewis]
> I'm surprised that you evaluate the present based on what I think
> about the future.

I hope everyone who evaluates any proposed language feature does their best
to consider its likely implications as far into the future as they can.  For
example, apart from explicitly identified unresolved issues in PEP 236, I did
my best to ensure that no addition, change, or deletion would *ever* be
needed to the rest of the PEP.  I may not have succeeded (only time will
tell), but I did my best toward that end.

I want to see the same done for "directive" too -- along with every other
language feature that ever gets proposed.

Besides, no new language feature gets in without a PEP -- that's the rule we
all play by now.  Why should "directive" be exempt?

> I think I've given a specification of the feature as implemented.

If true, it should be trivial to copy the text of your spec into a PEP.  I
haven't seen the spec that I know of (I have seen a blossoming collection of
now hard-to-find-again postings that suggest different variants of
"directive"), and I'm not going to try reverse-engineering it from the patch.
Neither are the Jython developers, who don't normally look at CPython patches
at all.  Etc.  It needs a PEP.

> I'd happily add a specificiation of the directive statement (including
> plans that I have how it should look in Python 2.5.2 :) into PEP 236,
> as a section "Alternative Proposal" - if you allow me to do that.

No, because it doesn't make sense:  if "directive nested_scopes" is accepted,
then PEP 236 moves into the Rejected section, along with all the text you put
into it.  Then you'll need a separate PEP anyway.  Play the game as if you
expect to succeed <wink>.

> I'd rather not author a new PEP, as the directive statement addresses
> the same problem as does PEP 236.

I don't know what the directive statement *is* in the absence of a PEP.  How
many "atoms" does it support in your mind today?  Is "atom" slang or a
deliberate reference to the Ref Man's grammar production of that name?  If
the latter, what sense is there in allowing constructs like

directive [x+1 for x, y in module.foo()
           if x > 3]

?  That is, don't you think the syntax is overly permissive?  Etc:  if you
refuse to spell out your intent in a PEP, nobody has a basis for discussing
it beyond wishful thinking (or reverse-engineering the patch du jour, which
is unacceptable).

What *would* make sense in PEP 236 is to say that PEP 240 (whatever -- the
"directive statement" PEP) subsumes this-and-that parts of a
future_statement's functionality.  The "directive statement" PEP needs to be
a separate entity not only for all the reasons given above, but because you
*don't* intend for it to stop here.  You need a separate PEP so that there's
one clear place to specify the additional bells and whistles that go beyond
the parts of future_statements it subsumes.

last-msg-from-me-on-this-until-there's-a-pep-ly y'rs  - tim




From NOstarfighterSPAM at freeuk.com  Sat Mar 17 11:37:01 2001
From: NOstarfighterSPAM at freeuk.com (Richard Townsend)
Date: Sat, 17 Mar 2001 16:37:01 -0000
Subject: Catching exceptions in Tkinter's mainloop
References: <%dws6.9947$Q47.2723350@news1.rdc1.tn.home.com>
Message-ID: <drMs6.32482$HR6.3694604@nnrp4.clara.net>

"Mike Callahan" <mcalla at home.com> wrote in message
news:%dws6.9947$Q47.2723350 at news1.rdc1.tn.home.com...
> I want to make sure that if any exception pops up inside a mainloop, that
> exception is printed to a log file and the mainloop quits. I tried to
> simulate this with the following code but failed:
>
> from Tkinter import *
>
> class Gui:
>     def __init__(self, master):
>         self.master = master
>         self.button = Button(master, text='Crash', command=self.crash)
>         self.button.pack()
>
>     def crash(self):
>         z = zz  # trigger exception
>
> root = Tk()
> app = Gui(root)
> try:
>     root.mainloop()
> except:
>     print 'caught'
>     root.quit()
>
> It doesn't work. The exception is printed out by standard traceback and
the
> mainloop continues. What am I doing wrong?
>
> Mike Callahan
>

Tkinter catches exceptions raised in callbacks and Tk by using the class
'CallWrapper' (defined in Tkinter.py). I copied the technique used by Pmw,
and created my own CallWrapper class (based on Tkinter.CallWrapper) and then
after initialising Tk, I use the following statement:

    Tkinter.CallWrapper = MyCallWrapper

MyCallWrapper's __call__ method invokes my error logging function in the
final except clause, you could get yours to print info to your log file and
quit.

Note, you must always have an except clause for SystemExit (like
Tkinter.CallWrapper does) otherwise calling sys.exit() in a callback won't
work!

e.g.

class MyCallWrapper:
    """Internal class. Stores function to call when some user
    defined Tcl function is called e.g. after an event occurred."""
    def __init__(self, func, subst, widget):
        """Store FUNC, SUBST and WIDGET as members."""
        self.func = func
        self.subst = subst
        self.widget = widget
    def __call__(self, *args):
        """Apply first function SUBST to arguments, than FUNC."""
        try:
            if self.subst:
                args = apply(self.subst, args)
            return apply(self.func, args)
        except SystemExit, msg:
            raise SystemExit, msg
        except:
            # substitute your functionality here...

Richard






From sholden at holdenweb.com  Tue Mar  6 19:35:53 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Tue, 6 Mar 2001 19:35:53 -0500
Subject: I come to praise .join, not to bury it...
References: <mailman.983890030.6303.python-list@python.org>
Message-ID: <trfp6.29423$1D5.1257189@e420r-atl1.usenetserver.com>

"Mike C. Fletcher" <mcfletch at home.com> wrote in message
news:mailman.983890030.6303.python-list at python.org...
> Here's a professional aesthetic's take:
>
A professional would call themself an aesthete :-)

> mystring.join( )
>
> This is a perfectly normal and good-looking construct.  It seems pythonic
> and beautiful, object orientation is part of Python, and this works
nicely.
>
Agreed.

> "'".join( )
> ";".join( )
> ".".join( )
> "!".join( )
> ",".join( )
>
> All look like executable line noise.  That is, the literal string's syntax
> makes the use of dotted attributes on such a string look jarring and
> off-putting.  You almost expect some weird function to jump out of this
> syntax.  It looks very Ruby/Perl-esque.
>
Just the same, a beginner can read the code as "some string's method is
applied to whatever argument is passed" just by knowing the syntax.

> Note particularly the amount of space between the " character and the .
> character.  Typographically, " characters serve to set the enclosing data
> off from the outer context, and that's the opposite of what you're trying
to
> do with dot notation.  Dot notation normally looks natural because almost
> every character that gets used right before the . comes fairly close to
the
> .  (even characters such as r ).  Also note: in the common cases of
> single-character joins, there's no significant visual weight difference
> between what's inside and outside the quotes).  Consider what happens when
> you alter the proportions of space and the size of the connector...
>
Well, if you like white space Python should be the perfect language for you!

> "."__join( )
> "."-->join( )
> "."..join( )
> " "__join( )
>
> You should find those slightly easier on the eyes (except possibly the " "
> one) because the entire "." entity can be understood as a single graphic
> with a connector to "join".
>
What about the currently syntactically-acceptable

>>> ", " . join(["big", "fat", "joint"])
'big, fat, joint'
>>>

Remember that technically the dot is an operator. Personally I find the
extra whitespace intrusive, but that is a matter of style and taste. Any
better for you?

> That said, I don't think we'll get this particular wart out of the
language,
> so we'll have to deal with the flaw in the way artists have through the
> ages, enhance it and accent it so people start to think it's a beauty
mark.
>
Not a wart, but a consistent application of object-oriented principles.
Sorry. I too wish the world looked more beautiful (to you).

but-if-wishes-were-horses-then-bill-gates-would-have-hoofmarks-in-his-ass-ly
y'rs  - steve





From bowman at montana.com  Fri Mar 30 10:06:20 2001
From: bowman at montana.com (bowman)
Date: Fri, 30 Mar 2001 08:06:20 -0700
Subject: Updated edition of Programming Python out
References: <pj4iat8kevh2fp52gkua87ig16h76oh3m0@4ax.com> <MPG.152d8de631a0ecbf98977b@news-enteract>
Message-ID: <cF0x6.406$bJ2.1654@newsfeed.slurp.net>

Alan Miller <ajm at enteract.com> wrote in message
news:MPG.152d8de631a0ecbf98977b at news-enteract...
>
>   Is there such a thing as a second edition of a computer (or other
>   technical) book that is _thinner_ than the original?

I don't think it is thinner than the first edition, but K&R hasn't gained
much weight in 20, 25 years, whatever the timespan is. The mark of a
successful language, imho. True, it doesn't document libraries and modules,
which seems to be where unchecked growth occurs.





From mcalla at home.com  Mon Mar  5 21:14:59 2001
From: mcalla at home.com (Mike Callahan)
Date: Tue, 06 Mar 2001 02:14:59 GMT
Subject: Simple Tkinter question
Message-ID: <DKXo6.12857$YF4.3052373@news1.rdc1.tn.home.com>

I know this is probably a stupid question but after reading the Tkinter
documentation I am still confused between root.quit() and root.destroy().
What are the differences between these two methods? Thanks.

Mike Callahan





From cce at clarkevans.com  Fri Mar  9 06:50:41 2001
From: cce at clarkevans.com (Clark C. Evans)
Date: Fri, 9 Mar 2001 06:50:41 -0500 (EST)
Subject: Static typing (was Re: Java guy interested in Python)
In-Reply-To: <ud7btyvsn.fsf_-_@advancedmp.net>
Message-ID: <Pine.LNX.4.21.0103090645530.12322-100000@clarkevans.com>

> So, are there any efforts in the Python community to provide something in
> this area?  I'm thinking of things like type inference, and of something
> like "permissive typing" (compiler/interpreter to programmer: "you don't have
> to tell me the type of a variable or argument, but if you do, I'll use that
> information to help you wherever I can").

I'd be interesting to know to what level of type infrencing
can be done with Python.  In particular, I wonder what the 
ambiguous cases are...

ML also does dynamic type infrencing, dynamically. When you 
type a  variable or define a function on the interactive
interpreter, ML returns with the type of the object.  I'm sure
something like this could be done in a Python interpreter,
perhaps as an option.   Of course, in ML, if a type is
ambiguous, it must be declared -- unlike python.

Clark





From aleaxit at yahoo.com  Tue Mar 13 04:05:22 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Tue, 13 Mar 2001 10:05:22 +0100
Subject: Einstein's Riddle
References: <oNPq6.396644$ge4.138479566@news2.rdc2.tx.home.com> <98glrd$1si1a$1@ID-75892.news.dfncis.de> <3AABE712.499CCF4C@home.com> <98gp5b$1u0oi$1@ID-75892.news.dfncis.de> <98i5880pl2@news2.newsguy.com> <98kfrl$pug$1@sshuraaa-i-1.production.compuserve.com>
Message-ID: <98kntc0o5l@news2.newsguy.com>

"Mark Yudkin" <myudkin at compuserve.com> wrote in message
news:98kfrl$pug$1 at sshuraaa-i-1.production.compuserve.com...
> Rather than looking at America sites for information about emu's, consider
> that it's an AUSTRALIAN flightless bird, and there's more to it than
eating
> it.

It's an animal of Australian _origin_, just as, say, cows are of old-world
origin; that doesn't imply it's wrong to look for emu information on
American
sites, any more than the species-origin of cows impedes Australian or
American
sites offering interesting bovine information.

And, yes, of course there's more than gastronomy about _any_ animal or
plant.  However, _most_ people's main interests in cows, tomatos, lettuce,
or emus, are quite likely to include culinary aspects (no matter how
fascinating it IS to study a tomato flower's sepals, petals, stamens,
carpels, anther, stigma, &c, for example, it is a biological as well
as a cultural fact that most people would be more interested in
ascertaining the amount of, say, sodium, vitamins, energy, minerals,
dietary fiber, etc, in a 100-gram serving of tomato juice...).


Alex





From me at nospam.net  Sat Mar 31 20:09:54 2001
From: me at nospam.net (Scottie)
Date: Sat, 31 Mar 2001 17:09:54 -0800
Subject: PEP 240 scares me
References: <15040.53203.949420.930912@beluga.mojam.com>, <Pine.LNX.4.21.0103222338530.29360-100000@kev.ioc.ee><99e2h61aao@news2.newsguy.com> <mailman.985731647.721.python-list@python.org>
Message-ID: <tccvvand35t509@corp.supernews.com>

def Farey(v, lim):
  '''Named after James Farey, an English surveyor.
  No error checking on args -- v >= 0, lim = max denominator,
  results are (numerator, denominator) in lowest terms.
  (1,0) is "infinity".
  '''
  lower, upper = (0L,1L), (1L,0L)
  while 1:
    mediant = (lower[0] + upper[0]), (lower[1]+upper[1])
    if v * mediant[1] > mediant[0]:
        if lim < mediant[1]: return upper
        lower = mediant
    else:
        if lim < mediant[1]: return lower
        upper = mediant

"Moshe Zadka" <moshez at zadka.site.co.il> wrote in message
news:mailman.985731647.721.python-list at python.org...
> On Tue, 27 Mar 2001 11:37:23 -0600 (CST), Skip Montanaro <skip at pobox.com>
wrote:
>
> > Okay, dumb question time.  Suppose the compiler sees the string "1.2" in
the
> > input stream.  How does it translate that into a rational number (6/5 or
> > 12/10 or 24/20 or ...)?  Does it simply move the decimal point over
> > sufficiently and create the corresponding rational number that contains
the
> > appropriate power-of-ten denominator?
>
> That's what Rational.py (in CVS nondist/sandbox/rational/) does.
>
> > What about rationals that can't be
> > represented correctly using decimal notation (e.g. how does one get from
> > 0.3333 to 1/3, or is it simply approximated by 3333/10000?)
>
> The later. If you want 1/3, you need to ask for it:
>
> third = rational(1)/3
>
> I guess I could be bribed to have
>
> rational("0.(3)") == rational(1)/3
>
> I just need to recall the formula for the sum of a geometric series and
> apply it. There is no rational which cannot be written as
>
> \d*.\d*(\d*)
>
> Where () means "repeat ad infinitum".
> --
> "I'll be ex-DPL soon anyway so I'm        |LUKE: Is Perl better than
Python?
> looking for someplace else to grab power."|YODA: No...no... no. Quicker,
>    -- Wichert Akkerman (on debian-private)|      easier, more seductive.
> For public key, finger moshez at debian.org
|http://www.{python,debian,gnu}.org
>




From admin at mulga.com.au  Sat Mar 17 20:41:00 2001
From: admin at mulga.com.au (Michael Hall)
Date: Sun, 18 Mar 2001 11:11:00 +0930 (CST)
Subject: reading stdinput
Message-ID: <Pine.LNX.4.10.10103181104580.12584-100000@localhost.localdomain>

Please excuse this ultra newbie question ...

How do I read standard input into a Python script? I'm trying to do the
same as 'read' in a shell script to create a variable. I've played about
with various combinations of 'sys.stdin.read()' and such but no go.

If there is a better list for newbie questions like this, please let me
know.

***********************************************
Michael Hall - mick at mulga.com.au - mulga.com.au
***********************************************





From francois.granger at free.fr  Sat Mar 17 07:47:33 2001
From: francois.granger at free.fr (=?ISO-8859-1?Q?Fran=E7ois_Granger?=)
Date: Sat, 17 Mar 2001 12:47:33 GMT
Subject: Python style questions
References: <98tb91$9t9$1@saltmine.radix.net> <98tdu1$cfk$1@panix2.panix.com> <1eqdo5u.1xxmzfi2kjk0qN%francois.granger@free.fr> <98v1vi$gre$1@news.udel.edu>
Message-ID: <1eqevwc.ydlvbi3ybxaoN%francois.granger@free.fr>

Terry Reedy <tjreedy at udel.edu> wrote:

> I posted an example about a month ago ... and about a year ago.
> Don't know what archive search facilities we currently have.

Sorry, found your post of Jan 24 throught groups.google.com.

Thanks

-- 
un jardin peut-il ?tre non naturel ?
ce qui n'est pas naturel peut-il ?tre un jardin ?
qu'est-ce qu'une mauvaise herbe ? part une herbe naturelle ?
un jardin sans herbe est-il un jardin ? - Richard


From gabriel_ambuehl-py at buz.ch  Fri Mar 30 13:55:23 2001
From: gabriel_ambuehl-py at buz.ch (Gabriel Ambuehl)
Date: Fri, 30 Mar 2001 20:55:23 +0200
Subject: Working with soaplib...
Message-ID: <86550378262.20010330205523@buz.ch>

-----BEGIN PGP SIGNED MESSAGE-----

Hello,
I've been trying to work with either soaplib or XMLRPC (I don't care
too much about which one gets used as long as my RPC works. As SOAP
looks to be a much broader standard, it would probably the one to
prefer, but this is largely theoretical as this is an inhouse project
and I plan to write a small wrapper function which should take care of
the underlying transport anyway)and while writing clients is simple, I
run into serious problems when it comes to writing servers (soapserver
got modified to not use any try/except constructions so it will dump
everything to
the screen which is to be preferred as error handling seems to be
broken, i.e. the client will just get some list containing other,
empty lists):

- ----------
import SocketServer
import soapserver

class TestRequestHandler(soapserver.RequestHandler):
    #Override method:
    def call(self, method, params):
        print "Dispatching: ", method, params
        try:
            server_method = getattr(self, method)
        except:
            raise AttributeError, "Server does not contain SOAP
procedure %s" % method
        return server_method(method, params)

    def equal_five(self, method, params):
        print "inside equal_five()"
        return 5

    def test(self, method, params):
        return params

    def add(self, method, params):
        c=params[0]+params[1]
        print "c: " + c
        return c

if __name__ == '__main__':
    server = SocketServer.TCPServer(('', 8000), TestRequestHandler)
    server.serve_forever()

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

And the client
>>> import soaplib
>>> server = soaplib.ServerProxy("http://local:8000")
>>> print server.test('test')

this works without any problems and yields
test
on the screen (as one would expect)
Now
>>> print server.equal_five()

won't work, the server dumps the following error:

Dispatching:  equal_five ()
inside equal_five()
- ----------------------------------------
Exception happened during processing of request from ('10.2.2.100',
3574)
Traceback (most recent call last):
  File "/usr/local/lib/python2.0/SocketServer.py", line 221, in
handle_request
    self.process_request(request, client_address)
  File "/usr/local/lib/python2.0/SocketServer.py", line 247, in
process_request
    self.finish_request(request, client_address)
  File "/usr/local/lib/python2.0/SocketServer.py", line 251, in
finish_request
    self.RequestHandlerClass(request, client_address, self)
  File "/usr/local/lib/python2.0/SocketServer.py", line 385, in
__init__
    self.handle()
  File "/usr/local/lib/python2.0/BaseHTTPServer.py", line 266, in
handle
    method()
  File "\\local\admin_net\soaplib-0.8-20000621\soapserver.py", line
37, in do_POST
  File "\\local\admin_net\soaplib-0.8-20000621\soaplib.py", line 654,
in dumps
AssertionError: argument must be tuple, method call/response, or fault
instance
- ----------------------------------------

(the script lives on a FreeBSD machine and the client gets loaded over
samba from it, that's why the strange UNCs are there)

I think there's something wrong with my return statements of the
not working functions but can't seem to figure out what to do about
it. What data type do I need to return so soaplib.dumps
(xmlrpclib.dumps appears to be having exactly the same problem, for
that matter) can handle it? A tuple like
return (value,)
doesn't appear to be working, neither does
return ([value,])

Any help would be greatly appreciated.




Thanks in advance and best regards,
 Gabriel

-----BEGIN PGP SIGNATURE-----
Version: PGP 6.0.2i

iQEVAwUBOsTIj8Za2WpymlDxAQGO+Af+Ms+cs0fY6Nu0njzyqUttJvs/Cb7o9QXi
6VL/7h3dxJ1eBxY+NNDgZeGlf1V0YgkjTjSYknksxARcW8wdILLK5WJPlbqBsZWb
TiUCWbNzCXVVLVF5DzEUmtlL5QMSR4cceRbufD/Kpfi/87xgcPTRzVoMC2vGMIKh
PXLn9f0L4MjvUWYcdXOWrZDRAQVfcT3ivaK4bQLeiRkI+lTWTcHSdjdKAKyHMvKh
owpUbzzwL7p4l1wYxjpgfZStNEulMzGIpw6FebdGRj8lwyCO1wiOyV+bBm40o6b1
RZ6ihz2Pvgl9Q+Q2Z40woOO/dTq9r1tFbQPT16WwL77xlXj+7PRcvg==
=NX/H
-----END PGP SIGNATURE-----





From ocaner at gmx.de  Sun Mar 11 06:55:45 2001
From: ocaner at gmx.de (Özgür Caner)
Date: Sun, 11 Mar 2001 12:55:45 +0100
Subject: RPM Module?
Message-ID: <3aab687b$1@news.teuto.net>

Does anybody know where to find a Documentation about the RPM Module for Python?




From phrxy at csv.warwick.ac.uk  Thu Mar 29 12:44:19 2001
From: phrxy at csv.warwick.ac.uk (John J. Lee)
Date: Thu, 29 Mar 2001 18:44:19 +0100
Subject: Defining class files
In-Reply-To: <3ac33343$0$12248$ed9e5944@reading.news.pipex.net>
References: <3ac33343$0$12248$ed9e5944@reading.news.pipex.net>
Message-ID: <Pine.SOL.4.30.0103291839430.13267-100000@primrose.csv.warwick.ac.uk>

On Wed, 28 Mar 2001, Neil Benn wrote:
[...]
> >>> YourClass
> <module YourClass at 2584541>
>
>     The text file is saved as YourClass.py - is this the problem, should
> class files have different terminaters in their filename??
[...]

The module name comes from the name of the file it is saved in.
YourClass.py gives you a module called YourClass.  This is fine -- using
the same name for both class and module is common practice (although not
when the module name includes the word 'class'!) -- but you'll have to
use:

import YourClass

foo = YourClass.YourClass()


or this:

from YourClass import YourClass

foo = YourClass()


to get at the class inside the module.


John



From Hawley_Randall_R at Lilly.com  Tue Mar 27 12:12:51 2001
From: Hawley_Randall_R at Lilly.com (Randy Hawley)
Date: Tue, 27 Mar 2001 17:12:51 GMT
Subject: newbie IDLE, Tkinter
References: <FHUu6.68719$A6.14534907@news4.rdc1.on.home.com>
Message-ID: <01c0b6e2$3d426740$4ecb0c28@PC0AB085S5A8>

Howdy Python Gurus.

I too am a newbie trying out Python on Win32 platforms.

I have the same problems as Nick describes below.  Also, somethimes when 
I hit a button that has "command=root.quit" defined as its action, not only
does my program's window quit, but so does IDLE.

What gives with this?  Anybody else ever run into this?  Is it a result of
poor coding in my script (saw it when I just ran the examples from the
Tkinter tutorial, too) ?

Thanks,
	Randy Hawley

Nick Perkins <nperkins7 at home.com> wrote in article
<FHUu6.68719$A6.14534907 at news4.rdc1.on.home.com>...
> I have been playing around with Python and Tkinter, using the IDLE prompt
> and edit windows.
> 
> I am a little confused about what happens when I run my tkinter programs,
> then close them (by closing their window).  Sometimes I come back to the
> prompt, sometimes the interactive window just seems to hang, and
sometimes
> the whole IDLE program just quits.
> 
> Can someone explain the best way to interactively edit and test a tkinter
> program with IDLE?  Do these window-closing behaviours depend on how I
> actually run my program?
> 
> 
> 
> 


From guido at digicool.com  Sat Mar  3 14:44:19 2001
From: guido at digicool.com (Guido van Rossum)
Date: Sat, 03 Mar 2001 14:44:19 -0500
Subject: Bug fix releases
In-Reply-To: Your message of "Sat, 03 Mar 2001 14:30:13 EST."
             <15009.17989.88203.844343@cj42289-a.reston1.va.home.com> 
References: <LNBBLJKPBEHFEDALKOLCMEEPJBAA.tim.one@home.com> <mailman.982897324.9109.python-list@python.org> <200103031821.NAA24060@panix3.panix.com> <200103031910.OAA21663@cj20424-a.reston1.va.home.com>  
            <15009.17989.88203.844343@cj42289-a.reston1.va.home.com> 
Message-ID: <200103031944.OAA21835@cj20424-a.reston1.va.home.com>

> Guido van Rossum writes:
>  > I wonder, does that extend to new library modules?  Is there also
>  > resistance against the growth there?  I don't think so -- if anything,
>  > people are clamoring for more stuff to become standard (while at the
> 
>   There is still the issue of name clashes; introducing a new module
> in the top-level namespace introduces a potential conflict with
> someone's application-specific modules.  This is a good reason for us
> to get the standard library packagized sooner rather than later
> (although this would have to be part of a "feature" release;).

But of course the library repackaging in itself would cause enormous
outcries, because in a very real sense it *does* break code.

>  > Wait a minute!  Now you're making it too complicated.  Betas of bugfix
>  > releases?  That seems to defeat the purpose.  What kind of
> 
>   Betas of the bugfix releases are important -- portability testing is
> fairly difficult to do when all we have are Windows and Linux/x86
> boxes.  There's definately a need for at least one beta.  We probably
> don't need to lengthy, multi-phase alpha/alpha/beta/beta/candidate
> cycle we're using for feature releases now.

OK, you can have *one* beta.  That's it.

>  > It's not too late, as I mentioned.  We'll also do this for 2.1.
> 
>   Managing the bugfix releases would also be an excellent task for
> someone who's expecting to use the bugfix releases more than the
> feature releases -- the mentality has to be right for the task.  I
> know I'm much more of a "features" person, and would have a hard time
> not crossing the line if it were up to me what went into a bugfix
> release.

That's how all of us here at PythonLabs are feeling...  I feel a
community task coming.  I'll bless a 2.0.1 release and the general
idea of bugfix releases, but doing the grunt work won't be a
PythonLabs task.  Someone else inside or outside Python-dev will have
to do some work.  Aahz?

>  > BTW, See you at the conference!
> 
>   If we don't get snowed in!

Good point.  East coasters flying to LA on Monday, watch your weather
forecast!

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



From erno-news at erno.iki.fi  Mon Mar  5 19:15:16 2001
From: erno-news at erno.iki.fi (Erno Kuusela)
Date: 06 Mar 2001 02:15:16 +0200
Subject: I come to praise .join, not to bury it...
References: <mailman.983485923.1136.python-list@python.org> <97p7iq$kk2$1@nntp6.u.washington.edu> <97rka002k9@news1.newsguy.com>
Message-ID: <kuu257kb1n.fsf@lasipalatsi.fi>

In article <97rka002k9 at news1.newsguy.com>, "Alex Martelli"
<aleaxit at yahoo.com> writes:

| I have not heard ANY _technical_ arguments opposed to
| this last time the discussion went around -- nothing but
| vague aesthetic, "it should be the other way", "I find it
| ugly" kind of complaints.

what's wrong with vague aesthetic?-) if lots of people share it,
it's not to be dismissed so lightly.

  -- erno


From francois.granger at free.fr  Fri Mar 30 18:18:56 2001
From: francois.granger at free.fr (=?ISO-8859-1?Q?Fran=E7ois_Granger?=)
Date: Fri, 30 Mar 2001 23:18:56 GMT
Subject: Creation d'un NG francophone sur Python
References: <1eqzyk6.gxd4jr1tkjykzN%francois.granger@free.fr> <9a1e6f011es@enews3.newsguy.com> <bS1x6.9225$SB2.838111@ruti.visi.com>
Message-ID: <1er40bo.b8vjsq6avf3gN%francois.granger@free.fr>

Grant Edwards <grante at visi.com> wrote:

> I certainly would never vote against a French Python newsgroup.
> But, I want to let you know that French language postings are
> quite welcome in this group -- as are threads in any other
> language.
> 
> You've even got a fair chance of getting replies to questions
> posted in Esperanto.

Thanks for this message and other nice messages.
And sorry to have started a political thread about linguistic, this was
not my aim.

[fr]
Merci pour ton messages et les autres messages accueillants.
D?sol? d'avoir d?marr? une enfilade politique ? propos de pr?f?rences
linguistiques, ce n'?tait pas le but.

-- 
"On jouit moins de ce qu'on obtient que de ce qu'on esp?re."
- Jean-Jacques Rousseau
- Cit? par Anne-Marie


From jmarshal at mathworks.com  Mon Mar  5 12:20:01 2001
From: jmarshal at mathworks.com (Joshua Marshall)
Date: 5 Mar 2001 17:20:01 GMT
Subject: Q: Operator precedence
References: <Pine.LNX.4.21.0103051051180.12138-100000@kev.ioc.ee> <3AA3BCB1.13CC1479@javanet.com>
Message-ID: <980hs1$c8t$1@news.mathworks.com>

Raymond Hettinger <othello at javanet.com> wrote:
> Pearu Peterson wrote:

>> Section 5.12 in Python reference manual summarizes operator precedence.
>>
>> I would like to have the following function:
>>
>> def getprec(s):
>>      return <the precedence number of a keyword/operator `s'>

> Here is an approach to rank ordering of precedence at run-itme:

> def higherOrEqual( op1, op2 ):
>     return eval('51 %s 83 %s 97 == ( 51 %s 83 ) %s 97' % (op1,op2,op1,op2))

> order = ['Lower', 'HigherOrEqual']

> binops = ['+', '-', '&', '^', '/']

> for x in binops:
>     for y in binops:
>         print x, order[higherOrEqual(x,y)], y

This will fail for '**', since it's right-associative.
(Ignoring integer overflow and computation time.)


From m.wilson at bibliocraft.com  Tue Mar  6 21:02:34 2001
From: m.wilson at bibliocraft.com (Mark Wilson)
Date: Wed, 7 Mar 2001 02:02:34 -0000
Subject: Windows98+Idle0.6+ 32 bit color HW or SW problem?
Message-ID: <010901c0a6aa$cbe04e60$7e824ac3@mwdell1>

Gerrit:

This may be a bit lower level than you want, but...

>When switching
>between 16 bit and 32 bit color mode the Tkinter windows are apparantly not
>updated, the contents of these windows becomes garbage.
>
>Microsoft applications, such as IE and outlook express do not suffer from
>these effects. Is there any known SW problem in Tkinter or related SW, which
>may cause these problems. The alternative is suspect HW (AMD Thunderbird 900
>MHz, ELSA Gladiac MX).

Windows sends a message with the id WM_DISPLAYCHANGE whenever the user applies
changes to screen resolution/depth. (In fact, with typical MS "panache", they
send the message twice - same params each time - and don't talk to me about
WM_ACTIVATE!). I have no experience with either Tkinter or Idle, but you could
check the Win sources and see if this message is at least handled by the Windows
message handling code. That might indicate a possible solution. I don't actually
know where this message is derived from, but I'd be surprised if it was the
graphics card driver.

>I now get horizontal disturbances in the idle and other
>TKinter windows, when using 32 bit color mode and 1280*1024.

Sounds like a different problem - perhaps that is the HW/driver. Perhaps you
could elaborate.

Have fun

wishing-he-was-at-ipc9-with-the-guys-who-could-give-you-better-answers-ly y'rs -
Mark







From chrisw at nipltd.com  Thu Mar  8 09:25:08 2001
From: chrisw at nipltd.com (Chris Withers)
Date: Thu, 08 Mar 2001 14:25:08 +0000
Subject: virtual hosting in Zope
References: <987lbo$119t$1@infocom.km.ua>
Message-ID: <3AA79644.58A556A8@nipltd.com>

"Anton A. Zotov" wrote:
> 
> Urgent: I'd appreciate a person who would tell me how to implement virtual
> hosting under Zope/Medusa environment

zope at zope.org is the mailing list to ask on.

SiteAccess is the key, it can be downloaded from www.zope.org and comes included
with Zope versions 2.3.0 and above.

cheers,

Chris



From bob_cannard at mentor.com  Fri Mar 23 17:09:14 2001
From: bob_cannard at mentor.com (Bob Cannard)
Date: Fri, 23 Mar 2001 14:09:14 -0800
Subject: re.match
References: <mailman.985198768.14569.python-list@python.org>
Message-ID: <3ABBC98A.EE24DE24@mentor.com>


> Dave Brueck wrote:
> Hi Bob,
> 
> '$' matches 'end of string', so you _can_ use re.match to test "Is this string a Thing".

Heh. I already knew that, Dave; apparently my original message
wasn't clear enough. My point was that there's no way to do it
without constructing a second regular expression, which is
inefficient and clumsy. What is wanted is something like:

def fullmatch(r, s):
    return re.match(r + '$', s)

but which, by analogy with re.match, does not have the overhead
of constructing and compiling a new regular expression; which
works equally well whether r is a string or a compiled re; which
works regardless of whether or not r already has a $ at the end.

I'm just surprised that Python has a separate function for matching
at the start of a string but seems to lack one for matching an
entire string. The issue is efficient re-use of a compiled regular
expression in different situations, not "how to do this".

Cheers,

       Bob.


From MarkH at ActiveState.com  Sat Mar 10 21:35:37 2001
From: MarkH at ActiveState.com (Mark Hammond)
Date: Sun, 11 Mar 2001 02:35:37 GMT
Subject: win32com:MakePy -- Early Binding without the size jitters?
References: <mailman.984043211.10915.python-list@python.org>
Message-ID: <3AAAE3BB.5090506@ActiveState.com>

Clark C. Evans wrote:

> Hello.  I'd like to use early-binding with about three COM 
> libraries, each one of the stub.pyc files over 100KB (with 
> one in particular over 300KB).  However, I'm only using
> about a dozen methods (including constructors, etc) from
> each library.  I was wondering... is it safe to delete
> methods/interfaces which I'm not using?  I've deleted
> a few, and it still works for my purposes, so I was 
> wondering if I'm shooting myself in the foot -- or if
> there is a methodological approach to this pruning?
> It'd be cool to get the .pyc alot smaller; no use
> toossing away 1MB of memory when I only need a few K...
> 

If you are using the gencache functions to create makepy files, try 
passing "bForDemand=1" to the functions.  This will only generate what 
is actually used by the program, as it is used.

Mark.




From fredp at mygale.org.nospam  Wed Mar 21 10:57:19 2001
From: fredp at mygale.org.nospam (Fred Pacquier)
Date: 21 Mar 2001 15:57:19 GMT
Subject: Accessing Oracle from Python remotely ?
References: <Xns906AB9C901807PaCmAnRDLM@161.48.76.20> <eE3u6.776$9%.203510@newsr1.maine.rr.com>
Message-ID: <Xns906BACA5C71FEPaCmAnRDLM@161.48.76.20>

"gkiffney" <gkiffney at my-deja.com> said :

>There's http://www.zope.org/Products/DCOracle which should get you
>started. There was an article by Uche Ogbuji in Linuxworld that helped
>me get going with this, but it's apparently fallen off the Internet map
>into the bit bucket.  Using the dcoracle module (which is a bit of a
>pain to get going, look at the Zope docs carefully, and if you can find
>the prebuilt dll's, use those), you can do things like this:

Thanks for confirming this. DCOracle definitely sounds like the right 
candidate.

Some will probably smile at this, but my first problem actually will be to 
get the Oracle client onto the Unix box. We have servers on NT and Unix, 
but unfortunately the "client = Windows" law applies here, too... :-)

-- 
YAFAP : http://www.multimania.com/fredp/


From greg at cosc.canterbury.ac.nz  Thu Mar  1 19:48:24 2001
From: greg at cosc.canterbury.ac.nz (Greg Ewing)
Date: Fri, 02 Mar 2001 13:48:24 +1300
Subject: Collection interfaces
References: <3a90fa97.0@news1.mweb.co.za> <3A933830.7F98D118@ix.netcom.com> <CgJk6.515$v77.66568@newsread2.prod.itd.earthlink.net> <220220010813598042%cmh@bDistributed.com> <ud7cakvi8.fsf@roundpoint.com> <MPG.150312ceb68f1ada98969f@news.earthlink.net> <8Gem6.473353$U46.14105062@news1.sttls1.wa.home.com> <MPG.1503548620e632919896a4@news.earthlink.net> <MCjm6.474652$U46.14163288@news1.sttls1.wa.home.com> <MPG.150394195f7f04a19896a7@news.earthlink.net> <3A9A86B1.9A8C69A2@ix.netcom.com> <MPG.1504d14563dce6499896af@news.earthlink.net> <97fjla128ir@news2.newsguy.com> <MPG.1505a76831c34b749896b3@news.earthlink.net> <97h5s40vhc@news2.newsguy.com> <MPG.1506007591cf818b9896bc@news.earthlink.net> <97ifd60sen@news1.newsguy.com>
Message-ID: <3A9EEDD8.9102C3A@cosc.canterbury.ac.nz>

Alex Martelli wrote:
> 
> the indisputable success
> of Smalltalk does show that single-inheritance is *practically
> workable* for many tasks

Smalltalk is like Python in that its inheritance hierarchy
is more about implementation inheritance than subtype
relationships. So I don't think you can conclude much
from Smalltalk about whether single inheritance is adequate
for classification purposes.

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


From akuchlin at mems-exchange.org  Fri Mar 23 14:29:04 2001
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 23 Mar 2001 14:29:04 -0500
Subject: pyncurses documentation??
References: <3abb8a08$0$16654@SSP1NO25.highway.telekom.at>
Message-ID: <3dn1acz3lr.fsf@ute.cnri.reston.va.us>

Alex Berger <aleb at gmx.net> writes:
> Does anybody know where I can get an introduction/documentation to 
> pyncurses??

The LibRef documentation is at
http://www.python.org/doc/lib/module-curses.html, and the HOWTO is at
http://py-howto.sourceforge.net/curses/ .

--amk


From woetzel at gmd.de  Mon Mar 26 05:19:26 2001
From: woetzel at gmd.de (Gerd Woetzel)
Date: 26 Mar 2001 10:19:26 GMT
Subject: Is setup.py a "good thing"? (long)
References: <3aba76e2.167083508@24.0.228.33> <3dvgp1zasw.fsf@ute.cnri.reston.va.us> <99fsdu$p2u$1@hop.gmd.de> <3dpuf8zb4t.fsf@ute.cnri.reston.va.us> <j4zoebmmpu.fsf@informatik.hu-berlin.de>
Message-ID: <99n53e$5s3$1@hop.gmd.de>

>Andrew Kuchling <akuchlin at mems-exchange.org> writes:

>> woetzel at gmd.de (Gerd Woetzel) writes:
>> > Modules/Setup has buit zlibmodule.so and socketmodule.so which worked
>> > fine during the test phase, but setup.py has (secretly?) created
>> > zlib.so (socket.so) too! Finally "make install" has installed all of them.
>>
>> That certainly shouldn't have happened.  Can you reproduce the problem?

Martin von Loewis <loewis at informatik.hu-berlin.de> writes:

>It could happen if you use *shared* in Modules/Setup and if setup.py
>is older than 1.30; before, it would use on builtin_module_names to
>exclude modules. So that was a problem once and should be fixed since
>2.1b1.

I have tried 2.1b1 and the newest 2.1b2a this morning.
It still installs both both _socket.so (wrong!) and _socketmodule.so (the
working module). Same with zlib.so and zlibmodule.so.

Note: After removing the installation directory /opt/python/lib/python2.1/
      I have done the Installation from scratch, starting with the extraction
      of Python-2.1b2a.tar.gz.

Here is my Setup.local which I have copied to Python-2.1b2a/Modules after
extraction of the sources (yes, I used  *shared*) :

   readline readline.c -I/opt/gnu/include -L/opt/gnu/lib -lreadline -ltermcap

   *shared*

   WZLIB=/home/woetzel/lib
   WZINCL=/home/woetzel/include
   _socket socketmodule.c \
	-DUSE_SSL -I$(WZINCL)/ssl -I$(WZINCL)/ssl/openssl \
	-L$(WZLIB)/ssl -R$(WZLIB)/ssl -lssl -lcrypto

   zlib zlibmodule.c -I$(WZINCL) -L$(WZLIB) -R$(WZLIB) -lz

Here is the output of "grep _socket make.out":

    Python-2.1b2a/Lib/test/output/test_socket
    Python-2.1b2a/Lib/test/test_socket.py
    Python-2.1b2a/PCbuild/_socket.dsp
    gcc -shared  Modules/socketmodule.o  -L/home/woetzel/lib/ssl -R/home/woetzel/lib/ssl -lssl -lcrypto  -o Modules/_socketmodule.so
    building '_socket' extension
    gcc -shared build/temp.solaris-2.8-sun4u-2.1/socketmodule.o -L/usr/local/ssl/lib -L/usr/local/lib -lssl -lcrypto -o build/lib.solaris-2.8-sun4u-2.1/_socket.so
    skipping '_socket' extension (up-to-date)
      from _socket import *
    test_socket
      from _socket import *
    test_socket
    /opt/gnu/gnubin/install -c -m 644 ./Lib/test/test_socket.py /opt/python/lib/python2.1/test
    /opt/gnu/gnubin/install -c -m 644 ./Lib/test/output/test_socket /opt/python/lib/python2.1/test/output
    Compiling /opt/python/lib/python2.1/test/test_socket.py ...
    Compiling /opt/python/lib/python2.1/test/test_socket.py ...
    skipping '_socket' extension (up-to-date)
    skipping '_socket' extension (up-to-date)
    copying build/lib.solaris-2.8-sun4u-2.1/_socket.so -> /opt/python/lib/python2.1/lib-dynload
    /opt/gnu/gnubin/install -c -m 555 Modules/_socketmodule.so /opt/python/lib/python2.1/lib-dynload/_socketmodule.so


ls -l /opt/python/lib/python2.1/lib-dynload/_socket*:

    -rwxrwxr-x  1 woetzel  woetzel     60704 Mar 26 11:24 /opt/python/lib/python2.1/lib-dynload/_socket.so
    -r-xr-xr-x  1 woetzel  woetzel     60668 Mar 26 11:35 /opt/python/lib/python2.1/lib-dynload/_socketmodule.so

Not a big Problem. "rm lib-dynload/_socket.so" will fix it.
Just for information ... 

Regards,
Gerd
--
Gerd Woetzel            | email: gerd.woetzel at gmd.de
GMD FIT.CSCW            | phone: ++49 2241 142648 (fax: 142084)
D-53754 Sankt Augustin  | www:   http://orgwis.gmd.de/~woetzel/


From aleb at gmx.net  Sat Mar 24 09:40:58 2001
From: aleb at gmx.net (Alex Berger)
Date: Sat, 24 Mar 2001 15:40:58 +0100
Subject: reading single bits?
Message-ID: <3abcb2da$0$20400@SSP1NO25.highway.telekom.at>

How can I read/write single bits of some binary data when I know the 
position??

alex.


From ws-news at gmx.at  Thu Mar  1 18:43:28 2001
From: ws-news at gmx.at (Werner Schiendl)
Date: Fri, 2 Mar 2001 00:43:28 +0100
Subject: catching a del
References: <j%zn6.322252$f36.11513719@news20.bellglobal.com>
Message-ID: <983490479.833644@newsmaster-04.atnet.at>

not sure how to implement such thing in python, but you could use an object
that has the reference to your object and allows the client software to
access the object's functionality
so you have references to the object in that 'reference' object and in the
dictionary
but only the client software has references to the 'reference' object (hate
that term, maybe someone more trained in english language can provide me
with a better one...)
so when all references to that new object are gone it will then be destroyed
and can remove the object from the dict

hth
werner

Dan Parisien <dan at eevolved.com> wrote in message
news:j%zn6.322252$f36.11513719 at news20.bellglobal.com...
> class obj:
>         def __init__(self, name, parent):
>                 parent.children[name] = self
>                 self.parent = parent
>                 self.name = name
>
>         def onDelete(self): #?!
>                 del self.parent[self.name]
>
> I want to be able to catch a del of an instance of the above class so I
can
> remove it from its parent dictionary. Is there a way? (note it cannot be
in
> a destructor because there will always be a reference inside the parent
> dict)
>
> Thanks,
> Dan




From pieter at equinox.co.za  Fri Mar 16 06:52:01 2001
From: pieter at equinox.co.za (Pieter Nagel)
Date: Fri, 16 Mar 2001 13:52:01 +0200
Subject: gah! I hate the new string syntax
In-Reply-To: <slrn9avqlm.8ee.quinn@retch.ugcs.caltech.edu>
References: <mailman.983485923.1136.python-list@python.org> <97sudi01i26@news1.newsguy.com> <98elrk$li8$1@nntp.Stanford.EDU> <98fc070thh@news1.newsguy.com> <98iqo2$56o$1@nntp.Stanford.EDU> <98iv3s01sc2@news2.newsguy.com> <slrn9avqlm.8ee.quinn@retch.ugcs.caltech.edu>
Message-ID: <Pine.LNX.4.21.0103161350510.2809-100000@bast.jhb.equinox.co.za>

On Mon, 12 Mar 2001 17:56:33 +0100, Alex Martelli <aleaxit at yahoo.com> wrote:

> Focusing on the polymorphism needs, I see the issue in the mirror
> way from you: it's a _weakness_ (of Python and most other languages)
> that I can't add methods 'post-facto, from the outside' to existing
> objects -- it impedes the most natural, elegant, typeswitch-free way
> to add functionality.

Not true. See my other message in this post for how to extend Python
classes with new methods at runtime.

-- 
     ,_
     /_)              /| /
    /   i e t e r    / |/ a g e l



From bill-bell at bill-bell.hamilton.on.ca  Tue Mar 20 11:49:36 2001
From: bill-bell at bill-bell.hamilton.on.ca (Bill Bell)
Date: Tue, 20 Mar 2001 11:49:36 -0500
Subject: Little red lines in PythonWin!
Message-ID: <3AB743D0.30906.DCA4AF@localhost>

OK, I withdraw my earlier stupid question. Let me ask another one!

What, in general, do you guys do about distributing sourcecodes in 
a language like Python that depends on indenting and where tabs 
can cause chaos. (And, no, you can't possibly shoot all of those of 
us who get caught this way, so that's not a solution.)

Please point me towards the accepted approach. Thanks.



From topmind at technologist.com  Fri Mar 23 00:26:08 2001
From: topmind at technologist.com (topmind)
Date: Fri, 23 Mar 2001 05:26:08 GMT
Subject: Collection interfaces
References: <3a90fa97.0@news1.mweb.co.za>   <3A9A86B1.9A8C69A2@ix.netcom.com> <MPG.1504d14563dce6499896af@news.earthlink.net> <MPG.1505a76831c34b749896b3@news.earthlink.net> <MPG.1506007591cf818b9896bc@news.earthlink.net> <slrn99opop.gr.amk@mira.erols.com> <MPG.1506265ebf6d934d9896be@news.earthlink.net> <97ibmc0k47@news1.newsguy.com> <MPG.1506cf3293b094b49896c1@news.earthlink.net> <97m0c7080c@news2.newsguy.com> s.earthlink.net> <zb%n6.31033$o%6.586026@news1.rdc1.bc.home.com> <MPG.150b383cea226d409896d0@news.earthlink.net> <kNko6.6986$eq3.707121@news.uswest.net> <MPG.150c46db9cec283e9896d4@news.earthlink.net> <3AA2A898.450B659E@mail.com> <gRCt6.12224$Im6.1308560@newsread1.prod.itd.earthlink.net> <3AB755F8.67E84126@mail.com> <%a%t6.17245$227.1601569@newsread2.prod.itd.earthlink.net> <3AB8A730.4BE13D22@mail.com> <B6DE17D2.D6B0%t*b*r*u*m*l*e*y@f*u*s*e.n*e*t> <anhu6.18497$Im6.1938203@newsread1.prod.itd.earthlink.net> <B6DF5D0D.D7AE%t*b*r*u*m*l*e*y@f*u*s*e.n*e*t>
Message-ID: <Q7Bu6.2721$P4.207644@newsread1.prod.itd.earthlink.net>

"Troy Brumley" <t*b*r*u*m*l*e*y at f*u*s*e.n*e*t> wrote in message
news:B6DF5D0D.D7AE%t*b*r*u*m*l*e*y at f*u*s*e.n*e*t...
> in article anhu6.18497$Im6.1938203 at newsread1.prod.itd.earthlink.net,
topmind
> at topmind at technologist.com wrote on 3/22/01 1:57 AM:
>
> >
> > "Troy Brumley" <t*b*r*u*m*l*e*y at f*u*s*e.n*e*t> wrote in message
> > news:B6DE17D2.D6B0%t*b*r*u*m*l*e*y at f*u*s*e.n*e*t...
> >> in article 3AB8A730.4BE13D22 at mail.com, James A. Robertson at
> >> jarober at mail.com wrote on 3/21/01 8:06 AM:
> >>
> >>> topmind wrote:
> >>>>
> >>>
> >>>>> http://www.cincom.com/smalltalk
> >>>>>
> >>>>
> >>>> I only see a front page, not a parsing example.
> >>>>
> >>>
> >>> Follow the tutorial link - it's right there on that page.
> >>>
> >>>
> >>>> Besides, what are you comparing it to so say
> >>>> it is better than a procedural/relational approach?
> >>>
> >>> It's vastly easier than similar code done in C or basic, for example
> >>
> >> User testimonial here ... last week I had to rip thorugh a pile of IIS
> > logs
> >> to get some statistics for some people, and I used the original
(non-gui)
> >> tutorial code as a starting point. I was once again impressed with how
> > much
> >> easier it was to do several common data processing tasks in Smalltalk
than
> >> it was in Assembly language or C or Pascal.
> >>
> >
> > Give me specific requirements and I bet I can match it on code size.
> >
> > C, Pascal, and Assembly are hardly what I would use for code-size
> > contests.
> >
> > Of course, there is more to comparing than code size, but those often
> > get very subjective or fuzzy.
>
> I rarely worry about code size except how it relates to code clarity. Even
> back in my assembly oriented youth I tried to keep functions under one
page
> on a listing for the sake of clarity.
>

Can you quantify "clarity" or is it a personal thing (subjective)?

What makes people confortable with code seems to vary
greatly per individual.

Do you have specific examples.

> I mention C, Pascal, and Assembly because they are my usual tool set, or
> were until I discovered Smalltalk.

I used similar stuff at first, but was glad to leave them in the dust when I
discovered the concepts of Table Oriented Programming.

I never wanted to see another fricken array again (accept as
hashes/dictionaries to pass single records or parameter
lists or attributes. That is their best use, and not data collection
holding.)

> I've done a lot of business programming
> over the years in procedural environments and used more languages than
those
> (even COBOL and RPG) but for the heavy lifting I would always stick with
> them.
>
> Again, until I discovered Smalltalk :)
>
> I've seen Perl and run away screaming.

Perl is a strong divisor. You either swear by it, or swear at it. There
are few in-betweeners.

> Python and Eiffel are on the to learn
> list, but Smalltalk seems to have the power I need for the tasks I want to
> do these days. Objects are becoming more and more natural as I use them,
and
> my code improves contantly.
>
> >
> >> I'm a long time procedural type, learning to really think in and use
> >> Smalltalk has been an interesting experience.
> >>
> >> I've also done a lot of database "fixing" and Smalltalk has usually
beat
> > out
> >> raw SQL for anything but the most trivial tasks.
> >>
> >
> > Fixing what? SQL is not good at string parsing.
>
> Usually sweeping changes, or dangling referential integrity problems.


Huh? Referential integrity is an option, not a forced feature.

If the lamp keeps you awake at night, then turn it off!

>
> >> All that said, I think it's the environment (interactive) that improves
> > the
> >> experience as much as the language itself. If I had a reasonable
> >> interpretive BASIC environment available, I could probably do many of
the
> >> things I did in Smalltalk almost as easily ... certainly more easily
than
> >> with C.
> >>
> >
> > I find interpretive environments potentially more compact code-wise than
> > static ones.
>
> No argument here.
>
> >
> >>
> >>>
> >>>>
> >>>>>>
> >>>>>>> [snip]





From scarblac at pino.selwerd.nl  Thu Mar 22 10:47:53 2001
From: scarblac at pino.selwerd.nl (Remco Gerlich)
Date: 22 Mar 2001 15:47:53 GMT
Subject: New PEP: The directive statement
References: <mailman.985085006.24236.python-list@python.org>
Message-ID: <slrn9bk7fk.fnm.scarblac@pino.selwerd.nl>

I'm not that happy with this PEP. I'll try to explain why. 

First I have to point out an inconsistency in the PEP, then I'll explain my
broader concerns.

For problem a) I think 'from __future__' is better, for b) I think
directives are too general, allow things we want to keep away from the
language, this problem is better solved with, say, a sys.options dict.

Martin von Loewis <loewis at informatik.hu-berlin.de> wrote in comp.lang.python:

(I'll leave some bits that I want to comment on)

(A)
>     In discussion of this PEP, readers commented that there are two
>     kinds of "settable" language features:
> 
>     a) those that are designed to eventually become the only option,
>        at which time specifying use of them is not necessary anymore.
>        The features for which the syntax of the "Back to the
>        __future__" PEP [3] was proposed fall into this category.  This
>        PEP supports declaring such features, and supports phasing out
>        the "old" meaning of constructs whose semantics has changed
>        under the new feature. However, it defines no policy as to what
>        features must be phased out eventually.
> 
>     b) those which are designed to stay optional forever, e.g. if they
>        change some default setting in the interpreter. An example for
>        such settings might be the request to always emit line-number
>        instructions for a certain module; no specific flags of that
>        kind are proposed in this specification.
> 
>     Since a primary goal of this PEP is to support new language
>     constructs without immediately breaking old libraries, special
>     care was taken not to break old libraries by introducing the new
>     syntax.

(B)
> Syntax
> 
>     A directive_statement is a statement of the form
> 
>         directive_statement: 'directive' NAME [atom] [';'] NEWLINE
> 
>     The name in the directive indicates the kind of the directive; it
>     defines whether the optional atom can be present, and whether
>     there are further syntactical or semantical restrictions to the
>     atom. In addition, depending on the name of the directive, certain
>     additional syntactical or semantical restrictions may be placed on
>     the directive (e.g. placement of the directive in the module may be
>     restricted to the top of the module).

(C)
> Backwards Compatibility
> 
>     Introducing 'directive' as a new keyword might cause
>     incompatibilities with existing code. Following the guideline in
>     [1], in the initial implementation of this specification,
>     directive is a new keyword only if it was used in a valid
>     directive_statement (i.e. if it appeared as the first non-string
>     token in a module).


First the inconsistency: the syntax definition in (C) claims that some
restrictions on the placement of directive may be in place, depending on
the directive. However, (D) notes that directives must appear as the first
non-string token in a file, for backwards compatibility. So in practice,
directive can't be used anywhere in a file.


Then the two kinds of language features, from (A). The transitional features
are your a), the features that will have a short transitional period before
they become just another part of the language. This is the part that the
__future__ PEP addresses. Your problem b) is for truly optional features, I
imagine that would be things like optimization, warnings, debugging output
(like your line numbers). These things are currently addressed by command
line options. I don't think these two problems are solved best by a single
solution.

What I miss most in the 'directive' approach is the introspection. Can a
program see which directives are on? List the current directives and their
docstrings? Also, if I can turn on a directive further down a file, how can
I turn it *off*?

Transitional features are transitional for a hopefully short time. They are
provided for people who need to test their existing software against an
upcoming incompatible feature, and for early adopters who want the new
feature now so they can tinker with it. The version in which the feature
will become part of the language proper is already decided (2.2, in the case
of nested scopes).

The 'from __future__ import' idea shows that a feature is being used from a
future version. It's also a realy module with real information, so that a
running program can see what is in there, and can see information like the
version when this will become mandatory and, not least, the feature's
docstring. What is ugly about that is the magical side effect of an import
statement at the first line of a file that has compile time side effects.
But since this is for testing code against upcoming changes and for early
adopters who are usually testers anyway, I don't see that as a big problem.
Besides, using __underscores__ always implies at least a little magic, as
should time machines.

The 'directive' idea still has the magical first line of the file bit, but
it doesn't provide information to the running program about what is turned
on. It's not possible to make a single tool once that removes unneeded
'directive' statements from files, or even prints a list of available
statements. Also, 'transitional' is odd. The feature is *in a transitional
period*, but neither the file nor the feature is transitional (it's either
on or off, not in some transitional state). So I don't even agree on the
clarity point.

And then the other half of the problem, use of the directive statement for
other things. Is this vapor? I can't think of any useful applications at the
moment; for systemwide features we have command line options (not that
that's perfect, but that's not the point here), and directives can only be
used on a file-wide level. I can imagine there would be options you want to
toggle on a smaller level, say, for only part of a file. But you can't
toggle directives, you can't even see what their current setting is, inside
the program.

So I feel the application domain for the second form of directive is very
small, basically compile time options that make sense for single files. I
haven't seen a pressing demand for those.

Besides, a dictionary 'sys.options' could control a lot of options. In
a much nicer and more Pythonic way. In the case of compile time options,
they'd have to be set before the file is compiled, say, before an 'import'
statement. Want to warn for mixed tabs and spaces for a single file, not
using the compile time option? set sys.options["mixedtabs"] to 1, then
import it.

What this means is that you can't have real (non-transitional) options that
change the semantics of the file and distribute the files, because you can't
rely on the option being turned on. This is a good thing! It keeps Python a
single language.

Option directives basically mean "This file is in Python, with directives 2,
4 and 6c turned on." We don't want that. Transitional features have a use,
but only because they'll be mandatory soon.



For problem a) I think 'from __future__' is better, for b) I think
directives are too general, allow things we want to keep away from the
language, this problem is better solved with, say, a sys.options dict.



I could go on, but noone reads this far anyway, I'll wait for the reactions
:-).

-- 
Remco Gerlich


From db3l at fitlinxx.com  Fri Mar  9 20:55:52 2001
From: db3l at fitlinxx.com (David Bolen)
Date: 09 Mar 2001 20:55:52 -0500
Subject: getting the field names from a resultset on Windows with mxODBC
References: <vf6q6.709$eH1.219552882@newssvr16.news.prodigy.com>
Message-ID: <u1ys68k0n.fsf@ctwd0143.fitlinxx.com>

"Scott Hathaway" <slhath at home.com.nospam> writes:

> I am using mxODBC for Windows.  When I do a query, how can I get the names
> of the fields that the resultset returns (without parsing the SQL myself or
> having the user supply them)?

Use <cursor>.description - it's a list of tuples corresponding to your
fields that map to each element of the result set you are getting
back.  It includes name and type information and some other stuff.
This is part of the standard Python DB-API, and is covered in both the
DB-API and mxODBC-specific documentation included with mxODBC.

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


From ransen_spam_me_not at nemo.it  Sun Mar  4 11:15:45 2001
From: ransen_spam_me_not at nemo.it (Owen F. Ransen)
Date: Sun, 04 Mar 2001 16:15:45 GMT
Subject: PyRun_SimpleFile - impossible to call it under Windows?
References: <3aa07704.2672581@news.newsguy.com>
Message-ID: <3aa40bea.3717542@news.newsguy.com>

On Fri, 02 Mar 2001 16:04:54 GMT, ransen_spam_me_not at nemo.it (Owen F.
Ransen) wrote:

>I've seen that there is an odd bug in PyRun_SimpleFile
>probably because of version incompatibilities in the
>FILE structure and using DEBUG modes in VC6

This bug or feature is still there in 2.1b1....


--
Owen F. Ransen
http://www.ransen.com/
Home of Gliftic & Repligator Image Generators


From cce at clarkevans.com  Fri Mar 23 03:02:54 2001
From: cce at clarkevans.com (Clark C. Evans)
Date: Fri, 23 Mar 2001 03:02:54 -0500 (EST)
Subject: PEP 245: Python interfaces
Message-ID: <Pine.LNX.4.21.0103230302200.19638-100000@clarkevans.com>

> http://python.sourceforge.net/peps/pep-0245.html

This is very good work  Michel.  Here are my comments
which I've been mulling over for the last few days.

  a)  First, I'm not sure a new type is needed, perhaps
      classes work just fine (see the PyXML library
      for current interface declaration practices...)
     
      In general, your PEP didn't convince me why
      the class construct couldn't be extended 
      in a reasonable way to satisify the goals
      you enumerated without introducing a new type.

  Assuming that the "class" type can be augmented:

  b)  If the class syntax was extended, perhaps it
      may be useful to add a new statement like
      pass, but could be used to mark "pure" 
      methods (like = 0 in C++), perhaps:

         class X:
            def func(self,z): pure
            
      Then, if any method in a class was marked
      "pure", the class could not be instantiated.
      
  c)  If the class syntax was extended, perhaps
      there might be a way to signify "inherit"
      from "implements", perhaps:

         class X (Y,#Z):
           
      The above class X "inherits" from the
      class Y, and "implements" the public 
      attributes found in class Z, but 
      without sharing the implementation.

   d) What is also missing from Python is
      an "auto-delegation".  An analysis
      of how this fits in or doesn't fit 
      in would be helpful.

   Assuming that a new "interface" type were
   to be added. 

   e) I personally feel that the syntax should
      be much different from the syntax of a 
      class to reduce confusion.  Note that
      maps and lists have different syntaxes.
      The different syntaxes give better 
      context for programmers...

   f) If a new type is added, why not go further?
      Although I don't like the idea of argument
      types in classes... they could very well
      make sence for interfaces.  This could help
      to differentiate the syntax, perhaps:

        interface Y:
           decl func( number as types.IntType, 
                      string as types.StringType,
                      any    as any,
                      cls    as IAnotherInterface ): 
               "This function does such and such"
         
   g)  I like the pre/post condition stuff, however,
       this shoudl be put in a seperate PEP as it
       is seperable and could be added later, no?

   h)  I don't like the string for pre/post conditions.  
       They should have special syntax, perhaps as
       these could be implemented as "local methods?"
   
         interface Y:
             decl func(...) 
                 def pre(self):
                    if ... then raise PreConditionFailure
                 def post(self): pass
                 def test(self): pass

 
   Summary:  I like what you have done, but given your
             requirements, I think we should extend class
             with a "pure" construct instead.

             If we are going to have a new type, then
             a new, more appropriate syntax should be
             created.


Wheu!  I hope this helps! 

Kind Regards,

Clark





From robin at stop.spam.alldunn.com  Fri Mar 16 21:49:31 2001
From: robin at stop.spam.alldunn.com (Robin Dunn)
Date: Fri, 16 Mar 2001 18:49:31 -0800
Subject: socket.ssl for win32
References: <mailman.984773016.25714.python-list@python.org> <Jqws6.22$quX.1114625@news.randori.com> <2_ys6.108066$__6.20130421@typhoon.southeast.rr.com>
Message-ID: <dhAs6.130$quX.18415973@news.randori.com>

"Don Tuttle" <tuttledon at hotmail.com> wrote in message
news:2_ys6.108066$__6.20130421 at typhoon.southeast.rr.com...
> "Robin Dunn"
> > I got impatient and just did it myself.  (Okay, I only waited about 5
> > minutes...)  I've put the binaries at http://alldunn.com/python/ in case
> > anyone else wants them.
>
> Thanks Robin! Would you check the integrity of your  zip file? I'm getting
> the following error when I try to extract socket.py :
> Error in file #1: bad Zip file offset (Error local header signature not
> found): disk #1 offset: 51766
>

It unzipped okay here, so I uploaded it again just in case.

> Do you know if any one has written a telnet server/client in Python that
run
> on Windows? I've been experimenting with SSH & SSHD for CYGWIN But Python
> would be nicer!

Not that I know of.


--
Robin Dunn
Software Craftsman
robin at AllDunn.com       Java give you jitters?
http://wxPython.org      Relax with wxPython!





From grante at visi.com  Fri Mar  9 22:52:56 2001
From: grante at visi.com (Grant Edwards)
Date: Sat, 10 Mar 2001 03:52:56 GMT
Subject: Python In A Nutshell
References: <nG5q6.54910$__6.11376601@typhoon.southeast.rr.com> <Ufcq6.140786$47.2078615@news.bc.tac.net> <Blfq6.60716$__6.11886688@typhoon.southeast.rr.com> <98c4ok$fhd$1@panix2.panix.com>
Message-ID: <slrn9aj98s.ock.grante@isis.visi.com>

In article <98c4ok$fhd$1 at panix2.panix.com>, Aahz Maruch wrote:
>
>>Planning stages?  We've all seen his posts (a.k.a. treatises) here.  I'd say
>>he should have the final draft to you in 3 weeks tops. ;-)
>
>ITYM "first draft".  It will take another three-to-six months for Alex
>to edit it down.  ;-)

"Sorry it's so long, I didn't have time to write it short."

No idea where that quote is from, but I noticed in grad school that
assignments stated maximum lengths allowed for papers while in undergrad
courses they specified minimum lengths for papers.  It sort of gives you an
idea about which types of people go on to grad school. ;)

-- 
Grant Edwards                   grante             Yow!  We have DIFFERENT
                                  at               amounts of HAIR --
                               visi.com            


From samschul at pacbell.net  Sat Mar 31 15:38:46 2001
From: samschul at pacbell.net (samschul at pacbell.net)
Date: 31 Mar 2001 20:38:46 GMT
Subject: I want to learn PYTHON!
References: <1iqx6.3744$aD4.234302@news2.atl>
Message-ID: <9a5f8m$7km$1@news.netmar.com>

In article <1iqx6.3744$aD4.234302 at news2.atl>, KEVIN <kapblp at bellsouth.nOt>
writes:
>Where do I start.
>
>I am not a programmer but have an intermediate level of computer knowledge
>in Windows only.
>
>--
>KEVIN
>
>"Si hoc legere scis, nimis eruditionis habes."
>-----------------------------------------------------------------
>Please repond in the newsgroup.
>My email address is altered for anti-spam
>Just change .nOt to .net
>
>

Try amazon.com and search on Python

Sam Schulenburg

 -----  Posted via NewsOne.Net: Free (anonymous) Usenet News via the Web  -----
  http://newsone.net/ -- Free reading and anonymous posting to 60,000+ groups
   NewsOne.Net prohibits users from posting spam.  If this or other posts
made through NewsOne.Net violate posting guidelines, email abuse at newsone.net


From louis at decipherinc.com  Fri Mar  9 12:46:36 2001
From: louis at decipherinc.com (prince.kattare.com)
Date: Fri, 9 Mar 2001 09:46:36 -0800
Subject: Python & Wireless Apps
Message-ID: <98b4ra$107$1@bashir.peak.org>

Hello!   I just got back from the Python Conference and glad to see that
there is a lot of motivation for the language.  I am just starting out with
python and was wondering how it's being used in the wireless space, meaning
mostly palm devices.  Has anyone ported an app to a palm pilot?  What kind
of restrictions should I be considering when thinking about this?

Thanks for any insight you might have!

Louis
louis at decipherinc.com




From aahz at panix.com  Thu Mar  8 20:33:51 2001
From: aahz at panix.com (Aahz Maruch)
Date: 8 Mar 2001 17:33:51 -0800
Subject: I come to praise .join, not to bury it...
References: <kuofvfjb77.fsf@lasipalatsi.fi> <000001c0a64c$497df250$a9807018@cr706570a> <mailman.983997159.4842.python-list@python.org> <Ntyp6.1479$y6.244056@ruti.visi.com>
Message-ID: <989btv$rd4$1@panix3.panix.com>

In article <Ntyp6.1479$y6.244056 at ruti.visi.com>,
Grant Edwards <grante at visi.com> wrote:
>In article <mailman.983997159.4842.python-list at python.org>, D-Man wrote:
>>
>>Even with my newfound understanding, provided by the Martellibot, I
>>agree with this.  Calling a method on an object is fine.  Calling it
>>on a literal looks weird.
>
>What makes you think a "literal" is not an object?

It is, but not at the aesthetic level.

>> Take for example something like this
>>
>>three = 1.__add__( 2 )
>
>Looks just fine to a ex-Smalltalker: everything is an object.

Python isn't Smalltalk.
-- 
                      --- Aahz (Copyright 2001 by aahz at pobox.com)

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

"..., and some of you may regard all women as evil traps that exist only
to tease, torture, and suck out your very soul."  --DrMax


From jwbnews at scandaroon.com  Fri Mar 16 10:21:25 2001
From: jwbnews at scandaroon.com (John W. Baxter)
Date: Fri, 16 Mar 2001 07:21:25 -0800
Subject: This math scares me
References: <984486395.407379@newsmaster-04.atnet.at> <slrn9asf1d.ja3.qrczak@qrnik.zagroda> <984518199.127140@newsmaster-04.atnet.at> <jwbnews-09D75F.13313313032001@corp.supernews.com> <98ohuc$du9$1@news.udel.edu> <7bQr6.644$lz1.17686@ruti.visi.com> <98oku1$f6a$1@news.udel.edu> <erRr6.659$lz1.20662@ruti.visi.com> <mailman.984664520.26463.python-list@python.org> <HR4s6.877$lz1.73512@ruti.visi.com> <mailman.984689603.19242.python-list@python.org> <locs6.79734$1D5.4121865@e420r-atl1.usenetserver.com>
Message-ID: <jwbnews-1886C5.07212416032001@corp.supernews.com>

In article <locs6.79734$1D5.4121865 at e420r-atl1.usenetserver.com>,
 "Steve Holden" <sholden at holdenweb.com> wrote:

> "Fabio Olive Leite" <olive at conectiva.com.br> wrote in message
> news:mailman.984689603.19242.python-list at python.org...
> > On Thu, Mar 15, 2001 at 03:04:07PM +0000, Grant Edwards wrote:
> > )
> > ) One step further is base-60, which is still used for many
> > ) things.  Time and angular measurement for example is partially
> > ) a base-60 system.  I believe that one or more of the ancient
> > ) Aztecs/Incans/Mayans used a base-60 number system.
> >
> > Indeed, they used it. And some other people too, as is seen even today in
> > the most common languages of the word. French, for example, has no "real
> > words" for seventy, eighty and ninety, and uses "sixty ten", "four twenty"
> > and "four twenty ten".
> >
> > Base 12 was used a lot for commerce, as it has more integer fractions than
> > 10. :)
> >
> What can you expect from a race who insist on calling seventeen, eighteen
> and nineteen "ten-seven", "ten-eight" and "ten-nine" respectively? And let's
> not get into punds, shillings and pence.
> 
> old-enough-to-remember-them-ly y'rs  - steve

And stone, furlong, and fortnight:  a valued employee at JPL in the 
early 1960s was tired of all the different measurement systems in which 
data arrived at the Lab from various field locations.  [At least four 
methods of expressing Azimuth, alone.]  He converted everything for one 
(formal) report into the stone/furlong/fortnight system.  ;-)

And let's not forget the 32-point circle used in [English-descended] 
sailing (STILL used in sailing, including iron jib sailing):  "two 
points off the starboard bow" is somehow easier to understand than "22.5 
degrees off the starboard bow".

This floating point thread has become more fun than any of the scores of 
others I've seen here and elsewhere.

  --John (probably bearing roughly WNW 1/4N from *someone* reading this)

-- 
John W. Baxter   Port Ludlow, WA USA  jwbnews at scandaroon.com


From bsb at winnegan.de  Fri Mar 30 05:56:47 2001
From: bsb at winnegan.de (Siggy Brentrup)
Date: 30 Mar 2001 12:56:47 +0200
Subject: HELP urllib POST can't get it to work
In-Reply-To: <9a1m91$fdg$1@pheidippides.axion.bt.co.uk>
References: <9a1m91$fdg$1@pheidippides.axion.bt.co.uk>
Message-ID: <87snjvtths.fsf@winnegan.de>

"mike mcleod" <mike.mcleod at bt.com> writes:

> My code:
> 
> import urllib
> params = urllib.urlencode({'address': '07', 'identity': 'me',
>                            'resultpg':'html/sendresult.html', 'message':
> 'test',
>                            'fPage':'true',
>                            })
> f = urllib.urlopen("http://www.samemachine.uk/bin/afunc.pl?", params)
> print f.read()
> 
> It looks like the order of the params come out different in the urlopen.
> But because the server will only accept in the order shown it fails.
> I have tried to do this using basic sockets but don't know how to get that
> work without lots of reading.

>From looking at 2.1b1 lib/urllib.py: use a list of pairs instead of a
dictionary (untested)

params = urllib.urlencode( [('address','07'), ('identity','me'),
                           ...

HTH
  Siggy

-- 
Siggy Brentrup - bsb at winnegan.de - http://www.winnegan.de/
****** ceterum censeo javascriptum esse restrictam *******




From gtalvola at nameconnector.com  Fri Mar  2 09:55:48 2001
From: gtalvola at nameconnector.com (Geoff Talvola)
Date: Fri, 02 Mar 2001 09:55:48 -0500
Subject: [ANNOUNCE] Webware for Python 0.5
References: <5.0.2.1.0.20010301205419.047bfd90@mail.mindspring.com>
Message-ID: <3A9FB474.A57363A4@NameConnector.com>

Chuck Esterbrook wrote:

> "Don Tuttle" <tuttledon at hotmail.com> wrote in message
> news:<fFCn6.13480$Z8.1902263 at typhoon.southeast.rr.com>...
> >"Chuck Esterbrook"
> > >    * Python Server Pages (PSP, similar to ASP, PHP and JSP)
> >I 'm trying to get a handle on why I would want to use PSP on WinNt with
> >IIS.  What's the difference/advantage to using PSP compared to using ASP
> >with embedded Python script?
> >Don
>
> [1] Webware's PSP offers more syntax options than ASP. If I remember
> correctly, using Python in ASP involves some funkiness because of the clash
> between Pythonic indentation and HTML. I don't remember the details, but
> I'm positive they are chronicled in the archives of c.l.p.
>
> PSP also allows you to inherit one or more base classes which can be good
> for reusing code. I don't know if ASP lets you do that or not.

This is a great feature of PSP that I make use of a lot, and I am pretty sure ASP offers nothing similar.

> In any case, if you check the docs for PSP, you can see what various
> options you have compared to ASP:
>
> http://webware.sourceforge.net/Webware/PSP/Docs/
>
> [2] Webware's PSP sits on top of WebKit, which means that when you use it,
> you can also create plain old Python servlets. You can also make your own
> "servlet factories". For example, you can write a handler that serves up
> *.sql, *.mytemplate, etc. There are also other components you can use with
> WebKit such as UserKit, MiddleKit, etc.  Of course, that doesn't matter if
> you're not interested in these options, although using Webware puts you in
> a position to use them any time later.
>
> [3] A colleague of mine has had "caching" problems in ASP regarding out of
> date code residing in memory despite file modifications and reload()s. I
> can't say I know the details, and I won't say that you are guaranteed to
> suffer this if you use ASP+Python.
>
> Hope that helps. More details might be available from people who have
> extensively used both.
>
> -Chuck

I recall that ASP with Python places limitations on what you can store in session variables.  For example, if you try to place a dictionary into a session variable you'll get an error like "TypeError: Objects of type 'dictionary' can not be converted to a COM VARIANT".  Generally, only "simple" types like integers, floats, strings, and lists of them can be placed into a session variable.  WebKit/PSP has no such restrictions.

One more feature of WebKit/PSP that I _think_ doesn't exist in ASP, but I might be wrong:  The ability to forward a request to another Servlet or PSP purely within the app server (as opposed to redirecting the browser).  This allows you to do things like write a "controller" servlet that examines the request and then forwards to the appropriate servlet or PSP.  It's a nice model for certain types of problems that is really simple to implement in WebKit/PSP, and I think it would be difficult or maybe even impossible in ASP.

On the flip side, one possible advantage of ASP with Python is that you might be able to take advantage of 3rd-party extensions that are designed to only work with ASP.  I don't know if such things exist.  I happen to be using a 3rd-party COM component in my web pages to draw dynamic graphs.  The component was designed for use with Visual Basic and Active Server Pages, but it works great with WebKit too.

ASP/Python also has the dubious advantage of being an easier sell within a corporation than WebKit/PSP, which may or may not be important to you...

Overall, I think WebKit/PSP is significantly more productive and powerful than ASP, and I was able to convince my coworkers of that, so it's working out great for me :-)

--


- Geoff Talvola
  Parlance Corporation
  gtalvola at NameConnector.com




From scarblac at pino.selwerd.nl  Tue Mar 20 04:00:58 2001
From: scarblac at pino.selwerd.nl (Remco Gerlich)
Date: 20 Mar 2001 09:00:58 GMT
Subject: Sequence-spreading
References: <3AB7167F.2816AA21@student.liu.se>
Message-ID: <slrn9be6tc.b0k.scarblac@pino.selwerd.nl>

Rikard Bosnjakovic <rikbo716 at student.liu.se> wrote in comp.lang.python:
> I made a small spreading-sequence with the following code:
> 
> ------
> arg = "some string"
> pm = 3
> offsets = []
> for i in xrange(pm):
>     everyX = filter(lambda y: not divmod(y, pm)[1], range(len(arg)))
>     added = map(lambda x: x+i, everyX)
>     offsets.append(filter(lambda x: x<len(arg), added))
> 
> print offsets
> ------
> 
> which gives the correct result:
> 
> >>> ## working on region in file /usr/tmp/python-3115DCp...
> [[0, 3, 6, 9], [1, 4, 7, 10], [2, 5, 8]]
> 
> "pm" is the number of sequences, and then the sequences are increasing.
> 0 in first, 1 in second, 2 in third, 3 in first and so on, until the
> length of the string 'arg' is reached.  However, the loop-code is pretty
> unreadable and awkward.
> 
> Anyone got a better solution for it?

arg = "some string"
pm = 3
offsets = [ [] for i in range(pm) ]
for i in range(len(arg)):
   offsets[i % pm].append(i)


-- 
Remco Gerlich


From cce at clarkevans.com  Tue Mar 27 04:15:34 2001
From: cce at clarkevans.com (Clark C. Evans)
Date: Tue, 27 Mar 2001 04:15:34 -0500 (EST)
Subject: Yet Another PEP:  Query Protocol Interface or __query__
In-Reply-To: <5.0.2.1.0.20010325210220.00a2cc00@mail.inet.com.br>
Message-ID: <Pine.LNX.4.21.0103270413201.17716-100000@clarkevans.com>

On Mon, 26 Mar 2001, Carlos Ribeiro wrote:
> 6) We can solve (5) using a 'proxy', or 'adapter'.

Ahh, it finally got though my dence skull what you
are proposing -- more of a proxy mechanism, which 
has a further constraint, not only must the expected
methods be mapped, but *only* the expected methos
are to be mapped.

> EXAMPLE
> 
> Assume that we have these protocols:
> 
> a) protocol 'IntegerDivision'
>     supports the __div__() method for integer division
> b) protocol 'FloatDivision'
>     supports the __div__() method for floating point division
> 
> #----------------------------------------------------------------------
> 
> class Protocol:
>    """ empty class that can be used for all protocol adaptation """
>    pass
> 
> def MakeProtocolDefault(protocol_methods, adaptee):
>    """ Returns the default protocol adapter for an object
>        The default uses the same name for all methods in the protocol
>        and in the object instance that is being adapted
>    """
>    p = Protocol()
>    for pm in protocol_methods:
>      om = getattr(adaptee, pm)
>      setattr(p, pm, om)
>    return p
> 
> def MakeProtocol(protocol_methods, object_methods):
>    """ Returns the a protocol adapter for an object, mapping
>        each protocol_method to a object_method
>    """
>    p = Protocol()
>    for pm, om in zip(protocol_methods, object_methods):
>      setattr(p, pm, om)
>    return p
> 
> class Number:
>    """ supports both protocols: 'IntegerDivision' & 'FloatDivision' """
>    def __init__(self, p_value):
>      self.value = p_value
> 
>    def div_int(self, b):
>      # integer division
>      return int(self.value)/int(b)
> 
>    def div_float(self, b):
>      # floating point division
>      return float(self.value)/float(b)
> 
>    def __adapt__(self, protocol):
>      if protocol == 'IntegerDivision':
>        return MakeProtocol(['__div__'], [self.div_int])
>      elif protocol == 'FloatDivision':
>        return MakeProtocol(['__div__'], [self.div_float])
>      else:
>        return None
> 
> #----------------------------------------------------------------------
> 
>  >>> n = Number(2)
>  >>> in = n.__adapt__('IntegerDivision')
>  >>> fn = n.__adapt__('FloatDivision')
>  >>> in/3
> 0
>  >>> fn/3
> 0.66666666666666663
>  >>> n.value = 5
>  >>> fn/3
> 1.6666666666666667
>  >>>
> 
> #----------------------------------------------------------------------

Very Neat. 

;) Clark




From fdrake at acm.org  Sat Mar 24 12:35:27 2001
From: fdrake at acm.org (Fred L. Drake, Jr.)
Date: Sat, 24 Mar 2001 12:35:27 -0500 (EST)
Subject: [Python-Dev] Anyone using weakrefs?
In-Reply-To: <018e01c0b456$6710b830$e46940d5@hagrid>
References: <15035.46175.599654.851399@localhost.localdomain>
	<018e01c0b456$6710b830$e46940d5@hagrid>
Message-ID: <15036.56031.221142.247490@cj42289-a.reston1.va.home.com>

Fredrik Lundh writes:
 > am I the only one who gets a bit nervous when the guy who
 > designed and implemented a feature has a hard time finding a
 > non-contrived use case? ;-)

  I had a non-contrived need for them 5 1/2 years ago!  But I worked
around it in C code back then, since I was building a C extension
anyway; the result was not generally applicable since it was tightly
tied to the particular C extension I needed.
  And I'm sure we could have used this in Grail as well.  ;-)


  -Fred

-- 
Fred L. Drake, Jr.  <fdrake at acm.org>
PythonLabs at Digital Creations




From loewis at informatik.hu-berlin.de  Sat Mar 24 05:32:16 2001
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 24 Mar 2001 11:32:16 +0100
Subject: New PEP: The directive statement
References: <mailman.985414952.22988.python-list@python.org>
Message-ID: <j466gzo3tb.fsf@informatik.hu-berlin.de>

"Tim Peters" <tim.one at home.com> writes:

> The entire point of the PEP is to give incompatible language changes
> a trial run of at least one full release cycle before saying "that's
> it".  In what respect does that come off as absolute certainty to
> you?  And in what respect does your PEP differ in relation to this
> point?  If you have a better idea, change your PEP to spell it out.

That was rather meant as a comment to PEP 236: Instead of giving a
mandatory release, it appears that all you can specify is a
is_mandatory (isMandatory, or mandatory_p) flag; this is also all you
said an application might need.

> > Tim claims that I invented the statement to declare file encodings,
> > e.g.
> >
> > directive encoding "koi8-r"
> 
> Gimme a break

Ok, I should have put a smiley in there, I guess :-) It is a fact that
you were claiming it, it is also a fact that I do like to see the
possibility to specify source encodings. 

It's just that I cannot hope to get that now - but I had the hope that
the future statement would be removed before the 2.1 release. Now that
the final beta still has it, that hope is vanishing...

Regards,
Martin


From C.Brewster at dcs.shef.ac.uk  Thu Mar 15 05:02:40 2001
From: C.Brewster at dcs.shef.ac.uk (Christopher Brewster)
Date: Thu, 15 Mar 2001 10:02:40 -0000
Subject: reversing a dictionary (newbie)
Message-ID: <007201c0ad37$12571f80$6809a78f@dcs.shef.ac.uk>

I have written a program to extract pairs of words from a corpus (the BNC).
It is my first Python program
Once it has processed the files I "reverse" the dictionary so that the keys
are frequencies and the values lists of word pairs of that frequency.

The processing of the files is quite quick (700 odd files (@ 100k mean size)
of directory 'A' in about 20 minutes)
but the reversing of the dictionary took hours and hours. I have run the
program successfully on a PC (500 MHz, 256 RAM, running Windows NT) and I am
still waiting for results from a Sun (1 Giga RAM) since yesterday. Here is
the code - what have I done wrong?

def revdict(diction):
	revdict = {}
	for wordpair in diction.keys():
		#print wordpair
		list = []
		freq = diction[wordpair]
		if revdict.has_key(freq):
			list.extend(revdict[freq])
			list.append(wordpair)
			revdict[freq] = list
		else:
			list.append(wordpair)
			revdict[freq] = list
	return revdict

The total number of 'wordpair's (i.e. keys in the original dictionary) were
567000. I need to be able to handle ten times that many.

Thank you,

Christopher Brewster

Department of Computer Science, University of Sheffield
Tel: +44(0)114-22.21872  Fax: +44 (0)114-22.21810
Regent Court, 211 Portobello Street
Sheffield   S1 4DP   UNITED KINGDOM





From danielk at aracnet.com  Tue Mar 13 21:49:44 2001
From: danielk at aracnet.com (Daniel Klein)
Date: Tue, 13 Mar 2001 18:49:44 -0800
Subject: Where can I find Docs on how to use the debugger please?
References: <3lltat8tgd86rknhgnqkij6bk9rkbqk824@4ax.com>
Message-ID: <u0ntatshpl6f5lp0eueksg1iuoa0hga3h4@4ax.com>

On Wed, 14 Mar 2001 13:35:45 +1100, cmfinlay at SPAMmagnet.com.au wrote:

>Where can I find Docs on how to use the debugger please?

Look up the 'pdb' module in the Library Reference.

Daniel Klein


From root at rainerdeyke.com  Sat Mar 17 12:25:04 2001
From: root at rainerdeyke.com (Rainer Deyke)
Date: Sat, 17 Mar 2001 17:25:04 GMT
Subject: [Very Long (11K)] Numeric PEPs, first public posts
References: <mailman.984748941.31248.python-list@python.org>
Message-ID: <Q5Ns6.432845$ge4.152574719@news2.rdc2.tx.home.com>

"Moshe Zadka" <moshez at zadka.site.co.il> wrote in message
news:mailman.984748941.31248.python-list at python.org...
> PEP: 237
> Title: Unifying Long Integers and Integers
> Rationale
>
>     Having the machine word size exposed to the language hinders
>     portability.  For examples Python source files and .pyc's are not
>     portable because of this.  Many programs find a need to deal with
>     larger numbers after the fact, and changing the algorithms later
>     is not only bothersome, but hinders performance in the normal
>     case.

A simpler way to solve this problem would be to standardize Python ints as
32 bit.

> Implementation
>
>     The PyInt type's slot for a C long will be turned into a
>
>         union {
>             long i;
>             struct {
>                 unsigned long length;
>                 digit digits[1];
>             } bignum;
>         };
>
>     Only the n-1 lower bits of the long have any meaning; the top bit
>     is always set.  This distinguishes the union.  All PyInt functions
>     will check this bit before deciding which types of operations to
>     use.

>From a performance point-of-view, it might be better to keep integers and
longs as separate types.  Not sure if this makes any real difference.

> PEP: 238
> Title: Non-integer Division

>         (a/b) * b = a (more or less)
>
>     The type of a/b will be either a float or a rational, depending on
>     other PEPs[2, 3].

Please, no floats.  This newsgroup is evidence that floating point numbers
are much more confusing than integer truncation.  Rationals I can live with,
but I think the operator '/' should keep its semantics and the new operator
should produce the rationals.  This would be consistent with the current
system where 'a / b' has the same type as 'a' and 'b' if 'a' and 'b' have
the same type.

> PEP: 239
> Title: Adding a Rational Type to Python

> RationalType
>
>     There will be a new numeric type added called RationalType.  Its
>     unary operators will do the obvious thing.  Binary operators will
>     coerce integers and long integers to rationals, and rationals to
>     floats and complexes.
>
>     The following attributes will be supported: .numerator and
>     .denominator.  The language definition will not define these other
>     then that:
>
>         r.denominator * r == r.numerator
>
>     In particular, no guarantees are made regarding the GCD or the
>     sign of the denominator, even though in the proposed
>     implementation, the GCD is always 1 and the denominator is always
>     positive.

I don't see the point in not guaranteeing this.  Undefined behavior is bad.

> PEP: 240
> Title: Adding a Rational Literal to Python
> Abstract
>
>     A different PEP[1] suggests adding a builtin rational type to
>     Python.  This PEP suggests changing the ddd.ddd float literal to a
>     rational in Python, and modifying non-integer division to return
>     it.

I very much like this idea (because I hate floats), but I suspect this will
break too much code.  A suffix similar to the 'L' after long literals might
work better.


--
Rainer Deyke (root at rainerdeyke.com)
Shareware computer games           -           http://rainerdeyke.com
"In ihren Reihen zu stehen heisst unter Feinden zu kaempfen" - Abigor




From fritz.heinrichmeyer at fernuni-hagen.de  Mon Mar 26 02:02:59 2001
From: fritz.heinrichmeyer at fernuni-hagen.de (Fritz Heinrichmeyer)
Date: 26 Mar 2001 09:02:59 +0200
Subject: Python ready to replace perl (was Re: Install Python on Apache)
References: <99fph5$a2n$1@rex.ip-plus.net> <wzelvnz4l5.fsf@sunshine.cs.uu.nl>
Message-ID: <86snk1qafw.fsf_-_@jfh00.fernuni-hagen.de>

A student who could help me with web support is more fluent with python
than with perl (that is no a feat :-).
So i consider using python in the future. For this python has to have

1. sql-interface like dbi, at least for postgres (i think this exists).

2. interface to openldap 

3. mod_perl beeing able to program authentification procedures using
   1. and 2. for apache and beeing able to program arbitrary URL-redirect
   procedures.

4. a template system beeing able to manage multiple language versions
   (to replace the perlish WML-System)?.

Are there solutions apart from using zope?

-- 
Fritz Heinrichmeyer mailto:fritz.heinrichmeyer at fernuni-hagen.de
FernUniversitaet Hagen, LG ES, 58084 Hagen (Germany)
tel:+49 2331/987-1166 fax:987-355 http://www-es.fernuni-hagen.de/~jfh


From inconnu at nul.net  Sat Mar 31 15:15:05 2001
From: inconnu at nul.net (Rémi yyyy)
Date: Sat, 31 Mar 2001 22:15:05 +0200
Subject: C++ and Python
References: <99q99l$g21$1@news.clinet.fi>
Message-ID: <9a5dr3$9fj$1@front7m.grolier.fr>

See in http://www.swig.org/ for C++ Wrapper
Ahto <ahto at jussila.as> a ?crit dans le message :
99q99l$g21$1 at news.clinet.fi...
> I'm very new to Python and haven't quite figured out how to use c++ and
> python together on windows...if someone could send me an example program
of
> tutorial webpage where this is explained?!?
>
> Thanks
>
>




From tim.hochberg at ieee.org  Sat Mar  3 11:24:30 2001
From: tim.hochberg at ieee.org (Tim Hochberg)
Date: Sat, 03 Mar 2001 16:24:30 GMT
Subject: the demise of 'from foo import * and its implications?
References: <3kVBAfATrDo6EwEM@jessikat.fsnet.co.uk> <mailman.983582584.8198.python-list@python.org> <z8dAGJAyqLo6EwlX@jessikat.fsnet.co.uk>
Message-ID: <2V8o6.1383$8q3.219969@news1.rdc1.az.home.com>

"Robin Becker" <robin at jessikat.fsnet.co.uk> wrote in message

[SNIP]

> If we really want to improve visibility we could try to make
>
> >>> def bingo(a):
> ...     if a<=1: return 0
> ...     return bingo(a-1)*a
> ...
>
> work naively, but I'm fairly sure this will be rejected as unpythonic or
> somesuch.


?????

In what sense doesn't this work now? Under 2.0 or 1.5.2 or even, I think,
under 0.9 this 'works' in the sense that it iterates for a while then
returns 0. If one replaces "return 0" with "return 1" it even returns the
factorial as I assume it should.

What doesn't work is the following:

def spam(n):
   def fact(a):
      if a <= 1: return 1
      else: return fact(a-1)*a
   return fact(n) + 42

With the current three level scope rule this fails, so a function that works
fine when defined at the top level fails miserably when defined at an
intermediate level. Given your earlier statement that:

>my preference would be that statements legal in one context should be
>legal in another where they make sense.

This seems like it would be of concern to you as well.

My understanding is we can have two of the following three options:

1. Decent performance (~current performance)
2. "Real" lexical scoping
3. Unrestricted use of  "from X import *" at other than module scope.

Given those choices I'd choose 1 and 2. Which is fortunate for me, since
that's what were going to get anyway.

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

Going off on a tangent: In listening to the various debates here, and when I
was spying on PythonDev, I've come to the conclusion that "from X import *"
was a mistake to begin with. It has it's uses, but there are other ways to
satisfy them without the downsides of "import *".

"import *" is used, in essence, to extend the list of __builtins__ that one
sees in a module or in interactive mode. This is commonly used with some
windowing toolkits for example, where importing the contents an item at a
time is too painful (using from X import a,b,c,d....) and where the names of
the toolkit's contents have been constructed not to stomp on the local
namespace.

The problems with "import *" include:
* May inadvertenly stomp on local namespace
* Can't easily reload module contents (OK, so it's not that hard,  but it's
hard to explain sometimes).
* Gives the compiler a headache in inner scope.

Let's consider another approach that I believe satisfies the current uses
for "import *" without its drawbacks.

<COUNTERFACTUAL WORLD>
__module_builtins__ is a list of modules, each of which is searched in turn
when lookup fails at the global (aka module) scope. Initally,
__module_builtins__ equals [__builtins__], so the behaviour is the same as
in the real world, except for going through an additional layer. But if
additional modules are appended to __module_builtins__, they are each
searched in turn to resolve a name lookup.

   import spam as * # I don't defend this syntax, it's only for illustration

Is shorthand for:

   import spam
   __module_builtins__.append(spam)
  del spam

So, to make all of the wxPython stuff available one types:

   import wxPython.wx as *
   class MyApp(wxApp):
       # ....

This looks first in the module dictionary, then in __builtins__ (since it's
the first entry in __module_builtins__) then in wxPython.wx, where it would
find wxApp. It involves extra lookups over "import *", but if anyone needs
speed in a particular place, they can import directly into the local scope,
as they do now.

</COUNTERFACUTAL WORLD>

The nice things about this approach include:

* It can't stomp on local variables
* Reload just works.
* You can use it wherever you want without giving the compiler fits since is
only affects __module_builtins__, not the local scope.
* You can cleanly remove imported  modules by deleting them from
_builtin_modules__, something that's difficult to do with import *
* Others to numerous to mention.

However, at this point I imaging it's many years to late, and "from spam
import *" is here to stay.

If you've put up with my rambling to this point I congratulate you.

Ramblin on,

-tim








.




From me at home.nl  Sun Mar 11 06:49:49 2001
From: me at home.nl (SonoBull)
Date: Sun, 11 Mar 2001 11:49:49 GMT
Subject: question , Does pytyon excist on a pda, like psion?
Message-ID: <xDJq6.28348$KE1.1127955@zwoll1.home.nl>

Hello all,

Is Python already ported to the psion or other pda, s?

greetings.

Sono.






From jepler at inetnebr.com  Sun Mar 11 21:12:54 2001
From: jepler at inetnebr.com (Jeff Epler)
Date: Mon, 12 Mar 2001 02:12:54 GMT
Subject: Subclass factories?
References: <mailman.984356655.27149.python-list@python.org>
Message-ID: <slrn9aoc56.2k6.jepler@potty.housenet>

On Mon, 12 Mar 2001 11:23:24 +1100, Tim CHURCHES
 <TCHUR at doh.health.nsw.gov.au> wrote:
>If the subclass factory approach is reasonable, what is the best way
>to do it? So far I been assembling the necessary statements for the
>new class definition into a string then exec'ing that string, but that
>seems rather inelegant. I'm sure there must be a better way.

Use the "new" module to create a class with a particular list of superclasses
and a particular dictionary of functions or attributes.

Example:
>>>import new
>>> class A:
...    pass
>>> class B:
...    pass
>>>def f(self):
...    print "it's f", self
>>> D = new.classobj("from new.classobj", (A, B), {"f": f})
>>> print D.__bases__
(<class __main__.A at 80c5080>, <class __main__.B at 80c6a10>)
>>> print dir(D)
['__doc__', '__module__', 'f']
>>> d = D()
>>> print d.f()
it's f <__main__.from new.classobj instance at 80c5b00>

Jeff


From jmarshal at mathworks.com  Fri Mar 16 11:40:10 2001
From: jmarshal at mathworks.com (Joshua Marshall)
Date: 16 Mar 2001 16:40:10 GMT
Subject: Python style questions
References: <98tb91$9t9$1@saltmine.radix.net>
Message-ID: <98tfla$gck$1@news.mathworks.com>

Cary O'Brien <cobrien at radix.net> wrote:
...
> 3. I really miss not having a "switch" or "case" statement.  Sniff.

Me too.

> 4. I need to change things like "88aa99bb" (hex string) into integers.  Should I

>    a. use expr("0x"+s) 
>    b. not use such things
>    c. ???

c. int(s, 16)

This works in Python 2.x.  If you're using an older version, use string.atoi.


From NoSpam at NoSpam.com  Sun Mar 11 11:16:10 2001
From: NoSpam at NoSpam.com (Tom)
Date: Sun, 11 Mar 2001 16:16:10 GMT
Subject: Memory leaks in VC6 2 line Python 2.1b1 apps?
References: <3aa88f52.5419185@news.newsguy.com> <hGPp6.53719$UZ4.13963947@news4.rdc1.on.home.com> <98b0a0$2g8i$1@shadow.skypoint.net> <3aa91ea5.645940@news.newsguy.com> <98dngs$2cmm$1@shadow.skypoint.net>
Message-ID: <exNq6.68746$UZ4.17890524@news4.rdc1.on.home.com>

Craig,

The debug CRT can give you this info.
Add the following to your program to enable this feature.  I put it in my
precompiled header.

// For memory debugging (VC++ specific)
// Must come after some STL includes (or conflicts with 'new')
#ifdef _DEBUG
//#include <stdlib.h> // seen this recom., but I don't see why
#define _CRTDBG_MAP_ALLOC
 // Causes CrtDbg to #define all the crt alloc routines to versions with
more debug info,
 // to replace operator new with one that adds debug info.
 //  (but in a stupid way that doesn't work - so we #define new ourselves
after CrtDbg.h.)
#include <crtdbg.h>
#define new new(_NORMAL_BLOCK, __FILE__, __LINE__)

For more info, see the crt docs in MSDN.  There are a number of options.

Let me know if you are using wxWindows - this requires more changes.

Tom.

"Craig Holman" <craig at patterncraft.com> wrote in message
news:98dngs$2cmm$1 at shadow.skypoint.net...
> I wish that Visual Studio's memory-leak reporter were smart enough to
> display where the leaked blocks of memory were allocated.  It only puts
out
> pairs of lines like the following :
>
> {1037} normal block at 0x0079CE80, 39 bytes long.
>  Data: <  y             > 20 EE 79 00 88 85 1F 1E 03 00 00 00 B8 17 20 1E
>
> The other lines in the annotated dump were added by me.  They contain the
> results of my efforts to figure out where the leaked blocks were
allocated.
> After each of the pair of lines produced by the memory-leak reporter, I've
> listed my best guess as to what kind of a block was leaked and why, the
line
> in the Python 2.1b1 source file that I placed a breakpoint at that
allocates
> the leaked memory block, and a snapshot of the relevant part of the
runtime
> stack at the point that the breakpoint was hit and execution was paused.
I
> haven't identified where all of the leaked blocks are being allocated yet.
>
> Hope this helps,
> Craig
>
> "Owen F. Ransen" <ransen_spam_me_not at nemo.it> wrote in message
> news:3aa91ea5.645940 at news.newsguy.com...
> > On Fri, 9 Mar 2001 10:28:33 -0600, "Craig Holman"
> > <craig at patterncraft.com> wrote:
> >
> > >Greetings,
> > >
> > >I'm afraid that both versions 2.0 and 2.1b1 leak memory for a single
> > >Py_Initialize, Py_Finalize sequence in a
> > >Visual C++ MFC application.
> >
> > Phew! I thought it was just me!  ;)
> >
> > I hope someone understands the rest of your post, I'm too new
> > to Python embedding to help. Also, for some reason, though I
> > did rebuild from source, VC gives me the leak "number" but not
> > the line in the library which caused it.
> >
> >
> > --
> > Owen F. Ransen
> > http://www.ransen.com/
> > Home of Gliftic & Repligator Image Generators
>
>




From scarblac at pino.selwerd.nl  Wed Mar 21 07:25:23 2001
From: scarblac at pino.selwerd.nl (Remco Gerlich)
Date: 21 Mar 2001 12:25:23 GMT
Subject: Checking the type
References: <99a493$bhl$1@sun.rhrk.uni-kl.de>
Message-ID: <slrn9bh78g.dju.scarblac@pino.selwerd.nl>

Markus Reitz <Markus_Reitz at yahoo.com> wrote in comp.lang.python:
> Hi,
> 
> is there a possibilty to check if a variable refers to an object of type x?
> Does a typeof-command exist in Python:
> 
>   if a typeof list:
>      #Commands only useable with list
> 
> I read the Python Tutorial, but have not found a hint to this topic.

There is a the type() function, and the list of types in the types module.

You can also use isinstance(), it's slightly more general.

>>> isinstance(3, types.IntType)
1
>>> class Spam:
...   pass
...
>>> isinstance(Spam(), Spam)
1

In many cases, the Python way is to just assume that your input is a list,
and use it as if it were one. If it's not, an exception will be raised as
appropriate. This make it possible to make classes that are not lists but
behave like one (like UserList does). Same for files.

-- 
Remco Gerlich


From bob_cannard at mentor.com  Wed Mar 28 13:10:21 2001
From: bob_cannard at mentor.com (Bob Cannard)
Date: Wed, 28 Mar 2001 10:10:21 -0800
Subject: R: unique items in lists
References: <_nZv6.2878$yU4.468236@news.infostrada.it> <slrn9c0ph1.o5h.scarblac@pino.selwerd.nl> <t5_v6.3466$yU4.477532@news.infostrada.it> <slrn9c1424.o5h.scarblac@pino.selwerd.nl> <3AC0E010.3CDC94FC@mentor.com> <slrn9c37vt.q06.scarblac@pino.selwerd.nl>
Message-ID: <3AC2290D.F79F1B70@mentor.com>


Remco Gerlich wrote:
> Dictionaries are hashes, I'm assuming a dictionary setattr is about constant
> time. The "dict.keys()" at the end must be linear, since its result is
> linear...

Looking back over the messages, it seems that I misread Remco's earlier
message - for some reason I read it as meaning the dictionary lookups
were linear, which would certainly be bad news! Oh well. At least
my analysis of the original algorithm appears to be correct.

Moral: never post while suffering from caffeine deficiency.

Cheers,

        Bob.


From ransen_spam_me_not at nemo.it  Wed Mar  7 13:02:58 2001
From: ransen_spam_me_not at nemo.it (Owen F. Ransen)
Date: Wed, 07 Mar 2001 18:02:58 GMT
Subject: ANNOUNCE: 2.1b1 SRPM
References: <mailman.983944149.1622.python-list@python.org>
Message-ID: <3aa861c3.1122387@news.newsguy.com>

On Tue, 6 Mar 2001 22:48:35 -0700, Sean Reifschneider
<jafo-pythonlist at tummy.com> wrote:

>The SRPM is available at:
What is an SRPM?


--
Owen F. Ransen
http://www.ransen.com/
Home of Gliftic & Repligator Image Generators


From jurgen.defurne at philips.com  Fri Mar 16 02:05:58 2001
From: jurgen.defurne at philips.com (jurgen.defurne at philips.com)
Date: Fri, 16 Mar 2001 08:05:58 +0100
Subject: This math scares me
Message-ID: <0056900016651515000002L052*@MHS>

It was the Babylonians who gave us the base-60 numbered
division of circular events.

Jurgen




grante at visi.com@SMTP at python.org on 15/03/2001 16:17:21
Sent by:	python-list-admin at python.org
To:	python-list at python.org@SMTP
cc:	 
Subject:	Re: This math scares me
Classification:	

In article <mailman.984664520.26463.python-list at python.org>, Fabio Olive Leite wrote:
>Hi there,
>
>On Wed, Mar 14, 2001 at 09:31:54PM +0000, Grant Edwards wrote:
>) No.  The binary floating point representation of many decimal
>) fraction values is not exact.  That's what started this whole
>) thread.  You can't represent 0.1 exactly in binary FP.  You can
>) in BCD.
>
>Just to make matters worse in this discussion, let me add some more noise.
>
>Since base 10 contains base 2 and base 5 (`factor 10`), every
>number coming from any of those bases is exactly represented in
>base 10. Conversely, base 2 contains only itself, and thus it
>cannot represent correctly some numbers that are exactly
>representable in other bases, like 0.1 in base 10.
>
>The usual 1/3 example is a problem for base 10, but not so for
>base 12, for example, since 12 contains 3 2 2. Thus:
>
>1_{12} / 3_{12} == 0.4_{12}
>
>And then base 30 would let us count exactly fractions of base
>2, 3, 5, 6, 10, 12, 15, 16 and all the others you can think of
>that have only 2, 3 and 5 as factors of the base. But since no
>one (usualy) has 30 fingers, that base is not terribly popular.
>:)

One step further is base-60, which is still used for many
things.  Time and angular measurement for example is partially
a base-60 system.  I believe that one or more of the ancient
Aztecs/Incans/Mayans used a base-60 number system.

--
Grant Edwards                   grante             Yow!  It's hard being
                                  at               an ARTIST!!
                               visi.com
--
http://mail.python.org/mailman/listinfo/python-list





From lac at cd.chalmers.se  Tue Mar  6 13:11:22 2001
From: lac at cd.chalmers.se (Laura Creighton)
Date: Tue, 6 Mar 2001 19:11:22 +0100 (MET)
Subject: Can anyone recomend a good intoduction to C...
Message-ID: <200103061811.TAA27998@boris.cd.chalmers.se>

Hello Werner

you write:

>I further think it is easier to move 'back' from C++ to C, when the need
>arises, than vice versa. This is simply because you are forced to use a
>different technique if the one you are used to does not work on that
>platform.
>Personally I started with procedural programming and moved to the object
>oriented technique afterwards. This is why I think it is hard to get used to
>objects when you are used to malloc, free, memcpy and friends.
>
>Did I miss an important point?

2 points.  First of all, while you may be correct that your first exposure
to procedural programming has made it hard for you to get used to
object oriented programming, it is also possible that your problem was that
C++ is a widely-avialable-but-not-particularily-good programming language. 
People whose first OO programming language is Smalltalk have, in my experience
a far easier time learning OO programming.  I do not know anybody whose 
first OO language was Eiffel, but I suspect they have an easier time as well. 
I think that the Smalltalk courses which I took at university, which
were optional, and I only took because I *like* learning languages ended
up being the most valuable courses I ever took.

So the question becomes, why is it that the original poster decided to
learn C or C++?  If he has an application which he thinks is suited to
one or the other, more power to him, though it is hard to make such
decisions until you have more experience with a host of computer languages.
Maybe he just has a compiler and wants to play with it.  Maybe he
has a linux or unix system and wants to find out exactly what it is
that his system is doing.  But finally, it may be that it is more
experience with computer languages in general which is what he wants,
in which case I would say, yes, by all means learn C.  Then go out and 
learn Smalltalk, or Lisp or Haskell.  And make sure you learn an  assembler 
as well on the way, because that too will make your brain go off and work 
in amazing new directions.   This is a keen source of personal pleasure.
They say that you cannot feel new neural connections being made in your 
brain, but I am not so sure about that.

Laura Creighton



From jmarshal at mathworks.com  Fri Mar 16 14:44:35 2001
From: jmarshal at mathworks.com (Joshua Marshall)
Date: 16 Mar 2001 19:44:35 GMT
Subject: list.sort(cmpfunc) question
References: <m1ae6lg5qt.fsf@pclabs188.labs.trema.com> <98t9dq$qvu$1@panix6.panix.com> <98tf9d$g3v$1@news.mathworks.com> <3AB249FB.131080CD@san.rr.com> <98thua$i40$1@news.mathworks.com> <Wnss6.186$sk3.53027@newsb.telia.net> <3AB25C59.284E58E3@san.rr.com>
Message-ID: <98tqf3$o4r$1@news.mathworks.com>

Darren New <dnew at san.rr.com> wrote:
> Fredrik Lundh wrote:
>> in other words, "recommended but not required"

> Even more strongly than "recommended", but rather "you better have a darn
> good reason and know every result of your change before you consider doing
> this." :-) It's like "Unix root accounts SHOULD have a hard-to-guess
> password."  It's possible to run UNIX with no root password, but you ought
> to have a good reason.

> And yes, there's an RFC defining what "SHOULD" and "MUST" and such mean. :-)

Of course, documentation is a very different thing than an RFC.


From jra at dorothy.msas.net  Thu Mar 29 17:43:01 2001
From: jra at dorothy.msas.net (Jay R. Ashworth)
Date: Thu, 29 Mar 2001 22:43:01 GMT
Subject: Linux GPS program
References: <3AC01637.4EF3A6F2@crynwr.com>    <slrn9c17k4.cfc.jra@dorothy.msas.net> <3AC0D915.83C2C0C3@crynwr.com>    <slrn9c1p9t.d3g.jra@dorothy.msas.net> <99suho$m7a$2@newsserver.rrzn.uni-hannover.de>
Message-ID: <slrn9c7ejl.sv.jra@dorothy.msas.net>

This one time, in band camp,
  Bernhard Reiter <bernhard at abnoba.intevation.de> wrote:
> In article <slrn9c1p9t.d3g.jra at dorothy.msas.net>,
> 	jra at dorothy.msas.net (Jay R. Ashworth) writes:
> > This one time, in band camp,
> >   Russell Nelson <nelson at crynwr.com> wrote:
> >> "Jay R. Ashworth" wrote:
> 
> >> > Any plans to do vector map data as well?  
> >> 
> >> I'm pretty ignorant here.  What sources of vector information are there,
> >> how big is the dataset, and how do I render it so I can use lat/lon or
> >> easting/northing to get my location?
> > 
> > There are *stunning* amounts of vector data, all in different formats. 
> 
> I agree, though most proprietory.

No, actually, I'd bet some cash that the largest collection by size is
the combined efforts of the US Govenment.  That's (almost) all free.

> > I'm trying to (get time cleared up to) sort some of that all out even
> > as I type.
> 
> The shapefile format is used for a couple of things.
> Check www.freegis.org for several libraries dealing with it
> (and other gps packages, btw).
> There is a python wrapper for the shapereading library.

Yeah, I saw that.

Cheers,
-- jra
-- 
Jay R. Ashworth                                                jra at baylink.com
Member of the Technical Staff     Baylink
The Suncoast Freenet         The Things I Think
Tampa Bay, Florida        http://baylink.pitas.com             +1 727 804 5015


From emile at fenx.com  Sat Mar 31 21:13:08 2001
From: emile at fenx.com (Emile van Sebille)
Date: Sat, 31 Mar 2001 18:13:08 -0800
Subject: Conditional operator in Python?
References: <3AC68CB8.22AAF7B8@alcyone.com>
Message-ID: <9a631k$3s8qb$1@ID-11957.news.dfncis.de>

Yes, it comes up.  Generally, except for short-circuiting, the solutions
follow:

>>> 1 and 2 or 3
2
>>> 0 and 2 or 3
3

HTH,

--

Emile van Sebille
emile at fenx.com

---------
"Erik Max Francis" <max at alcyone.com> wrote in message
news:3AC68CB8.22AAF7B8 at alcyone.com...
> Okay, okay, but I still have to ask.
>
> Why isn't there a conditional operator (a ? b : c) in Python?  2.0 now
> has augmented assignments, and the FAQ even acknowledges that the
> conditional operator is sometimes quite convenient and gives some (very
> awkward) Python alternatives.
>
> I've been reading comp.lang.python for a while (since just before 2.0
> shipped) and I can't really remember much discussion about it.  Is it a
> perennial request and I'm just missing it?
>
> --
>  Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
>  __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
> /  \ I sidestep the either
> \__/ or choices of logic and choose both. | Ken Feit
>     Product's Quake III Arena Tips / http://www.bosskey.net/
>  Tips and tricks from the absolute beginner to the Arena Master.




From jmarshal at mathworks.com  Mon Mar 19 15:19:48 2001
From: jmarshal at mathworks.com (Joshua Marshall)
Date: 19 Mar 2001 20:19:48 GMT
Subject: Why "from __future__" stinks; a counter-offer
References: <mailman.984949523.1896.python-list@python.org> <4Aft6.2171$pP2.372173@paloalto-snr1.gtei.net> <mailman.985029625.6313.python-list@python.org>
Message-ID: <995pl4$jta$1@news.mathworks.com>

Jeremy Hylton <jeremy at alum.mit.edu> wrote:
>>>>>> "VC" == Vadim Chugunov <chega_ at yahoo.com> writes:

>   >> might be OK -- alas, without a PEP, it's a non-starter).

>   VC> But, surely, the PEP can be updated!  This "from __future__"
>   VC> syntax is ugly, ugly, ugly!!!

> Vadim,

> If someone else wants to write a PEP and implement it this week, it
> would be considered.  Tim wrote the PEP.  I implemented it.  We're
> both going to be spending most of our waking hours this week getting
> ready for the beta release on Friday.  We have no time and less
> interest in implementing something other than the __future__
> mechanism.  We think it is a good solution to the problem it addresses.

> But neither of us is the BDFL.  If someone writes a PEP and implements
> it in the next day or two, he or she may be able to convince Guido
> that it is better than __future__.  In the absence of someone writing
> a new PEP, nothing is going to change.

> Jeremy

As responses to suggestions, there's a lot of "write up a PEP or it
won't get implemented" going around.  I can understand this, but it's
also important that good ideas don't go to waste just because the
proposer _didn't_ write a PEP.

I guess I just hope that the PEP-mechanism doesn't make it so that
good proposals get overlooked in favor of proposals with enthusiastic
proponents.


From parker at gol.com  Thu Mar  1 17:46:03 2001
From: parker at gol.com (Ian Parker)
Date: Thu, 01 Mar 2001 22:46:03 GMT
Subject: docs for mysqldb
References: <3A9D63BC.C3EB9246@iems.nwu.edu>
Message-ID: <gcxxuDAA4sn6Ewsc@gol.com>

In article <3A9D63BC.C3EB9246 at iems.nwu.edu>, Leonardo B Lopes
<leo at iems.nwu.edu> writes
>Sorry for the crossposting, but this actually interests both lists...
>
>Does anyone use the python module mysqldb? Does anyone have any decent
>docs for it? In particular, how do you do cursor.execute() with
>different parameters?

The MySQLdb-0.2.2.tar.gz kit that I downloaded includes both
documentation and examples.  IIRC correctly, I got it from
http://home.t-online.de/home/err666/

Checking again, it looks like there are later version available.

Regards

Ian
-- 
Ian Parker


From xyzmats at laplaza.org  Sun Mar  4 08:18:03 2001
From: xyzmats at laplaza.org (Mats Wichmann)
Date: Sun, 04 Mar 2001 13:18:03 GMT
Subject: jython question: accessing protected superclass methods
References: <jl66hqvzol.fsf@endor.bbb.caltech.edu>
Message-ID: <3aa23eb1.166345652@news.laplaza.org>

On 03 Mar 2001 22:03:06 -0800, Michael Vanier
<mvanier at endor.bbb.caltech.edu> wrote:

>So, I've been playing with jython for a few hours, and I'm *really*
>impressed.  This is a kick-ass system for doing portable graphics
>programming.
>
>However, the online docs are a bit sparse

Indeed.  Feel free to contribute to the docs, by the way, once you get
this worked out.  Jython is just like Python...except when it isn't,
and things work just like you'd expect, except when they can't because
of the way the JVM works... and those conditions are not always easy
to find out about. I'm sure any contributions would be welcome...

>Also, I was surprised to find that MyJPanel, just by subclassing JPanel, gets
>its own public version of the protected JPanel methods!  So if I don't
>override the method, I can call it directly.  I assume this is intentional,
>but maybe the jython developers can comment.

You'll get a quicker response, in general, on the jython-users (or for
appropriate topics, jython-developers) mailing list, if the question
is Jython-specific.  The core developers are very helpful, but don't
seem to be very active on this newsgroup.  You can signup for the
lists through the Jython project site, jython.sourceforge.net.

Mats
(a lurker on the jython lists, not a developer)


From bedge at troikanetworks.com  Thu Mar 15 11:52:58 2001
From: bedge at troikanetworks.com (Bruce Edge)
Date: Thu, 15 Mar 2001 08:52:58 -0800
Subject: building python for multiple targets
References: <3AB0F009.E27D9AD7@troikanetworks.com>
Message-ID: <3AB0F36A.F6B30A7F@troikanetworks.com>

Just a followup,

Is it possible to build python to be relacatable?
So the same runtime tree can be placed in /usr/local/python, or in
/home/anyone/target/usr/local/python?


Bruce Edge wrote:
> 
> I need to build Python for multiple targets from one host.
> 
> One target is the host, the other is for embedded Linux, with it's own
> compiler, libs, include, etc, all of which are installed in say
> /usr/local/rtlinux/...
> 
> The host build is simple, ./configure, make, make install.
> 
> The target one is ugly, currently I'd doing:
>   ./configure --include=<rtpath> --bindir=/usr/local/rtlinux/bin
> --sbindir=</usr/local/rtlinux/sbin ....<every other damn ./configure
> option there is>
> 
> Is there a better way?


From seth.shikora at intel.com  Fri Mar 23 14:35:14 2001
From: seth.shikora at intel.com (Seth Shikora)
Date: Fri, 23 Mar 2001 11:35:14 -0800
Subject: Good GUI for Python
References: <mailman.985355222.6126.python-list@python.org>
Message-ID: <99g8hj$mk4@news.or.intel.com>

I must agree. wxPython is a nice GUI toolkit. Much easier in my opinion than
working with native widows objects.

I'll also agree that wxPython will not work with PythonWin. I tried and it
crashes way too often. I've been using Boa Constructor for a week now and I
really like it.

Seth.

__________________________


"Rolander, Dan" <Dan.Rolander at marriott.com> wrote in message
news:mailman.985355222.6126.python-list at python.org...
I think wxPython is *very* easy to use, if you can maneuver your way through
the C++ documentation.

Dan
-----Original Message-----
From: winson at mdsserv.mds.com.tw [mailto:winson at mdsserv.mds.com.tw]
Sent: Thursday, March 22, 2001 9:19 PM
To: Python List
Subject: Good GUI for Python


Hi, All

I surveyed 3 GUI framework for Python including wxPython, PythonWin's
included win32ui and Tkinter. My development environment is Mark Hammond's
PythonWin downloaded from ActiveStates.

These're my issues:

First, ActiveStates' Python tool cannot support Tkinter and I confirmed with
their stuff. So Tkinter maybe good, but not compatible with PythonWin.

Second, Mark's masterpiece win32ui and total MFC solution is excellent, but
I'd wanna bind on Windows platform.

Third, wxPython sounds great but not so easy as win32ui.

Is there any suggestions ? Change my develop environment or is there another
good gui for me ??

--
Best Regards,

Winson Chang

3F, 2, 150 Lane, Sec. 5, Hsin I Rd., Taipei, Taiwan, R.O.C.

Tel: +886-2-27225333 ext 376
Fax: +886-2-27222330

Email: winson at mdsserv.mds.com.tw




From thygrrr at gmx.net  Fri Mar  9 12:30:37 2001
From: thygrrr at gmx.net (Moritz Voss)
Date: Fri, 9 Mar 2001 18:30:37 +0100
Subject: I don't get it (Extending C/C++)
Message-ID: <98b3cf$apu$02$1@news.t-online.com>

I'm somewhat stumped by all the weird function names in the Tutorial of the
2.0 Manual...

Is something like this possible:
I call a Python function from my C/C++ program. I pass it two pointers, one
to an array with data, and one to an empty space where it can put its
results....

I don't even remotely see a way how to do this in the Tutorial, it's just
full of (to me) irrelevant stuff about callbacks etc...I feel stupid here,
folks ;-)

Also, is it advisable to invoke the interpreter (do I even do that? what DO
I do when I execute a python function?) several hundred times? (running
about 500 instances of several, sometimes identical python scripts times
would be likely for medium load of my application).

I'm sorry, I just don't /understand/ the Python concept...yet.

It looks all so cool, but I don't have a clue about how to use it.
(I plan to use it as a scripting language to customize application component
behaviour)

--
--
Moritz "Thygrrr" Voss
http://www.thygrrr.com








From Patrick.Bothe at ePost.de  Sun Mar 25 10:17:01 2001
From: Patrick.Bothe at ePost.de (Patrick Bothe)
Date: 25 Mar 2001 17:17:01 +0200
Subject: Commercial Products in Python?
References: <3abb86ee.142862695@localhost> <3ABBD84B.9020404@ActiveState.com>
Message-ID: <87bsqp515e.fsf@gabriel.heim6.tu-clausthal.de>

Hi,


I think some pieces of code of "Severance - Blade of Darkness" 
(PC-GAME) are coded in Python 1.5.2.

bye,
    Patrick


-- 
Der Kopf ist rund,                                   (_ )
damit das Denken die Richtung ?ndern kann.        \\\@@ ) ^
                                                    \/, \(
[Francis Picabia]                                   cXc_/_)


From lsommerer at sewardweb.com  Tue Mar 13 15:49:44 2001
From: lsommerer at sewardweb.com (Lloyd Sommerer)
Date: Tue, 13 Mar 2001 14:49:44 -0600
Subject: Newbie: Is there a more pythonic approach?
Message-ID: <3AAE87E8.FE11B4F9@sewardweb.com>

I'm trying to reduce 404 errors (at least one that visitors see) on my
website by correcting problems with case and extensions on the fly.
This is one of my first python programs, and I was wondering if it:

a) made use of python language features in a way that seems natural to
"real" python programmers.
b) missed any language construction that makes the program "better"
(faster, easier to read...)

Thanks, the program follows below.

Lloyd Sommerer
=========================================
error_page = '/404.html'
base_url = 'http://churchsite.gatheringspot.com'
base_directory = '/home/gatheri/public_html/churchsite/'
extension_list = ['.html','.shtml','.htm','.txt','.php','.asp']

import string
import os

lowercase_url = string.lower(os.environ['REDIRECT_URL'])
url_path,url_extension = os.path.splitext(lowercase_url)
new_url = base_url+error_page
#
# Check to see if simply converting the URL to lowercase corrects the
problem.
# If it doesn't, check to see if the extension is incorrect and fix it
if it is.
#
if os.path.exists(base_directory+lowercase_url):
    new_url = base_url+lowercase_url
else:
    for extension in extension_list:
        if os.path.exists(base_directory+url_path+extension):
            new_url = base_url+url_path+extension
#
# Send HTTP header information to the browser so that it can request the
"correct" document.
#
print "Status: 301 Permanently moved"
print "Location: " + new_url
print



From bhk at dsl.co.uk  Wed Mar 14 11:30:17 2001
From: bhk at dsl.co.uk (Brian {Hamilton Kelly})
Date: Wed, 14 Mar 2001 16:30:17 GMT
Subject: Einstein's Riddle
References: <oNPq6.396644$ge4.138479566@news2.rdc2.tx.home.com> <UdSq6.2184$54.2166@www.newsranger.com> <7z9r6.2576$54.2837@www.newsranger.com>
Message-ID: <984587417snz@dsl.co.uk>

In article <7z9r6.2576$54.2837 at www.newsranger.com>
           gregj at pobox.com "Gregory Jorgense" writes:

> Clearly my statement that the puzzle can't be solved puts me in a minority. If
> you assume that SOMEONE keeps fish the puzzle can be solved with a little bit of> logic. Perhaps the intent of the riddle is to show how few people can devise a
> useful and logical methodology.

There is (or was) a British monthly (possibly bi-monthly) magazine
published under the title Logic Problems.  This had about fifty of this
genre of puzzle in every issue.  It is bought by a wide range of people,
not many of whom are likely to belong in the upper two-percentile of
intelligence, but all of whom (presumably, otherwise the magazine
wouldn't sell) have the right mindset to solve the problems.

Incidentally, I first saw a puzzle of this nature in McCall's Magazine in
the 1950s, when I was still at primary school.  That particular one was
*entitled* "Who owns the zebra", which was the only mention of that
particular species in the whole exposition of the puzzle.  It IS
"information" that is required for the puzzle to be soluble, however.

Isn't it about time this stupid thread came to a halt?  It was needlessly
cross-posted (perhaps also multiply to other groups than appear in the
headers of this post?) and has excited "interest" beyond all rationality.

IOW, YHBT.

-- 
Brian {Hamilton Kelly}                                          bhk at dsl.co.uk
    "We have gone from a world of concentrated knowledge and wisdom to one of
    distributed ignorance.  And we know and understand less while being incr-
    easingly capable."                          Prof. Peter Cochrane, BT Labs



From jmarshal at mathworks.com  Wed Mar 21 14:46:46 2001
From: jmarshal at mathworks.com (Joshua Marshall)
Date: 21 Mar 2001 19:46:46 GMT
Subject: learning python...
References: <3AB8F4E1.184DD9C@noaa.gov> <Z97u6.16733$Im6.1761313@newsread1.prod.itd.earthlink.net> <3ab90559_6@goliath.newsfeeds.com>
Message-ID: <99b0f6$ns1$1@news.mathworks.com>

Bubba Brains <chocolate at toblerone.yum> wrote:
> Forget Learning Python?  I think it's the best reference/tutorial around.  I
> use python once every couple months, and sometimes forget the simple python
> syntax.  The Learning Python book is great for getting up to speed fast, and
> to look up common stuff.  The other OReilly book (Programming Python) is
> absolutely annoying to use.  It's not even a good reference book (good luck
> finding the info you want quickly).  Don't know about the Prentice book.

I agree about Programming Python.  It _is_ a great book for
programmers to use to learn Python, but it can be tedious to try look
up something specific.


From fredrik at pythonware.com  Fri Mar 23 11:28:12 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 23 Mar 2001 16:28:12 GMT
Subject: Tkinter - Widget Colours
References: <mailman.985345890.22443.python-list@python.org>
Message-ID: <wQKu6.3197$4N4.475642@newsc.telia.net>

Clodagh O'Rourke wrote:
> I was wondering does anyone know how the bag/fag
> colours can be changed after the widget is packed.

as others have pointed out, you can use "config" (aka
"configure") to change options at any time.

a bit more info here:
http://www.pythonware.com/library/tkinter/introduction/widget-configuration.htm

Cheers /F




From andy at reportlab.com  Fri Mar 23 19:44:04 2001
From: andy at reportlab.com (Andy Robinson)
Date: Sat, 24 Mar 2001 00:44:04 +0000
Subject: Commercial Products in Python?
References: <3abb86ee.142862695@localhost> <slrn9bn8tr.avv.mmealman@loki.thebisgroup.com>
Message-ID: <k6rnbtselb11lg47l9p8sefa1ugscrn8ie@4ax.com>

On 23 Mar 2001 11:29:21 -0600, mmealman at loki.thebisgroup.com (Mark)
wrote:

>>3. What is the best way to go about it? Is it better to have customers
>>install Python and give them bytecode files or use one of the programs
>>that creates an executable?
>
>I'd try to compile Python down to an executable. It'd probably make it
>easier to deploy.

ReportLab have just put out two commercial demos for download
built around Gordon McMillan's Windows Installer, with some
mods of our own.   Once you get it set up, it works beautifully.
Have a quick look at
http://www.reportlab.com/pageCatcher/pagecatch.zip

I'm told it works just fine with GUI apps as well.  We will end
up advising the 'power users' to get Python, but the executable
is perfect for many users and for demonstrations.

- Andy Robinson
CEO/Chief Architect, ReportLab Inc.



From greg at cosc.canterbury.ac.nz  Mon Mar  5 23:01:56 2001
From: greg at cosc.canterbury.ac.nz (Greg Ewing)
Date: Tue, 06 Mar 2001 17:01:56 +1300
Subject: I come to praise .join, not to bury it...
References: <mailman.983485923.1136.python-list@python.org> <97p7iq$kk2$1@nntp6.u.washington.edu> <97rka002k9@news1.newsguy.com> <3AA2F4B3.A7496E76@cosc.canterbury.ac.nz> <97vmoh061p@news2.newsguy.com>
Message-ID: <3AA46134.D385208B@cosc.canterbury.ac.nz>

Alex Martelli wrote:
> 
> indeed, the string module supplies ready-made versions
> of such calls

For the time being, yes, but the intention has been
hinted at on python-dev to deprecate and eventually 
remove the string module. That worries me.

> if the string.join function had to typeswitch on the
> joiner-object to distinguish Unicode from single-byte --

I don't get this business about avoiding type switches
by dispatching on the joiner, because you still have to
typeswitch on all the OTHER strings that you're joining.
So you save one (very fast C-level) typeswitch out of 
quite a lot. Nowhere near a big enough gain to justify 
anything, to my mind.

(And if ordinary strings and Unicode strings are ever
unified, this argument will go away completely.)

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


From nanotech at europa.com  Thu Mar 29 13:38:51 2001
From: nanotech at europa.com (nanotech at europa.com)
Date: Thu, 29 Mar 2001 18:38:51 -0000
Subject: Accessing variables in __main__ from modules....
Message-ID: <99vvfr+8hq6@eGroups.com>

All:

When writing programs, I like to add print statements that are only 
displayed when the user provides a command line switch such as '--
debug'.

  Ex:  %myprog spam.txt ham
       Found the word 'ham' in 'spam.txt' 45 times

       %myprog --debug spam.txt ham
       -I-: Reading 'spam.txt'....has 213 lines....
       Found the word 'ham' in 'spam.txt' 45 times

Now, say most of the work is done in a module 'spam.py'. If I wrote 
code in this module like:

  lines=open(filename).readlines()
  if (DEBUG):
    print "-I-: Reading '%s'....has %s lines...."%(filename,len
(lines))

DEBUG would be searched for at the module scope. But I want it to 
look for DEBUG back in __main__, like:

  if (sys.modules["__main__"].DEBUG):
    print "-I-: Reading '%s'....has %s lines...."%len(filename,lines)

Is this the "best" way to access variable in the "__main__" scope? 
How else might I solve this (in the python way)??

Quentin Crain

Note: I might load many modules all wanting to know if DEBUG is 
turned on. I do not want to pass DEBUG to functions or classes and I 
do not want to set a module variable DEBUG to the same value 
(module1.DEBUG=DEBUG, module2.DEBUG=DEBUG, etc. in __main__). Unless 
these are "better" solutions! :)






From dfan at harmonixmusic.com  Mon Mar 26 10:45:24 2001
From: dfan at harmonixmusic.com (Dan Schmidt)
Date: 26 Mar 2001 10:45:24 -0500
Subject: Good Editor? (What's the big deal with syntax colouring?)
References: <mailman.985291829.27633.python-list@python.org> <99dqqo$gpi$1@news.mathworks.com> <99e9sv$usmp$1@ID-11957.news.dfncis.de> <99emg2$qif@gap.cco.caltech.edu> <ha3obts0kp05lonh3o25fsesr38bouo839@4ax.com> <slrn9bo559.48h.grey@teleute.dmiyu.org>
Message-ID: <wkelvk1qln.fsf@turangalila.harmonixmusic.com>

grey at despair.rpglink.com (Steve Lamb) writes:

|     Don't do much coding in the dark, do you?
| 
|     I never understood why people insist on it "looking like paper"
| when there is one major difference.  Paper is reflective, a monitor
| is projective.  I've never been blinded by a 19" piece of paper 3'
| from my face but I sure have been blinded by a 19" monitor 3' from
| my face because of this "looking like paper" bit.  :/

The last time I saw this debate, someone made the observation that,
in general, people who use computers in lit rooms prefer black on
white and people who use computers in dark rooms prefer something
light on black, and both of them have good ergonomic reasons for
their preference.

-- 
http://www.dfan.org


From sholden at holdenweb.com  Fri Mar 30 13:56:10 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Fri, 30 Mar 2001 13:56:10 -0500
Subject: email address verification
References: <99d5qi$95l$1@solaria.cc.gatech.edu> <99fjv8$bhf$1@solaria.cc.gatech.edu> <E006A65FA23EFF2C.2D74AEB35552E0B6.065EC0B29EF187C6@lp.airnews.net> <99fo48$d9h$1@solaria.cc.gatech.edu> <6491BB8AD25B69BC.0BFC7C13FCE3713E.F0B7BA5A511B6DBC@lp.airnews.net>
Message-ID: <SI4x6.37870$m04.1410919@e420r-atl1.usenetserver.com>

"Cameron Laird" <claird at starbase.neosoft.com> wrote in message
news:6491BB8AD25B69BC.0BFC7C13FCE3713E.F0B7BA5A511B6DBC at lp.airnews.net...
> In article <99fo48$d9h$1 at solaria.cc.gatech.edu>,
> Holland King  <insanc at cc.gatech.edu> wrote:
> >Cameron Laird <claird at starbase.neosoft.com> wrote:
> >: I don't understand your description.  In general,
> >
> >ok, we run a server that holds a lot of accounts, once an account expires
> >then we run a script that parses the .forward file found in the home
> >dir. the script then takes the addresses found and makes them into
aliases
> >so that hopefully the user will not lose any mail. we need to make sure
> >that the aliases are real addresses so that at worst mail is not bounced
> >and at best it doesn't go to a mailing list (i know i am being an
optimist)
> It's a good question to ask--but one which
> turns out to be quixotic.  Not only can one
> make no guarantees by a "static" examination
> of a purported e-mail address, my own exper-
> ience is that there are few heuristics which
> are worth implementing.  So much can and
> does happen at the recipient end of an e-mail
> transaction that it's nearly fruitless to
> try to automate.
> >
> >: comp.mail.misc might be a more productive place to
> >: ask these questions than c.l.p.
> >
> >ok thank you i will try there to, the reason i asked here is that i tried
> >comp.linux.misc and they didn't help and i am writing it in python (i
> >think my orignal post mention that i had found the smtp library and i
> >was wondering if there was something similar that would work)
> I understand.  It's a reasonable presumption.

You also need to remember that some domains (such as bellatlantic.net) won't
even raise a bounce message on an expired account, so it might appear mail
is getting through even when nobody will ever see it.

regards
 Steve





From sholden at holdenweb.com  Thu Mar 22 22:21:44 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 22 Mar 2001 22:21:44 -0500
Subject: Good Editor?
References: <3ab9a32b$1_1@news2.one.net> <Xns906C5F868E6B7PaCmAnRDLM@161.48.76.20> <GALn19.C1z@leeds.ac.uk>
Message-ID: <Vmzu6.8979$G72.208277@e420r-atl1.usenetserver.com>

Nope. That's two languages. No emacser worth their salt would leave elisp
out of their vocabulary.

(((elisp) anyone) for?)-ly y'rs  - steve

"Nick Efford" <nde at comp.leeds.ac.uk> wrote in message
news:GALn19.C1z at leeds.ac.uk...
> On 22 Mar 2001 08:22:26 GMT, Fred Pacquier <fredp at mygale.org.nospam>
wrote:
>
> > >Can someone recomend a good editor to write python in?  Im just getting
> > >started learning python and dont know which ones support python.
> >
> > You don't mention the platform. Under Unix/Linux there are a number of
> > options (including Vim, Emacs, etc.). Under Windows, I found WScite
> > (http://www.scintilla.org) the easiest to get started with...
>
> vim and emacs run on Windows, too!  So you can use one editor across
> platforms as well as one programming language :)
>
>
> N.
>




From cfelling at iae.nl  Tue Mar 27 09:10:26 2001
From: cfelling at iae.nl (Carel Fellinger)
Date: 27 Mar 2001 16:10:26 +0200
Subject: Is it possible to sort a dictionary?
References: <GEtv6.59583$Lm2.6378825@news0.telusplanet.net> <3ABE771B.7B9E1631@student.liu.se> <99o33k$aq4$0@216.39.170.247> <99o51301qgb@news1.newsguy.com> <mailman.985660594.13367.python-list@python.org>
Message-ID: <99q70i$1le$1@animus.fel.iae.nl>

Carlos Alberto Reis Ribeiro <cribeiro at mail.inet.com.br> wrote:
> At 03:09 27/03/01 +0200, you wrote:
...
>>similar and was surprised when I found the dict.items function to be
>>soo slow that going through the whole dict indexing each entry
>>one by one was faster.

> I was thinking about something similar today. It would be useful to have a 
> "xitems()" method, behaving exactly like items(), but *without* the need to 
> build the tuple in advance. I see two advantages:

I don't know.  One can only sort after the whole list exists, so a
non-sorting xitems() won't suffice.  In this case we wanted it to be
sorted on keys, in other cases one would like it sorted on values.
Are you proposing all kind of flavours for the xitem() method?

I just wonder why the items() method is slooow; values() and keys()
are both much faster, going over the whole dict indexing from keys()
is even faster.  Is tuple making/unpacking so expensive?  Or is a dict
optimized for key/value access on the expense of items()?  I know I
should dive into the python-code, but maybe someone can save me a few
minutes:)


-- 
groetjes, carel


From claird at starbase.neosoft.com  Tue Mar 27 10:45:34 2001
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 27 Mar 2001 09:45:34 -0600
Subject: dynamic web pages?
References: <99q701$jlr10@news-1.bank.dresdner.net>
Message-ID: <D65F64C08C1C1C31.AB463445171734BC.B3DA53B3C6E1C0F8@lp.airnews.net>

In article <99q701$jlr10 at news-1.bank.dresdner.net>,
Dr. Peter Dintelmann <Peter.Dintelmann at dresdner-bank.com> wrote:
>Hi,
>
>is there any module to create dynamic web pages
>using Python?
			.
			.
			.
There is an abundance of such schemes:  <URL:
http://starbase.neosoft.com/~claird/comp.lang.python/web_python.html >
-- 

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


From oddxcb at e.elim.com  Sat Mar 24 05:04:16 2001
From: oddxcb at e.elim.com (oddxcb at e.elim.com)
Date: Sat, 24 Mar 2001 10:04:16 GMT
Subject: Evidence on................YOUR COMPUTER................Get rid.............                                  .  1926
Message-ID: <Ai_u6.102154$KL5.101256@NewsReader>

I was shocked at what I found.
Files I shredded months earlier were clearly visible using this software.

It's a fact
Did you know.........
For example that every click you make on Windows 98 Start Menu 
is logged and stored for ever on a hidden encrypted database within your own computer?
your PC is keeping frightening records of both your online and off-line activity.

Evidence Eliminator.
http://www.evidence-eliminator.com/go.shtml?A653615
nyzktchuvgvmffcoudmgizrcnr



From mcherm_python at yahoo.com  Sun Mar 18 12:56:45 2001
From: mcherm_python at yahoo.com (Michael Chermside)
Date: Sun, 18 Mar 2001 11:56:45 -0600
Subject: Who's minister of propaganda this week?
References: <B98041329020E007.7DA84CAD9D0562AB.817BBC572F480869@lp.airnews.net> <98nu1m$12e@dispatch.concentric.net> <660ACCA1E4816F7E.C359CDF4168EF3FB.3B6B05538D619770@lp.airnews.net> <3AAFA13D.76E32C3F@divalsim.it> <98oa4o01i4v@news2.newsguy.com> <3AB0176D.8070903@yahoo.com> <3AB0C70A.C6A55DF5@ActiveState.com> <3AB40BE3.3080001@yahoo.com> <3AB417FD.1905AC3D@ActiveState.com>
Message-ID: <3AB4F6DD.10901@yahoo.com>

Paul Prescod wrote:

>                     ...[snip]...
> 
>> (But see Alex's response, where he argues that perhaps it DOES hurt,
>> for other reasons.)
> 
> 
> I think it takes a lot of work with both kinds of languages to recognize
> that you lose something significant in exchange for the earlier error
> messages. And I think that slowly over the next few years, programmers
> are going to to come to that realization.

Well, most of my experience has been in staticly typed languages. I
suspect that even in dynamically typed languages I tend to program
in a "static typed fashion". Do you have any suggestions of examples
of good code to read (or good books to read!) which would help me to
realize what kinds of advantages I can get from dynamic typing?

> It is highly akin to the GC situation. A few years back, garbage
> collection of any sort was considered unaffordable for real apps (even
> in the face of evidence to the contrary) merely because most programmers
> did not have real-world experience with it. Now, nobody designs a new
> language without GC. Of course it wasn't all just education. Moore's law
> helped too. The situation is the same with dynamic typing. Moore's law
> and education will bring programmers over.

Hmm... the first time I ever heard of GC I immediately said to myself,
"Aha! That's the way to go!". But I'm still working at figuring out the
advantages of dynamic typing.

-- Michael Chermside

> 


_________________________________________________________
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com




From stephen_purcell at yahoo.com  Tue Mar 27 12:43:43 2001
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Tue, 27 Mar 2001 19:43:43 +0200
Subject: large arbitrary binary data in python
In-Reply-To: <20010327192520.F1260@freedom.puma-ag.com>; from stephen_purcell@yahoo.com on Tue, Mar 27, 2001 at 07:25:20PM +0200
References: <slrn9c1hf7.m5q.wcuddy@lserv.ja10629.home> <20010327192520.F1260@freedom.puma-ag.com>
Message-ID: <20010327194343.A7081@freedom.puma-ag.com>

Steve Purcell wrote:
> Yes, see
>  --- snip ---
> Yes, but don't pipe it to md5sum if you can avoid it. Instead, use the
> 'md5' standard module:
>  --- snip ---
> Yep, see os.popen() and the popen2 module:
>  --- snip ---
> Yes. The limit is the amount of available virtual memory.


Python is a 'yes' language.

(Unlike Java, which is a YeSDK language)

-Steve

-- 
Steve Purcell, Pythangelist
Get testing at http://pyunit.sourceforge.net/
Any opinions expressed herein are my own and not necessarily those of Yahoo



From lac at cd.chalmers.se  Thu Mar  8 11:30:41 2001
From: lac at cd.chalmers.se (Laura Creighton)
Date: Thu, 8 Mar 2001 17:30:41 +0100 (MET)
Subject: Can Python 1.5.2 code run on Python 2.x?
Message-ID: <200103081630.RAA18168@boris.cd.chalmers.se>

Guess what I am finding out today.

Big change ... string conversion of long int no longer tacks an L on
the end.  Since I just ripped off the last digit wholesale all over
this code, I have many lines to change.

Laura Creighton



From costas at springmail.com  Mon Mar 12 15:38:42 2001
From: costas at springmail.com (costas at springmail.com)
Date: Mon, 12 Mar 2001 20:38:42 GMT
Subject: Crashed PythonWin. How can I fix it?
References: <3aa04cf5.34464140@News.CIS.DFN.DE> <3AAACCDE.80703@ActiveState.com>
Message-ID: <3aad3370.2161792618@News.CIS.DFN.DE>

On Sun, 11 Mar 2001 00:58:04 GMT, Mark Hammond <MarkH at ActiveState.com>
wrote:

>Costas Menico wrote:
>
>>     import glob
>>   File "glob.py", line 11, in ?
>>     glob2.what2()
>>   File "glob2.py", line 3, in what2
>>     print x
>> exceptions.NameError: There is no variable named 'x'
>
>It looks alot like a "glob.py" is being found that isnt the glob.py in 
>the standard library.

Mark,

I tried searching all of Python and could not find anywhere glob2 or
what2. I had to erase everything and reinstall Python not just
PythonWin. Hope I will never see this again :)

Costas



From aleaxit at yahoo.com  Tue Mar 13 08:45:16 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Tue, 13 Mar 2001 14:45:16 +0100
Subject: module importing and variable syntax
References: <3AADF3DC.1030003@dynworks.com>
Message-ID: <98l89901v7f@news1.newsguy.com>

"Jeff Davis" <jdavis at dynworks.com> wrote in message
news:3AADF3DC.1030003 at dynworks.com...
> I was previously a perl user, but now I am taking an interest in python.
> I am used to common perl methods such as using "variable variables"
> where the accessed variable depends on the situation (perl: "print
> ${$x};").

Isn't this style of ``soft'' referencing deprecated as dangerous
and unadvisable in Perl itself?

> Module importing doesn't take a string argument, but just a "bare word"
> (in perl terms). This means I have to know the name of the file in
> advance, and it needs to be in the right directory, or I need to use
eval().

Or, as other have suggested, __import__.  Also note that if the
module HAS already been imported, the module-object can also be
found at sys.modules[name] where name is a string that names
the module.


> What I am trying to do is have an arbitrary string with an arbitrary
> type (type with respect to my program, not python) access the function
> associated with the string in the module associated with the type.
>
> For example: someone writes an extension to my program creating a new
> object of name "O". Then some other part of the program (i.e. another
> extended module) tries to square an instance of that object, like:
> o = O()
> o.square()
>
> However, my program is completely unaware of either "O" or "square"
> until someone writes the module(s) and registers them with the program
> (not writing to the main program's source).
>
> It seems like I could get away with it if I used a lot of eval()
> statements, but talk about unreadable code (especially since you have to
> concatenate the variables in python to form a string)!

The "other extended module" can perfectly well import module O,
access its O attribute (a class, calling which creates an instance
thereof), and access (and call) the instance attribute called
'square', all on its own, without bothering "your program" -- if
I understand what you're saying.

If the call must happen in your main program, it need not be the
case that your main code has to be aware of anything -- it may
be passed the appropriate *objects*, rather than strings it's
supposed to use to fetch the objects.  Module-object, instance
object, bound-method object o.square -- whatever.  As I don't
exactly understand what you're doing, I can't tell whether this
is the appropriate route, but it often is -- people coming from
other programming languages often think in terms of passing
strings around, where in Python they might be better off passing
appropriate objects instead.


> I think I am having problems understanding the way python works in this
> regrard, which is why this question was probably way more complex than
> it needed to be.

Everything's object-centered; strings (names, variables, etc) are
basically 'tags' for the objects -- names (keys) whose corresponding
value is a reference to the object in some appropriate dictionary
('hash' for a Perlist).  Handy for human consumption, but programs
may have an easier time handly the objects directly.


> I would really appreciate some clarification. I like a lot about the way
> python is made and I think it has a lot of potential. Right now the
> concatenation of strings can be annoying compared to just (perl)
> '$string = "a $anotherstring be";', but for a lot of programs it is well
> worth the tradeoff.

Concatenation would be:
Perl:
    $string = "a " . $anotherstring . " be";
Python:
    astring = "a " + anotherstring + " be"
essentially equivalent.

What you're doing is variable-interpolation:

Perl:
    $string = "a $anotherstring be";
Python:
    astring = "a %(anotherstring)s be" % vars()

but note that in Python you can contextually pass
another dictionary to the %-operator (which does
the formatting, aka interpolation), AND use a
different format-specifier than %...s (which just
lets the object supply a string in whatever way
it wishes) much like in C's printf, if you wish.

Another equivalent approach:

    astring = "a %s be" % anotherstring

some find this preferable to having the names
mixed in with the format-string (mostly, ones
coming from C, I think:-).


Alex





From tim_one at email.msn.com  Tue Mar  6 21:22:21 2001
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 6 Mar 2001 18:22:21 -0800
Subject: IEEE 754 conformance in Python
In-Reply-To: <AW5p6.142249$My5.5890439@news.hananet.net>
Message-ID: <LNBBLJKPBEHFEDALKOLCAEEEJEAA.tim_one@email.msn.com>

[CY Jair]
> Python's floating-point numbers representations are done by C
> library, with 17 significant decimal digits preserved by the "%.17g"
> format in floatobject.c file.

All of Python's floating-point behavior is inherited from the platform C
compiler and libraries.

> It is advertised to be conformant to IEEE 754 double precision standard.

Unclear what "it" means here.  CPython does not promise anything wrt 754.

> Java is advertised as fully conformant to IEEE 754 by Sun. However, when I
> compare the two, I'm getting a little bit doubtful.

Note that string<->double behavior isn't fully specified by the 754 standard
(I/O conversions are unique among the operations covered by 754 in this way:
a conforming implementation need not round them properly).

> ...
> One more thing that makes me wonder is the way Python does rounding on
> floating-point numbers. In the IEEE standard, when rounding a "halfway"
> result to the nearest floating-point number, it picks the one
> that is even.

By default, yes, although it also depends on the rounding mode in effect.

> But Python's way of "%.17g" doesn't seem to work that way.

CPython's %.17g behavior is inherited from the platform C sprintf %.17g
behavior.  It can and does vary across platforms; e.g., "add a half and
chop" is normal under Windows:

C:\Code\python\dist\src\PCbuild>python
Python 2.1b1 (#11, Mar  3 2001, 05:56:55) [MSC 32 bit (Intel)] on win32
Type "copyright", "credits" or "license" for more information.
>>> "%.0g" % 2.5
'3'
>>>

Most (but not all) flavors of Unix produce '2' there.  Same thing if you
write the test in C.





From bedge at troikanetworks.com  Fri Mar  2 16:10:28 2001
From: bedge at troikanetworks.com (Bruce Edge)
Date: Fri, 02 Mar 2001 13:10:28 -0800
Subject: create new statement?
References: <3A9FE7C2.1F5525D2@troikanetworks.com> <mailman.983560384.31798.python-list@python.org>
Message-ID: <3AA00C44.FB588EAE@troikanetworks.com>

Steve Purcell wrote:
> 
> Bruce Edge wrote:
> > I want to add my own command to the interp, but not have to use object
> > calling syntax
> > eg:
> >     create obj opt
> > not
> >     create(obj, opt)
> >
> > How do I define new statements, or, a way of calling functions using the
> > former syntax?
> 
> You can get the effect you want by subclassing 'Cmd' in the 'cmd'
> standard library module and using your subclass in place of the usual
> interactive command interpreter.

I played around with that for a bit. I think it is useable. 
I would need to register an alternate completer for readline, as the
default
picks up the python global namespace.

Thanks.

> 
> -Steve
> 
> --
> Steve Purcell, Pythangelist
> Get testing at http://pyunit.sourceforge.net/
> Get servlets at http://pyserv.sourceforge.net/
> "Even snakes are afraid of snakes." -- Steven Wright


From ransen_spam_me_not at nemo.it  Sat Mar  3 05:02:28 2001
From: ransen_spam_me_not at nemo.it (Owen F. Ransen)
Date: Sat, 03 Mar 2001 10:02:28 GMT
Subject: PyRun_SimpleFile - impossible to call it under Windows?
References: <3aa07704.2672581@news.newsguy.com> <3AA08A11.95BC7745@bigfoot.de>
Message-ID: <3aa0bd3f.708696@news.newsguy.com>

On Sat, 03 Mar 2001 07:07:13 +0100, Gerhard H?ring
<gerhard.nospam at bigfoot.de> wrote:


>> I'd like to call PyRun_SimpleFile in the debug version
>> of my VC6 program, is it impossible currently with Python 2?
>I suggest that you use the debug version of the Python libraries,
>available at
>http://www.activestate.com/Products/ActivePython/Download.html
>Gerhard

Hi Gerhard,

I am already linking to Python20_d.lib, is that not the correct
version?


--
Owen F. Ransen
http://www.ransen.com/
Home of Gliftic & Repligator Image Generators


From jwbnews at scandaroon.com  Thu Mar 15 18:43:24 2001
From: jwbnews at scandaroon.com (John W. Baxter)
Date: Thu, 15 Mar 2001 15:43:24 -0800
Subject: Off-python-topic:  Cron jobs on linux...help
References: <mailman.984684081.4438.python-list@python.org>
Message-ID: <jwbnews-27AEA2.15432315032001@corp.supernews.com>

In article <mailman.984684081.4438.python-list at python.org>,
 "Chris Richard Adams" <chrisa at ASPATECH.COM.BR> wrote:

> I'd like to ask two things about the cron tool that comes with Linux -
> have RedHat 6.2.
>  
> 1.) I have a script that runs fine when I am logged in a the user who
> owns it.  I have env variables that are needed and these are loaded in
> the .bash profile as that user logs in.  As that user, I set a cron job
> putting the pathname of the script.  It runs, but fails because it
> appears the  environment vars are not set.  I can work around it by
> putting all the vars in the script of course, but what might be the
> problem?

Cron deliberately sets up a VERY limited set of environment variables 
when it starts a process.  [For example, the PATH is likely not to 
contain what one needs, so full path names are generally given.]

See
   man env

>  
> 2.) I want the script to run each 20 minutes.  With only the 5 time
> parameters to set, I can't do this?

This snippet from /etc/crontab runs the elided stuff every 5 minutes 
much of the day and every 15 minutes overnight.

0-55/5 7-23 * * * root nice /...
5-50/15 0-6 * * * root nice /...

You don't have to abbreviate that much...the minute can be shown as
0,5,10,15, and so on.

See
   man -S 5 crontab

(-s rather than -S on many Unix systems)

  --John

-- 
John W. Baxter   Port Ludlow, WA USA  jwbnews at scandaroon.com


From ego at macondo.visconti.home  Thu Mar 22 09:00:29 2001
From: ego at macondo.visconti.home (Gabriele Favalessa)
Date: 22 Mar 2001 14:00:29 GMT
Subject: Text printing under Windows
Message-ID: <slrn9bk130.10l.ego@macondo.visconti.home>

Hi,

how do I print using a dot matrix printer in text mode under Windows
98?  I have both a local and a remote printer.

I use python 2.0 and wxPython 2.2.5.  The latter seems to support
printing only in bitmap mode.

My knowledge of win32 APIs is poor, I'm looking for a simple
open/write/close API.  Something similar to:

   printer = os.popen ("lpr -Pprinter1")
   printer.write ("Hello world!\n")
   printer.close ()

that works fine under unix.

Thanks in advance.

   Gabriele


From pj at sgi.com  Sun Mar 18 03:26:45 2001
From: pj at sgi.com (Paul Jackson)
Date: 18 Mar 2001 08:26:45 GMT
Subject: How do I avoid the Tkinter 30 msec polling loop.
References: <98s781$7a8o$1@fido.engr.sgi.com> <3ab3d191.6695740@24.0.228.33>
Message-ID: <991rg5$14pol$1@fido.engr.sgi.com>

Thanks for the short answer.  I don't know tcl
(well, did a little once, but developed an
allergy to it).  Whether your long answer
provokes me into useful investigations remains
to be seen.  Guess it depends on how much I
am trying to avoid 'real' work sometime.

Anyhow - thanks for the prompt and most helpful
reply.
-- 
                          I won't rest till it's the best ...
                          Manager, Linux System Software
                          Paul Jackson <pj at sgi.com> 1.650.933.1373


From abg at saturnsys.com  Wed Mar 28 13:10:13 2001
From: abg at saturnsys.com (Adam Gurno)
Date: Wed, 28 Mar 2001 18:10:13 GMT
Subject: Python and Dock Apps...
Message-ID: <1103_985803013@stella>

Hi All,
	While browsing for new Dock Apps to use with Blackbox, I came across this:
http://www.kryogenix.org/tech/wm8ball.py.txt

	However, I can't find *any* other information on wmdock.py.

	Anyone know of any locations or can find anything else?

Adam
(mapson means no spam, if you knowhaddimean)



From jmarshal at mathworks.com  Mon Mar 19 22:12:20 2001
From: jmarshal at mathworks.com (Joshua Marshall)
Date: 20 Mar 2001 03:12:20 GMT
Subject: list.sort(cmpfunc) question
References: <m1ae6lg5qt.fsf@pclabs188.labs.trema.com> <98t9dq$qvu$1@panix6.panix.com> <98tf9d$g3v$1@news.mathworks.com> <Tjwt6.115997$ML1.5401185@e420r-atl2.usenetserver.com>
Message-ID: <996hqk$4p2$1@news.mathworks.com>

Steve Holden <sholden at holdenweb.com> wrote:

> "Joshua Marshall" <jmarshal at mathworks.com> wrote in message
> news:98tf9d$g3v$1 at news.mathworks.com...
> [ ... ]
>> Seems the word "should" should never appear in documentation.

> ... and '"should" should' should _definitely_ never appear in documentation!

I was hoping someone would bite :)


From jfontain at free.fr  Sun Mar 11 14:39:43 2001
From: jfontain at free.fr (Jean-Luc Fontaine)
Date: Sun, 11 Mar 2001 19:39:43 GMT
Subject: ANNOUNCE: tclpython-2.0
Message-ID: <3AABD472.EE38F075@free.fr>

### CHANGES ###

--- version 2.0 ---
allow multiple Python interpreters
no longer requires the Python print command to pass data to the Tcl
  interpreter
embedded Python interpreters can now send data to standard output or
  error (useful for debuging, using the print function, for example)
successfully tested against the Python thread library
added tclpython2 rpm for Redhat python2 compatibility
includes 1 line addition patch to the Python sources

### README ###

tclpython version 2.0: a Python package for Tcl

This package allows the execution of Python code from a Tcl
interpreter, as in:

    package require tclpython
    # or: package require tclpython2
    set interpreter [python::interp new]
    $interpreter eval {print("Hello World")}
    python::interp delete $interpreter

You can actually create several Python interpreters this way, if the
tclpython package was linked against a Python library compiled with
threads support, otherwise only 1 Python interpreter can exist at a
time.

Use the tclpython2 if you are using Python 2.0 or above.

This library is free software; you can redistribute it and/or modify
it under the terms of the GNU Library General Public License as
published by the Free Software Foundation; either version 2 of the
License, or (at your option) any later version.

This library is distributed in the hope that it will be useful, but
WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.

###

you may find it now at my homepage:

http://jfontain.free.fr/tclpython-2.0.tar.gz
http://jfontain.free.fr/tclpython-2.0-1.i386.rpm
http://jfontain.free.fr/tclpython-2.0-1.spec
http://jfontain.free.fr/tclpython2-2.0-1.i386.rpm
http://jfontain.free.fr/tclpython2-2.0-1.spec
http://jfontain.free.fr/tclpython.htm

Enjoy and please let me know what you think.

--
Jean-Luc Fontaine  mailto:jfontain at free.fr  http://jfontain.free.fr/





From ben.hutchings at roundpoint.com  Tue Mar  6 15:54:06 2001
From: ben.hutchings at roundpoint.com (Ben Hutchings)
Date: 06 Mar 2001 12:54:06 -0800
Subject: tokenizer.c crashes in 2.1b1
References: <3aa34d5c.1284101@news.newsguy.com>
Message-ID: <ulmqiobyp.fsf@roundpoint.com>

ransen_spam_me_not at nemo.it (Owen F. Ransen) writes:

> Embedding Python 2.1b1 I recreated the DLL and LIB from
> the sources, VC6 , and any call to PyRun_SimpleFile still
> crashes.
> 
> The debugger points to the guilty party:
> 
> tokenizer.c, the fgets call at line 247 fails.
> 
> My VC project uses Multithreaded Debug compiler settings,
> as does the Pythoncore project.
<snip>

Multithreaded Debug, or Multithreaded Debug *DLL*?  It sounds like you
picked the former, when you want the latter.  Linking with a static
run-time library tends to (but doesn't always) lead to disaster under
Windows.

-- 
Any opinions expressed are my own and not necessarily those of Roundpoint.


From francois.granger at free.fr  Wed Mar 28 15:15:14 2001
From: francois.granger at free.fr (=?ISO-8859-1?Q?Fran=E7ois_Granger?=)
Date: Wed, 28 Mar 2001 20:15:14 GMT
Subject: warrior geeks
References: <3ABF72F1.98CB78F@sage.att.com> <mailman.985634134.9497.python-list@python.org> <99okcr$mt1$1@panix3.panix.com>
Message-ID: <1er02ll.srnkf9j87ztzN%francois.granger@free.fr>

Aahz Maruch <aahz at panix.com> wrote:

> What are you?  Some kind of Commie?  Guns form the very foundation of
> the Pax Americana that gives you the freedom to use Python.

Is this a good location for such a phrase ?

Remember that the 'Rest Of The World' read this NG !

-- 
"La vie est une ?cole o? les le?ons co?tent cher, mais c'est la seule
 o? les insens?s puissent s'instruire."
- Kipling cit? par Bernard Hall?


From brian at sweetapp.com  Wed Mar 21 01:19:11 2001
From: brian at sweetapp.com (Brian Quinlan)
Date: Tue, 20 Mar 2001 22:19:11 -0800
Subject: Base 2 to long integers and back
In-Reply-To: <sab8b93d.007@dohprimo.ccf.health.nsw.gov.au>
Message-ID: <HPEJKHHCMJHPMNLDAPLMKEKPCEAA.brian@sweetapp.com>

I can't think of a library function that does this, but it should be pretty
easy to do it yourself.

String to int:

>>> test = '10110'
>>> val = 0L
>>> for i in test:
...     val <<= 1
...     if i == '1':
...             val += 1
>>> val
22

OR, for the anal and efficiency oriented:

>>> test = '10110'
>>> val = 0L
>>> for i in test:
...     val = ( val << 1 ) + ( ord( i ) & 1 )
>>> val
22

Int to string:

>>> out = ''
>>> val = 100
>>> while val:
...     out = str( val & 1 ) + out
...     val >>=1
>>> out
'1100100'

-----Original Message-----
From: python-list-admin at python.org
[mailto:python-list-admin at python.org]On Behalf Of Tim CHURCHES
Sent: Tuesday, March 20, 2001 7:23 PM
To: python-list at python.org
Subject: Base 2 to long integers and back


Can anyone suggest a fast method of converting base2 strings (e.g.
'0000101110') to Python long integers, and back? I couldn't locate any
functions in the standard library to do this, but no doubt I am overlooking
something obvious.

Cheers,

Tim Churches
Sydney, Australia



--
http://mail.python.org/mailman/listinfo/python-list




From jim.vickroy at noaa.gov  Fri Mar  9 17:11:43 2001
From: jim.vickroy at noaa.gov (j vickroy)
Date: Fri, 09 Mar 2001 15:11:43 -0700
Subject: Dictionary attributes and functions
References: <Pine.LNX.4.21.0103092014190.19640-100000@moroni.pp.asu.edu>
Message-ID: <3AA9551F.62F5F0F9@noaa.gov>

Hello David,

See if the following code fragment meets you needs.

mapping = {} # create an empty dictionary

for item in keys_list: # some list of keys you provide -- duplicates ok

  if not mapping.has_key (item):
    mapping [item] = [] # create empty list mapped to key item
  mapping [item] .append (some_value)

iddwb wrote:
> 
> I'm looking to read a list of items where their may be duplicates.  I'd
> like to use a dictionary to do this but am way to new to python to know
> fully how to do it.  I'm hoping there is some set of fuctions or
> attributes that can be reference regarding dictionaries... specifically if
> key(a) is already in dictionsary x I'd like to know prior to
> adding.  rather than change the value of key(a) I'd like to add the value
> of key(`a) to the value of key(a) so I end up with
> 
> {keya: [123]}
> on insert of keya againg have something like
> {keya: [123, 234]}
> 
> Any pointers on what to look for in the docs?
> 
> btw, I've been going through Mark Lutz's book on "programming python" but
> I'd like something with a little more langugage reference in it.
> 
> David Bear
> College of Public Programs/ASU


From loewis at informatik.hu-berlin.de  Sun Mar 11 15:50:15 2001
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 11 Mar 2001 21:50:15 +0100
Subject: Setting Exception data.
References: <3AAAEA46.5922B1BD@eso.org>
Message-ID: <j4ae6sm3nc.fsf@informatik.hu-berlin.de>

Ivan Munoz <imunoz at eso.org> writes:

> When raising a foo.error exception In my C code I set the error object
> to some data (an object defined on my extension module, an integer and a
> string) as follows:
> 
>  v=Py_BuildValue("Ois", e, errorNumber, errStr);
>  if (v != NULL) {
>    PyErr_SetObject(ErrorObject, v);
>    Py_DECREF(v);
>  }

That may not exactly do what you expect. It first builds a tuple, and
then sets it as the exception. Python normalizes the exception,
essentially by doing

   exc = ErrorObject(e,errorNumber,errStr)

i.e. it creates an instance by calling the class with the tuple as
arguments.

> so far so good, in Python then I catch this exception as:
> 
> except foo.error, (object, errnum, errstr):

Here, you have an implicit instance-unpacking:

  object,errnum,errstr = exc

Now, since exc is a class instance, this accesses exc[0],exc[1],exc[2]
subsequently, which then invokes __getitem__ three times.

> Now what is the right way to acomplish the same from Python?, this does
> not seem to work ....
> 
>     raise foo.error MyObject, errnum, 'You are toasted'

Please have a look at the raise syntax. It allows
  raise Class
  raise Class, Value
  raise Value
  raise Class, Value, Traceback

Value can be an instance of Class, a single value, or a tuple; in the
latter cases, an instance is constructed by calling the class. So you
can write either

     raise foo.error(MyObject, errnum, 'You are toasted')

or

     raise foo.error,(MyObject, errnum, 'You are toasted')

Hope this helps,
Martin


From delza at antarcti.ca  Mon Mar  5 13:54:57 2001
From: delza at antarcti.ca (Dethe Elza)
Date: Mon, 05 Mar 2001 18:54:57 GMT
Subject: Suffering For Your Art
References: <n2rl6.1686$v5.5388@newsfeeds.bigpond.com>
Message-ID: <5iRo6.44561$o%6.860997@news1.rdc1.bc.home.com>

But just think, if computer science was taught with python, the 4-year
curriculum would only take one semester, then all the students would drop
out to start writing code and making money.  C++ was designed to make things
hard to learn to keep comp sci instructors gainfully employed and off the
streets.

--Dethe

"Ben" <bencr at bigpond.com> wrote in message
news:n2rl6.1686$v5.5388 at newsfeeds.bigpond.com...
> # Just A Quick Observation
>
> Is it just me or is there a lot of animosity towards Python as a serious
> programming language ... it seems to be an old school attitude that  if
you
> do not have to struggle and suffer with an idiosyncratic, unintuitive
> programming syntax then it cannot possibly be a powerful language. All
these
> ascetics who would like to see everyone go back to the 'halcyon days' of
> assembly ... IMHO I think Python is a marvellously eloquant and beautiful
> language because it does not get in the road of programming. It pisses me
> off if I have to consult a tome of a help file/manual to find some obscure
> reference to do something that should have been quite simple ...
>
> # Would Anyone Like To Comment?
>
>




From f8dy at diveintopython.org  Sat Mar  3 02:06:52 2001
From: f8dy at diveintopython.org (Mark Pilgrim)
Date: Sat, 03 Mar 2001 02:06:52 -0500
Subject: Thinking Pythonically (was Re: gah! I hate the new string 
	syntax)
In-Reply-To: <97q14l$jbi$0@216.39.151.169>
References: <mailman.983485923.1136.python-list@python.org> <B7In6.16281$Qb7.2658902@newsb.telia.net> <97ohh2$polt0$1@ID-77331.news.dfncis.de> <97opsb$nhu$1@nntp6.u.washington.edu> <97oupq$qijrh$1@ID-77331.news.dfncis.de> <97p3mh$bkm$1@tyfon.itea.ntnu.no> <B6C58CBD.212F%f8dy@diveintopython.org> <97q14l$jbi$0@216.39.151.169>
Message-ID: <B6C6023C.48CD%f8dy@diveintopython.org>

in article 97q14l$jbi$0 at 216.39.151.169, Donn Cave at donn at oz.net wrote on
3/3/01 12:57 AM:
> I have no idea what it's like to use Visual Basic, I have never
> even heard of Powerbuilder.

Heh heh, I'd love to go back to my previous employer and quote you on that.
They thought Powerbuilder was the right tool for every job.

> verbose by comparison.  Many people love Perl.  You may find that
> you do, too, there's no reason to expect otherwise.  But it's

I have had just enough experience with Perl to know that I don't want any
more. ;)

> important to recognize that there is a huge crowd of people who
> swim upstream to use Python instead, and it's not just because
> we have to be different, it's not because we want to use Zope
> or our boss twisted our arm to write in Python.  It's certainly
> not because Python is more powerful, in the sense you have been
> talking about.

Oddly enough, my current boss did twist my arm to learn Python, but I
realize that that's the exception.  But I do now understand better why
people have called my coding style "dense" (meaning "too tightly packed"),
"line noise", etc.  Coming from Perl, I guess Python would be a breath of
fresh air precisely because it's more verbose, less dense, more readable.
But I came at Python from a different angle, after years of being forced to
use inferior, less powerful, less expressive languages which we were pushing
beyond their limits.  (We had to write code from scratch to parse the
command line, because there was nothing like sys.argv.  We had to write
Windows API calls to get a directory listing because we had no os.listdir
and the only built-in function to do it would only put the results into a
graphical listbox.  We had no regular expression parsing whatsoever.  Etc.)
I love that, in Python, I can write in three lines of code what would take
an entire page in Powerbuilder.  If you want to quibble and say that it
would be more Pythonic to write it in four lines of code, that's fine, we
can both happily co-exist.

> know, I guess it's not for everyone, but for myself, I want to
> challenge your notion of what it is to think Pythonically.

Actually, to me, rich data structures are a minor part of thinking
Pythonically; they're just what the original example was about in this
thread.  A much more important part is introspection, which I cover in depth
in chapter 2 of my book (http://diveintopython.org/apihelper_divein.html).
Looking back on it now, I can't believe I ever put up with a language where
you had to choose between dynamically constructing a method name and getting
a return value (*); where you couldn't find out what methods a class had;
where you couldn't get access to the stack trace when something went wrong.

(*) Fellow Powerbuilder survivors will remember that TriggerEvent, which
allows you to call an event by name, doesn't return the event's return
value, and all other ways of calling a method require you to know the method
name at design time.  If you were wondering why I poke fun at Powerbuilder
in my book, that's why.  That, and the
directory-listing-into-a-graphical-listbox function; that was classic.

-M
You're smart; why haven't you learned Python yet?
http://diveintopython.org/





From thomas at xs4all.net  Sun Mar  4 19:16:55 2001
From: thomas at xs4all.net (Thomas Wouters)
Date: Mon, 5 Mar 2001 01:16:55 +0100
Subject: [Python-Dev] Re: Bug fix releases
In-Reply-To: <200103031910.OAA21663@cj20424-a.reston1.va.home.com>; from guido@digicool.com on Sat, Mar 03, 2001 at 02:10:35PM -0500
References: <LNBBLJKPBEHFEDALKOLCMEEPJBAA.tim.one@home.com> <mailman.982897324.9109.python-list@python.org> <200103031821.NAA24060@panix3.panix.com> <200103031910.OAA21663@cj20424-a.reston1.va.home.com>
Message-ID: <20010305011655.V9678@xs4all.nl>

On Sat, Mar 03, 2001 at 02:10:35PM -0500, Guido van Rossum wrote:

> But I understand that you're saying that the community has grown so
> conservative that it can't stand new features even if they *are* fully
> backwards compatible.

There is an added dimension, especially with Python. Bugs in the new
features. If it entails changes in the compiler or VM (like import-as, which
changed the meaning of FROM_IMPORT and added a IMPORT_STAR opcode) or if
modules get augmented to use the new features, these changes can introduce
bugs into existing code that doesn't even use the new features itself.

> I wonder, does that extend to new library modules?  Is there also
> resistance against the growth there?  I don't think so -- if anything,
> people are clamoring for more stuff to become standard (while at the
> same time I feel some pressure to cut dead wood, like the old SGI
> multimedia modules).

No (yes), bugfix releases should fix bugs, not add features (nor remove
them). Modules in the std lib are just features.

> So that relegates us at PythonLabs to a number of things: coding new
> modules (boring), or trying to improve performance of the virtual
> machine (equally boring, and difficult to boot), or fixing bugs (did I
> mention boring? :-).

How can you say this ? Okay, so *fixing* bugs isn't terribly exciting, but
hunting them down is one of the best sports around. Same for optimizations:
rewriting the code might be boring (though if you are a fast typist, it
usually doesn't take long enough to get boring :) but thinking them up is
the fun part. 

But who said PythonLabs had to do all the work ? You guys didn't do all the
work in 2.0->2.1, did you ? Okay, so most of the major features are written
by PythonLabs, and most of the decisions are made there, but there's no real
reason for it. Consider the Linux kernel: Linus Torvalds releases the
kernels in the devel 'tree' and usually the first few kernels in the
'stable' tree, and then Alan Cox takes over the stable tree and continues
it. (Note that this analogy isn't quite correct: the stable tree often
introduces new features, new drivers, etc, but avoids real incompatibilites
and usually doesn't require extra upgrades of tools and such.)

I hope you don't think any less of me if I volunteer *again* :-) but I'm
perfectly willing to maintain the bugfix release(s). I also don't think we
should necessarily stay at a single bugfix release. Whether or not a 'beta'
for the bugfix release is necessary, I'm not sure. I don't think so, at
least not if you release multiple bugfix releases. 

Holiday-Greetings-from-Long-Beach-ly y'rs,
-- 
Thomas Wouters <thomas at xs4all.net>

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



From cfelling at iae.nl  Tue Mar 27 10:20:09 2001
From: cfelling at iae.nl (Carel Fellinger)
Date: 27 Mar 2001 17:20:09 +0200
Subject: Changing every other instance of <B> in a file
References: <8e41ctkk4l38jmj34070j15iotgv4is2cg@4ax.com>
Message-ID: <99qb39$2m0$1@animus.fel.iae.nl>

Lars Kl?boe <larskl at klassekampen.no> wrote:
...
> 	Blablabla <B> talktalk <B> blabla blabla balbalblabla

> The resulting file.html (html)

> 	Blablabla <B> talktalk </B> blabla blabla balbalblabla

> As you can tell, every other instance of <B> is to be changed into </B>.

Let's hope those B-tags aren't nested, then the following might work:

###define an input
file = """
<BB> talktalk<BB> talktalk
<BB> talktalk
  <CC> blabla blabla balbalblabla
    <B> talktalk
    <B> blabla blabla balbalblabla
  <CC> talktalk
<BB> blabla blabla balbalblabla
"""
###define the appropriate html equivalent
result = """
<BB> talktalk<\BB> talktalk
<BB> talktalk
  <CC> blabla blabla balbalblabla
    <B> talktalk
    <\B> blabla blabla balbalblabla
  <\CC> talktalk
<\BB> blabla blabla balbalblabla
"""

import re

### re.sub can deal with functions instead of simple strings to substitute
### but we need a function with state, so let's use a callable class instead
class Change:
    def __init__(self):
        self.dict = {}

    def __call__(self, matchobj):
        key = matchobj.group()
        val = self.dict.get(key, key)
        if key == val:
            self.dict[key] = val[:1] + '\\' + val[1:]
        else:
            self.dict[key] = key
        return val


assert result == re.sub(r'(<[^>]+>)', Change(), file)
-- 
groetjes, carel


From ssthapa at classes.cs.uchicago.edu  Fri Mar 23 13:09:09 2001
From: ssthapa at classes.cs.uchicago.edu (ssthapa at classes.cs.uchicago.edu)
Date: Fri, 23 Mar 2001 18:09:09 GMT
Subject: python suitability for large critical run forever apps
References: <3AB77CE4.78378862@troikanetworks.com> <memo.20010321220223.57471C@digitig.compulink.co.uk>
Message-ID: <slrn9bkcr0.9jv.ssthapa@hepcat.telocity.com>

Tim Rowe <digitig at cix.co.uk> wrote:
>It depends on what you mean by "large critical run forever apps". I 
>routinely deal with systems that can kill a few hundred people if they go 
>wrong, and that's what /I/ mean by "critical". But then I baulk at the 
>suggestion that such a system might be "large", never mind the suggestion 
>that it might be multithreaded! If you need something like IEC61508 SIL4 
>or the military equivalents then forget Python (and C/C++, and Perl, and 
>Ruby...). That doesn't make /any/ of those a "toy" language, just not 
>right for that application. I've said before -- here, I think -- that I 
>don't believe in general purpose languages; to misquote Michael Jackson, 
>if a language offers to helps you with /every/ problem, it won't help you 
>/much/ with /any/ problem.

  FWIW, Erlang which is used by Ericsson and Bluetail to control and operate
their telecom equipment has a fairly nice multithreaded model.  Basically, 
if an error causes a process to crash, a supervisor process just restarts
the process.  This may not be the best solution when you don't want any data to
be missed or lost but it gives fairly good performance on telecom switches.  


-- 
----------------------------------------------------------------------------
	   		              |
Suchandra Thapa                       | "There are only two kinds of math . 
s-thapa-11 at NOSPAMalumni.uchicago.edu  | books. Those you cannot read beyond 
			              | the first sentence, and those you 
			              | can not read beyond the first page."
			              |                     -C.N. Yang
----------------------------------------------------------------------------


From fredrik at pythonware.com  Fri Mar 30 02:14:30 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 30 Mar 2001 07:14:30 GMT
Subject: issues with mixing and matching python packages
References: <mailman.985916558.5517.python-list@python.org>
Message-ID: <qnWw6.4262$4N4.893536@newsc.telia.net>

John Schmitt wrote:
> Does this mean that I can't simply change the installation I'm using by
> simply changing some environment variables?  What's up?
>
> What happens when I upgrade to 2.1?  Is that going to be a problem?

2.0 and 2.1 can coexist.  registry-dependent 2.0 or 2.1 versions
cannot coexist with each other.  non-registry-dependent versions
(like the PythonWare distribution) can coexist with others.

Cheers /F




From colinmeeks at home.com  Fri Mar  2 08:35:17 2001
From: colinmeeks at home.com (Colin Meeks)
Date: Fri, 02 Mar 2001 13:35:17 GMT
Subject: CGI & Python (error)
References: <3A9F5FF5.34E01148@usa.net>
Message-ID: <pkNn6.4257$c7.1344814@news3.rdc1.on.home.com>

> When accessing  the  script I get the error:
>
> File "getdate2.py", line 12, in ?
>     day=form["day"].value
>   File "/usr/lib/python1.5/cgi.py", line 907, in __getitem__
>     raise KeyError, key
> KeyError: day
>
> The script works from the command line.
>
> My problems is:  the script I am trying to run is based on another one.
> This other script works. I can't  figure out the what is causing the
problem
> as the 2 scripts are identical  in every aspect execpt for 1 variable.
Both
> scripts are run in a page which has all the required parameters passed to
it.
> Help. Please...
If this is what I think it is (sorry it's not immediately clear), but when I
try
to parse anything from CGI, if nothing is entered in one of the boxes, this
seems to be especially for radio buttons, then the variable isn't
initialised.

The thing I always do is something like

try:
    day=form["day"].value
except:
    day=1 #or what ever you want the day variable initialised as

I hope this is of some help.

Regards

Colin





From fritz.heinrichmeyer at fernuni-hagen.de  Wed Mar 28 01:27:10 2001
From: fritz.heinrichmeyer at fernuni-hagen.de (Fritz Heinrichmeyer)
Date: 28 Mar 2001 08:27:10 +0200
Subject: Python ready to replace perl (was Re: Install Python on Apache)
References: <86snk1qafw.fsf_-_@jfh00.fernuni-hagen.de> <99fph5$a2n$1@rex.ip-plus.net> <wzelvnz4l5.fsf@sunshine.cs.uu.nl> <mailman.985604254.12448.python-list@python.org>
Message-ID: <86lmpq5ry9.fsf@jfh00.fernuni-hagen.de>

Moshe Zadka <moshez at zadka.site.co.il> writes:
> 
> > 3. mod_perl beeing able to program authentification procedures using

a stupid typo, i meant mod_python, mod_snake ..

Btw. all reactions i got indicate that the answer is "yes" (surprise
surprise) and where helpful and friendly.

-- 
Fritz Heinrichmeyer mailto:fritz.heinrichmeyer at fernuni-hagen.de
FernUniversitaet Hagen, LG ES, 58084 Hagen (Germany)
tel:+49 2331/987-1166 fax:987-355 http://www-es.fernuni-hagen.de/~jfh


From joelg at alum.mit.edu  Mon Mar 12 07:56:57 2001
From: joelg at alum.mit.edu (Joel Gould)
Date: Mon, 12 Mar 2001 12:56:57 GMT
Subject: Tkinter with PIL, Image Corruption - please help
Message-ID: <3aacc682.2989486@news.ne.mediaone.net>

I have noticed corruption of displayed imaged when using the
PhotoImage object from PIL (Python Image Library) under Tkinter.

I am using Windows 98 SE with 256 MB or RAM.  I see the effect under
Python 2.0 (using the latest Py20 Windows binaries).

I have attached a sample program which fails for me.  The failure
symptom is that the lower 10% of some of the images is corrupted with
noise as you cycle through the images (pressing the button).  Not
every image is corrupted and the corruption pattern changes even when
you get back to the same image.

You can get the images from:

http://www.gouldhome.com/Travel/StThomas00/images/001229_1106_charlotteAmalie1.jpg
http://www.gouldhome.com/Travel/StThomas00/images/001229_1106_charlotteAmalie2.jpg
http://www.gouldhome.com/Travel/StThomas00/images/001229_1106_charlotteAmalie3.jpg
http://www.gouldhome.com/Travel/StThomas00/images/001229_1106_charlotteAmalie4.jpg

Attached is the code sample.  I would appreciate it if anyone could
tell me what is wrong.

----- snip -----


import Tkinter
import Image
import ImageTk

imageList = [
    "001229_1106_charlotteAmalie1.jpg",
    "001229_1106_charlotteAmalie2.jpg",
    "001229_1106_charlotteAmalie3.jpg",
    "001229_1106_charlotteAmalie4.jpg",
    ]

class MainDialog:

    def __init__(self, parent):
        self.canvas = Tkinter.Canvas(parent,width=800,height=600)
        self.canvas.pack()

        self.images = []
        for name in imageList:
            image = Image.open(name)
            image.thumbnail( (800,600) )
            self.images.append(image)

        self.currentImage = 0
        self.tkim = ImageTk.PhotoImage(self.images[self.currentImage])
        self.item = self.canvas.create_image(
            0,0,image=self.tkim,anchor='nw')

        button = Tkinter.Button(parent,
            text='Next Image', command=self.onNewImage)
        button.pack()

    def onNewImage(self):
        self.currentImage = self.currentImage + 1
        if self.currentImage >= len(self.images):
            self.currentImage = 0
        self.tkim = ImageTk.PhotoImage(self.images[self.currentImage])
        self.item = self.canvas.create_image(
            0,0,image=self.tkim,anchor='nw')

if __name__ == '__main__':
    root = Tkinter.Tk()
    dialog = MainDialog(root)
    root.mainloop()



From jkraska1 at san.rr.com  Sat Mar 17 14:51:24 2001
From: jkraska1 at san.rr.com (Courageous)
Date: Sat, 17 Mar 2001 19:51:24 GMT
Subject: Pep 238 Why new // operator (Non-integer Division)
References: <98tc2s$qir$1@news2.ottawa.cyberus.ca> <mailman.984822021.13503.python-list@python.org>
Message-ID: <vag7bt09r54tv05nohtssd0ge891v3fa02@4ax.com>

>>All these changes make my head spin.  There is more to life than learning
>>all the fine details in a computer language.  The fewer number of ways to do
>>something the better.
>
>Just because the changes are being made doesn't mean you have to USE them.

This isn't true; the moment you encounter code written by someone else,
you DO have to use them. This simple principle is the magic of Python
minimalism, and the primary driving reason why new Python programmers
can become so effective so fast (IMO).

C//



From aahz at panix.com  Mon Mar 26 20:08:32 2001
From: aahz at panix.com (Aahz Maruch)
Date: 26 Mar 2001 17:08:32 -0800
Subject: PEP 1, PEP Purpose and Guidelines
References: <mailman.985196613.7994.python-list@python.org> <99ifgu$jei$1@panix3.panix.com> <mailman.985651416.18041.python-list@python.org>
Message-ID: <99op6g$dr7$1@panix2.panix.com>

In article <mailman.985651416.18041.python-list at python.org>,
Andrew Kuchling  <akuchlin at mems-exchange.org> wrote:
>aahz at panix.com (Aahz Maruch) writes:
>>
>> After taking a closer look at Moshe's numeric proposals, I suggest adding
>> a Depends-On: header.  (E.g. PEP 240 depends on PEP 239.)
>
>What would Depends-On mean?  Does it mean "To understand PEP 240, go
>read PEP 239 first"?  Or "To implement PEP 240, PEP 239 would first
>have to be implemented or accepted"?  Or something else?  

More that second one (dependency in the makefile sense, kinda).
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

"Boost the stock market -- fire someone"


From fpm at u.washington.edu  Thu Mar  1 00:27:25 2001
From: fpm at u.washington.edu (Frank Miles)
Date: 1 Mar 2001 05:27:25 GMT
Subject: PyQt file locations
References: <3A9AE155.FF8C3A08@ucdavis.edu> <3A9C6E6B.3B727A37@ucdavis.edu> <97j7vk$l2c$1@nntp6.u.washington.edu> <3A9DCE00.CD9BD4AD@ucdavis.edu>
Message-ID: <97kmjt$tom$1@nntp6.u.washington.edu>

Dear Arzhang:

	Thanks for the suggestion.  Unfortunately, it was already on
the path, as running ldconfig -v shows.

But wait... now it works!!!  But I didn't changed ld.so.conf!  Hmmn. I
haven't checked, but perhaps the 'make install' doesn't run ldconfig?
The one thing that was done was to run ldconfig.  I'd assumed that
the 'make install' did that.

	Now I can finally explore PyQt!  Great!

		-frank

In article <3A9DCE00.CD9BD4AD at ucdavis.edu>,
Bruce Wolk  <bawolk at ucdavis.edu> wrote:
>Frank,
>
>Here is the solution that solved my problem:
>-------------
>I had the same trouble with PyQt and RH6.2. The problem is that libsip
>installs itself in /usr/local/lib which is not in your library path by
>default. The best solution is to add the following line to the end of
>/etc/ld.so.conf
>
>/usr/local/lib
>
>and then run /sbin/ldconfig -v as root. You should have no problems after
>that.
>
>Good luck,
>Arzhang
>-------------------
>Frank Miles wrote:
>
>> In article <3A9C6E6B.3B727A37 at ucdavis.edu>,
>> Bruce Wolk  <bawolk at ucdavis.edu> wrote:
>>
>> >I have the same files in my site-packages directory, except for qt.pyc.
>> >
>> >make-install gave me some errors:
>>
>> [snip]
>>
>> >Traceback (most recent call last):
>> >  File "<string>", line 1, in ?
>> >  File "/usr/local/lib/python2.0/site-packages/qt.py", line 43, in ?
>> >    import libqtc
>> >ImportError: libsip.so.4: cannot open shared object file: No such file or
>> >directory
>> >make[3]: *** [install-data-hook] Error 1
>> >make[3]: Leaving directory `/home/bruce/down/PyQt-2.3/qt'
>> >make[2]: *** [install-data-am] Error 2
>> >make[2]: Leaving directory `/home/bruce/down/PyQt-2.3/qt'
>> >make[1]: *** [install-am] Error 2
>> >make[1]: Leaving directory `/home/bruce/down/PyQt-2.3/qt'
>> >make: *** [install-recursive] Error 1
>> >________________________
>> >
>> >When I try to run the examples, I get the same ImportError.
>> >
>> >Any ideas?
>>
>> I can only say that I am getting a nearly identical set of error messages.
>> I'm running Debian 'potato', with Python 1.5.2.  Sip seemed to install ok.
>> Anyone have a clue?
>>
>>         -frank
>> --
>


-- 


From xyzmats at laplaza.org  Wed Mar  7 10:37:35 2001
From: xyzmats at laplaza.org (Mats Wichmann)
Date: Wed, 07 Mar 2001 15:37:35 GMT
Subject: Can anyone recomend a good intoduction to C...
References: <3aa366c6_2@nnrp1.news.uk.psi.net> <8gNo6.440$y6.128357@ruti.visi.com>
Message-ID: <3aa6551e.212870101@news.laplaza.org>

On Mon, 05 Mar 2001 14:19:48 GMT, grante at visi.com (Grant Edwards)
wrote:

>In article <3aa366c6_2 at nnrp1.news.uk.psi.net>, Simon Brunning wrote:
>>... for a Python programmer?
>>
>>I'm going to have to bite the bullet and learn a bit of C. I'd at
>>least like to learn the *right* way to do it.
>
>The "right" way to do it is with a PDP-11, a Lear-Siegler ADM3a
>and the v7 source code.  ;)  Some would argue in favor of an
>ASR-33, but they tend to annoy the downstairs neighbors.

Sheesh, an ADM3a.  What luxury. On that you can run vi.  The ADM3, on
the other hand, only allowed "open" mode - a little better than a line
editor but not much.  (a totally pointless historical tidbit: it was
the 3a that allowed the vi mode of ex to creep out to a wider
population at UC Berkeley and thence to become inflicted on the rest
of the world. Closely followed by the much-less reliable HP 2621.)


(oops, I smell a flood of "I can top that" comments such as "you had a
terminal? I learned on a Teletype", "you had a Teletype? I had to
toggle code in using switches", "you had switches?  I had to use
Forth!".  Consider that joke already played out, eh).

Mats Wichmann

(Anti-spam stuff: to reply remove the "xyz" from the
address xyzmats at laplaza.org. Not that it helps much...)


From chris.gonnerman at usa.net  Mon Mar  5 00:35:31 2001
From: chris.gonnerman at usa.net (Chris Gonnerman)
Date: Sun, 4 Mar 2001 23:35:31 -0600
Subject: Overriding UserList methods
References: <b566atck0ruu4tae8n1srrkntbmme26gor@4ax.com>
Message-ID: <01b101c0a536$1908d480$a100000a@local>

----- Original Message -----
From: "Daniel Klein" <danielk at aracnet.com>
Subject: Overriding UserList methods


> I must be missing something, cos I can't figure out how to override a
method
> using the UserList module. Here is some code:
>
> import UserList
> class MyList(UserList.UserList):
>       __def __init__(self):
>         UserList.UserList.__init__(self)
>
>     def append(self, object):
>         self.append(object)
>
> Understandably, this goes into a recursive frenzy until it hits the stack
> limit. :-(

Like this:

import UserList
class MyList(UserList.UserList):
    def __init__(self):
        UserList.UserList.__init__(self)

    def append(self, object):
        UserList.UserList.append(self, object)

I think, anyway.  Good luck :-)





From MarkH at ActiveState.com  Tue Mar 13 08:31:50 2001
From: MarkH at ActiveState.com (Mark Hammond)
Date: Tue, 13 Mar 2001 13:31:50 GMT
Subject: Want to catch pywintypes.api_error
References: <98kftn$jf9$1@newsreaderm1.core.theplanet.net>
Message-ID: <3AAE208D.4010103@ActiveState.com>

Franz GEIGER wrote:

> When I start win32api.FindFiles(filespec) with a non-existing filespec it
> throws an exception of class pywintypes.api_error. But a "except
> pywintypes.api_error, e:" does not work. Using "except Exception, e:" does
> the job, because Exception is the base class. But how can I catch api_error?
> What do I miss?
> 

win32api.error

Mark.



From James_Althoff at i2.com  Mon Mar 19 19:58:42 2001
From: James_Althoff at i2.com (James_Althoff at i2.com)
Date: Mon, 19 Mar 2001 16:58:42 -0800
Subject: Why "from __future__" stinks; a counter-offer
Message-ID: <OF2F7EDB7A.C2B7955D-ON88256A15.0004FEC5@i2.com>

If folks are getting worked up about the name __future__  (seems fine to
me), perhaps __roadmap__ would be another possibility?

Jim




                                                                                                                   
                    "Tim Peters"                                                                                   
                    <tim_one at email.msn        To:     <python-list at python.org>                                     
                    .com>                     cc:                                                                  
                    Sent by:                  Subject:     RE: Why "from __future__" stinks; a counter-offer       
                    python-list-admin@                                                                             
                    python.org                                                                                     
                                                                                                                   
                                                                                                                   
                    03/18/01 01:04 PM                                                                              
                                                                                                                   
                                                                                                                   




[Jason Orendorff]
> Okay, nested scopes are a good idea.  Having the feature
> "disabled by default" in 2.1 is a good idea.  But this
> "from __future__" stuff isn't so hot.
>
> First, it's ugly.

Good -- then it will stick out <0.5 wink>.  Double double-underscore names
are intended to "stick out", btw, and future_statements are intended to be
temporary.  In that sense, the uglier they are, the more incentive to get
rid
of them ASAP.

> Second, owing mostly to its ugliness, it will be
> *really* embarrassing to try and explain this to my
> (non-Python-bitten) friends.

If they don't use Python, I don't know why you'd bother to mention this to
them regardless of spelling.

> Third, it's extremely silly to have a module named
> "__future__" that describes only the present and the past.

At the moment, __future__.py contains only one entry, which refers to 2.2
final.  Sounds like the future to me.

> Fourth, Python 2.0 and earlier report an
>     ImportError: No module named __future__
> when they encounter a future_statement.  Surely a
> compiler error, even a SyntaxError, would be more
> appropriate and less confusing.

I agree that would be better.  Over time, though, it becomes less
interesting, as people move to more recent releases (which will raise
exceptions at compile-time for future_statements they can't satisfy).

> Lastly and most importantly, it's not too late to do
> something saner.

Unclear, although in the absence of a competing PEP it's certain to remain
as
it is, and the last 2.1 beta release is scheduled for this coming Friday
(as
noted in PEP 226).

> I agree it's desirable to have the code itself advertise that it
> uses a certain nascent feature.

Since syntax and/or semantics change in the presence of a future_statement,
it's *mandatory* that modules desiring the new behavior explicitly say so
--
"desirable" has nothing to do with it.

> So, what if we do something like this:
>
>   1. Pick a name.  Suggestions follow:
>          require
>          using
>          pragma
>          this_code_uses
>          feature
> ...

Did you read Martin's messages on this?  He picked "directive", and
submitted
a patch to implement it (or some part of it).  But he didn't write a PEP,
so
you'll have to search c.l.py and SourceForge for the relevant bits.  He had
in mind a more general facility, and in particular wants to specify
source-file character encodings someday; I actively want not to conflate
future_statements with "general directives" (although "directive future
xxx"
might be OK -- alas, without a PEP, it's a non-starter).

BTW, from the names you suggested, I have to guess that you think a
future_statement is for an "optional feature".  It is not:  it's for an
incompatible language change that will become *mandatory* in a future
release
(and there's no going back then).  future_statements are a temporary aid
for
use during a transition period.  "require", "using", "this_code_uses" and
"feature" all send a wrong message in that respect.  "pragma" is
inappropriate because future_statements are about syntax and semantics, not
pragmatics.

patiently-awaiting-the-end-of-the-world-ly y'rs  - tim


--
http://mail.python.org/mailman/listinfo/python-list







From paulp at ActiveState.com  Thu Mar 22 16:39:09 2001
From: paulp at ActiveState.com (Paul Prescod)
Date: Thu, 22 Mar 2001 13:39:09 -0800
Subject: Automatic install of ActivePython
References: <elikbtker40ll94076q82tugve0jaqh264@4ax.com>
Message-ID: <3ABA70FD.67DA9BE3@ActiveState.com>

Dale Strickland-Clark wrote:
> 
> I've had a pretty good hunt around for an answer to this and I'm somewhat surprised to have drawn a
> blank.
> 
> How do I do an automated or quiet install of ActivePython 2.0?
> 
> It would be nice to be able to set the target directory but I can live without that.

ActivePython has a pretty sophisticated ability to do command-line
installs. This is documented in the release notes which should be in
your "start menu". Yes, you can control the target directory.

Remember: the install can download things off of the web. You should
turn those features off if you are going to be doing batch installs in
places without net connectivity.

-- 
Take a recipe. Leave a recipe.  
Python Cookbook!  http://www.activestate.com/pythoncookbook



From dima at localhost.localdomain  Wed Mar  7 21:20:48 2001
From: dima at localhost.localdomain (Dmitri A. Sergatskov)
Date: 8 Mar 2001 02:20:48 GMT
Subject: National Intruments GPIB
References: <3AA46593.753BAE1@newfocus.com> <ant0721560b02%2U@schwertberger.freenet.de>
Message-ID: <slrn9adr2g.48o.dima@localhost.localdomain>

On Wed, 7 Mar 2001 22:55:56 +0100, 
Dietmar Schwertberger <dietmar at schwertberger.de> wrote:
>Hi Ken!
>
>> GPIB API. Was hoping someone else had done it :)
>
>Instead of using C you could just call the DLL using windll/calldll.
>

I do not think it would work on Linux/x86 or Sparc/Solaris though...

You may want to look at 
ftp://ftp.llp.fu-berlin.de/pub/linux/LINUX-LAB/IEEE488/contrib/
file pythoninterface.tar.gz 
as a starting point also.

Dmitri.



From rdsteph at earthlink.net  Tue Mar 27 21:35:30 2001
From: rdsteph at earthlink.net (Ron Stephens)
Date: Wed, 28 Mar 2001 02:35:30 GMT
Subject: Jython and JPython Questions...How similar to Python, Tkinter?
Message-ID: <3ABFF14A.30D0096D@earthlink.net>

Does Jython support Tkinter? OK, maybe that question as worded is
nonsense, let me re-phrase it: If I code a program in Python and put a
GUI front end on it with Tkinter, how hard would it be to do the same
program in Jython? Or is there a better solution for Jython GUI's?

    By the way, as I am new to Python and learning it, but I am
interested in Jython for embedding (relatively simple) applets into
webpages, how different is Jython from Python? How steep will the
learning curve be for Jython, after learning a reasonable amount of
Python?

    How hard is it to translate Python programs into Jython?

    How good is Jython? Can I write Jython programs, and then turn them
into Java applets that easily embed into web pages? (I have a small but
working knowledge of embedding Java applets in web pages, despite my
overwhelming newbie status in all things coding related ;-)))

    Please forgive my newbieness and all replies will be deeply
appreciated!!!

Ron Stephens



From petar at metamarkets.com  Fri Mar  9 13:20:43 2001
From: petar at metamarkets.com (Petar Karafezov)
Date: Fri, 9 Mar 2001 10:20:43 -0800 
Subject: help needed with forms' parameters handling
Message-ID: <86FC34EA3C4F074AA6BBFFE52ED83FED27BE3E@METAEXCH.sfo.metamarkets.com>

Hello all -

I am running Apache/1.3.14 (Unix) mod_python/2.7.2 Python/2.0. At this point
I am not even sure that this is a mod_python issue, as I have seen this
same behavior before I ported my cgi to mod_python. I am not sure what to do
at this point..

Here's the issue:

When you do a POST to a python script you obtain the parameters using the
util.FieldStorage, when using cgi it's cgi.FieldStorage. The returned result
is a dictionary { key, value }. Everything is OK if you just what to work
with this dictionary. However, if you try to pass this dictionary object to
urllib.urlencode, we raise an exception:

Mod_python error: "PythonHandler parse2"

Traceback (most recent call last):

  File "/usr/local/lib/python2.0/site-packages/mod_python/apache.py", line
189, in Dispatch
    result = object(self.req)

  File "/usr/local/apache/htdocs/test/parse2.py", line 147, in handler
    params=urllib.urlencode(form)	

  File "/usr/local/lib/python2.0/urllib.py", line 1086, in urlencode
    for k, v in dict.items():

AttributeError: 'FieldStorage' instance has no attribute 'items'


Now, the funny part is that if I get this object 'form' - which was defined
as form=util.FieldStorage(req) and just get the keys and values and form a
new dictionary, I can pass that new one to urllib.urlencode with no
problems.

...
    newparams={}                       
    for key in form.keys():
          newparams[key]=form[key]
    params=urllib.urlencode(newparams) 
...

Again, I run into the same issue with both util.FieldStorage when working
with mod_python and with cgi.FieldStorage when using my cgi version.

I am quite happy with this workaround until I bumped into a different issue,
which I am not exactly sure if it's related or not. When there're multiple
checkboxes on a form:

<input type="checkbox"  name="frequencies" value="DEM" priority="6" checked>
<input type="checkbox"  name="frequencies" value="HLF" priority="6">
<input type="checkbox"  name="frequencies" value="DAY" priority="6">
<input type="checkbox"  name="frequencies" value="WEK" priority="6">

and we have selected more than 1, then the form parameter value gets passed
to the server as an array. For example, if I'd select DEM and DAY from the
checkboxed above, my params would be:

'frequences' : ['DEM','DAY']

This is all great, but when I pass the parameters to the server - my python
script is playing the role of a proxy, then I am passing as value for
'frequences' the STRING "['DEM','DAY']" and not the ARRAY ['DEM','DAY']. 

So, somewhere in what I do/use the array gets converted to a string. The way
I see it is:

1) It could be the python's FieldStorage when I originally get the keys and
values
2) It could be that funky dictionary rewriting that I am doing before I pass
the form parameters to urllib.urlencode.
3) I could be something in what urlencode does.


I am not sure where to look at this point. 

Any ideas, anyone ?? Any help would be greatly appreciated!

Thanks for your help!

regards,

Petar Karafezov

MetaMarkets.com
415-575-3015
-------------------------------------------
Investing Out Loud at
http://www.metamarkets.com
-------------------------------------------





From fredrik at pythonware.com  Sat Mar 17 10:33:18 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Sat, 17 Mar 2001 15:33:18 GMT
Subject: image handling
References: <mailman.984795862.9586.python-list@python.org>
Message-ID: <2tLs6.1193$4N4.107379@newsc.telia.net>

Michael Hall wrote:
> I'm brand new on this list and am looking for resources about image
> processing in a web environment in Python. I'm wondering if there exists a
> module that does things like the PHP image handling functions
> (creating, copying, resizing, etc).

here's one:
http://www.pythonware.com/products/pil

Cheers /F

<!-- buy 1.1 today, get 1.2 next week:
http://www.pythonware.com/products/works/index.htm
-->




From danielk at aracnet.com  Sun Mar  4 13:13:10 2001
From: danielk at aracnet.com (Daniel Klein)
Date: Sun, 04 Mar 2001 10:13:10 -0800
Subject: Subclassing builtin types
References: <i6u4atslg4euv732pvaujk6u036kemcidf@4ax.com> <obvo6.14838$1D5.510512@e420r-atl1.usenetserver.com>
Message-ID: <ia15at4undkk6lpv7kn8l2nl9s3bgt30ln@4ax.com>

On Sun, 4 Mar 2001 12:41:55 -0500, "Steve Holden" <sholden at holdenweb.com>
wrote:

>> 1) Are wrapper classes for the builtin types a feasible (and pythonic)
>thing to
>> do?
>>
>Yes and yes: so feasible the library includes a UserList class which
>provides esxactly what you seem to want: it wraps a list inside a class so
>you can inherit from it!

Thanks Steve, you have just saved me a ton of work, something Python has a
habit of doing consistently. :-)

> Delegation is also a useful trick.

I'm not sure what you mean by this. Are there any simple examples I can look
at?

Dan


From phrxy at csv.warwick.ac.uk  Sat Mar 31 09:04:51 2001
From: phrxy at csv.warwick.ac.uk (John J. Lee)
Date: Sat, 31 Mar 2001 15:04:51 +0100
Subject: Gaussian line profile using numeric python
In-Reply-To: <m3lmpn7dxw.fsf@chinon.cnrs-orleans.fr>
References: <3ac31698@zfree.co.nz> <m3puf0re81.fsf@chinon.cnrs-orleans.fr> <Pine.SOL.4.30.0103291850410.13267-100000@primrose.csv.warwick.ac.uk> <m3lmpn7dxw.fsf@chinon.cnrs-orleans.fr>
Message-ID: <Pine.SOL.4.30.0103311455540.4783-100000@mimosa.csv.warwick.ac.uk>

On 30 Mar 2001, Konrad Hinsen wrote:

> "John J. Lee" <phrxy at csv.warwick.ac.uk> writes:
[...]
> > but you probably don't need to know in detail, it boils down to
> >
> > inv_FFT(FFT(data)*FFT(kernel))
>
> Except for zero-padding in case of non-periodic signals. That's why
> I always recommend to look at a detailed description before doing
> this; some of my colleagues keep telling me that FFT techniques "don't
> work", and I suspect this is the reason.

:)

Yes.  And that's (IIRC) what Numeric's convolve() doesn't do quite as
generally as it might -- but there is (IIRC again) an XXX in the manual
somewhere that admits it.

Good old 'Numerical Recipes' has a reasonable description of the
accountancy for the various kinds of convolutions, as long as you've
grasped that you do inevitably lose some of your function to end effects
for non-periodic functions (as I didn't, to start with!).


John



From michael at stroeder.com  Wed Mar 14 02:56:05 2001
From: michael at stroeder.com (Michael =?iso-8859-1?Q?Str=F6der?=)
Date: Wed, 14 Mar 2001 08:56:05 +0100
Subject: Unpickling from strings
References: <qHxr6.19658$pR3.3830844@news2-win.server.ntlworld.com> <IHzr6.21597$Sl.962179@iad-read.news.verio.net>
Message-ID: <3AAF2415.5789AC4E@stroeder.com>

Doug Fort wrote:
> 
> Perhaps it is getting garbled in transmission.  We quote ours:
> 
>     urllib.quote_plus(cPickle.dumps(self._statsdict))

I would pickle it as binary and base64-encode that. The pickled
string might be shorter.

Ciao, Michael.


From jmg at ecs.soton.ac.uk  Wed Mar 28 12:38:32 2001
From: jmg at ecs.soton.ac.uk (Jacek Generowicz)
Date: 28 Mar 2001 18:38:32 +0100
Subject: Boost documentation.
References: <g0wv99x256.fsf@scumbag.ecs.soton.ac.uk>
Message-ID: <g0u24dx087.fsf@scumbag.ecs.soton.ac.uk>

Jacek Generowicz <jmg at ecs.soton.ac.uk> writes:

> Are there any sources of documentation for the Boost Python Library
> besides the source examples, getting_started<1-5>.cpp ?
> 
> How would I write a C++ function which receives a python list and
> converts it into a std::vector<int> for its own use ?
> 
> Thanks,
> 
> Jacek

Well, I've managed to get an attempt past the compiler now, so I'll
post it here for a critique.

It provides one function called `hmm', which takes a tuple (of
integers), and should print out those integers. Unfortunately it
thinks all the integers are zero (though it does get the correct
number of them).

When I try to do something similar with lists rather than tuples, the
get() method appears to be unavailable.

BTW, I'm using gcc version 2.95.2 on linux.
--------------------------------------------------------------
#include <vector>
#include <iterator>
#include <algorithm>

#include <boost/python/class_builder.hpp>
namespace python = boost::python;

namespace {
  void hmm( std::vector<int> v ) {
    std::copy( v.begin(), v.end(), std::ostream_iterator<int>( std::cout, " " ) );
    std::cout << std::endl;
  }
}

BOOST_PYTHON_BEGIN_CONVERSION_NAMESPACE

std::vector<int> from_python( PyObject * p, python::type< std::vector<int> > ) {
  python::tuple tup
    = python::tuple(python::ref(p, python::ref::increment_count));
  std::vector<int> v( tup.size() );
  for ( int i; i<tup.size(); i++ ) {
    int j = from_python( tup[i].get(), python::type<int>() ); 
    std::cout << j << std::endl;
    v[i] = j;
  }
  return v;
}

BOOST_PYTHON_END_CONVERSION_NAMESPACE

BOOST_PYTHON_MODULE_INIT(converting)
{
  try
  {
    python::module_builder this_module("converting");
    this_module.def( hmm, "hmm" );
  }
  catch(...)
  {
    python::handle_exception();
  }
}


From qrczak at knm.org.pl  Sun Mar 18 18:01:36 2001
From: qrczak at knm.org.pl (Marcin 'Qrczak' Kowalczyk)
Date: 18 Mar 2001 23:01:36 GMT
Subject: Who's minister of propaganda this week?
References: <B98041329020E007.7DA84CAD9D0562AB.817BBC572F480869@lp.airnews.net> <98nu1m$12e@dispatch.concentric.net> <660ACCA1E4816F7E.C359CDF4168EF3FB.3B6B05538D619770@lp.airnews.net> <3AAFA13D.76E32C3F@divalsim.it> <98oa4o01i4v@news2.newsguy.com> <mailman.984619039.7856.python-list@python.org> <98q9h60122v@news2.newsguy.com> <slrn9b52dj.8it.qrczak@qrnik.zagroda> <jqcabtsuvmc3io4pbn3ujeuaibg5b2hsbg@4ax.com>
Message-ID: <slrn9bafig.toq.qrczak@qrnik.zagroda>

Sun, 18 Mar 2001 14:24:54 -0800, Daniel Klein <danielk at aracnet.com> pisze:

> Having dealt with both Java and Python, I can confirm that once
> you are use to a dynamically typed language, variable declarations
> become like training wheels; once you learn how to 'ride the bike'
> they just get in the way.

Let me repeat that an important cause of complaints about static
typing are poor languages like C++ and Java.

In Haskell, OCaml, SML, Clean you don't have to write explicit type
declarations on variables or functions. They are usually written on
public functions for the purpose of documentation and for checking
that the body really has the type that the programmer has in mind,
but types can be inferred by the compiler and don't clutter the code.

-- 
 __("<  Marcin Kowalczyk * qrczak at knm.org.pl http://qrczak.ids.net.pl/
 \__/
  ^^                      SYGNATURA ZAST?PCZA
QRCZAK


From jwbnews at scandaroon.com  Sun Mar  4 13:30:28 2001
From: jwbnews at scandaroon.com (John W. Baxter)
Date: Sun, 04 Mar 2001 10:30:28 -0800
Subject: Bug fix releases
References: <LNBBLJKPBEHFEDALKOLCMEEPJBAA.tim.one@home.com> <mailman.982897324.9109.python-list@python.org> <200103031821.NAA24060@panix3.panix.com> <200103031910.OAA21663@cj20424-a.reston1.va.home.com> <15009.17989.88203.844343@cj42289-a.reston1.va.home.com> <mailman.983648765.842.python-list@python.org>
Message-ID: <jwbnews-30D883.10302704032001@corp.supernews.com>

In article <mailman.983648765.842.python-list at python.org>,
 Guido van Rossum <guido at digicool.com> wrote:

[Quoting Aasz, I think <JB>]
> >   Betas of the bugfix releases are important -- portability testing is
> > fairly difficult to do when all we have are Windows and Linux/x86
> > boxes.  There's definately a need for at least one beta.  We probably
> > don't need to lengthy, multi-phase alpha/alpha/beta/beta/candidate
> > cycle we're using for feature releases now.
> 
[Guido:]
> OK, you can have *one* beta.  That's it.

That covers the hoped for case, in which no disasters appear during that 
beta.  

If one does, it implies that the bug-fix release is dead, since even 
backing out the failed fix requires fresh testing of the remainder, just 
as fixing the fix would require fresh testing.

This could actually be a useful discipline:  it might improve the 
quality of the submitted fixes.

  --John (who comes from a time of slower release cycles than the world 
sees now)

-- 
John W. Baxter   Port Ludlow, WA USA  jwbnews at scandaroon.com


From db3l at fitlinxx.com  Fri Mar 16 18:48:19 2001
From: db3l at fitlinxx.com (David Bolen)
Date: 16 Mar 2001 18:48:19 -0500
Subject: Cookies
References: <mailman.984707839.22839.python-list@python.org> <lmq6m07j.fsf@propylon.com> <mailman.984773062.25819.python-list@python.org>
Message-ID: <uzoel8ed8.fsf@ctwd0143.fitlinxx.com>

"Ken Seehof" <kens at sightreader.com> writes:

> Does anyone know of another mailing list more specific to http questions?
> 
> 1. Client downloads an execuatable (frozen python application).
> 2. Client downloads a web page.
> 3. Web page javascript generates a statistically unique random identifier.
> 4. Web page sends identifier to the server, which uses it as database key.
> 5. User runs python application
> 6. *** Python application somehow gets the identifier from somewhere ***
> 7. Now the python application and the server both have the same identifier.

Because of the restrictive client Javascript environment you'll
probably be faced with, crossing the boundary between inside the
browser and outside is not trivial.  

Is there a reason that the identifier has to only be generated by 3?
It's not clear if your Python application is going to connect to the
server anyway, but if so couldn't it just do the same processing
you're putting into the javascript and generate a unique identifier
(that it thus knows about) at that point?

Or, if it isn't going to connect to the server, how about just letting
the Python application open up it's own web page internally to run
appropriate code on your web server to generate the identifier (which
could run as a server script), and then Python could parse it right
from the resulting HTML information.

> The only question I am asking is how to do step 6.
> 
> Alternative:
> 
> 1. Client downloads an execuatable (frozen python application).
> 2. Client downloads a web page.
> 3. *** Web page javascript makes a unique identifier for the client. ***
> 4. Web page sends identifier to the server, which uses it as database key.
> 5. User runs python application
> 6. *** Python application somehow generates the same identifier ***
> 7. Now the python application and the server both have the same identifier.
> 
> Again the only question I am asking is how to do steps 3 and 6 to get the
> same number.

This would probably prove even harder.  Algorithmic generation of a
truly unique identifier will almost always include some time based
deviation, so I'm not sure about an algorithm that will both be
globally unique and re-computable at different times - at least not
without still passing some information between the two compute engines
which puts you back at your same problem.

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


From lcortes at aodinc.com  Mon Mar 19 12:28:13 2001
From: lcortes at aodinc.com (Luis Cortes)
Date: Mon, 19 Mar 2001 10:28:13 -0700
Subject: What IDE are you'all using for Python 2.1 for Windows?
Message-ID: <995gai$2bv$1@news.chatlink.com>

Hi,

  Just wanted to know what "good" ide's are out there for python.  Please
let me know which ones you prefer.

Thanks,
Luis.





From aaldv97 at student.removethispart.vxu.se  Mon Mar 26 10:36:05 2001
From: aaldv97 at student.removethispart.vxu.se (David Andreas Alderud)
Date: Mon, 26 Mar 2001 17:36:05 +0200
Subject: Lists and Files
References: <99njdo$oa3$1@news.lth.se> <Xns90709FB7E5A4duncanrcpcouk@127.0.0.1>
Message-ID: <99nnm1$pak$1@news.lth.se>

Thank you very much Duncan Booth and Steve Purcell.
I really like Pythons fast prototyping capabilites, nothing short of
amazing, I'll never use LISP again :)

/Kind, regards
    David A. Alderud




From jwbnews at scandaroon.com  Tue Mar 13 20:09:04 2001
From: jwbnews at scandaroon.com (John W. Baxter)
Date: Tue, 13 Mar 2001 17:09:04 -0800
Subject: This math scares me
References: <3aad4062.1290021@News.CIS.DFN.DE> <20010312170356.B3513@better.net> <mailman.984480856.18097.python-list@python.org> <984486395.407379@newsmaster-04.atnet.at> <slrn9asf1d.ja3.qrczak@qrnik.zagroda> <984518199.127140@newsmaster-04.atnet.at> <jwbnews-09D75F.13313313032001@corp.supernews.com> <yTwr6.1235$Tg.167429@ruti.visi.com> <3AAEAB92.DC8E984A@san.rr.com>
Message-ID: <jwbnews-AABCA1.17090213032001@corp.supernews.com>

In article <3AAEAB92.DC8E984A at san.rr.com>, Darren New <dnew at san.rr.com> 
wrote:

> Grant Edwards wrote:
> > Of course we'd never imagined we'd ever see hardware FP and BCD
> > libraries were only marginally slower than binary ones. Now
> > that FP is usually in hardware, BCD FP (or fixed point) will be
> > pretty slow by comparison.
> 
> One mainframe I worked on (Xerox 560, AKA Sigma 9) had the "scientific
> module" which was hardware floating point, and the "business module" which
> was the hardware BCD. The latter added instructions like "Edit byte string"
> that would copy packed BCD into a string according to COBOL display-edit
> formatting. (Or whatever COBOL called the "print using" formatting. Too many
> years. Not enough COBOL. Or something like that.)

Sigmas were quite nice.

And the NCR C-315 line had a company-supplied Fortran with (obviously!) 
floating point routines, and a user group supplied set of floating point 
libraries which actually worked, which most sites used.  The 315 was a 
decimal machine (BCD)...a word was 12 bits...either 2 characters or 3 
digits.  The accumulator was as long as it needed to be (up to 12 slabs) 
to contain the current value...there was a T@ register (Tally of 
Accumulator) which held its current length.

Words were called syllables (ergo  slabs) because someone in engineering 
had thought 12 bits too short to be dignified as a word.

  --John

-- 
John W. Baxter   Port Ludlow, WA USA  jwbnews at scandaroon.com


From fredrik at pythonware.com  Thu Mar 29 10:04:36 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Thu, 29 Mar 2001 15:04:36 GMT
Subject: Basic Question: 2.0 --> 2.1 do extensions need to be recompiled?
References: <mailman.985875039.20340.python-list@python.org>
Message-ID: <8aIw6.4172$4N4.850578@newsc.telia.net>

Mike C. Fletcher wrote:
> I'm short on time this week, so I'd rather not have to get Python, Numeric,
> etceteras just to compile PyOpenGL if it isn't necessary.  So, is
> recompilation required for extension modules compiled for 2.0 to work with
> 2.1?

in theory, yes.

in practice, it may work on some platforms (but under protest).
windows is not such a platform.  unix might be.

Cheers /F




From bsass at freenet.edmonton.ab.ca  Fri Mar  2 01:49:48 2001
From: bsass at freenet.edmonton.ab.ca (Bruce Sass)
Date: Thu, 1 Mar 2001 23:49:48 -0700 (MST)
Subject: Lower bounds of lists
In-Reply-To: <00cb01c0a2b2$d1063fe0$a100000a@local>
Message-ID: <Pine.LNX.4.33.0103012331220.13596-100000@bms>

> > It seems an odd choice, as humans count from one, and Python in
> > other respects seems quite logical and well thought out.
>
> Humans count from 1.  Hackers count from 0.

Not really - it's an index (the offset from the start), not a count.

If you are at the start of something, the offset is 0.

<...>
> C does it that way.  So, in general, does assembler, and the internal
> representation of unsigned integer numbers runs from 0 to (2^bits)-1,
> where bits is the number of bits of data storable.
>
> Counting from 0 is counter-intuitive to humans, but so is much of
> computer science.  Ultimately, it's because Guido wanted it that way.

Consider this... maybe you are not really counting, you are actually
building a list then referring to the items in it.

Once you get your head in the right space it all becomes clear.

HTH


- Bruce




From bsb at winnegan.de  Tue Mar  6 13:05:35 2001
From: bsb at winnegan.de (Siggy Brentrup)
Date: 06 Mar 2001 19:05:35 +0100
Subject: urllib.urlopen and proxies
In-Reply-To: <BfakOm9OYyn2V0Hg9h1SSS=a7t7Y@4ax.com>
References: <vse4atgcgcfafth7c5tq3po7srd3l32f3h@4ax.com>
	<008301c0a57d$1331cde0$a100000a@local>
	<mailman.983804894.17415.python-list@python.org>
	<BfakOm9OYyn2V0Hg9h1SSS=a7t7Y@4ax.com>
Message-ID: <87g0gq6ads.fsf@winnegan.de>

Paul Moore <paul.moore at uk.origin-it.com> writes:

> On Mon, 5 Mar 2001 16:07:42 +0100 , Siggy Brentrup <bsb at winnegan.de>
> wrote:
> 

[...]

> >I suggest using urllib2, changing the proxy setup described in the
> >module's docstring to
> >
> >  proxy_support = urllib2.ProxyHandler({"http" :
> >                                        "http://user:pw at proxy:port"})
> >
> >Untested too, since my proxy doesn't require authorization.
> 
> I had a look at urllib2. The module docstring even includes an
> example!
> 

[...]

> 
> The only problem is that I don't know what the 'realm' and 'host'
> parameters to authinfo.add_password should be. I need to supply a
> username and password for everything...

The authinfo stuff is for remote sites requiring authentication, as
long as you visit unprotected sites only you can ignore it.

Modify the docstring example as follows:

--------
import urllib2

proxy_info = {
    'user' : 'your userid',
    'pass' : 'your password',
    'host' : "proxy.doma.in',
    'port' : 3128 # or 8080 or whatever
}
 
# build a new opener that uses a proxy requiring authorization
proxy_support = urllib2.ProxyHandler({"http" :
                "http://%(user)s:%(pass)s@%(host)s:%(port)d" % proxy_info})
opener = urllib2.build_opener(proxy_support)
 
# install it
urllib2.install_opener(opener)
 
f = urllib2.urlopen('http://www.python.org/')
--------

Proxy authorization still untested, please report if it works.

Siggy




From cwebster at nevada.edu  Tue Mar 13 13:16:55 2001
From: cwebster at nevada.edu (Corran Webster)
Date: Tue, 13 Mar 2001 10:16:55 -0800
Subject: Sortin dictionaries in python
References: <3aae4757.68617130@News.CIS.DFN.DE>
Message-ID: <cwebster-27E369.10165513032001@news.lvcablemodem.com>

In article <3aae4757.68617130 at News.CIS.DFN.DE>, costas at springmail.com 
wrote:

> Is there a simple way to sort a dictionary? It seems the only way to
> do this is by indirectly creating  a sorted list.

Dictionaries have no inherent order, but if you need to get at the items 
in a sorted order, you'd do something like:

keys = mydict.keys()
keys.sort()
for key in keys:
    # do stuff with mydict[key]

If the data that you are dealing with is something that is strongly 
ordered, you may be better off using someting like a list of tuples 
instead of a dictionary.

> And while on the topic of sorting lists. I tried sorting a list and
> wanted to to pass a compare function to the list.sort() method. The
> documentaion gives no example.

Have a look at the Sorting Mini-HOWTO from the main python site.

http://www.python.org/doc/howto/sorting/sorting.html

This has a solution to the problem below, as well.

> import string
> 
> x=['a','c','B']
> 
> def ignoreCase(left, right):
>      if string.upper(left) < string.upper(right):
>          return -1
>      else:
>          return 1

I think that this will work, but a comparison function should return 0 
if the two strings are equivalent, just in case the sort method changes 
in a future version of Python.  The following is a better way of doing 
this:

def ignorecase(left, right):
    return cmp(string.upper(left), string.upper(right))

In fact, if you are using a Python which has string methods, you could 
write this as:

def ignorecase(left, right):
    return cmp(left.upper(), right.upper())

which will be slightly more efficient.

> x.sort(ignoreCase(left, right))

When passing a function you only need to pass the name of the function, 
not the function parameters:

x.sort(ignorecase)

tells sort to use the comparison function "ignorecase".  Sort expects 
this function to have the right number of arguments, do the right kind 
of things to the arguments, and return the right kinds of results.

Finally, if you don't want to have to define a separate function and 
just want to do a one-off sort, you could do the following one-liner:

x.sort(lambda left, right: cmp(left.upper(), right.upper()))

This is quick if you only sort once, but slow if you need to sort many 
lists in this way.

Regards,
Corran


From ws-news at gmx.at  Wed Mar  7 04:06:22 2001
From: ws-news at gmx.at (Werner Schiendl)
Date: Wed, 7 Mar 2001 10:06:22 +0100
Subject: class object access in class definition?
References: <Pine.LNX.4.21.0103061556510.2015-100000@zero>
Message-ID: <983956262.417565@newsmaster-04.atnet.at>

Depends on what you really need this for

But you can of course access the class object in the __init__ method
like this

>>> class Foo:
...  def __init__(self):
...   print Foo
...

hth
werner

Lars Damerow <lars at pixar.com> wrote in message
news:Pine.LNX.4.21.0103061556510.2015-100000 at zero...
> Hi folks!
>
> Is it possible to reference a class from within its own definition?
Something
> along the lines of:
>
> class Foo:
>     print Foo
>
> This works for functions--that is,
>
> >>> def foo():
> ...     print foo
> ...
> >>> foo()
> <function foo at 0x8111024>
>
> works as expected, but the class definition above raises a NameError:
>
> >>> class Foo:
> ...     print Foo
> ...
> Traceback (most recent call last):
>   File "<stdin>", line 1, in ?
>   File "<stdin>", line 2, in Foo
> NameError: There is no variable named 'Foo'
>
> Thanks for any pointers!
> -lars
>
> ___________________________________________________________
> lars damerow
> senior unix administrator
> pixar animation studios
> lars at pixar.com
>
> "Nauseous.  Nauseated.  The first means 'sickening to contemplate'; the
second
> means 'sick at the stomach.'  Do not, therefore, say, "I feel nauseous,"
unless
> you are sure you have that effect on others."
>
> - William Strunk Jr., "The Elements of Style"
>




From Laura_01 at MailAndNews.com  Sun Mar 11 23:20:41 2001
From: Laura_01 at MailAndNews.com (Laura Lewin)
Date: Sun, 11 Mar 2001 23:20:41 -0500
Subject: Python In A Nutshell - suggestions
Message-ID: <3AAE7366@MailAndNews.com>

Hi!
Thanks for your suggestions.  We'll definitely take these into account.
Laura
LLewin at oreilly.om

>===== Original Message From Louis Luangkesorn <lluang at northwestern.edu> =====
>Laura,
>
>Well, since your asking for suggestions . . .
>
>One thing I would like the Nutshell book to include are sections on some of 
the commonly used modules that may not be included in the base
>distribution.  I say this because for the language itself and standard 
modules, there is already printed documentation in the form of the Python
>Essential Reference by David M. Beazley and presumably the The Standard 
Python Library by Fredrik Lundh that is coming out.  But in no place I've 
seen
>is there documentation on the third party modules.
>
>Of course this begs the question of "what are commonly used modules?"  There 
are a couple of lists out there.  One is the list of modules that is on the
>ActiveState page, 
http://www.activeState.com/Products/ActivePython/Extensions.html,
>These include:  tkinter, wxPython, expat, Numeric Python, and ODBC (they have 
mxODCB, but using DBI 2.0 compliant modules in general would make for a
>useful section) Another useful list are the modules that are mentioned in 
Python Programing for Windows32 by Hammond and Robinson in Chapter 3.  They
>only have space for a paragraph each as a teaser, but it would be good to 
have a few pages in a Nutshell book.  (and if you really want to know, the
>modules I use are Numeric Python and ODBC, and I'm going to be choosing 
between tkinter and wxPython soon)
>
>And while I'm at it, one useful book to have would be on wxWindows/wxPython.  
I see references to it in several places, and a few intros on the web, but
>nothing that looks like good documentation (besides help files), even on the 
internet.  (and this book would be useful to C, C++ as well as Python
>programmers.)
>
>Thank you. (as I look at my shelf with several binders of printed 
documentation)
>
>Louis
>LLewin at oreilly.com wrote:
>
>> Hi,
>> Yep--Python Nutshell is "forthcoming"...we're just in the planning stages 
at this point--very early on in the process.  Any suggestions are welcome!
>> Laura
>> LLewin at oreilly.com
>
>--
>K Louis Luangkesorn
>lluang at northwestern.edu  http://pubweb.nwu.edu/~kll560  PGP:0xF3E2D362
>Whatsoever things are true, ... honest, ... just, ... pure, ... lovely, ... 
of good report; if there be any virtue, and if there be any praise, think on
>these things.- motto - Northwestern University
>



From whisper at oz.net  Sat Mar 31 20:12:12 2001
From: whisper at oz.net (Dave LeBlanc)
Date: 1 Apr 2001 01:12:12 GMT
Subject: HTML Tidy and Python wrapper
References: <985945106.28359.0.nnrp-02.c1c3e154@news.demon.co.uk>
Message-ID: <9a5v9c$fc7$1@216.39.170.247>

Dunno about a python wrapper, but a wrapper for Tcl has been done and
some mods to tidy are done that allows it to be repeatedly called
instead of "once through" batching as it comes from the "factory" ;-)

Check out Tcl on SourceForge and look around for TclXML etc.

Regards,

Dave LeBlanc

On Fri, 30 Mar 2001 10:37:41 +0100, "Paul Brian" <pbrian at demon.net>
wrote:

>Dear all,
>
>I was sure that I recently read that there was a python wrapper around the
>HTML Tidy (http://www.w3.org/People/Raggett/tidy/). Unfortunately I cannot
>find the reference. Does any one know if such a thing exists, as simple
>searches through deja have come up nought.
>
>thank you
>
>



From loewis at informatik.hu-berlin.de  Fri Mar  2 10:56:42 2001
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 02 Mar 2001 16:56:42 +0100
Subject: A counter-counter-proposal for PEP 236: #pragma( nested_scopes)
References: <j47l29lpo8.fsf@informatik.hu-berlin.de> <mrwn6.16173$Qb7.2637930@newsb.telia.net> <0Swn6.3324$TW.17061@tor-nn1.netcom.ca>
Message-ID: <j4itlsjhad.fsf@informatik.hu-berlin.de>

"Warren Postma" <embed at geocities.com> writes:

> Has anyone considered borrowing a C notation:
> 
> #pragma( nested_scopes )
> 
> Benefits:
>     - It's a comment therefore it doesn't break anything else.
>     - It compiles fine and is ignored on pre-2.x systems.
>     - More pragmas can be added later, for warnings, code generation,
> etcetera.

I have considered it, but I dislike the idea of putting semantics into
comments. Comments are there for human readers, and not to be
interpreted by the language. If it has any formal meaning to the
language, it should have formal, outside-of-comments syntax as well.

Regards,
Martin



From wesc at deirdre.org  Tue Mar 20 15:19:39 2001
From: wesc at deirdre.org (Wesley J. Chun)
Date: Tue, 20 Mar 2001 12:19:39 -0800
Subject: Thread join example
Message-ID: <200103202019.MAA22128@emperor.deirdre.org>

<?????`	

    > From: seung-won hwang <shwang5 at students.uiuc.edu>
    > Subject: Thread join example
    > Date: Tue, 20 Mar 2001 13:59:08 -0600
    >
    > I am looking for a code example which contains thread join. I would
    > appreciate your help very much!


i put a small chapter on multithreaded programming in Core Python
Programming, including several simple code examples using the join()
method of Thread objects which come from the threading module.

click on the book URL below and hunt down the Chapter 17 code samples.

hope this helps!

-wesley

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 
Silicon Valley-SF Bay Area Python users group:  http://baypiggies.org

"Core Python Programming", Prentice Hall PTR, December 2000
    http://starship.python.net/crew/wesc/cpp/

wesley.j.chun :: wesc at baypiggies.org
cyberweb.consulting :: silicon.valley, ca
http://www.roadkill.com/~wesc/cyberweb/



From phd at phd.pp.ru  Thu Mar  1 12:41:41 2001
From: phd at phd.pp.ru (Oleg Broytmann)
Date: Thu, 1 Mar 2001 20:41:41 +0300 (MSK)
Subject: sre bug with \b in locale mode
Message-ID: <Pine.LNX.4.30.0103012037050.31158-100000@fep132.fep.ru>

Hi!

   The following program works fine under python 1.5.2 and has a problem
under python 2.0. It seems like a bug in sre in hadling \b in locale mode.
Is it a known bug? Can anyone test it with latest python/sre (adapt last
line to your locale, please)?

import re, locale
locale.setlocale(locale.LC_ALL, "")

def test(s):
   pattern = re.compile(r"\b%s\b" % s, re.LOCALE)

   match = pattern.search(s)

   if match:
      print s, "- found!"
   else:
      print s, "- not found..."

test("home") # english text
test("???")  # russian cyrillic, encoding koi8-r, locale is koi8-r

Oleg.
----
     Oleg Broytmann            http://phd.pp.ru/            phd at phd.pp.ru
           Programmers don't die, they just GOSUB without RETURN.




From phrxy at csv.warwick.ac.uk  Wed Mar 28 10:17:44 2001
From: phrxy at csv.warwick.ac.uk (John J. Lee)
Date: Wed, 28 Mar 2001 16:17:44 +0100
Subject: wavlets?
In-Reply-To: <tc27anovquqm79@corp.supernews.co.uk>
References: <tc27anovquqm79@corp.supernews.co.uk>
Message-ID: <Pine.SOL.4.30.0103281552540.7942-100000@mimosa.csv.warwick.ac.uk>

On Wed, 28 Mar 2001, Taz wrote:

> i'm about to write a wavelets module for python - does one already exist...
> couldn't find one?

Try asking on the Numerical Python mailing list.


John



From kcocaine at GamesandTheory.com  Sun Mar  4 19:51:23 2001
From: kcocaine at GamesandTheory.com (Kevin Douglas)
Date: Sun, 4 Mar 2001 19:51:23 -0500 (EST)
Subject: Need a free lancer python programmer - to do bots
In-Reply-To: <006e01c0a4fa$3c4e0820$52d0fea9@mario>
Message-ID: <Pine.LNX.4.21.0103041948210.19952-100000@critical.GamesandTheory.com>

On Sun, 4 Mar 2001, [iso-8859-1] M?rio Gilberto Eichler Junior wrote:

> I need a free lancer programmer to do a bots to search information in 35 sites based in a word dictionary and write the results in a mysql database
> 
> I will pay per a job.
> 
> I leave in Brazil.
> 
> All sites are locate in Brazil.
> 
> Will pay in American Dollars.
> 

hola!

i used to work for StarMedia Network - we rolled out a python product
in spanish and portugeuse - i'm sure i'd be perfect for this job

i really really really need the money too - u.s. corporate scene is all
whacky right now and i live in NYC - one of the most effected places

how much are you looking to spend and when do you need it done by?

i'm not busy at the moment i'm sure i can knock it out rather quick

dont really care how much the $$$ is either, as long as its a fair deal

been working with agent based technologies in python for 3 years now :)

multi-million transaction level type stuff

anyway - let me know - send me some details on the project - what kind
of server its on and if theres a development environment - etc


-kev






From aahz at panix.com  Wed Mar 14 00:06:47 2001
From: aahz at panix.com (Aahz Maruch)
Date: 13 Mar 2001 21:06:47 -0800
Subject: This math scares me
References: <3aad4062.1290021@News.CIS.DFN.DE> <984518199.127140@newsmaster-04.atnet.at> <jwbnews-09D75F.13313313032001@corp.supernews.com> <yTwr6.1235$Tg.167429@ruti.visi.com>
Message-ID: <98mu97$fd1$1@panix3.panix.com>

In article <yTwr6.1235$Tg.167429 at ruti.visi.com>,
Grant Edwards <grante at visi.com> wrote:
>
>Of course we'd never imagined we'd ever see hardware FP and BCD
>libraries were only marginally slower than binary ones. Now
>that FP is usually in hardware, BCD FP (or fixed point) will be
>pretty slow by comparison.

If anyone cares, it ought to be fairly easy to create an unbounded BCD
module using Python's longs.
-- 
                      --- Aahz <*>  (Copyright 2001 by aahz at pobox.com)

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

Python:
    Programming the way
    Guido indented it


From mertz at gnosis.cx  Sun Mar 11 22:46:30 2001
From: mertz at gnosis.cx (Dr. David Mertz)
Date: Sun, 11 Mar 2001 22:46:30 -0500
Subject: sql query results to xml (fwd)
Message-ID: <WaEr6kKkXguX092yn@bellatlantic.net>                         

Scott Hathaway wrote:
> I want to create a tool that will take a valid SQL query and return the
> results as valid xml.  Before I reinvent the wheel, is there a python class
> or code snippet that someone has that does this already?

Uche Ogbuji <uche at ogbuji.net> responded:
|Not that I know of.  Are you planning to use the syntax from Oracle's
|XSQL, or DB2's XML Extender, or any of the other hundreds of formats
|available (see
|http://www.linuxworld.com/lw-2001-02/lw-02-xml3databases.html
|But all that is just to point out that you probably don't have to
|reinvent a resultant XML format.

Before anyone *does* decide to roll-their-own version of this, I wanted
to mention that Scott Hathaway and I are working on a pair of tools in
answer to Scott's original inquiry.  Consider this preliminary, but my
little bit can be seen at:

  http://gnosis.cx/download/sql2dtd.py

A *likely* output file (partially hand done) can be found at:

  http://gnosis.cx/download/sql2xml_sample.xml

I've looked at a version of Scott's sql2xml.py portion of things, but am
not sure if he wants to make that available quite yet.

Anyway, it will be a matter of just a few days before we'll have a
fairly reasonable Python solution to the (small) problem above.
Hopefully, it will be useful to some developers.

However, contrary to Uche's no-doubt-sound advice, our stuff is a brand
new format, not any of those referenced by Uche's article.  The choices
are straightforward and obvious, but presumably not *identical* to other
systems.

Yours, David...



From sheila at spamcop.net  Tue Mar 27 08:46:05 2001
From: sheila at spamcop.net (Sheila King)
Date: Tue, 27 Mar 2001 13:46:05 GMT
Subject: Reading e-mail body (newbie)
References: <mailman.985697556.21712.python-list@python.org>
Message-ID: <a561ctgvf31s3evn9ogrdjr5fl0d7l0nfm@4ax.com>

On Tue, 27 Mar 2001 14:51:08 +0200, Jacobus van der Merwe <jacobus at geo.co.za>
wrote in comp.lang.python in article
<mailman.985697556.21712.python-list at python.org>:

:I have a script to which e-mail to a certain address is relayed.
:I need to write the body of the e-mail to a file.
:I use
:
:rfc822.Message(sys.stdin)
:
:to read the e-mail, but I can't figure out how to access the body.
:It seems that rfc822 only have methods to access the headers. ????
:Any help will be appreciated.


#! /usr/bin/env python

import rfc822, sys, os

inheaders=rfc822.Message(sys.stdin, 0)

for h in inheaders.headers:
	print h,

print 
for line in sys.stdin.readlines():
	print line,
print
print
print "This message was passed through a Python script"
print "passmailthrough.py"


--
Sheila King
http://www.thinkspot.net/sheila/
http://www.k12groups.org/


From andrew.bonello at metrics.co.uk  Wed Mar 28 04:22:17 2001
From: andrew.bonello at metrics.co.uk (Andrew Bonello)
Date: Wed, 28 Mar 2001 10:22:17 +0100
Subject: Python and SWIG - passing object pointer param to an embedded script.
Message-ID: <3ac1a999@pron.i-way.net.uk>

Good Day

I have a query as regards embedded Python scripts and the Python/C API.

I am invoking evaluation of a Python script from within my C++ code using
the PyRun_File(...) function. I want to be able to pass a pointer to one of
my PyObject-derived C++ objects into this script as a parameter. Then,
within the Python script, I want to be able to pass this parameter to
function calls in an imported module defined using a SWIG interface, which
invokes C++ code in a DLL that I have written.

What I am trying to do here is to be able to work in my wrapping C++ code,
creating and modifying the state of some (PyObject-based) C++ objects, then
invoke an embedded Python script, which in turn calls some C++ functions
defined in my SWIG module. Those functions may themselves modify the state
of my outer C++ objects, so clearly I need some way of linking the objects
into my script environment when the Python interpreter is run.

I have tried to do this as follows:

// intialise python interpreter
Py_Initialize();

std::string m_FileName = "./test.py";
FILE* fp = fopen( m_FileName.c_str(), "r" );
const char* fn = m_FileName.c_str();

PyObject *globals = PyDict_New();
PyObject *locals = PyDict_New();

// this is my outer C++ object which I want Python and any SWIF functions
called to have access to
PyObject* myobj = new MyPyObj();

PyDict_SetItemString( globals, "myobjptr", myobj);

PyObject* obj = PyRun_File( fp, (char*)fn, 0, globals, locals );

This just causes an access violation. (I am using Microsoft Visual C++ V6.0
with Python V2.0, Debug build).

It seems to me that what I really ought to be doing is getting a reference
to the *exisiting* global dictionary and adding my object pointer to that,
rather than constructing a new dict using PyDict_New() - can anyone offer
any help on this?

Furthermore, my "MyPyObject" derived class may not yet be a valid PyObject.
I have been unable to find examples of how to write code for and correctly
initialise a simple C++ object which inherits from the PyObject class and
can thus be accessed from within embedded Python scripts. If anyone is able
to offer any examples of doing this, I would be most grateful.

Thanks very much!


Andy Bonello




From matthias.oberlaender at daimlerchrysler.com  Thu Mar  8 05:11:11 2001
From: matthias.oberlaender at daimlerchrysler.com (matthias.oberlaender at daimlerchrysler.com)
Date: 8 Mar 2001 10:11:11 GMT
Subject: Who am I: can a class instance determine its own name?
References: <mailman.984034150.9635.python-list@python.org> <UPHp6.38461$1D5.1679137@e420r-atl1.usenetserver.com>
Message-ID: <987lrv$7v5$1@news.sns-felb.debis.de>

> "Tim CHURCHES" <TCHUR at doh.health.nsw.gov.au> wrote in message
> news:mailman.984034150.9635.python-list at python.org...
> This is probably an elementary question and the answer is probably writ
> large in multiple places in the Python documentation, but...
> 
> ...can an instance of a class determine the name of the variable to which 
it
> is assigned? For example:
> 
> ###########################
> class Foo:
>     def whoami(self):
>          return "You are a Foo() but I do not know your name"
> 
> FooBar = Foo()
> 
> print FooBar.whoami()
> ###########################
> 
> How does one define the method whoami() so that it returns "FooBar"? This
> sort of navel gazing is formally called introspection, I think (therefore I
> am)?
> 
> Tim Churches
> Sydney, Australia
> (where, due to the Coriolis effect, the Python prompt does indeed look like
> this: <<< - or maybe its because we are upside-down)

In <UPHp6.38461$1D5.1679137 at e420r-atl1.usenetserver.com> "Steve Holden" 
wrote:
> This question recurs so frequently I would appreciate knowing of any
> omprovement which could be made to FAQ entry 4.97: How can my code discover
> the name of an object?
> 
http://www.python.org/doc/FAQ.html seems to be out of data. I could not find 
4.97 on that page.
However,  the wizard finds it:
http://www.python.org/cgi-bin/faqw.py?query=4.97&querytype=simple&casefold=y
es&req=search

> This tries to address the question in a helpful manner.
> 
Yes, but here is something more constructive:

>>> def howDoYouCall(me, here):
..   # return a list of names whose value is identical to 'me' in namespace 
'here'
..   result = []
...   for (name, value) in here.items():
...     if value is me: result.append(name)
...   return result
...   
... 
>>> x = 5
>>> y = 3
>>> z = None
>>> 
>>> howDoYouCall(3, globals())
['y']
>>> howDoYouCall(5, globals())
['x']
>>> howDoYouCall(None, globals())
['__doc__', 'z']
>>> 

Does this help a bit?

--
 ____  __  _/_/ . 
( / / ( /  / / /  

=====================================================================
Matthias Oberlaender, DaimlerChrysler AG, Research Center Ulm
FT3/AB (Information Technology / Image Understanding)
Wilhelm-Runge-Str. 11,  P.O. Box 2360,  89013 Ulm, Germany
Phone: +49 731 505 2354       Fax: +49 731 505 4113
Email: matthias.oberlaender at daimlerchrysler.com
=====================================================================




From Padraig at Antefacto.com  Thu Mar 15 14:14:22 2001
From: Padraig at Antefacto.com (Padraig Brady)
Date: Thu, 15 Mar 2001 19:14:22 GMT
Subject: wxPython/Tkinter/pyGTK
Message-ID: <3AB11488.3010006@Antefacto.com>

Hi,
I'm just starting to use python, and my first
app will have the following widgets:

Tabbed interface, with tabs in horizontal position.
toolbar.
treeview.
listview (with embedded checkboxes)
the odd button.

like:

+--------------------------+
| MyApp                    |
+--------------------------+
|       toolbar            |
|--------------------------|
| tab1 +-----------------+ |
| tab2 |  treeview &/or  | |
| ..   |  listview       | |
| tabn +-----------------+ |
+--------------------------+

So

1. I'm wondering what to use?
    wxPython, Tkinter, PMW+Tkinter or pyGTK

2. Is there any documents comparing
    these options.

3. Are there other options I haven't considered.

4. What option do people consider as having
    the least dependencies (easiest to install),
    as this app is GPL'd and must be easy to install.

Thanks,
Padraig.



From thys at klaribel.netsys.co.za  Fri Mar 23 12:42:59 2001
From: thys at klaribel.netsys.co.za (Thys)
Date: Fri, 23 Mar 2001 19:42:59 +0200
Subject: wxImageList use in wxPanel
Message-ID: <01032319432603.09619@klaribel.feetjieland>

Hi All,

I need to display a few images on several places in a wxPanel, the obvious
soluttion would be to use a wxImageList, especially as the panel changes over
time (a status screen of sorts). 

I cant seem to get it to work though, no compilation problems but the graphics
refuse to be displayed. The documentation seems to concentrate on wxListCtrl
and wxtreeCtrl none of which is applicable in this case.

appreciated as always

-- 
Regards

Thys

--------------------
Thys Meintjes
thys at netsys.co.za

Netsys International(Pty) Ltd
http://www.netsys.co.za
+27 12 348-4246



From fredrik at pythonware.com  Tue Mar 20 16:10:54 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Tue, 20 Mar 2001 21:10:54 GMT
Subject: New PEP: The directive statement
References: <mailman.985117312.5806.python-list@python.org> <j4u24oi4e8.fsf@informatik.hu-berlin.de>
Message-ID: <yHPt6.2553$4N4.334967@newsc.telia.net>

Martin von Loewis wrote:
> That is also why the PEP 236 proposal (__future__ imports) finds that
> much resistance it uses an action statement to declare things.

except that "import" already is a declaration statement, of course.

> P.S. If "directive transitional nested_scopes" is approved

well, adding new, ugly, and utterly non-pythonic syntax for a one-
shot workaround strikes me as a really bad thing.  if I want Tcl
(random interpretation of barewords) or applescript (lame attemts
to make things look like english), I know where to find it.

Cheers /F




From fredrik at pythonware.com  Wed Mar 14 11:33:23 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 14 Mar 2001 16:33:23 GMT
Subject: [Python-Dev] Revive the types sig?
References: <jeremy@alum.mit.edu> <15020.9404.557943.164934@w221.z064000254.bwi-md.dsl.cnc.net> <200103120711.AAA09711@localhost.localdomain> <mailman.984439040.24419.python-list@python.org> <s44tatolisnipdfm4r4p1aqt2e09e458hk@4ax.com> <mailman.984581571.10100.python-list@python.org> <1pWvOlsMJ9Zxk0d4s9BhFmMCSusy@4ax.com>
Message-ID: <n3Nr6.263$4d3.55227@newsb.telia.net>

Paul Moore wrote:
> Type annotations, as described so far, sound like a sanitised and formalised
> version of this:
>
>     def f(n):
>         assert type(n) == type(1)
>
> I can imagine *very* few places where an assert like this is of genuine
> practical use.
>
> Of course, if type annotations will *actually* formalise and support
> interfaces in some form, that's entirely different - I'm mostly neutral, about
> something like this.
>
>     def f(n):
>         assert <n supports the "integer" interface>

footnote: the correct way to spell this is:

         assert isinstance(n, type(1)) # or types.IntType

as you say,

         assert type(n) is type(1)

is mostly evil.

Cheers /F




From sheila at spamcop.net  Mon Mar 19 20:08:39 2001
From: sheila at spamcop.net (Sheila King)
Date: Tue, 20 Mar 2001 01:08:39 GMT
Subject: Problems with 2.0 cgi scripts
References: <fs9bbtg68qne71t853lp89l44uarbb5jk5@4ax.com> <9956p1$8g0$1@panix3.panix.com>
Message-ID: <d9bdbtccsabg2sn9dqsnqanht080rger6r@4ax.com>

On 19 Mar 2001 06:57:37 -0800, aahz at panix.com (Aahz Maruch) wrote in
comp.lang.python in article <9956p1$8g0$1 at panix3.panix.com>:

:In article <fs9bbtg68qne71t853lp89l44uarbb5jk5 at 4ax.com>,
:Sheila King  <sheila at spamcop.net> wrote:
:>
:>So, basically, it cannot find the modules when I run it as a cgi-script. I
:>have all the paths set correctly, and the permissions set correctly. Grr, I'm
:>frustrated with this one by now...
:
:You can't set the path because you don't own the web server.  Simplest
:solution is to set sys.path yourself as the first line of the script.

Thanks, that was one solution.

However, the other solution was for me to re-install Python with the correct
exec_prefix option. (I had not done that correctly.)

Now I've fixed everything, and it's running smoothly. Thanks to D-Man, Michael
Reilly, and other regulars on the Python Tutor list!

More details here:
http://www.aota.net/ubb/Forum3/HTML/001564-1.html

--
Sheila King
http://www.thinkspot.net/sheila/
http://www.k12groups.org/



From jra at dorothy.msas.net  Tue Mar 27 09:27:04 2001
From: jra at dorothy.msas.net (Jay R. Ashworth)
Date: Tue, 27 Mar 2001 14:27:04 GMT
Subject: Linux GPS program
References: <3AC01637.4EF3A6F2@crynwr.com>
Message-ID: <slrn9c17k4.cfc.jra@dorothy.msas.net>

This one time, in band camp,
  Russell Nelson <nelson at crynwr.com> wrote:
> Hi. I have written a GPS program in Python for Linux. I'm sure that
> someone could port it to run under Windows, but I have no interest in
> doing so. I haven't done a real release of the software yet, since
> it's still under heavy development. But you're welcome to download a
> copy from handhelds.org's CVS. The location is "apps/pygps". For info
> on handhelds.org's CVS: http://www.handhelds.org/sources.html
>
> Right now, it does moving maps (UTM bitmaps), track recording, and
> playback. Will select the "best" of multiple maps.

No *schidt*.  Cool, Russ; thanks.  Any plans to do vector map data as
well?  (Yeah, I know, I know; "it's never enough". :-)

Did you use wxPython?

Cheers,
- jra
-- 
Jay R. Ashworth                                                jra at baylink.com
Member of the Technical Staff     Baylink
The Suncoast Freenet         The Things I Think
Tampa Bay, Florida        http://baylink.pitas.com             +1 727 804 5015


From marcin.chady at ntlworld.com  Tue Mar 13 18:03:46 2001
From: marcin.chady at ntlworld.com (Marcin Chady)
Date: Tue, 13 Mar 2001 23:03:46 -0000
Subject: Unpickling from strings
Message-ID: <qHxr6.19658$pR3.3830844@news2-win.server.ntlworld.com>

I have pickled an object into a string and saved it in an hidden HTML form
field, so that I can recover the state when the form is resubmitted.
However, when the form comes back and I try to unpickle my object I get an
error saying that the module in which the object's class was originally
defined is unknown, even though the CGI script which does the unpickling has
already imported the module!

Has anyone got an idea why it is?

Marcin




From topmind at technologist.com  Sun Mar 25 04:39:32 2001
From: topmind at technologist.com (topmind)
Date: Sun, 25 Mar 2001 09:39:32 GMT
Subject: Collection interfaces
References: <3a90fa97.0@news1.mweb.co.za>   <3A9A86B1.9A8C69A2@ix.netcom.com> <MPG.1504d14563dce6499896af@news.earthlink.net> <97fjla128ir@news2.newsguy.com> <97h5s40vhc@news2.newsguy.com> <MPG.1506007591cf818b9896bc@news.earthlink.net> <slrn99opop.gr.amk@mira.erols.com> <MPG.1506265ebf6d934d9896be@news.earthlink.net> <97ibmc0k47@news1.newsguy.com> <MPG.1506cf3293b094b49896c1@news.earthlink.net> <97m0c7080c@news2.newsguy.com> <MPG.15085517debf4b649896ca@news.earthlink.net> <zb%n6.31033$o%6.586026@news1.rdc1.bc.home.com> <MPG.150b383cea226d409896d0@news.earthlink.net> <kNko6.6986$eq3.707121@news.uswest.net> <MPG.150c46db9cec283e9896d4@news.earthlink.net> <3AA2A898.450B659E@mail.com> <gRCt6.12224$Im6.1308560@newsread1.prod.itd.earthlink.net> <3AB755F8.67E84126@mail.com> <%a%t6.17245$227.1601569@newsread2.prod.itd.earthlink.net> <3AB8A730.4BE13D22@mail.com> <61bv6.8693$P4.676800@newsread1.prod.itd.earthlink.net> <B6E2B64C.D8E3%t*b*r*u*m*l*e*y@f*u*s*e.n*e*t>
Message-ID: <o1jv6.9635$P4.789083@newsread1.prod.itd.earthlink.net>

>
>
http://www.cincom.com:80/scripts/smalltalk.exe/gettingstarted/index.asp?cont
> ent=trysmalltalknc
>

It seems to be wacked out right now.


> Pardon the word wrapping my news client does. There's no spaces in the
> above.
>
> That should get you to a place where you can download the lesson browser
> (thanks again to Mark Windholtz, great job!) and the tutorial (thanks to
Jim
> F. as well).
>

Lesson browser download?

The last thing I need is yet more foriegn software on my PC making it crash
more and confusing other DLL's.

Please, HTML.

>...site runs on
> Microsoft ASP VSscript instead of Smalltalk?
>
> Naw. Jim isn't afraid of anything. At least, nothing on Usenet!
>
> Me either :)
>
> The rest of the Cincom site, and the old Smalltalk portion of the site,
does
> indeed not run on Smalltalk. We'll get it all moved over eventually :)
>
> Heck, my own business unit, which is very into VisualWave, understands
that
> we have to use JavaScript on our pages to drive the browsers. There's
> nothing wrong with being multi-lingual.
>
> >
> > Do Chevy salesmen drive Fords?
>
> Some do. I work for Cincom and use both our Smalltalk products, I actually
> prefer ObjectStudio when I'm doing Windows specific work, but I also have
> Squeak and Dolphin around and have experimented with both.
>
> > -tmind-
>
>




From fredp at mygale.org.nospam  Thu Mar 22 08:04:09 2001
From: fredp at mygale.org.nospam (Fred Pacquier)
Date: 22 Mar 2001 13:04:09 GMT
Subject: Good Editor?
References: <3ab9a32b$1_1@news2.one.net> <Xns906C5F868E6B7PaCmAnRDLM@161.48.76.20> <GALn19.C1z@leeds.ac.uk>
Message-ID: <Xns906C8F49A8732PaCmAnRDLM@161.48.76.20>

nde at comp.leeds.ac.uk (Nick Efford) said :

>> You don't mention the platform. Under Unix/Linux there are a number of
>> options (including Vim, Emacs, etc.). Under Windows, I found WScite 
>> (http://www.scintilla.org) the easiest to get started with...
>
>vim and emacs run on Windows, too!  So you can use one editor across
>platforms as well as one programming language :)

This is totally true, I should have been more precise. I didn't mean to say 
that these solutions weren't cross-platform, just that for a Windows-only 
beginner the Scite editor seemed easier to set up and offers good 
functionality. BTW, Scite also works with Linux/GTK...

-- 
YAFAP : http://www.multimania.com/fredp/


From f8dy at diveintopython.org  Sat Mar 10 16:04:14 2001
From: f8dy at diveintopython.org (Mark Pilgrim)
Date: Sat, 10 Mar 2001 16:04:14 -0500
Subject: Private class and instance methods?
In-Reply-To: <3AAA73BF.9CDCF3B2@alcyone.com>
References: <3AAA73BF.9CDCF3B2@alcyone.com>
Message-ID: <B6D000FE.4D14%f8dy@diveintopython.org>

in article 3AAA73BF.9CDCF3B2 at alcyone.com, Erik Max Francis at
max at alcyone.com wrote on 3/10/01 1:34 PM:
> How often do people use methods beginning with __ in order to get some
> amount of privacy with the automatic name mangling that occurs, to get
> some kind of weak protection, with, say, accessor methods:
> 
> class C:
> def __init__(self, x):
> self.__x = x
> 
> def getX(self): return self.__x
> def setX(self, x): self.__x = x
> 
> [which would of course allow C.getX and C.setX to do something different
> in the future]?

There is a better way to do this in Python, using the __getattr__ and
__setattr__ special methods.

class foo:
    privatekeys = ('x',)
    
    def __init__(self, x):
        self.x = x
    def __getattr__(self, key):
        print "getting private value %s" % key
        return self.__dict__['__%s' % key]
    def __setattr__(self, key, value):
        if key in self.__class__.privatekeys:
            print "setting private value %(key)s to %(value)s" % locals()
            self.__dict__['__%s' % key] = value
        else:
            self.__dict__[key] = value

if __name__=='__main__':
    f = foo('abc')
    print "f.x = %s" % f.x
    f.y = []
    print "f.y = %s" % f.y
    f.x = 10
    print "f.x = %s" % f.x
    
Running the example script produces this output (minus the comments):

setting private value x to abc # from __init__ method
getting private value x # from getting f.x in first print in main block
f.x = abc # output of 1st print
f.y = [] # output of 2nd print
setting private value x to 10 # from setting f.x
getting private value x # from getting f.x in 3rd print
f.x = 10 # output of 3rd print

There is no f.x (insert gratuitous reference to "The Matrix" here), because
whenever you try to set it, Python calls __setattr__ and you end up setting
f.__x instead, and whenever you try to get it, Python calls __getattr__ and
you end up getting f.__x instead.

f.y does exist, just like a normal public attribute.

You can control which attributes are redirected by adding their names to the
class attribute "privatekeys".

More on special methods:
  http://diveintopython.org/fileinfo_specialmethods.html
  http://www.python.org/doc/current/ref/specialnames.html

More on class attributes:
  http://diveintopython.org/fileinfo_classattributes.html

Hope this helps.

-M
You're smart; why haven't you learned Python yet?
http://diveintopython.org/





From spahievi at vega.bg  Mon Mar 12 17:22:25 2001
From: spahievi at vega.bg (Niki Spahiev)
Date: Tue, 13 Mar 2001 00:22:25 +0200
Subject: Packages, modules and exceptions
In-Reply-To: <98ip5c$397$1@clematis.singnet.com.sg>
References: <98ip5c$397$1@clematis.singnet.com.sg>
Message-ID: <1514001711.20010313002225@vega.bg>

12.3.2001, 17:14:52, Ng Pheng Siong wrote:


NPS> Hi,

NPS> In M2Crypto, I have the following (abridged) directory structure:

NPS>     M2Crypto/
NPS>             SSL/
NPS>                 __init__.py
NPS>                 Connection.py

NPS> Within M2Crypto/SSL/__init__.py, I have the following:

NPS>     class SSLError(Exception): pass


Why not "class SSLError(StandardError)" so we can leave SystemExit
alone.

-- 
Best regards,
 Niki Spahiev





From urner at alumni.princeton.edu  Fri Mar 16 18:39:14 2001
From: urner at alumni.princeton.edu (Kirby Urner)
Date: Fri, 16 Mar 2001 15:39:14 -0800
Subject: revamping K-12 math curriculum
Message-ID: <ll55btc59tdho7jhto7dldvmtbh386330g@4ax.com>


     Exploring Integrated Math Topics + a Programming Language 
          with an Eye towards Developing the Content of a 
           Standard Pre-College Mathematics Curriculum

                          by Kirby Urner
                          March 15, 2001


     [slightly revised from first draft on math-teach at
                    the Swarthmore Math Forum          ]


I. CURRICULUM AS CONNECTED NETWORK OF "FABRICS"

I've used this forum (math-teach) as a scratch pad for evolving 
some curriculum segments.  A curriculum segment is a patch of 
related topics and activities that might be revisited at 
different levels of difficulty and sophistication.  Another, 
perhaps better, word for a segment is a "fabric" or a "weave".  
There's an emphasis on achieving a kind of seamless blend of 
themes and patterns, which assists students wanting a sense of 
the big picture (overview is what seems to be lacking in a lot 
of current math teaching).[1]

One example of a fabric is the whole patch of topics surrounding 
sequences and series, figurate numbers, polyhedra and Pascal's 
Triangle.  Conway & Guy develop this weave in 'The Book of Numbers' 
(Springer-Verlag, 1996), whereas I'd independently hit on a similar 
approach in my 'Numeracy and Computer Literacy' series.  I add 
the Buckminster Fuller element, with mention of the jitterbug 
transformation, geodesic spheres, viral protein sheaths and 
buckyballs.[2]

Another example of a fabric blends topics from cryptology, 
probability and group theory.  Here we develop the idea of a 
simple 'clubhouse code' based on letter substitution, and use 
this as a segue to the group theory concept of permutations, 
and multiplication as the composition of permutations. This 
framework allows us to discuss the properties of a group, in 
contrast to those of semi-groups, rings and fields.  The next 
step is to use the new concepts to develop more sophisticated 
enciphering strategies, wherein the substitution dictionary 
continues to change throughout the encryption process.  This 
provides a useful segue to historical threads, such as the 
storylines involving German U-boats, the Enigma code, Alan 
Turing and Bletchy Park ala Neal Stephenson's bestseller 
'Cryptonomicon'.[3]

Both of these fabrics contain material we could be phasing in 
as early as first or second grade...

Fabric #1:  Number and Geometry

Sphere packing and polyhedra (built with tooth picks and little 
marshmellows, for example) are a popular topic with smaller 
kids (I know from experience -- and not just because of the 
marshmellows -- modeling clay works too).  Euler's Law (V+F=E+2) 
makes sense to slightly older kids, as does Descartes' Deficit 
(720 degrees).  The idea of triangular and square numbers, 
which may be modeled by packing spheres (as in 'Pool Hall Math' 
-- see math-teach archives), makes sense early too, with 'flat 
shaped numbers' becoming 'spatial' (my 'beyond flatland' theme) 
rather early in the game (ala 'The Book of Numbers'). Fuller's 
10 F^2 +2, for the number of spheres in a cuboctahedral shell,
identical to the number in an icosahedral shell, fits into 
this context.  That the number of spheres is the same for both
shapes is shown by the jitterbug transformation, which hyperlinks 
to phi (the cuboctahedron's square cross-sections transform 
into the icosahedron's golden rectangles) and provides a segue 
to virology, buckyball chemistry, crystalography and architecture.[4]

Fabric #2:  Permutations and Cryptology

The idea of a 'clubhouse code' starts to make sense as soon as 
children have some mastery of the alphabet, and an understanding 
of why you might want to keep a message secret and therefore 
indecipherable by anyone but the intended recipient.  Games of 
chance, ideas about probability, enter the picture here, 
explorable with polyhedral dice (not just hexahedra) -- note 
the hyperlink to Fabric #1 here.  Using permutations to discuss 
the composition of functions makes sense at the middle school 
level, where we also should be developing the idea of modular 
arithmetic e.g. 240 mod 13.

II.  BEYOND CALCULATORS:  PYTHON AS 'FIRST COMPUTER LANGUAGE'

Aside from 'beyond flatland', another major theme promulgated 
by my Oregon Curriculum Network (a kind of Oriental Rug Factory 
for curriculum fabrics -- math-related especially), is 'beyond 
calculators'. Since the 1980s, programming languages have 
matured considerably, and since the 1990s we've had the option 
of teaching Python as a first language.  The synergetic blend 
of procedural, functional, and object-oriented programming 
styles available through Python provides a good nucleus of 
concepts which will serve a student well when branching out to 
other languages later, whether in the direction of C/C++/Java, 
LISP/Scheme, or any of several other well-traveled pathways.

Having some knowledge of a generic programming language, 
developing familiarity with it by coding around interwoven math 
topics, is going to yield many benefits later on in life.  
Early exposure (but not too early) lowers the chances of 
developing debilitating phobias later on, and gives students a 
vehicle for turning math topics into opportunities for 
exploration and portfolio-building (programs may be saved, 
revisited and improved over time). Using programs written in 
earlier classes to tackle topics in later classes helps provide 
a sense of continuity and relatedness -- something students 
need, but don't always get, in the current hodge-podge.

In 'Numeracy + Computer Literacy', I use Fabric #1 to develop 
some graphical output capability, by synergizing Python with a 
ray tracer (I use POV-Ray and VRML plug-ins, but many other 
graphics applications would be suitable, plus there's the 
VPython option, permitting an even more interactive approach 
based on OpenGL).  This link to computer graphics is likewise a 
segue to what I'll call Fabric #3, the weave on concepts 
relating coordinate systems, vectors, and geometric transformations 
as implemented by matrices or quaternions or some other apparatus 
(e.g. Clifford Algebra).  

Fabric #3 is where we use object-oriented programming to explore 
the Gibbs-Heaviside vector operations, bringing in trigonometric 
functions, rotation matrices and so on -- all of which makes 
more sense and is a lot more fun when one is rewarded with 
colorful, shadowed, perspective renderings, such as POV-Ray 
(or other software) provides (such colorful renderings might 
be transferred to T-shirts for those so inclined).

III.  TRAVERSING THE NET: FROM CRYPTOGRAPHY TO CALCULUS

With Fabric #2, students will be able to encipher and decipher 
text files, using their newfound/maturing computer language 
skills.  Given the opportunity is to encipher random passages, 
we have more opportunities to jump outside the math domain and 
capture topical paragraphs from other points in the curriculum.  
For example, we might choose to encipher Lincoln's Gettysberg 
Address, or a speech by Martin Luther King.  If students are 
on a network, they can practice passing ciphertext and secret 
keys back and forth.

Some educators may object to this focus on cryptography at an 
early age, suggesting that students will misuse this 
knowledge.  However, my experience is that many in 8th grade 
and above develop this interest on their own, picking up memes 
from popular culture, and, since this material is "avoided" in 
school, they develop the impression that cryptography is 
something the "establishment" is against, i.e. this knowledge 
takes on a "subversive" spin.  This is unfortunate, and it 
would be better if the schools worked in a more empowering 
mode, to help students develop skills and knowledge which is 
part of the mainstream commercial environment and net-based 
ecosphere.  Cryptography is not inherently "subversive" -- 
unless we craft an environment which makes it seem that way 
(which, perhaps inadvertently, is where we are today, in 
many school districts).[5]

The group theory aspect of permutations (as applied to 
cryptography) segues nicely to other kinds of group, such as 
sets of postivie integers less than and relatively prime to
some modulus (this is why modulo arithmetic was important). 
We can talk about CAIN (closure, associative, inverse, 
neutral element) and Abel(ian) groups.  The notion of coprimes 
segues to Euler's totient function, and Euler's Theorem, of 
which Fermat's Little Theorem is a special case -- more 
topics typically covered in group theory with applications 
to cryptography, in that huge primes are hard to "crack" 
into factors (what makes RSA and some other public key 
systems effective).  

Thanks to Python, we have access to big numbers (long 
integers) and don't have to suffer the trade-offs that 
plagued earlier "math through programming" forays into 
this realm (i.e. we don't trade away access to big numbers 
just because we choose to use a generic programming language, 
and not a strictly math-focussed or number theoretic 
package).[6]

We can loop back to Fabric #1 and talk about symmetry groups 
and polyhedra, or the group properties of quaternions under 
multiplication.  All of these concepts will have been developed 
concretely, in a hands-on, interactive context, using a 
computer language (I prefer Python) by the end of high school. 
The result will be a more sophistacted and non-math-phobic 
mindset among students, who have a well-connected set of math-
related concepts and a feel for the bigger picture.

I haven't forgotten the calculus, and will simply point out 
that we have a good basis for exploring limits starting with 
Fabric #1, with phi and Fibonacci numbers in particular (with 
a link between phi and the fractal-like, recursive nature of 
phi-based five-fold symmetric geometries). Fabric #1 was a lot 
about Sigma and Pi (summations and products of a sequences 
respectively), and about differences between successive terms 
in a sequence, the discrete math analog of the differential.  

With Fabric #3 (graphics/vectors), we have the standard 
graphing calculator topics of (x,f(x)) and might explore the 
calculus of the catenary for example (suspension bridges and 
electrical wires have this shape), which will include mention 
of hyperbolic trig functions.  I can talk more about Fabric #4 
in another post.[7]

Kirby

[1] these 'fabric' terminology is borrowed from memetics as 
embodied in Fluidiom, which is also an elastic interval geometry
application (http://www.fluidiom.com/)

[2] http://www.inetarena.com/~pdx4d/ocn/numeracy0.html

[3] http://www.inetarena.com/~pdx4d/ocn/clubhouse.html
    http://mail.python.org/pipermail/edu-sig/2001-March/001077.html

[4] Jay Kappraff's, Connections, the geometric bridge between
art and science (McGraw-Hill, 1996) is also good on the 
root(2)-based vs. phi-based aesthetics.  For an animated GIF
of the jitterbug, see "Getting Inventive with Vectors"
http://www.inetarena.com/~pdx4d/ocn/numeracy1.html

[5] Teachers might want to use the relevance/importance of a
cryptographic "fabric" as an argument for bringing more powerful
computing platforms into math classrooms, vs. "making do" with
less suitable calculators for this purpose.  For more along
these lines, see my post to k12.math.teach:
http://www.mathforum.com/epigone/k12.ed.math/glaterdphum/

[6] here's a post to math-learn quoting an earlier curriculum 
writer who had to contend with this trade-off:
http://groups.yahoo.com/group/math-learn/message/217


[7] or see http://www.inetarena.com/~pdx4d/ocn/catenary.html



From greg at cosc.canterbury.ac.nz  Thu Mar  1 19:40:38 2001
From: greg at cosc.canterbury.ac.nz (Greg Ewing)
Date: Fri, 02 Mar 2001 13:40:38 +1300
Subject: Best way to string compare a python object in a getattrofunc?
References: <mailman.983400200.18487.python-list@python.org>
Message-ID: <3A9EEC06.2180834D@cosc.canterbury.ac.nz>

Cayce Ullman wrote:
> I am
> thinking about interning the string "spam" on module initialization, then
> saving that pointer somewhere, and then just comparing the pointers on every
> fetch.

Interning your string on initialisation is a good idea,
BUT you should use PyObject_Compare() to compare it with
other strings. If the string you're comparing it with
also happens to be interned, PyObject_Compare() will do
a pointer comparison, otherwise it will use strcmp().
So you'll get the benefits of interning where possible,
while still doing the right thing in other cases.

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


From fredrik at pythonware.com  Tue Mar 20 02:26:38 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Tue, 20 Mar 2001 07:26:38 GMT
Subject: What IDE are you'all using for Python 2.1 for Windows?
References: <995gai$2bv$1@news.chatlink.com> <995in5$22s$1@panix2.panix.com>
Message-ID: <OCDt6.2395$4N4.291204@newsc.telia.net>

Luis Cortes wrote:
>  Just wanted to know what "good" ide's are out there for python.  Please
> let me know which ones you prefer.

pythonworks 1.2 (but I supposed I'm biased ;-)

    http://www.pythonware.com/products/works

Cheers /F




From zalivin at krona.obninsk.ru  Mon Mar 12 18:16:26 2001
From: zalivin at krona.obninsk.ru (Alexei Zalivine)
Date: Tue, 13 Mar 2001 02:16:26 +0300
Subject: IPC module
Message-ID: <984438966.381139@basic.maxnet.ru>

Hello,
Please help me to compile shm module for linux. I'm not a C programmer, and
I can't understand errors from compiler. You can get the source from
http://sirac.inrialpes.fr/~marangoz/python/shm/

Are there other IPC modules for Python?

Thank you,
Alexei.




From kalle at gnupung.net  Wed Mar 21 17:52:41 2001
From: kalle at gnupung.net (Kalle Svensson)
Date: Wed, 21 Mar 2001 23:52:41 +0100
Subject: Simple regex that has me banging my head against a wall
In-Reply-To: <20010321143144.K24341@trufflehunter.avalongroup.net>; from tjg@exceptionalminds.com on Wed, Mar 21, 2001 at 02:31:44PM -0800
References: <20010321143144.K24341@trufflehunter.avalongroup.net>
Message-ID: <20010321235241.D807@apone.network.loc>

Sez Timothy Grant:
> Hi again.
> 
> I actually thought I had this regex thing down pretty well, but
> I'm getting beat up by what I thought should be a simple one.
> 
> >>> x = '$1$asdfjke$lkjasdkjfsdasdfj'
> >>> re.sub(r'\$', r'\\\$', x)

One \ too many, it seems.  The second argument to re.sub is not a regex, so
you don't have to escape the $.

> '\\\\$1\\\\$asdfjke\\\\$lkjasdkjfsdasdfj'
> 
> All I want to do is substitute \$ for $.

Peace,
  Kalle
-- 
Email: kalle at gnupung.net     | You can tune a filesystem, but you
Web: http://www.gnupung.net/ | can't tune a fish. -- man tunefs(8)
PGP fingerprint: 0C56 B171 8159 327F 1824 F5DE 74D7 80D7 BF3B B1DD
 [ Not signed due to lossage.  Blame Microsoft Outlook Express. ]



From fredrik at pythonware.com  Fri Mar 16 11:59:14 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 16 Mar 2001 16:59:14 GMT
Subject: redefining stdout
References: <3AB240E4.1EB6E144@dvinc.com>
Message-ID: <CDrs6.175$sk3.51035@newsb.telia.net>

Mark Limber wrote:
> However, when I try:
>
> >>> import sys
> >>> import t # short file names = less typing!
> >>> sys.stdout = t.myPrint

>>> sys.stdout = t.myPrint()

> >>> print "hey"
> Traceback (most recent call last):
>   File "<stdin>", line 1, in ?
> TypeError: unbound method must be called with class instance
> 1st argument

Cheers /F




From loewis at informatik.hu-berlin.de  Wed Mar 14 06:05:54 2001
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 14 Mar 2001 12:05:54 +0100
Subject: This math scares me
References: <3aad4062.1290021@News.CIS.DFN.DE> <Hwbr6.1139$Tg.127571@ruti.visi.com> <3aad4485.2348499@News.CIS.DFN.DE> <jwbnews-7CC03F.19050312032001@corp.supernews.com> <yu993dcimjt8.fsf@europa.research.att.com>
Message-ID: <j41ys0mwz1.fsf@informatik.hu-berlin.de>

Andrew Koenig <ark at research.att.com> writes:

> I'm curious -- Does Python now follow either the IEEE or the more
> restrictive Scheme rules for floating-point conversion?

Somebody explained already that this is deferred to the underlying C
system. To be more specific, Python 2.0

- uses C double to represent floating point numbers

- when representing them as strings, uses the %.17g sprintf format,
  the commentary in the code says

   The repr() precision (17 significant decimal digits) is the minimal number
   that is guaranteed to have enough precision so that if the number is read
   back in the exact same binary value is recreated.  This is true for IEEE
   floating point by design, and also happens to work for all other modern
   hardware.

- uses strtod when converting a string to a double.

So, I think if you have a C99 compiler that defines __STDC_IEC_559__,
then the answer to your question is yes. In all other cases, the
answer is 'probably yes'.

Regards,
Martin


From cfelling at iae.nl  Sun Mar 11 10:37:13 2001
From: cfelling at iae.nl (Carel Fellinger)
Date: 11 Mar 2001 16:37:13 +0100
Subject: map - lambda - problem
References: <3AAB6078.EE50AE8D@aon.at>
Message-ID: <98g639$32f$1@animus.fel.iae.nl>

Gregor Lingl <aon.912502367 at aon.at> wrote:
...
> In direct mode I got:

>>>> r = []
>>>> for i in range(len(n[0])-1,-1,-1):
>         r.append(map(lambda x: x[i],n))
...
> If i put this into a function:

> def rotate(piece):
>     r = []
>     for i in range(len(piece[0])-1,-1,-1):
>         r.append(map(lambda x: x[i], piece))
>     return r
...
> Why, exactly, this behaviour. Is i now taken from some other namespace?

Bingo, you got it!

Python used to know only three scopes: local, global and builtin.
And lambda starts a new local scope, hiding rotate's local scope.
Newer versions of Python have nested function-scope, that is within
functions/lambdas scopes nest.

The work around used to be to pass the needed local vars explicitly to
the function/lambda like:

    def rotate(piece):
    	r = []
    	for i in range(len(piece[0])-1,-1,-1):
    	    r.append(map(lambda x, i=i: x[i], piece))
    	return r

You can do the same with named functions, like:

    def rotate(piece):
    	r = []
    	for i in range(len(piece[0])-1,-1,-1):
            def sub(x, i=i):
                return x[i]
    	    r.append(map(sub, piece))
    	return r

Or you could forgo that local i alltogether, like:

    def rotate(piece):
        r = apply(map, (None,) + list(piece))
        r.reverse()
        return r
-- 
groetjes, carel


From donotreply at interbulletin.bogus  Thu Mar  8 13:46:12 2001
From: donotreply at interbulletin.bogus (Rob Earl)
Date: Thu, 08 Mar 2001 18:46:12 +0000
Subject: cursor.execute fails under mxODBC
References: <mailman.983993842.26448.python-list@python.org> <tUzp6.29925$ML1.1507874@e420r-atl2.usenetserver.com>
Message-ID: <3AA7D374.7C72D4B2@interbulletin.com>

"Steve Holden" <sholden at holdenweb.com> wrote in article 
<tUzp6.29925$ML1.1507874 at e420r-atl2.usenetserver.com> : 
><bill_tolbert at bigfoot.com> wrote in message
>news:mailman.983993842.26448.python-list at python.org...
>> I just upgraded to the latest version of mxODBC. Calling execute on a
>> cursor object is hosing python. This code worked perfectly under 1.5.1
>> and version 1.1.1 of mxODBC.
>>
>> Having trouble on both Win98 and Win2000.
>>
>> I can stay on windows, so I'm gonna give ADO a try, but I'd still like
>> to know what's going on.
>>
>> Thanks,
>>
>> Bill
>>
>So would we! What error messages are you seeing, form what code?
>
>regadrs
> Steve
>
>
>
I have experienced something very similiar.  On my system running Windows 2000 the execute on a cursor object never completes.  Instead python takes 100% of the CPU and just sits there.  I haven't been able to find any logs that might help, but maybe I'm not looking in the right place.  Could you tell me where I should look?  Also, just so you are aware I'm using ActivePython not just the regular Python.

Thanks,
Rob

_______________________________________________
Submitted via WebNewsReader of http://www.interbulletin.com



From scarblac at pino.selwerd.nl  Thu Mar 29 15:24:59 2001
From: scarblac at pino.selwerd.nl (Remco Gerlich)
Date: 29 Mar 2001 20:24:59 GMT
Subject: setvbuf() ??
References: <slrn9c50mp.me7.wcuddy@lserv.ja10629.home> <3AC29961.9937EC59@ucdavis.edu>
Message-ID: <slrn9c76dv.98.scarblac@pino.selwerd.nl>

Bruce Wolk <bawolk at ucdavis.edu> wrote in comp.lang.python:
> Putting a trailing comma after the print statement eliminates the newline, but
> adds a trailing space.  Would periods separated by spaces work?
> 
> I would be curious if anyone knows how to eliminate the trailing space.

Either use sys.stdout.write() instead of print, or set sys.stdout.softspace
to 0 after the print command.

-- 
Remco Gerlich


From vlindberg at verio.net  Mon Mar 26 20:36:25 2001
From: vlindberg at verio.net (VanL)
Date: Mon, 26 Mar 2001 18:36:25 -0700
Subject: ftplib and zipfile question
Message-ID: <3ABFEE99.F5301392@verio.net>

Hello,
I am writing a program to download something from an ftp site.  I have
checked the ftplib documentation and I know that I need the retrbinary
command.  However, the docs say:

retrbinary (command, callback[, maxblocksize[, rest]]) Retrieve a file
in binary transfer mode. command should be an appropriate "RETR"
command, i.e. 'RETR filename'. The callback function is called for each
block of data received, with a single string argument giving the data
block ....

What is the callback command?  Could anyone give an example of the use
of this command to download a file and save it locally?

Further, does python have anything like the Unix pipe?  The file in
question is a zip file.  If possible, I would rather just pipe it
through to the zipfile module.

In short, is this possible?

++++++++++++++++++++++++++++++++++++++

import zlib, zipfile
from ftplib import FTP

ftp = FTP('mysite.com', 'anonymous', 'me at myaddress.com')
ftp.cwd('newdir')

myzip = Zipfile((ftp.retrbinary('binaryfile.zip', OTHER_ARGS_HERE), 'r')
for filename in myzip.namelist:
        file = open(filename, 'w')
        file.write(myzip.read(filename))

ftp.close()

+++++++++++++++++++++++++++++++++++++++++

If possible, I would like to process the file as it goes through,
without either incurring huge memory usage or having to save the file to
disk.

If I have to choose one or the other, tho, I would store the file to
disk.



From shaleh at valinux.com  Fri Mar  2 17:34:29 2001
From: shaleh at valinux.com (Sean 'Shaleh' Perry)
Date: Fri, 02 Mar 2001 14:34:29 -0800 (PST)
Subject: gah! I hate the new string syntax
In-Reply-To: <97p3sa$bpq$1@tyfon.itea.ntnu.no>
Message-ID: <XFMail.20010302143419.shaleh@valinux.com>

On 02-Mar-2001 Magnus Lie Hetland wrote:
> "Sean 'Shaleh' Perry" <shaleh at valinux.com> wrote in message
> news:mailman.983551204.27573.python-list at python.org...
> [...]
>> The point of all of this was 'join' especially,
>> but a few of the others (it
>> brother split) make for really ugly code.
>>
>> " ".split(list) is another I have seen recently.
>>
> 
> What does *this* mean, then? (AFAIK it isn't valid
> Python unless list is a character...)
> 

it was a snippet of code I saw somewhere.  It turns out the person has it
backwards (thankfully).

The actual code would be:

s = 'This is a string'
s.split(" ")

which makes some amount of sense, although I still like the string module
version better.



From sholden at holdenweb.com  Mon Mar  5 22:52:20 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Mon, 5 Mar 2001 22:52:20 -0500
Subject: PIL and Windows file formats?
References: <3AA45311.150DB844@alcyone.com>
Message-ID: <GdZo6.24467$1D5.1007863@e420r-atl1.usenetserver.com>

"Erik Max Francis" <max at alcyone.com> wrote in message
news:3AA45311.150DB844 at alcyone.com...
> I installed Python 2.0 and PIL for Python 2.0 on my machine and am
> trying to manipulate images.  I try to open a (valid) PNG image (with a
> standard .png file extension) with PIL.Image.open and get a "cannot
> identify image type" error.  I then debug through PIL.Image.py only to
> discover, much to my surprise, the only formats it seems to know about
> (in PIL.Image.ID) are GIF, JPEG, and TIFF.
>
> Any idea what in the world is going on here?
>
Nope. My PIL directory has a pngImagePlugin.py file. Does yours? Would this
make any difference?

regards
 Steve





From aahz at panix.com  Fri Mar  9 10:28:58 2001
From: aahz at panix.com (Aahz Maruch)
Date: 9 Mar 2001 07:28:58 -0800
Subject: gah! I hate the new string syntax
References: <mailman.983615170.23313.python-list@python.org> <mailman.984014353.5058.python-list@python.org> <BcCp6.37817$1D5.1582482@e420r-atl1.usenetserver.com> <98amlj01sp6@news2.newsguy.com>
Message-ID: <98asrq$645$1@panix6.panix.com>

In article <98amlj01sp6 at news2.newsguy.com>,
Alex Martelli <aleaxit at yahoo.com> wrote:
>"Steve Holden" <sholden at holdenweb.com> wrote in message
>news:BcCp6.37817$1D5.1582482 at e420r-atl1.usenetserver.com...
>    [snip]
>> > | > What about one who is both English and Scots, yet never set foot on
>> > | > that island ;-)?
>    [snip]
>> Either you were born in England, or you were born in Scotland, or you are
>> neither English nor a Scot. Part-either-or-both, maybe, but not fully
>
>Wrong!  You're talking about people whose seafaring skills and
>traditions are (deservedly) legendary.  It's perfectly acceptable
>for either an English or a Scot to be born aboard a ship...!!!

Since when are *women* allowed aboard ships?
-- 
                      --- Aahz (Copyright 2001 by aahz at pobox.com)

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

"..., and some of you may regard all women as evil traps that exist only
to tease, torture, and suck out your very soul."  --DrMax


From jwbnews at scandaroon.com  Thu Mar 15 10:39:05 2001
From: jwbnews at scandaroon.com (John W. Baxter)
Date: Thu, 15 Mar 2001 07:39:05 -0800
Subject: This math scares me
References: <20010312170356.B3513@better.net> <mailman.984480856.18097.python-list@python.org> <984486395.407379@newsmaster-04.atnet.at> <slrn9asf1d.ja3.qrczak@qrnik.zagroda> <984518199.127140@newsmaster-04.atnet.at> <jwbnews-09D75F.13313313032001@corp.supernews.com> <98ohuc$du9$1@news.udel.edu> <7bQr6.644$lz1.17686@ruti.visi.com> <98oku1$f6a$1@news.udel.edu> <erRr6.659$lz1.20662@ruti.visi.com> <mailman.984664520.26463.python-list@python.org> <HR4s6.877$lz1.73512@ruti.visi.com>
Message-ID: <jwbnews-8B4A2F.07390415032001@corp.supernews.com>

Had our Noble Ancestors believed their thumbs to be Sacred, and not to 
be used for counting or other Profane Purposes, we wouldn't be having 
these Problems.  

[Surely, another set of problems would have arisen to take their place.]

  --John

-- 
John W. Baxter   Port Ludlow, WA USA  jwbnews at scandaroon.com


From aleaxit at yahoo.com  Fri Mar 16 08:46:13 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 16 Mar 2001 14:46:13 +0100
Subject: Cookies
References: <mailman.984707839.22839.python-list@python.org> <lmq6m07j.fsf@propylon.com>
Message-ID: <98t5f60lb7@news1.newsguy.com>

"Martin Skott" <martin.skott at propylon.com> wrote in message
news:lmq6m07j.fsf at propylon.com...
    [snip]
> You should first of all be aware that there is no common unique
> identifier for a computer available from the net today[1]. IP
    [snip]
> [1]Why would somebody use cookies if there was?

Because cookies are more powerful -- they let you associate
state with, e.g., several different _users_ on a single
computer, rather than 'that computer as a whole'.  (That
was a trick question, right?-)


Alex





From dbrueck at edgix.com  Wed Mar 14 09:49:37 2001
From: dbrueck at edgix.com (Dave Brueck)
Date: Wed, 14 Mar 2001 07:49:37 -0700
Subject: "Scary Math" Solver
Message-ID: <BLEOJGEMBNBNEAKMIFDHGEDCPMAA.dbrueck@edgix.com>

Sorry if somebody already point this out but...

If it shakes your faith to know that many floating point numbers can't be
stored without slight rounding errors, in Python 2.1 you can always do this
to change how things work in interactive mode:

>>> 5.3
5.2999999999999998 # HELP! SAVE ME!!!
>>> def printstr(s):
>>>    print str(s)
>>> import sys
>>> sys.displayhook = printstr
>>> 5.3
5.3   # I BELIEVE!!!

And you could modify printstr so that other data types still use repr.

-Dave




From ixokai-nf at myNOSPAM.org  Thu Mar 29 05:34:24 2001
From: ixokai-nf at myNOSPAM.org (Ixokai)
Date: Thu, 29 Mar 2001 02:34:24 -0800
Subject: types.AnyType - describing interfaces with sequences of types
References: <mailman.985756894.12348.python-list@python.org> <mailman.985786896.14669.python-list@python.org>
Message-ID: <3ac31215$1_6@goliath.newsfeeds.com>

The reason there is no UserType is because Types are the exclusive domain
of the C side of Python. Under normal circumstances, extentions/additions of
this nature in C end up making "types", while extentions/additions in Python
end up making Classes, which are not, under normal circumstances, able to
be made in C.

This is my understanding, at least. The missing behavior that you wish is a
result of the fact that Classes and Types really don't have that much in
common
yet... Sure, some types have wrappers around them, but there is no ultimate
ancestor that is available...

The problem is, I think, is that there really is no functionality to wrap in
a
'TypeType'. And until classes and types get all combined into the same
thing,
where you'd then have an ultimate ancestor to inherit from and compare to, I
don't think you'll get a way to do what you want exactly like you want.

Then again, I may be talking out of my ass. I'm a semi newbie who just lurks
forever. :)

--Ix

(Replace 'NOSPAM' with 'seraph' to respond in email)

Carlos Alberto Reis Ribeiro wrote in message ...
>At 02:56 28/03/01 -0800, you wrote:
>>You get the idea. AnyType above will be eternally happy to be equivelent
to
>>anything you compare it to. Note that you can't do 'IntType is AnyType',
>>that tells you actual object identity. Now, if you wanted it to only show
up
>>as 'equivelent' to TypeTypes, and not say, strings, (e.g. "string" ==
>>AnyType) you could do a check to make sure what its being compared to is a
>>TypeType.....
>
>My first instinct it was to do something like this. However, it did not
>work, but because of a dumb mistake of mine :-) I tried again now, and this
>is the resulting code:
>
>class CAnyType:
>   def __cmp__(self, other):
>         return (type(other) != TypeType)
>   def __rcmp__(self, other):
>         return (type(other) != TypeType)
>
>AnyType = CAnyType()
>
>That was my fault - I was trying to compare the class to IntType, but I
>need an instance for the code to work.
>
>Anyway, I felt that should be a better way to make it, maybe be inheriting
>from some "UserType" object. However no such object exists. I don't know if
>"UserType" does not exist because it not make sense, or if it does not
>exist simply because no one else required it before. The way it is, it's
>inconsistent, because:
>
> >>> type(AnyType)        # should return "TypeType"
><type 'instance'>
>
>
>
>Carlos Ribeiro
>
>
>




-----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
-----==  Over 80,000 Newsgroups - 16 Different Servers! =-----


From shwang5 at students.uiuc.edu  Tue Mar 20 14:59:08 2001
From: shwang5 at students.uiuc.edu (seung-won hwang)
Date: Tue, 20 Mar 2001 13:59:08 -0600
Subject: Thread join example
Message-ID: <Pine.GSO.4.10.10103201358350.19653-100000@ux5.cso.uiuc.edu>

Hello,

I am looking for a code example which contains thread join. I would
appreciate your help very much!

Thanks!





From whisper at oz.net  Thu Mar 29 15:24:17 2001
From: whisper at oz.net (Dave LeBlanc)
Date: 29 Mar 2001 20:24:17 GMT
Subject: Which Python version is better on Win32
References: <20010329.3014838@mis.configured.host> <rcyw6.8737$m04.667866@e420r-atl1.usenetserver.com>
Message-ID: <9a05lh$ovk$0@216.39.170.247>

As far as I know, PythonWin isn't a full Python distribution, it's a
(large) add-on which, the last time I looked, is still available in
source from Mark Harrison's CVS. I think you can find it via either
The Vaults of Parnassus or ActiveState's site.

The only real varient I know of with respect to Python is Stackless
Python (which is supposedly about 10% faster, at least on Windows).

Dave LeBlanc

On Wed, 28 Mar 2001 22:40:27 -0500, "Steve Holden"
<sholden at holdenweb.com> wrote:

>"TM" <tm at kernelconsult.com> wrote in message
>news:20010329.3014838 at mis.configured.host...
>
>> I've seen recently that yhere were several releases of Python for the MS
>> Windows platform:
>> ActivePython is one of them there's also PythonWin which is not branded,
>> beOpenPython and maybe others.
>>
>> Which one would you advise me to use and why?
>
>BeOpen Python is the standard 2.0 release, as built by the development team
>while they were working at BeOpen (which looks, by the way, as though it
>might now BeClosed).
>
>PythonWin is actually just one component of ActiveState's ActivePython
>distribution. Unless anyone can tell you any better, it bundles BeOpen
>Python with a huge set of add-on goodies devised and produced mostly by Mark
>Hammond.
>
>IMHO you would be best advised to go for ActivePython unless you
>specifically want a small installation with just the Python language core on
>it.
>
>regards
> Steve
>
>
>



From p.herth at uni-koeln.de  Wed Mar  7 16:53:02 2001
From: p.herth at uni-koeln.de (Peter Herth)
Date: 07 Mar 2001 22:53:02 +0100
Subject: National Intruments GPIB
References: <3AA46593.753BAE1@newfocus.com> <EK6p6.2072$Xt3.397306@news1.rdc1.az.home.com>
Message-ID: <m3itlli6v5.fsf@uranus.uni-koeln.de>

"Tim Hochberg" <tim.hochberg at ieee.org> writes:

> I have a simple interface to NI44.2 GPIB that I wrote. I'll send you a copy
> via private email to avoid cluttering C.L.P. And if anyone else needs a
> copy, let me know. You're welcome to use it as a starting point.
> 
I would be very interested in it too, since I considered doing something
likewise too, getting fed up with using C for GPIB work...

Peter


From no at thank.you  Fri Mar 30 10:55:45 2001
From: no at thank.you (Owen Ap Owen)
Date: Fri, 30 Mar 2001 07:55:45 -0800
Subject: mxODBC chokes on dates! Any Help?
Message-ID: <MPG.152e4c579fa0baef9896a5@news.pacbell.net>

I'm using mxODBC from Python 2.0 under Win32. I have a functional ODBC 
connection but when I try to select data that includes date fields mxODBC 
chokes.

Here's what I'm doing.

>>> import mx.ODBC.Windows
>>> import mx.DateTime
>>> conn = mx.ODBC.Windows.connect('JTech',user='tech',password='tech')
>>> cur = conn.cursor()
>>> cur.execute('SELECT jci_num, test_date, weight, standard_25 FROM 
zwick_data')
>>> from pprint import pprint
>>> pprint(cur.description)
(('jci_num', 12, None, None, 30, 0, 0),
 ('test_date', 11, None, None, 23, 3, 1),
 ('weight', 5, None, None, 5, 0, 1),
 ('standard_25', 6, None, None, 15, 0, 1))
>>> data = cur.fetchall()
Traceback (innermost last):
  File "<interactive input>", line 1, in ?
InternalError: ('S1003', 0, '[Microsoft][ODBC Driver Manager] Program 
type out of range', 3013)

I've tried this using three seperate ODBC dsn connections. In every case 
I get a "Program type out of range" error whenever I try to select any 
data that includes a date field. 

If I connect to the ODBC connection using any other method besides mxODBC 
I can fetch the date fields without a problem.

Am I missing something obvious? Any suggestions?


From dsh8290 at rit.edu  Wed Mar  7 20:18:03 2001
From: dsh8290 at rit.edu (D-Man)
Date: Wed, 7 Mar 2001 20:18:03 -0500
Subject: gah! I hate the new string syntax
In-Reply-To: <meAp6.29961$ML1.1514835@e420r-atl2.usenetserver.com>; from sholden@holdenweb.com on Wed, Mar 07, 2001 at 07:15:48PM -0500
References: <mailman.983615170.23313.python-list@python.org> <tN5o6.2746$eH3.14913@NewsReader> <Koco6.7518$1D5.244602@e420r-atl1.usenetserver.com> <mailman.983997252.5628.python-list@python.org> <meAp6.29961$ML1.1514835@e420r-atl2.usenetserver.com>
Message-ID: <20010307201803.A277@harmony.cs.rit.edu>

On Wed, Mar 07, 2001 at 07:15:48PM -0500, Steve Holden wrote:
| "D-Man" <dsh8290 at rit.edu> wrote in message
| news:mailman.983997252.5628.python-list at python.org...
| > On Sat, Mar 03, 2001 at 03:19:05PM -0500, Steve Holden wrote:
| [ ... ]
| >
| > What about one who is both English and Scots, yet never set foot on
| > that island ;-)?
| >
| No such animal!

Oh.  Am I a bot too?  If so I must find a new maintainer so I can
match the intelligence of the other bots on the list.

;-)

-D




From sheila at spamcop.net  Mon Mar 26 08:56:45 2001
From: sheila at spamcop.net (Sheila King)
Date: Mon, 26 Mar 2001 13:56:45 GMT
Subject: How to set PYTHONHOME ???
References: <3ABF2743.ED5904F2@bevoux.grenoble.hp.com>
Message-ID: <vgiubt8a6t87ngrbs667cbjj78qehin504@4ax.com>



On Mon, 26 Mar 2001 12:25:55 +0100, Philippe Rodriguez
<Philippe_Rodriguez at bevoux.grenoble.hp.com> wrote in comp.lang.python in
article <3ABF2743.ED5904F2 at bevoux.grenoble.hp.com>:

:I'm running Python 1.6 on hpux 10.20. I got the standard depot binary
:release.
:
:Each time I'm trying to run Python, I have the following message :
:---
:Could not find platform dependent libraries <exec_prefix>
:Consider setting $PYTHONHOME to <prefix>[:<exec_prefix>]
:---
:I have set this to every conceivable option I can think of, but it still
:complains.
:The program seems to be OK, but my tool which use python runs a cron
:every minutes
: ans so I'm receiving a message every minute ... boring !

Yes, I have seen this error. It seems that the install is looking for files in
a place where you do not have them. While the best solution is a reinstall, if
you cannot do that, there is also a way to set the Python Path.

This link
http://www.aota.net/ubb/Forum3/HTML/001564-1.html

leads to a discussion board where I was seeking help, and also has links to
other places where I went for help.

--
Sheila King
http://www.thinkspot.net/sheila/
http://www.k12groups.org/




From fredrik at pythonware.com  Tue Mar 20 12:29:23 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Tue, 20 Mar 2001 17:29:23 GMT
Subject: re.match
References: <997skf$qoc$1@bmerhc5e.ca.nortel.com>
Message-ID: <TrMt6.2490$4N4.318602@newsc.telia.net>

Michael P. Soulier wrote:
>     Hey people. Could someone explain the purpose of the match function in the
> re module? If it's only difference from search is to match at the beginning of
> the line, why does it exist? That's what the ^ in the regexp is for.

no, it's not.  read the documentation again.

Cheers /F




From mda at idatar.com  Sat Mar 31 17:04:30 2001
From: mda at idatar.com (David Allen)
Date: Sat, 31 Mar 2001 22:04:30 GMT
Subject: Newbie Dictionary Question
References: <jfmx6.7861$iU.1496746@news1.rdc1.md.home.com>
Message-ID: <Ovsx6.1135$NU3.1093428@typhoon2.ba-dsg.net>

In article <jfmx6.7861$iU.1496746 at news1.rdc1.md.home.com>, "Gary Walker"
<borealis3 at home.com> wrote:

> Actually, they don't have to be ordered, so I suppose (thinking aloud here)
> that I could use a dictionary of dictionaries, but again - it seems to me
> that I'd need to dynamically create a dictionary whose name isn't known
> until runtime...

A dictionary of dictionaries doesn't sound like too
bad of an idea.  Each "key" in the toplevel dictionary
would be the name of another dictionary.  In that
way, since keys are just strings, you can have a 
bunch of dictionaries with variable names. Like:

metadict = {}  # Empty top level dictionary
user_input = get_user_input()
metadict[user_input] = {}

Now you can populate metadict[user_input] with
a bunch of keys and values.  User_input might be
something from the user or more broadly just anything
that isn't know until runtime.  (Like the value of
a config varaible or whatever)

-- 
David Allen
http://opop.nols.com/
----------------------------------------
It's that time of year again, Defcon has descended upon us. Time to venture
to SYN City to see the lights, meet strange new people, and create new
blurs we later dub memories.
	-- Attrition.org


From rdsteph at earthlink.net  Tue Mar 27 21:13:16 2001
From: rdsteph at earthlink.net (Ron Stephens)
Date: Wed, 28 Mar 2001 02:13:16 GMT
Subject: Completing the thread with latest program versions
References: <3ABCC757.6C6F17CA@earthlink.net> <3ABE73B2.CAFCA177@student.liu.se>
Message-ID: <3ABFEC14.BD3B0709@earthlink.net>

In the interest of completeness of the this thread, here is my latest
corrections and updates of my first Python program
(DecisionAnalysisBeta143.py) and a proposed GUI front end coded for me today
by another email correspondent (decisionaaron.py) in Tkinter. I have not
interfaced them yet. (Need to learn how to do that too ;-)))

    Thanks to all form my newbieness. Also don't forget to see Mr. Jeff
Elper'ss version of my idea below (he did in about 65 lines plus comments
what I took almost 400 lines to do so far ;-)))))



From jmarshal at mathworks.com  Mon Mar 19 14:51:58 2001
From: jmarshal at mathworks.com (Joshua Marshall)
Date: 19 Mar 2001 19:51:58 GMT
Subject: Python style questions
References: <98tb91$9t9$1@saltmine.radix.net> <m34rwpilxg.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <995o0u$itc$1@news.mathworks.com>

Michael Hudson <mwh21 at cam.ac.uk> wrote:
> cobrien at Radix.Net (Cary O'Brien) writes:
...
> The scoping can get a bit counter-intuitive
>> 3. I really miss not having a "switch" or "case" statement.  Sniff.

> -- 
> 3. Syntactic sugar causes cancer of the semicolon.
>   -- Alan Perlis, http://www.cs.yale.edu/homes/perlis-alan/quotes.html

A switch statement may just be syntactic sugar, but not necessarily in
a simple way.  A good compiler will compile switch statements into
jump-tables or, at worst, binary searches.  Cluttery to code by hand.

Sure this is "just" an efficiency thing, but switches can get quite big.


From cce at clarkevans.com  Thu Mar 22 05:14:25 2001
From: cce at clarkevans.com (Clark C. Evans)
Date: Thu, 22 Mar 2001 05:14:25 -0500 (EST)
Subject: PEP 1, PEP Purpose and Guidelines
In-Reply-To: <15032.59269.4520.961715@anthem.wooz.org>
Message-ID: <Pine.LNX.4.21.0103220504280.18700-100000@clarkevans.com>

Barry,

  If you don't mind, I'd like to apply for one of them
  there PEP numbers.  Sorry for not following the guidelines,
  it won't happen again.

  Also, I believe that this isn't just my work, but rather
  a first pass at concensus on this issue via the vocal and
  silent feeback from those on the main and type special
  interest group.  I hope that I have done their ideas
  and feedback justice (if not, I'm sure I'll hear about it).

Thank you so much,

Clark

...

PEP: XXX
Title: Protocol Checking and Adaptation
Version: $Revision$
Author: Clark Evans
Python-Version: 2.2
Status: Draft
Type: Standards Track
Created: 21-Mar-2001
Updated: 23-Mar-2001

Abstract

    This proposal puts forth a built-in, explicit method for
    the adaptation (including verification) of an object to a 
    context where a specific type, class, interface, or other 
    protocol is expected.  This proposal can leverage existing
    protocols such as the type system and class hierarchy and is
    orthogonal, if not complementary to the pending interface
    mechanism [1] and signature based type-checking system [2]

    This proposal allows an object to answer two questions.  First,
    are you a such and such?  Meaning, does this object have a 
    particular required behavior?  And second, if not, can you give
    me a handle which is?  Meaning, can the object construct an 
    appropriate wrapper object which can provide compliance with
    the protocol expected.  This proposal does not limit what 
    such and such (the protocol) is or what compliance to that
    protocol means, and it allows other query/adapter techniques 
    to be added later and utilized through the same interface 
    and infrastructure introduced here.

Motivation

    Currently there is no standardized mechanism in Python for 
    asking if an object supports a particular protocol. Typically,
    existence of particular methods, particularly those that are 
    built-in such as __getitem__, is used as an indicator of 
    support for a particular protocol.  This technique works for 
    protocols blessed by GvR, such as the new enumerator proposal
    identified by a new built-in __iter__.  However, this technique
    does not admit an infallible way to identify interfaces lacking 
    a unique, built-in signature method.

    More so, there is no standardized way to obtain an adapter 
    for an object.  Typically, with objects passed to a context
    expecting a particular protocol, either the object knows about 
    the context and provides its own wrapper or the context knows 
    about the object and automatically wraps it appropriately.  The 
    problem with this approach is that such adaptations are one-offs,
    are not centralized in a single place of the users code, and 
    are not executed with a common technique, etc.  This lack of
    standardization increases code duplication with the same 
    adapter occurring in more than one place or it encourages 
    classes to be re-written instead of adapted.  In both cases,
    maintainability suffers.

    In the recent type special interest group discussion [3], there
    were two complementary quotes which motivated this proposal:

       "The deep(er) part is whether the object passed in thinks of
        itself as implementing the Foo interface. This means that
        its author has (presumably) spent at least a little time
        about the invariants that a Foo should obey."  GvR [4]

    and

       "There is no concept of asking an object which interface it
        implements. There is no "the" interface it implements. It's
        not even a set of interfaces, because the object doesn't 
        know them in advance. Interfaces can be defined after objects
        conforming to them are created." -- Marcin Kowalczyk [5]

    The first quote focuses on the intent of a class, including 
    not only the existence of particular methods, but more 
    importantly the call sequence, behavior, and other invariants.
    Where the second quote focuses on the type signature of the
    class.  These quotes highlight a distinction between interface
    as a "declarative, I am a such-and-such" construct, as opposed
    to a "descriptive, It looks like a such-and-such" mechanism.

    Four positive cases for code-reuse include:

     a) It is obvious object has the same protocol that
        the context expects.  This occurs when the type or
        class expected happens to be the type of the object
        or class.  This is the simple and easiest case.

     b) When the object knows about the protocol that the
        context requires and knows how to adapt itself 
        appropriately.  Perhaps it already has the methods
        required, or it can make an appropriate wrapper

     c) When the protocol knows about the object and can
        adapt it on behalf of the context.  This is often
        the case with backwards-compatibility cases.

     d) When the context knows about the object and the 
        protocol and knows how to adapt the object so that
        the required protocol is satisfied.

    This proposal should allow each of these cases to be handled,
    however, the proposal only concentrates on the first two cases;
    leaving the latter two cases where the protocol adapts the 
    object and where the context adapts the object to other proposals.
    Furthermore, this proposal attempts to enable these four cases
    in a manner completely neutral to type checking or interface
    declaration and enforcement proposals.  

Specification

    For the purposes of this specification, let the word protocol
    signify any current or future method of stating requirements of 
    an object be it through type checking, class membership, interface 
    examination, explicit types, etc.  Also let the word compliance
    be dependent and defined by each specific protocol.

    This proposal initially supports one initial protocol, the
    type/class membership as defined by isinstance(object,protocol)
    Other types of protocols, such as interfaces can be added through
    another proposal without loss of generality of this proposal.  
    This proposal attempts to keep the first set of protocols small
    and relatively unobjectionable.

    This proposal would introduce a new binary operator "isa".
    The left hand side of this operator is the object to be checked
    ("self"), and the right hand side is the protocol to check this
    object against ("protocol").  The return value of the operator 
    will be either the left hand side if the object complies with 
    the protocol or None.

    Given an object and a protocol, the adaptation of the object is:
     a) self, if the object is already compliant with the protocol,
     b) a secondary object ("wrapper"), which provides a view of the
        object compliant with the protocol.  This is explicitly 
        vague, and wrappers are allowed to maintain their own 
        state as necessary.
     c) None, if the protocol is not understood, or if object 
        cannot be verified compliant with the protocol and/or
        if an appropriate wrapper cannot be constructed.

    Further, a new built-in function, adapt, is introduced.  This
    function takes two arguments, the object being adapted ("obj") 
    and the protocol requested of the object ("protocol").  This
    function returns the adaptation of the object for the protocol,
    either self, a wrapper, or None depending upon the circumstances.
    None may be returned if adapt does not understand the protocol,
    or if adapt cannot verify compliance or create a wrapper.

    For this machinery to work, two other components are required.
    First is a private, shared implementation of the adapt function
    and isa operator.  This private routine will have three 
    arguments: the object being adapted ("self"), the protocol 
    requested ("protocol"), and a flag ("can_wrap").  The flag
    specifies if the adaptation may be a wrapper, if the flag is not
    set, then the adaptation may only be self or None.  This flag is
    required to support the isa operator.  The obvious case 
    mentioned in the motivation, where the object easily complies 
    with the protocol, is implemented in this private routine.  

    To enable the second case mentioned in the motivation, when 
    the object knows about the protocol, a new method slot, __adapt__
    on each object is required.  This optional slot takes three
    arguments, the object being adapted ("self"), the protocol 
    requested ("protocol"), and a flag ("can_wrap").  And, like 
    the other functions, must return an adaptation, be it self, a
    wrapper if allowed, or None.  This method slot allows a class 
    to declare which protocols it supports in addition to those 
    which are part of the obvious case.

    This slot is called first before the obvious cases are examined, 
    if None is returned then the default processing proceeds.  If the
    default processing is wrong, then the AdaptForceNoneException
    can be thrown.  The private routine will catch this specific 
    exception and return None in this case.  This technique allows an
    class to subclass another class, but yet catch the cases where 
    it is considered as a substitutable for the base class.  Since 
    this is the exception, rather than the normal case, an exception 
    is warranted and is used to pass this information along.  The 
    caller of adapt or isa will be unaware of this particular exception
    as the private routine will return None in this particular case.

    Please note two important things.  First, this proposal does not
    preclude the addition of other protocols.  Second, this proposal 
    does not preclude other possible cases where adapter pattern may
    hold, such as the protocol knowing the object or the context 
    knowing the object and the protocol (cases c and d in the 
    motivation).  In fact, this proposal opens the gate for these 
    other mechanisms to be added; while keeping the change in 
    manageable chunks.

Reference Implementation and Example Usage

    -----------------------------------------------------------------
    adapter.py
    -----------------------------------------------------------------
        import types
        AdaptForceNoneException = "(private error for adapt and isa)"

        def interal_adapt(obj,protocol,can_wrap):

            # the obj may have the answer, so ask it about the ident
            adapt = getattr(obj, '__adapt__',None)
            if adapt:
                try:
                    retval = adapt(protocol,can_wrap)
                    # todo: if not can_wrap check retval for None or obj
                except AdaptForceNoneException:
                    return None
                if retval: return retval

            # the protocol may have the answer, so ask it about the obj
            pass

            # the context may have the answer, so ask it about the
            pass

            # check to see if the current object is ok as is
            if type(protocol) is types.TypeType or \
               type(protocol) is types.ClassType:
                if isinstance(obj,protocol):
                    return obj

            # ok... nothing matched, so return None
            return None

        def adapt(obj,protocol):
            return interal_adapt(obj,protocol,1)

        # imagine binary operator syntax
        def isa(obj,protocol):
            return interal_adapt(obj,protocol,0)

    -----------------------------------------------------------------
    test.py
    -----------------------------------------------------------------
        from adapter import adapt
        from adapter import isa
        from adapter import AdaptForceNoneException

        class KnightsWhoSayNi: pass  # shrubbry troubles

        class EggsOnly:  # an unrelated class/interface
            def eggs(self,str): print "eggs!" + str

        class HamOnly:  # used as an interface, no inhertance
            def ham(self,str): pass
            def _bugger(self): pass  # irritating a private member

        class SpamOnly: # a base class, inheritance used
            def spam(self,str): print "spam!" + str

        class EggsSpamAndHam (SpamOnly,KnightsWhoSayNi):
            def ham(self,str): print "ham!" + str
            def __adapt__(self,protocol,can_wrap):
                if protocol is HamOnly:
                    # implements HamOnly implicitly, no _bugger
                    return self
                if protocol is KnightsWhoSayNi:
                    # we are no longer the Knights who say Ni!
                    raise AdaptForceNoneException
                if protocol is EggsOnly and can_wrap:
                    # Knows how to create the eggs!
                    return EggsOnly()

        def test():
            x = EggsSpamAndHam()
            adapt(x,SpamOnly).spam("Ni!")
            adapt(x,EggsOnly).eggs("Ni!")
            adapt(x,HamOnly).ham("Ni!")
            adapt(x,EggsSpamAndHam).ham("Ni!")
            if None is adapt(x,KnightsWhoSayNi): print "IckIcky...!"
            if isa(x,SpamOnly): print "SpamOnly"
            if isa(x,EggsOnly): print "EggsOnly"
            if isa(x,HamOnly): print "HamOnly"
            if isa(x,EggsSpamAndHam): print "EggsAndSpam"
            if isa(x,KnightsWhoSayNi): print "NightsWhoSayNi"

    -----------------------------------------------------------------
    Example Run
    -----------------------------------------------------------------
        >>> import test
        >>> test.test()
        spam!Ni!
        eggs!Ni!
        ham!Ni!
        ham!Ni!
        IckIcky...!
        SpamOnly
        HamOnly
        EggsAndSpam

Relationship To Paul Prescod and Tim Hochbergs Type Assertion method

    The example syntax Paul put forth recently [2] was:

        interface Interface
            def __check__(self,obj)

    Pauls proposal adds the checking part to the third (3)
    case described in motiviation, when the protocol knows
    about the object.  As stated, this could be easily added
    as a step in the interal_adapt function:

            # the protocol may have the answer, so ask it about the obj

                if typ is types.Interface:
                    if typ__check__(obj):
                        return obj

    Further, and quite excitingly, if the syntax for this type 
    based assertion added an extra argument, "can_wrap", then this
    mechanism could be overloaded to also provide adapters to
    objects that the interface knows about.

    In short, the work put forth by Paul and company is great, and
    I dont see any problems why these two proposals couldnt work
    together in harmony, if not be completely complementary.

Relationship to Python Interfaces [1] by Michel Pelletier

    The relationship to this proposal is a bit less clear 
    to me, although an implements(obj,anInterface) built-in
    function was mentioned.  Thus, this could be added naively
    as a step in the interal_adapt function:

        if typ is types.Interface:
            if implements(obj,protocol):
                return obj

    However, there is a clear concern here.  Due to the 
    tight semantics being described in this specification,
    it is clear the isa operator proposed would have to have 
    a 1-1 correspondence with implements function, when the
    type of protocol is an Interface.  Thus, when can_wrap is
    true, __adapt__ may be called, however, it is clear that
    the return value would have to be double-checked.  Thus, 
    a more realistic change would be more like:

        def internal_interface_adapt(obj,interface)
            if implements(obj,interface):
                return obj
            else
                return None

        def interal_adapt(obj,protocol,can_wrap):

            # the obj may have the answer, so ask it about the ident
            adapt = getattr(obj, '__adapt__',None)
            if adapt:
                try:
                    retval = adapt(protocol,can_wrap)
                except AdaptForceNoneException:
                    if type(protocol) is types.Interface:
                        return internal_interface_adapt(obj,protocol)
                    else:
                        return None
                if retval: 
                    if type(protocol) is types.Interface:
                        if can_wrap and implements(retval,protocol):
                            return retval
                        return internal_interface_adapt(obj,protocol)
                    else:
                        return retval

            if type(protocol) is types.Interface:
                return internal_interface_adapt(obj,protocol)

            # remainder of function... 

    It is significantly more complicated, but doable.

Relationship To Iterator Proposal:
 
    The iterator special interest group is proposing a new built-in
    called "__iter__", which could be replaced with __adapt__ if an
    an Interator class is introduced.  Following is an example.

        class Iterator:
            def next(self):
                raise IndexError

        class IteratorTest:
            def __init__(self,max):
                self.max = max
            def __adapt__(self,protocol,can_wrap):
                if protocol is Iterator and can_wrap:
                    class IteratorTestIterator(Iterator):
                        def __init__(self,max):
                            self.max = max
                            self.count = 0
                        def next(self):
                            self.count = self.count + 1
                            if self.count < self.max:
                              return self.count
                            return Iterator.next(self)
                    return IteratorTestIterator(self.max)

Relationships To Microsofts Query Interface:

    Although this proposal may sounds similar to Microsofts 
    QueryInterface, it differs by a number of aspects.  First, 
    there is not a special "IUnknown" interface which can be used
    for object identity, although this could be proposed as one
    of those "special" blessed interface protocol identifiers.
    Second, with QueryInterface, once an object supports a particular
    interface it must always there after support this interface; 
    this proposal makes no such guarantee, although this may be 
    added at a later time. Third, implementations of Microsofts
    QueryInterface must support a kind of equivalence relation. 
    By reflexive they mean the querying an interface for itself 
    must always succeed.  By symmetrical they mean that if one 
    can successfully query an interface IA for a second interface 
    IB, then one must also be able to successfully query the 
    interface IB for IA.  And finally, by transitive they mean if 
    one can successfully query IA for IB and one can successfully
    query IB for IC, then one must be able to successfully query 
    IA for IC.  Ability to support this type of equivalence relation
    should be encouraged, but may not be possible.  Further research 
    on this topic (by someone familiar with Microsoft COM) would be
    helpful in further determining how compatible this proposal is.

Backwards Compatibility

    There should be no problem with backwards compatibility.  
    Indeed this proposal, save an built-in adapt() function, 
    could be tested without changes to the interpreter.

Questions and Answers

    Q:  Why was the name changed from __query__ to __adapt__ ?  

    A:  It was clear that significant QueryInterface assumptions were
        being laid upon the proposal, when the intent was more of an 
        adapter.  Of course, if an object does not need to be adapted
        then it can be used directly and this is the basic premise.

    Q:  Why is the checking mechansim mixed with the adapter
        mechanism.

    A:  Good question.  They could be seperated, however, there
        is significant overlap, if you consider the checking
        protocol as returning a compliant object (self) or
        not a compliant object (None).  In this way, adapting
        becomes a special case of checking, via the can_wrap.

        Really, this could be seperated out, but the two 
        concepts are very related so much duplicate work
        would be done, and the overall mechanism would feel
        quite a bit less unified.

    Q:  This is just a type-coercion proposal.

    A:  No. Certainly it could be used for type-coercion, such
        coercion would be explicit via __adapt__ or adapt function. 
        Of course, if this was used for iterator interface, then the
        for construct may do an implicit __adapt__(Iterator) but
        this would be an exception rather than the rule.

    Q:  Why did the author write this PEP?

    A:  He wanted a simple proposal that covered the "deep part" of
        interfaces without getting tied up in signature woes.  Also, it
        was clear that __iter__ proposal put forth is just an example
        of this type of interface.  Further, the author is doing XML 
        based client server work, and wants to write generic tree based
        algorithms that work on particular interfaces and would
        like these algorithms to be used by anyone willing to make
        an "adapter" having the interface required by the algorithm.

    Q:  Is this in opposition to the type special interest group?

    A:  No.  It is meant as a simple, need based solution that could
        easily complement the efforts by that group.

    Q:  Why was the identifier changed from a string to a class?

    A:  This was done on Michel Pelletiers suggestion.  This mechanism
        appears to be much cleaner than the DNS string proposal, which 
        caused a few eyebrows to rise.  

    Q:  Why not handle the case where instances are used to identify 
        protocols?  In other words, 6 isa 6 (where the 6 on the right
        is promoted to an types.Int

    A:  Sounds like someone might object, lets keep this in a
        separate proposal.

    Q:  Why not let obj isa obj be true?  or class isa baseclass?

    A:  Sounds like someone might object, lets keep this in a
        separate proposal.

    Q:  It seems that a reverse lookup could be used, why not add this?

    A:  There are many other lookup and/or checking mechanisms that
        could be used here.  However, the goal of this PEP is to be 
        small and sweet ... having any more functionality would make
        it more objectionable to some people.  However, this proposal
        was designed in large part to be completely orthogonal to other
        methods, so these mechanisms can be added later if needed

Credits

    This proposal was created in large part by the feedback 
    of the talented individuals on both the main mailing list
    and also the type signature list.  Specific contributors
    include (sorry if I missed someone).

        Robin Thomas, Paul Prescod, Michel Pelletier, 
        Alex Martelli, Jeremy Hylton, Carlos Ribeiro,
        Aahz Maruch, Fredrik Lundh, Rainer Deyke,
        Timothy Delaney, and Huaiyu Zhu

Copyright

    This document has been placed in the public domain.


References and Footnotes

    [1] http://python.sourceforge.net/peps/pep-0245.html
    [2] http://mail.python.org/pipermail/types-sig/2001-March/001223.html
    [3] http://www.zope.org/Members/michel/types-sig/TreasureTrove
    [4] http://mail.python.org/pipermail/types-sig/2001-March/001105.html
    [5] http://mail.python.org/pipermail/types-sig/2001-March/001206.html
    [6] http://mail.python.org/pipermail/types-sig/2001-March/001223.html





From greg at cosc.canterbury.ac.nz  Sun Mar  4 21:15:44 2001
From: greg at cosc.canterbury.ac.nz (Greg Ewing)
Date: Mon, 05 Mar 2001 15:15:44 +1300
Subject: gah! I hate the new string syntax
References: <mailman.983485923.1136.python-list@python.org> <97p7iq$kk2$1@nntp6.u.washington.edu> <97soak$qh9$1@nntp.Stanford.EDU>
Message-ID: <3AA2F6D0.5BF0268@cosc.canterbury.ac.nz>

Amit Patel wrote:
> 
> But certainly not capitalize or center.
> They feel like a user definable function that are in the string module
> for convenience.  At least in my mind, capitalize is NOT part of the
> basic functionality of a string class.

I second that. Capitalising, splitting, joining, etc. are
things that you do *to* strings, not things that strings
themselves do.

Another thing I'd like to mention is that, by keeping
as many things as possible *not* being methods, the modularity
of the core Python system is improved.

Suppose one day I wanted to make a cut-down version of
the Python interpreter for some specialised application
in which string processing wasn't going to be of much
importance. I might decide that I wanted to remove
most of the standard string-processing functions to
keep the size down.

If the string functions live in a separate module, I can
just recompile Python with that module disabled or replace
it with my own cut-down version.

But if they're methods of the string object, I have to
do surgery on one of the most basic built-in types, which
is a much hairier proposition.

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


From mitchchapman at earthlink.net  Sat Mar 17 11:13:14 2001
From: mitchchapman at earthlink.net (Mitch Chapman)
Date: Sat, 17 Mar 2001 16:13:14 GMT
Subject: Thoughts about extensions to the Python language
References: <mailman.984608060.6280.python-list@python.org> <3AB2FA0F.E61862B0@seebelow.org> <Pine.LNX.4.21.0103170930420.26659-100000@kev.ioc.ee>
Message-ID: <3AB38DE8.788C86D6@earthlink.net>

Pearu Peterson wrote:
> On Fri, 16 Mar 2001, Grant Griffin wrote:
> > Still, one can write FORTRAN in any language.  (But it takes so much
> > work to write FORTRAN in Python that lazy folks don't bother.)
> 
> What do you mean in the parenthesis? Wrapping Fortran programs to Python
> is now even easier than wrapping C programs. 

I think he meant "write Python code with the mindset of a
Fortran programmer."  This would entail lots of global variables, 
humongous function bodies and modules, etc.  (Dunno if this still holds; 
I'm basing it on the 20-30 year old code I used to maintain.)

--
Mitch


From aleaxit at yahoo.com  Thu Mar 22 03:24:52 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 22 Mar 2001 09:24:52 +0100
Subject: How to increase PythonWin v2.0: COM browser limits?
References: <994psh$so5$1@news1.xs4all.nl> <uy9u06vji.fsf@roundpoint.com> <3AB9444E.5040403@ActiveState.com>
Message-ID: <99ccsk015b7@news2.newsguy.com>

"Mark Hammond" <MarkH at ActiveState.com> wrote in message
news:3AB9444E.5040403 at ActiveState.com...
    [snip]
> Ahh - excellent.  Thats for tracking that down - it has always worked for
me.

Me too -- the categories part.  The one that breaks down is
the registered-libraries part (of combrowse.py).

The fix is in HLIHeadingRegisterdTypeLibs -- as the error
is an UnboundLocalError on 'name' at the line:

ret.append(HLIRegisteredTypeLibrary((keyName, versionStr), name))

I just set name to None right before the try/finally, and
guarded this specific line with an 'if name is not None:';
it now works fine (so I haven't looked deeper at WHAT was
causing the wrong behavior -- no doubt some craziness in
my registry, since other tools trying to systematically
parse it are also prone to funky breakdowns).


Alex





From bckfnn at worldonline.dk  Wed Mar 14 10:33:12 2001
From: bckfnn at worldonline.dk (Finn Bock)
Date: Wed, 14 Mar 2001 15:33:12 GMT
Subject: ANN: First alpha release of Jython-2.1
Message-ID: <3aaf8b98.20888005@smtp.worldonline.dk>

I am happy to announce the first alpha release of Jython 2.1.

Jython is a Java implementation of the Python programming
language.  It allows users to compile Python source code to
Java byte codes, and run the resulting bytecodes on any Java
Virtual Machine.  It is a very seamless and smooth
integration with Java:  from Python you have complete access
to all Java libraries, can build applets, can integrate with
Java beans, and can subclass Java classes in Python and vice
versa.  Like Python, and unlike Java, Jython can also be
used interactively:  just type some Jython code at the
prompt and see the results immediately.

A java installer is available for download at the Jython website:

    http://www.jython.org/

Installation of version 2.1a1 is similar to version 2.0. Further 
information and tips on installation is available at:

    http://www.jython.org/install.html

Jython 2.1 aims to be feature compatible with Python 2.1 and among
the new feature are:

  - Nested scopes
  - The warning framework
  - Rich comparison
  - The numeric coercion model have changed.
  - Function attributes.

The changes from Jython-2.0 include:

  New features.
   - Improve speed when indexing a string and iterating over a string
     in a for loop.
   - Reworked coercing model (pep-0208)
   - Added {}.popitem().
   - Improved speed for python files by buffering the RandomAccessFile.
   - Added function attributes (pep-0232)
   - Rich comparison (pep-0207)
   - Updated cPickle to handle CPython compatible unicode strings.
   - Updated sre to CPython-2.1a1
   - Added the "new" module
   - Added a PyServlet class to the util package.
   - Warning framework (pep-0230).
   - Added sys.add_classdir and sys.add_extdir.
   - Added nested scope to the interpreter (pep-0227).

  Bug fixes.
   - Allow self referencing adapters to be collected when using weak
     tables.
   - Added workaround for a reader bug in MRJ22Jitc.01.
   - Fixes #127340 where serializable parameter received a PyObject
     instead of the wrapped java object.
   - Include a LineNumberTable in the compiled class files.
   - Fixed a java version test bug in the installer.

A complete list of changes and differences are available here:

  http://www.jython.org/NEWS.html

Bugs can be reported to the bug manager on SourceForge:

  http://sourceforge.net/bugs/?group_id=12867

Cheers,
the jython-developers



From sheila at spamcop.net  Sun Mar 18 04:33:57 2001
From: sheila at spamcop.net (Sheila King)
Date: Sun, 18 Mar 2001 09:33:57 GMT
Subject: Help with 500 error on cgi script
Message-ID: <quv8bto7r8s6f1tckc7qpmuir0kvufp1nb@4ax.com>

OK, I'm at my wits end.

I have Apache running on my local Win98 machine, and I have one simple python
cgi script, equivalent to "Hello, World!", that I have been able to run.

Today I thought I would start to play with the cgi module and start doing some
forms and so forth.

Well, I was getting 500 internal server errors. I've checked everything. The
path, etc...

So, finally, I tried just running cgi.test(), as recommended in the Python
tutorial. It worked splendidly.

I also have the Python Core Programming book, from Wesley Chun, that comes
with a CD-ROM and I put up his friends.htm and friends1.py files from the CD
onto my local site, and it ran just fine, also.

I've compared my file to the other two files and just can't see what I'm doing
wrong.

So, here is my script. I'm sure someone here will see it right off?

--
Sheila King
http://www.thinkspot.net/sheila/
http://www.k12groups.org/


#!/usr/bin/env python


import cgi

form = cgi.FieldStorage()
form_ok = 0
    
if form.has_key("subscriberEmail") and form.has_key("command"):
    form_ok = 1
if not form_ok:
	output = '''Content-type: text/html\n\n
<html>
<Head>
<Title>Error in Form</Title>
</Head>\n
<Body>\n
<H1>Error</H1>
You have not filled out all necessary information to complete this request
You <b>MUST</b> fill in your subscriber e-mail in the From: box
and you <b>MUST</b> select some action to modify your delivery options.
Please use the back button on your browser to return to the form and fill
in the required information
</body>\n</html>
'''
	print output
else:
	for k in form.keys():
    	print k, " : ", form[k]
	print "All done!"




From aahz at panix.com  Tue Mar 20 10:44:18 2001
From: aahz at panix.com (Aahz Maruch)
Date: 20 Mar 2001 07:44:18 -0800
Subject: The most important modules
References: <Pine.GSO.4.21.0103191732210.22234-100000@laplace>
Message-ID: <997tsi$q91$1@panix3.panix.com>

In article <Pine.GSO.4.21.0103191732210.22234-100000 at laplace>,
Martyn Quick  <mrq at for.mat.bham.ac.uk> wrote:
>
>I've seen a few comments along the lines of "When you are a newbie it will
>take a while to realize which the important modules are."  As someone who
>is in the middle of wading through Chapter 3 of the Library Reference, it
>would be very helpful to know which ones are particularly worth
>noting.  (Presumably  sys, (c)Pickle and shelve, from the ones I've so far
>in the chapter - but what other treats are out there for me?)

Here's my list:

sys
string
re
os
os.path
shutil
zlib/gzip/zipfile
urllib
md5
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

There's a difference between a person who gets shit zie doesn't deserve
and a person who gets more shit than zie deserves.  --Aahz


From bkline at rksystems.com  Tue Mar 27 15:57:04 2001
From: bkline at rksystems.com (Bob Kline)
Date: Tue, 27 Mar 2001 15:57:04 -0500 (EST)
Subject: Changing every other instance of <B> in a file
In-Reply-To: <3e71cto4blbb71qdpp7v3h1kn9lousaq78@4ax.com>
Message-ID: <Pine.LNX.4.10.10103271555120.27878-100000@rksystems.com>

On Tue, 27 Mar 2001, Lars Kl?boe wrote:

> On Tue, 27 Mar 2001 05:56:26 -0800, "Emile van Sebille" <emile at fenx.com>
> wrote:
> >Here's one way (although I'm interested to see how re might do this):
> 
> [--- interresting code ---]
> 
> This is another (fancier ;)) implementation of the solution I've
> been working on, but I still refuse to believe that the script has
> to be this elaborate/big. *twisting my tired gray matter*

In the spirit of using the best tool for each job, why not use lex for
this?

%s GOT_B
%%
<GOT_B>\<B\>    { printf("</B>"); BEGIN(0); }
\<B\>           { printf("<B>"); BEGIN(GOT_B); }
%%
main()          { yylex(); }
yywrap()        { return 1; }

-- 
Bob Kline
mailto:bkline at rksystems.com
http://www.rksystems.com




From guido at digicool.com  Fri Mar  2 17:46:09 2001
From: guido at digicool.com (Guido van Rossum)
Date: Fri, 02 Mar 2001 17:46:09 -0500
Subject: Python 2.1b1 released
Message-ID: <200103022246.RAA18529@cj20424-a.reston1.va.home.com>

With great pleasure I announce the release of Python 2.1b1.  This is a
big step towards the release of Python 2.1; the final release is
expected to take place in mid April.

Find out all about 2.1b1, including docs and downloads (Windows
installer and source tarball), at the 2.1 release page:

    http://www.python.org/2.1/


WHAT'S NEW?
-----------

For the big picture, see Andrew Kuchling's What New in Python 2.1:

    http://www.amk.ca/python/2.1/

For more detailed release notes, see SourceForge:

    http://sourceforge.net/project/shownotes.php?release_id=25924

The big news since 2.1a2 was released a month ago:

- Nested Scopes (PEP 227)[*] are now optional.  They must be enabled
  by including the statement "from __future__ import nested_scopes" at
  the beginning of a module (PEP 236).  Nested scopes will be a
  standard feature in Python 2.2.

- Compile-time warnings are now generated for a number of conditions
  that will break or change in meaning when nested scopes are enabled.

- The new tool *pydoc* displays module documentation, extracted from
  doc strings.  It works in a text environment as well as in a GUI
  environment (where it cooperates with a web browser).  On Windows,
  this is in the Start menu as "Module Docs".

- Case-sensitive import.  On systems with case-insensitive but
  case-preserving file systems, such as Windows (including Cygwin) and
  MacOS, import now continues to search the next directory on sys.path
  when a case mismatch is detected.  See PEP 235 for the full scoop.

- New platforms.  Python 2.1 now fully supports MacOS X, Cygwin, and
  RISCOS.

[*] For PEPs (Python Enhancement Proposals), see the PEP index:

    http://python.sourceforge.net/peps/

I hope to see you all next week at the Python9 conference in Long
Beach, CA:

    http://www.python9.org

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



From mix77 at usa.net  Thu Mar 29 10:58:46 2001
From: mix77 at usa.net (Mix)
Date: Thu, 29 Mar 2001 17:58:46 +0200
Subject: Printing  "%"
Message-ID: <99vm4k$22l$1@ctb-nnrp2.saix.net>

How do you print a percenrage (%)sign in Python?
I tried :
  return "select * from fax where id LIKE '%s\%'" %ident

I get the following error
 File "/home/mixo/faxserv/cgi-bin/OutputStr.py", line 136, in GetFaxPages
    return "select * from fax where id LIKE '%s\%'" %ident
TypeError: not enough arguments for format string

Please help!




From thomas.heller at ion-tof.com  Thu Mar  8 10:43:30 2001
From: thomas.heller at ion-tof.com (Thomas Heller)
Date: Thu, 8 Mar 2001 16:43:30 +0100
Subject: Finding the TCL directory
Message-ID: <9889bb$mtr9$1@ID-59885.news.dfncis.de>

Hi,

I would like to update py2exe to fully support
Tkinter programs.
Fortunately, it's nearly done, the only thing
remaining to do is to copy the TCL installation
into the dist directory. But how to find it?

Looking into lib-tk\FixTk.py, I have some
impression how to do it.
But, isn't there an easier way than just to
copy all the code?

It should work for python 1.5.2, 2.0, ...

Thomas




From danielk at aracnet.com  Mon Mar 19 09:50:15 2001
From: danielk at aracnet.com (Daniel Klein)
Date: Mon, 19 Mar 2001 06:50:15 -0800
Subject: The order of classes
References: <9r3cbt04tg3c3li151416n9b2a0gg3eccb@4ax.com> <slrn9bc4fc.95q.scarblac@pino.selwerd.nl>
Message-ID: <a27cbt0aj8oqv3p85i00bnvbmb4592igu8@4ax.com>

On 19 Mar 2001 14:06:31 GMT, scarblac at pino.selwerd.nl (Remco Gerlich) wrote:

>Don't use default arguments like this. You do realize, that even if you put
>the definition of foo before bar, that every instance of bar will have 
>*the same foo instance* as its default argument? The 'f = foo()' happens
>only once, namely when the class is defined.
>
>Try something like this:
>
>class bar:
>   def __init__(self, f = None):
>      if f == None:
>         f = foo()
>
>class foo:
>   pass
>   
>
>Now the 'f = foo()' happens when a bar instance is created, and at that
>time, foo already exists.

Excellent answer Remco and it makes a heap of sense! I am already recoding this
as you advise.

Thanks,
Dan


From grante at visi.com  Wed Mar 14 16:31:54 2001
From: grante at visi.com (Grant Edwards)
Date: Wed, 14 Mar 2001 21:31:54 GMT
Subject: This math scares me
References: <3aad4062.1290021@News.CIS.DFN.DE> <20010312170356.B3513@better.net> <mailman.984480856.18097.python-list@python.org> <984486395.407379@newsmaster-04.atnet.at> <slrn9asf1d.ja3.qrczak@qrnik.zagroda> <984518199.127140@newsmaster-04.atnet.at> <jwbnews-09D75F.13313313032001@corp.supernews.com> <98ohuc$du9$1@news.udel.edu> <7bQr6.644$lz1.17686@ruti.visi.com> <98oku1$f6a$1@news.udel.edu>
Message-ID: <erRr6.659$lz1.20662@ruti.visi.com>

In article <98oku1$f6a$1 at news.udel.edu>, Terry Reedy wrote:
>
>"Grant Edwards" <grante at visi.com> wrote in message
>news:7bQr6.644$lz1.17686 at ruti.visi.com...
>> In article <98ohuc$du9$1 at news.udel.edu>, Terry Reedy wrote:
>>
>> >> Bill Gates got this right in the early days, delivering two
>> >> Basic interpreters: binary floating point for speed, and
>> >> decimal floating point for money and such
>> >
>> >If one counts pennies instead of dollars, then 32 bit ints (+-
>> >$20 million) will serve for many purposes.  Exact, faster, and
>> >more widely available than BCD
>>
>> In fiancial calculations you also need to represent numbers
>> that aren't amounts of money.  Interest rates for example.
>
>So?  

So 32 bit integers with a weighting of one penny per count
won't work for that.

>Floats (double precision) work well enough for this, I believe.

No.  The binary floating point representation of many decimal
fraction values is not exact.  That's what started this whole
thread.  You can't represent 0.1 exactly in binary FP.  You can
in BCD.

>Note that binary fraction stock prices are exact with floats
>and not with BCD.

Wrong.

1/16 is exactly represented in BCD as 0.0625.  1/32 is 0.03125,
and so on.  Some other fractions (e.g. 1/3) are not exactly
representable by BCD, but all the fractions used on exchanges
are all powers of 1/2 and can be represented exactly in BCD.

I believe that there are plans (on some exchanges) to switch
pricing from base-2 fractions to decimal fractions.

-- 
Grant Edwards                   grante             Yow!  Now I can join WEIGHT
                                  at               WATCHERS!
                               visi.com            


From stephen_purcell at yahoo.com  Thu Mar 29 11:31:28 2001
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Thu, 29 Mar 2001 18:31:28 +0200
Subject: Printing  "%"
In-Reply-To: <99vm4k$22l$1@ctb-nnrp2.saix.net>; from mix77@usa.net on Thu, Mar 29, 2001 at 05:58:46PM +0200
References: <99vm4k$22l$1@ctb-nnrp2.saix.net>
Message-ID: <20010329183128.B989@freedom.puma-ag.com>

Mix wrote:
> How do you print a percenrage (%)sign in Python?
> I tried :
>   return "select * from fax where id LIKE '%s\%'" %ident

Do the following:

    >>> "%i%%" % 85
    '85%'

Or, in your case,

    return "select * from fax where id LIKE '%s%%'" %ident

-Steve


-- 
Steve Purcell, Pythangelist
Get testing at http://pyunit.sourceforge.net/
Any opinions expressed herein are my own and not necessarily those of Yahoo



From dav-nospam-id at westco-nospam-ntrol.com  Thu Mar 22 10:59:00 2001
From: dav-nospam-id at westco-nospam-ntrol.com (David Brown)
Date: Thu, 22 Mar 2001 16:59:00 +0100
Subject: Cross-platform serial port access
References: <99d274$ogv$1@news.netpower.no> <PRou6.1592$SB2.181656@ruti.visi.com>
Message-ID: <99d7g3$qed$1@news.netpower.no>

Grant Edwards wrote in message ...
>In article <99d274$ogv$1 at news.netpower.no>, David Brown wrote:
>
>>I am relatively new to Python, but am planning to use it for a
>>number of small programs that should be able to run both on
>>Windows (for initial development and testing) and Linux (once
>>we get a server running properly). Some of these programs will
>>need to communicate with external devices using a serial port -
>>is this going to work in a portable fashion using Python?
>
>Currently, serial port stuff in Python is not portable between
>Mac/MS/Unix.

I suppose it should be easy enough to make a small wrapper around the
different platforms' code.

>
>>Does anyone know if Linux supports RTS toggle handshaking
>>(i.e., the RTS line is high while transmitting, for convenient
>>connection to RS-485 convertors),
>
>Not with the standard PC serial port driver.  I hacked the
>16x50 driver to do RTS toggle several years ago, but it never
>worked reliably.  It's very difficult to do it reliably in
>software at high baud rates.
>
>Your best bet is to buy a board that supports RTS toggle in
>hardware or use a 232-485 converter that handles it
>automatically by detecting the presence of data.  The only
>board I've used under Linux that supported hardware RTS toggle
>was Cyclades Y series 8-port board.  I believe that RTS toggle
>support doesn't work in their newer boards, so you might have
>to look around.
>

This is definitely a shame.  The RTS toggle works well under NT (no idea
about Win9x) - I have written a Delphi program that uses it to run a RS-485
convertor at 57600.  I don't know for sure how fast RTS is turned off after
the last byte - the other nodes on the bus wait 10 ms before answering
anyway.

I guess under Linux I would have to toggle RTS manually.  Turning it on
before sending is easy enough, it is turning it off at the right time that
is more of a hassle.  I suppose I would have to calculate how long the
sending would take, sleep for that time, then turn it off.  If the process
or thread in question was fairly high priority, this should work with
reasonable accuracy.

So-called "intelligent" RS-485 convertors that read the incomming data and
change direction appropriately have their own problems.  We have delivered
systems that communicated at 300 baud, and "intelligent" convertors
frequently do not work at such slow rates.  Additionally, they generally
need configured with DIP-switches and the like, which is an extra hassle if
you need to change the baud rate.  Simple RTS-controlled convertors are much
easier in that respect.

Although it is not really Python-specific, do you know if there are serial
drivers for Linux that simulate 9-bit UARTs (i.e., swapping the parity bit
from space to mark as appropriate)?

Thanks for your help.






From stevewilliams at wwc.com  Sat Mar 17 22:16:03 2001
From: stevewilliams at wwc.com (Steve Williams)
Date: Sun, 18 Mar 2001 03:16:03 GMT
Subject: Pep 238 Why new // operator (Non-integer Division)
References: <98tc2s$qir$1@news2.ottawa.cyberus.ca> <mailman.984822021.13503.python-list@python.org> <vag7bt09r54tv05nohtssd0ge891v3fa02@4ax.com>
Message-ID: <3AB42890.AD9E9FF0@wwc.com>

Courageous wrote:

> >>All these changes make my head spin.  There is more to life than learning
> >>all the fine details in a computer language.  The fewer number of ways to do
> >>something the better.
> >
> >Just because the changes are being made doesn't mean you have to USE them.
>
> This isn't true; the moment you encounter code written by someone else,
> you DO have to use them. This simple principle is the magic of Python
> minimalism, and the primary driving reason why new Python programmers
> can become so effective so fast (IMO).
>
> C//

[ simulated persona = "Kurt G?del", node #157, max search depth 49%, neural
variance 12.180 ]

           For simplicity's sake I do all my computations in the base-pi number
system. In
           this system pi=10.



From ajtodd at mpower.com.au  Fri Mar 30 21:23:42 2001
From: ajtodd at mpower.com.au (Andy Todd)
Date: Sat, 31 Mar 2001 12:23:42 +1000
Subject: best practice? null/None (correction)
References: <9a2b9v+lrgp@eGroups.com> <mailman.985976988.3926.python-list@python.org> <9a2rnd011au@news1.newsguy.com>
Message-ID: <3AC53FAE.8E94D410@mpower.com.au>

Having used Oracle for far too many years I can confirm that "" and NULL
are not the same, sorry Robin.

The Oracle RDBMS implements the standard ANSI SQL function 'is'/'is not'
for testing for null. 
eg. SELECT 'x' 
    FROM my_table 
    WHERE my_column IS NULL 
is NOT the same as 
    SELECT 'x' 
    FROM my_table 
    WHERE my_column=''
and will not be guaranteed to return the same results as specified in
"Oracle SQL Language Reference Manual".

What you *may* have observed is that some implementations of SQL (for
intance SQL*Plus) allow lazy programming and will converted NULL
VARCHAR2 column values into the empty string after selecting them from
the database.

Andy Todd
-- "Today I'm a part time Oracle DBA and full time head scratcher"

Alex Martelli wrote:
> 
> "Robin Thomas" <robin.thomas at starmedia.net> wrote in message
> news:mailman.985976988.3926.python-list at python.org...
> > At 11:58 AM 3/30/01 -0500, Robin Thomas wrote:
> > >HTTP query arguments are of one type: STRING. There is no concept of a
> > >"null" value; the closest thing you have is an "empty" string like "". In
> > >most database engines, there is a clear difference between NULL and "" in
> > >any textual or character data column.
> >
> > I'm sorry to be outright WRONG here, fluxent.
> >
> > Many databases consider "" to be NULL for textual columns. Oracle 7 and up
> > certainly does.
> 
> Wok?!  I _thought_ Oracle was an RDBMS -- fun to hear that it isn't
> (proper treatment of NULL as distinguished from any data value being
> one crucial requisite for being called an RDBMS).  Guess I'll stick with
> SQL Server/MSDE then...
> 
> Alex


From aleaxit at yahoo.com  Fri Mar 16 10:30:13 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 16 Mar 2001 16:30:13 +0100
Subject: Python with the VBA recording capabilities
References: <3AB21C3E.95DAB3D2@yahoo.fr>
Message-ID: <98tbi80r4l@news1.newsguy.com>

"Francois Forest" <francois_forest at yahoo.fr> wrote in message
news:3AB21C3E.95DAB3D2 at yahoo.fr...
> I compared Python to VBA (not for their syntax):
>
>     VBA gives the programming capability to Applications
>
>     Python gives the programming capability at the OS level
>     (as shell scripts traditionally did )

Python (with win32com) is basically on a par with VBA in terms
of its ability to script COM object models (the pluses and
minuses of each for this task are very subtle technical issues,
and we HAVE gone into them in the past on this forum).

Whether the 'programming' goes to an Application, to 'OS level',
or wherever, has zero to do with the language involved and
everything to do with the object-model involved.  For example,
with the 'Windows Scripting Host' object model, 'Scripting
Runtime', etc, VBA can just as well be used to automate OS
level tasks (though most users appear to prefer scripting
languages, such as VB Scripting Edition, for such tasks, still
VBA is quite able to perform them too).

Python has other, additional capabilities that do not rely on
scripting a COM/Automation object model (which VBA also does
to a very limited extent, for legacy-compatibility purposes,
but _almost_ all it does is scripting various object models --
that's how it interacts with its forms and the controls on
them, for example, although direct access to the Win32 API's
_is_ also feasible).


> VBA has the RECORD capability that transform any sequence of mouse clics
> into a program that can be edited and then adapted. This recording
> capability is amazingly efficient.

Actually (as you'd know if you had ever tried integrating full
VBA support into an application), it is the application that has
to do most of the work for this purpose; the support actually
provided by the VBA framework for this purpose is helpful, but
quite minimal.


> Can we imagine to add this feature to any current GUI (kde and others) ?

No, I can't imagine that; I don't see where the GUI gets into
the macro-recording process -- it's the _application_ (in VBA's
case, but I don't see much of an alternative?) that needs to
have a mode where keystroke, mouseclicks etc get 'recorded'
(by suitable calls that the application will make to the
recording utility) in terms that are semantically meaningful
on the application's object-model.

We do have (laying around somewhere) a generic COM recorder
we had developed to allow the 'capture and playback' mode of
our own applications to also produce editable scripts (at
first in our proprietary scripting language, with some idea
on how to generalize that), but we never invested the huge
amount of work to turn the proof-of-concept prototype into
production-level infrastructure -- our existing 'capture and
playback' having been deemed perfectly adequate for its
needed uses (basically, allowing some sequence of user actions
that reveals an application defect in some pre-release
version to be saved, then repeatedly replayed as a test to
prove the defect has indeed been fixed and does not come
up again in later pre-releases and releases).  The actual
recording infrastructure is the least of it -- the really
hard stuff is peppering the _application_ with suitable calls
to the recorder, to register relevant events at a semantically
appropriate level...


Alex





From paulp at ActiveState.com  Sat Mar 10 00:06:58 2001
From: paulp at ActiveState.com (Paul Prescod)
Date: Fri, 09 Mar 2001 21:06:58 -0800
Subject: class func & var name overlap
References: <3AA9AEC2.12BA1E95@nd.edu>
Message-ID: <3AA9B672.BB06B2C@ActiveState.com>

Changsen Xu wrote:
> 
> hi all,
> 
> I just now tried function name and variable name overlaping
> in Python's script (1) and (2):
> 
> (1)
> class x:
>         x = 3
>         def x(self): return 5
> y=x()
> print y.x
>
> ...
> 
> I'm here curious why Python doesn't even report an error
> for this obvious mistake ?

The code in a class is executed just as if it were at the top level. You
can have if statements, loops etc. in there. If Python disallowed
binding the same name twice in that particular context, it would not be
as consistent as it is today. The class context would be special whereas
today it is not.

-- 
Python:
    Programming the way
    Guido
    indented it.



From costas at meezon.com  Thu Mar  1 22:03:03 2001
From: costas at meezon.com (Costas Menico)
Date: Fri, 02 Mar 2001 03:03:03 GMT
Subject: define a new func on the fly?
References: <3A9EE927.8744154E@troikanetworks.com> <o6Cn6.846$45.1115@www.newsranger.com>
Message-ID: <3a9f0ce3.10347756@News.CIS.DFN.DE>

Gregory Jorgensen<gregj at pobox.com> wrote:

>eval() evaluates an expression. exec() executes a statement. Try this:
>
Here is the exact error I get when running it

>>>> exec("def f():\n\treturn 'hello'\t")
Traceback (innermost last):
  File "<interactive input>", line 1, in ?
  File "<string>", line 2
    return 'hello'
                  ^
SyntaxError: invalid syntax

I've trying all day to get this to work. No luck...

costas


From fredrik at pythonware.com  Thu Mar 22 18:39:40 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Thu, 22 Mar 2001 23:39:40 GMT
Subject: Checking the type
References: <99a493$bhl$1@sun.rhrk.uni-kl.de> <99daj70aeu@enews4.newsguy.com> <mailman.985287571.18194.python-list@python.org>
Message-ID: <03wu6.3116$4N4.451246@newsc.telia.net>

Ken Seehof wrote:
> The compare is a tiny bit slower, but looks nicer (I think).
>
> >>> a = [2,3,4]
> >>> b = "3,4,5"
> >>> def foo(x):
> ...  if type(x) == type([]):   # just use type() with a literal

for extra guru points, use "if type(x) is type([])" (there's only one
type object for each type), or better:

    if isinstance(x, type([])):
        ....

Cheers /F




From f8dy at diveintopython.org  Sat Mar  3 02:55:54 2001
From: f8dy at diveintopython.org (Mark Pilgrim)
Date: Sat, 03 Mar 2001 02:55:54 -0500
Subject: Thinking Pythonically (was Re: gah! I hate the new string 
	syntax)
References: <mailman.983485923.1136.python-list@python.org> <B7In6.16281$Qb7.2658902@newsb.telia.net> <97ohh2$polt0$1@ID-77331.news.dfncis.de> <97opsb$nhu$1@nntp6.u.washington.edu> <97oupq$qijrh$1@ID-77331.news.dfncis.de> <97p3mh$bkm$1@tyfon.itea.ntnu.no> <B6C58CBD.212F%f8dy@diveintopython.org> <97q14l$jbi$0@216.39.151.169> <7k31at0hbkuqgjtnsfts9nieskove1nli4@4ax.com>
Message-ID: <B6C60DB9.48DA%f8dy@diveintopython.org>

in article 7k31at0hbkuqgjtnsfts9nieskove1nli4 at 4ax.com, Kirby Urner at
urner at alumni.princeton.edu wrote on 3/3/01 1:41 AM:
> Pilgrim's use of list comprehensions wherever suitable doesn't
> seem a dive into an arcane or cryptic style.  The 'map' thing
> was cumbersome, because if your function had two or more
> parameters, but you wanted to hold one or more of them constant,

Heh heh, he said "Dive Into". :)

I agree completely with Kirby.  I mentioned this in passing earlier in this
thread, but it bears repeating: the map/lambda solution is awkward no matter
how you cut it.  Consider the example of getting a list of fully qualified
pathnames of the contents of a directory, something I do all the time.  You
could certainly do this with map:

>>> map(lambda f, dirname=dirname: os.path.join(dirname, f),
os.listdir(dirname))

That's awkward because of the "dirname=dirname" hack required by the lack of
nested scopes.  (Yes, yes, I know this will be solved in Python
2.1^H^H^H2.2^H^H^H2.3.)  Or you could break it up into two steps, like this:

>>> fileList = os.listdir(dirname)
>>> map(os.path.join, [dirname]*len(fileList), fileList)

That's awkward because of the list multiplication, and it has to be done in
two steps because you don't know how long the list will be ahead of time.
Or you could just use a list comprehension, like this:

>>> [os.path.join(dirname, f) for f in os.listdir(dirname)]

To me, that says "for each of the files in this directory, make a full
pathname out of it, and give me a list of all of those."  I don't know what
the map examples say, other than "I'm trying to do something simple and this
language is making it hard."  If I wanted to feel like that, I'd go back to
Powerbuilder.

-M
You're smart; why haven't you learned Python yet?
http://diveintopython.org/



From loewis at informatik.hu-berlin.de  Fri Mar  2 09:12:06 2001
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 02 Mar 2001 15:12:06 +0100
Subject: A counter-counter-proposal for PEP 236: #pragma( nested_scopes)
References: <j47l29lpo8.fsf@informatik.hu-berlin.de> <mrwn6.16173$Qb7.2637930@newsb.telia.net> <0Swn6.3324$TW.17061@tor-nn1.netcom.ca> <epxn6.16181$Qb7.2640372@newsb.telia.net>
Message-ID: <j4vgpsjm4p.fsf@informatik.hu-berlin.de>

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

> We're talking about two kinds of "directives" here:
> 
> -- Source code encoding, which needs to be specified at the at
> the top of the source file, since it affects the tokenizer/parser
> (cf. XML encodings).  As specified, "directive" doesn't cut it.

Why is that? A directive must appear at the beginning of the source
file.

> -- Experimental compiler features, which if successful will become
> standard in the next release.  Again, "directive" doesn't work, since
> it makes "experimental" stuff look "optional".

There are certainly other categories. For example, Tim Peters
suggested that users may want to declare "optimize for size, not
speed", on a per-module basis; that would not change the language at
all, and it won't become standard in the next release, either.

As another item, where would "all strings are Unicode" fit? It's not a
source encoding, and it won't become standard in the next release,
either. The -U option does not work for it, either, since it should be
a per-module setting.

Regards,
Martin


From aahz at panix.com  Fri Mar 23 23:38:06 2001
From: aahz at panix.com (Aahz Maruch)
Date: 23 Mar 2001 20:38:06 -0800
Subject: static methods
References: <mailman.985383828.2942.python-list@python.org>
Message-ID: <99h8be$om7$1@panix3.panix.com>

In article <mailman.985383828.2942.python-list at python.org>,
Clark C. Evans <cce at clarkevans.com> wrote:
>
>I was wondering if there is any rationale as to why 
>Python classes do not have static methods?  

Without debating the rationale, the standard answer is to stick the
class into a module with the same name and use module global functions
to simulate static methods.
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

"I won't accept a model of the universe in which free will, omniscient
gods, and atheism are simultaneously true."  -- M


From a2619725 at smail.uni-koeln.de  Wed Mar  7 15:33:28 2001
From: a2619725 at smail.uni-koeln.de (Oliver Hofmann)
Date: Wed, 7 Mar 2001 21:33:28 +0100
Subject: UserDict and items()
Message-ID: <98656v$cud$1@news.rrz.Uni-Koeln.DE>

Hello everyone!


Been toying around with UserDict lately and stumbled into a problem.
Here's
the relevant code:

---
from UserDict import UserDict

class ClText(UserDict):
    def __init__(self):
        UserDict.__init__(self)

        # Set initial value
        self.data['test'] = ''


    def __getitem__(self, key):
        if key == 'test':
            return 'found'

        return UserDict.__getitem__(self, key)
---

In short, whenever a few reserved keys are queried I'd like to return
default values or fetch needed values from a database. This works
quite nicely:

>>> a = class.ClText()

>>> a['test']
'found'

Unfortunately:

>>> a.items()
[('test', '')]

Is there a more elegant way to handle items(), values() etc than
overwriting each function?

One way would be to set data[key] to the required value during
init(); however I'd like to postpone fetching data from the database
until the time it is actually requested.

Any help would be appreciated,


     Oliver


--
Oliver Hofmann  -  University of Cologne  -  Department of Biochemistry
   o.hofmann at smail.uni-koeln.de - setar at gmx.de - connla at thewell.com

   You can present the material, but you can't make me care.
  -- Calvin





From emile at fenx.com  Thu Mar 22 08:57:50 2001
From: emile at fenx.com (Emile van Sebille)
Date: Thu, 22 Mar 2001 05:57:50 -0800
Subject: tempfile problem on win98
References: <g3nu6.369$hb.2141@news1.belrs1.nsw.optushome.com.au> <99d0dg$os92$1@ID-11957.news.dfncis.de>
Message-ID: <99d0is$qcv3$1@ID-11957.news.dfncis.de>

I just notice as well that the routine doesn't in fact create the file, so
that a delay in using the name, or accessing mktemp from another process
could also result in duplicates.

Emile


"Emile van Sebille" <emile at fenx.com> wrote in message
news:99d0dg$os92$1 at ID-11957.news.dfncis.de...
> The mktemp routine guarantees to return a name unique in the target
> directory.  Are you sure you're using fully qualified path names
throughout
> when subsequently discovering duplicate files?
>
> Emile van Sebille
> emile at fenx.com
>
>




From mariogilberto at eichlerjunior.inf.br  Sun Mar  4 17:27:01 2001
From: mariogilberto at eichlerjunior.inf.br (=?iso-8859-1?Q?M=E1rio_Gilberto_Eichler_Junior?=)
Date: Sun, 4 Mar 2001 19:27:01 -0300
Subject: Need a free lancer python programmer - to do bots
Message-ID: <006e01c0a4fa$3c4e0820$52d0fea9@mario>

I need a free lancer programmer to do a bots to search information in 35 sites based in a word dictionary and write the results in a mysql database

I will pay per a job.

I leave in Brazil.

All sites are locate in Brazil.

Will pay in American Dollars.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20010304/7755c685/attachment.html>

From duncan at rcp.co.uk  Mon Mar 26 06:02:12 2001
From: duncan at rcp.co.uk (Duncan Booth)
Date: Mon, 26 Mar 2001 11:02:12 +0000 (UTC)
Subject: Python T-shirts
References: <mailman.985019605.8781.python-list@python.org> <99g4ik$lb6$1@tyfon.itea.ntnu.no>
Message-ID: <Xns90707A0047E8duncanrcpcouk@127.0.0.1>

"Magnus Lie Hetland" <mlh at idi.ntnu.no> wrote in <99g4ik$lb6$1
@tyfon.itea.ntnu.no>:

>> At the conference we handed out T-shirts with the slogan on the back
>> "Python: programming the way Guido indented it".
>
>Just thought I'd mention...
>
>  http://www.cafepress.com/
>
>(I'm in no way affiliated with them.)
>
While I'm sure they provide a good service, I find their international 
shipping charges look a bit high to really justify. For example, shipping to 
the UK is charged at $20 for the first t-shirt.


From chrisa at ASPATECH.COM.BR  Wed Mar  7 07:08:40 2001
From: chrisa at ASPATECH.COM.BR (Chris Richard Adams)
Date: Wed, 7 Mar 2001 09:08:40 -0300
Subject: finding and listing the differences between two files
Message-ID: <BD674CDC0A68C642B60038986FBB56DF0138AD@mailsrv.ASPATECH.COM.BR>

Right now I am finding the differences using Bsh...like

diff file1 file2 > filediff

if [ $? = 0 ]
	then
		echo same
	else
		echo different
		read filediff...
fi



Anyone know the equivalent in Python or am I better off sticking to
shell.

Thanks,
Chris



From sandipan at vsnl.com  Sat Mar 31 04:06:56 2001
From: sandipan at vsnl.com (Sandipan Gangopadhyay)
Date: Sat, 31 Mar 2001 14:36:56 +0530
Subject: very basic, how do you find square roots?
References: <9a408a$1mpp$1@raewyn.paradise.net.nz>
Message-ID: <004201c0b9c1$f0b38f20$020000c3@node02>

>>> import math
>>> math.pow(2,4)
16.0
>>> math.pow(2, .5)
1.41421356237

Regards,

Sandipan
----- Original Message ----- 
From: "johnston112" <gordonj at paradise.net.nz>
Newsgroups: comp.lang.python
To: <python-list at python.org>
Sent: Saturday, March 31, 2001 12:35 PM
Subject: very basic, how do you find square roots?


> Yes, that is my question :)
> I am trying to make a program that dose the quadratic equation.
> 
> johnston112
> 
> 
> -- 
> http://mail.python.org/mailman/listinfo/python-list
> 





From nanotech at europa.com  Tue Mar 27 17:36:24 2001
From: nanotech at europa.com (nanotech at europa.com)
Date: Tue, 27 Mar 2001 22:36:24 -0000
Subject: Getting all the *files* from a directory -- A better way??
Message-ID: <99r4l8+kd3u@eGroups.com>

All:

Here is how I get all the files from a directory:

  dir="a/b/c"
  files=filter(os.path.isfile,["%s/%s"%(dir,f) for f in os.listdir
(dir)])

Is this *REALLY* the Pythonic way to do this? (OR, why does
os.listdir return only file names, and not the path given
prepended?)

Quentin Crain





From aleaxit at yahoo.com  Tue Mar 20 02:11:21 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Tue, 20 Mar 2001 08:11:21 +0100
Subject: Functionalism, aesthetics Was:(RE: I come to praise .join,  not
References: <mailman.985052844.17993.python-list@python.org>
Message-ID: <996vss013mp@news2.newsguy.com>

"Carlos Ribeiro" <carribeiro at yahoo.com> wrote in message
news:mailman.985052844.17993.python-list at python.org...
>
> I'm lost for words... This summarizes a lot of concerns that sometimes I
> find so hard to explain. Anyone who has actually *read* Christopher
> Alexander's books on design should know that the best solutions are the
> ones that take into account the inner sense of satisfaction of their end
> users. As Alexander puts it itself, it is a rewarding experience born out
> of the "completeness" of the solution. The best solutions also share
> another common trait: they look *and feel* good to the untrained eye.
> Beware of works that can only be appreciated by experts - they are way too
> trained to see the obvious.

Unfortunately, this does not generalize to fields in which the non-experts
lack a few thousand years of collective cultural adaptation.  In that case,
the superficial apperception of 'beauty' does need understanding of the
underlying technology to be grounded -- shared cultural background and
biology don't suffice.

Back when "Concorde" and "Jumbo" first appeared on the scene, for
example, non-experts found the former quite beautiful, the latter quite
plain.  Sure the sleek, fast-looking lines of Concorde were superior to
the goofy-looking fatness of Jumbo?

Unfortunately, this superficial perception had zero relevance for the
actual technical soundness of the designs.  A few decades later, we
can see with hindsight that Concorde was one of the costliest big
failures in airliner design, Jumbo the unqualified greatest success
ever in the same field.  The actual need was to carry huge masses
of humanity from A to B cheaply, NOT to wiz around a few upper
class poshies in half the time.

So why did the elitist design look better?  Perhaps because it's a
common subconscious mechanism to prefer the luxury good, even
when in actual fact one won't be able to take advantage of it, as
a wishful-thinking byproduct.  Remember that human beings are
quite prone to a few well-known systematic perception errors --
and it takes LOTS of training to teach one of us to stop making
them.  Thus, a design which happens to hit squarely on the button
one of those common perception errors may well be quite pleasing,
superficially, to the untrained masses -- only highly experienced
professionals really stand a chance to smell something fishy with
it.  Computers and peripherals without an on/off button ('software
will do it all' -- yeah, right, but what happens when due to a bug
you need to recycle power...?  unplug/replug the mains...?!) are
another example, closer to our field.

Buildings as such (and more generally dwelling-places) are old
enough that it's quite possible that we all share some basic
biological basis for sound judgment -- I'm not pronouncing
on that, but that's what Alexander is mostly writing about.

Other technologies don't share that characteristics, or may
even display nothing actually relevant to the end-user; I am
reminded of a huge brouhaha about a drug that happened
a few years ago here.  The color of the pills was changed from
vivid red to pale red (actually to reduce the amount of
artificial coloring after doubts were raised about it being
intrinsically harmful).  People were furious that the medicine
had been 'weakened' -- even though the actually active
chemicals in each pill were exactly the same as before!

We can endlessly discuss where computers and software
fit on this scale, but surely the example of pharmaceutics
does suffice to show conclusively that, with far too many
modern technologies, it's quite likely that what appears to
the untrained eye has *zero* relevance -- that, no matter
how ideologically unpleasant this may feel, we DO have to
rely on _expertise and training_ to be able to judge certain
issues.  Vox populi is NOT necessarily vox Dei -- not when
the underlying technology is too new (a few centuries) and/or
not reflected in immediately apperceivable traits.


> To summarize it: a  *technically* optimal solution for a problem is not
> always the *best* solution. If it does not look intuitively good for most
> users, its probably a bad solution.

So, you're arguing that the amount of artificial coloring should
have been kept high (who cares about giving users cancer, right?),
since that bright color was definitely what "looked intuitively
good for most users" as proved by the endless carping...?


> Its unfortunate that in many cases the non-expert user keeps quiet,
> assuming that whatever opinion the "master" haves, its going to be better
> than theirs - after all they are the so-called experts.

Ask the experts in any field whether untrained users keep too quiet,
or not enough, in their field, and you'll get a different opinion (of
course, self-serving conscious or unconscious biases are at work
as well!).


Alex





From Eugene.Leitl at lrz.uni-muenchen.de  Fri Mar 16 05:36:23 2001
From: Eugene.Leitl at lrz.uni-muenchen.de (Eugene Leitl)
Date: Fri, 16 Mar 2001 11:36:23 +0100 (MET)
Subject: OO-Browser
Message-ID: <Pine.GSO.4.03.10103161134540.11202-100000@sun1.lrz-muenchen.de>

I'm setting up a Python/C development environment, so I decided to go with
XEmacs/gdb/ddd/pdb and OO-Browser. However, I'm only running into dead
links when attempting to access BeOpen.

What's up with them?




From jsoons at juilliard.edu  Thu Mar 22 16:05:13 2001
From: jsoons at juilliard.edu (Jonathan Soons)
Date: Thu, 22 Mar 2001 16:05:13 -0500
Subject: expanding a variable
Message-ID: <B7C74E1A6FBAD411BFC60002B309DDF4E435@mailbox.juilliard.edu>

I am trying to give my cv group (who all have logins like cv???.)
all the same .profile. The variable in the last line of the script
doesn't seem to be expanding, even though this is very similar to
an example in "Core Python Progr..."


d = commands.getoutput('ls /u')
dirs = string.split(d, '\n')
for dir in dirs:
    if re.search('^cv', dir):
        os.system('cp  /u/test/.profile /u/dir/')
                                           ^^^

Should I stick to bash for this kind of thing?
Thanks for any advice.



From gregj at pobox.com  Fri Mar  9 01:58:25 2001
From: gregj at pobox.com (Gregory Jorgensen)
Date: Fri, 09 Mar 2001 06:58:25 GMT
Subject: Who am I: can a class instance determine its own name?
References: <mailman.984034150.9635.python-list@python.org> <slrn9af71n.d6p.scarblac@pino.selwerd.nl> <mailman.984081385.20244.python-list@python.org>
Message-ID: <la%p6.1337$54.1249@www.newsranger.com>

Emile van Sebille's very cool hack aside, I don't understand why you would do
this. I'm not criticizing; you've got me curious. I've never needed this in
years and years of programming. I don't think of a variable name as an attribute
of the object it refers to, because (as one poster pointed out) objects may have
no name, one name, or several names at any time, and the names referring to an
object may change during the object's lifetime through assignment, parameter
passing, etc.

Even Emile's hack has limited usefulness: it captures the name at the time the
Bar object is added to a Foo container, but consider these obvious problems:

def addbar(f, b):
f.addBar(b)

foo = Foo()
bar_a = Bar()
addbar(foo, bar_a)

bar_b = Bar()
foo.addBar(bar_b)
zzz = bar_b
del bar_b


foo.listBars() will yield ['b','bar_b'], not ['bar_a','zzz'].

You mention the "problem of composition" and give a container class example,
implying that this is a problem associated with those object-oriented designs.
Why?

Python has the "is" operator to test if two objects refer to the same thing;
that's safer than using names.

>>> b = Bar()
>>> z = b
>>> z is bar
1

I'm sure I'm not the only inquiring mind who wants to know.

Greg Jorgensen
Deschooling Society
Portland, Oregon, USA
gregj at pobox.com


From william.famy at noos.fr  Fri Mar 23 10:17:21 2001
From: william.famy at noos.fr (William Famy)
Date: Fri, 23 Mar 2001 16:17:21 +0100
Subject: newbe   how to read in a string? the solution
References: <3ABB4E70.54885C9F@noos.fr> <mailman.985355670.6600.python-list@python.org>
Message-ID: <3ABB6901.F714006A@noos.fr>

Thanks for you answer.

This a sumary of the answer i receive. 
_________________________________________________
import string
     d,n1,n2,n3 = string.split(a, ",")
     n1 = float(n1)
     n2 = float(n2)
     n3 = float(n3)
 
 If the list of numbers is of variable length, this might be cool:
 
     import string
     seq = string.split(a, ",")
     for i in range(1, len(seq)): # this skips the 0 index,
         seq[i] = float(seq)      # which needs to stay a string.
_______________________________________________
One possibility:

import string

a='2000/12/28,24.3,87.1,65.4'
b = string.split(a, ',')

Then b[2] is 87.1,written as string: str(b[2])
_________________________________________________

strFields = a.split(",")
thirdField = strFields[2]
_________________________________________________



-- 
 			     \\\|///
                           \\  - -  //
                            (  @ @  )
+-------------------------oOOo-(_)-oOOo-------------------------+
|  Famy William  36 avenue des barattes  74000 Annecy  France   |
|  email:william.famy at noos.fr william.famy at mail-enseignant.com  |
+---------------------------------Oooo--------------------------+
                           oooO   (   )
                          (   )    ) /
                           \ (    (_/
                            \_)


From grante at visi.com  Fri Mar 30 16:57:29 2001
From: grante at visi.com (Grant Edwards)
Date: Fri, 30 Mar 2001 21:57:29 GMT
Subject: Know of Substantial Apps Written in Python?
References: <3ac4e4a8_1@news4.newsfeeds.com> <KE6x6.415$SR6.235983@typhoon1.ba-dsg.net>
Message-ID: <dj7x6.9561$SB2.850896@ruti.visi.com>

In article <KE6x6.415$SR6.235983 at typhoon1.ba-dsg.net>, David Allen wrote:

>What's the definition of "substantial"?
>
>There's Grail, a web browser
>http://grail.sourceforge.net

Anybody get Grail to work lately?  Last time I tried it
wouldn't run.

>There's ZOPE, an internet publishing thingy
>(thingy is a technical term  :)

Sketch is one of my favorites:

  http://sketch.sourceforge.net/

And there's teh PySol solitare game

 http://wildsau.idv.uni-linz.ac.at/mfx/pysol.html



-- 
Grant Edwards                   grante             Yow!  Yes, Private
                                  at               DOBERMAN!!
                               visi.com            


From aahz at panix.com  Sat Mar 31 17:08:11 2001
From: aahz at panix.com (Aahz Maruch)
Date: 31 Mar 2001 14:08:11 -0800
Subject: Threading legality/morality
References: <fsDv6.1247$NR.100201@news3.oke.nextra.no> <Pine.SOL.4.30.0103271851370.7646-100000@mimosa.csv.warwick.ac.uk> <99qogp$o6g$2@troll.powertech.no> <9a2r6n$mlq$1@troll.powertech.no>
Message-ID: <9a5kgb$3kk$1@panix3.panix.com>

In article <9a2r6n$mlq$1 at troll.powertech.no>,
Syver Enstad <syver at NOSPAMcyberwatcher.com> wrote:
>
>I have a confession to make here. It seems that this behaviour is a feature
>rather than a bug and has to do with the os rather than the app or python
>itself for that matter. (Inserts foot in mouth).
>
>If you click the mouse in the client area of a console window, the os
>freezes the application running there until you press a key. I guess it's
>for the convenience of copying text off of the console but it sure tripped
>me up.

Oh, that one.  Forgot about it.  Change your CMD preferences to not allow
one-click select.  (I'd be more precise, but I don't have access to NT
anymore.)
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

"I put people on my humor list as a form of revenge, you realize."  --SJM


From bsb at winnegan.de  Mon Mar  5 10:14:37 2001
From: bsb at winnegan.de (Siggy Brentrup)
Date: 05 Mar 2001 16:14:37 +0100
Subject: Little problem with the "+" operator
In-Reply-To: <983802310.155838463@news.t-online.de>
References: <983802310.155838463@news.t-online.de>
Message-ID: <87vgpo6yea.fsf@winnegan.de>

Zamurai <zamurai at gmx.net> writes:

> I want to make the following:
> 
> variableOne = 500
> myVariable = variableOne + "cool"
> 
> But I get the following error: "unsupported operand types for +"
> 
> Does someone know how to solve the problem?

That depends on what your problem is :-)

You're trying to add a string to an integer, which isn't supported.
If you expect '500cool' as a result, try

myVariable = str(variableOne) + 'cool'


Siggy




From lex at cc.gatech.edu  Wed Mar 28 11:26:12 2001
From: lex at cc.gatech.edu (Lex Spoon)
Date: 28 Mar 2001 11:26:12 -0500
Subject: Collection interfaces
References: <3a90fa97.0@news1.mweb.co.za> <MPG.15085517debf4b649896ca@news.earthlink.net> <zb%n6.31033$o%6.586026@news1.rdc1.bc.home.com> <o1jv6.9635$P4.789083@newsread1.prod.itd.earthlink.net> <99l6s8$gpu$1@nyheter.chalmers.se> <m38zlq4w34.fsf@chaos.resnet.gatech.edu> <lv13ct8qbni7f3h7q9cujj6d3iequ8to2s@4ax.com>
Message-ID: <m3g0fxg8qt.fsf@chaos.resnet.gatech.edu>

Phil Earnhardt <pae at dim.com> writes:

> On 27 Mar 2001 18:42:49 -0500, Lex Spoon <lex at cc.gatech.edu> wrote:
> 
> >matso at dtek.chalmers.se (Mats Olsson) writes:
> >
> >>     Yea, use a Java applet instead. Guaranted not to fiddle with your
> >> DLL's :-)
> >
> >Oh great, so we get to fiddle with JAR's instead of DLL's!  At least
> >JAR is easier to pronounce.  :)
> 
> Can you explain your comment, Lex? How is downloading and running
> applets from a jarfile have anything to do with fiddling with DLLs?


Java programs tend to require you to have various extensions installed
as JAR files, just the same way that Windows EXE's tend to require
that you have various extensions installed as DLL's.  I'd hazard that
JAR's are *worse* than DLL's -- DLL's at least *pretend* to do
something about versions.


Admittedly, I missed the "applet" part.  Applets run with a standard
set of classes available, so they don't have that problem.  (Instead,
they have the problem that *no* extensions are available.  Easy to set
up, but not as powerful!  :))


-Lex



From robin at jessikat.fsnet.co.uk  Mon Mar 12 19:05:06 2001
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Tue, 13 Mar 2001 00:05:06 +0000
Subject: ANNOUNCE Mailman 2.0.3
References: <mailman.984437116.20063.python-list@python.org>
Message-ID: <CCaB3KAyQWr6Ewo5@jessikat.fsnet.co.uk>

In article <mailman.984437116.20063.python-list at python.org>, Barry A.
Warsaw <barry at digicool.com> writes
>
>[I think I forgot to announce the Mailman 2.0.2 release on these
>mailing lists.  Apologies.  -Barry]
>
>I've just announced the 2.0.3 release of Mailman, the GNU Mailing List
>Manager.  Mailman is released under the GNU General Public License
>(GPL).
>
>I highly recommend that all sites running earlier versions of Mailman
>upgrade to 2.0.3.  This release includes the important privacy patch
>from Mailman 2.0.2.  Sadly, Mailman 2.0.2 contained a small but nasty
>typo requiring this new version.  Mailman 2.0.3 is being released as
>both a gzip'd source tarball and as a patchfile.
....
>Enjoy,
>-Barry
>
thanks, I've just installed 2.0.2 and set up my first list. It's not a
problem for me now, but is it easy to install over an existing
installation?
-- 
Robin Becker


From michael at lindrum.de  Thu Mar 15 17:01:58 2001
From: michael at lindrum.de (Michael Lindrum)
Date: Thu, 15 Mar 2001 23:01:58 +0100
Subject: Python2C
References: <3AAE98EA.2754A358@lindrum.de> <4gyr6.160$4d3.34825@newsb.telia.net>
Message-ID: <3AB13BD5.5831BD4A@lindrum.de>

Thanks, in the meantime I got the lib, but it needs libpython.a from Python
2.0

Fredrik Lundh schrieb:

> Michael Lindrum wrote:
> > I'm searching for the lib that is necessary to compile C-Code generated
> > by Python2C.
> > OS: SunOS
> > Who knows where I can find it?
>
> assuming you mean libpython, all you have to do is to
> download and build python itself:
>
>     http://www.python.org
>
> (if that's not it, consider telling us what library you think
> you're missing)
>
> Cheers /F



From bedge at troikanetworks.com  Fri Mar  2 10:37:34 2001
From: bedge at troikanetworks.com (Bruce Edge)
Date: Fri, 02 Mar 2001 07:37:34 -0800
Subject: define a new func on the fly?
References: <mailman.983498523.8355.python-list@python.org>
Message-ID: <3A9FBE3E.5CD3BB69@troikanetworks.com>

Christopher Li wrote:
> 
> On Thu, 1 Mar 2001, Bruce Edge wrote:
> 
> > >>> class xx:
> > ...   def __init__(self):
> > ...     exec( "def yyy():  print \"yyyyyyyyy\"" )
> > ...
> > >>> x=xx()
> > >>>
> > >>> x.__dict__
> > {}
> >
> > Where is yyy defined?
> In local namespace of __init__.

Ahhh. I'm coming from Tcl, it's a wee bit different.

> >
> > Ideally, I'd like to be able to create it in the global namespace.
> Do yon mean in class xx?
> How about
> 
> >>> class xx:
> ...     def __init__(self):
> ...         exec( "def tmp(self): print 'yyyyyy'")
> ...         self.yyy = tmp
> ...
> >>> x = xx()
> >>> x.__dict__
> {'yyy': <function tmp at 0x81df40c>}

OK, so that kicks it up to the class namespace.
Now, just for completeness, I wanted it in the global namespace, so:

>>> class xx:
...  def __init__(self):
...    global g
...    exec( "def tmp(): print 'yyyyyy'")
...    self.yyy = tmp
...    g = tmp
... 
>>> g=dir  
>>> g
<built-in function dir>

>>> t=xx()
>>> t.__dict__
{'yyy': <function tmp at 0x81e63ec>}
>>> type(g)
<type 'function'>
>>> g()
yyyyyy


Thanks to all for the help. Guess I won't be going back to Tcl after all
:-)

> Chris


From DavidlNo at spammyraqia.com  Wed Mar 28 13:28:01 2001
From: DavidlNo at spammyraqia.com (David Lees)
Date: Wed, 28 Mar 2001 13:28:01 -0500
Subject: Setting sleep interval less than 10 mS on a Linux PC?
Message-ID: <3AC22D31.FD4644B1@spammyraqia.com>

I suspect this is a system question, but I will ask anyway.  I am using
the Python sleep method (time.sleep) and on a PC running RedHat 7.0
Linux,  it has a 10 millisecond lower limit.  Below that it still wakes
up every 10 milliseconds, regardless of the argument.  I gather from a
little reading, that 10 mS is the default miniumumlinux sleep interval. 
Is there a way from Python to set this to a shorter interval (1 mS would
be nice), without recompiling the linux kernel?

david lees


From fdrake at localhost.localdomain  Fri Mar 23 14:11:52 2001
From: fdrake at localhost.localdomain (Fred Drake)
Date: Fri, 23 Mar 2001 14:11:52 -0500 (EST)
Subject: [development doc updates]
Message-ID: <20010323191152.3019628995@localhost.localdomain>

The development version of the documentation has been updated:

	http://python.sourceforge.net/devel-docs/


Documentation for the second beta release of Python 2.1.

This includes information on future statements and lexical scoping,
and weak references.  Much of the module documentation has been
improved as well.




From jeremy at alum.mit.edu  Mon Mar 19 14:19:09 2001
From: jeremy at alum.mit.edu (Jeremy Hylton)
Date: Mon, 19 Mar 2001 14:19:09 -0500 (EST)
Subject: Why "from __future__" stinks; a counter-offer
In-Reply-To: <4Aft6.2171$pP2.372173@paloalto-snr1.gtei.net>
References: <mailman.984949523.1896.python-list@python.org>
	<4Aft6.2171$pP2.372173@paloalto-snr1.gtei.net>
Message-ID: <15030.23469.813235.630703@w221.z064000254.bwi-md.dsl.cnc.net>

>>>>> "VC" == Vadim Chugunov <chega_ at yahoo.com> writes:

  >> might be OK -- alas, without a PEP, it's a non-starter).

  VC> But, surely, the PEP can be updated!  This "from __future__"
  VC> syntax is ugly, ugly, ugly!!!

Vadim,

If someone else wants to write a PEP and implement it this week, it
would be considered.  Tim wrote the PEP.  I implemented it.  We're
both going to be spending most of our waking hours this week getting
ready for the beta release on Friday.  We have no time and less
interest in implementing something other than the __future__
mechanism.  We think it is a good solution to the problem it addresses.

But neither of us is the BDFL.  If someone writes a PEP and implements
it in the next day or two, he or she may be able to convince Guido
that it is better than __future__.  In the absence of someone writing
a new PEP, nothing is going to change.

Jeremy



From embed at geocities.com  Thu Mar  8 12:43:56 2001
From: embed at geocities.com (Warren Postma)
Date: Thu, 8 Mar 2001 12:43:56 -0500
Subject: Ldap / Python help
References: <mailman.984067391.2391.python-list@python.org>
Message-ID: <IyPp6.6645$TW.28719@tor-nn1.netcom.ca>

"Pablo Pernot" <ppe at albert.com> wrote in message
news:mailman.984067391.2391.python-list at python.org...
> Hello all,
>
> I'm new in python
>
> I found an ldap module.
>
> Where can I find documentation about ldap/python
>
> my modify instruction won't work => modify_s is waiting for a string (dn)
> and a list (mylist) formed by tuple (mytuple) and this tuple is formed for
3 strings.
>
> So I can't understand the error message about integer.
>
> mytuple=("MOD_REPLACE","sn","test")
> mylist=[mytuple]
>
> >>> l.modify_s(dn,mylist)
> Traceback (most recent call last):
>   File "<stdin>", line 1, in ?
> TypeError: an integer is required
>
>

To convert a tuple to a list, try:

mylist = list(mytuple)

To see what mylist=[mytuple] does, try this:
print repr( [mytuple] )

You see the difference now?

Warren




From qrczak at knm.org.pl  Mon Mar 19 14:50:13 2001
From: qrczak at knm.org.pl (Marcin 'Qrczak' Kowalczyk)
Date: 19 Mar 2001 19:50:13 GMT
Subject: [Very Long (11K)] Numeric PEPs, first public posts
References: <mailman.984748941.31248.python-list@python.org>
Message-ID: <slrn9bconk.ns2.qrczak@qrnik.zagroda>

Fri, 16 Mar 2001 15:18:43 +0200, Moshe Zadka <moshez at zadka.site.co.il> pisze:

> PEP: 237
> Title: Unifying Long Integers and Integers

I like it!

> Overflows
> 
>     When an arithmetic operation on two numbers whose internal
>     representation is as machine-level integers returns something
>     whose internal representation is a bignum, a warning which is
>     turned off by default will be issued.  This is only a debugging
>     aid, and has no guaranteed semantics.

I would not bother with this warning.

> Implementation
> 
>     The PyInt type's slot for a C long will be turned into a 
> 
>         union {
>             long i;
>             struct {
>                 unsigned long length;
>                 digit digits[1];
>             } bignum;
>         };
> 
>     Only the n-1 lower bits of the long have any meaning; the top bit
>     is always set.  This distinguishes the union.

The i is unified with length, not with digits[].

Isn't it simpler to have Int as Long as separate types? If they behave
in the same way, arithmetic on Ints yields longs when needed etc.,
it should be OK and nice. They can even return the same type(x)
and have a unified PyInt_Check, but operations are more naturally
dispatched as currently. It's probably unnecessary to normalize the
large form into the small form if the value gets smaller.

That way there are fewer places which have to dispatch on the
representation.

Wouldn't using gmp be an advantage?

> Open Issues

Let me add: I would like to have PyInt_AsString. I know I can convert
to a Python string object and get the C string from it, but it looks
like an unnecessary indirection.

I wanted PyLong_AsString to convert them for another bignum library.
There might be faster ways to do this, but every library will
understand decimal strings.

Unfortunately it's unclear how to obtain and free the storage, as
usual in C.

Well, maybe it's not that important.

>     What to do about sys.maxint?

Deprecate like the long function etc., unless some important libraries
would still be limited to that.

>     What to do about PyInt_AS_LONG failures?

A good question.

>     What do do about %u, %o, %x formatting operators?

If portability here is not that important, let them output negative
numbers as negative. One can use "%x" % (x % 2**32) to get the old
behavior.

If it is, I don't know.

>     How to warn about << not cutting integers?

If there is a portability switch like from __future__ import foo,
use that; it's bad that it doesn't happen at compile time, but maybe
it's all that we can get. If not, I don't know.

>     Should the overflow warning be on a portable maximum size?

I would drop it.

>     Will unification of types and classes help with a more
>     straightforward implementations?

I don't know.

> PEP: 238
> Title: Non-integer Division

I like it!

> Open Issues
> 
>     Should the // operator be renamed to "div"?

If % is % and not "mod", I would probably spell it //. But I would
call % "mod", except that it's too late.

> PEP: 239
> Title: Adding a Rational Type to Python

I like it!

I guess that sin(1.5) converts the argument to float?

How many digits does str(1/3) return, assuming it prints the number
in decimal?

Should str(5.0) return '5.0' or '5'? Note that it finally doesn't
hurt if you have an integer instead of a rational, so there is more
sense to let it return '5' than currently: integers can be treated
as a special case of rationals (implemented as a separate type).
But maybe accepting 4.0 as list indices etc. is too much.

What does repr(1/3) return? '1/3'?

> PEP: 240
> Title: Adding a Rational Literal to Python

I like it!

All PEPs together make arithmetic on literals exact, independent of
the platform's integer size and float representation. Moreover one can
always write 42 instead of 42.0 and get the same result (numerically).

BTW. A nice thing about static typing is that it allowed Haskell's
numeric literals and functions to have types deduced from the context
of usage. 5+1/3 can be computed on either Rational or Double or
Complex Float or any other (perhaps user defined) fractional type.
Literals are internally converted from Integer (infinite precision)
or Rational (= Ratio Integer). In Python it can't be done that way,
but the conversion can be delayed such that the results are the same
(modulo the *lack* of *loss* of precision if numbers are converted
to float too late).

-- 
 __("<  Marcin Kowalczyk * qrczak at knm.org.pl http://qrczak.ids.net.pl/
 \__/
  ^^                      SYGNATURA ZAST?PCZA
QRCZAK


From weiqigao at networkusa.net  Wed Mar  7 20:53:03 2001
From: weiqigao at networkusa.net (Weiqi Gao)
Date: Wed, 07 Mar 2001 19:53:03 -0600
Subject: Article on Rhino and Jython
Message-ID: <3AA6E5FF.FDAA74C@networkusa.net>

Hi,

I just though some of the Jython fans on this list might find the
following article useful in trying to convince your Java-knowledgable
but Jython-ignorant colleagues to start using Jython (or Rhino):

http://www.ociweb.com/jnb/archive/jnbMar2001.html

Feel free to send comments to me (weiqigao at networkusa.net), link to it,
show it to your friend.  My goal is to make the average Java programmer
curious about Jython and Rhino (or any of the other languages mentioned)
enough to actually try them.

--
Weiqi Gao
networkusa.net


From syver at NOSPAMcyberwatcher.com  Tue Mar 20 19:38:46 2001
From: syver at NOSPAMcyberwatcher.com (Syver Enstad)
Date: Wed, 21 Mar 2001 01:38:46 +0100
Subject: urllib.urlopen doesn't accept Unicode strings.
Message-ID: <998tau$q6i$1@troll.powertech.no>

The offending line is at 241 in urllib.py (python 2.0)

    def open_http(self, url, data=None):
        """Use HTTP protocol."""
        import httplib
        user_passwd = None
        # next line patched, original line was: if type(url) is type(""):
        if type(url) is type("") or type(url) is type(u""):
            host, selector = splithost(url)
            if host:

Or maybe this has been fixed in Python 2.1?




From not.this at seebelow.org  Mon Mar 19 16:23:04 2001
From: not.this at seebelow.org (Grant Griffin)
Date: 19 Mar 2001 13:23:04 -0800
Subject: Pep 238 Why new // operator (Non-integer Division)
References: <mailman.984811161.25962.python-list@python.org>
Message-ID: <995tbo0oce@drn.newsguy.com>

In article <mailman.984811161.25962.python-list at python.org>, "Tim says...
>
>[Grant Griffin]
>> BTW, I guess I'm not the first one to notice that this new operator
>> looks an awful lot like a C/C++ comment.
>
>No, you're the first! 329 people did note that // looks an awful lot like
>Fortran's string concatenation operator.  I expect these were the same 329
>who complained to the C++ committee about abusing *their* prior art.

Eureka!--I get it now.  It's designed to make it easier to write Fortran in
Python (er, I mean "Fortran _as_ Python") <wink>

>> So maybe some other symbol should be used.

Now that you mention it, did the previous line inspire "print >>"?

>
>Why bother?  Python's # looks an awful lot like C++'s preprocessor directive
>flag too.

Now don't get me start on that: to me, it had looked more like _Perl's_ #!  
(That has too many sharp edges!)

anything-that-python-and-perl-agree-on-can't-be
   -*all*-good-<wink>-ly y'rs,

=g2

_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation            http://www.iowegian.com



From root at [127.0.0.1]  Wed Mar 28 09:35:08 2001
From: root at [127.0.0.1] (deadmeat)
Date: Wed, 28 Mar 2001 14:35:08 GMT
Subject: list vs tuple
References: <hGlw6.782$p5.3210@news1.rivrw1.nsw.optushome.com.au> <3AC1EB93.7BC87292@hack.org>
Message-ID: <wEmw6.785$p5.3426@news1.rivrw1.nsw.optushome.com.au>

> Tuples are immutable, lists are not.

As are strings which I find a tad irritating (I come from a Pascal
background)...  it just means more work for me to split & rejoin.

so is that the only difference?





From brangdon at cix.co.uk  Thu Mar  1 05:16:00 2001
From: brangdon at cix.co.uk (Dave Harris)
Date: Thu, 1 Mar 2001 10:16 +0000 (GMT Standard Time)
Subject: Stream interfaces
References: <3A9BC3F9.1929EB9A@ix.netcom.com>
Message-ID: <memo.20010301101637.55617A@brangdon.madasafish.com>

tgagne at ix.netcom.com (Thomas Gagne) wrote (abridged):
>                 ExternalStream
>                     BufferedExternalStream
>                         ExternalReadStream
>                             ExternalReadWriteStream
>                         ExternalWriteStream
>                 InternalStream
>                     WriteStream
>                         ReadWriteStream

Do you agree this hierarchy is a bit duff? Notice how "Write" appears in 
3 leaf classes: ExternalReadWriteStream, ExternalWriteStream, 
ReadWriteStream. These apparently share no implementation or interface. 
Notice also that in one place we get a trio of classes:

>                         ExternalReadStream
>                             ExternalReadWriteStream
>                         ExternalWriteStream

and in another we get a pair:

>                     WriteStream
>                         ReadWriteStream

yet they would seem to be doing the same kind of thing. How do I get an 
internal stream which is readable but not writable? If ReadWriteStream 
inherits from WriteStream, why doesn't ExternalReadWriteStream inherit 
from ExternalWriteStream?

Something has gone wrong here.

  Dave Harris, Nottingham, UK | "Weave a circle round him thrice,
      brangdon at cix.co.uk      |   And close your eyes with holy dread,
                              |  For he on honey dew hath fed
 http://www.bhresearch.co.uk/ |   And drunk the milk of Paradise."


From syver at NOSPAMcyberwatcher.com  Tue Mar 27 06:43:02 2001
From: syver at NOSPAMcyberwatcher.com (Syver Enstad)
Date: Tue, 27 Mar 2001 13:43:02 +0200
Subject: Threading legality/morality
References: <fsDv6.1247$NR.100201@news3.oke.nextra.no> <99nnbv$phs$1@panix6.panix.com>
Message-ID: <h0%v6.1842$NR.144463@news3.oke.nextra.no>


"Aahz Maruch" <aahz at panix.com> wrote:
> Syver Enstad <syver at NOSPAMcyberwatcher.com> wrote:
> >
> >How is calling print from more than one thread looked upon?
>
> Depends what you're trying to do.  If it's debugging output and if you
> concatenate strings before trying to do a print (don't use multiple
> prints in succession), you're safe.  Anything other than that is
> probably risky -- but unlikely to do actual damage (you'll just get
> messed up output).

Yes, that is to be expected. Are you aware of any material on which Python
modules that are threadsafe, and the definition of threadsafety I am talking
about is using functions and objects from the same class in different
threads, but not one object from more than one thread. To those familiar
with COM, threadsafe like apartment threaded objects.

I have seen some strange behaviour in some multi threaded programs I have
made in python, where they seem to lockup, but a push on the the return key
(or another key, I don't know) keeps them going. These are console programs,
running under Win2000 server and professional, Python 2.0.

As I was writing this I checked out my program that has been running for
about 24 hours, downloading files from the internet via urllib, and using
ADO via pythoncom. It has been running ever so happily until about 10 clock
this morning when I inspected some output it had given. Then it froze for
about three hours, when I pushed a key on the keyboard it continued again.
Strange.




From bernhard at abnoba.intevation.de  Tue Mar 27 08:14:51 2001
From: bernhard at abnoba.intevation.de (Bernhard Reiter)
Date: 27 Mar 2001 13:14:51 GMT
Subject: Linux GPS program
References: <3AC01637.4EF3A6F2@crynwr.com>
Message-ID: <99q3ob$ehr$1@newsserver.rrzn.uni-hannover.de>

[Posted and mailed]

Hi Russell,

will pygps be Free Software? I could not find anything about a
license in the cvs.

If it was I will be happy to add it to the www.freegis.org
link collection.

	Bernhard

In article <3AC01637.4EF3A6F2 at crynwr.com>,
	Russell Nelson <nelson at crynwr.com> writes:
> Hi.  I have written a GPS program in Python for Linux.  I'm sure that
> someone 
> could port it to run under Windows, but I have no interest in doing so. 
> I haven't
> done a real release of the software yet, since it's still under heavy
> development.
> But you're welcome to download a copy from handhelds.org's CVS.  The
> location is "apps/pygps".  For info on handhelds.org's CVS:
> http://www.handhelds.org/sources.html
> 
> Right now, it does moving maps (UTM bitmaps), track recording, and
> playback.
> Will select the "best" of multiple maps.
> 

-- 
Professional Service around Free Software                (intevation.net)  
The FreeGIS Project                                         (freegis.org)
Association for a Free Informational Infrastructure            (ffii.org)
FSF Europe                                            	  (fsfeurope.org)


From greg at cosc.canterbury.ac.nz  Tue Mar 20 21:22:30 2001
From: greg at cosc.canterbury.ac.nz (Greg Ewing)
Date: Wed, 21 Mar 2001 14:22:30 +1200
Subject: relational/boolean operator overloading -> PEP 207
References: <98tjq6$qre$1@news.flashnet.it> <994h71$ohq$1@news.flashnet.it> <996r1u$67f$1@panix6.panix.com> <9977ap$l05$1@news.flashnet.it> <997t73$p7i$1@panix3.panix.com>
Message-ID: <3AB81066.38E5DFE9@cosc.canterbury.ac.nz>

Aahz Maruch wrote:
> 
>     5 In the current proposal, when A<B returns an array of
>       elementwise comparisons ... its interpretation as Boolean by 
>       the shortcut operators raises an exception.

That's a special case, though. In general, obj1 and obj2 will do 
the same as it does now on arbitrary objects -- i.e. evaluate to 
obj2 if obj1 is considered "true", otherwise obj1.

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


From greg at cosc.canterbury.ac.nz  Thu Mar  1 19:55:47 2001
From: greg at cosc.canterbury.ac.nz (Greg Ewing)
Date: Fri, 02 Mar 2001 13:55:47 +1300
Subject: Collection interfaces
References: <3a90fa97.0@news1.mweb.co.za> <3A933830.7F98D118@ix.netcom.com> <CgJk6.515$v77.66568@newsread2.prod.itd.earthlink.net> <220220010813598042%cmh@bDistributed.com> <ud7cakvi8.fsf@roundpoint.com> <MPG.150312ceb68f1ada98969f@news.earthlink.net> <8Gem6.473353$U46.14105062@news1.sttls1.wa.home.com> <MPG.1503548620e632919896a4@news.earthlink.net> <MCjm6.474652$U46.14163288@news1.sttls1.wa.home.com> <MPG.150394195f7f04a19896a7@news.earthlink.net> <3A9A86B1.9A8C69A2@ix.netcom.com> <MPG.1504d14563dce6499896af@news.earthlink.net> <97fjla128ir@news2.newsguy.com> <MPG.1505a76831c34b749896b3@news.earthlink.net> <97h5s40vhc@news2.newsguy.com> <MPG.1506007591cf818b9896bc@news.earthlink.net> <97ifd60sen@news1.newsguy.com> <MPG.1506c16edd0cb6159896bf@news.earthlink.net>
Message-ID: <3A9EEF93.4B18FCA@cosc.canterbury.ac.nz>

Topmind wrote:
> 
> There is no polymorphism if there is
> only one "type" with a bunch of independant attributes -
> HAS-A reigns and sub-IS-A is rare.

Polymorphism is not inherently dependent on any 
kind of inheritance or classification system.
In Python, for instance, most functions are quite
promiscuous and will happily operate on any
objects which have the right methods, without there
having to be any inheritance relationship between
them.

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


From chris.gonnerman at usa.net  Wed Mar 21 22:51:28 2001
From: chris.gonnerman at usa.net (Chris Gonnerman)
Date: Wed, 21 Mar 2001 21:51:28 -0600
Subject: CGIPython on BSD/OS (BSDI) ?
Message-ID: <005801c0b283$69e8d3a0$a100000a@local>

I have recently set up a personal web space account on Prohosting.com.  They
permit CGI but only support /bin/sh and perl.  According to the FAQ they are
running BSDI systems (meaning the "commercial" BSD/OS kernel).

Has anyone on this list ported mxCGIPython to BSD/OS?  Lemberg's site
(http://www.lemburg.com/files/python/mxCGIPython.html) doesn't mention BSDI
so obviously no one has sent it to him.

Alternately, is there anyone who can build for this target?  I'd do it
myself
but I've never before seen a BSDI system.





From ws-news at gmx.at  Mon Mar 12 05:51:14 2001
From: ws-news at gmx.at (Werner Schiendl)
Date: Mon, 12 Mar 2001 11:51:14 +0100
Subject: Can anyone recomend a good intoduction to C...
References: <mailman.984256456.30652.python-list@python.org>
Message-ID: <984394558.795085@newsmaster-04.atnet.at>

The string module and methods are a perfect example that it is somewhat hard
to have _only_ one passibility of doing things.
What is perfectly fine and state-of-the-art today will probably look poor
and old fashioned in some years.

But I agree that there should be at least one _obvius_and_easy_ way to
perform every task.

regards
werner

Tim Peters <tim.one at home.com> wrote in message
news:mailman.984256456.30652.python-list at python.org...
> [Werner Schiendl]
> > What's TMTOWTDI, please?
>
> [Fran?ois Granger]
> > There is More Than One Way To Do It
> >
> > which is a Perl motto as far as I know.
>
> Right!  My automobile license plate reads
>
>     TOOWTDI
>
> Honest.  I will know that my work on Python is finally complete when the
> average American looks at my license plate and nods in silent agreement.
I'm
> not there yet.  So far, approximately 100 out of 100 look at it, ponder,
then
> start making noises like some demented 170-pound bird with a broken beak.
>
> perhaps-i'd-have-better-luck-if-i-moved-to-california-ly y'rs  - tim
>
>




From dsh8290 at rit.edu  Thu Mar  8 09:52:38 2001
From: dsh8290 at rit.edu (D-Man)
Date: Thu, 8 Mar 2001 09:52:38 -0500
Subject: gah! I hate the new string syntax
In-Reply-To: <m2vgpkyfjo.fsf@mycroft.actrix.gen.nz>; from mycroft@actrix.gen.nz on Thu, Mar 08, 2001 at 06:51:07PM +1300
References: <mailman.983615170.23313.python-list@python.org> <tN5o6.2746$eH3.14913@NewsReader> <Koco6.7518$1D5.244602@e420r-atl1.usenetserver.com> <mailman.983997252.5628.python-list@python.org> <m2vgpkyfjo.fsf@mycroft.actrix.gen.nz>
Message-ID: <20010308095238.A2645@harmony.cs.rit.edu>

On Thu, Mar 08, 2001 at 06:51:07PM +1300, Paul Foley wrote:
| On Wed, 7 Mar 2001 15:33:30 -0500, D-Man  wrote:
| 
| > What about one who is both English and Scots, yet never set foot on
| > that island ;-)?
| 
| AFAIK, nobody outside the US would admit the existence of such a
| thing.  The American habit of calling themselves "Irish" or "Italian"
| or whatever when they're not is *very* strange.  I've never heard
| anything like it anywhere else.

Heh.  Maybe Americans are very odd.  Wouldn't surprise me at all.  ;-)
I was born in America and have lived in America all my life.  I do
have English, Scots, Irish and Welsh ancestry though.  One could argue
that I am not really American since none of my ancestors came from
America.  "Real" Americans would be the so-called "Indians", or more
appropriately "Native Americans".  (Also Aztecs, Inca, etc., and their
descendants such as the Quiche)

-D




From tuttledon at hotmail.com  Thu Mar 22 20:50:23 2001
From: tuttledon at hotmail.com (Don Tuttle)
Date: Fri, 23 Mar 2001 01:50:23 GMT
Subject: PyUnit in Python 2.1 standard library
References: <3ABA8A23.85EA76CE@downright.com>
Message-ID: <zZxu6.150312$__6.27726222@typhoon.southeast.rr.com>

"Doug Fort"
> Just recieved word that Steve Purcell has checked PyUnit  into the
> standard library.  Aside from being a useful tool to me PyUnit typifies
> the kind of development that ESR advocates in 'The Cathedral and the
> Bazaar'.   A project started to 'scratch an itch', then shared with the
> communityto evolve and grow.  None of the stuff I  work on seems to come
> out so clean.  Congratulations Steve!  PyUnit is an example to us all.

So what kind of itch does PyUnit scratch?

Don




From cce at clarkevans.com  Tue Mar  6 14:20:40 2001
From: cce at clarkevans.com (Clark C. Evans)
Date: Tue, 6 Mar 2001 14:20:40 -0500 (EST)
Subject: Can Python replace Visual Basic? Should it?
In-Reply-To: <pW9p6.12818$hn5.1804827@news1.rdc1.mb.home.com>
Message-ID: <Pine.LNX.4.21.0103061410570.21280-100000@clarkevans.com>

> > Face it, most VB users aren't exactly star programmers. If I were
> 
> That's a sweeping generalization, that I'd be very careful about 
> making, to say the least.

There is one thing that I would point out.  In the corporate
environment actually 'laying-in-code is really a minor part
of an individuals job.  Getting the requirements, talking to 
the users, managing expectations, building test plans, 
drawing up a design, working out impacts on other applications,
getting buy-in for the extra field you need on a common 
table, getting your data model approved, supporting the
previous version of the program, etc., all take up considerable
time.  When it comes down to it, when I was in a corporate
setting, I only got to 'code about 10 hours a week.
What GUIs do is let you 'jump into the process and continue
to work.  In other words, they are good at helping to remind
you where your context is... you just press play and see
what breaks!  

> I'll certainly admit that I've only got about 7 months worth of experience
> in "the industry", but of all the programmers I know of that had more than a
> passing interest in VB (in other words, they liked and used it enough that
> one might refer to them as "A VB Programmer(TM)"), at LEAST three (of the
> six that I'm thinking of) were extremely bright and creative hackers. From
> my perspective, I would be proud to be as good as they are!

In general, VB is a good "put-the-components-together" tool.
Quite a few, although not the majority of VB programmers that 
I know are very good at many other languages.  And ones like
me use VB to prototype a feature, and then I re-write it as 
an MSVC OCX so that it can be shared across the enterprise.

> That said though, the stereotype of VB programmers all being idiots is
> extremely common among Linux and Unix users, who, in many cases have almost
> no knowledge of the tool, and would look silly themselves if throw into a
> large project that used it ("silly" as in far from the creative geniuses
> they would seem to indicate they are).

Hear Hear!

But, alas, no Holy Wars.  I see Python as my new
"put-the-components-together" tool.  And possibly
even a "make-the-components-tool", but I that remains
to be proven to me.

Kind Regards,

Clark




From scarblac at pino.selwerd.nl  Mon Mar 19 09:06:31 2001
From: scarblac at pino.selwerd.nl (Remco Gerlich)
Date: 19 Mar 2001 14:06:31 GMT
Subject: The order of classes
References: <9r3cbt04tg3c3li151416n9b2a0gg3eccb@4ax.com>
Message-ID: <slrn9bc4fc.95q.scarblac@pino.selwerd.nl>

Daniel Klein <danielk at aracnet.com> wrote in comp.lang.python:
> Is it a requirement to place classes in a particular order if one class uses
> another in its __init__ method?

*In* its method, no. But below, you use it as a default argument for the
method, and those are evaluated when the 'def' is executed - and at that
moment, foo isn't defined yet, so you get an error.

> To illustrate this in its simplest form...
> 
> class bar:
>     def __init__(self, f = foo()):
>         pass
> 
> class foo:
>     def __init__(self):
>         pass
> 
> ...then at the interactive prompt...
> 
> >>> import foobar
> Traceback (innermost last):
>   File "<pyshell#0>", line 1, in ?
>     import foobar
>   File "C:\DanielK\python\foobar.py", line 1, in ?
>     class bar:
>   File "C:\DanielK\python\foobar.py", line 2, in bar
>     def __init__(self, x = foo()):
> NameError: There is no variable named 'foo'
> 
> ...however, it works if I reverse the order of the classes and put 'foo' before
> 'bar'.
> 
> I would like to keep my classes in a particular order but it does not appear
> that I can do this in all cases.

Don't use default arguments like this. You do realize, that even if you put
the definition of foo before bar, that every instance of bar will have 
*the same foo instance* as its default argument? The 'f = foo()' happens
only once, namely when the class is defined.

Try something like this:

class bar:
   def __init__(self, f = None):
      if f == None:
         f = foo()

class foo:
   pass
   

Now the 'f = foo()' happens when a bar instance is created, and at that
time, foo already exists.

-- 
Remco Gerlich


From mcfletch at home.com  Thu Mar 29 09:11:18 2001
From: mcfletch at home.com (Mike C. Fletcher)
Date: Thu, 29 Mar 2001 09:11:18 -0500
Subject: Basic Question: 2.0 --> 2.1 do extensions need to be recompiled?
Message-ID: <000001c0b85a$1fec6f50$a9807018@cr706570a>

I'm short on time this week, so I'd rather not have to get Python, Numeric,
etceteras just to compile PyOpenGL if it isn't necessary.  So, is
recompilation required for extension modules compiled for 2.0 to work with
2.1?

Thanks all,
Mike
__________________________________
 Mike C. Fletcher
 Designer, VR Plumber
 http://members.home.com/mcfletch




From loredo at spacenet.tn.cornell.edu  Mon Mar 12 18:34:24 2001
From: loredo at spacenet.tn.cornell.edu (Tom Loredo)
Date: Mon, 12 Mar 2001 18:34:24 -0500
Subject: ZODB and OSX
References: <mailman.984324988.25505.python-list@python.org> <m3wv9w3x6h.fsf@gnu.org> <slrn9aoi0i.rt.amk@mira.erols.com> <m3bsr7h9rt.fsf@gnu.org> <B6D236BD.2310%bobhicks@adelphia.net>
Message-ID: <3AAD5D00.C91265FB@spacenet.tn.cornell.edu>

Howdy-

Robert L Hicks wrote:
> 
> The current beta says that it will compile on OSX. ZODB would be a great
> thing to have as well!

I compiled the ZODB parts of Zope for Python1.5.2 under OS9 last summer,
and I have a similar port for Python2.0.  However, I'm only just
learning ZODB myself and haven't exercised either of these ports (which
is why I haven't put them on the web yet).  If there are Mac users
interested in these, feel free to drop me a line.

-Tom Loredo


From akuchlin at mems-exchange.org  Thu Mar 22 17:41:19 2001
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 22 Mar 2001 17:41:19 -0500
Subject: Is setup.py a "good thing"? (long)
References: <3aba76e2.167083508@24.0.228.33>
Message-ID: <3dvgp1zasw.fsf@ute.cnri.reston.va.us>

support @ internetdiscovery.com (Mike Clarkson) writes:
> #1) Someone documented what setup.py is supposed to do, and how it is
> supposed to interact with configure and make.

I've written most of setup.py, so I'll comment.  It arose from the
observation that most people don't edit the Module/Setup file and turn
on all the modules they could possibly run.  This is annoying because
there are bunches of modules there that are useful, but you can't
really count on them being very widespread -- how many people
uncomment, say, the syslog module when they compile from source.

There was some work done for 2.0, and during the 2.0->2.1 development
process, on making configure check for the BSDDB library, several DBM
variants, and curses, but this didn't approach the number of modules
available in Setup, and the process of supporting a module -- hack
configure.in, which in turn built a Setup.config file -- was messy and
complicated.  (This was before the Makefiles were rewritten to be
non-recursive and the build process tidied up in various ways.  It's
quite possible that perhaps this will make adding modules to configure
and setup.py is no longer such a burning need.)

> #3) Python configure had the  configure options --with-tcl-config and
> --with-tk-config so I could say explicitly where what I want is. This

If a module is enabled in Modules/Setup, then setup.py won't try to
build it, so you could just enable Tkinter there with whatever options
are required.  

--amk


From rodrigc at mediaone.net  Sun Mar 25 01:47:55 2001
From: rodrigc at mediaone.net (rodrigc at mediaone.net)
Date: Sun, 25 Mar 2001 06:47:55 GMT
Subject: Python Mega Widgets rpm spec file?
Message-ID: <vwgv6.1085$107.1233792@typhoon.ne.mediaone.net>

Hi,

I am using John Grayson's book "Python and Tkinter Programming"
to learn how to develop GUI's under Python.
I developing under Redhat Linux 7.0 which comes with
Python 1.52 and Tkinter.

Grayson's book refers to Python Mega Widgets (Pmw) quite
a bit, but this does not come standard with Redhat distributions.
Has anyone made an rpm spec file for Pmw, so that it
can be installed as an rpm package?

I like to work on Redhat systems, and would like to deploy
my GUI to other Redhat systems, without having to package
up 3rd party libraries which are not included as part of
the base distribution.

Thanks for any help.
-- 
Craig Rodrigues        
http://www.gis.net/~craigr    
rodrigc at mediaone.net          


From aleaxit at yahoo.com  Fri Mar 30 09:21:49 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 30 Mar 2001 16:21:49 +0200
Subject: Is inheritance broken?
References: <mailman.985883539.13017.python-list@python.org>
Message-ID: <9a24q50684@news1.newsguy.com>

"Jonathan Claggett" <jcc.ugm at ix.netcom.com> wrote in message
news:mailman.985883539.13017.python-list at python.org...
    [snip]
> styles (OOP Fu :-). I just checked* and it appears Scott Meyers has a book
> on effective programming which I now intend to purchase. Obviously, there
is

Be warned, though -- Scott's excellent book is specifically on _C++_.

If that's what you want, fine (just be sure to get the CD version,
it's MUCH richer & better value for money) -- but it's NOT a general
book on programming, nor, alas, is it about Python.


Alex





From hungjunglu at yahoo.com  Sun Mar 11 17:59:37 2001
From: hungjunglu at yahoo.com (hungjunglu at yahoo.com)
Date: Sun, 11 Mar 2001 22:59:37 -0000
Subject: Chinese text GIF file generation
Message-ID: <98h00p+bqc2@eGroups.com>

--- In python-list at y..., William Park <parkw at b...> wrote: 
> Because there is so many variables (ie. font, style, size, ...), I 
don't
> think there is package that does what you want.  However, you can 
use
> LaTeX to produce Postscript format, and then convert it into GIF.

There surely are existing packages. Just visit any of the Chinese 
news portals (e.g: www.sinanet.com) and you'll see what I mean. 
Unlike Korean or Japanese, or even Vietnamese, for Chinese websites, 
it is usually a better idea to present text in GIF than in multi-byte 
encoding schemes. The thing is, there is a big need out there for 
real-time or quasi-real-time on-the-fly GIF generation of Chinese 
text in GIF format. Anything that involves an intermediate manual 
process is unacceptable. (Besides, it's easy to do it manually on any 
PC: screen capture + format conversion and you are done. It's just 
time-consuming.) For websites containing multiple foreign languages 
in the USA, text-to-gif converters is a real necessity: you don't 
want to show jibberish to your visitors.

Writing these packages in Java is a bit involved, but not that hard. 

I am pretty sure that the C library has been written, since someone 
else has mentioned it to me a long time ago. Imagine the number of 
Chinese computer geeks out there: it is unthinkable that this thing 
has not been done. (Of course there might be legal issues with GIF, 
but PNG will be fine.)

Hung Jung





From paulp at ActiveState.com  Tue Mar 13 22:22:51 2001
From: paulp at ActiveState.com (Paul Prescod)
Date: Tue, 13 Mar 2001 19:22:51 -0800
Subject: Where can I find Docs on how to use the debugger please?
References: <3lltat8tgd86rknhgnqkij6bk9rkbqk824@4ax.com>
Message-ID: <3AAEE40B.4ACD89D7@ActiveState.com>

cmfinlay at SPAMmagnet.com.au wrote:
> 
> Where can I find Docs on how to use the debugger please?
> 
> I am using the Win32 version of Python 2.1
> 
> I would like to set a breakpoint
> and then later step down the source code line by line.
> and view the source code as it is executed.

There are actually several Python debuggers and it is hard to know which
one you are asking about. How did you get into the debugger? What is on
the menu bar? Or are you talking about the command line debugger?
-- 
Python:
    Programming the way
    Guido
    indented it.
       - (originated with Skip Montanaro?)



From bernhard at abnoba.intevation.de  Wed Mar 28 10:01:52 2001
From: bernhard at abnoba.intevation.de (Bernhard Reiter)
Date: 28 Mar 2001 15:01:52 GMT
Subject: Linux GPS program
References: <3AC01637.4EF3A6F2@crynwr.com>
    <99q3ob$ehr$1@newsserver.rrzn.uni-hannover.de>
    <3AC0D79E.ADED2832@crynwr.com>
Message-ID: <99sud0$m7a$1@newsserver.rrzn.uni-hannover.de>

In article <3AC0D79E.ADED2832 at crynwr.com>,
	Russell Nelson <nelson at crynwr.com> writes:
> Bernhard Reiter wrote:
>> will pygps be Free Software? I could not find anything about a
>> license in the cvs.
> 
> Yes, it will.  Right now, I want people to fetch it from the CVS so
> they have the latest-n-greatest.

Well right now there is no usage term blurb attached, so the default
would be that the user does not have the necessary freedoms to make
it Free Software.

If you want it to be Free Software, 
attach a license (maybe MIT style or LGPL).
Just put it into CVS. People will still have no tarball.
If you ask them for not distribution their changes on their own I do
not see why somebody would want to do this.

If they don't care for your copyright, they can still produce
"rogue" versions now...

>> If it was I will be happy to add it to the www.freegis.org
>> link collection.
> 
> It's free software now -- I just don't want rogue versions that would be
> hard for me to integrate back into the mainline branch.  Feel free
> to suggest changes.

	Bernhard

-- 
Professional Service around Free Software                (intevation.net)  
The FreeGIS Project                                         (freegis.org)
Association for a Free Informational Infrastructure            (ffii.org)
FSF Europe                                            	  (fsfeurope.org)


From corour01 at motorola.com  Wed Mar 21 06:57:15 2001
From: corour01 at motorola.com (O'Rourke Clodagh-corour01)
Date: Wed, 21 Mar 2001 11:57:15 -0000
Subject: Tkinter Problem - NameError
Message-ID: <13EE655665F4D311AB4D0008C789498A01EE2C91@zei02exm02.cork.cig.mot.com>

From: Fredrik Lundh [mailto:fredrik at pythonware.com]
>Looks like recursive from-import problems

Ahh yes thanks a million.  I was using import-* instead of import.
However now my functions which close the current window and open a new one don't work

it was for example

def newCellWindow
    self.destroy()
    Cell.(self.parent)

How do I call the constructor of the Cell class from another class to which Cell is imported?

Thanks again,
Clodagh.



From smnordby at yahoo.com  Sat Mar 31 19:07:43 2001
From: smnordby at yahoo.com (Steve)
Date: Sat, 31 Mar 2001 16:07:43 -0800
Subject: I want to learn PYTHON!
References: <1iqx6.3744$aD4.234302@news2.atl>
Message-ID: <3AC6714F.F3EAC690@yahoo.com>

Step one:
Go to http://www.python.org and download Python version 2.0 for Windows.

Step two:
Run it to install Python.

Step three:
Go to http://www.crosswinds.net/~agauld and follow his programming
tutorial.

Step four:
Join the Python Tutor mailing list (you'll find info on the python.org
website).

Step five:
Read this newsgroup regularly.

HTH
-SteveN-




KEVIN wrote:
> 
> Where do I start.
> 
> I am not a programmer but have an intermediate level of computer knowledge
> in Windows only.
> 
> --
> KEVIN
> 
> "Si hoc legere scis, nimis eruditionis habes."
> -----------------------------------------------------------------
> Please repond in the newsgroup.
> My email address is altered for anti-spam
> Just change .nOt to .net


From vecernik at aon.at  Wed Mar 28 09:32:35 2001
From: vecernik at aon.at (Oliver Vecernik)
Date: Wed, 28 Mar 2001 16:32:35 +0200
Subject: diff lists
References: <mailman.985787369.15447.python-list@python.org>
Message-ID: <3AC1F603.A1074A7B@aon.at>

"Mike C. Fletcher" wrote:
> >>> [ element for element in a if element not in b ] # python 2.0+

Works great! Thank you.

Regards,
Oliver


From max at alcyone.com  Wed Mar  7 23:53:15 2001
From: max at alcyone.com (Erik Max Francis)
Date: Wed, 07 Mar 2001 20:53:15 -0800
Subject: is it possible to use Tkinter on linux
References: <98714n$3p8$1@news2-2.kornet.net>
Message-ID: <3AA7103B.CE819252@alcyone.com>

rambo wrote:

> someone knows that it is possible to use Tkinter on linux?

Yes.

> if it is, what program I should need to use Tkinter on linux,
> how to install , which environment variable should I set.

An RPM is downloadable from

    http://www.python.org/2.0/

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ My life was better before I knew you.
\__/ Edith Wharton (to Morton Fullerton)
    Polly Wanna Cracka? / http://www.pollywannacracka.com/
 The Internet resource for interracial relationships.


From edcjones at erols.com  Sat Mar 17 12:29:46 2001
From: edcjones at erols.com (Edward C. Jones)
Date: Sat, 17 Mar 2001 12:29:46 -0500
Subject: Debugging Python extensions (was Re: image handling)
References: <mailman.984795862.9586.python-list@python.org> <2tLs6.1193$4N4.107379@newsc.telia.net>
Message-ID: <3AB39F0A.6EB4E1C5@erols.com>

Fredrik Lundh wrote:

<!-- buy 1.1 today, get 1.2 next week:
  http://www.pythonware.com/products/works/index.htm
  -->

Can this product or something like it be used to debug both the Python and the C
of a Python extesnion module?




From erno-news at erno.iki.fi  Tue Mar 27 05:45:43 2001
From: erno-news at erno.iki.fi (Erno Kuusela)
Date: 27 Mar 2001 13:45:43 +0300
Subject: Namespaces question
References: <mailman.985627029.26291.python-list@python.org>
Message-ID: <kubsqnwkvc.fsf@lasipalatsi.fi>

In article <mailman.985627029.26291.python-list at python.org>, Chris
Withers <chrisw at nipltd.com> writes:

| What should I be using instead of locals()?

you aren't supposed to alter the dictionary returned by locals.
try using exec if you really want to do this, although it would
probably be better to just put the values in a regular dictionary.

  -- erno



From tjg at exceptionalminds.com  Sat Mar 17 15:24:53 2001
From: tjg at exceptionalminds.com (Timothy Grant)
Date: Sat, 17 Mar 2001 12:24:53 -0800
Subject: md5 passwords
In-Reply-To: <87elvwvq59.fsf@psyche.dnsalias.org>; from c.evans@clear.net.nz on Sat, Mar 17, 2001 at 07:57:38PM +1300
References: <mailman.984779122.3483.python-list@python.org> <87elvwvq59.fsf@psyche.dnsalias.org>
Message-ID: <20010317122453.B24992@trufflehunter.avalongroup.net>

I will do. Thanks. Thanks to all who have been kind enought to
respond to this question.

On Sat, Mar 17, 2001 at 07:57:38PM +1300, Carey Evans wrote:
> Timothy Grant <tjg at exceptionalminds.com> writes:
> > I would like to be able to be able to os.system('useradd -p
> > md5/crypt username')
> 
> Have a look at Michal Wallace's `md5crypt.py' module, which should do
> this for you:
> 
>     http://www.sabren.net/code/python/crypt/

-- 
Stand Fast,
    tjg.

Timothy Grant                         tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.         <><       (503) 246-3630
>>>>>>>>>>>>>Linux, because rebooting is *NOT* normal<<<<<<<<<
>>>>This machine was last rebooted:  60 days     3 hours ago<<



From danielk at aracnet.com  Tue Mar 27 09:02:43 2001
From: danielk at aracnet.com (Daniel Klein)
Date: Tue, 27 Mar 2001 06:02:43 -0800
Subject: Executing external programs
References: <vhZv6.4205$45.21324@newsfeeds.bigpond.com>
Message-ID: <3971ct8tq60d44mc65b1758fkfho9jbfac@4ax.com>

On Tue, 27 Mar 2001 21:13:28 +1000, "Ben" <bencr at bigpond.com> wrote:

>Hi,
>
>this appears to be a simple problem ... but i can not find any documentation
>on this ... could anyone point me to a rescource that would have an answer
>to this solution ...
>
>All I want to do is
>
> run program1.exe -args
>
> ... when program1.exe is finished
>
>run program2.exe -args ...
>
>run program3.exe -args
>
>etc etc
>
> ie the next process waits for the preceding processes exit code before
>starting ... if I can avoid the overhead of having the command shell open
>everytime I run an exe that would be good.

Methinks you should look at os.system().

Daniel Klein


From guido at digicool.com  Sat Mar  3 15:07:09 2001
From: guido at digicool.com (Guido van Rossum)
Date: Sat, 03 Mar 2001 15:07:09 -0500
Subject: [Python-Dev] Re: Bug fix releases
In-Reply-To: Your message of "Sat, 03 Mar 2001 14:55:09 EST."
             <LNBBLJKPBEHFEDALKOLCEELHJDAA.tim.one@home.com> 
References: <LNBBLJKPBEHFEDALKOLCEELHJDAA.tim.one@home.com> 
Message-ID: <200103032007.PAA21925@cj20424-a.reston1.va.home.com>

> [Fred L. Drake, Jr.]
> > ...
> >   Managing the bugfix releases would also be an excellent task for
> > someone who's expecting to use the bugfix releases more than the
> > feature releases -- the mentality has to be right for the task.  I
> > know I'm much more of a "features" person, and would have a hard time
> > not crossing the line if it were up to me what went into a bugfix
> > release.

[Uncle Timmy]
> Note there was never a bugfix release for 1.5.2, despite that 1.5.2 had some
> serious bugs, and that 1.5.2 was current for an unprecedentedly long time.
> Guido put out a call for volunteers to produce a 1.5.2 bugfix release, but
> nobody responded.  Past is prelude ...
> 
> everyone-is-generous-with-everyone-else's-time-ly y'rs  - tim

I understand the warning.  How about the following (and then I really
have to go write my keynote speech :-).  PythonLabs will make sure
that it will happen.  But how much stuff goes into the bugfix release
is up to the community.

We'll give SourceForge commit privileges to individuals who want to do
serious work on the bugfix branch -- but before you get commit
privileges, you must first show that you know what you are doing by
submitting useful patches through the SourceForge patch mananger.

Since a lot of the 2.0.1 effort will be deciding which code from 2.1
to merge back into 2.0.1, it may not make sense to upload context
diffs to SourceForge.  Instead, we'll accept reasoned instructions for
specific patches to be merged back.  Instructions like "cvs update
-j<rev1> -j<rev2> <file>" are very helpful; please also explain why!

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



From aahz at panix.com  Mon Mar 26 18:46:35 2001
From: aahz at panix.com (Aahz Maruch)
Date: 26 Mar 2001 15:46:35 -0800
Subject: warrior geeks
References: <3ABF72F1.98CB78F@sage.att.com> <mailman.985634134.9497.python-list@python.org>
Message-ID: <99okcr$mt1$1@panix3.panix.com>

In article <mailman.985634134.9497.python-list at python.org>,
Tim Churches  <tchur at optushome.com.au> wrote:
>
>I'll use our time zone advantage to jump in early on this thread and
>opine that it is way off-topic. Not that I am opposed to
>marital...whoops...martial arts (although they are not my cup of tea),
>but I am concerned that if this threads continues, sooner or later
>someone will mention the whole "Geeks with Guns" thing that Eric S.
>Raymond promotes (see http://www.tuxedo.org/~esr/geeks-with-guns/ or for
>even worse IMHO see http://www.tuxedo.org/~esr/guns/ ), especially given
>ESR's apparent fondness for Python. That sort of thing gives me a fit of
>the shudders (playing with guns, not ESR's fondness for Python). 

What are you?  Some kind of Commie?  Guns form the very foundation of
the Pax Americana that gives you the freedom to use Python.
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

"Boost the stock market -- fire someone"


From mda at idatar.com  Sat Mar 17 13:36:13 2001
From: mda at idatar.com (David Allen)
Date: Sat, 17 Mar 2001 18:36:13 GMT
Subject: image handling
References: <mailman.984795862.9586.python-list@python.org>
Message-ID: <x8Os6.501$Fb.771198@typhoon2.ba-dsg.net>

In article <mailman.984795862.9586.python-list at python.org>, "Michael Hall"
<admin at mulga.com.au> wrote:

> G'day people:
> 
> I'm brand new on this list and am looking for resources about image
> processing in a web environment in Python. I'm wondering if there exists a
> module that does things like the PHP image handling functions
> (creating, copying, resizing, etc). 
> 
> I've had a look at the Python Functions Library but there doesn't appear to
> be much in there. I've started to use Python a bit in a web environment and
> like what I've seen so far, and from what I've heard I'm sure there must be
> some pretty flash image handling capabilities hanging around somewhere.

Sounds like you're looking for the Python Imaging 
Library (PIL)

http://www.pythonware.com/library/the-pythonware-image-library.htm
-- 
David Allen
http://opop.nols.com/
----------------------------------------
Be wary of strong drink. It can make you shoot at tax collectors and 
miss 
- Lazarus Long, "Time Enough for Love"


From aleaxit at yahoo.com  Thu Mar 15 07:11:39 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 15 Mar 2001 13:11:39 +0100
Subject: This math scares me
References: <3aad4062.1290021@News.CIS.DFN.DE> <20010312170356.B3513@better.net> <mailman.984480856.18097.python-list@python.org> <984486395.407379@newsmaster-04.atnet.at> <slrn9asf1d.ja3.qrczak@qrnik.zagroda> <984518199.127140@newsmaster-04.atnet.at> <jwbnews-09D75F.13313313032001@corp.supernews.com> <98ohuc$du9$1@news.udel.edu> <7bQr6.644$lz1.17686@ruti.visi.com> <984610248.564504@newsmaster-04.atnet.at>
Message-ID: <98qbhk02dck@news1.newsguy.com>

"Werner Schiendl" <ws-news at gmx.at> wrote in message
news:984610248.564504 at newsmaster-04.atnet.at...
> > In fiancial calculations you also need to represent numbers
> > that aren't amounts of money.  Interest rates for example.
>
> How about 1/100 of a percent?

(Normally called a "basis point", btw -- at least when one is
discussing _changes_ in a security's yield).


Alex





From winkler at abaqus-sn.com  Thu Mar 29 12:49:09 2001
From: winkler at abaqus-sn.com (david winkler)
Date: Thu, 29 Mar 2001 17:49:09 GMT
Subject: how query function args
Message-ID: <3AC375A5.507C4D73@abaqus-sn.com>

I need to query the names and values of a function's keyword arguments.
I
suspect that there is an easy way to do this but cannot find it.  Would
someone care to make a suggestion?



From aleaxit at yahoo.com  Fri Mar 30 15:46:44 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 30 Mar 2001 22:46:44 +0200
Subject: best practice? null/None (correction)
References: <9a2b9v+lrgp@eGroups.com> <mailman.985976988.3926.python-list@python.org>
Message-ID: <9a2rnd011au@news1.newsguy.com>

"Robin Thomas" <robin.thomas at starmedia.net> wrote in message
news:mailman.985976988.3926.python-list at python.org...
> At 11:58 AM 3/30/01 -0500, Robin Thomas wrote:
> >HTTP query arguments are of one type: STRING. There is no concept of a
> >"null" value; the closest thing you have is an "empty" string like "". In
> >most database engines, there is a clear difference between NULL and "" in
> >any textual or character data column.
>
> I'm sorry to be outright WRONG here, fluxent.
>
> Many databases consider "" to be NULL for textual columns. Oracle 7 and up
> certainly does.

Wok?!  I _thought_ Oracle was an RDBMS -- fun to hear that it isn't
(proper treatment of NULL as distinguished from any data value being
one crucial requisite for being called an RDBMS).  Guess I'll stick with
SQL Server/MSDE then...


Alex





From neotaku at free.fr  Mon Mar  5 10:23:00 2001
From: neotaku at free.fr (Neotaku)
Date: Mon, 05 Mar 2001 15:23:00 GMT
Subject: Access to sybase API
Message-ID: <obOo6.1158$f07.1696858@nnrp6.proxad.net>

Hello everybody

I search a code source for python to access Sybase thru API connexion the
Sybase DB-Library API or Sybase CT-Library (aka the Client Library) API

Thanks a lot
see ya



From jmarshal at mathworks.com  Wed Mar 14 12:49:08 2001
From: jmarshal at mathworks.com (Joshua Marshall)
Date: 14 Mar 2001 17:49:08 GMT
Subject: Who's minister of propaganda this week?
References: <B98041329020E007.7DA84CAD9D0562AB.817BBC572F480869@lp.airnews.net> <98nu1m$12e@dispatch.concentric.net> <3AAFA861.C773D977@san.rr.com>
Message-ID: <98oauk$j8j$1@news.mathworks.com>

Darren New <dnew at san.rr.com> wrote:
> Phlip wrote:
>> Ruby trumps Python in the simple matter of
>> closures (imagine my shock when I discovered Python didn't have them). 

> Can someone summarize the important difference between closures and object
> instances? Other than convenience, I mean, it would seem that something like

> class yadda:
>   def __init__(self, a, b):
>     self.a = a ; self.b = b
>   def __call__(self, x, y):
>     return (self.a + x * self.b * y)

> would seem to be quite a lot like a closure with a and b bound and x and y
> free?

Yup.  Object instances and closures are very similar.  Scoping issues
aside, you can think of closures as classless instances with only one
method.


From whisper at oz.net  Sat Mar 10 20:14:35 2001
From: whisper at oz.net (Dave LeBlanc)
Date: 11 Mar 2001 01:14:35 GMT
Subject: Urgent Question about Python licensing
References: <mailman.984254413.27254.python-list@python.org>
Message-ID: <98ejhr$38q$0@216.39.170.247>

Dear Tim;

When UCITA first came out, there was a lot of press about it's evils.
Statements where made that a licensor could impose new license clauses
after the fact and unilaterally, that "shrink wrap" licenses became
automagically binding, that a licensor could make a choice of venue
for resolution of legal issues anywhere in the world, and that a
licensor could create software that the licensor could remotely
disable (presumably via the internet) the software you've licensed.

After your statement that that was a "most concentrated pile of
absurdist FUD", I went back and did some further research intent on
proving you incorrect. After all, this was reported by some regular
news organizations (InfoWord, CNET etc.) so how could they be wrong?!
However, after reviewing information at http://www.ucitaonline.com/
and at http://www.2bguide.com, I came to the conclusion that i'd been
mislead. Imagine! being mislead by the press! <g>.  (I should have
known better - i'm a firm believer in the "believe none of what you
hear and only half of what you read" school of thought.)

As it turns out, after the fact license changes are NOT binding,
"shrink wrap" or "click wrap" licenses ARE binding and have been
upheld by the courts, can't make a bogus venue choice, and that a
licensor CAN unilaterally disable software but only after a 2 week
warning in many cases.

http://www.2bguide.com/newsart.html does have a lot of links to the
news articles I recall about the evils of the UCITA, but
http://www.ucitaonline.com has a good FAQ which addresses and rebuts
the claims of it's evilness. FWIW, this FAQ is worth reading,
particularly if you're a small software business entrepeneur.  UCITA
adds some implicit waranties worth checking into.

It also turns out that Virginia's law adopting UCTIA doesn't take
effect until July 2001, so whatever laws Virginia currently has would
apply to Python 1.6+ and not the UCTIA.

So, I was incorrect wrt to my statement about unilateraly chages etc.
My apologies to any who where mislead and alarmed by my post.

Moving along, I find it ironic that the Open Software Foundation calls
the GPL an open source license since it violates the rules given for
such. GPL software has source available, but it's use is constrained
and contaminating (unlike the GLL library license).

Regards,

Dave LeBlanc

P.S Guido is welcome to my first born if I ever have one - but he has
to do his own cooking.

On Sat, 10 Mar 2001 14:59:21 -0500, "Tim Peters" <tim.one at home.com>
wrote:
<snip>
>> 3. Is Python 2.0 covered by that evil CNRI 1.6 license?
>
>2.0 is covered by CNRI's 1.6 license, as well as by BeOpen.com's minor
>variation on that.  I disagree that it's evil, although the way it was
>imposed on the community was exceedingly ham-handed, and it has created
>problems due to its purported incompatibility with the GPL.
>
>> I avoided 1.6 mostly because of the license, and thought that 2.0 had
>> "skipped" 1.6 and used a far more equitable license then the 1.6 one.
>
>I suggest you read the licenses in question (you clearly have not yet).
>
>> 1.6's license permits the copyright holder to suddenly decide that
>> anyone using the covered software must pay a fee (or any other thing they
>> so desire - this is only an example), and by virtue of your even so much
>> as having read the license are implicitly bound to their unilateral
>> declaration - lovely Virginia laws).
>
>No offense intended, Dave, but this is the most concentrated pile of
>absurdist FUD I've seen in a solid year of ill-informed license rants.  Even
>by the FSF's standards, the CNRI 1.6 license is a "free software" license
>(see "The License of Python 1.6b1 and later versions", at:
>
>    http://www.fsf.org/philosophy/license-list.html
>
>).
>
>the-2.1-license-will-require-giving-your-firstborn-to-guido-ly y'rs  - tim
>
>



From porter at et.byu.edu  Tue Mar 20 13:53:00 2001
From: porter at et.byu.edu (C. Porter Bassett)
Date: Tue, 20 Mar 2001 11:53:00 -0700 (MST)
Subject: Insertin **keywords into a class
In-Reply-To: <99877b$uss$1@nntp6.u.washington.edu>
References: <mailman.984971990.29857.python-list@python.org> <9960i1$tp0$1@nntp6.u.washington.edu> <99661m$scc$1@panix6.panix.com> <99877b$uss$1@nntp6.u.washington.edu>
Message-ID: <Pine.GHP.4.21.0103201151090.4991-100000@don.et.byu.edu>

cCan you please tell me why the following code snippet doesn't update the
value of radius?


class myClass:
   def __init__(self, *arguments, **keywords):
      self.radius = 0.0
      for kw in keywords.keys():
         print kw, ":", keywords[kw]
         self.kw = keywords[kw]
      print "self.radius =", self.radius


b = myClass(radius = 1.0)
                   

--------------------------------------------------------------------------
"Pretend like this is a really witty saying." - Anonymous
--------------------------------------------------------------------------





From rtexier at elikya.com  Fri Mar 30 03:11:12 2001
From: rtexier at elikya.com (Romuald Texier)
Date: Fri, 30 Mar 2001 10:11:12 +0200
Subject: htmllib samples
References: <3AC2F48E.4EF77C71@aon.at> <3AC3296D.E24DBA02@downright.com>
Message-ID: <9a1f2i$72v$1@news.irisa.fr>

Doug Fort wrote:


> I've attached a simple parser we use.  It may help.
> 
> Note that Frederick  Lundh modestly failed to mention his upcoming book on
> 'The Python Library'.  There's an ebook version of his 'eff-bot' Guide,
> which is the  first place I look for code  examples.
> 

It looks very interesting but what is this webnudge module you are using? I 
can't find it anywhere...

Thank you,

-- 
Romuald Texier


From gkiffney at my-deja.com  Thu Mar 22 11:45:13 2001
From: gkiffney at my-deja.com (gkiffney)
Date: Thu, 22 Mar 2001 11:45:13 -0500
Subject: os.system output
References: <mailman.985192771.32528.python-list@python.org>
Message-ID: <9%pu6.792$9%.208326@newsr1.maine.rr.com>

If  you're on Unix, you can use 'commands'.
    import commands
    res = commands.getoutput('ls')

and res will be a long string with the output of 'ls'.
To make a list out of res, you'd just do
    import string
    dirlist = string.split(res, '\n')

"Jonathan Soons" <jsoons at juilliard.edu> wrote in message
news:mailman.985192771.32528.python-list at python.org...
> I am trying to save the output of
>
> >>>dirs = os.system('ls')
>
> into a variable but I do not need any output to
> the screen (altho it doesn't hurt). How do I
> go about this?
> My fat book only explains how to get output to the screen.
>
> I could do
>
> >>>os.system('ls > file')
> >>>f = open('file')
> >>>dirs = f.readlines()
>
> but it looks crude. Is there a better way?
>




From fredrik at pythonware.com  Fri Mar 16 12:41:56 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 16 Mar 2001 17:41:56 GMT
Subject: Python style questions
References: <98tb91$9t9$1@saltmine.radix.net> <98tclh$ap6$2@lahti.tasking.nl> <98th1k$ka2$1@saltmine.radix.net>
Message-ID: <Efss6.183$sk3.52641@newsb.telia.net>

Cary O'Brien wrote:
> Dick Streefland: <dick_streefland at tasking.com> wrote:
> >c. use string.atoi(s, 16), or for Python >= 2.0, the builtin int(s, 16).
>
> Hmm...
>
> >>> int('abcd',16)
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
> TypeError: int requires exactly 1 argument; 2 given
> >>>
>
> Is this a 2.0 feature?

that's probably what Dick meant with "Python >= 2.0", don't
you think?

Cheers /F




From db3l at fitlinxx.com  Tue Mar 13 18:08:24 2001
From: db3l at fitlinxx.com (David Bolen)
Date: 13 Mar 2001 18:08:24 -0500
Subject: Divorcing Command Line and GUI
References: <mailman.984160633.14724.python-list@python.org> <98dpoi32ckn@news1.newsguy.com>
Message-ID: <usnkh1d3r.fsf@ctwd0143.fitlinxx.com>

"Alex Martelli" <aleaxit at yahoo.com> writes:

> Piece of cake -- just make sure your CL tools take a switch
> (-u or --unbuffered) to work without any input buffers on
> stdin/stdout, and similarly-unbuffered Python GUI front-ends
> will be easily able to drive them (after an os.popen2).

Or just pass the "-u" to python itself when running the command line
script - no changes to scripts required.

(Something I find extremely helpful to work with pre-existing scripts,
unlike Perl where you have to modify the script itself to enable
autoflushing.)

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


From chrislamb at btinternet.com  Sat Mar 24 05:48:42 2001
From: chrislamb at btinternet.com (chris)
Date: Sat, 24 Mar 2001 10:48:42 +0000
Subject: learning python...
References: <3AB8F4E1.184DD9C@noaa.gov>
Message-ID: <a2uh99.o71.ln@pingu>

Benjamin.Altman wrote:

> Hello.  From the perspective of someone who already has programming
> experience, would anyone know if O'reillys "Learning Python" book would
> be better than say Prentice Hall's "Core Python Programming"?  Or would
> it be better to go straight into something like "Programming Python"?
> 
> Thanks,
> Ben

"Core Python Programming "from PH is easily the better book for easing into 
Python. I am an O'Reilly fan and generally buy their books blind if it is 
on a subject I am interested in (since they very rarely disappoint). 
However on this occassion the PH book is a genuinely good book and worth 
buying, considerably more than the Learning Python book and the 2e version 
of Programming Python.

My opinion!

Chris


From emile at fenx.com  Tue Mar 27 10:22:15 2001
From: emile at fenx.com (Emile van Sebille)
Date: Tue, 27 Mar 2001 07:22:15 -0800
Subject: who is my caller?
References: <mailman.985700854.31956.python-list@python.org> <slrn9c18o0.ojm.scarblac@pino.selwerd.nl> <99q8nv$2aibn$1@ID-11957.news.dfncis.de>
Message-ID: <99qbdu$2a0et$1@ID-11957.news.dfncis.de>

Well yes, but traceback gets its info by raising an exception anyway.

--

Emile van Sebille
emile at fenx.com

---------
"Emile van Sebille" <emile at fenx.com> wrote in message
news:99q8nv$2aibn$1 at ID-11957.news.dfncis.de...
>
> "Remco Gerlich" <scarblac at pino.selwerd.nl> wrote in message
> news:slrn9c18o0.ojm.scarblac at pino.selwerd.nl...
> <snip>
> >
> > Anyways, yes, it's more like black magic in Python. There is a call
stack
> > you can inspect, but only if you just caught an exception.
> >
>
> If I'm not mistaken, with 2.0 you can directly import traceback without
> needing to generate an exception first.
>
>
> --
>
> Emile van Sebille
> emile at fenx.com
>
>




From shredwheat at mediaone.net  Mon Mar 19 10:40:51 2001
From: shredwheat at mediaone.net (Pete Shinners)
Date: Mon, 19 Mar 2001 15:40:51 GMT
Subject: newbie - infinite loop 
References: <985012513.18971.0.nnrp-12.c1c3e154@news.demon.co.uk>
Message-ID: <7Mpt6.72006$Ch.9139356@typhoon.we.rr.com>

"Paul Brian" <pbrian at demon.net> wrote
> Surely this should not happen.  Surely the line 3 condition "for every
> element in this list", should not constantly expand but should be set solid
> till the next pass?
>
> What have I done wrong - is it the mutable list v the immutable tuple
> problem (but it should work..surely?) Or is this just the way things are -
> "live with it and do not try anything like this again?"
>
> thanks in advance
> ==============
> uniqueList = ['q']
> line = 'q'
>
> for listElement in uniqueList:   #infinite loop
>     if listElement == line:
>         uniqueList.append(line)
>
> print len(uniqueList)

hello paul. this code should indeed loop forever. let's look and see why...

it sounded like in your message, you only wanted to add items to the
list if they weren't already in it. this section of code is doing
something different. this is going through the list, and everytime it
finds a match for a word, it is adding that word again. the way this
loop is setup, the word will be added to the end of the list, and it
will loop over the appended elements in the list. because of this, the
code will keep finding the same word on the list and keep appending the
same word to the end... infinitely.

obviously this is not the best solution. there are several other ways
you could do this. the way i would probably do it (not the best, likely)
would replace the loop with this;

if line not in uniqueList:
    uniqueList.append(line)

it ends up being somewhat trickier to loop through a list and
make sure an element isn't found. since it sounded like you wanted
to see your loop fixed, i'll take a stab at it. remember, an 'else'
statement in a for loop is executed when the loop is not 'break'ed

for listElement in uniqueList:
    if listElement == list:
        break
else:
    uniqueList.append(list)


good luck, hopefully this was able to teach you a little something
about python. btw, the looping method will definitely be slower than
the "in" version.





From stephen_purcell at yahoo.com  Mon Mar 19 03:51:56 2001
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Mon, 19 Mar 2001 09:51:56 +0100
Subject: How to find out about free space on a given drive?
In-Reply-To: <3AB5C19E.FA40DDA9@darwin.in-berlin.de>; from gherman@darwin.in-berlin.de on Mon, Mar 19, 2001 at 09:21:50AM +0100
References: <3AB5C19E.FA40DDA9@darwin.in-berlin.de>
Message-ID: <20010319095156.A12242@freedom.puma-ag.com>

Dinu Gherman wrote:
> is there a simple way (or any way at all) to find out for 
> any given hard disk how much free space is left on that
> device? I looked into the os module, but either not hard
> enough or there is no such function. Of course, the ideal
> solution would be platform-independant, too... :)

os.statvfs() seems to be what you want for UNIX. I don't think you'll find
a cross-platform solution.

-Steve

-- 
Steve Purcell, Pythangelist
Get testing at http://pyunit.sourceforge.net/
Any opinions expressed herein are my own and not necessarily those of Yahoo



From aleaxit at yahoo.com  Mon Mar 26 14:22:13 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Mon, 26 Mar 2001 21:22:13 +0200
Subject: Is it possible to sort a dictionary?
References: <GEtv6.59583$Lm2.6378825@news0.telusplanet.net> <3ABE771B.7B9E1631@student.liu.se> <99o33k$aq4$0@216.39.170.247>
Message-ID: <99o51301qgb@news1.newsguy.com>

"Dave LeBlanc" <whisper at oz.net> wrote in message
news:99o33k$aq4$0 at 216.39.170.247...
> How about:
>
> # prints the elements of myDictionary in sort order
> # error handling needed (myKeys != none for example)
> myKeys = myDictionary.keys()
> myKeys.sort()
> for key in myKeys:
> print myDictionary[key]

Fine!  Even faster & simpler:

    myItems = myDictionary.Items()
    myItems.sort()
    for key, value in myItems:
        print value


Alex





From dan at eevolved.com  Thu Mar  1 18:07:36 2001
From: dan at eevolved.com (Dan Parisien)
Date: Thu, 01 Mar 2001 23:07:36 GMT
Subject: Lower bounds of lists
References: <2mjt9tcfa0o55ap99p452trjj3u7h3qst6@4ax.com>
Message-ID: <YCAn6.269034$Pm2.4216141@news20.bellglobal.com>

Michael Prager wrote:

> As a newcomer to Python, I'm curious about the choice to number
> list elements starting at zero, rather than one.  My impression
> is that there is no user option to change that in a specific
> program.  Correct?

correct.
 
> It seems an odd choice, as humans count from one, and Python in
> other respects seems quite logical and well thought out.

The reason this is so is historical. Writing a program involved having data 
contained in addresses in memory. Data's addresses were always offsets from 
a set address (say 5 bytes from the begining). Because of that it starts at 
zero (0 bytes from the beginning _is_ the beginning)

It's the case in a lot of programming languages as well. 

Hope this helps,
Dan


From aleaxit at yahoo.com  Wed Mar  7 05:25:46 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Wed, 7 Mar 2001 11:25:46 +0100
Subject: Can anyone recomend a good intoduction to C...
References: <mailman.983894288.19965.python-list@python.org> <983896344.701456@newsmaster-04.atnet.at> <m3lmqibw0a.fsf@atrus.jesus.cam.ac.uk> <983911749.217622@newsmaster-04.atnet.at> <UScp6.1097$y6.196347@ruti.visi.com> <983916572.286918@newsmaster-04.atnet.at>
Message-ID: <9852b8015jm@news1.newsguy.com>

"Werner Schiendl" <ws-news at gmx.at> wrote in message
news:983916572.286918 at newsmaster-04.atnet.at...
> > C++ is far more complex that C without being much safer.  Since
> > the safety has to be "inspected" into the program, you want
> > to make things as easy to inspect as possible: a nice simple
> > language.
>
> That is of course a point, if you need to let the program inspect you will
> try to keep it as simple as possible.
> But given the same functionality, I think a C++ program will not be more
> complex.

The program will not be, the language is.  It's a hard-to-judge tradeoff.

> ...as long as it is well designed of course, nobody forces you to use any
> number of features at all possible
>
> don't you think so?

I do not: C++ *forces* "any number of features" on you -- you can't
easily subset it and ENSURE features outside of the subset are not
intruding.

> Think about string operations: I think about 50 % of all security issues I
> read about are somehow related with a (string) buffer overflow. The string
> family of classes does not have that kind of problem.

No, but it does have other issues -- e.g.,
    std::string oneletter('A', 1);
does NOT do what one might expect, nor does it give any warning
in the implementations I have at hand (oneletter becomes a string
of 65 characters all equal to '\001', not one of 1 character
equal to 'A' -- oops, arguments the wrong way around...!-).

The trade-off is no doubt a win for std::string here -- it
removes many more issues than it introduces.  But it's not
so clear overall -- considering that you can't subset, not
really.  I think that C++ (used properly, particularly with
STRONG reliance on GOOD libraries) is a definite net win
over C, but I can well see somebody might disagree, most
particularly when security-crucial applications are being
targeted.  (If I _could_ throw a switch to subset C++,
_ensuring_ at least warnings for a zillion potential errors
and complexities, the picture would of course change).


Alex





From aleaxit at yahoo.com  Mon Mar  5 14:54:08 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Mon, 5 Mar 2001 20:54:08 +0100
Subject: Name of the Script
References: <20010304.185354.2114937732.8277@localhost.localdomain>    <NJCo6.6720$qv3.2834377@nnrp5-w.sbc.net> <u0i089.7it.ln@127.0.0.1>
Message-ID: <980qsk02m1g@news1.newsguy.com>

"David Given" <dg at pearl.tao.co.uk> wrote in message
news:u0i089.7it.ln at 127.0.0.1...
    [snip]
> in argv[0]. (This is the same as a the generic Unix question: `How do I
> find out where my executable lives?' It's basically not solveable.) I
> don't know about Windows, though.

In Windows, location of the executable and/or dll IS solvable (there's
a system-call, or API if you will, expressly for that), but you still would
not learn 'exactly how' it was invoked (through ShellExecute, or through
CreateProcess [which underlies it], through a shortcut, etc) -- and it
does not help for an interpreter (you'd only learn about Python.Exe or
thereabouts, not which script it started out executing!).


Alex





From s713221 at student.gu.edu.au  Mon Mar 19 19:04:05 2001
From: s713221 at student.gu.edu.au (s713221 at student.gu.edu.au)
Date: Tue, 20 Mar 2001 10:04:05 +1000
Subject: Organizing Python
References: <995ogm$1u6$0@216.39.170.247> <zoeh4e4m.fsf@propylon.com>
Message-ID: <3AB69E75.CEDBC8AE@student.gu.edu.au>

> from the module foo. A module is usually represented by a file, but it
> is also possible to have it be a directory which I once read something
> about on python.org, but I can't find it now. 

If you place a file called __init__ into a directory on your path,
python imports that as the name of the directory.

/python/path/FFT
		|
		----->__init__.py #contains python instructions to load code from
other
		|		files
		|
		----->other/files/and/directories which make up FFT

>>>import FFT	#Imports FFT/__init__.py
>>>print "Handy, ain't it?"

Another method is to put a file called module_name.pth into your python
path, containing the name of a directory that the module is in. (What
I've used to get Numeric working without fiddling with my /etc/profile
or ~/.bashrc files.).

less /usr/lib/python2.0/site-packages/Numeric.pth

Numeric

In this case, python loads the file with the name of the module

/usr/lib/python2.0/site-packages/Numeric
					|
					----> Numeric.py
					|
					----> other numeric files

>>>import Numeric	# Imports Numeric/Numeric.py
>>>print "A little bit messier, but maybe what the original poster was looking for?"

Joal Heagney/AncientHart

Joal Heagney/AncientHart


From fredrik at pythonware.com  Tue Mar 20 17:45:10 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Tue, 20 Mar 2001 22:45:10 GMT
Subject: ASCII encoding error: ordinal not in range(128)
References: <91o5ih$2kv$1@news.wrc.xerox.com> <20001219172933.B29727@ActiveState.com> <mailman.985123475.18421.python-list@python.org>
Message-ID: <W3Rt6.2582$4N4.344041@newsc.telia.net>

Jose Isaias Cabrera wrote:

> UnicodeError: ASCII encoding error: ordinal not in range(128)

translated to english, this overly cute error message attempts
to say something like "cannot convert this Unicode string to
ASCII as it contains non-ASCII characters"

> I've search on python sites for this, and though there is
> something about Unicode, there is no python example code
> to use.
>
> Does anyone have any idea how to fix this?

summary: someone's handing you unicode strings (probably
the COM interface), and you're trying to use them as 8-bit
strings (probably by passing them to a method that expects
8-bit data) without specifying an encoding.

(if you want me to be more precise, you have to show us
more code).

to solve this, you have to add an explicit conversion to the
right place.  e.g:

    s = u.encode("latin-1")

where "u" is a unicode string and "s" an 8-bit ISO Latin string.

(if you don't do this, Python will convert the string anyway when
you pass it to an 8-bit function.  the default conversion assumes
ASCII, which is why you get that exception)

Hope this helps!

Cheers /F




From thomas at xs4all.net  Mon Mar  5 10:33:44 2001
From: thomas at xs4all.net (Thomas Wouters)
Date: Mon, 5 Mar 2001 16:33:44 +0100
Subject: Overriding UserList methods
In-Reply-To: <2oc6atgkn0ror77rj3b6j8d8d2s0gaa5k7@4ax.com>; from danielk@aracnet.com on Sun, Mar 04, 2001 at 10:35:27PM -0800
References: <b566atck0ruu4tae8n1srrkntbmme26gor@4ax.com> <2oc6atgkn0ror77rj3b6j8d8d2s0gaa5k7@4ax.com>
Message-ID: <20010305163344.A9678@xs4all.nl>

On Sun, Mar 04, 2001 at 10:35:27PM -0800, Daniel Klein wrote:
> I found the answer in the Beazley book. It's

> def append(self, object):
>         self.data.append(object)

> using the 'data' instance variable automatically set up in the UserList
> class. But I'll try the other suggestions anyway just to get a better
> understanding.

Note that though this method is 'allowed' with UserList instances (because
'data' is not a private variable) it does break the encapsulation. If you
were to change your class to derive from OtherModule.ExtendedUserList, which
did some extra funky stuff before/after really appending to the data, you
wouldn't get the new functionality. If you just used
'UserList.UserList.append(self, data)', you would, though you would have to
remember to change all references from 'UserList.UserList' to
'OtherModule.ExtendedUserList'.

-- 
Thomas Wouters <thomas at xs4all.net>

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



From grante at visi.com  Sat Mar 31 01:22:40 2001
From: grante at visi.com (Grant Edwards)
Date: Sat, 31 Mar 2001 06:22:40 GMT
Subject: Know of Substantial Apps Written in Python?
References: <3ac4e4a8_1@news4.newsfeeds.com> <KE6x6.415$SR6.235983@typhoon1.ba-dsg.net> <dj7x6.9561$SB2.850896@ruti.visi.com> <3AC5529D.8D7BCC63@seebelow.org>
Message-ID: <slrn9cb52o.qq.grante@tuxtop.visi.com>

On Fri, 30 Mar 2001 21:44:29 -0600, Grant Griffin <not.this at seebelow.org> wrote:

Hi Grant,

>Hey, I looked at that, and it looks cool.  But I was somewhat
>puzzled by the following statement:
>
>   Sketch is an interactive vector drawing program for Linux
>   and other UNIX compatible systems. <snip>
>   
>   A somewhat unique(for a drawing program) feature of Sketch
>   is that it is implemented almost completely in a very
>   high-level, interpreted language, Python.
>
>If it's implemented almost completely in Python, why does it
>only run on "UNIX-compatible" systems?  (BTW, does it delete
>hyphens? <wink>) Is there something about its "almost" part
>that can't be done in Windows.

I believe it uses standard Tcl/Tk, PIL, and one module with
support functions written in C.  I would guess that the C stuff
has never been ported to non-Unix platforms.  It looks like
some of the low-level drawing and rendering stuff is written in
C (I presume for performance reasons), as well as some
file-parsing and font manipulation routines.

Hr. Herzog (Sketch's author) has been known to read c.l.p, and
knows more about it that I do (I'm just a happy user -- since
before I even knew what Python was.)

-- 
Grant Edwards                   grante             Yow!  I smell a RANCID
                                  at               CORN DOG!
                               visi.com            


From syver at NOSPAMcyberwatcher.com  Fri Mar 30 15:41:55 2001
From: syver at NOSPAMcyberwatcher.com (Syver Enstad)
Date: Fri, 30 Mar 2001 22:41:55 +0200
Subject: Threading legality/morality
References: <fsDv6.1247$NR.100201@news3.oke.nextra.no> <99nnbv$phs$1@panix6.panix.com> <h0%v6.1842$NR.144463@news3.oke.nextra.no> <99qa3v$6e0$1@panix6.panix.com> <Pine.SOL.4.30.0103271851370.7646-100000@mimosa.csv.warwick.ac.uk> <99qogp$o6g$2@troll.powertech.no>
Message-ID: <9a2r6n$mlq$1@troll.powertech.no>

I have a confession to make here. It seems that this behaviour is a feature
rather than a bug and has to do with the os rather than the app or python
itself for that matter. (Inserts foot in mouth).

If you click the mouse in the client area of a console window, the os
freezes the application running there until you press a key. I guess it's
for the convenience of copying text off of the console but it sure tripped
me up.


"Syver Enstad" <syver at NOSPAMcyberwatcher.com> wrote in message
news:99qogp$o6g$2 at troll.powertech.no...
>
> "John J. Lee" <phrxy at csv.warwick.ac.uk> wrote in message > Perhaps you're
> right, but I've seen the same behaiviour on win95, without
> > multi-threading.  When you run command-line (do we mean the same thing
by
> > console and command-line?) apps it freezes until a key gets pressed.
Not
> > using urllib either.
> >
> > It is possible (likely, even) that I was using cygwin, so maybe that's
> > where the blame lies.  I'll see if I can reproduce it with the standard
> > DOS shell.
>
> I am using console program/command line whatever you like, and I am not
> using Cygwin. I've experienced this both on the BeOpen and ActiveState
> distros.
>
>




From mwh21 at cam.ac.uk  Mon Mar  5 14:53:23 2001
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 05 Mar 2001 19:53:23 +0000
Subject: ANNOUNCE: Python - The RPG
References: <97he34$jhc$1@newshost.accu.uu.nl>
Message-ID: <m3zof0at70.fsf@atrus.jesus.cam.ac.uk>

m.faassen at vet.uu.nl (Martijn Faassen) writes:

>   Jim Fulton
[snip]
>     Disadvantages:
> 
>       * Can inadvertantly make people's heads explode when attempting
>         to explain how something works, rendering them unable to think
>         very well for a while.

This is a disadvantage *how*?

Cheers,
M.

-- 
  Hmmm... its Sunday afternoon: I could do my work, or I could do a
  Fourier analysis of my computer's fan noise.
       -- Amit Muthu, ucam.chat (from Owen Dunn's summary of the year)


From bobhicks at nospam.adelphia.net  Sun Mar 25 16:22:16 2001
From: bobhicks at nospam.adelphia.net (Robert Hicks)
Date: Sun, 25 Mar 2001 21:22:16 GMT
Subject: Python classes on east coast?
Message-ID: <bobhicks-2503011622160001@va-frontroyal1a-202.wch.adelphia.net>

Is there or are there any plans for Python classes on the east coast
(preferrably around DC)?

- Bob

-- 
Bob

Remove: the nospam. in my address if you want to email me


From akuchlin at mems-exchange.org  Thu Mar 29 15:43:42 2001
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 29 Mar 2001 15:43:42 -0500
Subject: Looking for Python equivalent to a Perl'ism
References: <3AC375A7.A4BA87F1@spammraqia.com>
Message-ID: <3d3dbwwbk1.fsf@ute.cnri.reston.va.us>

David Lees <DavidlNo at spammraqia.com> writes:
> ----------
> my %dict;
> my $string="this is xxfooxx this is junk xxmumblexx and more stuff";
> %dict=(foo=>"yechfoo",mumble=>"yukkomumble");
> 
> $string =~ s/xx(\S+)xx/$dict{$1}/g;
> 
> print $string;
> ------------

To simply substitute the contents of a single group, you can 
include \N, or \g<N>:

p=re.compile(r'zz(\w+)zz')
s='this is and so is zzxxxzz abc that an zzyyyzz abc zzxxxzz or else'
print p.sub('\g<1>', s)

This produces "this is and so is xxx abc that an yyy abc xxx or else"
You want to do a dictionary lookup, though; instead of passing a
replacement string, you can pass a function which will be handed the
match object, and can return whatever string it likes:

def sub_func(match):
    return dict[ match.group(1) ]

print p.sub(sub_func, s)

This will output "this is and so is foo abc that an barf abc foo or
else".

Note that this raises an exception if the contents of group 1 aren't
found in the dictionary, so 'zzrandomzz' will trigger a KeyError.
Perl would probably just assume undef or the empty string or something
like that, so sub_func() would have to be changed accordingly --
to dict.get(match.group[1], ""), perhaps.

--amk




From aleaxit at yahoo.com  Mon Mar  5 12:25:20 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Mon, 5 Mar 2001 18:25:20 +0100
Subject: the demise of 'from foo import * and its implications?
References: <3kVBAfATrDo6EwEM@jessikat.fsnet.co.uk><mailman.983582584.8198.python-list@python.org><z8dAGJAyqLo6EwlX@jessikat.fsnet.co.uk> <mailman.983746566.27513.python-list@python.org> <suPo6.16824$ML1.841861@e420r-atl2.usenetserver.com>
Message-ID: <980i6t017ag@news2.newsguy.com>

"Steve Holden" <sholden at holdenweb.com> wrote in message
news:suPo6.16824$ML1.841861 at e420r-atl2.usenetserver.com...
    [snip]
> I did begin to wonder whether it might be possible to specify a name
prefix,
> along the lines of
>
>     from modulewithaverylongname import * prefix sh_
>
> to allow
>
>     sh_i

But why not just
    import modulewithaverylongname as sh
and then use
      sh.i
...?

You can do that today, and the detailed semantics seem
likelier to be what is desired (e.g., should attribute
i of the module ever be re-bound, it's probably better
to automatically pick up the new reference, as happens
by using sh.i, than keep using the old one, as any
'from/import' approach would do -- marginal stuff, sure).


Alex





From qrczak at knm.org.pl  Sun Mar 25 16:21:24 2001
From: qrczak at knm.org.pl (Marcin 'Qrczak' Kowalczyk)
Date: 25 Mar 2001 21:21:24 GMT
Subject: static methods
References: <Pine.LNX.4.21.0103231640070.5553-100000@clarkevans.com> <mailman.985386272.7736.python-list@python.org> <slrn9bpcb0.bm7.qrczak@qrnik.zagroda> <mailman.985552474.12440.python-list@python.org>
Message-ID: <slrn9bsoak.f1i.qrczak@qrnik.zagroda>

Sun, 25 Mar 2001 12:37:43 -0800, Paul Prescod <paulp at ActiveState.com> pisze:

> Doh! Okay, it will work if you add it to the instances in the __init__:
> 
> class Class:
> 	def __init__(self):
> 		self.staticmethod = _static

What is static in this method?

(You can't get it from the class for example, but need an instance.)

-- 
 __("<  Marcin Kowalczyk * qrczak at knm.org.pl http://qrczak.ids.net.pl/
 \__/
  ^^                      SYGNATURA ZAST?PCZA
QRCZAK


From quickdry at users.sourceforge.net  Thu Mar 22 08:25:32 2001
From: quickdry at users.sourceforge.net (Steven Adams)
Date: Thu, 22 Mar 2001 13:25:32 GMT
Subject: tempfile problem on win98
Message-ID: <g3nu6.369$hb.2141@news1.belrs1.nsw.optushome.com.au>

Hi,

I'm using the tempfile module in Python 1.5.2 and it keeps giving me
duplicate filenames.

It's being called only once each time the script runs, (small cgi
messageboard) with this is the code related to the tempfile:

 import tempfile
 tempfile.tempdir=SOME_DIRECTORY
 tempfile.template=None
 id=tempfile.mktemp()

shouldn't this give me a unique value for id every time? I initially
had it running without the tempfile.template being assigned None, but
when I noticed the duplicate filenames, so I thought I'd try that.

Currently I'm making it recompute tempfile names until it has one that
does not already exist, but why does this happen?once I get a lot of
messages in the system I don't want to be possibly iterating through
the whole lot checking their names (unfortunately I don't have
permissions or the space to use a proper database)

thanks for any help,

Steven




From shonn at shonn.com  Fri Mar  2 18:17:35 2001
From: shonn at shonn.com (Shonn Gilson)
Date: Fri, 2 Mar 2001 15:17:35 -0800
Subject: Python, ASP, ADO, and Image string/binary problems
Message-ID: <ta0amgebq0sn0b@corp.supernews.com>

Okay, here it goes.
I have a site in ASP where I need to upload an image (already working) and
then make a thumbnail of it. The image is stored in a MS SQL Server
database, I'm using IIS5 (Win2k), and the python is Active State's with all
the COM functionallity added.
Here's what I'm trying to do:

<script runat="server" language="Python">
import os, sys
import PIL.Image
import StringIO

strSQL = "select * from [FILE] where FILE_Guid =
'406D58D8C4EE4FA28DF1CFC988A49C11'"
objConn = Server.CreateObject("ADODB.Connection")
objRS = Server.CreateObject("ADODB.Recordset")
objConn.Open (connection_string)
objRS.Open (strSQL,objConn,3,1)
bstrIn = objRS("FILE_File").value # returns a buffer, not a string
im = PIL.Image.open(StringIO.StringIO(bstrIn))
im.thumbnail((128, 128))
bstrOut = StringIO.StringIO()
im.save(bstrOut, "JPEG",optimize=1,progressive=1)
Response.ContentType = 'text/html' #'image/jpeg'
Response.Write(" ")
Response.BinaryWrite(bstrOut.getvalue())
#Response.Write ('size')
#Response.Write (`im.size`)
objRS.Close()
objConn.Close()
</script>


If I do a Response.BinaryWrite(bstrIn) I get the image in the browser but if
I do Response.BinaryWrite(bstrOut.getvalue()) I get spaces between every
character (I've seen this with unicode problems). It's not just an
incompatibility with Response.BinaryWrite because I've written the results
back to the database with the same results. It seems like it's a problem
with how python deals with strings.

Anyone have any ideas?

Thanks in advance

Shonn




From aahz at panix.com  Fri Mar  9 10:01:43 2001
From: aahz at panix.com (Aahz Maruch)
Date: 9 Mar 2001 07:01:43 -0800
Subject: Can anyone recomend a good intoduction to C...
References: <mailman.983894288.19965.python-list@python.org> <989cvj$sba$1@panix3.panix.com> <984132875.222868@newsmaster-04.atnet.at> <lX5q6.45269$1D5.2047774@e420r-atl1.usenetserver.com>
Message-ID: <98ar8n$1rt$1@panix6.panix.com>

In article <lX5q6.45269$1D5.2047774 at e420r-atl1.usenetserver.com>,
Steve Holden <sholden at holdenweb.com> wrote:
>"Werner Schiendl" <ws-news at gmx.at> wrote in message
>news:984132875.222868 at newsmaster-04.atnet.at...
>>
>> What's TMTOWTDI, please?
>
>There's another programming language whose name begins with P, whose primary
>philosophy is demonstrated in the well-known phrase"
>
>    "There's More Than One Way to Do It"
>
>This means that it's more difficult to establish recognisable programming
>idioms, because for every task there are between three and seventeen ways to
>accomplish it, and the choice is usually made on efficiency grounds rather
>than those of program readability.

More often than that, the idiom chosen is simply the one the programmer
happens to know, because zie succeeded in learning that idiom from
another person's program and therefore sticks to it like a leech because
zie *knows* the idiom will work.
-- 
                      --- Aahz (Copyright 2001 by aahz at pobox.com)

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

"..., and some of you may regard all women as evil traps that exist only
to tease, torture, and suck out your very soul."  --DrMax


From piet at cs.uu.nl  Wed Mar 28 04:47:51 2001
From: piet at cs.uu.nl (piet at cs.uu.nl)
Date: 28 Mar 2001 11:47:51 +0200
Subject: I'm sure there's a good reason!
References: <mailman.985214563.25964.python-list@python.org> <99cl87$3nq$1@inputplus.demon.co.uk>
Message-ID: <usnjyb4xk.fsf@cs.uu.nl>

>>>>> ralph at inputplus.demon.co.uk (Ralph Corderoy) (RC) writes:

>> if-one-string-gimmick-were-enough-for-all-purposes-there-wouldn't-
>> be-two<wink>-ly y'rs  - tim

RC> A PEP for Hollerith constants anyone?

LOL!
-- 
Piet van Oostrum <piet at cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: P.van.Oostrum at hccnet.nl


From ckrill at qvlinc.com  Fri Mar 23 18:47:42 2001
From: ckrill at qvlinc.com (Coy Krill)
Date: Fri, 23 Mar 2001 15:47:42 -0800
Subject: Daemonizing a python programme.
References: <mailman.985376338.19672.python-list@python.org>
Message-ID: <3ABBE09E.3287DEC5@qvlinc.com>

Timothy Grant wrote:

> Hi all,
>
> I have been investigating daemonizing a python programme, and
> have discovered to bits of code that assist in this endeavour:
> daemonize.py and Daemon.py.
>
> I think that I like Daemon.py's method of doing things, but I
> can't get it to work as it imports a module named delay that I
> can't find anywhere.
>
> Does anyone have any success stories with either of these
> tools?
>
> Thanks
>
> --
> Stand Fast,
>     tjg.
>
> Timothy Grant                         tjg at exceptionalminds.com
> Red Hat Certified Engineer            www.exceptionalminds.com
> Avalon Technology Group, Inc.         <><       (503) 246-3630
> >>>>>>>>>>>>>Linux, because rebooting is *NOT* normal<<<<<<<<<
> >>>>This machine was last rebooted:  65 days 23:52 hours ago<<

I combined ideas from both and rolled this one, that works for me.

#  Daemon Module - basic facilities for becoming a daemon process
#
#  Combines ideas from Steinar Knutsens daemonize.py and
#  Jeff Kunces demonize.py

"""Facilities for Creating Python Daemons"""

import os
import time
import sys

class NullDevice:
    def write(self, s):
        pass

def daemonize():
    if (not os.fork()):
        # get our own session and fixup std[in,out,err]
        os.setsid()
        sys.stdin.close()
        sys.stdout = NullDevice()
        sys.stderr = NullDevice()
        if (not os.fork()):
            # hang around till adopted by init
            ppid = os.getppid()
            while (ppid != 1):
                time.sleep(0.5)
                ppid = os.getppid()
        else:
            # time for child to die
            os._exit(0)
    else:
        # wait for child to die and then bail
        os.wait()
        sys.exit()



From fredrik at pythonware.com  Wed Mar 28 11:34:28 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 28 Mar 2001 16:34:28 GMT
Subject: help with passing arguments to published objects a la Zope
References: <mailman.985795124.31453.python-list@python.org>
Message-ID: <ooow6.1975$sk3.667679@newsb.telia.net>

Eugene Leitl wrote:
> However, is there a simple mechanism, allowing me to pass arguments
> via the URI, as well as use the CGI module (i.e. to parse arguments into a
> dictionary)?

how about XML-RPC?

http://mail.python.org/pipermail/python-announce-list/2001-March/000698.html
http://www.pythonware.com/products/xmlrpc/index.htm
http://www.xmlrpc.com

Cheers /F




From aleaxit at yahoo.com  Fri Mar  9 06:56:29 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 9 Mar 2001 12:56:29 +0100
Subject: Who am I: can a class instance determine its own name?
References: <mailman.984034150.9635.python-list@python.org>
Message-ID: <98agdc01mi6@news2.newsguy.com>

"Tim CHURCHES" <TCHUR at doh.health.nsw.gov.au> wrote in message
news:mailman.984034150.9635.python-list at python.org...
> This is probably an elementary question and the answer is probably writ
large in
> multiple places in the Python documentation, but...

> ...can an instance of a class determine the name of the variable to which
it is
> assigned? For example:

As many respondents have already stated, the answer has to be
No; any object, at any given moment, may be referred-to by any
number N of references [[an aside: in the current version of
CPython, it must be the case that N>0, or the object would be
destroyed; in Jython, in Python .NET, and possibly in future
versions of CPython, this guarantee does not hold]], including
maybe both ones that are named, and other ones that are not
named (but, rather, indexed).

For example,

    adict = {}
    adict['!}@}!'] = []
    adict['!}@}!'].append(MyClass())

now, what is "the name of the variable" you would wish to
determine for this instance, which can at this time be
referred to only as adict['!}@}!'][0] but 'at a drop of a
hat' might also be seen through other references, not be
seen through this one any more, etc, etc...?


Asking an object "under what names [and/or unnamed refs]
are you known" is akin to asking an individual "on what
web sites (URLs) are you mentioned".  The individual most
likely has no idea: he or she is typically not informed
each and every time a reference to him/her is added to
some URL or removed from it, nor does the existence of
such reference have any effect directly observable on/by
him/her.

If you wanted to find all such URLs, you would have to
grep through ALL URLs of the web, and, for each one,
determine if it refers to the individual of interest --
this is basically what search engines such as Google do.

Similarly, if you want to find all references to some
Python object, you will have to grep through all the
objects currently existing, recursively determining all
the references held, and see which of these references
are the ones of your interest (use the 'is' operator
for this purpose).  It makes no real sense to delegate
this job to each individual/object you might be looking
for -- do it in a global function (or utility class) so
you can reuse it for any object, regardless of type.


Supposing you're not interested in unnamed (indexed)
references, local variables, or attributes of instance
objects, but only in attributes of modules (module-level
'global variables'), for example, here's a start...:

def namesOf(anobject):
    result = []
    import sys
    for modname,modobj in sys.modules.items():
        for varname,varobj in modobj.__dict__.items():
            if varobj is anobject:
                result.append("%s.%s" % (
                    modname, varname))
    return result

import sys, re, string

sys.blab = re.blob = string.blub = "fakaoplod"

names = namesOf(sys.blab)
for name in names:
    print name


Alex





From grante at visi.com  Sat Mar 24 11:38:12 2001
From: grante at visi.com (Grant Edwards)
Date: Sat, 24 Mar 2001 16:38:12 GMT
Subject: static methods
References: <mailman.985383828.2942.python-list@python.org>
Message-ID: <slrn9bpqgo.15q.grante@tuxtop.visi.com>

On Fri, 23 Mar 2001 16:51:10 -0500 (EST), Clark C. Evans <cce at clarkevans.com> wrote:

>I was wondering if there is any rationale as to why 
>Python classes do not have static methods?  

What's a "static method?"  I assume it's some C++ contrivance?

After using Modula-3 and Smalltalk, I tried to learn C++ once.
It was mind boggling.  When I got to "virtual" methods, I just
shook my head and gave up.  Apparently in C++ you have to know
ahead of time what methods somebody might want to override at
some point in the future.  Bizarre.

-- 
Grant Edwards                   grante             Yow!  I hope I
                                  at               bought the right
                               visi.com            relish... zzzzzzzzz...


From gmcm at hypernet.com  Sat Mar 10 19:53:33 2001
From: gmcm at hypernet.com (Gordon McMillan)
Date: 11 Mar 2001 00:53:33 GMT
Subject: import headaches/gripes
References: <3aaa3f94.15940553@News.CIS.DFN.DE> <mailman.984250586.21661.python-list@python.org> <Pine.LNX.4.21.0103102113210.7964-100000@kev.ioc.ee>
Message-ID: <9060CBC10gmcmhypernetcom@199.171.54.155>

pearu at cens.ioc.ee (Pearu Peterson) wrote:

>
>On Sat, 10 Mar 2001, Paul Prescod wrote:
>
>> I think there are two reasons:
>> 
>>  1. Even checking file dates has a performance cost. It involves going
>> to the hard disk after all. Code like this isn't rare in Python
>
>I think this is not true in Linux where recently read files stay in
>memory buffer so that later access to the files does not involve
>(physical) actions on the hard disk. Don't know if Windows has such a
>feature..

Windows caches that stuff too. But Linux doesn't cache everything you need 
to do this:

 for dir in sys.path:
    for ext in ('.py', '.pyc') + c_extensions:
        testnm = os.path.join(dir, modulename) + ext
        if os.path.exists(testnm):
           .....

so Paul's point stands.

- Gordon


From steven at haryan.to  Fri Mar  2 10:27:58 2001
From: steven at haryan.to (Steven Haryanto)
Date: Fri, 02 Mar 2001 22:27:58 +0700
Subject: Editing Python source as XML?
Message-ID: <4.3.2.7.2.20010302222446.00c84360@202.53.255.51>

A recent experience of having to modify a long C source file has
led me to wonder whether there are any tool/effort towards editing
Python source code as XML or other form of structured document.

What I would like is an editor that presents the source file as a
tree outline, where I can collapse selected class or function
definition body, certain long if's/for's, some section of code,
view/hide the asserts, etc. So then I only have to deal with the
section I am concentrating with and don't have look at other
distracting parts (especially in C with all those #ifdef/#endif's,
argh...).

I visited conglomerate.org some months ago, and it seems the
project has been abandoned.

Thanks,
Steve




From db3l at fitlinxx.com  Mon Mar 26 18:08:59 2001
From: db3l at fitlinxx.com (David Bolen)
Date: 26 Mar 2001 18:08:59 -0500
Subject: win32net, documentation
References: <3AB8CD29.C69CC1E2@Bibliothek.Uni-Augsburg.de>
Message-ID: <un1a82kms.fsf@ctwd0143.fitlinxx.com>

Sibylle Koczian <Sibylle.Koczian at Bibliothek.Uni-Augsburg.de> writes:

> Hello, all,
> 
> I want to do the Python equivalent of 
> 
> net use k: \\server\share
> 
> so I suppose I must call
> 
> win32net.NetUseAdd("\\server", level, data).

NetUseAdd is the right function, yes, but not called quite this way.
Also, it will only work under NT 3.1 or later or 2000, but not Windows
9x.

> Level = 1 should be right, I think, but I can't get the syntax right.
> I'm trying to translate the api_error messages (they seem to come
> directly from my german Windows version):
> 
> - "server" without the "\\": api_error 50, the network request isn't
> supported
>
> - "\\server": api_error 123, the syntax for the file name, path name or
> the drive name is wrong. I've tried:

The NetUseAdd function is documented (in MSDN) as reserving this
parameter and it should be NULL (None in Python).

> data['local'] = 'k:', 'k:\\' or 'k'
> data['remote'] = 'share', '\\server\share', '\\\\server\\share',
> '\\share'
> data['password'] = my normal domain password
> data['asg_type'] = STYPE_DISKTREE
> 
> It's a network ressource I'm using constantly, so I'm quite sure it
> exists, I've got the name right and I can connect to it.

In your data block, just use the drive letter for local (with ":"), the
full UNC to the sharename for the remote.  The appropriate asg_type
value would be USE_DISKDEV (assuming a shared file resource).  I don't
see USE_DISKDEV actually included in win32con (it's off in an LMUSE
header file rather than the WIN header files which may be why), but
it's value is 0.

> So what's the right syntax? The help file for the Win32 extensions
> hasn't got anything about the USE_INFO_* dictionaries, and the MSDN
> library doesn't help either.

The MSDN reference for NetUseAdd should have everything you need,
including detailed documentation on the USE_INFO structures for each
supported level.

Check out http://msdn.microsoft.com/library/psdk/network/ntlmapi2_1cdg.htm,
and follow the link to the documentation on the USE_INFO_1 structure
(or just look for NetUseAdd if you have a local MSDN CD).

The following should work to map your local K: drive to a resource
\\server\share:

    import win32net

    data = {}
    data['local']    ='k:'
    data['remote']   = r'\\server\share'
    data['asg_type'] = 0

    win32net.NetUseAdd(None,1,data)

It actually works also even if you leave out the asg_type, but it's
probably safer to include it.  You only need to supply a password if
it's going to be different than your current authentication

Note also the use of the raw (r'') string syntax for the share name.  That
way you can use the backslashes normally - if you don't do that then you'd
have to double each backslash for the Python parser.

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


From sholden at holdenweb.com  Thu Mar 22 22:19:54 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 22 Mar 2001 22:19:54 -0500
Subject: Please rate these Python books
References: <20010322.1445209@mis.configured.host> <_Meu6.8736$8O6.1818817@typhoon.austin.rr.com> <slrn9bjuha.4qv.grey@teleute.dmiyu.org>
Message-ID: <blzu6.8974$G72.207211@e420r-atl1.usenetserver.com>

"Steve Lamb" <grey at despair.rpglink.com> wrote in message
news:slrn9bjuha.4qv.grey at teleute.dmiyu.org...
> On Thu, 22 Mar 2001 03:59:54 GMT, alan runyan <runyaga at thisbox.com> wrote:
> >how are you approaching python?  what is your level of experience w/
python?
> >what kinda books do you like? reference, explanatory, or 'by example' ?
My
> >favorite is Essential Reference, David Beazley did a incredible job.
>
>     No disrespect towards Beazley, he's a large reason why I work in
Python as
> much as I can now but his book semmed to be pretty much a dump of the
> postscript of the website to a printer.  Not that that is a bad thing as
it is
> the /only/ Python book I own and one I recommend to anyone who has passing
> familiarity with other scripting languages (Perl comes to mind).  You're
> correct in stating it is sparse on the examples but it is a rare case
where
> I've really needed the examples.  Python being as interactive as it is I
tend
> to learn by firing up Python in a separate window and then playing with
the
> syntax until it works out in that session before plopping it into my
script.
>
I would disagree with that (just so no-one can say this is a "me too" post
:-).

Beazley travels with me so I don't need to look up the ActiveState help
files as often as I otherwise would. If I can find it quicker offline, I'll
spend the money on the dead trees.

Prime collection is currently "Learning Python" (despite its age), "Win32
Programming in Python" and "Python Essential Reference". The rest have their
place, but it's in the bookshelf, not the briefcase.

regards
 Steve





From olczyk at interaccess.com  Tue Mar 13 10:57:19 2001
From: olczyk at interaccess.com (Thaddeus L Olczyk)
Date: Tue, 13 Mar 2001 15:57:19 GMT
Subject: Einstein's Riddle
References: <oNPq6.396644$ge4.138479566@news2.rdc2.tx.home.com> <UdSq6.2184$54.2166@www.newsranger.com> <984380185.768920@news.aei.ca> <uP%q6.2375$54.2207@www.newsranger.com> <4w6r6.193645$df5.5192565@news1.crdva1.bc.home.com>
Message-ID: <3aae4114.63604234@nntp.interaccess.com>

On Mon, 12 Mar 2001 16:08:32 GMT, "David Murphy" <dwmurphy at home.com>
wrote:

>This reminds me of a test my class was given WAY back in grade 6.  It was
>two sheets of paper long, and at the very top, it was clearly stated that
>the student should 'Read the entire test carefully before beginning'.
>
>The point was, at the bottom of the second page, the student was told
>something to the effect of: 'After reading through the test, place your
>pencil on you desk and wait for further instructions'.
>
>Well, 90%+ of the class immediately went to work on this test.  There were a
>couple of questions where students had to get up and do things, and it was
>rather amusing to watch them run around.  About 30 minutes later, they
>started to get to the end, and they realized that they had been running
>around and scribbling down answers for nothing.  It was a rather profound
>lesson, at least for me.
>
The version I saw the last question was sign your name and turn it in.
There were ( AFAIK ) no questions about getting up and running around.
We were given an hour told it was very important. I gave up after
reading the test halfway in the first thirty minutes and started
answering questions.

It didn't help that we had a strange test two weeks.
For the first time the HS I was in had set up a special schedual for
midterms/finals. In this particular class we would meet for two
seperate hours. THe teacher gave us an exam that should have lasted
one hour ( by her estimation ). She was going to use the second hour
for a class. It turned out that most fiished about 3/4 of the test in
the two hours. So she had a reputation for giving long tests.

>*IF* Einstein was just talking about people's ability to deal with just the
>facts, and not to make assumptions, I think 2% +/- 5% is probably a good
>estimate.   Take a look at the first draft of the next project specification
>you get *grin*
>
The problem is typical of problems published in many magazines ( here
in Dell Crosswords ). I suspect that 2%-5% will have the motivation to
sit through solving the puzzle unless there were some other reward,
but I figure half the population could actually solve it given the
proper motivation.

I also figure the fish thing was simply a typo, and that this guy (
who hasn't resdponded ) was probably given this as a homework
assignement, and is simply looking for a cheap way to get it done.



From carribeiro at yahoo.com  Sun Mar 18 22:19:18 2001
From: carribeiro at yahoo.com (Carlos Ribeiro)
Date: Mon, 19 Mar 2001 00:19:18 -0300
Subject: Wrong reasons to enhancing the language
Message-ID: <5.0.2.1.0.20010318232654.02243ba0@mail.inet.com.br>

Le me try to summarize what I've seen on the past ten days or so on this 
list, regarding the awful lot of "enhancements" proposed to the Python 
language.

1) Proposing "syntactic sugar" changes to make some particular code 
constructions look better;
2) Proposing new operators because of speed concerns;
3) Fixing the semantic of the language to fix incompatibilities with other 
languages/modules/theoretical model;
4) Including new constructions because some other language support them;
5) Avoiding the inclusion of some constructions because some other language 
support them (yes, that also happens sometimes).

In my opinion, none of the above reasons is good enough to justify alone 
changes to the language. There are other criteria that need to be met. Are 
the extensions useful to many people? Are they generic and extensible? Do 
they cause any compatibility issue?

I dont think that we'll ever see a perfect language. Every one will have 
some issues or limitations, and thats part of the design decisions. Trying 
to make too much inside of the constraints of a single language is 
overkill. Small incompatibilities are impossible to avoid, partly because 
there are so many different things around. Some problems are just not worth 
the time to "fix" them. Maybe a different language is the answer for them. 
I think that is better to focus the effort where Python is already good, to 
make it stronger. There is no sense in trying to make Python the best tool 
for all jobs - it is an unattainable goal.

Synctatic sugar may make things clear for some, but makes it also a little 
bit harder for the others. If some code is speed critical, let's do it in 
C, C++, whatever compiled language you want. Python makes it extremely 
simple both ways, either to have embedded Python in a compiled language 
application, or to call compiled modules from Python code. Alternatively, 
I'm sure there is a lot of enhancements that could be done to the compiler 
to shave off precious CPU cycles. Why not spend more energy on this? 
Sometimes Python cant do what you want - maybe it's better to use another tool.

Note that I'm not against all change. I found the list comprehensions to be 
extremely effective. Also for the nested scopes - we all know that it 
didn't make it before for practical reasons. But some PEPs... sound silly 
at best.





_________________________________________________________
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com




From tjg at exceptionalminds.com  Fri Mar 23 14:38:33 2001
From: tjg at exceptionalminds.com (Timothy Grant)
Date: Fri, 23 Mar 2001 11:38:33 -0800
Subject: Daemonizing a python programme.
Message-ID: <20010323113833.F7911@trufflehunter.avalongroup.net>

Hi all,

I have been investigating daemonizing a python programme, and
have discovered to bits of code that assist in this endeavour:
daemonize.py and Daemon.py.

I think that I like Daemon.py's method of doing things, but I
can't get it to work as it imports a module named delay that I
can't find anywhere.

Does anyone have any success stories with either of these
tools?

Thanks


-- 
Stand Fast,
    tjg.

Timothy Grant                         tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.         <><       (503) 246-3630
>>>>>>>>>>>>>Linux, because rebooting is *NOT* normal<<<<<<<<<
>>>>This machine was last rebooted:  65 days 23:52 hours ago<<



From brendhanhorne at bellsouth.net  Mon Mar 12 07:10:44 2001
From: brendhanhorne at bellsouth.net (Brendhan Horne)
Date: Mon, 12 Mar 2001 07:10:44 -0500
Subject: I have no idea
Message-ID: <g13r6.201$qt5.10920@news3.mia>

In basic you use to be able to do this:
10 print "Hello World"
20 goto 10
Run

If you did that it would run up the screen in an endless loop until you hit
the ESC key. If you added   ; it would cover the whole screen. How do you
write the equivalent in python?

--
Thanks,
Brendhan




From grey at despair.rpglink.com  Thu Mar  8 14:14:02 2001
From: grey at despair.rpglink.com (Steve Lamb)
Date: Thu, 08 Mar 2001 19:14:02 -0000
Subject: Get terminal size in perl and python
References: <mailman.984077291.3849.python-list@python.org>
Message-ID: <slrn9afmfq.ik7.grey@teleute.dmiyu.org>

On Thu, 8 Mar 2001 19:40:29 +0100, Egbert Bouwman <egbert at bork.demon.nl>
wrote:
>My perl correspondent uses:
>    use Term::ReadKey;
>    ($cols,$lines)=GetTerminalSize()
>What is the python equivalent ?
>If I don't come up with one, python looses ten points.

    If I do in under 5 minutes (having never needing it before) does that mean
you lose 10 instead?

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


From stephen_purcell at yahoo.com  Mon Mar 12 10:39:07 2001
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Mon, 12 Mar 2001 16:39:07 +0100
Subject: Packages, modules and exceptions
In-Reply-To: <98ip5c$397$1@clematis.singnet.com.sg>; from ngps@madcap.dyndns.org on Mon, Mar 12, 2001 at 03:14:52PM +0000
References: <98ip5c$397$1@clematis.singnet.com.sg>
Message-ID: <20010312163907.A23991@freedom.puma-ag.com>

Ng Pheng Siong wrote:
> 
> In M2Crypto, I have the following (abridged) directory structure:
> 
>     M2Crypto/
>             SSL/
>                 __init__.py
>                 Connection.py
> 

I assume you also have a file 'M2Crypto/__init__.py'...


> Now, various parts of M2Crypto/SSL/Connection.py wish to raise SSLError 
> when it is appropriate to do so. 
> 
> For M2Crypto/SSL/Connection.py to get access to SSLError which is defined
> in M2Crypto/SSL/__init__.py, it (Connection.py) contains the following:
> 
>     SSLError = 
>      __import__('M2Crypto.SSL', globals(), locals(), 'SSLError').SSLError
> 
> This works, but looks kinda funny.
> 
> What's the best way to do this?


from M2Crypto.SSL import SSLError
raise SSLError, "there's a problem"


-Steve


-- 
Steve Purcell, Pythangelist
Get testing at http://pyunit.sourceforge.net/
Any opinions expressed herein are my own and not necessarily those of Yahoo



From sholden at holdenweb.com  Wed Mar  7 21:30:10 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Wed, 7 Mar 2001 21:30:10 -0500
Subject: gah! I hate the new string syntax
References: <mailman.983615170.23313.python-list@python.org> <tN5o6.2746$eH3.14913@NewsReader> <Koco6.7518$1D5.244602@e420r-atl1.usenetserver.com> <mailman.983997252.5628.python-list@python.org> <meAp6.29961$ML1.1514835@e420r-atl2.usenetserver.com> <mailman.984014353.5058.python-list@python.org>
Message-ID: <BcCp6.37817$1D5.1582482@e420r-atl1.usenetserver.com>

"D-Man" <dsh8290 at rit.edu> wrote in message
news:mailman.984014353.5058.python-list at python.org...
> On Wed, Mar 07, 2001 at 07:15:48PM -0500, Steve Holden wrote:
> | "D-Man" <dsh8290 at rit.edu> wrote in message
> | news:mailman.983997252.5628.python-list at python.org...
> | > On Sat, Mar 03, 2001 at 03:19:05PM -0500, Steve Holden wrote:
> | [ ... ]
> | >
> | > What about one who is both English and Scots, yet never set foot on
> | > that island ;-)?
> | >
> | No such animal!
>
> Oh.  Am I a bot too?  If so I must find a new maintainer so I can
> match the intelligence of the other bots on the list.
>
Well, that would be nice, but I digress. ;-)

By "that island" I presume you refer to Great Britain?

Either you were born in England, or you were born in Scotland, or you are
neither English nor a Scot. Part-either-or-both, maybe, but not fully
either.

Or perhaps you were born on the border, and carried away before you could
walk?

Or maybe you just have no feet?!?

regards
 Steve






From piet at cs.uu.nl  Wed Mar 21 06:18:24 2001
From: piet at cs.uu.nl (piet at cs.uu.nl)
Date: 21 Mar 2001 12:18:24 +0100
Subject: New PEP: The directive statement
References: <200103201042.LAA20248@pandora.informatik.hu-berlin.de> <mailman.985107140.4586.python-list@python.org> <9988ie02k7u@drn.newsguy.com>
Message-ID: <ulmpzjrpb.fsf@cs.uu.nl>

>>>>> Grant Griffin <not.this at seebelow.org> (GG) writes:

GG> No kidding, Gang, when I first ran across posts about the __future__
GG> directive, I thought it was another one of Tim's beloved witicisms.
GG> (That's a polite way of saying I thought it was a joke. <wink>) It took
GG> me awhile to figure out it was serious, because it _reads_ like a
GG> joke!: "from __future__ import..."

It would be much less of a joke when __future__ would be replaced by
__features__, I think.
-- 
Piet van Oostrum <piet at cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: P.van.Oostrum at hccnet.nl


From mwh21 at cam.ac.uk  Sun Mar 11 14:44:13 2001
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 11 Mar 2001 19:44:13 +0000
Subject: Questions about no more nested imports
References: <98gi4l$n9n$0@216.39.170.247>
Message-ID: <m3n1as850y.fsf@atrus.jesus.cam.ac.uk>

whisper at oz.net (Dave LeBlanc) writes:

> I'm a bit confused by this. 

 | <- the right end                        the wrond end -> |
 +----------------------------------------------------------+
 |                         the stick                        |
 *----------------------------------------------------------+
                                                         ^
                                            you're about here

> It appears that if this is enforced, one can no longer dynamically
> load extensions/whatever dynamically and all imports must be done at
> the top of a module prior to anticipated use - whether used or
> not. If this is true, then load times for python apps could go up
> drastically.

The only thing that's getting banned is "from blah import *" and
"exec" without "in" in a set of fairly restricted circumstances.

Read the PEP:

  http://python.sourceforge.net/peps/pep-0227.html

(specifically the section "Backwards compatibility").

Cheers,
M.

-- 
  please realize that the Common  Lisp community is more than 40 
  years old.  collectively, the community has already been where 
  every clueless newbie  will be going for the next three years.  
  so relax, please.                     -- Erik Naggum, comp.lang.lisp


From brendhanhorne at bellsouth.net  Tue Mar 13 07:27:25 2001
From: brendhanhorne at bellsouth.net (Brendhan Horne)
Date: Tue, 13 Mar 2001 07:27:25 -0500
Subject: Very Cool
References: <g13r6.201$qt5.10920@news3.mia> <OYer6.280$qt5.17754@news3.mia> <slrn9arlfg.bjg.scarblac@pino.selwerd.nl>
Message-ID: <Gsor6.652$Ee5.816464@news1.mia>

That did it . I must have tried every other space combinationin the world
except that one.

--
Thanks,
Brendhan




From micampe at f2s.com  Tue Mar 27 05:39:21 2001
From: micampe at f2s.com (comp.lang.python)
Date: Tue, 27 Mar 2001 10:39:21 GMT
Subject: R: unique items in lists
References: <_nZv6.2878$yU4.468236@news.infostrada.it> <slrn9c0ph1.o5h.scarblac@pino.selwerd.nl>
Message-ID: <t5_v6.3466$yU4.477532@news.infostrada.it>

Here is what I'm doing now:

        for w in words:
            while words.count(w) > 1:
                words.remove(w)

comments?




Remco Gerlich <scarblac at pino.selwerd.nl> wrote in message
slrn9c0ph1.o5h.scarblac at pino.selwerd.nl...
> comp.lang.python <micampe at f2s.com> wrote in comp.lang.python:
> > Which is the best (read: fastest) way to remove duplicated items from a
list
> > of strings?
>
> The standard idiom is to put the items in a dictionary, then get the
> dictionary keys:
>
> def unique_strings(l):
>    dict = {}
>    for s in l:
>       dict[s] = 1
>    return dict.keys()
>
> I think this is also very fast, at least lots faster than comparing all
the
> strings in a double for loop or something.
>
> --
> Remco Gerlich




From tim at vegeta.ath.cx  Fri Mar  9 18:37:11 2001
From: tim at vegeta.ath.cx (Tim Hammerquist)
Date: Fri, 09 Mar 2001 23:37:11 GMT
Subject: "!=" is bad form. Re: sorry....never mind
References: <DZkp6.18811$pF4.3015652@news1.telusplanet.net> <fdlp6.16982$go.2157953@news0.telusplanet.net> <3AA6E308.5B63ECC6@cosc.canterbury.ac.nz> <3AA72253.FB021815@Lugoj.Com> <slrn9aehgr.33d.tim@vegeta.ath.cx> <984047188.296438@newsmaster-04.atnet.at> <slrn9agdkv.9ms.tim@vegeta.ath.cx> <3AA86D0F.E36B0BDC@Lugoj.Com>
Message-ID: <slrn9aiqfq.hia.tim@vegeta.ath.cx>

James Logajan <JamesL at Lugoj.Com> wrote:
> Tim Hammerquist wrote:
> > I realize I seem to be in the minority in this group when I say I like
> > C/C++.  NB: I like them as _languages_, but would prefer NOT to do
> > everyday programming in them, as they're just a pain in the ass most of
> > the time.
> 
> Ironically, I like C and have been coding in C on and off for it seems just
> about forever (well, 18 some years I think). And I find programming in C a
> mostly pleasurable experience. Yet as I stated earlier, for Python "<>"
> feels "righter" to me than "!=". I suspect it is a triviality for both
> programmers to understand, and the language to support, both forms. While
> deprecating "<>" in Python would not mean the end of civilization as we know
> it, it would be a damn shame anyway.

Likewise, I would be disappointed if '!=' were deprecated, but, as you
said, not the end of civilisation.  ;)

> (a > b) or (a == b)  --->  (a >= b)
> (a < b) or (a == b)  --->  (a <= b)
> (a < b) or (a >  b)  --->  (a <> b)

Thank you, BTW for this explanation.  I just don't think of it like
that, but it makes a little more sense to me now.  Thx.

-- 
-Tim Hammerquist <timmy at cpan.org>
A university is what a college becomes when
the faculty loses interest in students.
	-- John Ciardi


From loriendNOSPAM at bigpond.com  Sat Mar 17 21:04:03 2001
From: loriendNOSPAM at bigpond.com (Lorien Dunn)
Date: Sun, 18 Mar 2001 13:04:03 +1100
Subject: Meta-class inheritance problem (Python2.0 bug?)
References: <6Kzs6.35062$v5.73895@newsfeeds.bigpond.com> <5nDs6.35198$v5.74638@newsfeeds.bigpond.com> <90678B1B1gmcmhypernetcom@199.171.54.194>
Message-ID: <SrUs6.35681$v5.76094@newsfeeds.bigpond.com>

Gordon McMillan wrote:

> 
> That's the same code as Python 1.5. Did this work before?

        I don't know. I don't have python 1.5 anymore, and its a new extension I'm 
writing. David Abrahams (the major author of the boost::python lib) said 
that isinstance() and issubclass() have been fixed to deal with metaclasses 
in Python 2.0, but it seems this check of self hasn't.
> 
> From your previous post, it does not appear to be using the Don Beaudry
> hook, but then again, I don't know what type(MyBoostClass) is.

        A boost class is a class that uses the boost::python library to reflect a 
C++ class in python. It does use the Don Beaudry hook. IMHO Boost seems to 
be the only reasonable way to wrap C++ (and I've tried SWIG, Siloon, SIP, 
CXX, and Boost), in that it supports exceptions, overidable virtual 
functions, derivation from C++ classes without shadow classes, automatic 
type conversion, function overloading, etc. It is also MUCH easier to use 
than Zope. However it puts a lot of stress on the compiler in terms of 
adherance to the ansi standard. And a lot of stress on the system. The 
wrapping code of my extension takes about  10 minutes 160 megs of ram to 
compile (and its not that large). 
        Boost as a whole is a C++ library that seems to be almost a draft for the 
next ansi standard. If you use C++ it will make your life more pleasant. 
Have a look at http://www.boost.org.

Thanks,
Lorien Dunn


From loewis at informatik.hu-berlin.de  Sat Mar 24 06:44:43 2001
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 24 Mar 2001 12:44:43 +0100
Subject: XML/XSLT for newbies
References: <3abbbb0f.69122775@nnrp.wiznet.ca>
Message-ID: <j4u24jmlw4.fsf@informatik.hu-berlin.de>

lance.hill at removespamfilter.olsinc.net writes:

> All the documentation I have been able to find on how to do xsl
> transformation assumes you have ASP,VB,JAVA or such. Does anyone know
> of learning resources for xml/xsl in python?

For XML, I recommend the Python/XML tutorial:

http://py-howto.sourceforge.net/xml-howto/xml-howto.html

and the API documentation:

http://python.sourceforge.net/devel-docs/lib/markup.html

For XSLT, I recommend the 4Suite API documentation:

http://services.4Suite.org/topics/rdf/Top/Documentation/APIs

Regards,
Martin


From arkanea at yahoo.com  Fri Mar 30 08:25:05 2001
From: arkanea at yahoo.com (Olivier Laurent)
Date: Fri, 30 Mar 2001 15:25:05 +0200
Subject: converting python to latex
Message-ID: <9a21er$pa4$1@news.brutele.be>

I'd like to convert some little Python scripts into Latex.
This Latex source will be converted into html and pdf with
colored syntax highligting
(something like py2html but outputing in Latex instead of
html)

I've tried an old program called py2tex but it doesn't
convert into Latex (it converts int o a file that can be
imported by Latex).

Does someone know of a module or program that can do it. If
not : what's the best an d simple way to do it.

Thank you.

Olivier Laurent.
--
P3B.org : Association des programmeurs Python de Belgique
francophone.


From bsass at freenet.edmonton.ab.ca  Thu Mar 22 00:21:50 2001
From: bsass at freenet.edmonton.ab.ca (Bruce Sass)
Date: Wed, 21 Mar 2001 22:21:50 -0700 (MST)
Subject: New PEP: The directive statement
In-Reply-To: <slrn9bi7q3.2sa.hzhu@rocket.knowledgetrack.com>
References: <mailman.985085006.24236.python-list@python.org>
 <slrn9bi7q3.2sa.hzhu@rocket.knowledgetrack.com>
Message-ID: <Pine.LNX.4.33.0103212154070.6632-100000@bms>

On Wed, 21 Mar 2001, Huaiyu Zhu wrote:
> With the __future__ PEP, I envision there could also be things like
>
> from __future__ import nested_scope
> from __experimentatal__ import stackless
> from __optional__ import static_type
> from __mandatory__ import string_methods
<...>
[an alternative]
> directive transitional nested_scope
> directive experimentatal stackless
> directive optional static_type
> directive mandatory string_methods
[or whatever]
<...>

Why was (something along the lines of)...

use:
    __future__.nested_scopes
    __experimental__.stackless
    __optional__.static_type
    __mandatory__.string_methods
except FutureError:
    # compensate for lack of nested scopes
except ExperimentalError:
    .
    .

...and allowing...

use:
    __future__.nested_scopes
    # code that relies on nested scopes
except FutureError:
    # code that does not use nested scopes

...rejected?  [It looks like a natural to me <shrug>...

<...>

> IMHO, reusing existing syntax is closer to Python than Perl, which
> introduces new syntax for many things that are only slightly different.

...and seems the most "Pythonic" of all.]


- Bruce




From prairiesasquatch at crosswinds.net  Thu Mar 15 22:13:34 2001
From: prairiesasquatch at crosswinds.net (Kevin Russell)
Date: Thu, 15 Mar 2001 19:13:34 -0800
Subject: Umlaut and Tkinter
References: <3AB0C38F.34218EEB@teleatlas.com>
Message-ID: <3AB184DE.8CF1682E@crosswinds.net>


Michael Husmann wrote:

> typing in idle.py using Python 2.1b1:
>
> >>> print "??"
>
> results in an:
>
> UnicodeError: ASCII encoding error: ordinal not in
> range(128)
>
> Is there someone who can help me.
>
> Michael

This looks like a messy incompatibility between Python 2's
not-quite-fully-integrated-yet Unicode support and Tcl's
even-less-quite-fully-integrated-yet Unicode support, and the
desire of one or both of the parties to communicate in the
lowest common denominator (ASCII) unless beaten over the head
with a stick and forced not to.

Until the Tkinter and Unicode gurus can sort the mess out,
in their abundant free time, you'll probably have to use
one of the other Python interfaces if you want to be able to
use upper-128 Latin-1 characters at an interactive prompt.
I just tried your '''print "??"''' line in both PythonWin
and the Python mode of Emacs (with Python 2.0) and it does
what you'd expect.  It also works fine under in a Windows
MS-DOS box (abstracting away from the whole different passel
of character set problems there), so I suspect it should
also work OK at a Unix command line.

Good luck.

-- Kevin






From chrisw at nipltd.com  Fri Mar 23 10:46:08 2001
From: chrisw at nipltd.com (Chris Withers)
Date: Fri, 23 Mar 2001 15:46:08 +0000
Subject: Slightly Off-Topic: Python featured in Severance: Blade of Darkness
References: <NMHu6.144$NR.1628@news3.oke.nextra.no>
Message-ID: <3ABB6FC0.97A8E10B@nipltd.com>

Thomas Weholt wrote:
> 
> A friend of mine just informed me that while starting up his copy of the
> game Blade of Darkness he saw a reference to Python 1.5.2 during the loading
> process. Anybody know anything about that? What has Python been used for?
> Writing mods/extensions in Python would be cool in any game. :->

IIRC, Severance looks like it's 90% written in python :-)
(the bladex engine looks like it's been wrapped up as a DLL, I wonder if they
used SWIG? ;-)

Anyway, take a look in the scripts and lib folders of the games install and
you'll find lots of python you can play with. I believe there's even one called
cheats.py...

happy hunting,

Chris



From tuttledon at hotmail.com  Thu Mar 22 11:55:33 2001
From: tuttledon at hotmail.com (Don Tuttle)
Date: Thu, 22 Mar 2001 16:55:33 GMT
Subject: Good Editor?
References: <3ab9a32b$1_1@news2.one.net> <3AB9FAC9.97EA420E@student.liu.se> <slrn9bjv7m.4qv.grey@teleute.dmiyu.org> <99d780$l2g$1@laurel.tc.umn.edu> <3ABA20F7.AB71D3B4@student.liu.se>
Message-ID: <98qu6.166420$Z8.35271037@typhoon.southeast.rr.com>

"Rikard Bosnjakovic"
> Anyone sending unwanted advertising e-mail to my address will be
> charged $250 for network traffic and computing time. By extracting my
> address from this message or its header, you agree to these terms.

love it...Love It...LOVE IT!

Don




From chris.gonnerman at usa.net  Sun Mar  4 09:46:51 2001
From: chris.gonnerman at usa.net (Chris Gonnerman)
Date: Sun, 4 Mar 2001 08:46:51 -0600
Subject: How to broadcast socket
References: <NLpo6.21534$AH6.3012743@newsc.telia.net>
Message-ID: <003b01c0a4b9$fa467380$a100000a@local>

So, *exactly* what *does* it do?  Throw an exception?  Appear to work but
not send a packet?  How do you know if it sent or not (i.e. are you testing
with
another application or did you actually use a packet sniffer?)

If you are testing with an application and the application shows no sign of
receiving the packet, get a packet sniffer and see if anything is actually
being sent.  I use Ethereal (don't remember the URL but look on
freshmeat.net)
but I believe there are many others.

----- Original Message -----
From: "J?rgen Tapani" <jota at telia.com>
Subject: How to broadcast socket


> I'm new to Python, and I have achived this
> def test():
>  s = socket(AF_INET, SOCK_DGRAM)
>  s.setsockopt(SOL_SOCKET,SO_BROADCAST,1)
>  s.sendto('Broadcast test',INADDR_BROADCAST)
>  s.close()
>
> And it do not run!
> Help!
> Jorgen Tapani





From fredrik at pythonware.com  Wed Mar 21 10:29:38 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 21 Mar 2001 15:29:38 GMT
Subject: Access Issues
References: <mailman.985182567.26602.python-list@python.org>
Message-ID: <CN3u6.896$sk3.291165@newsb.telia.net>

Clodagh O'Rourke wrote:
> I wonder if anyone can help.
>
> I am trying call the constructor of the EnvVar class from the
> newEnvWin function of the Cell class.  But I am getting the
> following error
>
> Exception in Tkinter callback
> Traceback (innermost last):
>   File "/usr/vob/omc_reference/ThirdParty/scriptSources/lib/python1.5/lib-tk/Tkinter.py", line 764, in __call__
>     return apply(self.func, args)
>   File "Cell.py", line 63, in newCellMgt
>     self.newEnvWin('cellmgt')
>   File "Cell.py", line 60, in newEnvWin
>     instance = EnvVar.__init__(g, self.parent)
> AttributeError: __init__
>
> I have include some extracts of the code.

__init__ is not a constructor; it's the function called by the
constructor to initialize an instance.

(and in your case, EnvVar appears to be a module, not a class)

try replacing

       instance = EnvVar.__init__(g, self.parent)

with

       instance = EnvVar.EnvVar(g, self.parent)

Cheers /F




From barry at digicool.com  Sat Mar 10 13:50:31 2001
From: barry at digicool.com (Barry A. Warsaw)
Date: Sat, 10 Mar 2001 13:50:31 -0500
Subject: "!=" is bad form. Re: sorry....never mind
References: <DZkp6.18811$pF4.3015652@news1.telusplanet.net>
	<fdlp6.16982$go.2157953@news0.telusplanet.net>
	<3AA6E308.5B63ECC6@cosc.canterbury.ac.nz>
	<3AA72253.FB021815@Lugoj.Com>
Message-ID: <15018.30583.852611.836468@anthem.wooz.org>

Personally, I agree.  I hate "!="; to me <> looks much better and I
use it all the time.  Guido has been known to disagree with me so
violently that he'll make a patch to my code just to change that. :)

http://cvs.sourceforge.net/cgi-bin/cvsweb.cgi/python/dist/src/Lib/smtpd.py?rev=1.4&content-type=text/x-cvsweb-markup&cvsroot=python

-Barry



From gregj at pobox.com  Sun Mar 11 00:47:01 2001
From: gregj at pobox.com (Gregory Jorgensen)
Date: Sun, 11 Mar 2001 05:47:01 GMT
Subject: grabbing return codes from os.system() call
References: <%wFp6.9481$dL4.126404@vixen.cso.uiuc.edu> <x8Gp6.899$54.943@www.newsranger.com>
Message-ID: <pjEq6.2038$54.2165@www.newsranger.com>

In article <x8Gp6.899$54.943 at www.newsranger.com>, Gregory Jorgensen says...
>
>os.system returns the exit status. The low-order byte of the exit status is the
>signal number that killed the process, and the high-order byte is the exit
>status (if the signal number is 0). This only works under Unix; on Windows the
>exit status is always 0.

Actually the current docs at python.org say that Windows 95 and 98 always return
0. Apparently Windows NT and 2000 do return something. I don't know about Win
ME. The Beazley book "Python Essential Reference" (which I used) says "On
Windows, the exit code is always 0" in the description of os.system.

Greg Jorgensen
Deschooling Society
Portland, Oregon, USA
gregj at pobox.com


From tuttledon at hotmail.com  Fri Mar 16 22:23:23 2001
From: tuttledon at hotmail.com (Don Tuttle)
Date: Sat, 17 Mar 2001 03:23:23 GMT
Subject: socket.ssl for win32
References: <mailman.984773016.25714.python-list@python.org> <Jqws6.22$quX.1114625@news.randori.com> <2_ys6.108066$__6.20130421@typhoon.southeast.rr.com> <dhAs6.130$quX.18415973@news.randori.com>
Message-ID: <LMAs6.108308$__6.20270544@typhoon.southeast.rr.com>

Got it. Thanks.

Don




From tjg at exceptionalminds.com  Mon Mar 12 18:22:16 2001
From: tjg at exceptionalminds.com (Timothy Grant)
Date: Mon, 12 Mar 2001 15:22:16 -0800
Subject: Process details
Message-ID: <20010312152216.N6836@trufflehunter.avalongroup.net>

Hi all,

I hope everyone is enjoying there first week back after the
events of P9.

I need to create a list of processes and process information. I
have created a class that contains much of the information in
/proc/'pid'/status.

It works quite well, however, it takes about four seconds to
populate a list of 90 procs. This seems a bit long. Is there a
faster way to get ahold of the information than reading /proc?

-- 
Stand Fast,
    tjg.

Timothy Grant                         tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.         <><       (503) 246-3630
>>>>>>>>>>>>>Linux, because rebooting is *NOT* normal<<<<<<<<<
>>>>This machine was last rebooted:  55 days  3:34 hours ago<<



From spam at spam.com  Sat Mar 31 01:02:38 2001
From: spam at spam.com (Penfold)
Date: Sat, 31 Mar 2001 07:02:38 +0100
Subject: mxODBC chokes on dates! Any Help?
References: <MPG.152e4c579fa0baef9896a5@news.pacbell.net> <ulmpnnmte.fsf@ctwd0143.fitlinxx.com> <MPG.152e9338351a7e1b9896a7@news.pacbell.net>
Message-ID: <tcastvrc90ihfd@corp.supernews.com>

"Owen Ap Owen" <no at thank.you> wrote in message
news:MPG.152e9338351a7e1b9896a7 at news.pacbell.net...
> Thanks for the info. It is SQL Server that I am connecting to. But I got
> the same error with date fields when I set up a test connection to
> Access.
>
> It's nice that a fix is in the works. But this seems like something of a
> show stopper (at least for the win32 side of things).
>
> Considering that the creator of mxODBC is trying to charge a laughably
> exorbitant license fee for commercial use he really needs to fix this
> kind of stuff very quickly. He also needs to add a lot more functionality
> and even then his licensing concept is pretty extreme.

Considering lots of developers in commercial settings such as finance and
banking
currently free ride off what is in general a fantastic package, its not that
laughable.

Questions:
(i) Did you try to contact him directly ?  He has always been very prompt in
replying
to me when I have, and I do *not* pay him a licence fee
(ii) Have you actually paid any money for mxODBC.  If not, why do you expect
that it
should be perfect and that he should fix it for you quick smart if it aint ?

I imagine you probably think that mxODBC and mxDateTime, a fantastically
useful package
as well, should be free for you to use and abuse simply because you expect
to be paid to
do development work for a commercial company, but you think that others
should provide
for free the tools you need in such work.

If its a problem, why not write your own DB interface and datetime support,
I'm sure you could
manage it in a jiffy ;-)

>
> In article <ulmpnnmte.fsf at ctwd0143.fitlinxx.com>, db3l at fitlinxx.com
> says...
>
> > I ran into the same problem with some date fields in an MS SQL Server
> > database, and Marc-Andre was able to revert back to an earlier way of
> > handling date fields (he preferred a different C data type if defined
> > in 2.0.0 than in earlier versions) to resolve it.  I was traveling and
> > was a little tardy getting back to verify that the fix worked, but I
> > expect it will show up in the next release of mxODBC.




From fredrik at pythonware.com  Tue Mar 20 15:37:22 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Tue, 20 Mar 2001 20:37:22 GMT
Subject: re.match
References: <997skf$qoc$1@bmerhc5e.ca.nortel.com> <TrMt6.2490$4N4.318602@newsc.telia.net> <mailman.985114106.26896.python-list@python.org> <998bke$88i$1@bmerhc5e.ca.nortel.com>
Message-ID: <6cPt6.2545$4N4.332824@newsc.telia.net>

Michael P. Soulier wrote:
> If you do not include anchors, then by the definition of the regexp syntax,
> the entire line should be searched.

by your definition, perhaps.  obviously not by Python's definition.

> No biggy, but surprising

if you find it surprising that "match" matches and "search" searches,
I'm sure Python has more surprises in store for you...

Cheers /F




From gmcm at hypernet.com  Sat Mar 10 14:19:00 2001
From: gmcm at hypernet.com (Gordon McMillan)
Date: Sat, 10 Mar 2001 14:19:00 -0500
Subject: ANN: Linux version of Installer for Python 2
Message-ID: <3AAA37D4.21892.6BB45937@localhost>

The Linux version of Installer for Python 2 is here:
 http://www.mcmillan-inc.com/installer_dnld.html

Background:
 The Installer is a sort of compilerless Freeze - a way 
of distributing Python apps with all the required support built 
in. The Linux version collects all dependencies and packages 
the pure Python into an archive attached to the executable.
Just tar, gzip and ship.


Old-Python-style license.

Contact: gmcm at hypernet.com

<P><A 
HREF="http://www.mcmillan-inc.com/installer_dnld.html">
Linux Installer release 3j</A> A compiler-less way 
of distributing Python 2.0 apps on Windows. (10-Mar-01)



- Gordon



From joconnor at cybermesa.com  Fri Mar  9 15:23:25 2001
From: joconnor at cybermesa.com (Jay O'Connor)
Date: Fri, 09 Mar 2001 20:23:25 GMT
Subject: "!=" is bad form. Re: sorry....never mind
References: <DZkp6.18811$pF4.3015652@news1.telusplanet.net> <fdlp6.16982$go.2157953@news0.telusplanet.net> <3AA6E308.5B63ECC6@cosc.canterbury.ac.nz> <3AA72253.FB021815@Lugoj.Com> <slrn9afcb5.lou.qrczak@qrnik.zagroda>
Message-ID: <3aa93b8e.775214008@news.uswest.net>

On 8 Mar 2001 16:20:53 GMT, qrczak at knm.org.pl (Marcin 'Qrczak'
Kowalczyk) wrote:

>Wed, 07 Mar 2001 22:10:27 -0800, James Logajan <JamesL at Lugoj.Com> pisze:
>
>> The operations "<", ">", "<=", ">=", "==", "<>" form a nice
>> consistent and predictable set.
>
>Except "==". Using "=" instead would be consistent.
>
>PS. Lisp, Ada, Eiffel and Haskell use "/=" for inequality. Icon uses
>"~=" for numbers and "~==" for strings. Simula uses NE for value
>inequality and "=/=" for identity inequality.  OCaml uses "=" and "<>"
>for value (in)equality, and "==" and "!=" for identity (in)equality.
>

I prefer 

	if not x == y:

myself

Take care,
Jay


From python at commandprompt.com  Thu Mar 15 18:35:00 2001
From: python at commandprompt.com (python at commandprompt.com)
Date: Thu, 15 Mar 2001 15:35:00 -0800 (PST)
Subject: Announcement: Python Programming: Taming the Snake.
Message-ID: <Pine.LNX.4.30.0103151534060.14049-100000@crazypenguins.commandprompt.com>

Open Docs, LLC has announced today "Python Programming: Taming the Snake".
The book, their fifth title is a comprehensive programmer's guide to the
Python Programming Language. With this release, Open Docs has managed to
keep pace with the open source's cutting edge development. The title is
ensured to be one of the most comprehensive resource books available
today.

About Python:

 The first beta version of Python 2.1, a rapid application development
language, was released on March 2, 2001. As developers have grown to use
Python on a number of different platforms (Unix, Linux, Windows, Mac )
many companies have embraced Python and are now using it for their
products. The final release of Python 2.1 is due in April 2001; with the
added platforms and technical advances it will bring a new level of
functionality to the Python community.

 Python is being found just about anywhere you look, from animation for
the big screen (Disney animation) to advising users through Linux
installations. You'll find it managing internet discussion forums, dishing
up directories and maps on the internet. They have even gone as far as
scripting wireless products, and online games. Python is striking every
where you look and striking fast.

About the Book:

 Open Docs is set to release "Python Programming: Taming the Snake" at the
end of June. It will cover 2.0 including features found in 2.1. It will
also cover the new Python tools and applications, including Jython (a
version of Python that compiles as Java byte code); support; the CGI and
Internet protocol modules; and database support, for databases such as
PostgreSQL . Open Docs will also be including 2 CD-ROM's loaded with
examples from the book and Python-related packages. These include the full
Python 2.1 source code distribution, the Zope web framework, and the
Vaults of Parnassus. We are sure this book will prove to be invaluable to
any serious Python programmer.

About Open Docs Publishing:

 Open Docs is a publishing company specializing in technical books. They
are a full service publisher that offers authoring, editing, along with
special programs that allow companies to maintain their focus on
development. Open Docs strives in maintaining tight quality control to
ensure that they put out only the best books possible. They take pride in
the work that others find numbing; documentation.

 Open Docs currently has three books in print. The third book, "Securing
and Optimizing Linux; Red Hat Edition" is the number one selling book on
LinuxCentral.com, a premiere online Linux retailer, and has been for six
months. They are currently preparing there fourth title "Getting Stated
with Yellow Dog Linux".

 Additionally Open Docs is striving to help support the Open Source
Documentation Community by donating 10% of all net profits from the sales
of "Python Programming; Taming the Snake" to the Python Software
Foundation. As part of their Open Source Documentation Fund they are
setting a new standard. If you would like to read more on Open Docs
Publishing or their Open Source Documentation Fund please visit there web
site at www.opendocspublishing.com.

Contact Info:
WWW: http://www.opendocspublishing.com/
Voice: 503-963-8316
Email: Christine Parmer.
Postal:
OpenDocs, LLC.
2746 Front St. NE
Salem, OR 97303





From tjreedy at udel.edu  Sun Mar 18 21:46:51 2001
From: tjreedy at udel.edu (Terry Reedy)
Date: Sun, 18 Mar 2001 21:46:51 -0500
Subject: Help Rewriting/Refactoring/Rethinking Parsing Algorythm
References: <3AB5675A.8363C9F2@mindspring.com>
Message-ID: <993s09$8p7$1@news.udel.edu>

...
> print "what's wrong with %s?" % first name     ->     print "what's
> wrong with %s?" % firstName
...
> I've gotten all but the last case to work.  But once I started trying to
...
You need to add 'quote' or sum such to your list of special words.  You
will need a variable 'inquote' to keep track of when you are in quotes (and
should therefor turn off all other special processing).
To do this without making code spaghetti, you might jump to another
within-quote function.

There have been  versions of WordPerfect bundled with Dragon N. S.  This
might give you more ideas on how to direct text editing with DNS.






From menscher+python at uiuc.edu  Thu Mar  8 18:07:57 2001
From: menscher+python at uiuc.edu (Damian Menscher)
Date: Thu, 08 Mar 2001 23:07:57 GMT
Subject: grabbing return codes from os.system() call
References: <%wFp6.9481$dL4.126404@vixen.cso.uiuc.edu> <98795o$kq8$0@216.39.151.169> <45BCEEDEDA54E206.15154B99753B2A0C.82A8F3BAA10D6E91@lp.airnews.net>
Message-ID: <hhUp6.9715$dL4.129738@vixen.cso.uiuc.edu>

Cameron Laird <claird at starbase.neosoft.com> wrote:
> In article <98795o$kq8$0 at 216.39.151.169>, Donn Cave  <donn at oz.net> wrote:
>>Quoth Damian Menscher <menscher+python at uiuc.edu>:
>>| I'm new to Python (as in, my experience is essentially the tutorial),
>>| but I've already come up with an interesting question:
>>|
>>| How do I get the return code from an os.system call?  I would have
>>| expected I could do something like
>>|
>>| ---returncode---
>>| #/bin/csh
>>| echo do stuff
>>| exit 3
>>|
>>| and then in my python program I could do
>>|
>>| print os.system('./returncode')
>>|
>>| But it prints out 768.  Not particularly useful, even after I recognize
>>| the trick of dividing by 256 (byte-swapping going on?  No, because a
>>| return code of 768 reports as 0).  Given that my real return codes
>>| will be (possibly large) integers, this limitation will likely cause
>>| some serious problems down the line.
>>|
>>| Even better would be a way of returning a string (the script I run
>>| can be something other than csh, but it has to be a separate script).
>>|
>>| Ideas?  I'm trying to avoid writing the string out to a file and then
>>| read the file back in to the python program....

> Guys, guys; you are making it too hard on Mr. Menscher.  My guess
> is that he'd appreciate being told that, if he creates /tmp/other
> with contents
>   #!/bin/sh
>   echo "This is a string from an external process."
> he can then have fun with
>   import os
>   print os.popen("/tmp/other").read()

Thank you much -- this solves my problem.  I'm sure I'll have more
questions in the near future, though....

Damian Menscher
-- 
--==## Grad. student & Sys. Admin. @ U. Illinois at Urbana-Champaign ##==--
--==## <menscher at uiuc.edu> www.uiuc.edu/~menscher/ Ofc:(217)333-0038 ##==--
--==## Physics Dept, 1110 W Green, Urbana IL 61801 Fax:(217)333-9819 ##==--


From moshez at zadka.site.co.il  Sat Mar 31 05:42:09 2001
From: moshez at zadka.site.co.il (Moshe Zadka)
Date: Sat, 31 Mar 2001 12:42:09 +0200
Subject: list vs tuple
In-Reply-To: <MVhx6.1021$p5.4157@news1.rivrw1.nsw.optushome.com.au>
References: <MVhx6.1021$p5.4157@news1.rivrw1.nsw.optushome.com.au>, <hGlw6.782$p5.3210@news1.rivrw1.nsw.optushome.com.au> <3AC20CD3.265A2D94@alcyone.com> <ru69cto61jtmrqsf809e38hb3pbm2lhi7g@4ax.com> <bC1x6.4319$4N4.905946@newsc.telia.net> <71q9ct4le63anefmiajj5cbh3mg04qu1ga@4ax.com> <mailman.985991202.5578.python-list@python.org> <7Bfx6.989$p5.3791@news1.rivrw1.nsw.optushome.com.au> <Ukhx6.4421$4N4.950603@newsc.telia.net>
Message-ID: <E14jIpd-0001ML-00@darjeeling>

On Sat, 31 Mar 2001 10:01:16 GMT, "deadmeat" <root@[127.0.0.1]> wrote:

> Ooh, poor diddums got proven wrong. I blew the lid on your little baby.
> <cries>

You're proving yourself less knowledgable with every post.

> What python DOES is irrelevant

Of course. Completely irrelevant.

> The same
> statement (a = b) behaves entirely differently in each case.

The same statement behaves *exactly the same*. It's the next statements
(b[5] = 9, b = 9) that behave entirely differently. There are valid
arguments for not making = do such wildly different things (the Scheme
world occasionally get into these flamewars, only they spell = as set!)

b = 9 binds the variable
b[5] = 9 is equivalent to b.__setitem__(5, 9)

Fredrik's explanation makes sure to explain that *this* is the gotcha.
 
-- 
"I'll be ex-DPL soon anyway so I'm        |LUKE: Is Perl better than Python?
looking for someplace else to grab power."|YODA: No...no... no. Quicker,
   -- Wichert Akkerman (on debian-private)|      easier, more seductive.
For public key, finger moshez at debian.org  |http://www.{python,debian,gnu}.org



From ljohnson at resgen.com  Thu Mar 15 18:37:16 2001
From: ljohnson at resgen.com (Lyle Johnson)
Date: Thu, 15 Mar 2001 17:37:16 -0600
Subject: wxPython/Tkinter/pyGTK
References: <3AB11488.3010006@Antefacto.com>
Message-ID: <tb2kd5c5s30l13@corp.supernews.com>

> 3. Are there other options I haven't considered.

FXPy (the Python bindings for FOX) should be able to handle everything
you've described here. Here's a link:

    http://fxpy.sourceforge.net

Hope this helps,

Lyle




From scarblac at pino.selwerd.nl  Tue Mar 20 10:22:31 2001
From: scarblac at pino.selwerd.nl (Remco Gerlich)
Date: 20 Mar 2001 15:22:31 GMT
Subject: The most important modules
References: <Pine.GSO.4.21.0103191732210.22234-100000@laplace> <tber6micgl2hd3@corp.supernews.com>
Message-ID: <slrn9bet8p.bap.scarblac@pino.selwerd.nl>

Lyle Johnson <ljohnson at resgen.com> wrote in comp.lang.python:
> > I've seen a few comments along the lines of "When you are a newbie it will
> > take a while to realize which the important modules are."  As someone who
> > is in the middle of wading through Chapter 3 of the Library Reference, it
> > would be very helpful to know which ones are particularly worth
> > noting.  (Presumably  sys, (c)Pickle and shelve, from the ones I've so far
> > in the chapter - but what other treats are out there for me?)
> 
> For writing scripts that deal with files (whether for system administration
> or other tasks), you should get familiar with the "fileinput", "glob" and
> "os" modules.

And "shutil" if you have to copy or remove files. People forget about that
one.

> All of Python's support for regular expressions comes in
> through the "re" module, so that will be one you'll want for some
> text-processing needs. And I've been meaning to teach myself how to use the
> source-level Python debugger module, "pdb", but haven't gotten around to
> that yet;
> 
> A lot of the other modules get into more application-specific needs, whether
> it's for network support ("ftplib", "httplib", ...), XML processing
> ("xml.*") or others.

"urllib" is always so easy to use :)

> But it's all good stuff. And of course the standard
> Python library doesn't even scratch the surface of what's available; you'll
> also want to check some of the other sources for contributed software, like
> the Vaults of Parnassus:
> 
>     http://www.vex.net/parnassus/

Especially win32all for Windows, PIL for images, NumPy for numerical things,
and the mx* modules like mxDateTime (that should be in the standard library).

It doesn't hurt to just read through the library reference now and then, try
some things out, etc. I've never read it back to back, I would forget most
of it anyway...

-- 
Remco Gerlich


From sholden at holdenweb.com  Thu Mar 22 22:41:55 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 22 Mar 2001 22:41:55 -0500
Subject: email address verification
References: <99d5qi$95l$1@solaria.cc.gatech.edu>
Message-ID: <NFzu6.9043$G72.216162@e420r-atl1.usenetserver.com>

"Holland King" <insanc at cc.gatech.edu> wrote in message
news:99d5qi$95l$1 at solaria.cc.gatech.edu...
> does python provide quick/simple way to verify that an email address
exist?
> i looked at the smtp library and the verify method but that does not seem
> to be very standard anymore. is there anything else that is simalar?
> thank you.
>
The classic way would be to send it mail containing a link, which begins:
"To remove yourself from this mailing list, all you have to do is click
here". Naturally the URL clicked on adds the relevant email address to at
least several hundred salable lists, including "gullible individuals who
believe what they read in their inbox".

do-i-seem-a-bit-cynical-today-or-what?-ly y'rs  - steve





From shaleh at valinux.com  Fri Mar  2 11:42:09 2001
From: shaleh at valinux.com (Sean 'Shaleh' Perry)
Date: Fri, 02 Mar 2001 08:42:09 -0800 (PST)
Subject: gah! I hate the new string syntax
In-Reply-To: <UY5kBKANG8n6EwqG@jessikat.demon.co.uk>
Message-ID: <XFMail.20010302084209.shaleh@valinux.com>

On 02-Mar-2001 Robin Becker wrote:
> In article <0vpn79.331.ln at 127.0.0.1>, Carl Banks <idot at vt.edu> writes
>>Sean 'Shaleh' Perry <shaleh at valinux.com> wrote:
>>> return ";".join(["%s=%s" % (k, params[k]) for k in params.keys()])
>>>
>>> every day python seems to be moving closer to the line noise aspect of
>>> coding.
>>
>>I find it ok for methodlike functions such as split.
>>
>>For unmethodlike functions such as join, I would use the old
>>string.join syntax (for regular strings, anyways).
>>
>>"".join is just creepy.
>>
>>
> it looks plain wrong, but would ['a','b','c'].join(';') look any better?
> 
> In fact there may be some benefit in being able to say
> 
> pathJoiner = (sys.platform=='win32' and ';' or ':').join
> .....
> path1 = pathJoiner([d1,d2,.....])
> path2 = pathJoiner([e1,e2,.....])
> 

good ole:

from string import join

l = join(';', ['a','b','c']) # thanks, this is clear and obvious

Please o' gods of python, do not deprecate the clean and obvious syntax that is
found in the string module.



From mvanier at endor.bbb.caltech.edu  Sun Mar  4 00:49:58 2001
From: mvanier at endor.bbb.caltech.edu (Michael Vanier)
Date: 03 Mar 2001 21:49:58 -0800
Subject: writing a C function to manipulate python lists
References: <jllmqpnclz.fsf@endor.bbb.caltech.edu> <EIwn6.16175$Qb7.2638512@newsb.telia.net>
Message-ID: <jl7l26w0ah.fsf@endor.bbb.caltech.edu>

I should have RTFM before I posted.  I worked my way through the
embedded/extending parts of the online docs and got the function I needed
working in no time.  I think SWIG had conditioned me into thinking that
extending python at the C level was really hard (my fault, not SWIG's), but
it's actually quite easy, especially with the newer releases.

Thanks to all who replied!

Mike

--------------------------------------------------------------
Mike Vanier	mvanier at bbb.caltech.edu
Department of Computation and Neural Systems, Caltech 216-76
GNU/Linux: We can't lose; we're on a mission from God.


From will at nospam.misconception.org.uk  Mon Mar 19 13:10:45 2001
From: will at nospam.misconception.org.uk (Will Newton)
Date: Mon, 19 Mar 2001 18:10:45 +0000
Subject: Subsetting a dictionary
References: <993dkm$ttr$1@news6.svr.pol.co.uk> <993sif$92d$1@news.udel.edu>
Message-ID: <995hu4$ea7$1@newsg3.svr.pol.co.uk>

Terry Reedy wrote:

> Hmmm.  What you need is a currently imaginary feature that we might call a
> dict comprehension,
> with syntax paralleling that for list comprehensions:
> D2 = {key:val for (key,val) in D if constraint(key,val)}
 
That would indeed be cool. The worry I have with this syntax is the 
constraint will often be on only the value or the key (as in my case) and 
accessing the dict as a list of key-value tuples may be very slow and not 
much faster than the current way of doing things. (And why not use a list 
of tuples in this case?)

I'm not sure exactly how it would be best do this, but I think it could be 
a useful feature.




From xeno at bigger.aa.net  Sat Mar 10 01:01:28 2001
From: xeno at bigger.aa.net (Xeno Campanoli)
Date: 9 Mar 2001 22:01:28 -0800
Subject: LWP session problems
Message-ID: <3aa9c338$1_3@huge.aa.net>

I'm trying to make some Perlscripts work that need to make web client requests.
LWP, it turns out, is really hard (possibly impossible) to make work over multiple
requests for an SSL session because you can't seem to make it maintain the SessionId...
or at least, I haven't been able to make it do that, and nobody on the mailing lists
seems to have a clear answer.  

So, there must be a good (probably superior) web robot that does everything without
problems or obscurity or breakage in Python...Is there a class in Python I can use
effectively to program multiple requests in an SSL session?  If so, I'll try to use
it soon.  I keep hearing that Python and Python People do it better..so?

Sincerely, Xeno

-- 
Xeno Campanoli (erstwhile Xeno Whitenack, and Rick Burgess)
Email:	xeno at aa.net	(Web pages:  http://www.aa.net/~xeno)


From stephen_purcell at yahoo.com  Wed Mar 14 09:19:41 2001
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Wed, 14 Mar 2001 15:19:41 +0100
Subject: Better string.translate?
In-Reply-To: <3aaf758f$1_2@news.zet.net>; from stowasser.h@idowa.de on Wed, Mar 14, 2001 at 02:44:42PM +0100
References: <3aaf758f$1_2@news.zet.net>
Message-ID: <20010314151941.B30007@freedom.puma-ag.com>

Stowasser Harald wrote:
> Hello,
> i am very new in Python. I don't know how i can do this better:
> 
> _deltable=string.maketrans('.,:;-_?!.=()?[]{}\'\"\r\n\f\t\v','
> ')
> ret = string.translate(s,_deltable)
> 
> This works fine, but i need a 'anti-maketrans'.
> I want to define what characters NOT to change in a space-character.


Use the 'sub' function in the 're' module:

>> import re
>> re.sub('[^abcd]','', "string from which to remove all but a, b, c or d")
'cababcd'


-Steve


-- 
Steve Purcell, Pythangelist
Get testing at http://pyunit.sourceforge.net/
Any opinions expressed herein are my own and not necessarily those of Yahoo



From romberg at smaug.fsl.noaa.gov  Thu Mar 22 17:47:00 2001
From: romberg at smaug.fsl.noaa.gov (Mike Romberg)
Date: 22 Mar 2001 15:47:00 -0700
Subject: Is setup.py a "good thing"? (long)
References: <3aba76e2.167083508@24.0.228.33>
Message-ID: <ul3snk5a0bf.fsf@smaug.fsl.noaa.gov>

>>>>> " " == Mike Clarkson <support> writes:

     > setup.py was introduced in 2.1, presumably to help configure
     > and compile Python, so presumably it's a "good thing".  But I
     > have an uneasy feeling which I will try to explain.

[snip]

  I too have these same concerns about setup.py.  To me it seems that
distutils (which is what I think setup.py uses) does not yet work very
well with C extension modules.  So, why not use configure to deal with
these and let distutils do the pure python stuff.  I suppose distutils
could even call configure.  A setup like this would allow users to
specify Tcl/Tk and other libraries using one of the methods supported
by configure.  As of now I'm not real comfortable with having to hack
setup.py to get _tkinter to build.  I can do this but describing what
needs to be done to others is difficult.  It is more complicated that
a makefile or a simple command line argument to configure.

Mike Romberg (romberg at fsl.noaa.gov)



From grante at visi.com  Sat Mar 24 20:09:15 2001
From: grante at visi.com (Grant Edwards)
Date: Sun, 25 Mar 2001 01:09:15 GMT
Subject: static methods
References: <mailman.985383828.2942.python-list@python.org> <slrn9bpqgo.15q.grante@tuxtop.visi.com> <mailman.985457492.29908.python-list@python.org> <7h1qbt8h8aos1parc090s8tqu2ooc04t4d@4ax.com> <slrn9bqc2r.7nq.grante@isis.visi.com> <4lgqbto46c4c6i307lnq0e614uujd2fsss@4ax.com>
Message-ID: <slrn9bqoet.1bs.grante@tuxtop.visi.com>

On Sun, 25 Mar 2001 00:52:06 GMT, Courageous <jkraska1 at san.rr.com> wrote:

>>>Hrm. I'd forgotten that. In languages where there are no namespaces,
>>>it's useful to be able to make certain that symbols don't go out and
>>>pollute the global namespace in which ALL symbols reside. This is
>>>what that's for.
>>
>>I'm confused -- are you saying that C doesn't have namespaces?
>
>Erm, well, unless C99 defined them, no.

There's a global namespace, a "file" namespace, and each set of
{} defines a namespace (they're all nested).  Somehow I don't
think that's what you're talking about...

-- 
Grant Edwards                   grante             Yow!  World War III? No
                                  at               thanks!
                               visi.com            


From JohnH at PHM.GOV.AU  Thu Mar  1 23:13:20 2001
From: JohnH at PHM.GOV.AU (Hirsch, John)
Date: Fri, 2 Mar 2001 15:13:20 +1100 
Subject: Cannot find win32gui.IsIconic()
Message-ID: <A90DD8D7B82FD311B4D00008C7DBD6EAB5EBDE@phm-exch.phm.gov.au>

Hi All
I am trying to use win32gui.IsIconic() to check if a window has been
minimized.
But Python gives me an AttributeError when I try to call it.
According to the win32gui documentation it should be there.
Am I doing something wrong.

Thanks
Jon




From jra at dorothy.msas.net  Thu Mar 29 17:44:39 2001
From: jra at dorothy.msas.net (Jay R. Ashworth)
Date: Thu, 29 Mar 2001 22:44:39 GMT
Subject: Linux GPS program
References: <3AC01637.4EF3A6F2@crynwr.com> <slrn9c17k4.cfc.jra@dorothy.msas.net> <3AC0D915.83C2C0C3@crynwr.com> <slrn9c1p9t.d3g.jra@dorothy.msas.net> <Pine.SOL.4.30.0103291940140.13267-100000@primrose.csv.warwick.ac.uk>
Message-ID: <slrn9c7emm.sv.jra@dorothy.msas.net>

This one time, in band camp,
  John J. Lee <phrxy at csv.warwick.ac.uk> wrote:
> On Tue, 27 Mar 2001, Jay R. Ashworth wrote:
> > This one time, in band camp,
> >   Russell Nelson <nelson at crynwr.com> wrote:
> [...]
> > > No.  I used pygtk, which is a set of Python bindings to the gtk
> [...]
> 
> Okay, he does know what it (free software) means, I was wrong.

You may have missed that this is *Russ Nelson*, who's been doing
commercial support for the *free* Clarkson/Crynwyr packet drivers for
about 300 years...  Yeah, he knows what "free" means.

Cheers,
-- jra
-- 
Jay R. Ashworth                                                jra at baylink.com
Member of the Technical Staff     Baylink
The Suncoast Freenet         The Things I Think
Tampa Bay, Florida        http://baylink.pitas.com             +1 727 804 5015


From claird at starbase.neosoft.com  Mon Mar 12 12:40:31 2001
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 12 Mar 2001 11:40:31 -0600
Subject: Magnitude of the wx* market (was: Python In A Nutshell - suggestions)
References: <nG5q6.54910$__6.11376601@typhoon.southeast.rr.com> <Ufcq6.140786$47.2078615@news.bc.tac.net> <3AAA4FE3.B471B689@northwestern.edu>
Message-ID: <A909AD29CBD32AFC.4F1641C135749F87.7149617C7C166896@lp.airnews.net>

In article <3AAA4FE3.B471B689 at northwestern.edu>,
Louis Luangkesorn  <lluang at northwestern.edu> wrote:
			.
			.
			.
>And while I'm at it, one useful book to have would be on wxWindows/wxPython.  I see references to it in several places, and a few intros on the web, but
			.
			.
			.
This is an interesting point.  It's typical in computer
book publishing that publishers want more than authors
can produce; roughly, someone like ORA would happily
print many more titles than they can find authors willing
and able to write.

I happen to know, though, that several publishers have
rejected wxWindows proposals.

I doubt, though, that there'd be much trouble making the
case that the subject deserves a chapter in a larger work.
-- 

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


From whisper at oz.net  Thu Mar 29 15:29:19 2001
From: whisper at oz.net (Dave LeBlanc)
Date: 29 Mar 2001 20:29:19 GMT
Subject: htmllib samples
References: <3AC2F48E.4EF77C71@aon.at> <gKCw6.4130$4N4.837469@newsc.telia.net>
Message-ID: <9a05uv$png$0@216.39.170.247>

Ok, i'll bite - why would you want to use the significantly greater
complexity of SGML if you think you want to do HTML?

Regards,

Dave LeBlanc

P.S Looked at the book niblet on O'Reilly - looks good. Can't wait to
see it in print!

On Thu, 29 Mar 2001 08:53:32 GMT, "Fredrik Lundh"
<fredrik at pythonware.com> wrote:

<snip>
>note that when you think you need the htmllib module, you
>usually want the sgmllib module:
<snip>
>Cheers /F



From sholden at holdenweb.com  Tue Mar 13 02:33:10 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Tue, 13 Mar 2001 02:33:10 -0500
Subject: learning C
References: <mailman.983972829.23481.python-list@python.org> <983975751.18833@newsmaster-04.atnet.at> <slrn9adklr.bv9.scarblac@pino.selwerd.nl> <984047775.592522@newsmaster-04.atnet.at> <9896hk$ig8$1@verence.demon.co.uk> <984134612.781779@newsmaster-04.atnet.at> <3AACE22D.3A9EEB12@bt.com>
Message-ID: <j6kr6.68748$ML1.3282888@e420r-atl2.usenetserver.com>

"Alan Gauld" <alan.gauld at bt.com> wrote in message
news:3AACE22D.3A9EEB12 at bt.com...
> Werner Schiendl wrote:
> > Unsafe is somewhat the same topic I think: It's a question
> > of disciplined use.
>
> I think one problem with C++ is the number of temporary objects
> it creates.
[ ... ]
>
> In C, by virtue of its fixed primitive types, is much easier
> to gurantee what will happen by way of memory allocation etc.
>
> > > There's the feeling that you never quite know what it's
> > > going to do.
>
> Temporary objects being one such case. Others include the use of
> operator overloading, function overloading vv default parameters
> etc etc. C++ is a maintenance nightmare compared to C - I speak
> from 18 months painful experience maintaining a large C++ system!
>
> > You can have the same in C++ (using one or more static objects,
> > or construct the objects just once when booting)
>
> Even here the compiler will cause tempoiray object copies to
> occur when passing object references around. Static objects are
> not as 'static' as we might like to think!
>
>
> > This is nothing more difficult than in C.
>
> Sorry but it is. There's just so much less lurking under
> the hood in C.
>
But by this argument, assembly code is even better!  It's always a question
of selecting language features because they are appropriate to your problem
domain.

or-doing-what-the-pointy-haired-managers-tell-you-ly y'rs  - steve





From robin.thomas at starmedia.net  Thu Mar  1 01:58:49 2001
From: robin.thomas at starmedia.net (Robin Thomas)
Date: Thu, 01 Mar 2001 01:58:49 -0500
Subject: d = {}; d[0:1] = 1; d[0:1] = 2; print d[0:1]
In-Reply-To: <97kotr$su9$0@216.39.151.169>
References: <mailman.983419142.24027.python-list@python.org>
Message-ID: <4.3.1.2.20010301014119.00cf3490@exchange.starmedia.net>

At 06:06 AM 3/1/01 +0000, Donn Cave wrote:

>If it helps, you ruined my day.

Now I'm not alone!


>I think we might be able to do better.  I hacked in a quick fix
>in ceval.c that looks to me like it has the desired effect without
>closing the door to intentional slice keys (however unlikely.)

Cool. I think that the abstract object layer saves us, though:

- slice assignment is handled by PySequence_SetSlice
- *intentional* assignment of slice object is handled by PyObject_SetItem

PySequence_SetSlice, strangely, has a big chunk of code expressly written 
to support slice assignment to mapping objects. From Objects/abstract.c 
below. If we remove that code, we get the same effect.

int
PySequence_SetSlice(PyObject *s, int i1, int i2, PyObject *o)
{
         PySequenceMethods *m;
         PyMappingMethods *mp;

         if (s == NULL) {
                 null_error();
                 return -1;
         }

         m = s->ob_type->tp_as_sequence;
         if (m && m->sq_ass_slice) {
                 if (i1 < 0 || i2 < 0) {
                         if (m->sq_length) {
                                 int l = (*m->sq_length)(s);
                                 if (l < 0)
                                         return -1;
                                 if (i1 < 0)
                                         i1 += l;
                                 if (i2 < 0)
                                         i2 += l;
                         }
                 }
                 return m->sq_ass_slice(s, i1, i2, o);
         /* the evil begins */
         } else if ((mp = s->ob_type->tp_as_mapping) && mp->mp_ass_subscript) {
                 int res;
                 PyObject *slice = sliceobj_from_intint(i1, i2);
                 if (!slice)
                         return -1;
                 res = mp->mp_ass_subscript(s, slice, o);
                 Py_DECREF(slice);
                 return res;
         }

         type_error("object doesn't support slice assignment");
         return -1;
}

Anybody have an explanation why this behavior was desired?

>I don't know if this is the answer, but it at least demonstrates
>that [:] and slice() can be separate issues at some level.

They're separate issues even for lists:

 >>> l = []
 >>> l[0:1] = [1]
 >>> f = slice(0,1)
 >>> l[f] = [1]
Traceback (innermost last):
   File "<pyshell#21>", line 1, in ?
     l[f]= [1]
TypeError: sequence index must be integer

The question remains: should slice objects be hashable/comparable so that

 >>> slice(0,1) == slice(0,1)
1

I say YES YES YES. What do you say?


--
Robin Thomas
Engineering
StarMedia Network, Inc.
robin.thomas at starmedia.net




From pfortin at pfortin.com  Tue Mar 13 19:35:22 2001
From: pfortin at pfortin.com (Pierre Fortin)
Date: Tue, 13 Mar 2001 19:35:22 -0500
Subject: PyQT-2.3 compile fails...
Message-ID: <3AAEBCCA.4E53671@pfortin.com>

Hi,

Anyone have a clue as to why others can compile PyQT-2.3; but I get errors...? 
I've also tried the CVS files...

I'd hate to think this is a symlink related problem:
/usr/local/src --> /hd/b6/src

Thanks,
Pierre

# make
make  all-recursive
make[1]: Entering directory `/hd/b6/src/Python/PyQt-2.3'
Making all in qt
make[2]: Entering directory `/hd/b6/src/Python/PyQt-2.3/qt'
/bin/sh ../libtool --mode=compile c++ -DHAVE_CONFIG_H -I. -I. -I.. 
-I/usr/local/include/python2.0 -I/usr/local/include/sip -I/usr/local/qt/include 
-I/usr/X11R6/include  -fno-exceptions -O2 -c sipqtQString.cpp
c++ -DHAVE_CONFIG_H -I. -I. -I.. -I/usr/local/include/python2.0
-I/usr/local/include/sip -I/usr/local/qt/include -I/usr/X11R6/include
-fno-exceptions -O2 -Wp,-MD,.deps/sipqtQString.pp -c  -fPIC -DPIC
sipqtQString.cpp -o sipqtQString.lo
sip/qstring.sip: In function `int sipConvertTo_QString(PyObject *, QString **,
int, int *)':
sip/qstring.sip:456: no matching function for call to `QString::setUnicodeCodes
(short unsigned int *&, int &)'
make[2]: *** [sipqtQString.lo] Error 1
make[2]: Leaving directory `/hd/b6/src/Python/PyQt-2.3/qt'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/hd/b6/src/Python/PyQt-2.3'
make: *** [all-recursive-am] Error 2


From thomas.heller at ion-tof.com  Thu Mar 22 04:01:48 2001
From: thomas.heller at ion-tof.com (Thomas Heller)
Date: Thu, 22 Mar 2001 10:01:48 +0100
Subject: Standalone executables
References: <998igr$4al$1@niagara.nonet> <iXWt6.156032$Z8.33024358@typhoon.southeast.rr.com> <99aldn$7dh$1@niagara.nonet>
Message-ID: <99cf1s$hrsb$1@ID-59885.news.dfncis.de>

> I wonder, for Linux, what if I wrote in Jython, compiled with jythonc,
> then used a Java native-code compiler? Would that be too insane? (-:
>
With virtually every Linux coming with python, why is this needed at all?

Thomas




From dmaas at dcine.com  Mon Mar 12 00:55:04 2001
From: dmaas at dcine.com (Dan Maas)
Date: Mon, 12 Mar 2001 05:55:04 GMT
Subject: Poll'able Queue objects.
References: <txkr904gwu9.fsf@vortex.ukshells.co.uk> <cuRq6.240807$o91.31035696@typhoon.nyroc.rr.com> <slrn9aolc9.sci.grante@isis.visi.com>
Message-ID: <YwZq6.241551$o91.31418266@typhoon.nyroc.rr.com>

> The mind wobbles.  One of the things that Unix got right is that you can
> select() (or poll()) on pipes, sockets, serial ports (any properly written
> character device that's properly written), or, in degenerate-case sort of
> way, "regular" files.

As much of a Unix fan as I am, we've got it worse... Pthreads condition
variables. SysV semaphores and message queues. Asynchronous disk I/O (POSIX
AIO). UNIX signals. None of these "waitable objects" can be multiplexed
together with file descriptors*... Also the "stateless" paradigm of select()
and poll() is not always preferable to "stateful" Windows message loops -
consider the problems involved with using two unrelated Unix libraries that
each want control of the main loop...

AFAIK MsgWaitForMultipleObjects() handles Win32 message queues, semaphores,
mutexes, async disk I/O, network I/O, etc. The API is not as clean but the
functionality is there =)

Dan

* well FreeBSD has kqueue, those lucky bastards =)




From dwig at advancedmp.net  Mon Mar 19 12:39:47 2001
From: dwig at advancedmp.net (Don Dwiggins)
Date: 19 Mar 2001 09:39:47 -0800
Subject: Pep 238 Why new // operator (Non-integer Division)
In-Reply-To: "Tim Peters"'s message of "Sat, 17 Mar 2001 04:50:30 -0500"
References: <LNBBLJKPBEHFEDALKOLCIEEJJGAA.tim.one@home.com>
Message-ID: <usnk9ac9o.fsf@advancedmp.net>

Tim Peters writes:
> [Alex Martelli]
>> So what's so terrible about non-infix notation, a builtin div(a,b) [and
>> mod(a,b) too since we're at it...?].  Sure, infix can be handy --

> Infix is also a speed issue in Python, since syntax is fixed but builtin
> names are not reserved; so the speed freaks will surely argue for syntax.

>> but then this might be the occasion to steal a leaf from Haskell's
>> book... let any 2-operand function be used as an infix operator
>> (possibly with some stropping if that is needed to disambiguate
>> syntax).

Many Prologs also have this capability, with a way to define prefix, infix,
and postfix operators, along with precedence values to determine grouping.

-- 
Don Dwiggins                    "Solvitur Ambulando"
Advanced MP Technology
dwig at advancedmp.net




From annis at biostat.wisc.edu  Mon Mar 26 09:24:28 2001
From: annis at biostat.wisc.edu (William Annis)
Date: 26 Mar 2001 08:24:28 -0600
Subject: Daemonizing a python programme.
References: <mailman.985376338.19672.python-list@python.org> <3ABBE09E.3287DEC5@qvlinc.com>
Message-ID: <jtofuo1ucj.fsf@wazor.biostat.wisc.edu>

Coy Krill <ckrill at qvlinc.com> writes:
> class NullDevice:
>     def write(self, s):
>         pass
> 
> def daemonize():
>     if (not os.fork()):
>         # get our own session and fixup std[in,out,err]
>         os.setsid()

        I include these here:

        os.chdir(some_sensible_root_dir_for_core_files)
        os.umask(000)     # or whatever seems sensible

        See chapter 13, "Daemon Processes," pp.415-425 of "Advanced
Programming in the Unix Environment" by W. Richard Stevens for a full
treatment.

>         sys.stdin.close()
>         sys.stdout = NullDevice()
>         sys.stderr = NullDevice()

        This is curious.  Why not just close down STDOUT and STDERR,
too? 

-- 
William Annis - System Administrator - Biomedical Computing Group
annis at biostat.wisc.edu                       PGP ID:1024/FBF64031
Mi parolas Esperanton - La Internacian Lingvon  www.esperanto.org


From jmarshal at mathworks.com  Mon Mar  5 12:25:46 2001
From: jmarshal at mathworks.com (Joshua Marshall)
Date: 5 Mar 2001 17:25:46 GMT
Subject: Little problem with the "+" operator
References: <983802310.155838463@news.t-online.de> <mailman.983804468.16428.python-list@python.org>
Message-ID: <980i6q$crs$1@news.mathworks.com>

Steve Purcell <stephen_purcell at yahoo.com> wrote:
> Zamurai wrote:
>> variableOne = 500
>> myVariable = variableOne + "cool"
>> 
>> But I get the following error: "unsupported operand types for +"
>> 
>> Does someone know how to solve the problem?

> You can't add variables of different types together in Python and expect them
> to magically agree on a sensible result.

> So, you have to explicitly say:

>>>> variableOne = 500
>>>> myVariable = str(variableOne) + "cool"
>>>> myVariable 
> '500cool'
>>>> 

> Or,

>>>> variableOne = 500
>>>> myVariable = variableOne + int("cool")
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
> ValueError: invalid literal for int(): cool
>>>> 


> This is infinitely more sane than:

> % perl -e 'print 500 + "cool"'
> 500

Comparing Python to Perl here doesn't seem very fair.  They're very
different languages.  In general, I'm surprised by the urgency with
which people often jump to compare them.


From sholden at holdenweb.com  Tue Mar 27 23:39:02 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Tue, 27 Mar 2001 23:39:02 -0500
Subject: newbie IDLE, Tkinter
References: <FHUu6.68719$A6.14534907@news4.rdc1.on.home.com> <01c0b6e2$3d426740$4ecb0c28@PC0AB085S5A8>
Message-ID: <mZdw6.1046$m04.190645@e420r-atl1.usenetserver.com>

"Randy Hawley" <Hawley_Randall_R at Lilly.com> wrote in message
news:01c0b6e2$3d426740$4ecb0c28 at PC0AB085S5A8...
> Howdy Python Gurus.
>
> I too am a newbie trying out Python on Win32 platforms.
>
> I have the same problems as Nick describes below.  Also, somethimes when
> I hit a button that has "command=root.quit" defined as its action, not
only
> does my program's window quit, but so does IDLE.
>
> What gives with this?  Anybody else ever run into this?  Is it a result of
> poor coding in my script (saw it when I just ran the examples from the
> Tkinter tutorial, too) ?
>
> Thanks,
> Randy Hawley
>
> Nick Perkins <nperkins7 at home.com> wrote in article
> <FHUu6.68719$A6.14534907 at news4.rdc1.on.home.com>...
[ ... similar problems ... ]

The difficulty you are experiencing is to do with IDLE being a Tkinter
application itself, I suspect. I normally run Tkinter-based programs from a
command window to avoid the unpredictable interactions you describe. I guess
one day IDLE may be able to run programs as a separate process.

regards
 Steve





From bbollenbach at homenospam.com  Tue Mar  6 15:41:34 2001
From: bbollenbach at homenospam.com (Brad Bollenbach)
Date: Tue, 06 Mar 2001 20:41:34 GMT
Subject: Can Python replace Visual Basic? Should it?...and generalities : )
References: <mailman.983905329.22236.python-list@python.org>
Message-ID: <2Ybp6.12845$hn5.1818206@news1.rdc1.mb.home.com>

This reply is interesting, but still contains a lot of the "I've started
invistigating Python and these are the things I like about it" comments,
which I was hoping to avoid.

[snip]

>   A Python "OCX" control needs to be written.  This control
>   would simply encapsulate Python for us Visual Basic idiots.
>   This way we could 'pop in the control in our VB project and
>   start using Python for parts of the program.

That's a neat idea. I hope someone smarter than me decides to do it. :)

>   At first, this control could be non-visual.  The OCX would
>   have to expose through IDispatch the objects managed within
>   the embedded python system. Then, it  would be cool if it
>   became a wxWindows frame, so that small client areas could
>   be re-written in Python, etc.  For extra bonus points,
>   the visual control would allow OCX components to be dropped.
>
>   Eventually the VB programmer could 'drop the VB shell after
>   their entire program is re-written.  In the corporate world,
>   it will be damn near impossible to get funding for re-writing
>   a program in Python.  However, it is very possible to have
>   this conversion be INCREMENTAL, one sub-routine at a time!
>   In short, a complete re-write is just too high of risk for
>   most corporate progammers to do...

"Damn near impossible", but not "impossible". I know, because where I work,
we ARE almost done a complete rewrite of an app that was written in
Powerbuilder 4.x (the exact version, I'm not sure), now being completely
rewritten (with a totally new database) in Powerbuilder 6.5.1. To give you
an idea of the size of the app, there are 200+ stored procs behind it, and
about 30,000-40,000 lines of Powerscript code (but that's totally ballpark,
I'm just guessing). And with the dime-a-dozen comments that keep saying how
ridiculously simple Python is, you'd think that maybe we should have chosen
Python instead of a newer version of Powerbuilder?! Hmm...the more I think
about it, the more I think we chose the right tool for the job.

But then, as seems to be often the case, people seem to blur the
differentiation between "easy programming languages" and "easy programming
tasks". How many people have written 30,000 lines out of _any_ programming
language and still refer to it as being simple to work with?

As, you mentioned in a reply to the other thread of this thread (heh),
coding really is the most insignificant part of a large programming project
in a coporate environment. In theory at least, you should have a clear idea
of what you're going to do before you do it, in which case the coding
(almost irregardless of the language) is often times just grunt work.




From topmind at technologist.com  Wed Mar 21 05:07:26 2001
From: topmind at technologist.com (topmind)
Date: Wed, 21 Mar 2001 10:07:26 GMT
Subject: Collection interfaces
References: <3a90fa97.0@news1.mweb.co.za>   <MPG.1503548620e632919896a4@news.earthlink.net> <MCjm6.474652$U46.14163288@news1.sttls1.wa.home.com> <MPG.150394195f7f04a19896a7@news.earthlink.net> <3A9A86B1.9A8C69A2@ix.netcom.com> <MPG.1504d14563dce6499896af@news.earthlink.net> <97fjla128ir@news2.newsguy.com> <MPG.1505a76831c34b749896b3@news.earthlink.net> <97h5s40vhc@news2.newsguy.com> <MPG.1506007591cf818b9896bc@news.earthlink.net> <slrn99opop.gr.amk@mira.erols.com> <MPG.1506265ebf6d934d9896be@news.earthlink.net> <97ibmc0k47@news1.newsguy.com> <MPG.1506cf3293b094b49896c1@news.earthlink.net> <97m0c7080c@news2.newsguy.com> <MPG.15085517debf4b649896ca@news.earthlink.net> <zb%n6.31033$o%6.586026@news1.rdc1.bc.home.com> <MPG.150b383cea226d409896d0@news.earthlink.net> <kNko6.6986$eq3.707121@news.uswest.net> <MPG.150c46db9cec283e9896d4@news.earthlink.net> <L9Go6.2720$hZ4.875436@news.uswest.net> <DWCt6.12229$Im6.1309770@newsread1.prod.itd.earthlink.net> <jWEt6.7867$3_.998377@news.uswest.net>
Message-ID: <y3%t6.17161$227.1600545@newsread2.prod.itd.earthlink.net>

----- Original Message -----
From: "Randy A. Ynchausti" <mickeyz at qwest.net>
Newsgroups:
comp.object,comp.lang.smalltalk,comp.lang.python,comp.lang.java.advocacy
Sent: Tuesday, March 20, 2001 1:01 AM
Subject: Re: Collection interfaces


> topmind,
>
> > Okay.
> >
> > For the sake of argument, let's say that OO shines if the app is very
> mathy.
> >
> > But, what about the rest?
>
> What the rest any different from the very mathy?  The classes and objects
> don't care what their behaviors are.
>

They do if their behavior is mathy.

> It use to be that people said you should never implement a compiler using
OO
> because for all of the "obvious" reasons.  Then someone did it and found
out
> that their solution was essentially as fast, easier to maintain, and
> produced a smaller code base.  As far as I am concerned the same
experience
> was had by the "data-visualization" industry.
>
> Now -- where is that procedural code -- I have the OO code in hand?
>
> Regards,
>
> Randy
>
>

I don't consider that a "typical custom biz app" because it requires a
certain
math *specialty* and I don't know enough about that narrow specialty to
comment.

Most apps I work with use very little math beyond adding and multiplying
and dividing. Thus, my math skills have greatly rusted over the years, as
have
many other biz app programmer's.

This is NOT saying that biz apps are simple, but just not complexity in the
math sense.

-tmind-





From daniel at dittmar.net  Wed Mar 14 15:04:26 2001
From: daniel at dittmar.net (Daniel Dittmar)
Date: Wed, 14 Mar 2001 21:04:26 +0100
Subject: Dynamic construction of a print format string
References: <3AAFB663.57FCCE7@troikanetworks.com>
Message-ID: <3AAFCECA.18663E5D@dittmar.net>

If you want to set some format size dynamically, you can use the *
option.

"%*s" % (namelen, data)

Not quite the asnwer to your question, but maybe the solution to your
problem.

Daniel

Bruce Edge wrote:
> 
> I'm using:
> 
>   fmtstr = "%s%s%s" % ( "    %", namelen, "s - %s" )
> 
> to get:
> 
>   "    %19s - %s"
> 
> Is there a better way?
> 
> TIA, Bruce.


From tchur at optushome.com.au  Tue Mar 13 13:39:06 2001
From: tchur at optushome.com.au (Tim Churches)
Date: Wed, 14 Mar 2001 05:39:06 +1100
Subject: This math scares me
References: <3aad4062.1290021@News.CIS.DFN.DE> <slrn9aqkth.aki.scarblac@pino.selwerd.nl> <3aad6ffe.3692710@News.CIS.DFN.DE>
Message-ID: <3AAE694A.765076D3@optushome.com.au>

Costas Menico wrote:
> >> Ok, I can see maybe division having problems. But why does addition of
> >> the two numbers below:
> >>
> >> 5.01+5.54
> >>
> >> give me this?
> >>
> >> 10.550000000000001
> >>
> >> I hope its some setting that can take care of this. I am using Python
> >> under Windows 98.
> >
> >It's worse than that. Try:
> >
> >>>> 5.01
> >5.0099999999999998
> >

I don't think anyone has mentioned Jurjen Bos' real-accurate module (see
ftp://ftp.python.org/pub/www.python.org/ftp/python/contrib-09-Dec-1999/DataStructures/).
Alas it doesn't fix the finite precision limitation which these
newfangled electric digital computer gizmos have when it comes to
floating point numbers, but it does make the limitation explicit, viz:

>>> import real
>>> a = real.r(5.01)
>>> b = real.r(5.54)
>>> a
5.010000000000000+-1
>>> b
5.540000000000000+-1
>>> a + b
10.549999999999997+-4

give-or-take-a-bit-ly y'rs - Tim C



From brian_zhou at agilentNOSPAM.com  Sun Mar 18 18:22:35 2001
From: brian_zhou at agilentNOSPAM.com (Brian Zhou)
Date: Sun, 18 Mar 2001 15:22:35 -0800
Subject: [Announce] jywiki
Message-ID: <984957750.523849@emperor.labs.agilent.com>

This is a wiki clone ported from Martin Pool's PikiPiki to Jython Servlet.

The project is hosted here: http://sourceforge.net/projects/jywiki

And there is a wiki (using phpwiki for now): http://jywiki.sourceforge.net
with information about Jython Servlet in general and jywiki in specific.

It's still work in progress since I don't have a lot of time to spare.

Cheers,

-Brian Zhou




From aleaxit at yahoo.com  Mon Mar 12 04:28:07 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Mon, 12 Mar 2001 10:28:07 +0100
Subject: any interest in type-scoped static constants?
References: <3AAB9578.4C8754E4@fxtech.com> <98gs2i$2vtp$1@news1.simtel.ru>
Message-ID: <98i4r70p4r@news2.newsguy.com>

"Alexander Semenov" <sav at ulmen.mv.ru> wrote in message
news:98gs2i$2vtp$1 at news1.simtel.ru...
    [snip]
> > If I create an extension type for Point, I can't do this:
> >
> > p = Point.ZERO
>
> In Python:
>
> class Point:
>     def __init(self, x, y): pass
>
> Point.ZERO = Point(0,0)
>
> Then you can write p = Point.ZERO.
> It's even simpler and clearer then C++

Yes, but it doesn't apply to *extension types* (which you write
in C or C++, then *use* from Python), only to classes you write
in Python itself.

One solution is, indeed, to split the extension into two parts:
one, written in C or C++, is in some _foo.pyd (the leading _
suggests it's internal-use-only), and it's wrapped by a foo.py
which adds the niceties -- a rather popular solution (you only
use C or C++ for truly speed-critical parts).

Another is to use the Boost Python Library (www.boost.org) to
write your extension (in C++, only -- C is not supported).  BPL
uses the Beaudry Hook so the 'types' it exports are actually
'extension classes', so you can do a lot with them that would
not be feasible with 'just-types', including using attributes.


Alex





From bfleming at bserv.com  Thu Mar  1 15:03:56 2001
From: bfleming at bserv.com (Bruce Fleming)
Date: Thu, 01 Mar 2001 20:03:56 GMT
Subject: Can SAX parser return <!-- COMMENTS -->  ???
Message-ID: <MWxn6.15038$x27.906254@news1.rdc2.on.home.com>

How can I expose comments in the content handler ?
It does not return comments by default, and I cannot
find where to hook into the class..?




From ben.hutchings at roundpoint.com  Tue Mar  6 16:42:31 2001
From: ben.hutchings at roundpoint.com (Ben Hutchings)
Date: 06 Mar 2001 13:42:31 -0800
Subject: Little problem with the "+" operator
References: <983802310.155838463@news.t-online.de> <mailman.983804468.16428.python-list@python.org> <980i6q$crs$1@news.mathworks.com> <mt1yscm7yu.fsf@astron.berkeley.edu> <3aa3d3a7.420988249@news.uswest.net>
Message-ID: <uvgpmlgl4.fsf@roundpoint.com>

joconnor at cybermesa.com (Jay O'Connor) writes:
<snip>
> Actually, JavaScript will allow you to add numbers to strings, with
> weird results
> 
> '100' + 200 = '100200'
> 100 + '200' = 300
> 
> I *think* that's how it works, I just know it's not very safe	

So my mathematics teachers were lying when they said addition was
commutative!

Seriously, that's just awful behaviour.  I now have this vision of
what someone at Netscape must have written when hacking JavaScript
(ne? LiveScript) together:

    class LiveScriptObject {
        virtual LiveScriptObject * add(LiveScriptObject const * other) = 0;
    };
    class LiveScriptInteger {
        virtual LiveScriptObject * add(LiveScriptObject const * other) {
            return new LiveScriptInteger(integer_ + other->toInteger());
        }
    };
    class LiveScriptString {
        virtual LiveScriptObject * add(LiveScriptObject const * other) {
            return new LiveScriptString(string_ + other->toString());
        }
    };

Yes, addition is *obviously* just a function that's dynamically
dispatched on its left argument.  Any web page author should know
*that*.

-- 
Any opinions expressed are my own and not necessarily those of Roundpoint.


From niessink at serc.nl  Mon Mar 19 07:56:53 2001
From: niessink at serc.nl (Frank Niessink)
Date: 19 Mar 2001 12:56:53 GMT
Subject: Tkinter Canvas inheritance
References: <3AB5EC65.DF767FDC@ira.uka.de>
Message-ID: <994vml$c20$1@newshost.accu.uu.nl>

Markus von Ehr <vonehr at ira.uka.de> wrote:
> 
> When I change the Code and create the image in the main function,
> there's no problem, it works well. How can I do it in the constructor
> and why doesn't it work????

You need to keep a reference to the image else it gets garbage collected.

Change:
>        slider = PhotoImage(file="button.gif")

into:
>        self.slider = slider = PhotoImage(file="button.gif")

Cheers, Frank
-- 
See first, think later, then test. But always see first. Otherwise you will
only see what you were expecting. Most scientists forget that.
	-- Douglas Adams, 'So long, and thanks for all the fish'


From parkw at better.net  Sat Mar 17 19:49:29 2001
From: parkw at better.net (William Park)
Date: Sat, 17 Mar 2001 19:49:29 -0500
Subject: Interactive startup file
In-Reply-To: <3dTs6.2144$ge.55105@news3.mia>; from brendhanhorne@bellsouth.net on Sat, Mar 17, 2001 at 07:22:15PM -0500
References: <3dTs6.2144$ge.55105@news3.mia>
Message-ID: <20010317194929.A9033@better.net>

On Sat, Mar 17, 2001 at 07:22:15PM -0500, Brendhan Horne wrote:
> In the tutorial it  says in section 2.2.3
...
> import os
> filename = os.environ.get('PYTHONSTARTUP')
> if filename and os.path.isfile(filename):
>     execfile(filename)
...
> Can some one please explain this to me in novice english and for a guy
> who is using windows98 with his python.

Well, it's exactly what it says.

If you set environment variable
    PYTHONSTARTUP='.pythonrc'
then commands in file '.pythonrc' will be executed everytime you run
python in interactive mode (ie. when you get '>>>' prompt).  For
example, mine contains
    import rlcompleter
    rlcompleter.readline.parse_and_bind('tab: complete')

If you want to run those commands from a script, then you have to call
the file explicitly, using
    execfile('.pythonrc')

:wq --William Park, Open Geometry Consulting, Linux/Python, 8 CPUs.



From t*b*r*u*m*l*e*y at f*u*s*e.n*e*t  Fri Mar 23 08:04:34 2001
From: t*b*r*u*m*l*e*y at f*u*s*e.n*e*t (Troy Brumley)
Date: Fri, 23 Mar 2001 08:04:34 -0500
Subject: Collection interfaces
References: <3a90fa97.0@news1.mweb.co.za>   <3A9A86B1.9A8C69A2@ix.netcom.com> <MPG.1504d14563dce6499896af@news.earthlink.net> <MPG.1505a76831c34b749896b3@news.earthlink.net> <MPG.1506007591cf818b9896bc@news.earthlink.net> <MPG.1506265ebf6d934d9896be@news.earthlink.net> <97ibmc0k47@news1.newsguy.com> <MPG.1506cf3293b094b49896c1@news.earthlink.net> <97m0c7080c@news2.newsguy.com> s.earthlink.net> <zb%n6.31033$o%6.586026@news1.rdc1.bc.home.com> <MPG.150b383cea226d409896d0@news.earthlink.net> <kNko6.6986$eq3.707121@news.uswest.net> <MPG.150c46db9cec283e9896d4@news.earthlink.net> <3AA2A898.450B659E@mail.com> <gRCt6.12224$Im6.1308560@newsread1.prod.itd.earthlink.net> <3AB755F8.67E84126@mail.com> <%a%t6.17245$227.1601569@newsread2.prod.itd.earthlink.net> <3AB8A730.4BE13D22@mail.com> <B6DE17D2.D6B0%t*b*r*u*m*l*e*y@f*u*s*e.n*e*t> <anhu6.18497$Im6.1938203@newsread1.prod.itd.earthlink.net> <B6DF5D0D.D7AE%t*b*r*u*m*l*e*y@f*u*s*e.n*e*t> <Q7Bu6.2721$P4.207644@newsread1.prod.itd.earthlink.net>
Message-ID: <B6E0B412.D887%t*b*r*u*m*l*e*y@f*u*s*e.n*e*t>

in article Q7Bu6.2721$P4.207644 at newsread1.prod.itd.earthlink.net, topmind at
topmind at technologist.com wrote on 3/23/01 12:26 AM:

> 
> "Troy Brumley" <t*b*r*u*m*l*e*y at f*u*s*e.n*e*t> wrote in message
> news:B6DF5D0D.D7AE%t*b*r*u*m*l*e*y at f*u*s*e.n*e*t...
>> in article anhu6.18497$Im6.1938203 at newsread1.prod.itd.earthlink.net,
> topmind
>> at topmind at technologist.com wrote on 3/22/01 1:57 AM:
>> 
>>> 
>>> "Troy Brumley" <t*b*r*u*m*l*e*y at f*u*s*e.n*e*t> wrote in message
>>> news:B6DE17D2.D6B0%t*b*r*u*m*l*e*y at f*u*s*e.n*e*t...
>>>> in article 3AB8A730.4BE13D22 at mail.com, James A. Robertson at
>>>> jarober at mail.com wrote on 3/21/01 8:06 AM:
>>>> 
>>>>> topmind wrote:
>>>>>> 
>>>>> 
>>>>>>> http://www.cincom.com/smalltalk
>>>>>>> 
>>>>>> 
>>>>>> I only see a front page, not a parsing example.
>>>>>> 
>>>>> 
>>>>> Follow the tutorial link - it's right there on that page.
>>>>> 
>>>>> 
>>>>>> Besides, what are you comparing it to so say
>>>>>> it is better than a procedural/relational approach?
>>>>> 
>>>>> It's vastly easier than similar code done in C or basic, for example
>>>> 
>>>> User testimonial here ... last week I had to rip thorugh a pile of IIS
>>> logs
>>>> to get some statistics for some people, and I used the original
> (non-gui)
>>>> tutorial code as a starting point. I was once again impressed with how
>>> much
>>>> easier it was to do several common data processing tasks in Smalltalk
> than
>>>> it was in Assembly language or C or Pascal.
>>>> 
>>> 
>>> Give me specific requirements and I bet I can match it on code size.
>>> 
>>> C, Pascal, and Assembly are hardly what I would use for code-size
>>> contests.
>>> 
>>> Of course, there is more to comparing than code size, but those often
>>> get very subjective or fuzzy.
>> 
>> I rarely worry about code size except how it relates to code clarity. Even
>> back in my assembly oriented youth I tried to keep functions under one
> page
>> on a listing for the sake of clarity.
>> 
> 
> Can you quantify "clarity" or is it a personal thing (subjective)?

My rule of thumb definition back in assembly days was not having to have a
separate note pad to keep track of what was going on. When I saw many
coworkers running out of fingers to mark pages in fanfold listings as they
flipped back and forth, I realized there was an easier solution -- move code
around and rewrite it. Yet another reason why refactoring comes natural to
me, I've never been afraid to rewrite code.

> What makes people confortable with code seems to vary
> greatly per individual.

True.

> Do you have specific examples.

Not anymore. That system is retired. I still use those principles when I am
forced into a procedural environment because I know they work and people
have repeatedly said "thanks for cleaning that code up so I could understand
it" back in those old days. Unlike some, my personal software process has
not developed to include record keeping.

> 
>> I mention C, Pascal, and Assembly because they are my usual tool set, or
>> were until I discovered Smalltalk.
> 
> I used similar stuff at first, but was glad to leave them in the dust when I
> discovered the concepts of Table Oriented Programming.
> 
> I never wanted to see another fricken array again (accept as
> hashes/dictionaries to pass single records or parameter
> lists or attributes. That is their best use, and not data collection
> holding.)

While I'm not familiar with table oriented programming, a long time ago I
did read a book on decision tables in software engineering.

I wonder if your advocacy of TOP is because it's the paradigm that works
best with how your mind works? I know people who can do wonderful things in
RPG that I could never do in the language, and indeed I found that working
in it gave me headaches. These same people, bright and capable, couldn't
work in my prefered environment of assembly language. Some assembly language
programmers I worked with could not transition to 4GLs like Cincom's MANTIS.

We all have different thinking styles, and I think language/environment
preference relates directly to this. I'm a non-linear non-verbal intuitive
thinker and I like Smalltalk. I know one verbal linear type who I couldn't
imagine working in Smalltalk, but she's a very sharp gal.

> 
>> I've done a lot of business programming
>> over the years in procedural environments and used more languages than
> those
>> (even COBOL and RPG) but for the heavy lifting I would always stick with
>> them.
>> 
>> Again, until I discovered Smalltalk :)
>> 
>> I've seen Perl and run away screaming.
> 
> Perl is a strong divisor. You either swear by it, or swear at it. There
> are few in-betweeners.

Scripting tools are wonderful, but why write one with an ugly looking C type
of code? Yuck! :)

> 
>> Python and Eiffel are on the to learn
>> list, but Smalltalk seems to have the power I need for the tasks I want to
>> do these days. Objects are becoming more and more natural as I use them,
> and
>> my code improves contantly.
>> 
>>> 
>>>> I'm a long time procedural type, learning to really think in and use
>>>> Smalltalk has been an interesting experience.
>>>> 
>>>> I've also done a lot of database "fixing" and Smalltalk has usually
> beat
>>> out
>>>> raw SQL for anything but the most trivial tasks.
>>>> 
>>> 
>>> Fixing what? SQL is not good at string parsing.
>> 
>> Usually sweeping changes, or dangling referential integrity problems.
> 
> Huh? Referential integrity is an option, not a forced feature.

Peculiar to our stage in development. We have both database enforced and
some program enforced integrity to manage. I prefer to have the integrity
turned on and would like more of it managed in the database, but in a couple
of cases there's a chicken and egg cycle between tables.

Some of our deletes needed to cascade and did not. I botched some things on
a migration and had to go in and fix them up as well.

Data does not refactor as well as code :)

> If the lamp keeps you awake at night, then turn it off!
> 
>> 
>>>> All that said, I think it's the environment (interactive) that improves
>>> the
>>>> experience as much as the language itself. If I had a reasonable
>>>> interpretive BASIC environment available, I could probably do many of
> the
>>>> things I did in Smalltalk almost as easily ... certainly more easily
> than
>>>> with C.
>>>> 
>>> 
>>> I find interpretive environments potentially more compact code-wise than
>>> static ones.
>> 
>> No argument here.
>> 
>>> 
>>>> 
>>>>> 
>>>>>> 
>>>>>>>> 
>>>>>>>>> [snip]
> 
> 
> 



From aleaxit at yahoo.com  Thu Mar 22 03:10:11 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 22 Mar 2001 09:10:11 +0100
Subject: urllib.urlopen doesn't accept Unicode strings.
References: <998tau$q6i$1@troll.powertech.no> <j4n1aeycsl.fsf@informatik.hu-berlin.de> <99bk6c$hh7$1@panix6.panix.com>
Message-ID: <99cc13014ic@news2.newsguy.com>

"Aahz Maruch" <aahz at panix.com> wrote in message
news:99bk6c$hh7$1 at panix6.panix.com...
> In article <j4n1aeycsl.fsf at informatik.hu-berlin.de>,
> Martin von Loewis  <loewis at informatik.hu-berlin.de> wrote:
> >
> >using types.UnicodeType is much better than type(u"")
>
> Really?  I haven't seen any strong consensus on this issue.

Me neither.  Personally, I generally find type(aconstant)
more directly usable than types.WhateverType, though I do
not think it's that big a deal either way.


Alex





From michael at stroeder.com  Thu Mar  8 14:34:20 2001
From: michael at stroeder.com (Michael =?iso-8859-1?Q?Str=F6der?=)
Date: Thu, 08 Mar 2001 20:34:20 +0100
Subject: Ldap / Python help
References: <mailman.984067391.2391.python-list@python.org>
Message-ID: <3AA7DEBC.555BA9A5@stroeder.com>

Pablo Pernot wrote:
> 
> Where can I find documentation about ldap/python

Hmm, David sent a new PDF to the mailing list, 28 Jan 2001. Hmm, it
does not seem to be on the web page...

> my modify instruction won't work => modify_s is waiting for a string (dn)
> and a list (mylist) formed by tuple (mytuple) and this tuple is formed for 3 strings.
> 
> So I can't understand the error message about integer.
> 
> mytuple=("MOD_REPLACE","sn","test")

Should be mytuple=(ldap.MOD_REPLACE,"sn",["test"]). ldap.MOD_REPLACE
is an integer constant hence the...

> >>> l.modify_s(dn,mylist)
> Traceback (most recent call last):
>   File "<stdin>", line 1, in ?
> TypeError: an integer is required

Ciao, Michael.


From bill-bell at bill-bell.hamilton.on.ca  Thu Mar 22 21:08:27 2001
From: bill-bell at bill-bell.hamilton.on.ca (Bill Bell)
Date: Thu, 22 Mar 2001 21:08:27 -0500
Subject: Digest Number 1946
In-Reply-To: <985302234.525.51095.l7@yahoogroups.com>
Message-ID: <3ABA69CB.10176.7159B28@localhost>

I'm new to Python, have been playing with someone's whois client, 
which contains the following lines:

        s = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
        s.setblocking(1)
        s.connect((self.whoisserver, 43))

The code works fine (as long as the requested whois server is 
available) if the argument to 'setblocking' is 1. However, if I change 
this to 0 I receive the diagnostic,

"The socket operation could not complete without blocking."

I cannot find the code that contains this literal. Would someone 
please suggest what might be going wrong.

Thanks for any help!

Bill Bell, MSc, Software Developer



From parkw at better.net  Mon Mar 12 15:45:06 2001
From: parkw at better.net (William Park)
Date: Mon, 12 Mar 2001 15:45:06 -0500
Subject: continue/else functionality
In-Reply-To: <BD674CDC0A68C642B60038986FBB56DF0138B7@mailsrv.ASPATECH.COM.BR>; from chrisa@ASPATECH.COM.BR on Mon, Mar 12, 2001 at 05:27:39PM -0300
References: <BD674CDC0A68C642B60038986FBB56DF0138B7@mailsrv.ASPATECH.COM.BR>
Message-ID: <20010312154506.A2934@better.net>

On Mon, Mar 12, 2001 at 05:27:39PM -0300, Chris Richard Adams wrote:
> I have a script that reads lines out of a text file. The first line in
> the file serves as a "flag-line" stating whether or not the file should
> be read: 1 yes, 0 no. The rest of the file contains data lines to be
> read. How can I get my script to read on the first line and upon getting
> a 1 go back to the for, get the next line and read the rest from there?
> For example, I tried:
> 
> 
> 	for line in f.readlines():
> 
> 	    datalist = string.split(line)
>           if datalist[0][0] == 1:
>              continue
>           elif datalist[0][0] == 0:
>              break
> 	    
> 	    ## Do a bunch of string manipulation here...

As it is written, you're checking first non-whitespace
character on every line.  If it's 1, you read next line; if it's 0, you
exit the loop; only if it's neither 1 or 0, do you go down further.

Solution:
    Check the first line.  Then, do the loop for 2nd, 3rd, ... lines.
    You may want to check for '1' or '0' (ie. entire string) instead of
    '1...' or '0...' (ie. first character of string).

---William Park, Open Geometry Consulting, Linux/Python, 8 CPUs.



From hungjunglu at yahoo.com  Sun Mar 11 19:24:33 2001
From: hungjunglu at yahoo.com (hungjunglu at yahoo.com)
Date: Mon, 12 Mar 2001 00:24:33 -0000
Subject: Chinese text GIF file generation
Message-ID: <98h501+8kmu@eGroups.com>

--- In python-list at y..., William Park <parkw at b...> wrote:
> They are using blocks of image in 
>     <map> <area...> <area...> ...  </map>.
> So, what is your question?  Do you simply want to duplicate it using
> Python?  Or, do you want to display each character as separate 
image?

Character-by-charater would not be efficient, since this implies 
downloading thousands of files to view just one single webpage. 
Usually people do it by chunks. A chunk is typically a few 
paragraphs. A small article can fit into one single chunk.

> Once these images are stored in /icons/chinese/, you would have to 
write
> stand-alone script to match your input characters to corresponding
> images.  It would spit out <img src=...> for each chinese 
character.  

Usually people would use an application server like Zope to serve 
this kind of requests. LaTeX is good for printed documents, but not 
for webpages utilities.

By the way, I just found that that they revived the IFCSS website: 
http://www.ifcss.org/ (There were some internal political problems 
for sometime, and the website was dead for a long time.) They have 
text-to-bmp, text-to-PostScript converters, but they are kind of old. 
I bet their fonts are missing quite a few Chinese characters, too. 
It's just sad how politics has so deeply ruined this website.

I'll search around a bit more. If I don't find anything good, then 
I'll use Java, or if I have time, I'll write one myself. :) Might 
have to play with font format and with PIL a bit.

regards,

Hung Jung





From thomas at xs4all.net  Tue Mar 13 09:05:42 2001
From: thomas at xs4all.net (Thomas Wouters)
Date: Tue, 13 Mar 2001 15:05:42 +0100
Subject: Einstein's Riddle
In-Reply-To: <98kpaj0p6m@news2.newsguy.com>; from aleaxit@yahoo.com on Tue, Mar 13, 2001 at 10:29:54AM +0100
References: <mailman.984369618.12493.python-list@python.org> <NI9r6.2583$54.2812@www.newsranger.com> <98kpaj0p6m@news2.newsguy.com>
Message-ID: <20010313150542.B404@xs4all.nl>

On Tue, Mar 13, 2001 at 10:29:54AM +0100, Alex Martelli wrote:
> "Gregory Jorgensen" <gregj at pobox.com> wrote in message
> news:NI9r6.2583$54.2812 at www.newsranger.com...

> > So I'm stupid, is that it? I've taken lots of IQ tests and none of them
> > had this question on them; that's probably why I don't score so well!
> > Actually I did figure out that the German owned the fish. Furthermore I
> > was able to determine that the fish is named Eric.

> Actually, I'm pretty sure she was called Wanda.

Only if the german didn't have a fish licence for it.

-- 
Thomas Wouters <thomas at xs4all.net>

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



From glenfant.nospam at bigfoot.com  Fri Mar 30 11:38:58 2001
From: glenfant.nospam at bigfoot.com (Gilles Lenfant)
Date: Fri, 30 Mar 2001 18:38:58 +0200
Subject: HELP urllib POST can't get it to work
References: <9a1m91$fdg$1@pheidippides.axion.bt.co.uk> <9a1puq$gnf$1@pheidippides.axion.bt.co.uk>
Message-ID: <9a2civ$1jr3$1@news4.isdnet.net>

Remove the "?" at the end of the URL (useless with POST)
"mike mcleod" <mike.mcleod at bt.com> a ?crit dans le message news:
9a1puq$gnf$1 at pheidippides.axion.bt.co.uk...
> Also if I try to add the params as as a string:
>
> params = 'address==07&message=test&...
> f = urllib.urlopen("http://www.samemachine.uk/bin/afunc.pl?", params)
>
> I get:
> <HTML><body><p><b>Error occured on server. Couldnt open result file,
>
> Mike McLeod.
>
>




From neilh at scintilla.org  Fri Mar  2 16:09:27 2001
From: neilh at scintilla.org (Neil Hodgson)
Date: Fri, 02 Mar 2001 21:09:27 GMT
Subject: Editing Python source as XML?
References: <mailman.983546704.10789.python-list@python.org> <3a9fe475_1@news.nwlink.com>
Message-ID: <b_Tn6.57540$o85.322150@news-server.bigpond.net.au>

Adam Logghe:
> www.activestate.com has a project called "Komodo" which is written with a
> combination of the Mozilla codebase and Python (my understanding is that
> originally this was also to have a substantial portion written in Perl but
> the developers could code so much faster in Python that Perl ended up as a
> minor part).

    It'd be excellent to hear from ActiveState about their use of Python
versus Perl here. Just a company statement that "Perl sucks" would do fine.
;)

   Neil





From bedge at troikanetworks.com  Thu Mar  1 20:10:51 2001
From: bedge at troikanetworks.com (Bruce Edge)
Date: Thu, 01 Mar 2001 17:10:51 -0800
Subject: define a new func on the fly?
References: <3A9EE927.8744154E@troikanetworks.com> <o6Cn6.846$45.1115@www.newsranger.com>
Message-ID: <3A9EF31B.4105F05C@troikanetworks.com>

Gregory Jorgensen wrote:
> 
> eval() evaluates an expression. exec() executes a statement. Try this:

Yup, that was it! thanks!

Now, how can I get it to create this function in the global namespace?
I'm doing this inside of a class method, so I have no access to the new
func after I leave the class method.

>>> exec( "def xxx():  print \"xxxxxxxx\"" )
>>> xxx()
xxxxxxxx

Woo Hoo! works!

now:

>>> class xx:
...   def __init__(self):
...     exec( "def yyy():  print \"yyyyyyyyy\"" )
... 
>>> x=xx()
>>>     
>>> x.__dict__
{}

Where is yyy defined?

Ideally, I'd like to be able to create it in the global namespace.


-Bruce.

> 
> >>> exec("def f():\n\treturn 'hello'\t")
> >>> f()
> 'hello'
> 
> >>> eval("f()")
> 'hello'
> 
> In article <3A9EE927.8744154E at troikanetworks.com>, Bruce Edge says...
> >
> >I'm trying to define a new function on the at runtime,
> >
> >this fails:
> >
> >>>> eval( "def global_func_name():\n\tpass" )
> >Traceback (most recent call last):
> >  File "<stdin>", line 1, in ?
> >  File "<string>", line 1
> >    def global_func_name():
> >      ^
> >SyntaxError: invalid syntax
> >
> >any idea how I can do this?
> >
> >
> >
> >Thanks, Bruce.
> 
> Greg Jorgensen
> Deschooling Society
> Portland, Oregon, USA
> gregj at pobox.com


From TCHUR at doh.health.nsw.gov.au  Tue Mar 27 03:15:29 2001
From: TCHUR at doh.health.nsw.gov.au (Tim CHURCHES)
Date: Tue, 27 Mar 2001 18:15:29 +1000
Subject: Pickling functions
Message-ID: <sac0d8c7.078@dohprimo.ccf.health.nsw.gov.au>

How does one go about pickling and unpickling a function object? All my attempts so far
result in a reference to the function object being pickled but not the function itself.
For example:

################################################
def maketransform(name,code=None,path="/somewhere/"):
	import zlib, cPickle, new	filename =   path + name + "_transform.gsz"	f = open(filename, 'w+')	func = new.function(compile(code,'blah','exec'),{},name,(v,))	print func(1)	f.write(zlib.compress(cPickle.dumps(func, 1)))	f.close()
maketransform("sexlabel",code="""def sexlabel(v):	ldict = {1:"Male",2:"Female"}	try:		return ldict[v]	except:		return v""")#########################################

and then in a separate Python session

#########################################
def loadtransform(name,path="/somewhere/"):	import zlib, cPickle	filename =   path + name + "_transform.gsz"	f = open(filename, 'r+')	func = cPickle.loads(zlib.decompress(f.read()))	f.close()	return func		sexlabel = loadtransform("sexlabel")print sexlabel(1)##########################################

does not work - Python reports a NameError because function sexlabel can't be found in __main__
What am I doing wrong? Is there a simpler way of doing this?

Tim C





From tim.one at home.com  Sun Mar 11 12:35:42 2001
From: tim.one at home.com (Tim Peters)
Date: Sun, 11 Mar 2001 12:35:42 -0500
Subject: "!=" is bad form. Re: sorry....never mind
In-Reply-To: <11032001.1@wanderer.local.jae.ddns.org>
Message-ID: <LNBBLJKPBEHFEDALKOLCMENPJEAA.tim.one@home.com>

[J?rgen A. Erhard, reacting to the deprecation of <> in favor of !=]
> <curious mode="slightly confused">
> So when can we expect `||' and `&&' to replace `or' and `and'?
> Shouldn't we `Beware of the slippery slope'?
> </curious>

The deprecation of <> occurred about a decade ago, before Python's first
public release.  It surprises people today only because (a) contrary to some
opinion, Guido is very keen on preserving backward compatibility so never
took <> away; and, (b) people generally don't read the language Reference
Manual, where this has been documented all along.

How slippery is this slope?  Well, one other class of syntax change did
tumble down the same path:  the shift (<< >>) and bitwise logical (& | ^ ~)
operators were added after <> was deprecated.

wondering-whether-micro-changes-per-decade-is-a-useful-measure-of-
    slipperiness-ly y'rs  - tim




From runyaga at thisbox.com  Thu Mar 29 23:57:37 2001
From: runyaga at thisbox.com (alan runyan)
Date: Fri, 30 Mar 2001 04:57:37 GMT
Subject: dtml-with and dtml-sqltest
References: <9a0b2s$16to$1@node17.cwnet.frontiernet.net>
Message-ID: <5nUw6.8000$wx.1368177@typhoon.austin.rr.com>

Lee,

the zope mailing list is more appropriate for ZOPE inquiries.  there is a
larger community focused solely on that domain.  please direct email to
zope at zope.org.  I dont really know what sqltest does, I have never used it.

you can try this:

sqlMethod -
arguement - aaa=1
SELECT * FROM table WHERE id=<dtml-sqlvar aaa type="int">

dtmlMethod:

<dtml-let aaalist="[1,2,3]">
<dtml-in aaalist>
 <dtml-call "REQUEST.set('aaa', _['sequence-item'])">
 <dtml-in getPhById>
  <dtml-var PH>
 </dtml-in>
</dtml-in>
</dtml-let>


the dtmlMethod should probably be in a PythonMethod that looks like:

aaalist=[1,2,3]
stuff = []

for aaa in aaalist:
    for sql in container.getPhById({'aaa': aaa}): # _TRICKY_ ZSQLMethods *DO
NOT* use DTML namespace
       stuff.append(sql[0])                       # position in sql row
returned

return stuff


Lessons learned: PythonMethods are cleaner (on a whole), ZSQL doesnt get a
DTML namespace, Dieter Maurer has most of the answers on the ZOPE mailing
list, just search through it @ egroups.

~runyaga





From shaleh at valinux.com  Fri Mar  2 11:54:06 2001
From: shaleh at valinux.com (Sean 'Shaleh' Perry)
Date: Fri, 02 Mar 2001 08:54:06 -0800 (PST)
Subject: gah! I hate the new string syntax
In-Reply-To: <97ohh2$polt0$1@ID-77331.news.dfncis.de>
Message-ID: <XFMail.20010302085406.shaleh@valinux.com>

> 3. I *like* string methods, even .join().  I am aware that this puts me in a
> very small minority on comp.lang.python.  But I would maintain that
> ";".join(...) only seems odd to you because you see it as the result of a
> strained migration from string.join(..., ";"), instead of seeing it with
> fresh eyes.  However, I may add a footnote explaining the old style
> string.join, if only because new programmers will eventually come across old
> code that uses it.
> 

if I were learning python today, I would still hate code like the above. 
Turning a string literal (";") suddenly into a string object and then using its
methods is not an easy read.  While trying to scan that line, my eye wants to
read the list comprehension (which is a nice feature, if compact).  Then I see
the join function and it takes a moment to realize it belongs to the semicolon.

The other problem I have with this is I know too many languages.  I am forced
to deal with quite a bit of perl and I actually read your book immediately
after a week's worth of working on some really evil perl code that was dumped
in my lap.  SO what do I see?

";" . join() # in perl, this means concat ";" with the result of join()

Yes, I know this is an issue you learn to deal with.  However, the nice, pretty
string module at least makes python look like python.

Mark, I like a lot of what I read in your book.  The apihelper is a nifty idea
(if only it worked on the re module )-:).  However, the coding style and
decisions seem counter to the python way.  But maybe that is just me and my
hyper sensitivity to hard to read code after way too many lines of perl.



From gahanr at gofree.indigo.ie  Tue Mar  6 15:25:23 2001
From: gahanr at gofree.indigo.ie (Robert Gahan)
Date: Tue, 6 Mar 2001 20:25:23 -0000
Subject: Remote script execution with Python ...
Message-ID: <7Ybp6.7729$r4.7336@news.indigo.ie>

Hi there,

I have installed python on a Windows PC and I am looking to
remotely execute a script on a UNIX box from a python program.
I need to be able to pipe a "y" to stdin also. Is there a windows
equivalent of the unix "rexec" that could help me ?. What is the
most appropriate way of doing this ?.

Thanks again,
Rob.




From fuess at att.net  Mon Mar 26 08:01:29 2001
From: fuess at att.net (David Fuess)
Date: Mon, 26 Mar 2001 13:01:29 GMT
Subject: Help how do I read `word' variables from a binary file?
References: <3abf3b0b.1487830@news.is.co.za>
Message-ID: <aafubtkrem3o3fu5j5sbf9vnh2e3dnrea1@4ax.com>

Read the data in recordsize groups like you're doing now, then use
struct.py do decompose the records into Python variables.

Dave

On Mon, 26 Mar 2001 12:57:31 GMT, nomad***@***freemail.absa.co.za
(nomad) wrote:

>Howzit,
>
>I'm having trouble with reading a binary file properly in Python 2.0
>on win32.  I'm stating explicitly that I'm using Win32 as I know that
>you have to state when opening the file that it is for `br' as in
>binary reading (I'm think this is due to intels little-endian
>approach, but I could be wrong).
>
>I can open the file without problems, and I can read from the file
>without a hitch, _BUT_ the trouble comes in that some of the data
>records in the file have fields that would be a double word or even a
>normal word in C++, so in effect the data may be 0x80002310, but
>python reads the data directly from the file, and as such is reading
>hex bytes as (in this example) `10 23 00 80' etc.  I'm sure that there
>must be a simple way to read in data like this, but I can't figure it
>out.
>
>Any help would be greatfully recieved.
>
>TIA.



From chris.arndt at gmx.net  Wed Mar  7 12:14:43 2001
From: chris.arndt at gmx.net (Christopher Arndt)
Date: Wed, 07 Mar 2001 18:14:43 +0100
Subject: Easy way to get the X selection (as text)?
Message-ID: <3AA66C83.29412EE0@gmx.net>

I skimmed the docs of most X GUI modules for python on how to retrieve
the X selection, but this seems to be a non-trivial task. Is there an
easy way to do this, without having to use tkinter (don't want a useless
window) or is there a ready-to-use module for this?

I want to write a little utility for openening urls I have selected in
some window.

Thanx in advance. Please email me. I'm no list member, nor do I read the
news regulary

Cheers, Chris



From grante at visi.com  Tue Mar 13 12:40:51 2001
From: grante at visi.com (Grant Edwards)
Date: Tue, 13 Mar 2001 17:40:51 GMT
Subject: learning C
References: <mailman.983972829.23481.python-list@python.org> <983975751.18833@newsmaster-04.atnet.at> <slrn9adklr.bv9.scarblac@pino.selwerd.nl> <984047775.592522@newsmaster-04.atnet.at> <9896hk$ig8$1@verence.demon.co.uk> <984134612.781779@newsmaster-04.atnet.at> <3AACE22D.3A9EEB12@bt.com> <j6kr6.68748$ML1.3282888@e420r-atl2.usenetserver.com>
Message-ID: <DYsr6.1213$Tg.160308@ruti.visi.com>

In article <j6kr6.68748$ML1.3282888 at e420r-atl2.usenetserver.com>, Steve Holden wrote:

>> > This is nothing more difficult than in C.
>>
>> Sorry but it is. There's just so much less lurking under
>> the hood in C.
>>
>But by this argument, assembly code is even better!  

Yes, but not much.  C is little more than a portable assembly
language.  If we're talking about embedded systems, a competent
programmer pretty much knows what assembly language the C
compiler is going to generate for each line of code he writes.
Example: for a particular processor/compiler combination he
should know whether it's faster to increment a pointer through
an array or to increment an index through an array.

>It's always a question of selecting language features because
>they are appropriate to your problem domain.

If there were some way to disable C++ features, then yes, that
might be feasible.

-- 
Grant Edwards                   grante             Yow!  Thank god!!... It's
                                  at               HENNY YOUNGMAN!!
                               visi.com            


From jeremy at alum.mit.edu  Thu Mar  1 11:37:53 2001
From: jeremy at alum.mit.edu (Jeremy Hylton)
Date: Thu, 1 Mar 2001 11:37:53 -0500 (EST)
Subject: d = {}; d[0:1] = 1; d[0:1] = 2; print d[:]
In-Reply-To: <4.3.1.2.20010228223714.00d29430@exchange.starmedia.net>
References: <4.3.1.2.20010228223714.00d29430@exchange.starmedia.net>
Message-ID: <15006.31457.377477.65547@w221.z064000254.bwi-md.dsl.cnc.net>

>>>>> "RT" == Robin Thomas <robin.thomas at starmedia.net> writes:

  RT> Using Python 2.0 on Win32. Am I the only person to be depressed
  RT> by the following behavior now that __getitem__ does the work of
  RT> __getslice__?

You may the only person to have tried it :-).

  RT> Python 2.0 (#8, Oct 16 2000, 17:27:58) [MSC 32 bit (Intel)] on win32
  >>> d = {}
  >>> d[0:1] = 1
  >>> d
  {slice(0, 1, None): 1}

I think this should raise a TypeError (as you suggested later).

>>> del d[0:1]
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
TypeError: object doesn't support slice deletion

Jeremy



From fredrik at pythonware.com  Tue Mar 20 17:26:28 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Tue, 20 Mar 2001 22:26:28 GMT
Subject: New PEP: The directive statement
References: <mailman.985117312.5806.python-list@python.org> <j4u24oi4e8.fsf@informatik.hu-berlin.de> <yHPt6.2553$4N4.334967@newsc.telia.net> <998hv2$5ub$1@panix3.panix.com>
Message-ID: <oOQt6.2576$4N4.342378@newsc.telia.net>

Aahz Maruch wrote:
> >except that "import" already is a declaration statement, of course.
>
> Depends how you define "already".  There is no current released version
> of Python (alphas and betas do not count as "released versions") in
> which "import" is a declaration.

import declares things, in all existing Python versions (hint: it's
an assignment statement)

the future hack overloads the meaning of a specific declaration,
inspired by similar solutions in other languages.

(fwiw, the original proposal was "import __nested_scopes__", and
the BDFL didn't seem to find that very unpythonic, given that it's
only there to allow people to do prerelease testing.  pre-2.2, that
is)

> So you're saying that "from __future__" is only to be used for this one
> issue of nested scopes?  I don't think so....

well, I do -- with the possible exception of cleaning up the numerics
model (also an age-old issue).

Cheers /F




From urner at alumni.princeton.edu  Fri Mar  2 21:07:16 2001
From: urner at alumni.princeton.edu (Kirby Urner)
Date: Fri, 02 Mar 2001 18:07:16 -0800
Subject: Hitting two targets:  OO + group theory (pedagogy)
Message-ID: <u2k0atkgkiatdp9ukj3mpd1ej0b648ldsm@4ax.com>

============== OREGON CURRICULUM NETWORK =====================

RE:  HITTING TWO TARGETS:  OO + MATH TOPIC

OO + POLYHEDRA

In earlier writings, I've discussed how object-oriented (OO) 
syntax might be melded with math notations with regard to 
polyhedra.[see links]  A Polyhedron superclass, subclassed 
as Tetrahedron, Octahedron etc., would demonstrate the concepts 
of class hierarchy and instantiated selves, or objects.  

Many intro-to-OO books use already flatlander shapes 
(superclass: shape; subclasses: triangle, circle, square...). 
However the Polyhedron case is mathematically more interesting 
and takes better advantage of the computer's graphical display 
capabilities.  Rotation, translation and scaling may be 
implemented using matrices, or we might swap in quaternions
to accomplish rotation, or we might develop use some other 
Clifford Algebra ala Hestenes et al.

OO + GROUP THEORY 

More recently, I've been looking to group theory as another 
paradigm case for hitting two targets with one stone:  OO 
concepts and number-theoretic math concepts. Each set of 
concepts helps to reinforce the other, thereby bootstrapping 
students into a synergetic mindset which sets the stage for 
elaboration both in programming and in abstract mathematics.

Whereas this might at first sound like a college level 
approach, my contention is that newcomers to programming will 
leap-frog those who cut their teeth in the hay day of 
procedural programming -- which doesn't mean the procedural 
approach has gone out of style, only that it's no longer 
relevant to consider the object-oriented as "advanced", with 
mastery of earlier forms a prerequisite.  Functional programming
methods will also make their debut earlier in a student's 
career.

And group theory has always been accessible at a variety of 
levels.  It makes sense to start phasing in some of these 
concepts early, so that later plantings of these ideas find 
fertile ground.

PYTHON AS INTRO LANGUAGE

The computer language of choice in these explorations is 
Python.  We want a language which is interactive at the command 
line, like Logo, such that programmed methods may be accessed 
and combined on the fly, with the screen itself serving as a 
principal site for i/o.  Whereas we'll want the option to read 
in text files for processing, there should be no demand on 
students to compile standalone executables which must either 
support their own GUI for interactivity, or rely on command 
line arguments pointing to files containing the raw data of 
interest.  

ISETL, DrScheme, APL and others share these positive features 
with Python and there's no stipulation here that only Python is 
the appropriate vehicle for these kinds of activities (given 
the OO focus, some of these others do not lend themselves to 
hitting our double-target) -- only that this author has found 
the Python environment highly congenial for this kind of 
undertaking.

Let's define our sandbox to be groups of relative primes less
than a particular prime modulus n.  These are groups under 
multiplication, where a*b is defined as (a*b) mod n.  In 
Python, the mod operator is %, such that a%b returns the 
remainder a (mod b).  

SYNTAX BASICS 

The primitive built-in function pow() takes a mininum of two 
arguments a,b such that pow(a,b) = a^b (or a**b in Python, 
as exponentiation is signified with a double-asterisk).  The 
optional 3rd argument provides a modulus, such that pow(a,b,n) 
= (a**b) mod n.  This 3-argument form of pow() is internally 
optimized such that the modulus is used to reduce the product 
as powering proceeds -- far more efficient than applying the 
syntax (a**b)%n, which can take indefinitely longer to complete 
for large numbers.

Where object-oriented syntax shines in this application is 
when we go to define multiplication.  We're able to override
the meaning of * (multiplication operator) in such a way 
that operator.mul will likewise behave consistently, i.e. will 
respect the definition we provide for multiplication.  This 
is done be defining __mul__ in the class definition.  Similar 
overriding is done with respect to ** and pow(), by defining
__pow__, and for relational operators <, >, and ==, by 
defining __lt__, __gt__, and __eq__ respectively.

I won't go into a full-blown dissection of source code here,
but rather will conclude with an interactive command line 
session which gives the flavor of the final result.  In class,
or over the web (if a distance learning opportunity), students
would be boning up on Python and some of the basics of group
theory via supplementary materials.  Some of the preliminary
prototypical writing is available in draft form on math-learn,
an egroup focussed on K-12 math education, with emphasis on the
middle school years. [see links below]

Further forays into group theory + Python may be found in the
'For further reading' section of my intro page on cryptography.

COMMAND LINE INTERFACE (CLI)

Here's from the Python command line:

 >>> from groups import *
 >>> residues17 = factory(17)  # make objects using coprimes of 17
 >>> residues17                # display as list of integers
 [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16]
 >>> table(residues17)         # generate a multiplication table
 
       1  2  3  4  5  6  7  8  9 10 11 12 13 14 15 16
    ----------------------------------------------------
  1|   1  2  3  4  5  6  7  8  9 10 11 12 13 14 15 16
  2|   2  4  6  8 10 12 14 16  1  3  5  7  9 11 13 15
  3|   3  6  9 12 15  1  4  7 10 13 16  2  5  8 11 14
  4|   4  8 12 16  3  7 11 15  2  6 10 14  1  5  9 13
  5|   5 10 15  3  8 13  1  6 11 16  4  9 14  2  7 12
  6|   6 12  1  7 13  2  8 14  3  9 15  4 10 16  5 11
  7|   7 14  4 11  1  8 15  5 12  2  9 16  6 13  3 10
  8|   8 16  7 15  6 14  5 13  4 12  3 11  2 10  1  9
  9|   9  1 10  2 11  3 12  4 13  5 14  6 15  7 16  8
 10|  10  3 13  6 16  9  2 12  5 15  8  1 11  4 14  7
 11|  11  5 16 10  4 15  9  3 14  8  2 13  7  1 12  6
 12|  12  7  2 14  9  4 16 11  6  1 13  8  3 15 10  5
 13|  13  9  5  1 14 10  6  2 15 11  7  3 16 12  8  4
 14|  14 11  8  5  2 16 13 10  7  4  1 15 12  9  6  3
 15|  15 13 11  9  7  5  3  1 16 14 12 10  8  6  4  2
 16|  16 15 14 13 12 11 10  9  8  7  6  5  4  3  2  1
 >>> a = residues17[7]        # select a = specific object
 >>> a 
 8
 >>> a.inv()                  # compute a's inverse
 15
 >>> a*a.inv()                # a * a^-1 = 1
 1
 >>> powers(residues17)       # compute g to successive powers
   1 = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]
   2 = [1, 2, 4, 8, 16, 15, 13, 9, 1, 2, 4, 8, 16, 15, 13, 9]
   3 = [1, 3, 9, 10, 13, 5, 15, 11, 16, 14, 8, 7, 4, 12, 2, 6]
   4 = [1, 4, 16, 13, 1, 4, 16, 13, 1, 4, 16, 13, 1, 4, 16, 13]
   5 = [1, 5, 8, 6, 13, 14, 2, 10, 16, 12, 9, 11, 4, 3, 15, 7]
   6 = [1, 6, 2, 12, 4, 7, 8, 14, 16, 11, 15, 5, 13, 10, 9, 3]
   7 = [1, 7, 15, 3, 4, 11, 9, 12, 16, 10, 2, 14, 13, 6, 8, 5]
   8 = [1, 8, 13, 2, 16, 9, 4, 15, 1, 8, 13, 2, 16, 9, 4, 15]
   9 = [1, 9, 13, 15, 16, 8, 4, 2, 1, 9, 13, 15, 16, 8, 4, 2]
  10 = [1, 10, 15, 14, 4, 6, 9, 5, 16, 7, 2, 3, 13, 11, 8, 12]
  11 = [1, 11, 2, 5, 4, 10, 8, 3, 16, 6, 15, 12, 13, 7, 9, 14]
  12 = [1, 12, 8, 11, 13, 3, 2, 7, 16, 5, 9, 6, 4, 14, 15, 10]
  13 = [1, 13, 16, 4, 1, 13, 16, 4, 1, 13, 16, 4, 1, 13, 16, 4]
  14 = [1, 14, 9, 7, 13, 12, 15, 6, 16, 3, 8, 10, 4, 5, 2, 11]
  15 = [1, 15, 4, 9, 16, 2, 13, 8, 1, 15, 4, 9, 16, 2, 13, 8]
  16 = [1, 16, 1, 16, 1, 16, 1, 16, 1, 16, 1, 16, 1, 16, 1, 16]
 >>> b = residues17[12]        # pick a g the period < totient
 >>> b
 13
 >>> sg = subgroup(b)          # generate the subgroup using b
 >>> sg
 [1, 13, 16, 4]
 >>> table(sg)                 # generate the multiplication table
 
       1  4 13 16
    ----------------
  1|   1  4 13 16
  4|   4 16  1 13
 13|  13  1 16  4
 16|  16 13  4  1


The source code I'm using is here:
http://www.inetarena.com/~pdx4d/ocn/python/groups_draft1.html

Links:

OO + Polyhedron:  http://www.inetarena.com/~pdx4d/ocn/oop.html
also:             http://www.inetarena.com/~pdx4d/ocn/trends2000.html

See bottom of http://www.inetarena.com/~pdx4d/ocn/overcome.html
for links to the math-learn postings (these were rough drafts --
a better version for the web is in the works)

Also see:  http://www.inetarena.com/~pdx4d/ocn/clubhouse.html

============== OREGON CURRICULUM NETWORK =====================

   By Kirby Urner, 4D Solutions, Portland, Oregon, USA  




From sholden at holdenweb.com  Tue Mar 13 02:48:30 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Tue, 13 Mar 2001 02:48:30 -0500
Subject: Python In A Nutshell - Numpy essential
References: <3AAE7E0D@MailAndNews.com> <slrn9aolph.ccf.jimhill@meldrick.swcp.com> <3AACE20C.AC8D2EA2@northwestern.edu>
Message-ID: <Ikkr6.68807$ML1.3285992@e420r-atl2.usenetserver.com>

"Louis Luangkesorn" <lluang at northwestern.edu> wrote in message
news:3AACE20C.AC8D2EA2 at northwestern.edu...
> I have to second that Numeric Python belongs in the ESSENTIAL column.  For
me,
> the fact that Python can replace Matlab was the reason I started using
Python.
> And when my university discontinued its campus-wide license, and every
> installation of Matlab must be paid for, that it is an open-source
replacement
> for Matlab makes it uniquely valuable (Octave was too slow).
>
Well, just for balance, I still haven't downloaded NumPy, and don't ever
expect to.

one-man's-essential-is-another-man's-redundant-ly y'rs  - steve





From gregj at pobox.com  Thu Mar  8 02:02:53 2001
From: gregj at pobox.com (Gregory Jorgensen)
Date: Thu, 08 Mar 2001 07:02:53 GMT
Subject: grabbing return codes from os.system() call
References: <%wFp6.9481$dL4.126404@vixen.cso.uiuc.edu>
Message-ID: <x8Gp6.899$54.943@www.newsranger.com>

os.system returns the exit status. The low-order byte of the exit status is the
signal number that killed the process, and the high-order byte is the exit
status (if the signal number is 0). This only works under Unix; on Windows the
exit status is always 0.

768 = 0x0300 -- there's your 3 return value.

This is documented in the Python Library Docs, at:

http://www.python.org/doc/current/lib/os-process.html

You can separate the bytes in many ways. One simple technique:

x = os.system('./returncode')
signal = x & 0xFF
exitcode = (x >> 8) & 0xFF


In article <%wFp6.9481$dL4.126404 at vixen.cso.uiuc.edu>, Damian Menscher says...
>
>I'm new to Python (as in, my experience is essentially the tutorial),
>but I've already come up with an interesting question:
>
>How do I get the return code from an os.system call?  I would have
>expected I could do something like
>
>---returncode---
>#/bin/csh
>echo do stuff
>exit 3
>
>and then in my python program I could do
>
>print os.system('./returncode')

Greg Jorgensen
Deschooling Society
Portland, Oregon, USA
gregj at pobox.com


From jafo at tummy.com  Thu Mar 15 01:29:46 2001
From: jafo at tummy.com (Sean Reifschneider)
Date: Wed, 14 Mar 2001 23:29:46 -0700
Subject: RPM Module?
In-Reply-To: <3aab687b$1@news.teuto.net>; from ocaner@gmx.de on Sun, Mar 11, 2001 at 12:55:45PM +0100
References: <3aab687b$1@news.teuto.net>
Message-ID: <20010314232946.N30951@tummy.com>

On Sun, Mar 11, 2001 at 12:55:45PM +0100, ?zg?r Caner wrote:
>Does anybody know where to find a Documentation about the RPM Module
>for Python?

Unfortunately, I've mostly figured out what I use it for by just groping
around in the module.  Looking at the symbolx in the module, etc...
I seem to recall Donnie Barnes replying to somone on usenet at one point
about a program that used most of the functionality of it, but I can't
seem to find a reference to it.  I may be thinking of the "snack"
module instead though...

Sean
-- 
 If I were an Indian, my name would be "Running With No Swap".
                 -- Sean Reifschneider, 1999
Sean Reifschneider, Inimitably Superfluous <jafo at tummy.com>
tummy.com - Linux Consulting since 1995. Qmail, KRUD, Firewalls, Python



From phd at phd.pp.ru  Mon Mar 26 03:31:27 2001
From: phd at phd.pp.ru (Oleg Broytmann)
Date: Mon, 26 Mar 2001 12:31:27 +0400 (MSD)
Subject: Python ready to replace perl
In-Reply-To: <86snk1qafw.fsf_-_@jfh00.fernuni-hagen.de>
Message-ID: <Pine.LNX.4.30.0103261228530.15258-100000@fep132.fep.ru>

On 26 Mar 2001, Fritz Heinrichmeyer wrote:
> So i consider using python in the future. For this python has to have
>
> 1. sql-interface like dbi, at least for postgres (i think this exists).

   Exists. There are few different implementations. There is PyGreSQL,
PoPy, psycopg.

> 2. interface to openldap

   Exists.

> 3. mod_perl beeing able to program authentification procedures using
>    1. and 2. for apache and beeing able to program arbitrary URL-redirect
>    procedures.

   mod_python or mod_snake, perhaps.

> 4. a template system beeing able to manage multiple language versions
>    (to replace the perlish WML-System)?.

   DTML can be used outside of Zope. I use it in pure Python CGIs pretty
happily.

Oleg.
----
     Oleg Broytmann            http://phd.pp.ru/            phd at phd.pp.ru
           Programmers don't die, they just GOSUB without RETURN.




From aleaxit at yahoo.com  Sun Mar 11 03:08:41 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sun, 11 Mar 2001 09:08:41 +0100
Subject: gah! I hate the new string syntax
References: <mailman.983485923.1136.python-list@python.org> <97p7iq$kk2$1@nntp6.u.washington.edu> <97soak$qh9$1@nntp.Stanford.EDU> <97sudi01i26@news1.newsguy.com> <98elrk$li8$1@nntp.Stanford.EDU>
Message-ID: <98fc070thh@news1.newsguy.com>

"Amit Patel" <amitp at Xenon.Stanford.EDU> wrote in message
news:98elrk$li8$1 at nntp.Stanford.EDU...
    [snip]
> What I mean to say is that I cannot write new methods on strings.  So

Perfectly true, just as you cannot on dictionaries, lists, tuples,
file objects, and so on.  Only classes and extension-classes
are extensible (by inheritance and overriding), not types.

> has control over the string class.  Since I don't control the string
> class, I have to use a different syntax than someone who does have
> control over the string class gets to use.  That seems okay if the
> function is some essential part of that class.  But it seems "wrong"
> if the function is there merely for convenience.

string is not a class -- it's a type.  To many, the split seems wrong
_all_ of the time.  I may be one of the few that sees some pluses
in the distinction (!), but it does complicate things, and is a strong
candidate for "aspect of Python fewest Pythonistas _like_":-).


Alex





From a2619725 at uni-koeln.de  Sun Mar 18 12:17:04 2001
From: a2619725 at uni-koeln.de (Oliver Hofmann)
Date: 18 Mar 2001 17:17:04 GMT
Subject: Efficiency of UserString
Message-ID: <992qig$96$1@news.rrz.Uni-Koeln.DE>

This buffer is for notes you don't want to save, and for Lisp evaluation.
If you want to create a file, visit that file with C-x C-f,
then enter the text in that file's own buffer.

'lo everyone!


Got a question regarding UserString's efficiency. I am parsing text
and would like to store a word's position (sentence, absolute position)
somewhere; this would allow turning the document into a list of
words, then removing stopwords or stemming the word and still having
the information about it's original position in the document.

Figured instead of creating several lists to store and update that info
I could put it into the string itself by using UserString; however
the Python library reference states:

"It should be noted that these classes are highly inefficient compared 
to real string or Unicode objects; this is especially the case for
MutableString."

I've tried two basic operations below but couldn't spot much of a 
difference. Could someone please point me at an error I've made or
tell me which operations are slower with UserString?

----

import UserString
import profile, pstats

text = """ The compound ppp(A2p)3A3[32P]pCp is a commercially available
radioactive analogue of the 2,5 oligoadenylate series ppp(A2p)nA, n
greater than or equal to 2, commonly referred to as 2-5A. It is used
as a probe for measuring concentrations in competition radiobinding and
radioimmune assays. We have found that incubation of the probe with
extracts from HeLa, CV1, or neuroblastoma cells results in its covalent
attachment to two size classes of RNA: the first includes a major species
with a molecular weight of approximately 350,000, the second is much
smaller (40 +/- 5 nucleotides in length) and could represent tRNA
half-molecules. Ligation is to the 3 end of the probe molecule with
formation of a 3,5-phosphodiester bond. Thus, probe ligation provides a
sensitive and convenient assay for the detection not only of RNA
ligase(s) but also of ligatable RNAs (such as the putative tRNA
half-molecules) in mammalian cell extracts. """

text = ' '.join(text.split())
text2 = UserString.UserString(text)


def fun1():
    global text
    
    for a in range(0, 1000):
        text = ' '.join(text.split())
        
def fun2():
    global text
    
    for a in range(0, 1000):
        text.find('neuro')
        
def fun3():
    global text2
    
    for a in range(0, 1000):
        text2 = ' '.join(text2.split())
        
def fun4():
    global text2
    
    for a in range(0, 1000):
        text2.find('neuro')

def main():
    for a in range(0, 100):
        fun1()
        fun2()
        fun3()
        fun4()
    
    
profile.run('main()', 'profile.tmp')
p = pstats.Stats('profile.tmp')
p.sort_stats('cumulative').print_stats(10)

----

Here are the stats:


         404 function calls in 80.010 CPU seconds

   Ordered by: cumulative time

   ncalls  tottime  percall  cumtime  percall filename:lineno(function)
        1    0.000    0.000   80.010   80.010 <string>:1(?)
        1    0.000    0.000   80.010   80.010 profile:0(main())
        1    0.000    0.000   80.010   80.010 test3.py:48(main)
      100   39.050    0.390   39.050    0.390 test3.py:24(fun1)
      100   39.040    0.390   39.040    0.390 test3.py:36(fun3)
      100    0.960    0.010    0.960    0.010 test3.py:42(fun4)
      100    0.960    0.010    0.960    0.010 test3.py:30(fun2)


Many thanks once again!


    Oliver

--
Oliver Hofmann  -  University of Cologne  -  Department of Biochemistry
   o.hofmann at smail.uni-koeln.de - setar at gmx.de - connla at thewell.com

   If you care, you just get disappointed all the time. If you don't care
nothing matters so you are never upset.   -- Calvin



From phrxy at csv.warwick.ac.uk  Thu Mar 29 13:01:44 2001
From: phrxy at csv.warwick.ac.uk (John J. Lee)
Date: Thu, 29 Mar 2001 19:01:44 +0100
Subject: Gaussian line profile using numeric python
In-Reply-To: <m3puf0re81.fsf@chinon.cnrs-orleans.fr>
References: <3ac31698@zfree.co.nz> <m3puf0re81.fsf@chinon.cnrs-orleans.fr>
Message-ID: <Pine.SOL.4.30.0103291850410.13267-100000@primrose.csv.warwick.ac.uk>

On 29 Mar 2001, Konrad Hinsen wrote:

> "Geoff Low" <geoff at ou043085.otago.ac.nz> writes:
>
> > I have an extensive set of xy data that I want to convolute with a gaussian
> > profile.  Below is my algorithm,  which doesn't work.  Can anyone see where
> > I'm being dumb?

What goes wrong?  Couldn't see your original message.

> To be honest, I am too lazy to look through this. However,

Me too.  :)  I do have some working routines that will do gaussian
convolution, using the standard Numeric FFT module -- very simple.  Or at
least would be if I didn't have a lot of stuff for using the FFTW library
in there, which doesn't actually work (the docs for the rfftw part of the
FFTW python wrapper seem to be wrong:  anybody got this working?).

> 1) The function Numeric.convolve is certainly a lot more efficient
>    than your code, and should make it simpler.

I seem to recall that it doesn't cover all the possibilities, though.  I
don't remember exactly why, but since I wrote my own functions to do this
there must have been a reason!

> 2) If you work with large data sets, you ought to use FFTs to
>    compute the convolution, that's O(N*log(N)) in the size of the
>    data set, instead of O(N**2) for the straightforward method.
>    Any good book on FFTs should explain how this works in detail.

but you probably don't need to know in detail, it boils down to

inv_FFT(FFT(data)*FFT(kernel))

But if you have small data sets it might be slower, of course.

> > P.S.  I'm no math wizard but this is my interpretation of the gaussian lineshape
> > function
> > G(v) = 1/(sqrt(2*PI)*FWHM)*exp(-(v-vo)^2/(2*FWHM^2))
>
> Assuming that FWHM stands for full width at half minimum, no, it's not
> correct. What you call FWHM is the variance sigma of the Gaussian, the
> FWHM is around 2.35*sigma (quick calculation, no guarantee).

That number is about right, from memory.


John



From stephen_purcell at yahoo.com  Wed Mar 28 04:47:44 2001
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Wed, 28 Mar 2001 11:47:44 +0200
Subject: Help with regular expression backreferences
In-Reply-To: <3AC16D0F.6E33377E@nospammytheworld.com>; from deblNO@nospammytheworld.com on Wed, Mar 28, 2001 at 04:48:08AM +0000
References: <3AC16D0F.6E33377E@nospammytheworld.com>
Message-ID: <20010328114744.F8325@freedom.puma-ag.com>

David Lees wrote:
> I would like to use regular expression back references as keys into a
> dictionary when doing substitution.  So far all I have is:
> 
> s='This xxMYJunkkxx the string that xxYourJunkxx the thing'
> p=re.compile(r'xx(\w+)xx')
> new_s=re.sub(p,'junk',s)
> print new_s

Don't use 're.sub' -- use the 'sub' method of your compiled regex 'p':

>>> import re
>>> s='This xxMYJunkkxx the string that xxYourJunkxx the thing'
>>> p=re.compile(r'xx(\w+)xx')
>>> new_s = p.sub(r'\1', s)
>>> new_s
'This MYJunkk the string that YourJunk the thing'


-Steve


-- 
Steve Purcell, Pythangelist
Get testing at http://pyunit.sourceforge.net/
Any opinions expressed herein are my own and not necessarily those of Yahoo



From gmunsey at adobe.com  Sat Mar 24 14:08:44 2001
From: gmunsey at adobe.com (Grant Munsey)
Date: Sat, 24 Mar 2001 11:08:44 -0800
Subject: [Ann] Python module for Perforce Source Code Management System API/Windows
Message-ID: <0h6v6.24$Vu1.40771@news.pacbell.net>

I have coded up a C level module that lets Python 2.0 callers access the
Perfoce SCM API interface.

http://opensource.adobe.com/project6/project.shtml

One thing to note ... while Perforce costs $$ per
seat you can run a server for two users for free.
You can also run a Perforce server for free if you
are maintaining an open source site.






From gradha at iname.com  Sat Mar 10 17:36:41 2001
From: gradha at iname.com (gradha at iname.com)
Date: Sat, 10 Mar 2001 23:36:41 +0100
Subject: Problems writting a wrapper around linux' ssh command
References: <31rd89.lki.ln@127.0.0.1> <fdwq6.1906$54.1822@www.newsranger.com>
Message-ID: <p9ae89.imr.ln@127.0.0.1>

Gregory Jorgensen <gregj at pobox.com> wrote:
> ssh explicitly opens /dev/tty to read the password, so it's not looking on stdin
> (see the readpass.c file in your ssh sources). In a normal terminal session
> stdin is mapped to /dev/tty, but a child process started with popen2 will have
> stdin mapped to the Popen3.tochild file descriptor. Can you use one of the other
> secure authentication methods ssh supports?

Er... I am new to ssh, so I don't even know what other methods there are...
So far, I want to write a wrapper on top of sitecopy (a program which allows
updating remote web sites through ftp) using ssh and scp to upload files
and make directory modifications. If there's any other way of doing this,
any help is very welcome. The only restriction is that the program I write
from commandline or in background, accepts password only once, and logs
everything to files, so I thought wrapping around ssh and scp would be the
easiest way.

-- 

 Grzegorz Adam Hankiewicz   gradha at iname.com - http://gradha.infierno.org
 Other web pages:  http://glub.ehu.es/  -  http://welcome.to/gogosoftware/


From hgg9140 at cola.ca.boeing.com  Wed Mar 14 13:53:47 2001
From: hgg9140 at cola.ca.boeing.com (Harry George)
Date: Wed, 14 Mar 2001 18:53:47 GMT
Subject: Who am I: can a class instance determine its own name?
References: <mailman.984034150.9635.python-list@python.org> <DYV4H3fqAHA.187@typhoon.i3s.net> <mailman.984324974.25480.python-list@python.org> <vZJg10qqAHA.210@typhoon.i3s.net> <xqxzoeol3ju.fsf@cola.ca.boeing.com> <98o8fu$2s4dj$1@ID-11957.news.dfncis.de>
Message-ID: <xqxhf0wtc5g.fsf@cola.ca.boeing.com>

I'd like to know too.  I never actually use that kind of code, but it
would be handy.

"Emile van Sebille" <emile at fenx.com> writes:

> Has locals changed so that it may be written to reliably?  My documentation
> still warns:
> 
> locals ()
> Return a dictionary representing the current local symbol table. Warning:
> The contents of this dictionary should not be modified; changes may not
> affect the values of local variables used by the interpreter.
> 
> Emile van Sebille
> emile at fenx.com
> 
> 
> 
> "Harry George" <hgg9140 at cola.ca.boeing.com> wrote in message
> news:xqxzoeol3ju.fsf at cola.ca.boeing.com...
> > class MyName:
> >     #--------------------------
> >     def init(self,name):
> >         self.name=name
> >         try:
> >             raise None
> >         except:
> >             c=sys.exc_info()[2].tb_frame.f_back.f_back
> >         c.f_locals[name]=self
> > ...
> >
> > MyName("me")
> > print me.name
> >
> > "Kevin Cazabon" <kevin_cazabon at hotmail.nospamplease!.com> writes:
> >
> > > The reason I say it's not elegant is that there is the possibility (read
> > > "probability") that the assigned name and the actual name can differ.
> > >
> > > We're assuming that the programmer assigns exactly the same name to
> both,
> > > which doesn't seem like a hard task, but it introduces the possibility
> of
> > > problems, and probably hard ones to debug too.
> > >
> > > If there were an automatic method, it would be MUCH more reliable.
> > >
> > > Just MHO.
> > >
> > > Kevin.
> > >
> > >
> > >
> > > "Paul Prescod" <paulp at ActiveState.com> wrote in message
> > > news:mailman.984324974.25480.python-list at python.org...
> > > | Kevin Cazabon wrote:
> > > | >
> > > | > ...
> > > | > You could also
> > > | > make it an initialization parameter.
> > > | >
> > > | > FooBar = Foo(name="FooBar")
> > > | >
> > > | > Just trying to help... not elegantly though.
> > > |
> > > | I disagree Kevin. This is not only the canonical way to give an
> instance
> > > | a name, it is also the elegant way. It gives you a lot more
> flexibility
> > > | because the name it has at runtime is unrelated to the variable name
> you
> > > | happened to have given it in your code. If you need to change one or
> the
> > > | other for any reason, you can do so safely.
> > > |
> > > |  Paul Prescod
> > > |
> > >
> > >
> >
> > --
> > Harry George                E-mail:  harry.g.george at boeing.com
> > The Boeing Company          Renton:  (425) 237-6915
> > P. O. Box 3707  02-CA       Everett: (425) 266-3868
> > Seattle, WA 98124-2207      Page:    (425) 631-8803
> 
> 

-- 
Harry George                E-mail:  harry.g.george at boeing.com
The Boeing Company          Renton:  (425) 237-6915
P. O. Box 3707  02-CA       Everett: (425) 266-3868
Seattle, WA 98124-2207      Page:    (425) 631-8803  


From aahz at panix.com  Sat Mar  3 16:55:28 2001
From: aahz at panix.com (Aahz Maruch)
Date: 3 Mar 2001 13:55:28 -0800
Subject: Bug fix releases
References: <LNBBLJKPBEHFEDALKOLCMEEPJBAA.tim.one@home.com> <mailman.982897324.9109.python-list@python.org> <200103031821.NAA24060@panix3.panix.com> <mailman.983646726.27322.python-list@python.org>
Message-ID: <97rp8g$4th$1@panix3.panix.com>

In article <mailman.983646726.27322.python-list at python.org>,
Guido van Rossum  <guido at digicool.com> wrote:
>Aahz writes:
>>
>> I think so, yes, on that latter clause.  I think perhaps it wasn't clear
>> at the time, but I believe that much of the yelling over "print >>" was
>> less over the specific design but because it came so close to the
>> release of 2.0 that there wasn't *time* to sit down and talk things
>> over rationally.
>
>In my eyes the issues are somewhat different: "print >>" couldn't
>possibly break existing code; nested scopes clearly do, and that's why
>we decided to use the __future__ statement.
>
>But I understand that you're saying that the community has grown so
>conservative that it can't stand new features even if they *are* fully
>backwards compatible.

Then you understand incorrectly.  There's a reason why I emphasized
"*time*" up above.  It takes time to grok a new feature, time to think
about whether and how we should argue in favor or against it, time to
write comprehensible and useful arguments.  In hindsight, I think you
probably did make the right design decision on "print >>", no matter how
ugly I think it looks.  But I still think you made absolutely the wrong
decision to include it in 2.0.

>So that relegates us at PythonLabs to a number of things: coding new
>modules (boring), or trying to improve performance of the virtual
>machine (equally boring, and difficult to boot), or fixing bugs (did I
>mention boring? :-).
>
>So what can we do for fun?  (Besides redesigning Zope, which is lots
>of fun, but runs into the same issues.)

Write new versions of Python.  You've come up with a specific protocol
in a later post that I think I approve of; I was trying to suggest a
balance between lots of grunt work maintenance and what I see as
perpetual language instability in the absence of any bug fix releases.

>Your math at first confused the hell out of me, but I see what you
>mean.  You want us to spend time on 2.0.1 which should be a bugfix
>release for 2.0, while at the same time working on 2.1 which is a new
>feature release.

Yup.  The idea is that because it's always an N and N-1 pair, the base
code is the same for both and applying patches to both should be
(relatively speaking) a small amount of extra work.  Most of the work
lies in deciding *which* patches should go into N-1.

>Guess what -- I am secretly (together with the PSU) planning a 2.0.1
>release.  I'm waiting however for obtaining the ownership rights to
>the 2.0 release, so we can fix the GPL incompatibility issue in the
>license at the same time.  (See the 1.6.1 release.)  I promise that
>2.0.1, unlike 1.6.1, will contain more than a token set of real
>bugfixes.  Hey, we already have a branch in the CVS tree for 2.0.1
>development!  (Tagged "release20-maint".)

Yay!  (Sorry, I'm not much of a CVS person; the one time I tried using
it, I couldn't even figure out where to download the software.  Call me
stupid.)

>We could use some checkins on that branch though.

Fair enough.

>> This means that each feature-based release gets one-and-only-one pure
>> bugfix release.  I think this will do much to promote the idea of Python
>> as a stable platform for application development.
>
>Anything we can do to please those republicans! :-)

<grin>

>> There are a number of ways I can see this working, including setting up
>> a separate project at SourceForge (e.g. pythonpatch.sourceforge.net).
>> But I don't think this will work at all unless the PythonLabs team is at
>> least willing to "bless" the bugfix release.  Uncle Timmy has been known
>> to make snarky comments about forever maintaining 1.5.2; I think this is
>> a usable compromise that will take relatively little effort to keep
>> going once it's set up.
>
>With the CVS branch it's *trivial* to keep it going.  We should have
>learned from the Tcl folks, they've had 8.NpM releases for a while.

I'm suggesting having one official PythonLabs-created bug fix release as
being a small incremental effort over the work in the feature release.
But if you want it to be an entirely community-driven effort, I can't
argue with that.

My one central point is that I think this will fail if PythonLabs
doesn't agree to formally certify each release.

>> I think one key advantage of this approach is that a lot more people
>> will be willing to try out a beta of a strict bugfix release, so the
>> release N bugfixes will get more testing than they otherwise would.
>
>Wait a minute!  Now you're making it too complicated.  Betas of bugfix
>releases?  That seems to defeat the purpose.  What kind of
>beta-testing does a pure bugfix release need?  Presumably each
>individual bugfix applied has already been tested before it is checked
>in!  

"The difference between theory and practice is that in theory, there is
no difference, but in practice, there is."

I've seen too many cases where a bugfix introduced new bugs somewhere
else.  Even if "tested", there might be a border case where an
unexpected result shows up.  Finally, there's the issue of system
testing, making sure the entire package of bugfixes works correctly.

The main reason I suggested two betas was to "lockstep" the bugfix
release to the next version's feature release.

>Or are you thinking of adding small new features to a "bugfix"
>release?  That ought to be a no-no according to your own philosophy!

That's correct.  One problem, though, is that sometimes it's a little
difficult to agree on whether a particular piece of code is a feature or
a bugfix.  For example, the recent work to resolve case-sensitive
imports could be argued either way -- and if we want Python 2.0 to run
on OS X, we'd better decide that it's a bugfix.  ;-)

>> If there's interest in this idea, I'll write it up as a formal PEP.
>
>Please do.

Okay, I'll do it after the conference.  I've e-mailed Barry to ask for a
PEP number.
-- 
                      --- Aahz (Copyright 2001 by aahz at pobox.com)

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

Nostalgia just ain't what it used to be


From juergen.erhard at gmx.net  Mon Mar 26 10:53:27 2001
From: juergen.erhard at gmx.net (=?ISO-8859-1?Q?=22J=FCrgen_A=2E_Erhard=22?=)
Date: Mon, 26 Mar 2001 17:53:27 +0200
Subject: PEP 240 scares me
In-Reply-To: <Pine.LNX.4.21.0103230727440.31371-100000@kev.ioc.ee> (message
	from Pearu Peterson on Fri, 23 Mar 2001 07:48:14 +0200)
References: <Pine.LNX.4.21.0103222338530.29360-100000@kev.ioc.ee> <99e2h61aao@news2.newsguy.com> <EjAu6.13052$_46.499367@e420r-atl2.usenetserver.com> <Pine.LNX.4.21.0103230727440.31371-100000@kev.ioc.ee>
Message-ID: <26032001.4@wanderer.local.jae.ddns.org>

>>>>> "Pearu" == Pearu Peterson <pearu at cens.ioc.ee> writes:

Hi Pearu,

    Pearu> Thanks for your replies. I now got it right what's in PEP 240.

I was drafting a message very similar to yours... that \d+\.\d+
confused me too.

Fortunately I read python-dev before posting it... and it slowly
(*sloooowly* ;-) dawned on me that what was really intended was not
1.4 == 1/4.  Which would have been royally stupid indeed.

So, I can feel with you (I made damn sure I pressed *delete* on that
draft and not *send* accidentially! ;-)

Bye, J

PS: I'm feeding python-dev daily from the pipermail archive... I'm not
posting this code anywhere as a) it's not really that impressive and
b) anyone who wants it can ask me... TPTB might possibly frown upon
such a thwarting of the marvellous "no-one can read the list directly,
but anyone can post" ;-P

-- 
J?rgen A. Erhard    juergen.erhard at gmx.net   phone: (GERMANY) 0721 27326
     MARS: http://members.tripod.com/Juergen_Erhard/mars_index.html
                  Advent, Advent, der Faschist brennt.
  Erst das ?rmchen, nun das Beinchen, bald das ganze Nazischweinchen.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 248 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/20010326/18d6a9e6/attachment.sig>

From drtimcouper at yahoo.co.uk  Thu Mar 29 03:11:08 2001
From: drtimcouper at yahoo.co.uk (=?iso-8859-1?q?Tim=20Couper?=)
Date: Thu, 29 Mar 2001 09:11:08 +0100 (BST)
Subject: OO Windows Registry API 
In-Reply-To: <E14iKb8-000473-00@mail.python.org>
Message-ID: <20010329081108.12569.qmail@web6405.mail.yahoo.com>

Benji

I would prefer this style of registry access. 

Comments:

1. reg[HKEY_LOCAL_MACHINE]['Software']['My App']

I'd prefer to see:

reg[HKEY_LOCAL_MACHINE]['Software\My App'] or,
perhaps, more Pythonesquely
 
reg[HKEY_LOCAL_MACHINE]['Software','My App']

2. What about a AddKey-type of method:

CurrentKey=reg[HKEY_LOCAL_MACHINE] ....
CurrentKey.AddKey('foo')

> 
> After using the _winreg module I was pining for the
> fjords -- oh, sorry -- I was pining for a more
> Pythonesc API.  I noticed that the Python
> documentation mentions that a better API may be
> written in the future, but none has been
> forthcoming.  So -- while up at 2AM with my
> 6-week-old son last night -- I wrote a wrapper that
> exposes the registry as a dictionary-like object. 
> It seems to work really well, and more importantly
> it feels more like a Python module should.
> 
> I decided to search Python-Dev for mentions of a
> higher-level API and found a squabble over various
> proposals.  Did the idea of an OO interface die last
> August, or is there any interest in what I've done?
> 
> (Untested) Example code:
> 
> from winreg import Registry, NewKey,
> HKEY_CURRENT_USER
> reg = Registry()
> 
> # Create a new "My App" key under
> HKEY_CURRENT_USER/Software.
> reg[HKEY_CURRENT_USER]['Software']['My App'] =
> NewKey()
> 
> # Add some values to the newly created key.
> # Note: the module only supports strings right now,
> but that's
> #  easy to change.
> reg[HKEY_CURRENT_USER]['Software']['My App']['foo']
> = 'value 1'
> reg[HKEY_CURRENT_USER]['Software']['My App']['bar']
> = 'value 2'
> 
> # Create a new subkey.
> reg[HKEY_CURRENT_USER]['Software']['My
> App']['subkey'] = NewKey()
> 
> # Delete a value.
> del reg[HKEY_CURRENT_USER]['Software']['My
> App']['foo']
> 
> # Delete a key.
> del reg[HKEY_CURRENT_USER]['Software']['My App']
> 
> 
> Any comments or suggestions?  If anyone is
> interested in the code, I'll post it on the web.
> 
> --
> Benji York
> york at ficom.net


____________________________________________________________
Do You Yahoo!?
Get your free @yahoo.co.uk address at http://mail.yahoo.co.uk
or your free @yahoo.ie address at http://mail.yahoo.ie



From hgg9140 at cola.ca.boeing.com  Fri Mar  9 13:59:20 2001
From: hgg9140 at cola.ca.boeing.com (Harry George)
Date: Fri, 9 Mar 2001 18:59:20 GMT
Subject: Divorcing Command Line and GUI
References: <mailman.984160633.14724.python-list@python.org>
Message-ID: <xqxae6ulqev.fsf@cola.ca.boeing.com>

A couple of options:

1. The Model-View-Controller pattern, where the model is your
underlying application engine and you have two views: commandline and
gui.

You will probably need as abstract "user interface" class which
defines the available callbacks, and then subclass the comline and gui
off that.  In the abstract case, you need ways to let the engine get
the data it wants (modal dialogs), and for the ui to trigger actions
in the engine (modeless actions).  

2. Keep the comline app as is.  Run it from the gui (e.g., via pipes).

Timothy Grant <tjg at exceptionalminds.com> writes:

> Hello Everyone.
> 
> This is something of a strange question, but please bear with
> me.
> 
> I have a series of tools that I have created that work from the
> command line. I have now been tasked with creating GUIs for
> those tools. The problem I'm faced with is that the command
> line tools and the GUI tools will probably be licensed
> differently (e.g, command line will be Open Source, GUI will be
> something undetermined as of yet.) This means that I can't put
> my GUI code in with my CL code.
> 
> My first cut at this demarcation sort of worked, but was also
> troublesome. So while I have some new ideas, I'm looking to see
> if anyone else in the python community has ever had to do
> something like this, and if there are any easily maintainable
> solutions to this problem.
> 
> 
> 
> -- 
> Stand Fast,
>     tjg.
> 
> Timothy Grant                         tjg at exceptionalminds.com
> Red Hat Certified Engineer            www.exceptionalminds.com
> Avalon Technology Group, Inc.         <><       (503) 246-3630
> >>>>>>>>>>>>>Linux, because rebooting is *NOT* normal<<<<<<<<<
> >>>>This machine was last rebooted:  51 days 22:09 hours ago<<
> 

-- 
Harry George                E-mail:  harry.g.george at boeing.com
The Boeing Company          Renton:  (425) 237-6915
P. O. Box 3707  02-CA       Everett: (425) 266-3868
Seattle, WA 98124-2207      Page:    (425) 631-8803  


From bbollenbach at homenospam.com  Tue Mar  6 11:40:31 2001
From: bbollenbach at homenospam.com (Brad Bollenbach)
Date: Tue, 06 Mar 2001 16:40:31 GMT
Subject: Can Python replace Visual Basic? Should it?
References: <5DZo6.11126$hn5.1578410@news1.rdc1.mb.home.com> <98266k0p5h@news1.newsguy.com>
Message-ID: <3q8p6.12338$hn5.1798569@news1.rdc1.mb.home.com>

"Alex Martelli" <aleaxit at yahoo.com> wrote in message
news:98266k0p5h at news1.newsguy.com...
> "Brad Bollenbach" <bbollenbach at homenospam.com> wrote in message
> news:5DZo6.11126$hn5.1578410 at news1.rdc1.mb.home.com...
> > Again, I'm much more interested in the real world experience of those
who
> > used to use the big visual development tools in Windows like VB, Delphi
or
> > Powerbuilder rather than theoretical answers.
>
> Why do you focus on 'used to use'?  One may perfectly well keep using
> some VB (&c) for tasks it's particularly well-suited to, while using
> Python wherever feasible (and, indeed, using both in the same project

Certainly. And for all intents and purposes "used to use" isn't necessary,
but the question happened to be directed (as its subject suggests) at those
who _have_ replaced VB with Python. :)

But realistically, I just wanted to emphasize my interest in hearing answers
from those who have or currently are using VB or one of the other Win32
visual development tools. Personally, I'm a former VB programmer now working
in Powerbuilder and am unhappy with both, again based on my dislike of IDE's
that obscure the programmer's view of what's going on.

I could explain _why_ I don't like the 40,000 foot views of IDE's, but
that's another thread.

Brad




From mwh21 at cam.ac.uk  Tue Mar  6 10:24:10 2001
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 06 Mar 2001 15:24:10 +0000
Subject: I come to praise .join, not to bury it...
References: <mailman.983485923.1136.python-list@python.org> <97p7iq$kk2$1@nntp6.u.washington.edu> <97rka002k9@news1.newsguy.com> <kuu257kb1n.fsf@lasipalatsi.fi> <982a4f0su3@news1.newsguy.com> <kuofvfjb77.fsf@lasipalatsi.fi>
Message-ID: <m3ofvec44l.fsf@atrus.jesus.cam.ac.uk>

Erno Kuusela <erno-news at erno.iki.fi> writes:

> In article <982a4f0su3 at news1.newsguy.com>, "Alex Martelli"
> <aleaxit at yahoo.com> writes:
> 
> | "Erno Kuusela" <erno-news at erno.iki.fi> wrote in message
> 
> || what's wrong with vague aesthetic?-) if lots of people share it,
> || it's not to be dismissed so lightly.
> 
> | Lots of people share the 'vague aesthetic' of desiring
> | to use braces instead of whitespace to delimit blocks
> | of statements, for example.
> 
> | Python dismisses that lightly, and that is definitely
> | a part of what makes it such a wonderful language.
> 
> it could also be argued that python uses whitespace instead
> of braces because of vague aesthetic shared by lots of people. :)

Some people don't have a problem with, or even quite like, ''.join,
too.

They tend to be quieter, though.

Cheers,
M.

-- 
  For their next act, they'll no doubt be buying a firewall running
  under NT, which makes about as much sense as building a prison out
  of meringue.                                           -- -:Tanuki:-
               -- http://home.xnet.com/~raven/Sysadmin/ASR.Quotes.html


From bedge at troikanetworks.com  Tue Mar 20 20:04:07 2001
From: bedge at troikanetworks.com (Bruce Edge)
Date: Tue, 20 Mar 2001 17:04:07 -0800
Subject: InteractiveConsole locals parameter
Message-ID: <3AB7FE07.5F7584F5@troikanetworks.com>

I can't get my __main__.__dict__ passed to an InteractiveConsole:

def run_python():
	import code
	import __main__
	locals = __main__.__dict__
	x = code.InteractiveConsole(locals)
	x.interact("")

Shouldn't this have the same namespace as my main interpreter?

If not, how can I get my main interp namespace propagated down to my
InteractiveConsole?

Are there any detailed docs on this? 
I've read all the ones I can find.

Thanks, Bruce.


Here's my test:

Python 2.0 (#42, Mar 15 2001, 13:24:46) 
[GCC 2.95.3 20010219 (prerelease)] on linux2
Type "copyright", "credits" or "license" for more information.

>>> from CLI import *	<--run my CLI app

k-> 
k-> python 		<-- command to start subshell, calls run_python, above
>>> dir()		<-- this is all that's defined
['__builtins__', '__doc__', '__name__', 'atexit', 'histfile', 'os',
'readline', 'rlcompleter', 'savehist', 'sys']
>>> 
>>> 			<-- exit subshell
k-> 			<-- exit app
>>> dir()		<-- top level, this is what I want in my subshell:
['CommandException', 'KCompleter', 'Kcmd', 'Port', 'Zone',
'__builtins__', '__doc__', '__name__', 'atexit', 'basedir', 'cNode',
'cOpt', 'cType', 'cli', 'cmd', 'cmds', 'copy', 'ctree', 'dprintf',
'dprintf_exception', 'dprintf_init', 'dprintf_level', 'histfile',
'histname', 'kcmd', 'kcompl', 'kcomplMode', 'kcompleter', 'kinterp',
'kuser', 'mType', 'main', 'modules', 'nType', 'os', 'port', 're',
'readline', 'register_new_command', 'rights', 'rlcompleter', 'savehist',
'sl', 'string', 'sys', 'tokens', 'topcmds', 'traceback', 'types',
'util']


From alf at leo.logilab.fr  Fri Mar 16 04:49:37 2001
From: alf at leo.logilab.fr (Alexandre Fayolle)
Date: Fri, 16 Mar 2001 09:49:37 +0000 (UTC)
Subject: Basic information requested
References: <MHks6.144$lC5.52172@news2-win.server.ntlworld.com>
Message-ID: <98snjh$hsu$1@norfair.nerim.net>

Jamie MacIsaac <jamie at cctech.co.uk> wrote:
> Hi,

> I've just started looking at Python and would like to use it to perform some
> simple admin tasks on a Windows network. However, I'm having difficulty
> finding out how to accomplish elementary tasks.

> Can someone please point me to sample code that shows how to map a network
> share and copy some files? Is there a central resource of code samples?


You may want to purchase a copy of 'Python Programming on Win32' (O'Reilly, 
by Mark Hammond and Andy Robinson, ISBN 1-56592-621-8). 
Chapter 16 covers this point. You can download the code samples from 
http://starship.python.net/crew/mhammond/ppw32/

Cheers
 
Alexandre Fayolle
-- 
http://www.logilab.com 
Narval is the first software agent available as free software (GPL).
LOGILAB, Paris (France).


From ben.hutchings at roundpoint.com  Tue Mar 20 17:24:59 2001
From: ben.hutchings at roundpoint.com (Ben Hutchings)
Date: 20 Mar 2001 14:24:59 -0800
Subject: How to increase PythonWin v2.0: COM browser limits?
References: <994psh$so5$1@news1.xs4all.nl> <uy9u06vji.fsf@roundpoint.com>
Message-ID: <uhf0o6ptw.fsf@roundpoint.com>

I wrote:

> The wrapper for COM enumerators, in true Pythonic style, does not
> check the size of the collection in advance, but attempts to read
> items until this yields an error.  I think that the category
> enumerator implementation is reporting the no-more-items error in the
> wrong way (returning E_OUTOFMEMORY rather than S_FALSE), so the
> wrapper does not translate it into an IndexError exception which
> would cause the loop to terminate successfully.  OLEView probably
> just doesn't check for the error but displays what it already got.

Let me revise that:

A Python iterator for a COM 'container' is an instance of
win32com.client.util.Enumerator, which wraps a PyIEnumVARIANT object
that in turn wraps an IEnumVARIANT interface pointer obtained from the
container.  When an IEnumVARIANT user requests more items than there
are left, the implementation should return any items that are left,
with a status code of S_FALSE (not an error code).

In the case of iteration, Enumerator will request only 1 item each
time.  PyIEnumVARIANT's wrapper function returns a list of items,
which at the end of iteration will have length 0.  Enumerator checks
for this case and throws an IndexError when it occurs, which is the
normal way for iteration to end.  Unfortunately it appears that this
specific implementation of IEnumVARIANT returns a status code of
E_OUTOFMEMORY (an error code), which PyIEnumVARIANT turns into a
win32types.com_error exception.

OLEView probably treats this error in the same way as a normal end
of enumeration, and displays the results it already obtained.

-- 
Any opinions expressed are my own and not necessarily those of Roundpoint.


From sholden at holdenweb.com  Fri Mar  9 23:33:23 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Fri, 9 Mar 2001 23:33:23 -0500
Subject: true newbie
References: <3aa9a9e9$0$65277$38c90ffe@news1.twtelecom.net>
Message-ID: <8ciq6.46945$1D5.2247691@e420r-atl1.usenetserver.com>

Try the Non-Programmer's Tutorial, at

    http://www.honors.montana.edu/~jjc/easytut/easytut/easytut.html

regards
 Steve

"Elaine Sweeny" <bobbo at milwpc.con> wrote in message
news:3aa9a9e9$0$65277$38c90ffe at news1.twtelecom.net...
> can anyone help me find a tutorial for python thats for people just
starting
> out in programming. The one one that comes with it is ok but its still too
> difficult.
>         thanks
>
>
>




From mcfletch at home.com  Fri Mar 16 15:04:38 2001
From: mcfletch at home.com (Mike C. Fletcher)
Date: Fri, 16 Mar 2001 15:04:38 -0500
Subject: can a class intantiate itself to None?
In-Reply-To: <20010316112211.E3109@trufflehunter.avalongroup.net>
Message-ID: <000001c0ae54$550f0ef0$a9807018@cr706570a>

Classic approach is to use a factory function...

def foo( bar ):
	if type(bar) == StringType:
		return _Foo( bar )
	else:
		return None

-----Original Message-----
From: Timothy Grant [mailto:tjg at exceptionalminds.com]
Sent: Friday, March 16, 2001 2:22 PM
To: Python People
Subject: can a class intantiate itself to None?


I'd like to create a class that can instantiate itself to None
when instantiation fails...

class foo:
	def __init__(self, bar):
		if type(bar) == StringType:
				self.data = 'This is a String'
				

x = foo('a')

y = foo(1)  # In this instance I would like y to be None

So, can it be done, or do I need to raise an exception, catch
it and then set y to None?

-- 
Stand Fast,
    tjg.

Timothy Grant                         tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.         <><       (503) 246-3630
>>>>>>>>>>>>>Linux, because rebooting is *NOT* normal<<<<<<<<<
>>>>This machine was last rebooted:  58 days 23:33 hours ago<<

-- 
http://mail.python.org/mailman/listinfo/python-list




From dnew at san.rr.com  Fri Mar 16 13:32:56 2001
From: dnew at san.rr.com (Darren New)
Date: Fri, 16 Mar 2001 18:32:56 GMT
Subject: list.sort(cmpfunc) question
References: <m1ae6lg5qt.fsf@pclabs188.labs.trema.com> <98t9dq$qvu$1@panix6.panix.com> <98tf9d$g3v$1@news.mathworks.com> <3AB249FB.131080CD@san.rr.com> <98thua$i40$1@news.mathworks.com> <Wnss6.186$sk3.53027@newsb.telia.net>
Message-ID: <3AB25C59.284E58E3@san.rr.com>

Fredrik Lundh wrote:
> in other words, "recommended but not required"

Even more strongly than "recommended", but rather "you better have a darn
good reason and know every result of your change before you consider doing
this." :-) It's like "Unix root accounts SHOULD have a hard-to-guess
password."  It's possible to run UNIX with no root password, but you ought
to have a good reason.

And yes, there's an RFC defining what "SHOULD" and "MUST" and such mean. :-)

-- 
Darren New / Senior MTS & Free Radical / Invisible Worlds Inc.
San Diego, CA, USA (PST).  Cryptokeys on demand.


From qrczak at knm.org.pl  Sat Mar 31 16:17:07 2001
From: qrczak at knm.org.pl (Marcin 'Qrczak' Kowalczyk)
Date: 31 Mar 2001 21:17:07 GMT
Subject: list vs tuple
References: <mailman.986066384.30712.python-list@python.org> <d1rx6.1053$p5.4024@news1.rivrw1.nsw.optushome.com.au>
Message-ID: <slrn9cciaj.3ug.qrczak@qrnik.zagroda>

Sat, 31 Mar 2001 20:23:37 GMT, deadmeat <root@[127.0.0.1]> pisze:

> In Pascal the programmer defines what is a pointer and what is
> not, and the difference is specified in the syntax. In Python does
> not. The difference is based on what the name refers to, and that
> difference is not specified in the syntax - which is my point.

THERE IS NO DIFFERENCE.

-- 
 __("<  Marcin Kowalczyk * qrczak at knm.org.pl http://qrczak.ids.net.pl/
 \__/
  ^^                      SYGNATURA ZAST?PCZA
QRCZAK


From ewilson at nospam.jamdat.com  Wed Mar 14 18:57:38 2001
From: ewilson at nospam.jamdat.com (eric wilson)
Date: Wed, 14 Mar 2001 23:57:38 GMT
Subject: compile / exec
Message-ID: <SzTr6.14658$R_6.1511179@newsread2.prod.itd.earthlink.net>

hi -
i am new to python and i'm trying to use the compile / exec feature for
speeding up a loop.  i'm using python v2.0 with pythonwin.  i don't
understand why this doesn't work....
thanks in advance for the help.
//eric

the code snippet looks like this:
    exec_code = compile("""
    for eachLine in map(strip, allLines):
        mySequence = split('\s\s+|\t', eachLine)
        foundFields = str(gm2local(DateTimeFromString(mySequence[0])))[:-3]
+ '\t' + mySequence[1]+ '\t' + mySequence[4]+ '\t' + mySequence[5]+ '\t' +
mySequence[9]+ '\t' + mySequence[16] + '\n'
        fw.write(foundFields)
        count = count + 1
    """, '', 'exec')
    exec exec_code

the error message that i get when this piece of code runs is:
Traceback (most recent call last):
  File "C:\Python20\Pythonwin\pywin\framework\scriptutils.py", line 301, in
RunScript
    exec codeObject in __main__.__dict__
  File "E:\W2\pytest\logman.py", line 114, in ?
    checkArgs()
  File "E:\W2\pytest\logman.py", line 110, in checkArgs
    parmFind(sys.argv[1], sys.argv[2])
  File "E:\W2\pytest\logman.py", line 55, in parmFindRAM
    exec_code = compile("""
  File "<string>", line 2
     for eachLine in map(strip, allLines):
     ^
 SyntaxError: invalid syntax






From bryan at eevolved.com  Sat Mar  3 00:12:06 2001
From: bryan at eevolved.com (Bryan Mongeau)
Date: Sat, 03 Mar 2001 05:12:06 GMT
Subject: Adding a method to an instance
Message-ID: <G2%n6.360714$JT5.11606234@news20.bellglobal.com>

Quick one for you guys:

I would like to add a method to an instantiated class. 

class foo:
  def __init__(self):
    self.bar = 123

f = foo()

Now to add a method to the instance f of class foo:

def newMethod(self):
  print self.bar

f.__class__.newMethod = newMethod

This works, except that it modifies the definition of class foo!

>>> dir(foo)
['__doc__', '__init__', '__module__']
>>> f.__class__.newMethod = newMethod
>>> f.newMethod()
123
>>> dir(foo)
['__doc__', '__init__', '__module__', 'newMethod']

Is there a way to add the method only to the instance?  Oh, and please 
don't ask me *why* I want to do this. Just trust me :)

Regards,
-- 
<=====================================>
Bryan Mongeau
Lead Developer, Director
eEvolved Real-Time Technologies Inc.
Website:    http://www.eevolved.com
Public key: http://eevolved.com/bcm.pk
<=====================================>

"The eternal mystery of the world is its comprehensibility."-- Einstein



From chris at voodooland.net  Sat Mar 10 19:40:31 2001
From: chris at voodooland.net (Chris Watson)
Date: Sat, 10 Mar 2001 18:40:31 -0600 (CST)
Subject: Poll'able Queue objects.
In-Reply-To: <txkpufpxlgu.fsf@vortex.ukshells.co.uk>
Message-ID: <Pine.BSF.4.21.0103101827050.33320-100000@open-systems.net>

As an aside. Granted this is *NON PORTABLE* and BSD specific. BSD has a
*really* cool replacement for select/poll called kqueue. It is currently
in FreeBSD and NetBSD AFAIK.

DESCRIPTION
     kqueue() provides a generic method of notifying the user when an
     event happens or a condition holds, based on the results of small
     pieces of kernel code termed filters.  A kevent is identified by the
     (ident,filter) pair; there may only be one unique kevent per kqueue.

     The filter is executed upon the initial registration of a kevent in
     order to detect whether a preexisting condition is present, and is
     also executed whenever an event is passed to the filter for evaluation.  
     If the filter determines that the condition should be reported, then 
     the kevent is placed on the kqueue for the user to retrieve.

     The filter is also run when the user attempts to retrieve the kevent
     from the kqueue.  If the filter indicates that the condition that
     triggered the event no longer holds, the kevent is removed from the
     kqueue and is not returned.

<...snip...>

It scales like a fantasy :) 
http://www.kegel.com/dkftpbench/Poller_bench.html
Very very nice. For those of use using BSD that is. Also Doug White wrote
a Python wrapper for kqueue. It's in FreeBSD's ports tree as Py-Kqueue.

--
=============================================================================
-Chris Watson         (316) 326-3862 | FreeBSD Consultant, FreeBSD Geek 
Work:              scanner at jurai.net | Open Systems Inc., Wellington, Kansas
Home:  scanner at deceptively.shady.org | http://open-systems.net
=============================================================================
WINDOWS: "Where do you want to go today?"
LINUX: "Where do you want to go tomorrow?"
BSD: "Are you guys coming or what?"
=============================================================================
irc.openprojects.net #FreeBSD -Join the revolution!
ICQ: 20016186




From aweis at gmx.de  Wed Mar 21 09:09:31 2001
From: aweis at gmx.de (Arnold Weis)
Date: Wed, 21 Mar 2001 14:09:31 GMT
Subject: HTML escape in Python 1.5.2
References: <3fQTii$l4N@openbazaar.net>
Message-ID: <3ab8ae34.14307242@News.CIS.DFN.DE>

Pahud.bbs at openbazaar.net (???J?w) wrote:

>I want to do the html escaping in python, however, HTMLgen class
>does not ship with python 1.5.2. Is there any alternative?
>Thanks in advance.

Hi,

is cgi.escape() what you are searching for?

Regards

	Arnold


From tim.one at home.com  Fri Mar  2 18:06:25 2001
From: tim.one at home.com (Tim Peters)
Date: Fri, 2 Mar 2001 18:06:25 -0500
Subject: the demise of 'from foo import * and its implications?
In-Reply-To: <XFMail.20010302143902.shaleh@valinux.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCCEJAJDAA.tim.one@home.com>

[Tim]
> What gave you the idea that "import *" is going away?

[Sean 'Shaleh' Perry]
> /me needs to take a week off apparently.  Either that or stop believing
> anything he reads.
>
> I would have SWORN I read it here, and a fellow python hacker said
> something to that affect as well.

Oh, that.  That's just people panicking.  Don't listen to people -- they're
always wrong <wink>.  Here's an item from the 2.1b1 NEWS file, which you can
pick up from the 2.1b1 release later today:

  - Using "from...import *" or "exec" without in-clause in a function
    scope that also defines a lambda or nested function with one or
    more free (non-local) variables.  The presence of the import* or
    bare exec makes it impossible for the compiler to determine the
    exact set of local variables in the outer scope, which makes it
    impossible to determine the bindings for free variables in the
    inner scope.  To avoid the warning about import *, change it into
    an import of explicitly name object, or move the import* statement
    to the global scope; to avoid the warning about bare exec, use
    exec...in... (a good idea anyway -- there's a possibility that
    bare exec will be deprecated in the future).

So, yes, by the time 2.2 rolls around,

    using "from ... import *" in a
         function scope that also contains a
             lambda or nested function
                definition
                whose body contains at least one free variable
    will go away.

It's quite an ambitious act of compression to read that as "import * is going
away" <wink>.  And, according to the Reference Manual, "import *" at other
than module scope has never been legitimate (nested functions or not).

There's been no hint that import* will ever go away at module scope, despite
that it's excellent advice to avoid it whenever *reasonably* possible.
Binding to massive, flat systems (like windowing toolkits) can make avoiding
it unreasonable.

> Whichever, I will promptly use a gag as I am just digging myself a hole
> today.

Holes are comfy!  Take in a blanket and some lavender eye pillows.  import*
will still be here when you come out again <wink>.

jealously y'rs  - tim




From no at thank.you  Tue Mar 27 14:59:17 2001
From: no at thank.you (Owen Ap Owen)
Date: Tue, 27 Mar 2001 11:59:17 -0800
Subject: Need Help with mxODBC Connection String
Message-ID: <MPG.152a90eb9b8dbc09896a3@news.pacbell.net>

I'm using Python 2.0 and the latest mxODBC. I'm trying to connect to SQL 
Server 2000. 

When I try to connect using this connection string:

db 
=mx.ODBC.Windows.DriverConnect('DSN=JTech_Connection;UID=tech;PWD=tech')

I get this error message: "[Microsoft][ODBC SQL Server Driver][SQL 
Server] Changed database context to 'JTech'

That's fine with me because I want the database context to be 'JTech'. 
But mxODBC does not deal well with this message and the db connection is 
not created. 

Nothing I do seems to change this. I've tried specifying the default 
database in the connection string. I've also tried specifying and not 
specifying the database to change to in the ODBC driver setup. No matter 
what I do I get the message about the database context being changed.

Does anybody have any insights into how I can modify my connection string 
so that it is accepted?


From gmunsey at adobe.com  Mon Mar 12 17:10:00 2001
From: gmunsey at adobe.com (Grant Munsey)
Date: Mon, 12 Mar 2001 14:10:00 -0800
Subject: PyArg_ParseTuple on steroids?
Message-ID: <YObr6.839$dq1.77192@news.pacbell.net>

I have been doing a lot of Python to C library adapters recently.
Currently I'm generating the code for each Python cover routine with
a Python program which uses a crude IDL to drive the generator. I
build the IDL descriptors by hand or with another Python program in
the case where the API is described in a "regular" way.

Currently the C source that is output by the generator looks pretty
much like code that would be generated by hand  ... with PyArg_ParseTuple()
at the beginning and Py_BuildValue() at the end.

I was thinking of changing this to have the generator generate a table
of parameter descriptions for each procedure and one call to an "uber"
argument handler ... that would check the parameters, get the C values,
call the function, and convert the results back into a Python return value.
Kind of a PyArg_ParseTuple(), call function, Py_BuildValue() all rolled
into one but that could handle many more parameter types than the standard
routines do.

This would have the benefit of smaller overall code generation and the
various parameter handlers would be in one place instead of spread all
over a large Python program which generates code.

Anyone done this in the past and have any advice?




From doughellmann at bigfoot.com  Wed Mar 14 06:37:01 2001
From: doughellmann at bigfoot.com (Doug Hellmann)
Date: 14 Mar 2001 06:37:01 -0500
Subject: main in mac?
In-Reply-To: <20010313123230.19201.00000153@ng-dh1.aol.com>
References: <mailman.984497176.29067.python-list@python.org> 
	<20010313123230.19201.00000153@ng-dh1.aol.com>
Message-ID: <20010314113434.MJDD2709.femail4.sdc1.sfba.home.com@[192.168.1.6]>

On 13 Mar 2001 17:32:30 +0000, Park997 wrote:
> >I don't know what the run all button is (not on mac), but the whole reason
> >for this idiom is so it wont 'do anything' when imported. If a file named
> >x.py is imported, then it's __name__ is x. If it is being run as a program,
> >then it's __name__ is __main__. 
> >
> >
> >David
> 
> I knew that was the expected behavior, and the
> disadvantage of replacing if __name__ is __main__. 
> with if 1: is that it does run when imported. Turning the run as main option on
> or off produces the expected behavior on the mac.

It's not that you're running on a Mac, it's that you're using an IDE. If
you double click the saved file in the Finder the __name__ would be set
to '__main__' as you expect -- just like if a Windows or UN*X user runs
the file directly.

Doug





From aleaxit at yahoo.com  Fri Mar 23 01:48:30 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 23 Mar 2001 07:48:30 +0100
Subject: Why Python/Jython?
References: <99cl9e$b2b$1@Merlin.et.bocholt.fh-gelsenkirchen.de> <mailman.985260089.28444.python-list@python.org> <srzu6.12868$_46.475705@e420r-atl2.usenetserver.com>
Message-ID: <99erof0173p@news2.newsguy.com>

"Steve Holden" <sholden at holdenweb.com> wrote in message
news:srzu6.12868$_46.475705 at e420r-atl2.usenetserver.com...
    [snip]
> > Programmer performance is usually gauged by coding time and
> > number of bugs, both of which are vaguely proportional to the number of
> > lines of code in most sane languages*.
> >
> > * Perl is a noteworthy exception
> >
> I saw no inconsistency here. Why would Perl be an exception to your rule?

Lines of code may not be proportional to 'code size' in languages
which afford many styles, one of which is 'insanely compact'.  I
am not sure this really changes things wrt Perl -- a given
individual coder is likely to have roughly constant density of
'code size'/line in his or herl Perl code -- but it surely applies to
APL.  In any case, #bugs and code time depend on code-size
(e.g., number of 'operators' in some vague sense), which in turn
may be only indirectly proportional to measures such as SLOC.


Alex





From duncan at rcp.co.uk  Thu Mar 15 05:09:06 2001
From: duncan at rcp.co.uk (Duncan Booth)
Date: Thu, 15 Mar 2001 10:09:06 +0000 (UTC)
Subject: Regular Expressions and Python 1.5.2
References: <98q0k8$2b77$1@norfair.nerim.net>
Message-ID: <Xns90656694BD35duncanrcpcouk@127.0.0.1>

Alexandre Fayolle <alf at leo.logilab.fr> wrote in <98q0k8$2b77$1
@norfair.nerim.net>:

>Is this a known bug in the module re of python 1.5.2, or am I missing 
>something obvious?
>
>>>> import re
>>>> t = "m&n"
>>>> re.sub('[^/_ -]','*',t)
>'***'
>>>> re.sub('[^_ -/]','*',t)
>'*&*'
>
I think you are missing something obvious. You specified a range from space 
to '/', and the ampersand is in that range, so it doesn't get replaced. In 
the first example the dash doesn't form part of a range so it is treated 
literally.


From timr at probo.com  Mon Mar 19 23:58:29 2001
From: timr at probo.com (Tim Roberts)
Date: Mon, 19 Mar 2001 20:58:29 -0800
Subject: Python style questions
References: <98tb91$9t9$1@saltmine.radix.net> <m34rwpilxg.fsf@atrus.jesus.cam.ac.uk> <995o0u$itc$1@news.mathworks.com>
Message-ID: <0oodbtgdgc3tv452iiqs29n2rd3534dc2k@4ax.com>

Joshua Marshall <jmarshal at mathworks.com> wrote:
>
>A switch statement may just be syntactic sugar, but not necessarily in
>a simple way.  A good compiler will compile switch statements into
>jump-tables or, at worst, binary searches.  Cluttery to code by hand.
>
>Sure this is "just" an efficiency thing, but switches can get quite big.

But Python, at least today, is not compiled.  Switch statements in an
interpreted language are not as big of a "win".
--
- Tim Roberts, timr at probo.com
  Providenza & Boekelheide, Inc.


From aleaxit at yahoo.com  Tue Mar 13 04:29:54 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Tue, 13 Mar 2001 10:29:54 +0100
Subject: Einstein's Riddle
References: <mailman.984369618.12493.python-list@python.org> <NI9r6.2583$54.2812@www.newsranger.com>
Message-ID: <98kpaj0p6m@news2.newsguy.com>

"Gregory Jorgensen" <gregj at pobox.com> wrote in message
news:NI9r6.2583$54.2812 at www.newsranger.com...
> So I'm stupid, is that it? I've taken lots of IQ tests and none of them
had this
> question on them; that's probably why I don't score so well! Actually I
did
> figure out that the German owned the fish. Furthermore I was able to
determine
> that the fish is named Eric.

Actually, I'm pretty sure she was called Wanda.


Alex





From tim.hochberg at ieee.org  Fri Mar  9 15:11:12 2001
From: tim.hochberg at ieee.org (Tim Hochberg)
Date: Fri, 09 Mar 2001 20:11:12 GMT
Subject: "!=" is bad form. Re: sorry....never mind
References: <BLEOJGEMBNBNEAKMIFDHEELCPLAA.dbrueck@edgix.com> <mailman.984164594.2536.python-list@python.org>
Message-ID: <ANaq6.11508$Xt3.1922910@news1.rdc1.az.home.com>

"Peter Schneider-Kamp" <peter at schneider-kamp.de> wrote in message
[SNIP]
> Well, Python *does* define an ordering of a and b, no matter what
> a and b may be. E.g.:
>
> >>> 1 > (1,2)
> 0
> >>> 1 < (1,2)
> 1
[SNIP]

This used to be true, but for better or for worse this hasn't been true for
a while. It's true for most builtin objects still, but for instance:

>>> class Uncomparable:
 def __cmp__(self, other):
  raise RuntimeError("I can't be compared")

>>> u = Uncomparable()
>>> u < 5
Traceback (innermost last):
  File "<pyshell#10>", line 1, in ?
    u < 5
  File "<pyshell#8>", line 3, in __cmp__
    raise RuntimeError("I can't be compared")
RuntimeError: I can't be compared

The guarantee of all types being ordered is likely to erode further with the
advent of rich comparisons in Python 2.1 (PEP 207).

-tim





From bedge at troikanetworks.com  Thu Mar 15 13:41:44 2001
From: bedge at troikanetworks.com (Bruce Edge)
Date: Thu, 15 Mar 2001 10:41:44 -0800
Subject: redirecting stdio
Message-ID: <3AB10CE8.C4978B35@troikanetworks.com>

I'm trying to use pdb to debug an app that uses stdio.
I'd like to get pdb, or my app to use another terminal for it's stdio.

Can either the pdb class or the cmd class be vectored to use another
terminal for I/O?


From m.hadfield at niwa.cri.nz  Thu Mar  8 15:29:45 2001
From: m.hadfield at niwa.cri.nz (Mark Hadfield)
Date: Fri, 9 Mar 2001 09:29:45 +1300
Subject: grabbing return codes from os.system() call
References: <%wFp6.9481$dL4.126404@vixen.cso.uiuc.edu> <x8Gp6.899$54.943@www.newsranger.com>
Message-ID: <984083439.472914@clam-ext>

"Gregory Jorgensen" <gregj at pobox.com> wrote in message
news:x8Gp6.899$54.943 at www.newsranger.com...
> os.system returns the exit status. ... on Windows the
> exit status is always 0.

Not so.

On my Win 2000 box with Python 2.0, successful commands return 0 and
unsuccessful commands (where the command is not found) return 1.

Of course there may be more to it, but that's what I found in 20 s of
experimentation...

---
Mark Hadfield
m.hadfield at niwa.cri.nz  http://katipo.niwa.cri.nz/~hadfield
National Institute for Water and Atmospheric Research




From ej at ee.duke.edu  Fri Mar 30 03:23:35 2001
From: ej at ee.duke.edu (eric jones)
Date: Fri, 30 Mar 2001 08:23:35 GMT
Subject: Performance hit for ALLOW_THREADS macros? 
Message-ID: <boXw6.211304$__6.36867117@typhoon.southeast.rr.com>

Hello,

Adding Py_BEGIN_ALLOW_THREADS and Py_END_ALLOW_THREADS appears to have
significant negative impact on my extension module's speed -- even when not
using threads. (Linux RH6.1, Python2.1a2)

I have a numeric routine that fills a large matrix.  The routine is written
in FORTRAN and lives in an extension module (wrapped by f2py).  The routine
has be written such that it can be called twice in parallel with one call
filling the upper half of the matrix and one filling the
lower half of the matrix (nice for multiprocessors).  Placing
Py_BEGIN_ALLOW_THREADS
and Py_END_ALLOW_THREADS around the call to the Fortran routine in the
extension module allows it to be called independently from two Python
threads.  Doing this provided very marginal speed-up (10-20% maybe).
Because of the run time of the routine (10s of seconds to days), I expected
a factor much closer to 100%. I played with sys.setcheckinterval() a little
to see if it helped things and it didn't.  This is what I expected reasoning
that all CPU cycles should be burning in the Fortran routines.

Looking for the issue, I ran the routine in single threaded mode, once
compiled with (1) Py_BEGIN_ALLOW_THREADS/Py_END_ALLOW_THREADS, and (2) once
without.  On a typical small problem, single threaded times showed the
following:

    (1) using ALLOW_THREADS macros   20.54 seconds
    (2) no macros                                         13.78 seconds

Why would simply adding these macros affect the run time so much?  I didn't
expect them to have any affect (especially when threads aren't even used).
Here's where I put the macros in the code.

    (1)
    static PyObject *foo(PyObject *capi_self, PyObject *capi_args, PyObject
*capi_keywds,     void (*f2py_func)()) {
    {
          ...
          Py_BEGIN_ALLOW_THREADS
          (*f2py_func)(&alpha,&i12,&cfreq,epstj,muej,...);
          Py_END_ALLOW_THREADS
          ...
    }

    (2)
    static PyObject *foo(PyObject *capi_self, PyObject *capi_args, PyObject
*capi_keywds,     void (*f2py_func)()) {
    {
          ...
          (*f2py_func)(&alpha,&i12,&cfreq,epstj,muej,...);
          ...
    }

Any hints as to what is happening here?

thanks,
eric





From nas at python.ca  Tue Mar 27 00:15:52 2001
From: nas at python.ca (Neil Schemenauer)
Date: Mon, 26 Mar 2001 21:15:52 -0800
Subject: fixing map(None, [1,2,3]) ?
In-Reply-To: <873dc040ny.fsf@mathdogs.com>; from mkc@mathdogs.com on Mon, Mar 26, 2001 at 04:37:21PM -0600
References: <873dc040ny.fsf@mathdogs.com>
Message-ID: <20010326211552.C17390@glacier.fnational.com>

Mike Coleman wrote:
> Shouldn't "map(None, [1,2,3])" return "[(1,), (2,), (3,)]"?
> 
> Is there any hope that this could be fixed?

Probably not.  Use zip().

  Neil



From hans.kristian.ruud at inenco.no  Wed Mar 28 12:01:52 2001
From: hans.kristian.ruud at inenco.no (Hans Kristian Ruud)
Date: Wed, 28 Mar 2001 17:01:52 GMT
Subject: The Scale widget
Message-ID: <3AC216C0.F38935C1@inenco.no>

Is there a way to modify the scale widget so that the slider
only can move in one direction?

-- 
Hans Kristian


From tim.one at home.com  Mon Mar 19 21:33:07 2001
From: tim.one at home.com (Tim Peters)
Date: Mon, 19 Mar 2001 21:33:07 -0500
Subject: Wrong reasons to enhancing the language
In-Reply-To: <99661m$scc$1@panix6.panix.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCEEMJJGAA.tim.one@home.com>

[Aahz]
> ...
> Nested scopes are an optional feature of Python 2.1

I wouldn't phrase it that way.  Nested scopes are a mandatory feature of
Python 2.2, and 2.1 is giving people a one-release grace period to convert
their code to the new semantics (if they indeed have any code in need of
converting ... seems that conflicts are mighty rare so far).

> and the others are simply *proposed* features.
>
> This is a *great* time to be debating the numerics issues.

Indeed it is.  So far, though, IEEE-754 aficionados are conspicuous by
absence.

and-we-really-*need*-a-fifth-group-that-doesn't-agree-with-any-of-the-
    others<wink>-ly y'rs  - tim




From tjg at exceptionalminds.com  Fri Mar 16 14:22:11 2001
From: tjg at exceptionalminds.com (Timothy Grant)
Date: Fri, 16 Mar 2001 11:22:11 -0800
Subject: can a class intantiate itself to None?
Message-ID: <20010316112211.E3109@trufflehunter.avalongroup.net>

I'd like to create a class that can instantiate itself to None
when instantiation fails...

class foo:
	def __init__(self, bar):
		if type(bar) == StringType:
				self.data = 'This is a String'
				

x = foo('a')

y = foo(1)  # In this instance I would like y to be None

So, can it be done, or do I need to raise an exception, catch
it and then set y to None?

-- 
Stand Fast,
    tjg.

Timothy Grant                         tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.         <><       (503) 246-3630
>>>>>>>>>>>>>Linux, because rebooting is *NOT* normal<<<<<<<<<
>>>>This machine was last rebooted:  58 days 23:33 hours ago<<



From db3l at fitlinxx.com  Thu Mar  8 04:20:04 2001
From: db3l at fitlinxx.com (David Bolen)
Date: 08 Mar 2001 04:20:04 -0500
Subject: Name of the Script
References: <20010304.185354.2114937732.8277@localhost.localdomain> <NJCo6.6720$qv3.2834377@nnrp5-w.sbc.net> <FpmjOjvwe5bm0Bg09LDTMPhh5vji@4ax.com> <Pine.LNX.4.21.0103051715150.18263-100000@kev.ioc.ee> <cbKjOqPRwNza00jOm7A1HBDkRYj7@4ax.com> <upuft9z0s.fsf@ctwd0143.fitlinxx.com> <J1xp6.6661$Xt3.1009487@news1.rdc1.az.home.com>
Message-ID: <ug0gor517.fsf@ctwd0143.fitlinxx.com>

"Tim Hochberg" <tim.hochberg at ieee.org> writes:

> Another approach, if __file__ is what you want is to have a module import
> itself and grab the __file__ from the imported copy. Here's a (contrived)
> example:
> 
> # file test.py
> if __name__ == "__main__":
>     import test
>     print `test.__file__`
> 
> Unfortunately, this whole __file__ approach fails when using files from an
> archive (such as produced with Gordon McMillan's installer), so this isn't
> always useful.

And isn't this risky in general since it depends on the fact that the
module itself is going to be somewhere properly on sys.path?  I've had
cases where the main script was in a system bin location that wasn't
on sys.path since it only contained top level scripts meant to be
directly executed.

I think I also had some issues with this approach when running a
module that resided within a package directory, but my memory is a bit
vague.

At a minimum you'd want to protect the import with a try/except block
to catch a possible ImportError and have some fallback scenario in
that case.

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


From emile at fenx.com  Sat Mar 24 19:52:23 2001
From: emile at fenx.com (Emile van Sebille)
Date: Sat, 24 Mar 2001 16:52:23 -0800
Subject: time to invest ???
References: <3abd3a17.7103358@news.vif.com>
Message-ID: <99jfmh$1i58c$1@ID-11957.news.dfncis.de>

Pick a simple admin task, something you do regularly and feel you
understand, work the tutorial, and then implement in python.

Two ideal days.  If it takes a week, you've started with too big a project.
Stop at two days and ask for help/clarification.

--

Emile van Sebille
emile at fenx.com

---------
"Yvon Boulianne" <mysticsguy at yahoo.ca> wrote in message
news:3abd3a17.7103358 at news.vif.com...
> Hello, i start to learn Python some day ago, with lot of knowledge as
> a system admin but little as programmer and i wonder what is the
> average time/hours someone have to invest to be able to use a language
> like that easily.  i talk as well about devlopping the way of thinking
> of a programmer and the knowledge of the language. please don't tell
> me theres no average, at least take the idea of a normal guy, not a
> genius and not stupid  ;-)
>
> Thanks verry much and sorry for my english, my first language is
> french.
>
> Yvon Boulianne ; Saphi
> mysticsguy at yahoo.ca
> www.monunivers.net
> Humour,Paradox & Changements




From tim.one at home.com  Fri Mar  2 14:18:20 2001
From: tim.one at home.com (Tim Peters)
Date: Fri, 2 Mar 2001 14:18:20 -0500
Subject: create new statement?
In-Reply-To: <3A9FE7C2.1F5525D2@troikanetworks.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCIEHLJDAA.tim.one@home.com>

[Bruce Edge]
> Can you add a new statement to python?

Yes, but you probably can't <wink>.  That is, the set of statements is fixed,
and to add a new statement you would need to change the grammar of the
language, rebuild the parse tables, and alter the compiler to generate
appropriate code for it.

> ...
> I want to add my own command to the interp, but not have to use object
> calling syntax
> eg:
>     create obj opt
> not
>     create(obj, opt)
>
> How do I define new statements, or, a way of calling functions using the
> former syntax?

You don't, in Python.  "It's a feature" that everyone's code looks the same
here.  You can do that in Perl, though.  Without changing the language
implementation, the closest you could get in Python would be to write a
source-to-source (or source-to-bytecode) translator of your own, that output
legitimate Python source (or bytecode).

python's-syntax-is-neither-extensible-nor-overridable-ly y'rs  - tim




From DanielK at jBASE.com  Fri Mar 16 16:34:54 2001
From: DanielK at jBASE.com (Daniel Klein)
Date: Fri, 16 Mar 2001 13:34:54 -0800
Subject: Python creator: Perl users are moving to Python  
Message-ID: <2Gvs6.11725$a3.27986@typhoon.aracnet.com>

http://searchdatabase.techtarget.com/searchDatabase_News_Item/0,281943,53162
9,00.html

I apologize if this has already been posted to the group.

Dan




From thys at klaribel.netsys.co.za  Fri Mar 23 12:28:45 2001
From: thys at klaribel.netsys.co.za (Thys)
Date: Fri, 23 Mar 2001 19:28:45 +0200
Subject: wxImageList use in wxPanel
Message-ID: <01032319371802.09619@klaribel.feetjieland>

Hi All,

I need to display a few images on several places in a wxPanel, the obvious
soluttion would be to use a wxImageList, especially as the panel changes over
time (a status screen of sorts). 

I cant seem to get it to work though, no compilation problems but the graphics
refuse to be displayed. The documentation seems to concentrate on wxListCtrl
and wxtreeCtrl none of which is applicable in this case.

appreciated as always

-- 
Regards

Thys

--------------------
Thys Meintjes
thys at netsys.co.za

Netsys International(Pty) Ltd
http://www.netsys.co.za
+27 12 348-4246




From kens at sightreader.com  Tue Mar 20 13:53:16 2001
From: kens at sightreader.com (Ken Seehof)
Date: Tue, 20 Mar 2001 10:53:16 -0800
Subject: re.match
References: <997skf$qoc$1@bmerhc5e.ca.nortel.com> <TrMt6.2490$4N4.318602@newsc.telia.net>
Message-ID: <005901c0b16f$068d1790$04090a0a@upcast.com>

re.match tests whether or not a string -matches- a regular expression.
re.search -searches- a string for the first occurrence of a regular
expression.

"Is this a dead parrot?" vs "Does anyone know where my parrot is?"

----- Original Message -----
From: "Fredrik Lundh" <fredrik at pythonware.com>
Newsgroups: comp.lang.python
To: <python-list at python.org>
Sent: Tuesday, March 20, 2001 9:29 AM
Subject: Re: re.match


> Michael P. Soulier wrote:
> >     Hey people. Could someone explain the purpose of the match function
in the
> > re module? If it's only difference from search is to match at the
beginning of
> > the line, why does it exist? That's what the ^ in the regexp is for.
>
> no, it's not.  read the documentation again.
>
> Cheers /F
>
>
> --
> http://mail.python.org/mailman/listinfo/python-list
>




From mwh21 at cam.ac.uk  Wed Mar 14 03:47:19 2001
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 14 Mar 2001 08:47:19 +0000
Subject: Console column Width
References: <hhAr6.3889$L7.27822@news1.sshe1.sk.home.com>
Message-ID: <m3snkg7n54.fsf@atrus.jesus.cam.ac.uk>

"Nathan Heagy" <gervaserules at yahoo.com> writes:

> I can't figure out how to learn the (linux) console's column width from
> within a script? That is, without using curses.

This works for me:

def getwidth():
    return getheightwidth()[1]

# Eep!  this next should be made rather more portable (ie. fallback to
# env vars, then tigetstr("lines"), etc).
def getheightwidth():
    height, width = struct.unpack(
        "hhhh", ioctl(0, TERMIOS.TIOCGWINSZ ,"\000"*8))[0:2]
    return height, width

Cheers,
M.

-- 
  Not only does the English Language borrow words from other
  languages, it sometimes chases them down dark alleys, hits 
  them over the head, and goes through their pockets.   -- Eddy Peters


From topmind at technologist.com  Thu Mar  1 22:01:00 2001
From: topmind at technologist.com (Topmind)
Date: Fri, 02 Mar 2001 03:01:00 GMT
Subject: Collection interfaces
References: <3a90fa97.0@news1.mweb.co.za> <3A933830.7F98D118@ix.netcom.com> <CgJk6.515$v77.66568@newsread2.prod.itd.earthlink.net> <220220010813598042%cmh@bDistributed.com> <ud7cakvi8.fsf@roundpoint.com> <MPG.150312ceb68f1ada98969f@news.earthlink.net> <8Gem6.473353$U46.14105062@news1.sttls1.wa.home.com> <MPG.1503548620e632919896a4@news.earthlink.net> <MCjm6.474652$U46.14163288@news1.sttls1.wa.home.com> <MPG.150394195f7f04a19896a7@news.earthlink.net> <3A9A86B1.9A8C69A2@ix.netcom.com> <MPG.1504d14563dce6499896af@news.earthlink.net> <97fjla128ir@news2.newsguy.com> <MPG.1505a76831c34b749896b3@news.earthlink.net> <97h5s40vhc@news2.newsguy.com> <MPG.1506007591cf818b9896bc@news.earthlink.net> <97ifd60sen@news1.newsguy.com> <MPG.1506c16edd0cb6159896bf@news.earthlink.net> <3A9EEF93.4B18FCA@cosc.canterbury.ac.nz>
Message-ID: <MPG.1508af2fe4fa17149896cb@news.earthlink.net>

> Topmind wrote:
> > 
> > There is no polymorphism if there is
> > only one "type" with a bunch of independant attributes -
> > HAS-A reigns and sub-IS-A is rare.
> 
> Polymorphism is not inherently dependent on any 
> kind of inheritance or classification system.
> In Python, for instance, most functions are quite
> promiscuous and will happily operate on any
> objects which have the right methods, without there
> having to be any inheritance relationship between
> them.
> 
> -- 
> Greg Ewing, Computer Science Dept, University of Canterbury,	  
> Christchurch, New Zealand
> To get my email address, please visit my web page:	  
> http://www.cosc.canterbury.ac.nz/~greg
> 

I kind of had (having) this discussion with Jay recently
WRT Smalltalk. I would
give you a message ID but the designers of usenet were
drunk when they designed that part.

-tmind-


From donn at oz.net  Fri Mar 23 01:46:38 2001
From: donn at oz.net (Donn Cave)
Date: 23 Mar 2001 06:46:38 GMT
Subject: Multiple Processes on Unix with Python
References: <3ABA8919.835D8BED@centropolisfx.com> <99eic9$i06$0@216.39.151.169> <3ABADF0E.A1F9E93A@centropolisfx.com>
Message-ID: <99erge$bj5$0@216.39.151.169>

Quoth Nickson Fong <nickson at centropolisfx.com>:
| this is helpful - what i really want to do is>
| for example: i have a 2 processor irix machine and i want to launch 2
| processes. 
|
| proc #1 = render frames 1-10
| proc #2 = render frames 11-20
|
|
| os.fork os.execv os.wait are things that i have been reading.....
|
| question: why do you os.spawnv?

Just to be different.  Seriously, I have always thought that a
fork+exec function would be a worthwhile convenience, and here
it is.  You can see for yourself in os.py, it's no more than
a fork and exec, plus waitpid depending on the option, but
notice that it has better than average exception handling.
And it's actually a Windows function, so it's a little more
cross platform portable than fork & exec.

Really I think the big payoff is to use it instead of system(),
with generated arguments.  Like,
   os.spawnv(os.P_WAIT, '/usr/bin/lpr', ('lpr', '-P', printer, file))
instead of
   os.system('lpr -P %s %s' % (printer, file))

If the input is bad (printer or file), the first is much less
scary than the second.

	Donn Cave, donn at oz.net


From jkahanpa at pcu.helsinki.fi.invalid  Tue Mar 13 06:48:51 2001
From: jkahanpa at pcu.helsinki.fi.invalid (Jere Kahanpaa)
Date: 13 Mar 2001 11:48:51 GMT
Subject: What's wrong with me/Python ?
References: <3AAE0652.D68F253E@daimlerchrysler.com>
Message-ID: <98l1f3$fnn$1@oravannahka.helsinki.fi>

Alexander Stirzel <alexander.stirzel at daimlerchrysler.com> wrote:
: 8<----------- error message -----------------------------
: Traceback (most recent call last):
:   File "/usr/sbin/netscape-killer.py", line 11, in ?
:     splitted = string.split(line)
: NameError: There is no variable named 'string'
: 8<----------- error message -----------------------------

: Now here's the script:

: 8<------------ script -----------------------------------
: #!/usr/local/bin/python
: import os
: import time

Add 'import string'. The 'string' module is not automatically loaded. You 
can also import several modules on a single line:

import os, time, string 

Happy Hacking
	Jere
-- 
Lord, make my words as sweet as honey, for one day I may have to eat them
								
							Daryl Benson


From tim.one at home.com  Thu Mar  1 15:58:41 2001
From: tim.one at home.com (Tim Peters)
Date: Thu, 1 Mar 2001 15:58:41 -0500
Subject: A counter-counter-proposal for PEP 236: #pragma( nested_scopes)
In-Reply-To: <15006.46745.206128.996289@w221.z064000254.bwi-md.dsl.cnc.net>
Message-ID: <LNBBLJKPBEHFEDALKOLCEECDJDAA.tim.one@home.com>

[Jeremy Hylton]
> ...
> A word of clarification: __future__ isn't about experimental
> features.  Instead it is for bleeding edge users to use a new feature
> in release X even though the feature won't be standard until release
> Y, where X < Y.
>
> If we want to do experimental stuff, we should probably do something
> like:
>
> from __possible_world__ import __future__
> from __future__ import ...

Hmm.  Maybe that's why I so dislike Martin's attempt to hijack PEP 236:  his
current formulation of an alternative leaves only the raw names, without the
__future__ or __possible_world__ qualifiers.  We have a clear taxonomy in
mind when we're doing this stuff, but that's lost in an all-purpose
"directive" statement.  So, over time, only the experts who introduced
various directive gimmicks will remember where their particular raw names fit
in the grander scheme.

Back to your point <wink>, I expect that until release Y happens, it probably
*is* most correct to view a __future__ feature as experimental.  Indeed,
that's why PEP 236 says a MandatoryRelease value may become None (meaning
that the feature was withdrawn).  But it's experimental in the sense that
while it *may* be withdrawn, the presumption is that it won't be -- it's not
just random "trying things out" (which is how I suspect you read /F's use of
the word, and for which __possible_world__ (whatever) would be more
appropriate).

words-words-and-more-words-ly y'rs  - tim




From thomas at cintra.no  Tue Mar  6 09:11:57 2001
From: thomas at cintra.no (Thomas Weholt)
Date: Tue, 6 Mar 2001 15:11:57 +0100
Subject: [ Cry for help ] Please bugtest my full-text indexer
Message-ID: <Pe6p6.8119$t21.206104@news3.oke.nextra.no>

Hi,

Just finished my own full-text indexer which is easy to use and seem to be
quite fast. BUT, when tested against a lot of data, my archive of RFCs, the
thing croaked after inserting approx. 500.000 keys ( size of data stored was
~40 MB ), giving me an error-message from the bsddb-module, (0,'error').

I'm terribly sorry for posting this, with attachments and all, but I need
this one to work with HUGE amounts of data. Please let me know if there are
anything I can do.

The module works fine for smaller amounts of data and it's the easiest
full-text indexer I've seen in Python so far ( ok, so I'm biased ;-> ) and
it could be useful for other folks here too.


The module is located here
http://download.sourceforge.net/snipthis/hsindexer.zip

Best regards,
Thomas Weholt




From dnew at san.rr.com  Thu Mar 15 18:15:28 2001
From: dnew at san.rr.com (Darren New)
Date: Thu, 15 Mar 2001 23:15:28 GMT
Subject: I come to praise .join, not to bury it...
References: <mailman.983485923.1136.python-list@python.org> <97p7iq$kk2$1@nntp6.u.washington.edu> <97rka002k9@news1.newsguy.com> <3AAF7E47.294617E1@sage.att.com> <98qeco02hn7@news1.newsguy.com> <3AB106CD.7D259EBC@san.rr.com> <mailman.984685357.7864.python-list@python.org>
Message-ID: <3AB14D10.7F3A31F@san.rr.com>

> Hmm ... I wonder what I'd do if I could overload __if__ ...

That's how Smalltalk works. It has no "if" statement. There's an "IfElse"
method that takes two blocks. True evaluates its first argument, False its
second, in a polymorphic way. So

bool if: [yes yes] else: [no no]

doesn't contain any branching code. Just a polymorphic dispatch. Very
strange to wrap your head around. :-)

(Of course, in reality, it branches for efficiency, recognised by the
compiler, but in theory...)

-- 
Darren New / Senior MTS & Free Radical / Invisible Worlds Inc.
San Diego, CA, USA (PST).  Cryptokeys on demand.


From ransen_spam_me_not at nemo.it  Sat Mar  3 06:12:29 2001
From: ransen_spam_me_not at nemo.it (Owen F. Ransen)
Date: Sat, 03 Mar 2001 11:12:29 GMT
Subject: PyRun_SimpleFile - impossible to call it under Windows?
References: <3aa07704.2672581@news.newsguy.com>
Message-ID: <3aa0ccdf.4708595@news.newsguy.com>

On Fri, 02 Mar 2001 16:04:54 GMT, ransen_spam_me_not at nemo.it (Owen F.
Ransen) wrote:

>I've seen that there is an odd bug in PyRun_SimpleFile
>probably because of version incompatibilities in the
>FILE structure and using DEBUG modes in VC6
>
>I'd like to call PyRun_SimpleFile in the debug version
>of my VC6 program, is it impossible currently with Python 2?
>TIA
>Owen

Following up my own message, is a possibility that I
recompile and create my Python lib using the
sources of 2.1b1 ? 

I'd have to link statically to make sure I do not
interfere with outher embedders extenders...

?


--
Owen F. Ransen
http://www.ransen.com/
Home of Gliftic & Repligator Image Generators


From aahz at panix.com  Mon Mar 19 18:51:18 2001
From: aahz at panix.com (Aahz Maruch)
Date: 19 Mar 2001 15:51:18 -0800
Subject: Wrong reasons to enhancing the language
References: <mailman.984971990.29857.python-list@python.org> <9960i1$tp0$1@nntp6.u.washington.edu>
Message-ID: <99661m$scc$1@panix6.panix.com>

In article <9960i1$tp0$1 at nntp6.u.washington.edu>,
Russell E. Owen <owen at astrono.junkwashington.emu> wrote:
>
>I believe many of the improvements in Python 2.0 fall into the category 
>of helpful improvements. I am more doubtful about some of the stuff 
>coming up (including nested scopes, // and rational numbers), but we'll 
>see how it works out.

None of those are Python 2.0 improvements.  Nested scopes are an
optional feature of Python 2.1 and the others are simply *proposed*
features.  This is a *great* time to be debating the numerics issues.
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

There's a difference between a person who gets shit zie doesn't deserve
and a person who gets more shit than zie deserves.  --Aahz


From qrczak at knm.org.pl  Sat Mar 24 10:21:48 2001
From: qrczak at knm.org.pl (Marcin 'Qrczak' Kowalczyk)
Date: 24 Mar 2001 15:21:48 GMT
Subject: python, perl, lisp, tcl / got a decision to make, can you give me a view point please ? and a joke for python users :)
References: <3aba3076.14395287@news.vif.com>
Message-ID: <slrn9bpesb.bs7.qrczak@qrnik.zagroda>

Ok, it seems that you can't stand Perl's success and everybody here
must from time to time say that Perl is evil :-)

So to make you laugh, here is my favorite Perl silliness
(perl-5.6.0/toke.c). Enjoy the algorithm of disambiguation of the
last case!

/* S_intuit_more
 * Returns TRUE if there's more to the expression (e.g., a subscript),
 * FALSE otherwise.
 *
 * It deals with "$foo[3]" and /$foo[3]/ and /$foo[0123456789$]+/
 *
 * ->[ and ->{ return TRUE
 * { and [ outside a pattern are always subscripts, so return TRUE
 * if we're outside a pattern and it's not { or [, then return FALSE
 * if we're in a pattern and the first char is a {
 *   {4,5} (any digits around the comma) returns FALSE
 * if we're in a pattern and the first char is a [
 *   [] returns FALSE
 *   [SOMETHING] has a funky algorithm to decide whether it's a
 *      character class or not.  It has to deal with things like
 *      /$foo[-3]/ and /$foo[$bar]/ as well as /$foo[$\d]+/
 * anything else returns TRUE
 */

/* This is the one truly awful dwimmer necessary to conflate C and sed. */

STATIC int
S_intuit_more(pTHX_ register char *s)
{
    if (PL_lex_brackets)
	return TRUE;
    if (*s == '-' && s[1] == '>' && (s[2] == '[' || s[2] == '{'))
	return TRUE;
    if (*s != '{' && *s != '[')
	return FALSE;
    if (!PL_lex_inpat)
	return TRUE;

    /* In a pattern, so maybe we have {n,m}. */
    if (*s == '{') {
	s++;
	if (!isDIGIT(*s))
	    return TRUE;
	while (isDIGIT(*s))
	    s++;
	if (*s == ',')
	    s++;
	while (isDIGIT(*s))
	    s++;
	if (*s == '}')
	    return FALSE;
	return TRUE;
	
    }

    /* On the other hand, maybe we have a character class */

    s++;
    if (*s == ']' || *s == '^')
	return FALSE;
    else {
        /* this is terrifying, and it works */
	int weight = 2;		/* let's weigh the evidence */
	char seen[256];
	unsigned char un_char = 255, last_un_char;
	char *send = strchr(s,']');
	char tmpbuf[sizeof PL_tokenbuf * 4];

	if (!send)		/* has to be an expression */
	    return TRUE;

	Zero(seen,256,char);
	if (*s == '$')
	    weight -= 3;
	else if (isDIGIT(*s)) {
	    if (s[1] != ']') {
		if (isDIGIT(s[1]) && s[2] == ']')
		    weight -= 10;
	    }
	    else
		weight -= 100;
	}
	for (; s < send; s++) {
	    last_un_char = un_char;
	    un_char = (unsigned char)*s;
	    switch (*s) {
	    case '@':
	    case '&':
	    case '$':
		weight -= seen[un_char] * 10;
		if (isALNUM_lazy_if(s+1,UTF)) {
		    scan_ident(s, send, tmpbuf, sizeof tmpbuf, FALSE);
		    if ((int)strlen(tmpbuf) > 1 && gv_fetchpv(tmpbuf,FALSE, SVt_PV))
			weight -= 100;
		    else
			weight -= 10;
		}
		else if (*s == '$' && s[1] &&
		  strchr("[#!%*<>()-=",s[1])) {
		    if (/*{*/ strchr("])} =",s[2]))
			weight -= 10;
		    else
			weight -= 1;
		}
		break;
	    case '\\':
		un_char = 254;
		if (s[1]) {
		    if (strchr("wds]",s[1]))
			weight += 100;
		    else if (seen['\''] || seen['"'])
			weight += 1;
		    else if (strchr("rnftbxcav",s[1]))
			weight += 40;
		    else if (isDIGIT(s[1])) {
			weight += 40;
			while (s[1] && isDIGIT(s[1]))
			    s++;
		    }
		}
		else
		    weight += 100;
		break;
	    case '-':
		if (s[1] == '\\')
		    weight += 50;
		if (strchr("aA01! ",last_un_char))
		    weight += 30;
		if (strchr("zZ79~",s[1]))
		    weight += 30;
		if (last_un_char == 255 && (isDIGIT(s[1]) || s[1] == '$'))
		    weight -= 5;	/* cope with negative subscript */
		break;
	    default:
		if (!isALNUM(last_un_char) && !strchr("$@&",last_un_char) &&
			isALPHA(*s) && s[1] && isALPHA(s[1])) {
		    char *d = tmpbuf;
		    while (isALPHA(*s))
			*d++ = *s++;
		    *d = '\0';
		    if (keyword(tmpbuf, d - tmpbuf))
			weight -= 150;
		}
		if (un_char == last_un_char + 1)
		    weight += 5;
		weight -= seen[un_char];
		break;
	    }
	    seen[un_char]++;
	}
	if (weight >= 0)	/* probably a character class */
	    return FALSE;
    }

    return TRUE;
}

-- 
 __("<  Marcin Kowalczyk * qrczak at knm.org.pl http://qrczak.ids.net.pl/
 \__/
  ^^                      SYGNATURA ZAST?PCZA
QRCZAK


From mcfletch at home.com  Fri Mar  9 10:03:04 2001
From: mcfletch at home.com (Mike C. Fletcher)
Date: Fri, 9 Mar 2001 10:03:04 -0500
Subject: Can anyone recomend a good intoduction to C...
In-Reply-To: <984132875.222868@newsmaster-04.atnet.at>
Message-ID: <000001c0a8aa$0b9143b0$a9807018@cr706570a>

Perl-ism, http://www.drbbs.com/jsw/jargon/jargon_35.html#TAG1944

Enjoy,
Mike

-----Original Message-----
From: Werner Schiendl [mailto:ws-news at gmx.at]
Sent: Friday, March 09, 2001 5:10 AM
To: python-list at python.org
Subject: Re: Can anyone recomend a good intoduction to C...


What's TMTOWTDI, please?


-- 
http://mail.python.org/mailman/listinfo/python-list




From emile at fenx.com  Sat Mar 31 16:57:35 2001
From: emile at fenx.com (Emile van Sebille)
Date: Sat, 31 Mar 2001 13:57:35 -0800
Subject: list vs tuple
References: <mailman.986066384.30712.python-list@python.org> <d1rx6.1053$p5.4024@news1.rivrw1.nsw.optushome.com.au> <slrn9cciaj.3ug.qrczak@qrnik.zagroda> <lWrx6.1057$p5.4152@news1.rivrw1.nsw.optushome.com.au>
Message-ID: <9a5k2e$3mk79$1@ID-11957.news.dfncis.de>

"""Grant me the serenity to accept the things I cannot change, the courage
to change the things I can, and, above all, grant me the wisdom to know the
difference."""

--

Emile van Sebille
emile at fenx.com

---------
"deadmeat" <root@[127.0.0.1]> wrote in message
news:lWrx6.1057$p5.4152 at news1.rivrw1.nsw.optushome.com.au...
> > THERE IS NO DIFFERENCE.
>
> What happens to the objects is not the point, what happens to what I put
> into a and b is the point.
>
> >>> a = 1
a points somewhere
> >>> b = a
b points to the same place
> >>> a = 2
a now points elsewhere
> >>> b
what else would you expect?
> 1
>

> >>> a = [1,2,3]
a points somewhere
> >>> b = a
b points to the same place
> >>> a[0] = 0
the thing a points to is changed
> >>> b
what else would you expect?
> [0, 2, 3]
>


> Are you saying there is no difference (IN THE *RESULT*) between these?
Are
> you able to read?
>
> What goes on in the background is NOT THE POINT. I only care about the
> values I put into them, and what happens after b = a.
>
>
>
>




From whisper at oz.net  Mon Mar 26 13:41:18 2001
From: whisper at oz.net (Dave LeBlanc)
Date: 26 Mar 2001 18:41:18 GMT
Subject: daemon app in win32
References: <99nk49$a2q$02$1@news.t-online.com> <mailman.985619314.10299.python-list@python.org> <99nsn3$q39$06$1@news.t-online.com>
Message-ID: <99o2ge$9on$0@216.39.170.247>

AFAIK, Every Windows app gets sent a WM message when it's about to be
terminated. Most apps just let this fall through to the default
message processor (which you have to do anyway) without doiing
anything to clean up. You can add a message handler for this such that
the app can gracefully (make out it's will?) prepare to end it all.
Just make sure that in the end, it passes the message on to the
default message handler so that windows can do it's own cleanup of
your app.

Regards,

Dave LeBlanc

On Mon, 26 Mar 2001 19:02:27 +0200, Michael Bauer <Mike_B at T-Online.de>
wrote:

>Steve Purcell wrote:
>
>> Michael Bauer wrote:
>>> i am writing a daemon application that runs both under windows and unix.
>>> Since i am not very experienced in win32 - programming, i really would
>>> appreciate some hints how to register my app in win32 that i can do some
>>> cleanup-functions when windows is shut down.
>>> 
>>> At the moment it is just a standard console application that simply gets
>>> killed when windows shuts down. The app does not have/need a GUI, it
>>> should just run in the background.
>> 
>> 
>> If you're using NT, take a look at the win32service API. I think it
>> provides hooks like SvcStop that Windows will call for you (if you ask it
>> nicely).
>
>Hmm, seems like this is only available under NT. Is there a more general 
>way to accomplish this task independently of the underlying Windows-version?
>
>But maybe i should just get a life and write different wrappers for Win95, 
>Win98, WinME, Win200, WinNT... :-(
> 
>still hoping... Mike



From Karl.Bellve at umassmed.edu  Fri Mar 16 11:29:06 2001
From: Karl.Bellve at umassmed.edu (Karl Bellve)
Date: Fri, 16 Mar 2001 11:29:06 -0500
Subject: C++ and Python
Message-ID: <3AB23F52.FB2E087C@umassmed.edu>


I have a class that will use Python and Numpy. There could be several
instances of this class, each using and initializing Python.

The problem I am having is with PyMethoDef which passes a table of
methods and their descriptions to Python. As far as I can tell, this
can't be a static structure. Anyone build a PyMethodDef structure inside
a C++ class so the appropiate function pointers point to the appropiate
instance of the class?

Otherwise...I am stuck using static variables and functions inside a C++
class?


-- 
Cheers,



Karl Bellve		             ICQ # 13956200
Biomedical Imaging Group             TLCA# 7938 		
University of Massachusetts
Email: Karl.Bellve at umassmed.edu
Phone: (508) 856-6514
Fax:   (508) 856-1840
PGP Public key: finger kdb at molmed.umassmed.edu



From hernan.foffani at iname.com  Fri Mar 16 14:18:54 2001
From: hernan.foffani at iname.com (hernan.foffani at iname.com)
Date: 16 Mar 2001 19:18:54 GMT
Subject: Extending Python: rewriting a single method in C
References: <g0wv9vxms0.fsf@scumbag.ecs.soton.ac.uk> <slrn9apn8f.382.jepler@potty.housenet> <g0ofv6xbj4.fsf@scumbag.ecs.soton.ac.uk> <98l1i901k4u@news1.newsguy.com> <g0lmq9xxz0.fsf@scumbag.ecs.soton.ac.uk> <98m6g20fgv@news1.newsguy.com> <g0g0ggy7ct.fsf@scumbag.ecs.soton.ac.uk> <98ntl80vrp@news2.newsguy.com>
Message-ID: <20010316141854.145$sP@newsreader.com>

"Alex Martelli" <aleaxit at yahoo.com> wrote:
>...
> With the HTML-Help form of the docs (I forget the helpful person
> from whose site I downloaded that from), searching throughout the
> docs is instantaneous -- which roughly makes up for defects in
> the index and/or organizational problems in the docs themselves.

You can find it here:
http://alldunn.com/python/
or here:
http://www.orgmf.com.ar/condor/pytstuff.html

regards,
-Hernan

-- 
-------------------- http://NewsReader.Com/ --------------------
                       Usenet for the Web


From kens at sightreader.com  Sat Mar  3 16:34:18 2001
From: kens at sightreader.com (Ken Seehof)
Date: Sat, 3 Mar 2001 13:34:18 -0800
Subject: IPC9 hotel room share
Message-ID: <007b01c0a429$b6cb8720$a325fea9@his>

If you are interested in saving money by sharing a room at the conference, call me.

(650) 575-9186 (cell, any time)
(408) 353-5670 (if the cell phone doesn't work)

Or email me before 6:00 today, March 3.

I have a res at Holiday Inn (2.5 km from IPC9) which can be cancelled by 6pm.  Your half is about $50 including tax.  I'd also consider splitting a room at the Hilton if you already have a reservation there.

- Ken

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20010303/bfbf0015/attachment.html>

From bhk at dsl.co.uk  Wed Mar 14 20:56:46 2001
From: bhk at dsl.co.uk (Brian {Hamilton Kelly})
Date: Thu, 15 Mar 2001 01:56:46 GMT
Subject: Einstein's Riddle
References: <984587984snz@dsl.co.uk>
Message-ID: <984621406snz@dsl.co.uk>

In article <984587984snz at dsl.co.uk>
           bhk at dsl.co.uk "Brian {Hamilton Kelly}" writes:

> "No fair", starting a troll by cross-posting to EIGHT groups.  One's
> supposed to limit the initial posting to no more than eight groups.
                                                        ^^^^^
                                                        
s/eight/six/

Sigh.

-- 
Brian {Hamilton Kelly}                                          bhk at dsl.co.uk
    "We have gone from a world of concentrated knowledge and wisdom to one of
    distributed ignorance.  And we know and understand less while being incr-
    easingly capable."                          Prof. Peter Cochrane, BT Labs



From stephen_purcell at yahoo.com  Fri Mar  9 05:21:17 2001
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Fri, 9 Mar 2001 11:21:17 +0100
Subject: OODB question
In-Reply-To: <WM1q6.7817$Z%5.6737@NewsReader>; from arinagour@yahoo.co.uk on Fri, Mar 09, 2001 at 09:59:16AM -0000
References: <WM1q6.7817$Z%5.6737@NewsReader>
Message-ID: <20010309112117.A12238@freedom.puma-ag.com>

chris lamb wrote:
> I am sure that I read somewhere of an OODB implementation developed
> specifically with Python or having a Python interface.
> 
> I am beginning to wonder if I dreamt it up since I cannot remember where I
> read about it!
> 
> Can anyone help me ? (I doubt it!)


You're probably thinking of ZODB. Take a look at http://www.amk.ca/zodb/

-Steve


-- 
Steve Purcell, Pythangelist
Get testing at http://pyunit.sourceforge.net/
Any opinions expressed herein are my own and not necessarily those of Yahoo



From akuchlin at mems-exchange.org  Tue Mar 20 18:31:31 2001
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 20 Mar 2001 18:31:31 -0500
Subject: ASCII encoding error: ordinal not in range(128)
References: <91o5ih$2kv$1@news.wrc.xerox.com> <20001219172933.B29727@ActiveState.com> <mailman.985123475.18421.python-list@python.org> <W3Rt6.2582$4N4.344041@newsc.telia.net>
Message-ID: <3d4rwo10h8.fsf@ute.cnri.reston.va.us>

"Fredrik Lundh" <fredrik at pythonware.com> writes:
> > UnicodeError: ASCII encoding error: ordinal not in range(128)
> 
> translated to english, this overly cute error message attempts
> to say something like "cannot convert this Unicode string to
> ASCII as it contains non-ASCII characters"

Hmm... hasn't changing this message before been discussed someplace?
Was there some reason it wasn't changed?  How about: "UnicodeError:
ASCII encoding error: character not in ASCII range 0-127" ?  (I
vaguely recall Marc-Andre arguing for "ordinal", since that's used in
the Unicode documents, but if it's less clear...)

--amk


From max at alcyone.com  Thu Mar 29 11:03:20 2001
From: max at alcyone.com (Erik Max Francis)
Date: Thu, 29 Mar 2001 08:03:20 -0800
Subject: Printing  "%"
References: <99vm4k$22l$1@ctb-nnrp2.saix.net>
Message-ID: <3AC35CC8.AD0B71E2@alcyone.com>

Mix wrote:

> How do you print a percenrage (%)sign in Python?
> I tried :
>   return "select * from fax where id LIKE '%s\%'" %ident

Use %% instead:

Python 2.0 (#5, Nov 10 2000, 21:47:15) 
[GCC egcs-2.91.66 19990314/Linux (egcs-1.1.2 release)] on linux2
Type "copyright", "credits" or "license" for more information.
>>> "We're at %f%%" % 200
"We're at 200.000000%"

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ People are taught to be racists.
\__/ Jose Abad
    Esperanto reference / http://mirror/alcyone/max/lang/esperanto/
 An Esperanto reference for English speakers.


From rambo at linuxsecurity.co.kr  Thu Mar  8 18:08:46 2001
From: rambo at linuxsecurity.co.kr (rambo)
Date: Thu, 8 Mar 2001 15:08:46 -0800
Subject: about environment variable.
Message-ID: <9877ho$kqn$1@news2-2.kornet.net>


I installed Tkinter ,as follows
rpm -Uvh Tkinter-2.0-8.0-3.i386.rpm

after that,
   == python prompt ===
1. >>>from Tkinter import *
2. >>> w=Button (text="hello" , command='exit' )
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
  File "/usr/local/lib/python2.0/lib-tk/Tkinter.py", line 1819, in __init__
    Widget.__init__(self, master, 'button', cnf, kw)
  File "/usr/local/lib/python2.0/lib-tk/Tkinter.py", line 1752, in __init__
    BaseWidget._setup(self, master, cnf)
  File "/usr/local/lib/python2.0/lib-tk/Tkinter.py", line 1727, in _setup
    _default_root = Tk()
  File "/usr/local/lib/python2.0/lib-tk/Tkinter.py", line 1482, in __init__
    self.tk = _tkinter.create(screenName, baseName, className)

after type 2th line , I got error message as above,
what environment variable  should I set  to fix above problem.

thanks for your time.







From sholden at holdenweb.com  Sat Mar  3 15:19:05 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Sat, 3 Mar 2001 15:19:05 -0500
Subject: gah! I hate the new string syntax
References: <mailman.983615170.23313.python-list@python.org> <tN5o6.2746$eH3.14913@NewsReader>
Message-ID: <Koco6.7518$1D5.244602@e420r-atl1.usenetserver.com>

"chris" <chrislamb at btinternet.com> wrote in message
news:tN5o6.2746$eH3.14913 at NewsReader...
> > at latest since the 19th century; in America, it appears this is
normally
> > introduced in the 6th or 7th grades (ages 11-12, for those of you in
> > inferior countries <wink>
>
> In Scotland that kind of stuff is pre-School! <wink-back>
>
> Chris

Ha! As an Englishman married to a Scot I'm not going to let you get away
with that one: this is classic Scottish braggadocio [I expect Italian
education form tne martellibot about that]. Brass neck, as we simple
Yorkshire folk would term it.

Still, at least we can remind all that Scotland and England are different
countries on c.l.py.

Lest people might think this is intended as criticism rather than humo(u)r,
I hasten to add that I lived in Scotland for three years before moving to
the USA, and found the Scots to be extraordinarily warm and generous people.

As I put it to a taxi-driver on Scotland recently who asked me how I got on
in Scotland "I find that once we get past 'Engl***h' and 'bas***d' we can
get on all right". He correctly pointed out that this was a racist remark,
but we got on fine and agreed that nobody was offended.

So, for the record, they don't get their mathmatics degrees until fourth
grade in Scotland.

coming-from-the socialist-people's-republic-of-yorkshire-and-therefore-
    not-quite-as-english-as-others-we-won't-mention-y'rs  - steve





From ajung-ml at andreas-jung.com  Wed Mar 28 08:13:29 2001
From: ajung-ml at andreas-jung.com (Andreas Jung)
Date: Wed, 28 Mar 2001 08:13:29 -0500
Subject: dynamic web pages?
Message-ID: <003001c0b788$e2c65d60$9865fea9@SUXLAP>

And we have the small gun: PMZ (Poor Men's Zope): http://pmz.sourceforge.net

Andreas
"Robert Kiendl" <rkiendl at gmx.net> schrieb im Newsbeitrag
news:3AC0B262.6F0E30DE at gmx.net...
> we have the BIG GUN here: Zope (www.zope.org) dynamic app.server in
> native python logic including an OODBMS!
>
> or use cgi programming with apache








From tuttledon at hotmail.com  Thu Mar 15 17:40:27 2001
From: tuttledon at hotmail.com (Don Tuttle)
Date: Thu, 15 Mar 2001 22:40:27 GMT
Subject: Python Programming on Win32 (was: Magnitude of the wx* market (was: Python In A Nutshell - suggestions))
References: <3AB16A79@MailAndNews.com>
Message-ID: <vxbs6.118341$Z8.23943431@typhoon.southeast.rr.com>

"Laura Lewin" <Laura_01 at MailAndNews.com> wrote in message
news:3AB16A79 at MailAndNews.com...
<snip>
> To try to address your other concerns (your mini rant), we are planning
> application-oriented books in the Python space, but these will be for
> developers.  I hear your concerns about books that cover Python apps, but
> ORA's main focus has always been for the developer, and that's where
you'll
> see most of our books.  I can't speak for all of the other
publishers/authors
> out there.

Hmm... I guess I didn't make myself clear.  I'm not really interested in
Python apps.  I'm interested in more specific topics on Python Development.
In short, don't make me learn VB, Perl, or C++ just so I can THEN figure out
how to do it in Python.  (This is a general statement to all publishers, not
just O'Reilly.)

The void in how to program in Python has been filled by several good books.
The basic language is no longer an issue.  How to do usefull tasks is. This
is where I have to go to current books written for other languages (And yes,
I just got my copy yesterday Programming Python 2E. ;-)

In Python Programming On Win32's case, it suffers from too much breadth, not
enough depth. It would be much better to chop the cookbook sections up into
a seperate series of books.  As for .NET, if the Common Language Runtime
becomes cross platform, you won't have to call this a Win32 series anymore!

> Please do check out The Python Cookbook
> (http://pythoncookbook.activestate.com/)though.  You mention an interest
in
> system administration, databases, network programming.  These are all
> sections > in the Cookbook.  Feel free to contribute recipes, or take any
with you!!

It's about time! <wink> I was complaining about a need for a Python Cookbook
here several months ago!

Don




From mcalla at home.com  Fri Mar 16 17:13:15 2001
From: mcalla at home.com (Mike Callahan)
Date: Fri, 16 Mar 2001 22:13:15 GMT
Subject: Catching exceptions in Tkinter's mainloop
Message-ID: <%dws6.9947$Q47.2723350@news1.rdc1.tn.home.com>

I want to make sure that if any exception pops up inside a mainloop, that
exception is printed to a log file and the mainloop quits. I tried to
simulate this with the following code but failed:

from Tkinter import *

class Gui:
    def __init__(self, master):
        self.master = master
        self.button = Button(master, text='Crash', command=self.crash)
        self.button.pack()

    def crash(self):
        z = zz  # trigger exception

root = Tk()
app = Gui(root)
try:
    root.mainloop()
except:
    print 'caught'
    root.quit()

It doesn't work. The exception is printed out by standard traceback and the
mainloop continues. What am I doing wrong?

Mike Callahan






From dsavitsk at e-coli.net  Mon Mar 12 18:26:15 2001
From: dsavitsk at e-coli.net (dsavitsk)
Date: Mon, 12 Mar 2001 17:26:15 -0600
Subject: tkinter entry index
References: <iHcr6.27192$tV.5424761@newsrump.sjc.telocity.net> <SJcr6.17936$Qb7.3120517@newsb.telia.net>
Message-ID: <vWcr6.27211$tV.5434235@newsrump.sjc.telocity.net>

thanks.

"Fredrik Lundh" <fredrik at pythonware.com> wrote in message
news:SJcr6.17936$Qb7.3120517 at newsb.telia.net...
> "dsavitsk" wrote:
> > how can i get the position of the cursor in an entry widget? .index and
> > .icursor don't seem to it. That is, if "foo" is in an entry, and the
cursor
> > is between the "o" and the other "o", I could get a return of 2.
>
>     position = w.index(INSERT)
>
> Cheers /F
>
> <!-- pythonworks 1.2 coming soon:
> http://www.pythonware.com/products/works/index.htm
> -->
>
>




From l_hill at mindspring.com  Mon Mar  5 22:32:54 2001
From: l_hill at mindspring.com (Laura Hill)
Date: Mon, 05 Mar 2001 19:32:54 -0800
Subject: OOPSLA 2001 call for participation
Message-ID: <3AA45A66.8A4B07F9@mindspring.com>

The OOPSLA 2001 submission system is now active at:
oopsla.acm.org

**************************************************************
OOPSLA 2001 Call for Participation

OOPSLA 2001
October 14-18, 2001
Tampa Bay Convention Center
Tampa Bay, Florida USA

The ACM OOPSLA Conference (Object-Oriented Programming,
Systems, Languages, and Applications) is the world's
premier forum for object-oriented technology.
If you're involved in object technology, OOPSLA 2001
is the place to be.

Submissions are solicited in the following categories:

Technical Papers,  Tutorials,  Practitioner Reports,
Panels, Demos,  Educators' Symposium,  Workshops,  Posters,
Doctoral Symposium,  DesignFest,  Vendor Exhibits

For more info, check out our Web site:
    http://oopsla.acm.org

Important Dates
March 23, 2001:  Due date for Technical Papers,
    Practitioner Reports, Educators' Symposium,
    Tutorials, Workshops, and Panel proposals,
    and DesignFest problems
July 20, 2001:  Due date for Posters, Demonstrations,
    Doctoral Symposium, and Student Volunteers






From hinsen at cnrs-orleans.fr  Fri Mar 30 05:23:39 2001
From: hinsen at cnrs-orleans.fr (Konrad Hinsen)
Date: 30 Mar 2001 12:23:39 +0200
Subject: Gaussian line profile using numeric python
References: <3ac31698@zfree.co.nz> <m3puf0re81.fsf@chinon.cnrs-orleans.fr> <Pine.SOL.4.30.0103291850410.13267-100000@primrose.csv.warwick.ac.uk>
Message-ID: <m3lmpn7dxw.fsf@chinon.cnrs-orleans.fr>

"John J. Lee" <phrxy at csv.warwick.ac.uk> writes:

> > 2) If you work with large data sets, you ought to use FFTs to
> >    compute the convolution, that's O(N*log(N)) in the size of the
> >    data set, instead of O(N**2) for the straightforward method.
> >    Any good book on FFTs should explain how this works in detail.
> 
> but you probably don't need to know in detail, it boils down to
> 
> inv_FFT(FFT(data)*FFT(kernel))

Except for zero-padding in case of non-periodic signals. That's why
I always recommend to look at a detailed description before doing
this; some of my colleagues keep telling me that FFT techniques "don't
work", and I suspect this is the reason.
-- 
-------------------------------------------------------------------------------
Konrad Hinsen                            | E-Mail: hinsen at cnrs-orleans.fr
Centre de Biophysique Moleculaire (CNRS) | Tel.: +33-2.38.25.56.24
Rue Charles Sadron                       | Fax:  +33-2.38.63.15.17
45071 Orleans Cedex 2                    | Deutsch/Esperanto/English/
France                                   | Nederlands/Francais
-------------------------------------------------------------------------------


From mwh21 at cam.ac.uk  Mon Mar  5 14:49:20 2001
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 05 Mar 2001 19:49:20 +0000
Subject: Dr. Dobb's Python-URL! - weekly Python news and links (Mar  5)
References: <FFE41EE542E94336.B7A7E90E01745DB0.38760D529048CCC3@lp.airnews.net>
Message-ID: <m33dcsc7y7.fsf@atrus.jesus.cam.ac.uk>

Oleg Broytmann <phd at phd.russ.ru> writes:

>       Michael Hudson continues Andrew Kuchling's marvelous tradition
>       of summarizing action on the python-dev mailing list twice a
>       month.

FWIW, I'm planning on doing them every other week.  This isn't very
different from twice a month (not at all in Feb!), but for the sake of
accuracy...

Cheers,
M.

-- 
 The rapid establishment of social ties, even of a fleeting nature,
 advance not only that goal but its standing in the uberconscious 
 mesh of communal psychic, subjective, and algorithmic interbeing. 
 But I fear I'm restating the obvious.  -- Will Ware, comp.lang.python


From qrczak at knm.org.pl  Tue Mar 13 10:26:37 2001
From: qrczak at knm.org.pl (Marcin 'Qrczak' Kowalczyk)
Date: 13 Mar 2001 15:26:37 GMT
Subject: This math scares me
References: <3aad4062.1290021@News.CIS.DFN.DE> <20010312170356.B3513@better.net> <mailman.984480856.18097.python-list@python.org> <984486395.407379@newsmaster-04.atnet.at>
Message-ID: <slrn9asf1d.ja3.qrczak@qrnik.zagroda>

Tue, 13 Mar 2001 13:21:50 +0100, Werner Schiendl <ws-news at gmx.at> pisze:

> The most useful approach is to round the numbers to 15 digits for display,
> because this is the precision double values can really correctly represent.
> The drawback is, that the rounded value does not discover all the
> information actually stored in the variable.

The most useful approach is to display as much as is needed to read
the value back, but no more. I.e. repr(float("0.2")) == "0.2", and
there is no other float value whose repr is "0.2".

This doesn't solve the problem from the beginning of this thread,
because float("0.1") + float("0.2") needs not be equal to float("0.3").
But at least all short numbers entered at the prompt don't have those
unnecessary 00000s or 99999s, and conversion to a string and back
preserves the value.

-- 
 __("<  Marcin Kowalczyk * qrczak at knm.org.pl http://qrczak.ids.net.pl/
 \__/
  ^^                      SYGNATURA ZAST?PCZA
QRCZAK


From rkiendl at gmx.net  Tue Mar 27 09:28:45 2001
From: rkiendl at gmx.net (Robert Kiendl)
Date: Tue, 27 Mar 2001 16:28:45 +0200
Subject: Bayes Tools for Python ?
Message-ID: <3AC0A39D.389A7F6F@gmx.net>

anybody know sources for Bayes Net computing with Python (NumPy) ?

-robert


From dougfort at downright.com  Tue Mar 20 12:50:34 2001
From: dougfort at downright.com (Doug Fort)
Date: Tue, 20 Mar 2001 12:50:34 -0500
Subject: python suitability for large critical run forever apps
References: <3AB77CE4.78378862@troikanetworks.com>
Message-ID: <3AB7986A.206EFB37@downright.com>

Our website loadtesting application http://www.stressmy.com has backend

agents that 'run forever'. (Actually, we're doing frequent upgrades,

so most agents get restarted at least once a week, but we've

never had anything fail from memory or resource leaks).

I find multithreaded apps much easier to develop and debug in

Python than in C/C++ or Java.  However, I'm becoming convinced

that it's better to use select() than threads for waiting on I/O.

--
Doug Fort (dougfort at downright.com)
Senior Meat Manager
Downright Software LLC
http://www.dougfort.net


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20010320/c17de350/attachment.html>

From tim.one at home.com  Thu Mar 22 23:39:49 2001
From: tim.one at home.com (Tim Peters)
Date: Thu, 22 Mar 2001 23:39:49 -0500
Subject: New PEP: The directive statement
In-Reply-To: <j4hf0mywbn.fsf@informatik.hu-berlin.de>
Message-ID: <LNBBLJKPBEHFEDALKOLCKELHJHAA.tim.one@home.com>

[Martin von Loewis]
> ...
> I have my problems with that part of PEP 236 also. While it is
> certainly possible to provide the release when a feature was first
> released, it is not possible to give an accurate answer to the
> question when it will become mandatory. The intent to make it
> mandatory in the future should go into the documentation, IMO.

If the mandatory release is greater than the release you're running under,
it's certain that the feature in question is not yet "the rule".

If the mandatory release is less than or equal to the release you're running
under, it's certain that the feature in question *is* "the rule".

Those are the only two interesting questions at runtime, and they're answered
precisely.  It doesn't matter if the first prediction is wrong, because
__future__.py ships *with* the compiler.  This gives the pair of them a
consistent view of what the truth is for that version of the compiler; if the
prediction is wrong, no problem, it's updated in the next release and then
*that* gets a self-consistent view of the world too.  The two questions above
are always answered correctly for the version you're running under.

Of course the docs for a release should explain the state of features too,
but info in the docs isn't accessible to programs.

> Furthermore, what is the use of that information to a program?

Primarily, it's what makes it possible to write a simple tool that nukes
obsolete future statements.  The tool can't know whether a future_statement
is obsolete wrt the release it's being run under unless it has a repository
of release info shipped with that release, and in program-accessible form
(see above).  __future__.py makes that not just possible, but *easy*.

In another vein, in my own code I don't hesitate to use new features, and
couldn't care less if rare incompatible changes require me to rewrite some
code.  So I'll likely use future_statements much more than most people.  I
already have code that does

    from __future__ import nested_scopes
    import sys
    assert nested_scopes.getMandatoryRelease() > sys.version_info

This will assert-fail as soon as I run it under a release where nested_scopes
is the rule, and at that point I'll simply delete the future_statement and
move on.

Code like this also works fine:

import __future__
if __future__.nested_scopes.getMandatoryRelease() > sys.version_info:
    import old_version_of_module as amodule
else:
    import amodule

I don't know that I'll have an acute need for that wrt nested_scopes
specifically (neither do I know that I won't), but there is an endless
sequence of proposals for, e.g., changing Python in incompatible ways that
would (supposedly) allow it to run faster.  In that case I may very well have
an acute need to ship one version of my source that exploits the new
semantics but falls back to an older & slower module if the compiler it's
running under can't play along.

Etc.  Introspective features *always* find good uses.  I'm sure the Python
community will think of others that haven't even dimly occurred to me yet.
That's why it's there:  not because someone already thought far enough ahead
to ask for it, but because it's thoroughly predictable that somebody *will*.

when-you-don't-have-a-hammer-everything-looks-like-a-brick-wall-ly y'rs
    - tim




From root at rainerdeyke.com  Wed Mar  7 21:07:06 2001
From: root at rainerdeyke.com (Rainer Deyke)
Date: Thu, 08 Mar 2001 02:07:06 GMT
Subject: I come to praise .join, not to bury it...
References: <kuofvfjb77.fsf@lasipalatsi.fi> <000001c0a64c$497df250$a9807018@cr706570a> <mailman.983997159.4842.python-list@python.org>
Message-ID: <ePBp6.367559$ge4.129409870@news2.rdc2.tx.home.com>

"D-Man" <dsh8290 at rit.edu> wrote in message
news:mailman.983997159.4842.python-list at python.org...
> Even with my newfound understanding, provided by the Martellibot, I
> agree with this.  Calling a method on an object is fine.  Calling it
> on a literal looks weird.  Take for example something like this
>
> three = 1.__add__( 2 )

How about 'three = (1).__add__(2)'?  Adding parentheses around the literal
should make this look right; otherwise you are not calling methods on the
results of complex expressions often enough.


--
Rainer Deyke (root at rainerdeyke.com)
Shareware computer games           -           http://rainerdeyke.com
"In ihren Reihen zu stehen heisst unter Feinden zu kaempfen" - Abigor




From phrxy at csv.warwick.ac.uk  Tue Mar 27 11:58:40 2001
From: phrxy at csv.warwick.ac.uk (John J. Lee)
Date: Tue, 27 Mar 2001 17:58:40 +0100
Subject: boolean values in COM / VB / Python
In-Reply-To: <3ABFE1F6.4232DE4E@venix.com>
Message-ID: <Pine.SOL.4.30.0103271748490.3595-100000@mimosa.csv.warwick.ac.uk>

On Mon, 26 Mar 2001, Lloyd Kvam wrote:

> My guess is that you don't have a COM wrapper in place for the
> Document.  In other words:
>
> import win32com.client
> ie = win32com.client.Dispatch("InternetExplorer.Application")
> ie.Visible = 1	# toggles window visible
> ie.Visible = 0	# toggles window invisible
> ie.GetHome()	# loads the home page
>
> Now when you refer to
> ie.Document
> you are actually referring to a new object.

Well, yes.  And?

> I believe that you will
> need to do some work to provide the wrapper so that Python values get
> translated to the equivalent COM values.
[...]

But what would the equivalent COM values be?  That's the problem!  In
particular, what would be the Python for 'false' in VB?  I've tried 0, -1,
'0', None, 1 and other less likely objects, and they all either don't do
anything, or pop up a box asking where to save, which would be contrary to
the docs if I were succeeding in passing something meaning 'false'.

When you talk about a 'COM wrapper', is this in the general sense, or in
some specific COM sense?  I'm new to COM and windows stuff.


John



From jcc.ugm at ix.netcom.com  Wed Mar 28 14:29:23 2001
From: jcc.ugm at ix.netcom.com (Jonathan Claggett)
Date: Wed, 28 Mar 2001 14:29:23 -0500
Subject: Is inheritance broken?
In-Reply-To: <15042.11167.848530.564353@beluga.mojam.com>
Message-ID: <NDBBLAKACLGJJBDNGGCMOEJBCCAA.jcc.ugm@ix.netcom.com>

>    Jonathan> Isn't this wrong?
>
>Nope.  It's a feature of multiple inheritance, not a bug.  The search
>through base classes is left-to-right, depth first, thus
>parallelogram.angle is found before rectangle.angle.

Thank you, I think I understand now. I think the problem is that the depth
first searching works fine so long as you don't have a common ancestor in
your class tree. If you do have a common ancestor, it will be searched ahead
of some of its descendant classes. I wonder if there is a (quick) way to do
a depth first search that accounts for common ancestors...

If I really wanted this to work, guess one thing I could do is to cheat by
creating a 'flat' list of classes and then manually creating my class tree.
Something like this:

<code>

# 'flat' classes
class parallelogram_flat:
	def angle(self): return "variable"
	def side(self): return "variable"

class rhombus_flat:
	def side(self): return "same"

class rectangle_flat:
	def angle(self): return "90 degree"

class square_flat:
	pass

# 'real' classes
class parallelogram(parallelogram_flat):
	pass
class rhombus(rhombus_flat, parallelogram_flat):
	pass
class rectangle(rectangle_flat, parallelogram_flat):
	pass
class square(square_flat, rhombus_flat, rectangle_flat, parallelogram_flat):
	pass

# usage

s = square()
print "squares have", s.angle(), "angles"
print "squares have", s.side(), "sides"

</code>

Ugly and error prone, no?

>Someone with more Python years under their belt will have to answer why
>Guido chose depth first instead of breadth first search for methods when
>multiple inheritance is involved.  I'm pretty sure it's a choice that can't
>be changed now, however.  There's probably too much code that relies on the
>current semantics.

I second your question (does Guido read this list?) :-)




From aleaxit at yahoo.com  Fri Mar  9 17:47:53 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 9 Mar 2001 23:47:53 +0100
Subject: gah! I hate the new string syntax
References: <mailman.983485923.1136.python-list@python.org> <B7In6.16281$Qb7.2658902@newsb.telia.net> <97ohh2$polt0$1@ID-77331.news.dfncis.de> <97opsb$nhu$1@nntp6.u.washington.edu> <mailman.983615170.23313.python-list@python.org> <tN5o6.2746$eH3.14913@NewsReader> <Koco6.7518$1D5.244602@e420r-atl1.usenetserver.com> <mailman.983997252.5628.python-list@python.org> <m2vgpkyfjo.fsf@mycroft.actrix.gen.nz> <mailman.984063199.18779.python-list@python.org> <98an7i01tgp@news2.newsguy.com> <1eq0wih.1s0r5163x0lfkN%francois.granger@free.fr>
Message-ID: <98dpoc02ckn@news1.newsguy.com>

"Fran?ois Granger" <francois.granger at free.fr> wrote in message
news:1eq0wih.1s0r5163x0lfkN%francois.granger at free.fr...
    [snip]
> There are a lot of french citizen wich are descendant from italian. At

Not a new development either -- there's a continuity between
(notice the Italian surnames of French icons!-) Mazarino, Bonaparte,
Gambetta, Platini...:-).  [With any luck, one day my descendants
will be able to give a similar list of Albanian-surnamed Italian
greats...!-)].


> > -- few are aware, for example,
> > that such an icon of French popular culture as Yves Montand was
> > Italian -- real name Ivo Livi, born of just-as-Italian parents
    [snip]
> And lots of french people are really proud of Yves Montant, knowing or
> not his origin as Ivo Livi.

(And, joking aside, who could blame them -- besides being a
great singer and actor, he was a _true_ human being... his being
buried right next to his wife of many decades, such a rare event
in the show-business world, being a typical symptom...).


Alex





From ssartor at bellatlantic.net  Wed Mar 28 13:15:29 2001
From: ssartor at bellatlantic.net (Steven Sartorius)
Date: Wed, 28 Mar 2001 18:15:29 GMT
Subject: How to update status bar in wxPython?
Message-ID: <5Tpw6.2172$K%.667185@typhoon1.ba-dsg.net>

I'm a newbie to GUI programming in general so apologies if this is a FAQ....

I've got a Python script that does some heavy duty number crunching and file
I/O.  To track the progress of the script I print a series of messages to the
console/x-term ("Working on file #....", etc,etc).  As a project, I decided to
guify the script with wxPython and Boa-constructor.  I've successfully built a
window with a single button, a couple of text input fields and a status bar;
clicking the button fires off the script.  What I'm trying to do (so far
unsuccessfully) is redirect the progress messages from the console to the
status bar on the window.  Because I'm retrofitting an existing script, I've
just replaced the original 'print' statements with
'statusBar.SetStatusText("Working on file....") and (for good measure)
'statusBar.Refresh()'.  This approach kind of works....the status bar shows
the final progress message ("All data processed")  but only after I force the
window to redraw (moving it or covering and then uncovering it).  I'm sure I'm
missing something basic here so any help would be much appreciated.

Thanks,

Steve


From scarblac at pino.selwerd.nl  Tue Mar 13 09:21:35 2001
From: scarblac at pino.selwerd.nl (Remco Gerlich)
Date: 13 Mar 2001 14:21:35 GMT
Subject: What's wrong with me/Python ?
References: <3AAE0652.D68F253E@daimlerchrysler.com>
Message-ID: <slrn9asb38.c80.scarblac@pino.selwerd.nl>

Alexander Stirzel wrote in comp.lang.python:
> Hi!
> 
> I just wrote a small script for killing some hanging netscape processes,
> the source is listed below.
> Now if I run this script I get this error message:
> 
> 8<----------- error message -----------------------------
> Traceback (most recent call last):
>   File "/usr/sbin/netscape-killer.py", line 11, in ?
>     splitted = string.split(line)
> NameError: There is no variable named 'string'
> 8<----------- error message -----------------------------

You forgot to import string. Have a nice day.

-- 
Remco Gerlich


From spencerm at cruzio.com  Fri Mar 30 02:54:01 2001
From: spencerm at cruzio.com (Michael Spencer)
Date: Thu, 29 Mar 2001 23:54:01 -0800
Subject: Creation d'un NG francophone sur Python
References: <1eqzyk6.gxd4jr1tkjykzN%francois.granger@free.fr>
Message-ID: <9a1e6f011es@enews3.newsguy.com>

Pardon, mais pourquoi un NG pour Python en fran?ais?

Python a cr?? par un nederlandais qui a ?crit en anglais - pas sa langue
natif. Il a ?crit en anglais ? assurer la plus grande distribution de ses
id?es. Pourquoi on marcherait en arri?re aux temps quand chaque groupe
a maintenu sa <<independance>> par s'isoler des autres?


Fran?ois Granger <francois.granger at free.fr> wrote in message
news:1eqzyk6.gxd4jr1tkjykzN%francois.granger at free.fr...
> Sorry for this french message. Translation follow.
>
> Il est encore temps de voter pour la cr?ation d'un NG en fran?ais sur
> Python. Allez dans 'fr.usenet.forums.annonces' et cherchez le message
> '[AAV 3] Creation de fr.comp.lang.python (non-modere)'
> votez en r?pondant ? ce message.
>
> [en]
> It is time for voting to create the french speaking newsgroup
> 'fr.comp.lang.python'. Go to 'fr.usenet.forums.annonces' and look for
> the message '[AAV 3] Creation de fr.comp.lang.python (non-modere)'.
> Votes by replying to this message.
>
> --
> "La connaissance est le chemin de la tol?rance, c'est valable pour
> tous, en toutes saisons."
> - Raymond Page




From bill-bell at bill-bell.hamilton.on.ca  Tue Mar 20 07:17:23 2001
From: bill-bell at bill-bell.hamilton.on.ca (Bill Bell)
Date: Tue, 20 Mar 2001 07:17:23 -0500
Subject: Whois client?
In-Reply-To: <985032471.1506.6177.l8@yahoogroups.com>
Message-ID: <3AB70403.26884.2645462@localhost>

Has anyone seen a Whois client coded in Python?

Thanks for any leads.

Bill Bell, MSc, Software Developer



From jschmitt at vmlabs.com  Thu Mar 29 20:42:19 2001
From: jschmitt at vmlabs.com (John Schmitt)
Date: Thu, 29 Mar 2001 17:42:19 -0800
Subject: issues with mixing and matching python packages
Message-ID: <008F0A63472BD311AF9800104BCD102561CD63@minirex.vmlabs.com>

I'm talking about the PythonLabs 2.0 versus ActiveState versus the installer
on Python.org.  I just glanced at the notes for the ActiveState installer
and it says:

- On Windows, ActivePython 2.0 cannot coexist with other Python 2.0
installations (for example, BeOpen's Python 2.0).  Whichever installer is
run last will determine what version of Python is used (the registry can
only point to one executable).  The ActivePython windows installer will warn
you if you try to install over an existing Python 2.0 installation. 

- You must ensure your PYTHONPATH is clean before installation. You should
ensure that your machine does not have PYTHONPATH, PYTHONHOME, or other
relevant variables set, and that your installation directories are clean. If
you have problems importing certain modules after installation, please check
these things before submitting a bug report. 

Does this mean that I can't simply change the installation I'm using by
simply changing some environment variables?  What's up?

What happens when I upgrade to 2.1?  Is that going to be a problem?

John



From erno-news at erno.iki.fi  Fri Mar  9 03:02:08 2001
From: erno-news at erno.iki.fi (Erno Kuusela)
Date: 09 Mar 2001 10:02:08 +0200
Subject: Apache and os.fork() in CGI-Binaries
References: <3aa7ded0@si-nic.hrz.uni-siegen.de> <3AA84BF3.E19873DB@ninenet.com>
Message-ID: <kuwv9zid4v.fsf@lasipalatsi.fi>

In article <3AA84BF3.E19873DB at ninenet.com>, Matt Harden
<matth at ninenet.com> writes:

| Simon,
| You need os.dup2().  Like this:
|    logfile = os.open("/path/to/logfile",os.O_CREAT|os.O_WRONLY)
|    os.lseek(logfile, 0, 2)  /* seek to end */
|    os.dup2(logfile, sys.stdout.fileno())
|    os.dup2(logfile, sys.stderr.fileno())
|    os.close(logfile)

it is better to use O_APPEND with log files. you may also need
to lock it so that many processes do not write to it at the same time.

  -- erno


From quinn at retch.ugcs.caltech.edu  Wed Mar 14 17:03:34 2001
From: quinn at retch.ugcs.caltech.edu (Quinn Dunkan)
Date: 14 Mar 2001 22:03:34 GMT
Subject: gah! I hate the new string syntax
References: <mailman.983485923.1136.python-list@python.org> <97sudi01i26@news1.newsguy.com> <98elrk$li8$1@nntp.Stanford.EDU> <98fc070thh@news1.newsguy.com> <98iqo2$56o$1@nntp.Stanford.EDU> <98iv3s01sc2@news2.newsguy.com>
Message-ID: <slrn9avqlm.8ee.quinn@retch.ugcs.caltech.edu>

On Mon, 12 Mar 2001 17:56:33 +0100, Alex Martelli <aleaxit at yahoo.com> wrote:
>Focusing on the polymorphism needs, I see the issue in the mirror
>way from you: it's a _weakness_ (of Python and most other languages)
>that I can't add methods 'post-facto, from the outside' to existing
>objects -- it impedes the most natural, elegant, typeswitch-free way
>to add functionality.  If you pursued your "capitalize as a function"
>idea, you'd be unable to properly capitalize Unicode (e.g.) without
>a typeswitch -- a very unadvisable construct.  Something like Haskell's
>typeclasses, on the other hand, would provide the best of both worlds
>(multimethods would be an even more powerful alternative) -- let the
>compiler/interpreter/runtime do the dispatching, and just code in
>the most natural, polymorphic way.

Note that in ruby, any time you write

class Foo
    ...
end

you are adding to, not overriding, 'Foo'.  Thus,

class String
    def cap
        s = (if self[0].betwwen?(97, 122) then
                self[0] - 32
            else
                self[0]
            end).chr
        s + self[1 .. self.length]
    end
end

is perfectly valid ruby that does what he wanted.  I thought this sort of
thing was cute when I first saw it, but it makes me a bit nervous: I've
introduced a global change in a system-level class.  The problem is that, in
an imperative language, that change occurs at a pont in time.  It makes the
order I load modules of possible importance.  It could break client modules
who are expecting a "clean slate".

Naturally, in a statically-typed functional language like haskell, these
issues don't come up.  But in an imperative language, it's like frobbing
__builtins__ ---probably not a very good idea.

For classes that are not completely under your control, or if the
functionality is only of local interest, I think either a subclass or a
"external" function is better style (ruby provides "singleton methods", but
I'm happier with a plain non-OO function).

For example, I had a set of Cell objects that were to be displayed
graphically.  The display mechanism required that they each have a special tag
to keep track of their position on the screen.  The tag was useless to all
other clients of Cells, and not even computable by the Cell itself.  Unsure of
what the "official" solution would be, I wrote the tag computation as a
function of the display module, and had the display object cache the tags
internally.  It seemed more logical that way: the display mechanism should
keep track of its own administrivia.  But then I wound up with 'tagof(c)',
where 'tagof()' was a typeswitch on 'c', which is the sort of thing you're not
supposed to have in OO.

In haskell, I could have simply defined a 'Taggable' class and instances
implementing 'tagof' in the Display module... no problem.  But "traditional"
OO seems to make the solution much less clear...


From martin.skott at propylon.com  Wed Mar  7 13:03:41 2001
From: martin.skott at propylon.com (Martin Skott)
Date: 07 Mar 2001 18:03:41 +0000
Subject: XML Module problems
References: <mailman.982853044.8344.python-list@python.org> <memo.20010306181753.34635C@digitig.compulink.co.uk> <y9uhy8cf.fsf@propylon.com>
Message-ID: <u255xxqa.fsf@propylon.com>

Martin Skott <martin.skott at propylon.com> writes:

> Sean isn't currently reading newsgroups, but I hope my answer will do

Damn finger getting to fast typing. What I meant is that Seand is away
and doesn't read news currently :-)

-- 
Martin Skott
Software Developer
Propylon - Enabling Universal Mobility
http://www.propylon.com
Tel: +353 096 37245
Mobile: +353 087 9680370


From tim.one at home.com  Sun Mar 25 17:12:15 2001
From: tim.one at home.com (Tim Peters)
Date: Sun, 25 Mar 2001 17:12:15 -0500
Subject: Python classes on east coast?
In-Reply-To: <bobhicks-2503011622160001@va-frontroyal1a-202.wch.adelphia.net>
Message-ID: <LNBBLJKPBEHFEDALKOLCKEFBJIAA.tim.one@home.com>

[Robert Hicks]
> Is there or are there any plans for Python classes on the east coast
> (preferrably around DC)?

Learning Tree has DC-area classes:

    http://www.learningtree.com/us/ilt/courses/430.htm

some-even-in-guido's-home-town<wink>-ly y'rs  - tim




From andreas at andreas-jung.com  Thu Mar 22 07:55:49 2001
From: andreas at andreas-jung.com (Andreas Jung)
Date: Thu, 22 Mar 2001 07:55:49 -0500
Subject: Why Python/Jython?
References: <99cl9e$b2b$1@Merlin.et.bocholt.fh-gelsenkirchen.de>
Message-ID: <99csmm$bbf$1@slb6.atl.mindspring.net>

In a former project I used JPython in combinatination with CPython
to access an Oracle 8i database. Because the Python bindings for Oracle 8
were incomplete  - the combination JPython/JDBC was very cool :-) Working
with Java APIs in Pythons interactive mode is nice :-) Performance does not
matter in any case.

Andreas


"Andreas Freier" <freier at xtramind.com> schrieb im Newsbeitrag
news:99cl9e$b2b$1 at Merlin.et.bocholt.fh-gelsenkirchen.de...
> In several articles I read that Python/Jython imporoves the programmers
> performance many times. Why?
> I am considering to use Jython for my job as Java programmer. How should I
> use it to achieve this performance?
>





From hgg9140 at cola.ca.boeing.com  Mon Mar  5 18:09:58 2001
From: hgg9140 at cola.ca.boeing.com (Harry George)
Date: Mon, 5 Mar 2001 23:09:58 GMT
Subject: MySQL-python build griefs
References: <G9qvt6.L0I@world.std.com>
Message-ID: <xqxvgpn4xtl.fsf@cola.ca.boeing.com>

I installed no problem (MySql 3.23.32, Redhat 6.2) with:

export PKGVER=MySQL-python-0.3.1
ln -s ../${PKGVER}.tar.gz .
gzip -cd  ${PKGVER}.tar.gz | tar xvf -
#---1.5.2---
cd  ${PKGVER}
#===edit Setup===
if sys.platform == "linux-i386": # Red Hat
    include_dirs = ['/usr/local/mysql/include/mysql']
    library_dirs = ['/usr/local/mysql/lib/mysql']
#===end edits===
python setup.py build
su -c "python setup.py install"                                                      
wware at world.std.com (Will Ware) writes:

> I'm trying to build MySQL-python on a Red Hat 6.2 box. In order
> not to confuse the Red Hat installation, which depends heavily
> on Python, I've left Python at 1.5.2. I have Distutils 1.0.1
> installed. The version of MySQL-python is 0.3.0. To get things
> to work at all, I needed to build my own copy of Python-1.5.2/*
> to populate the /usr/include/python1.5 tree, running "configure"
> to produce a suitable "config.h". Now I'm getting an error on
> "python setup.py build" that looks like this:
> 
> running build
> running build_py
> not copying MySQLdb.py (output up-to-date)
> not copying CompatMysqldb.py (output up-to-date)
> running build_ext
> Traceback (innermost last):
>   File "setup.py", line 90, in ?
>     extra_objects=extra_objects,
>   File "/usr/lib/python1.5/site-packages/distutils/core.py", line 138, in setup
>     dist.run_commands()
>   File "/usr/lib/python1.5/site-packages/distutils/dist.py", line 829, in run_commands
>     self.run_command(cmd)
>   File "/usr/lib/python1.5/site-packages/distutils/dist.py", line 849, in run_command
>     cmd_obj.run()
>   File "/usr/lib/python1.5/site-packages/distutils/command/build.py", line 106, in run
>     self.run_command(cmd_name)
>   File "/usr/lib/python1.5/site-packages/distutils/cmd.py", line 328, in run_command
>     self.distribution.run_command(command)
>   File "/usr/lib/python1.5/site-packages/distutils/dist.py", line 849, in run_command
>     cmd_obj.run()
>   File "/usr/lib/python1.5/site-packages/distutils/command/build_ext.py", line 200, in run
>     customize_compiler(self.compiler)
>   File "/usr/lib/python1.5/site-packages/distutils/sysconfig.py", line 106, in customize_compiler
>     (cc, opt, ccshared, ldshared, so_ext) = \
>   File "/usr/lib/python1.5/site-packages/distutils/sysconfig.py", line 368, in get_config_vars
>     func()
>   File "/usr/lib/python1.5/site-packages/distutils/sysconfig.py", line 274, in _init_posix
>     parse_makefile(filename, g)
>   File "/usr/lib/python1.5/site-packages/distutils/sysconfig.py", line 207, in parse_makefile
>     value = value[:m.start()] + done[n] + after
> TypeError: illegal argument type for built-in operation
> 
> Am I using the wrong color duct tape here? Or just not enough
> bubblegum?
> 
> -- 
> -----------------------------------+---------------------
>  22nd century: Esperanto, geodesic | Will Ware
>  domes, hovercrafts, metric system | wware at world.std.com

-- 
Harry George                E-mail:  harry.g.george at boeing.com
The Boeing Company          Renton:  (425) 237-6915
P. O. Box 3707  02-CA       Everett: (425) 266-3868
Seattle, WA 98124-2207      Page:    (425) 631-8803  


From m_debusk at deja.com  Mon Mar 12 03:18:13 2001
From: m_debusk at deja.com (Michael DeBusk)
Date: Mon, 12 Mar 2001 03:18:13 -0500
Subject: Einstein's Riddle
References: <oNPq6.396644$ge4.138479566@news2.rdc2.tx.home.com> <UdSq6.2184$54.2166@www.newsranger.com>
Message-ID: <3AAC8645.48990315@eclipsetel.com>

Gregory Jorgensen wrote:

> We have no evidence that any of the five owns fish.

I know very little about transformational grammar, but I've picked up
enough to respond to this. There's a thing called "presupposition" at
play here.

A presupposition is an idea one has to pre-suppose so one can make sense
of a given sentence. For example, the sentence "tell the dog to get off
the sofa" contains presuppositions that such a thing as a dog exists,
such a thing as a sofa exists, that both the dog and the sofa exist in
such a way that dogs can be on sofas, that dogs can obey commands...
even that the speaker believes that the listener speaks and understands
English. All the things native speakers of a language intuitively know
when they go through the process of understanding a sentence, in other
words.

The question "Who owns the fish?" contains a linguistic presupposition
indicating *someone* owns a fish; the question is not whether, but whom.
That presupposition is the evidence.

If the question was "Does someone own a fish?" one could then assert
there was no available information.

Someone else in the thread pointed out the problems involved in
translating the puzzle to other languages. I imagine that would create
some roadblocks, because the structure of presuppositions would be
different in each.

-- 
Use the munged address below to reply:
debu4335 [shift-2] eclipsetel [dot] com



From dsh8290 at rit.edu  Fri Mar  9 11:27:34 2001
From: dsh8290 at rit.edu (D-Man)
Date: Fri, 9 Mar 2001 11:27:34 -0500
Subject: Static typing (was Re: Java guy interested in Python)
In-Reply-To: <20010309095209.24339.qmail@web206.mail.yahoo.com>; from hamish_lawson@yahoo.co.uk on Fri, Mar 09, 2001 at 09:52:09AM +0000
References: <20010309095209.24339.qmail@web206.mail.yahoo.com>
Message-ID: <20010309112734.G4546@harmony.cs.rit.edu>

On Fri, Mar 09, 2001 at 09:52:09AM +0000, Hamish Lawson wrote:
| Michael Hudson wrote:
| 
| > If you type-annotate this, you might write (using entirely made up
| > syntax):
| > 
| > def write16(file : <file>, int : <int>):
| >     file.write(struct.pack("h",int))
| > 
| > but then what about passing in a StringIO object, or some user 
| > wrapper of a file object?  What makes a "<file>"?  At the moment, 
| > it's just a loose collection of methods, some of which are optional 
| > in some circumstances.
| 
| Wouldn't there be a hierarchy of interface types for file-like objects,
| with some interfaces specifying more methods than others? You'd then
| choose the interface type that specified the collection of methods that
| you require of your passed object.

Yes.  It's called Java ;-).

| 
| Not-that-I'd-necessarily-want-this-all-of-the-time-ly yours,

I certainly don't.


I do like the ERR side of compile-time type checking however.  Perhaps
someone should create a Py-lint that will check python source for this
sort of thing.  Then again maybe it's already called PyUnit or
doctest.

-D




From dmaas at dcine.com  Thu Mar  8 00:19:45 2001
From: dmaas at dcine.com (Dan Maas)
Date: Thu, 08 Mar 2001 05:19:45 GMT
Subject: Unicode getting in the way...
Message-ID: <RDEp6.199688$v.18718576@typhoon.nyroc.rr.com>

Python 2.0. All I want to do is construct and parse simple
binary packets. e.g.

foo = "abcd" + struct.pack("B", 254)
foo += "efg%s" % "lalala"

At some random point in my code, foo changes from being a
'string' object to a  'unicode' object. Subsequent attempts
to manipulate foo fail with a UnicodeError (ASCII decoding
error, ordinal not in range(128)).

How can I stop this from happening? I just need raw 8-bit
strings everywhere!

Thanks,
Dan




From harri at trema.com  Fri Mar 16 09:15:06 2001
From: harri at trema.com (Harri Pasanen)
Date: 16 Mar 2001 15:15:06 +0100
Subject: list.sort(cmpfunc) question
Message-ID: <m1ae6lg5qt.fsf@pclabs188.labs.trema.com>

The list.sort(cmp) docs for Python 2.0 state:

   The sort() method takes an optional argument specifying a
   comparison function of two arguments (list items) which should
   return -1, 0 or 1 depending on whether the first argument is
   considered smaller than, equal to, or larger than the second
   argument.


The build in cmp() function docs state:


cmp (x, y)

   Compare the two objects x and y and return an integer according to
   the outcome. The return value is negative if x < y, zero if x == y
   and strictly positive if x > y.


So interpreted strictly the built in cmp() could not be used as
list.sort() argument, as it can apparently (and does) occasionally
return for instance the value 2.

In practice list.sort() seems to be happy with having a cmp function
that behaves as built in cmp(), but can this be relied upon?  And if
yes, maybe the doc should be changed?

Regards,

Harri





From grante at visi.com  Thu Mar  8 20:50:15 2001
From: grante at visi.com (Grant Edwards)
Date: Fri, 09 Mar 2001 01:50:15 GMT
Subject: I come to praise .join, not to bury it...
References: <kuofvfjb77.fsf@lasipalatsi.fi> <000001c0a64c$497df250$a9807018@cr706570a> <mailman.983997159.4842.python-list@python.org> <Ntyp6.1479$y6.244056@ruti.visi.com> <989btv$rd4$1@panix3.panix.com>
Message-ID: <rFWp6.212$Tg.13491@ruti.visi.com>

In article <989btv$rd4$1 at panix3.panix.com>, Aahz Maruch wrote:

>>>Even with my newfound understanding, provided by the Martellibot, I
>>>agree with this.  Calling a method on an object is fine.  Calling it
>>>on a literal looks weird.

>>What makes you think a "literal" is not an object?
>
>It is, but not at the aesthetic level.

But, aesthetics is determined largely by cultural influences:
i.e. background.

";" isn't an object at the aesthetic level *for a C
programmer*.  For others with different backgrounds, ";" is
just as much an object as anything else -- both at an aesthetic
level and at a technical level.

>>> Take for example something like this
>>>
>>>three = 1.__add__( 2 )
>>
>>Looks just fine to a ex-Smalltalker: everything is an object.
>
>Python isn't Smalltalk.

Neither is it C. ;)

-- 
Grant Edwards                   grante             Yow!  Is it FUN to be
                                  at               a MIDGET?
                               visi.com            


From aleaxit at yahoo.com  Thu Mar 15 07:35:12 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 15 Mar 2001 13:35:12 +0100
Subject: tiny extension crashes
References: <Pine.SOL.4.30.0103141957240.10526-100000@mimosa.csv.warwick.ac.uk>
Message-ID: <98qcto02fde@news1.newsguy.com>

"John J. Lee" <phrxy at csv.warwick.ac.uk> wrote in message
news:Pine.SOL.4.30.0103141957240.10526-100000 at mimosa.csv.warwick.ac.uk...
>
> I've been using a tiny little C extension module that I wrote for a while
    [snip]
>     double number;
>     double *intp;
>
>     if (!PyArg_ParseTuple(args, "d", &number))
>         return NULL;
>     if (fabs(modf(number, intp)) >= 0.5) {

As others have noted, THIS is your problem -- I just wanted
to underscore that this is a very common C-newbie error: when
a function takes a "whatever*" (pointer-to-whatever) argument,
you need to pass *the address of something* (or possibly, for
some functions, an explicit NULL), *NEVER* a generic and not
initialized variable of type "whatever*".  It's easy to remember:
C arguments are by-value -- all you ARE passing is the VALUE
of the actual argument; passing as actual-argument a variable
is never right unless it's been previously initialized by some
suitable expression being assigned to it (for some reason, this
programming error typically affects the passing of arguments
which are declared to have pointer-type).


Alex





From jra at dorothy.msas.net  Sat Mar 31 01:31:29 2001
From: jra at dorothy.msas.net (Jay R. Ashworth)
Date: Sat, 31 Mar 2001 06:31:29 GMT
Subject: Linux GPS program
References: <3AC01637.4EF3A6F2@crynwr.com>    <slrn9c17k4.cfc.jra@dorothy.msas.net> <3AC0D915.83C2C0C3@crynwr.com>    <slrn9c1p9t.d3g.jra@dorothy.msas.net> <99suho$m7a$2@newsserver.rrzn.uni-hannover.de> <slrn9c7ejl.sv.jra@dorothy.msas.net> <3AC4ADE4.DD629457@crynwr.com>
Message-ID: <slrn9caudh.224.jra@dorothy.msas.net>

This one time, at band camp,
  Russell Nelson <nelson at crynwr.com> wrote:
> "Jay R. Ashworth" wrote:
> > No, actually, I'd bet some cash that the largest collection by size is
> > the combined efforts of the US Govenment.  That's (almost) all free.
> 
> So, do you have any suggestions on what vector file format and dataset I
> should aim at first?

No.

:-)

Depends on whether the target is hikers, road warriors, boaters, or
flyers, really.  The USGS DRG-O and Shapefile formats seem to be the
most common, but the former is scheduled to be replaced by a format I
mentioned last week sometime (my laptop went for a swim the other day,
and the mouse circuitry is still mildly horked, so I can't find the
reference easily for you), which is likely to displace the latter too,
in time.

Check out freegis.org and nationalatlas.gov, for a couple of useful
starting points, I guess.  More When I Can Get To More<tm>.

Cheers,
-- jra
-- 
Jay R. Ashworth                                                jra at baylink.com
Member of the Technical Staff     Baylink
The Suncoast Freenet         The Things I Think
Tampa Bay, Florida        http://baylink.pitas.com             +1 727 804 5015


From jkraska1 at san.rr.com  Thu Mar 22 02:37:33 2001
From: jkraska1 at san.rr.com (Courageous)
Date: Thu, 22 Mar 2001 07:37:33 GMT
Subject: Good Editor?
References: <3ab9a32b$1_1@news2.one.net>
Message-ID: <q5bjbt8sd2602hsutc7hfm3nvj7f281lg8@4ax.com>

On Wed, 21 Mar 2001 23:00:58 -0800, "fjs" <fjs6 at usa.net> wrote:

>Can someone recomend a good editor to write python in?  Im just getting
>started learning python and dont know which ones support python.

Vim has a language highliting for python and has friendly tab control.
I have mine set, for example, so that "every tab fills in enough spaces
to take you to the next tab stop." On all my python projects, actual tab
characters are completely forbidden.

C//



From johnson2 at hiwaay.net  Sun Mar 25 23:06:23 2001
From: johnson2 at hiwaay.net (Lyle Johnson)
Date: Sun, 25 Mar 2001 22:06:23 -0600
Subject: Python script to create HTML gallery, thumbnails
References: <3ABCAC94.B095CB9A@ozemail.com.au>
Message-ID: <tbtfklt02kj0a5@corp.supernews.com>

> I recently got a digital camera (Kodak DC4800) and wanted to
> put all the crappy photos I shoot on the web.

No answer for your question, I was just thinking about how Kodak needs to
get in touch with you for their next ad campaign:

    "With Kodak's new DC4800 digital camera, now you too can put all the
crappy photos you shoot on the web!"






From ransen_spam_me_not at nemo.it  Mon Mar  5 03:32:14 2001
From: ransen_spam_me_not at nemo.it (Owen F. Ransen)
Date: Mon, 05 Mar 2001 08:32:14 GMT
Subject: tokenizer.c crashes in 2.1b1
Message-ID: <3aa34d5c.1284101@news.newsguy.com>

Embedding Python 2.1b1 I recreated the DLL and LIB from
the sources, VC6 , and any call to PyRun_SimpleFile still
crashes.

The debugger points to the guilty party:

tokenizer.c, the fgets call at line 247 fails.

My VC project uses Multithreaded Debug compiler settings,
as does the Pythoncore project.

What do I do now?




--
Owen F. Ransen
http://www.ransen.com/
Home of Gliftic & Repligator Image Generators


From claird at starbase.neosoft.com  Tue Mar 27 11:31:40 2001
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 27 Mar 2001 10:31:40 -0600
Subject: Why not Tcl/Tk?
References: <4211ctgvusgcma0c6ok1cpmjou98lv9714@4ax.com> <8J2w6.3336$SB2.606281@ruti.visi.com>
Message-ID: <B55136AA6FACFC33.F1A4F29279C28E0D.924F8B9490EDACDF@lp.airnews.net>

In article <8J2w6.3336$SB2.606281 at ruti.visi.com>,
Grant Edwards <grante at visi.com> wrote:
			.
			.
			.
>Tcl 
> Flaws:  
>     Only one data type: string
>     Incomprehensible quoting semantics
>     Roll-your-own control flow
>     No OO support
> Strengths:
>     Tk integration
>     Lots of code out there
>     Roll-your-own control flow
>     
>Python:
> Flaws:
>     No roll-your-own control flow
>     Lots of GUI options
> Strengths:
>     No roll-your-own control flow
>     Good OO support
>     Lots of GUI options
>     Simple syntax and semantics
>     Library modules 
>
>>-Why do you prefer the one you use over the other?
>
>I gave up on Tcl after one program and switch to Scheme.  Now I
>use Python a lot more than Scheme.
>
>[I don't know much about eithr on Win32, so I'll let others
>answer those questions.]
			.
			.
			.
Grant of course does fine on his own.  I think a few
footnotes are pertinent.

We mention in <URL: 
http://www.sunworld.com/unixinsideronline/swol-02-2001/swol-0202-regex.html >
that Grant's footnote alludes to the single most crucial
difference between Tcl and Python:  the latter's current
support of and integration with Win* is far superior.

Grant is absolutely right in repeating that, in case after
case, the same characteristic has both advantages and
disadvantages.  To further complexify his calculations,
I'll testify that many Tcl and Python application developers
aren't even aware of the control-flow philosophical
differences between the two languages.  Many programmers
spend their time with rather conventional procedural or
OO concerns, and are quite indifferent to metaprogramma-
bility or introspective capabilities.

I respect that Grant doesn't like Tcl quoting.  There
are, however, equally sane people who applaud it.  It
seems to be an ... ambiguous distinction.

Among the benefits Tcl has over Python that are easiest
to appreciate by the mass of application developers are
the former's [exec], [open], and [socket].  They allow
for succinct, portable interprocess communication that's
simply not as polished (at that level) in Python.
-- 

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


From tim.one at home.com  Sat Mar  3 04:54:01 2001
From: tim.one at home.com (Tim Peters)
Date: Sat, 3 Mar 2001 04:54:01 -0500
Subject: Bug in Py_Finalize() ?
In-Reply-To: <3aa1889e.2213634@news.newsguy.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCKEKFJDAA.tim.one@home.com>

[Owen F. Ransen, getting
    Fatal Python error: UNREF invalid object
 when running
    for (UINT i = 0 ; i < 10 ; i++) {
        Py_Initialize();
        Py_Finalize() ;
    }
]

[Tim]
> Please try this again with the lastest CVS version, or pick up
> 2.1b1 (we'll be releasing that later today).  Something "very much like
> this" got fixed last month; ...

[Owen]
> Thanks for the pointer.
>
> Now I am going to reveal my ignorance, what does CVS stand for?

The latest development version of the Python source code is maintained under
CVS-- a popular source-control system --and available to the public at
SourceForge.  Doesn't matter unless you're willing to compile Python from
source.  Else pick up a 2.1b1 installer (you said you were running on
Windows) and just try it:

    http://www.python.org/2.1/

You didn't give a complete program so it's impossible for me to say whether
your problem is fixed.  I did put your loop in an extension module and
boosted the loop limit to 1000.  No problem under Windows Python 2.1b1, under
either release or debug builds.




From slhath at home.com  Sun Mar  4 11:46:09 2001
From: slhath at home.com (Scott Hathaway)
Date: Sun, 04 Mar 2001 16:46:09 GMT
Subject: sql query results to xml
Message-ID: <ljuo6.134776$bb.1432827@news1.rdc1.tx.home.com>

I want to create a tool that will take a valid SQL query and return the
results as valid xml.  Before I reinvent the wheel, is there a python class
or code snippet that someone has that does this already?

Thanks,
Scott




From xu.20 at nd.edu  Mon Mar 26 16:58:32 2001
From: xu.20 at nd.edu (C. S. Xu)
Date: Mon, 26 Mar 2001 16:58:32 -0500
Subject: Biggest float number?
Message-ID: <3ABFBB88.E718B2D@nd.edu>

Hi all,

It seems the biggest floating number in Python can be:
    2.**1024 - 1.
which is about 1.8E308. Is there any way to get bigger value than this,
just like long intergers almost have no limits?

Thanks in advance for any replies.

xcs



From morelli at alpha.cerm.unifi.it  Thu Mar  8 06:31:06 2001
From: morelli at alpha.cerm.unifi.it (Morelli Enrico)
Date: Thu, 8 Mar 2001 10:31:06 -0100
Subject: cdrecord output capture
Message-ID: <Pine.LNX.4.30.0103081022410.23372-100000@alpha.cerm.unifi.it>

Dear all,

I am a new user of python.

I trying to write a small python program under linux to create and
save ISO files using mkisofs, cdrecord and dialog.

I'm able to capture mkisofs output using popen2, but I'm not able to do
the same thing with cdrecord. I had see the source code of eroaster
(graphic python program that use mkisofs and cdrecord) but this program
use a difficult way for me, I don't understand all code.

Someone explain me what I must to do to capture the cdrecord output?

Thanks in advance

-- 
                           \\\ //
                           (0 0)
------------------------ooO-(_)-Ooo-------------------------------
#============================#=====================================#
|     ENRICO MORELLI         |  email: morelli at CERM.UNIFI.IT       |
| *     *       *       *    |  phone: +39 055 4574269             |
|  University of Florence    |  fax  : +39 055 4574253             |
|  CERM - via Sacconi, 6 -  50019 Sesto Fiorentino (FI) - ITALY    |
#============================#=====================================#




From mertz at gnosis.cx  Sat Mar  3 02:20:51 2001
From: mertz at gnosis.cx (Dr. David Mertz)
Date: Sat, 03 Mar 2001 02:20:51 -0500
Subject: cataloging words in text file (fwd)
Message-ID: <TtJo6kKkXA4D092yn@bellatlantic.net>                         

Stephen Boulet <stepheb at comm.mot.com> wrote:
| I remember this homework assignment for my data structures (c++)
| class: read in a large file, and create a data structure containing
| every word in the file and the number of times it appears.

Funny you should ask.  I just finished a program called 'indexer.py'
that does EXACTLY this, and then uses the data structure (a dictionary)
to perform searches.  Find the program at:

  http://gnosis.cx/download/indexer.py

I wrote this module as part of my _Charming Python_ series of articles,
and there is an article discussing the design of the module.  However, I
have been a bit naughty lately in putting up articles to my own site
(thanks all my readers) before IBM gets around to publishing them (they
pay for them, so should have some rights here[*]).  So I don't actually
have the accompanying article on my website now.  If anyone begs me for
it in email, I'll cough it up individually... otherwise, wait a couple
weeks (and read it at IBM developerWorks, especially... and excellent
site for fine programming information).

That said, the above module itself is rather extensively documented...
and I very much welcome feedback.

Yours, David...

------------------------------------------------------------------------
[*] Actually...  I'm not sure they *should* have rights.  Information
wants to be free, and everything should be available to everyone all the
time, gratis.  After the revolution, writers, programmers, artists, and
us whole merry lot will be paid endowments, and all intellectual
creation will be part of the common lot of humankind... and so on...

But for now, IBM pays me money, and I agree to a variety of conditions
about publications, legal restrictions, and stuff like that.

P.S. Strained joke along these lines can be found at gnosis.cx/.





From tim.one at home.com  Thu Mar 29 18:28:52 2001
From: tim.one at home.com (Tim Peters)
Date: Thu, 29 Mar 2001 18:28:52 -0500
Subject: diff lists
In-Reply-To: <m3zoe4b2tu.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <LNBBLJKPBEHFEDALKOLCOEDAJJAA.tim.one@home.com>

[Michael Hudson]
> Tangentially, does anyone know of any good algorithms for "edit
> distance" between two sequences?  E.g. if I have
>
> "abcdef"
>
> and want to get to
>
> "abQUACKcde"
>
> I want to get the answer back "insert 'QUACK' at position 3 and delete
> a character at position 11".

See std library module difflib.py (new in 2.1).  This packages the
SequenceMatcher class from the heart of the Tools/scripts/ndiff.py tool for
easy reuse.

Python 2.1b2 (#12, Mar 23 2001, 14:01:30) [MSC 32 bit (Intel)] on win32
Type "copyright", "credits" or "license" for more information.
IDLE 0.8 -- press F1 for help
>>> import difflib
>>> a = "abcdef"
>>> b = "abQUACKcde"
>>> m = difflib.SequenceMatcher(None, a, b)
>>> for tag, i1, i2, j1, j2 in m.get_opcodes():
	print ("%7s a[%d:%d] (%s) b[%d:%d] (%s)" %
                   (tag, i1, i2, a[i1:i2], j1, j2, b[j1:j2]))

  equal a[0:2] (ab) b[0:2] (ab)
 insert a[2:2] () b[2:7] (QUACK)
  equal a[2:5] (cde) b[7:10] (cde)
 delete a[5:6] (f) b[10:10] ()
>>>

It does not attempt to produce minimal edit sequences, but does attempt to
produce edit sequences that "look right" to people (read the comments in
ndiff.py for more on that).

> "Good" means "pretty quick", here.

For suitably large values of "pretty", sure <wink>.

not-for-all-though-ly y'rs  - tim




From qrczak at knm.org.pl  Tue Mar 13 17:15:25 2001
From: qrczak at knm.org.pl (Marcin 'Qrczak' Kowalczyk)
Date: 13 Mar 2001 22:15:25 GMT
Subject: This math scares me
References: <3aad4062.1290021@News.CIS.DFN.DE> <20010312170356.B3513@better.net> <mailman.984480856.18097.python-list@python.org> <984486395.407379@newsmaster-04.atnet.at> <slrn9asf1d.ja3.qrczak@qrnik.zagroda> <984518199.127140@newsmaster-04.atnet.at>
Message-ID: <slrn9at6vt.h63.qrczak@qrnik.zagroda>

Tue, 13 Mar 2001 22:11:53 +0100, Werner Schiendl <ws-news at gmx.at> pisze:

> 15 to 16 digits are the precision double values will provide. So
> rounding to 15 digits will normally display the value you have
> entered. There is however no guarantee that your value is displayed
> exactly.

That's why I am asking for not fixing the number of digits but having
this functionality:

    def most_useful_str_or_repr_variant(x):
        s = str(x)
        if float(s) == x: return s
        return repr(x)

-- 
 __("<  Marcin Kowalczyk * qrczak at knm.org.pl http://qrczak.ids.net.pl/
 \__/
  ^^                      SYGNATURA ZAST?PCZA
QRCZAK


From NoSpam at NoSpam.com  Fri Mar  2 18:40:52 2001
From: NoSpam at NoSpam.com (Tom)
Date: Fri, 02 Mar 2001 23:40:52 GMT
Subject: Unresolved symbols in MSVCRT.DLL when running a py2exe packaged program.
References: <mailman.983571904.9828.python-list@python.org>
Message-ID: <8cWn6.16791$UZ4.4554807@news4.rdc1.on.home.com>

"Tim Peters" <tim.one at home.com> wrote in message
news:mailman.983571904.9828.python-list at python.org...
> [Tom, on System File Protection in recent flavors of Windows]
> > I didn't mean to suggest 'fooling' it - I agree that safeguards
> > like SFP are needed - I meant that one had to do it according to
> > the rules.
>
> What rules?  Only MS can replace system files now -- there is no provision
> for "one" to *ever* do it anymore, if one != MS (no API, no magic dance,
not
> even an option to turn SFP off).  One of the links I posted before
suggests
> that they *may* make an exception for developers of new hardware.

The API that I was referring to was the Setup API (see Platform SDK).  I
must admit that I've only used it for replacing drivers (I used
InstallShield for replacing DLL's), but I had thought it could be used for
replacing DLL's.

Tom.

> > As for not being able to update the DLL's, this would be a problem.  If
a
> > service pack for VC6 updates the MFC or CRT DLL's (which SP's usually
do),
> > and you ship an app build with it, then you must update these DLL's on
your
> > target system.  Perhaps this is where the new ability to run 'personal'
> > instances of DLL's comes in: ie. you can't update the system32 version
the
> > DLL, but you can install your own personal copy.
>
> That's my understanding, yes.  As one of the links pointed out, .inf files
> don't have conditionals, so MS expects you to ship different infs for
> different MS OSes now (depending on whether you can or cannot exploit
private
> system DLLs on the OS in question).
>
> Python hasn't needed to deal with that yet.  Wrestling with Windows
esoterica
> isn't really a passion of mine, so I'm not looking forward to the day when
it
> must.
>
> luckily-i'm-told-everyone-will-be-on-linux-then-anyway-ly y'rs  - tim
>
>




From mrq at for.mat.bham.ac.uk  Fri Mar 30 04:50:54 2001
From: mrq at for.mat.bham.ac.uk (Martyn Quick)
Date: Fri, 30 Mar 2001 10:50:54 +0100
Subject: BLT extensions with BeOpen Python 2.0
In-Reply-To: <Pine.SOL.4.30.0103291832080.13267-100000@primrose.csv.warwick.ac.uk>
References: <Pine.GSO.4.21.0103291300160.11408-100000@babbage> <Pine.SOL.4.30.0103291832080.13267-100000@primrose.csv.warwick.ac.uk>
Message-ID: <Pine.GSO.4.21.0103301049410.16891-100000@babbage>

On Thu, 29 Mar 2001, John J. Lee wrote:

> On Thu, 29 Mar 2001, Martyn Quick wrote:
> 
> > I've been trying to install the BLT extensions for Tk into my BeOpen
> > distribution of Python 2.0.  However, it seems that the directory
> > structure that the BLT distribution expects (see www.tcltk.com/blt) is
> > different to that provided by the python distribution.
> >
> > Has anyone succeeded in installing BLT with the distribution, or do I need
> > to switch distributions?  (and to which one?)
> 
> I'm pretty sure I had BLT working with BeOpen 2.0 at some point, at least
> on linux, and perhaps windows too.  I couldn't tell you exactly where I
> put it, though.  Make sure you have the right BLT version for the tcl that
> comes with python 2.0 .

This reply has pointed out to me that I ought to say what platform I was
using.  I'm running on Windows 95, so downloaded the precompiled version
of BLT for Tcl8.3 (which is the version in BeOpen 2.0).

Martyn

--------------------------------------------------------
Dr. Martyn Quick  (Research Fellow in Pure Mathematics)
University of Birmingham, Edgbaston, Birmingham, UK.
http://www.mat.bham.ac.uk/M.R.Quick





From mcfletch at home.com  Sat Mar 17 22:01:10 2001
From: mcfletch at home.com (Mike C. Fletcher)
Date: Sat, 17 Mar 2001 22:01:10 -0500
Subject: module to do euler path and topological sorting in python
In-Reply-To: <CUTs6.80$gB1.365@typhoon.nyu.edu>
Message-ID: <000001c0af57$b005a770$a9807018@cr706570a>

Not sure what a euler path is, but here's two algos that do topological
sorting.  Guts of the second (toposort) are from Tim Peters, first (sort) is
one I built myself, so expect it to have problems :) .  I actually built the
second after the first without realising I was creating a toposort (I was
trying to determine parallelism in a "content compiler"), it's just included
for posterity :o) .

Enjoy,
Mike

8<______________ toposort.py _________________
class RecursionError( OverflowError, ValueError ):
    '''Unable to calculate result because of recursive structure'''


def sort(nodes, routes, noRecursion=1):
    '''Passed a list of node IDs and a list of source,dest ID routes
    attempt to create a list of stages where each sub list
    is one stage in a process.
    '''
    children, parents = _buildChildrenLists(routes)
    # first stage is those nodes
    # having no incoming routes...
    stage = []
    stages = [stage]
    taken = []
    for node in nodes:
        if (not parents.get(node)):
            stage.append (node)
    if nodes and not stage:
        # there is no element which does not depend on
        # some other element!!!
        stage.append( nodes[0])
    taken.extend( stage )
    nodes = filter ( lambda x, l=stage: x not in l, nodes )
    while nodes:
        previousStageChildren = []
        nodelen = len(nodes)
        # second stage are those nodes
        # which are direct children of the first stage
        for node in stage:
            for child in children.get (node, []):
                if child not in previousStageChildren and child not in
taken:
                    previousStageChildren.append(child)
                elif child in taken and noRecursion:
                    raise RecursionError( (child, node) )
        # unless they are children of other direct children...
        # TODO, actually do that...
        stage = previousStageChildren
        removes = []
        for current in stage:
            currentParents = parents.get( current, [] )
            for parent in currentParents:
                if parent in stage and parent != current:
                    # might wind up removing current...
                    if not current in parents.get(parent, []):
                        # is not mutually dependent...
                        removes.append( current )
        for remove in removes:
            while remove in stage:
                stage.remove( remove )
        stages.append( stage)
        taken.extend( stage )
        nodes = filter ( lambda x, l=stage: x not in l, nodes )
        if nodelen == len(nodes):
            if noRecursion:
                raise RecursionError( nodes )
            else:
                stages.append( nodes[:] )
                nodes = []
    return stages

def _buildChildrenLists (routes):
    childrenTable = {}
    parentTable = {}
    for sourceID,destinationID in routes:
        currentChildren = childrenTable.get( sourceID, [])
        currentParents = parentTable.get( destinationID, [])
        if not destinationID in currentChildren:
            currentChildren.append ( destinationID)
        if not sourceID in currentParents:
            currentParents.append ( sourceID)
        childrenTable[sourceID] = currentChildren
        parentTable[destinationID] = currentParents
    return childrenTable, parentTable


def toposort (nodes, routes, noRecursion=1):
    '''Topological sort from Tim Peters, fairly efficient
    in comparison (it seems).'''
    #first calculate the recursion depth

    dependencies = {}
    inversedependencies = {}
    if not nodes:
        return []
    if not routes:
        return [nodes]
    for node in nodes:
        dependencies[ node ] = (0, node)
        inversedependencies[ node ] = []


    for depended, depends in routes:
        # is it a null rule
        try:
            newdependencylevel, object = dependencies.get ( depends, (0,
depends))
        except TypeError:
            print depends
            raise
        dependencies[ depends ] = (newdependencylevel + 1,  depends)
        # "dependency (existence) of depended-on"
        newdependencylevel,object = dependencies.get ( depended, (0,
depended) )
        dependencies[ depended ] = (newdependencylevel, depended)
        # Inverse dependency set up
        dependencieslist = inversedependencies.get ( depended, [])
        dependencieslist.append (depends)
        inversedependencies[depended] = dependencieslist
    ### Now we do the actual sorting
    # The first task is to create the sortable
    # list of dependency-levels
    sortinglist = dependencies.values()
    sortinglist.sort ()
    output = []
    while sortinglist:
        deletelist = []
        generation = []
        output.append( generation)
        while sortinglist and sortinglist[0][0] == 0:
            number, object = sortinglist[0]
            generation.append ( object )
            deletelist.append( object )
            for inverse in inversedependencies.get(object, () ):
                try:
                    oldcount, inverse = dependencies [ inverse]
                    if oldcount > 0:
                        # will be dealt with on later pass
                        dependencies [ inverse] = (oldcount-1, inverse)
                    else:
                        # will be dealt with on this pass,
                        # so needs not to be in the sorting list next time
                        deletelist.append( inverse )
                    # just in case a loop comes through
                    inversedependencies[object] = []
                except KeyError:
                    # dealing with a recursion-breaking run...
                    pass
            del sortinglist [0]
        # if no elements could be deleted, then
        # there is something which depends upon itself
        if not deletelist:
            if noRecursion:
                raise RecursionError( sortinglist )
            else:
                # hack so that something gets deleted...
##              import pdb
##              pdb.set_trace()
                dependencies[sortinglist[0][1]] = (0,sortinglist[0][1])
        # delete the items that were dealt with
        for item in deletelist:
            try:
                del dependencies [ item ]
            except KeyError:
                pass
        # need to recreate the sortinglist
        sortinglist = dependencies.values()
        if not generation:
            output.remove( generation )
        sortinglist.sort ()
    return output





if __name__ == "__main__":

    import pprint, traceback
    nodes= [ 0,1,2,3,4,5 ]
    testingValues = [
        [ (0,1),(1,2),(2,3),(3,4),(4,5)],
        [ (0,1),(0,2),(1,2),(3,4),(4,5)],
        [
        (0,1),
        (0,2),
        (0,2),
                    (2,4),
                    (2,5),
                (3,2),
        (0,3)],
        [
        (0,1), # 3-element cycle test, no orphan nodes
        (1,2),
        (2,0),
                    (2,4),
                    (2,5),
                (3,2),
        (0,3)],
        [
        (0,1),
        (1,1),
        (1,1),
                (1,4),
                (1,5),
                (1,2),
        (3,1),
        (2,1),
        (2,0)],
        [
            (0,1),
            (1,0),
            (0,2),
            (0,3),
        ],
        [
            (0,1),
            (1,0),
            (0,2),
            (3,1),
        ],
    ]
    print 'sort, no recursion allowed'
    for index in range(len(testingValues)):
##      print '    %s -- %s'%( index, testingValues[index])
        try:
            print '        ', sort( nodes, testingValues[index] )
        except:
            print 'exception raised'
    print 'toposort, no recursion allowed'
    for index in range(len(testingValues)):
##      print '    %s -- %s'%( index, testingValues[index])
        try:
            print '        ', toposort( nodes, testingValues[index] )
        except:
            print 'exception raised'
    print 'sort, recursion allowed'
    for index in range(len(testingValues)):
##      print '    %s -- %s'%( index, testingValues[index])
        try:
            print '        ', sort( nodes, testingValues[index],0 )
        except:
            print 'exception raised'
    print 'toposort, recursion allowed'
    for index in range(len(testingValues)):
##      print '    %s -- %s'%( index, testingValues[index])
        try:
            print '        ', toposort( nodes, testingValues[index],0 )
        except:
            print 'exception raised'



-----Original Message-----
From: cyberian bear [mailto:cyberian_bear at hotmail.com]
Sent: Saturday, March 17, 2001 8:08 PM
To: python-list at python.org
Subject: module to do euler path and topological sorting in python


can anyone suggest where i can find the modules which would do those two
things.
I have found a  website which had a link to topological sorting but it says
that access is forbidden. www.python.org doesn't have any matching reuslts.
thanx
cb


--
http://mail.python.org/mailman/listinfo/python-list




From loewis at informatik.hu-berlin.de  Thu Mar  8 05:30:06 2001
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 08 Mar 2001 11:30:06 +0100
Subject: LONG_BIT error when compiling python2.0
References: <m14rx4d4wi.fsf@gepc4223.it.marconicomms.com>
Message-ID: <j4pufsmu35.fsf@informatik.hu-berlin.de>

Enrico Spinielli <enrico.spinielli at marconi.com> writes:

> I am trying to install python2.0 on RedHat 7.0 Linux.
[...]
> Any idea about a possible solution?
> In the meanwhile I'll install gcc2.96-69 and retry...

Upgrade glibc. When doing so, complain to Redhat for shipping
prerelease glibc snapshots with Redhat 7.

Regards,
Martin



From aleaxit at yahoo.com  Fri Mar 30 07:42:27 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 30 Mar 2001 14:42:27 +0200
Subject: PEP 245: Python interfaces
References: <Pine.LNX.4.32.0103222253510.26302-100000@localhost.localdomain> <99o9g4$nh3$0@216.39.170.247> <Pine.LNX.4.21.0103261430360.25260-100000@localhost.localdomain> <j4wv99m503.fsf@informatik.hu-berlin.de> <mailman.985899166.8262.python-list@python.org>
Message-ID: <9a1v0r030up@news1.newsguy.com>

"Carlos Alberto Reis Ribeiro" <cribeiro at mail.inet.com.br> wrote in message
news:mailman.985899166.8262.python-list at python.org...
    [snip]
> For example, there is multiple inheritance. I know very few legitimate
> cases where it is a good idea. Composition usually attains the same
> results, but with less side effects and much better control over the
> process. Interfaces are exceptionally well suited for this kind of
construct.

Multiple inheritance of implementation (mixins) is a highly
convenient and practical programming technique; dunno what
you mean by "legitimate" cases in this context?

Implementation by one object of multiple interfaces (which
is also expressed as multiple-inheritance in C++) is more
than just convenient -- it's a must to enable most decent
modeling techniques ("not relying on absolute taxonomy" is
what I mean by "decent" here:-).

The two issues are quite separate, of course.


Alex





From bill-bell at bill-bell.hamilton.on.ca  Fri Mar 23 13:11:45 2001
From: bill-bell at bill-bell.hamilton.on.ca (Bill Bell)
Date: Fri, 23 Mar 2001 13:11:45 -0500
Subject: TimeoutSocket exceptions handling
In-Reply-To: <985369104.693.34698.l7@yahoogroups.com>
Message-ID: <3ABB4B91.23301.370A1C@localhost>

I'm new to Python. In an effort to understand why I can't get hold of 
an exception from O'Malley's TimeoutSocket I wrote the following two 
teeny-tiny scripts.

Script2.py:

	import Script1

	try:
	    myFunc()
	except myErr, (reason):
	    print 'myErr trapped: ', reason

Script1.py:

	class myErr ( Exception ):
	    pass

	def myFunc():
	    raise myErr ( 'myErr reason' )

Couldn't be much simpler, huh? And they even seem to work.

Timeoutsocket.py (looks fantastic, thanks Alex Martelli), contains the 
following:

...
class Timeout(Exception):
    pass
...
        raise Timeout("Attempted connect to %s timed out." % 
str(addr) )

My code that uses TimeoutSocket has

        try:        
            s = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
            s.connect((self.whoisserver, 43))
            s.send("%s\n" % self.domain)
            page = ""
            while 1:
                data = s.recv(8196)
                if not data: break
                page = page + data
                pass
            s.close()
        except Timeout, (reason):
            raise 'TimedOut', 'Whois server timedout'

If my sanity has not departed me this combination appears similar to 
my would-be autodidactic efforts.

However, not so: Python says "NameError: There is no variable 
named 'Timeout'" for my 'except' statement.

I'd appreciate a word of advice, or a pointer to code that works.

Bill



From paulp at ActiveState.com  Mon Mar 12 20:45:51 2001
From: paulp at ActiveState.com (Paul Prescod)
Date: Mon, 12 Mar 2001 17:45:51 -0800
Subject: FOLLOWUPS!!!!!!!
References: <Pine.LNX.4.10.10103121801530.7351-100000@falcon.physics.wisc.edu>
Message-ID: <3AAD7BCF.4D4F69B7@ActiveState.com>

Please keep follow-ups to just types-sig. I'm very sorry I cross-posted
in the beginning and I apologize to everyone on multiple lists. I did
direct people to follow up only to types-sig but I should have used a
header....or separate posts!

-- 
Python:
    Programming the way
    Guido
    indented it.
       - (originated with Skip Montanaro?)



From gerhard.nospam at bigfoot.de  Tue Mar  6 01:29:05 2001
From: gerhard.nospam at bigfoot.de (Gerhard =?iso-8859-1?Q?H=E4ring?=)
Date: Tue, 6 Mar 2001 07:29:05 +0100
Subject: Python & mySQL
References: <DJQo6.27114$c7.7825529@news3.rdc1.on.home.com>
Message-ID: <slrn9a91p8.id.gerhard.nospam@lilith.hqd-internal>

On Mon, 05 Mar 2001 18:16:03 GMT, Colin Meeks <colinmeeks at home.com> wrote:
>Does anybody have any pointers for integrating mySQL with Python.  I've been
>using Python a while and am looking to develop a web backend using Python
>with mySQL.  My site is hosted by a Linux based ISP who has both Python and
>mySQL but they don't currently have a link between the two.  I'm using
>Windows2000 as my development base, so I want something that is usuable on
>both, so it's easier for me to track any problems.

You'll need the MySQLdb python module. A prebuilt Windows version is
available (at my homepage, see the sig).

If your ISP isn't so friendly and won't install the MySQL module for you,
you can still do that yourself if you have the correct binary version. For
Linux, you should only need to copy the following files in your cgi-bin
directory:

CompatMysqldb.py
CompatMysqldb.pyc
MySQLdb.py
MySQLdb.pyc
_mysql.so

I have done this successfully under FreeBSD and SunOS, where the ISP were
not flexible enough to install a Python module.

>
>Ideally I want something that doesn't involved recompiling Python if that's
>possible, 

You should never need to recompile Python for an extension module.

>as i want the installation to be as straightforward for my ISP and
>of course me :-)

As simple as
    python setup.py build
    python setup.py install
?

Gerhard
-- 
mail:   gerhard <at> bigfoot <dot> de
web:    http://highqualdev.com


From carribeiro at yahoo.com  Sun Mar 18 21:02:56 2001
From: carribeiro at yahoo.com (Carlos Ribeiro)
Date: Sun, 18 Mar 2001 23:02:56 -0300
Subject: Who am I: can a class instance determine its own name?
Message-ID: <5.0.2.1.0.20010318225613.022cc670@pop.mail.yahoo.com>

Alex Martelli wrote:
>(My dream is actually of a collection of bits in an attribute...

While I agree on the *principle*, I disagree on the details. Thinking about 
this as a bitmap is an artifact of historic optimization techniques. Worse 
- the performance gain is highly improbable in scripting languages like 
Python, and should be negligibe in most cases.  Any exception to this 
probably deserve to be implemented in C or some similar, highly optimized 
compiled language.

OTOH, having a set of attributes would make the code more readable and 
extensible. (the set of attributes is itself in a dictionary and would also 
be subject to the same control).


-------------------------------
Carlos Ribeiro
cribeiro-AT-mail-inet-com-br



_________________________________________________________
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com




From bowman at montana.com  Sun Mar 11 08:29:29 2001
From: bowman at montana.com (bowman)
Date: Sun, 11 Mar 2001 06:29:29 -0700
Subject: "!=" is bad form. Re: sorry....never mind
References: <LNBBLJKPBEHFEDALKOLCCELJJEAA.tim.one@home.com> <mailman.984315558.26130.python-list@python.org>
Message-ID: <l6Mq6.175$VG2.1073@newsfeed.slurp.net>

J?rgen A. Erhard <juergen.erhard at gmx.net> wrote in message
>
> So when can we expect `||' and `&&' to replace `or' and `and'?
> Shouldn't we `Beware of the slippery slope'?

only if 'or' has lower precedence than '||'  to help in evaluating 'die'
expressions.




From ransen_spam_me_not at nemo.it  Fri Mar  2 13:40:37 2001
From: ransen_spam_me_not at nemo.it (Owen F. Ransen)
Date: Fri, 02 Mar 2001 18:40:37 GMT
Subject: Bug in Py_Finalize() ?
References: <3a9f636d.2908314@news.newsguy.com>
Message-ID: <3aa0c6e5.809467@news.newsguy.com>

I tried to report this bug to sorgeforge, signed up, but
it kept on giving me page not found errors after I logged
in....


On Fri, 02 Mar 2001 09:32:12 GMT, ransen_spam_me_not at nemo.it (Owen F.
Ransen) wrote:

>Trying to find a bug in my C extension I ended up writing
>this loop:
>
>    for (UINT i = 0 ; i < 10 ; i++) {
>        Py_Initialize();
>        Py_Finalize() ;
>    }
>
>which crashes Python when i=8 with the message:
>Windows 32, VC++ program.
>
>Fatal Python error: UNREF invalid object
>
>My understanding is that I can use Py_Initialize() and
>Py_Finalize() in pairs to free up memory and to start
>with a clean interpreter each time I need it. And that
>is why I'd like to use Py_Finalise....

--
Owen F. Ransen
http://www.ransen.com/
Home of Gliftic & Repligator Image Generators


From pearu at cens.ioc.ee  Mon Mar  5 10:17:38 2001
From: pearu at cens.ioc.ee (Pearu Peterson)
Date: Mon, 5 Mar 2001 17:17:38 +0200
Subject: Name of the Script
In-Reply-To: <FpmjOjvwe5bm0Bg09LDTMPhh5vji@4ax.com>
References: <20010304.185354.2114937732.8277@localhost.localdomain> <NJCo6.6720$qv3.2834377@nnrp5-w.sbc.net> <FpmjOjvwe5bm0Bg09LDTMPhh5vji@4ax.com>
Message-ID: <Pine.LNX.4.21.0103051715150.18263-100000@kev.ioc.ee>

On Mon, 5 Mar 2001, Paul Moore wrote:

> 
> This doesn't always include full pathname information. On Windows,
> 
> --- a.py ---
> import sys
> print sys.argv[0]
> ---
> 
> python a.py
> 
> just prints "a.py". Is there a way of getting the *full* script name.
> For example, so that the script can locate a data file which is held
> in the same directory as the script itself.

--- a.py ---
import os,sys
print os.path.abspath(sys.argv[0])
---
might be what you want.

Pearu



From claird at starbase.neosoft.com  Thu Mar 22 17:28:50 2001
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 22 Mar 2001 16:28:50 -0600
Subject: Why Python/Jython?
References: <99cl9e$b2b$1@Merlin.et.bocholt.fh-gelsenkirchen.de> <mailman.985260089.28444.python-list@python.org> <1901D50DC30D30E4.16551C1AB0FB4363.C0D72E208759F8A7@lp.airnews.net> <99dkg301894@news1.newsguy.com>
Message-ID: <D3BEEB21C25EDB86.8F1F23B32E6A0452.393DE27BCBA8899F@lp.airnews.net>

In article <99dkg301894 at news1.newsguy.com>,
Alex Martelli <aleaxit at yahoo.com> wrote:
>"Cameron Laird" <claird at starbase.neosoft.com> wrote in message
>news:1901D50DC30D30E4.16551C1AB0FB4363.C0D72E208759F8A7 at lp.airnews.net...
>    [snip]
>>     meaningful.  People do NOT start learning
>>     Java lightly; it often takes hours just to
>>     install and configure a minimal JDK, let
>
>*blink* -- I'm hardly a Java-ite, myself, but
>"install and configure a minimal JDK" I would
>estimate as typically a 10-minute job (on, say,
>NT -- is it an order of magnitude harder on,
>say, Linux?!).  Maybe I misunderstand what
>you mean...?
>
>I'll underwrite the rest of your post (haven't
>had occasion to use Jython 'in anger', yet, but
>from just playing with it it does seem a great
>way to exploit the huge libraries available in
>and for Java), but this point seems strange.
>
>
>Alex
>
>
>

As Andrew mentioned nearby, as confusing as it can be for
a newcomer to Python to figure out who and where the players
are, Java manages to dwarf this.  Java is MUCH less compat-
ible between versions, Java downloads are much, MUCH bigger,
and Sun makes it (unless something's changed recently) a
serious nuisance to locate binaries for anything other than
Win*.  Those little annoyances do indeed multiply out to an
order of magnitude, in many common situations ("Oh, you only
have 128 Mb of main memory?  Too bad; start over.") (my main
machine continues to do fine with 32 Mb, by the way).
-- 

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


From sdm7g at virginia.edu  Tue Mar 13 17:50:33 2001
From: sdm7g at virginia.edu (Steven D. Majewski)
Date: Tue, 13 Mar 2001 17:50:33 -0500 (EST)
Subject: This math scares me
In-Reply-To: <yTwr6.1235$Tg.167429@ruti.visi.com>
Message-ID: <Pine.NXT.4.21.0103131738310.262-100000@localhost.virginia.edu>


On Tue, 13 Mar 2001, Grant Edwards wrote:

> In article <jwbnews-09D75F.13313313032001 at corp.supernews.com>, John W. Baxter wrote:
> >In article <984518199.127140 at newsmaster-04.atnet.at>,
> > "Werner Schiendl" <ws-news at gmx.at> wrote:
> >
> >Bill Gates got this right in the early days, delivering two
> >Basic interpreters: binary floating point for speed, and
> >decimal floating point for money and such.  
> 
> I don't know if you can credit Gates with doing anything other
> than following the pack.  Back when I first started doing
> software (CP/M 1.7), all language packages for "personal"
> computers came with both BCD and binary FP math libraries so
> you could do exact monetary calculations if you wanted.

Exactly what pack do you believe Gates was following ? 

You do know (don't you?) that Bill's BASIC predates not just
MS-DOS (which microsoft bought from another company) but CP/M
and just about any other "PC" software (and most of the hardware)
since the first version was written to run on the Altair computer --
which was the first microprocessor based personal computer. 
( I think you probably had to boost the memory up to 4K to get 
  Basic to run. ) 

> Of course we'd never imagined we'd ever see hardware FP and BCD
> libraries were only marginally slower than binary ones. Now
> that FP is usually in hardware, BCD FP (or fixed point) will be
> pretty slow by comparison.

-- Steve Majewski





From Simon.Budig at unix-ag.org  Fri Mar 16 18:26:42 2001
From: Simon.Budig at unix-ag.org (Simon Budig)
Date: 17 Mar 2001 00:26:42 +0100
Subject: list.sort(cmpfunc) question
References: <m1ae6lg5qt.fsf@pclabs188.labs.trema.com> <98t9dq$qvu$1@panix6.panix.com>
Message-ID: <3ab2a132@si-nic.hrz.uni-siegen.de>

Aahz Maruch <aahz at panix.com> wrote:
> In article <m1ae6lg5qt.fsf at pclabs188.labs.trema.com>,
> Harri Pasanen  <harri at trema.com> wrote:
>>
>>The list.sort(cmp) docs for Python 2.0 state:
>>
>>   The sort() method takes an optional argument specifying a
>>   comparison function of two arguments (list items) which should
>>   return -1, 0 or 1 depending on whether the first argument is
>>   considered smaller than, equal to, or larger than the second
>>   argument.
> 
> Note that "should" carefully.  The current implementation of sort() only
> checks for a negative value, but that's not a behavior that should be
> relied on in your coding.  You can be sure that any anamolies in the
> behavior of cmp() will be fixed if the sort() algorithm ever changes.

As a related question since I did not find it in the documentation:

Is the sort-function guaranteed to be stable? Does it keep the order,
when cmpfunc returns 0 for a pair of items?

Thanks,
        Simon
-- 
      Simon.Budig at unix-ag.org       http://www.home.unix-ag.org/simon/


From jwbnews at scandaroon.com  Fri Mar  2 17:50:22 2001
From: jwbnews at scandaroon.com (John W. Baxter)
Date: Fri, 02 Mar 2001 14:50:22 -0800
Subject: gah! I hate the new string syntax
References: <mailman.983485923.1136.python-list@python.org> <B7In6.16281$Qb7.2658902@newsb.telia.net> <97ohh2$polt0$1@ID-77331.news.dfncis.de> <97opsb$nhu$1@nntp6.u.washington.edu>
Message-ID: <jwbnews-7D01E3.14502002032001@corp.supernews.com>

In article <97opsb$nhu$1 at nntp6.u.washington.edu>,
 Donn Cave <donn at u.washington.edu> wrote:

> I bet when people see ";".join(...) with fresh eyes, they still
> find it strained.  This is not about the old school vs. the new,
> it's a serious question of how people understand computer programming
> languages.  Is it like natural language?  Can we think about it in
> sort of mathematical terms, like symmetry etc.?  Does everyone handle
> it basically the same way?

Take two APLs, or a liberal dose of John Iverson, and all will be well.  
;-)

Fortunately, it's not the only way to join.

  --John

-- 
John W. Baxter   Port Ludlow, WA USA  jwbnews at scandaroon.com


From gregor at mediasupervision.de  Thu Mar  8 10:14:15 2001
From: gregor at mediasupervision.de (Gregor Hoffleit)
Date: Thu, 8 Mar 2001 16:14:15 +0100
Subject: News from Python9!!!
In-Reply-To: <NUCp6.65206$Z8.11697516@typhoon.southeast.rr.com>; from tuttledon@hotmail.com on Thu, Mar 08, 2001 at 03:21:17AM +0000
References: <NUCp6.65206$Z8.11697516@typhoon.southeast.rr.com>
Message-ID: <20010308161415.C30277@mediasupervision.de>

On Thu, Mar 08, 2001 at 03:21:17AM +0000, Don Tuttle wrote:
> 4)And a bugfix release of Python 2.0 has been released. It includes PPM
> (Programmer's Package Manager ) to make finding and installing Python
> modules simple.

Shouldn't this read 'bugfix release of ActivePython 2.0' ?

    Gregor



From ws-news at gmx.at  Thu Mar  8 05:26:19 2001
From: ws-news at gmx.at (Werner Schiendl)
Date: Thu, 8 Mar 2001 11:26:19 +0100
Subject: learning C
References: <mailman.983972829.23481.python-list@python.org> <983975751.18833@newsmaster-04.atnet.at> <slrn9adklr.bv9.scarblac@pino.selwerd.nl> <mailman.984014483.6974.python-list@python.org>
Message-ID: <984047460.29247@newsmaster-04.atnet.at>

D-Man <dsh8290 at rit.edu> wrote in message
news:mailman.984014483.6974.python-list at python.org...
> On Thu, Mar 08, 2001 at 12:32:47AM +0000, Remco Gerlich wrote:
> | Werner Schiendl <ws-news at gmx.at> wrote in comp.lang.python:
> | > If it is not obsolete, it should be at least considered deprecated for
new
> | > projects.
> |
> | A baseless statement. There are many types of project for which C is
still
> | the best choice. For one thing, think embedded systems.
>
> C does have it strong points.  After all, python is essentially a C
> program.
>

And it will compile with a C++ compiler, I think this is in fact done.
At least on Windows, where the MS-DEV environment is used, AFAIK.

So I think, in common use C++ _did_ already replace C, whether you use the
new features or not.
Some platforms even allow C++ statements in Files ending in .c.
Otherwise the same compiler is used, just with some internal things disabled
to make it appear C.

> | I can imagine something like Java deprecating C++, but C isn't going to
go
> | away soon.
>
> Java?  I thought Python took care of that.
>
> ;-)
>
> -D
>
>




From barry at digicool.com  Mon Mar 12 18:15:15 2001
From: barry at digicool.com (Barry A. Warsaw)
Date: Mon, 12 Mar 2001 18:15:15 -0500
Subject: [Python-Dev] Revive the types sig? 
References: <jeremy@alum.mit.edu>
	<15020.9404.557943.164934@w221.z064000254.bwi-md.dsl.cnc.net>
	<200103120711.AAA09711@localhost.localdomain>
Message-ID: <15021.22659.616556.298360@anthem.wooz.org>

>>>>> "UO" == Uche Ogbuji <uche.ogbuji at fourthought.com> writes:

    UO> I know this isn't the types SIG and all, but since it has come
    UO> up here, I'd like to (once again) express my violent
    UO> disagreement with the efforts to add static typing to Python.
    UO> After this, I won't pursue the thread further here.

Thank you Uche!  I couldn't agree more, and will also try to follow
your example, at least until we see much more concrete proposals from
the types-sig.  I just want to make a few comments for the record.

First, it seemed to me that the greatest push for static type
annotations at IPC9 was from the folks implementing Python on top of
frameworks other than C.  I know from my own experiences that there is
the allure of improved performance, e.g. JPython, given type hints
available to the compiler.  While perhaps a laudable goal, this
doesn't seem to be a stated top priority of Paul's.

Second, if type annotations are to be seriously considered for
inclusion in Python, I think we as a community need considerable
experience with a working implementation.  Yes, we need PEPs and specs
and such, but we need something real and complete that we can play
with, /without/ having to commit to its acceptance in mainstream
Python.  Therefore, I think it'll be very important for type
annotation proponents to figure out a way to allow people to see and
play with an implementation in an experimental way.

This might mean an extensive set of patches, a la Stackless.  After
seeing and talking to Neil and Andrew about PTL and Quixote, I think
there might be another way.  It seems that their approach might serve
as a framework for experimental Python syntaxes with minimal overhead.
If I understand their work correctly, they have their own compiler
which is built on Jeremy's tools, and which accepts a modified Python
grammar, generating different but compatible bytecode sequences.
E.g., their syntax has a "template" keyword approximately equivalent
to "def" and they do something different with bare strings left on the
stack.

The key trick is that it all hooks together with an import hook so
normal Python code doesn't need to know anything about the mechanics
of PTL compilation.  Given a homepage.ptl file, they just do an
"import homepage" and this gets magically transformed into a .ptlc
file and normal Python objects.

If I've got this correct, it seems like it would be a powerful tool
for playing with alternative Python syntaxes.  Ideally, the same
technique would allow the types-sig folks to create a working
implementation that would require only the installation of an import
hook.  This would let them build their systems with type annotation
and prove to the skeptical among us of their overwhelming benefit.

Cheers,
-Barry



From michael at lindrum.de  Tue Mar 13 17:02:18 2001
From: michael at lindrum.de (Michael Lindrum)
Date: Tue, 13 Mar 2001 23:02:18 +0100
Subject: Python2C
Message-ID: <3AAE98EA.2754A358@lindrum.de>

I'm searching for the lib that is necessary to compile C-Code generated
by Python2C.
OS: SunOS
Who knows where I can find it?



From digitig at cix.co.uk  Wed Mar 14 19:42:00 2001
From: digitig at cix.co.uk (Tim Rowe)
Date: Thu, 15 Mar 2001 00:42 +0000 (GMT Standard Time)
Subject: Magnitude of the wx* market (was: Python In A Nutshell - suggestions)
References: <A909AD29CBD32AFC.4F1641C135749F87.7149617C7C166896@lp.airnews.net>
Message-ID: <memo.20010315004230.1527B@digitig.compulink.co.uk>

In article 
<A909AD29CBD32AFC.4F1641C135749F87.7149617C7C166896 at lp.airnews.net>, 
claird at starbase.neosoft.com (Cameron Laird) wrote:

> In article <3AAA4FE3.B471B689 at northwestern.edu>,
> Louis Luangkesorn  <lluang at northwestern.edu> wrote:
> 			.
> 			.
> 			.
> >And while I'm at it, one useful book to have would be on 
> wxWindows/wxPython.  I see references to it in several places, and a 
> few intros on the web, but
> 			.
> 			.
> 			.
> This is an interesting point.  It's typical in computer
> book publishing that publishers want more than authors
> can produce; roughly, someone like ORA would happily
> print many more titles than they can find authors willing
> and able to write.
> 
> I happen to know, though, that several publishers have
> rejected wxWindows proposals.
> 
> I doubt, though, that there'd be much trouble making the
> case that the subject deserves a chapter in a larger work.

And the larger work in question is "Python Programming on Win32" by Mark 
Hammond & Andy Robinson, already published by ORA, so I doubt they'd be 
particularly interested in another! It doesn't have full documentation of 
the wxPython classes, though, just a tutorial introduction. Maybe the way 
to go is to get them to extend it in the next edition?


From bsb at winnegan.de  Sun Mar 18 18:06:30 2001
From: bsb at winnegan.de (Siggy Brentrup)
Date: 19 Mar 2001 00:06:30 +0100
Subject: NOT A scope problem
In-Reply-To: <993bar$7h5$1@news.chatlink.com>
References: <993bar$7h5$1@news.chatlink.com>
Message-ID: <8766h665jd.fsf@winnegan.de>

"Luis Cortes" <lcortes at aodinc.com> writes:

> Hello,
> 
> I have a piece of code that I think should work fine, but I believe that
> I've hit a scope problem (Python 2.1b1).  Does anyone out there have a hint
> as to how to fix it??

This is definitively no scope problem, you are dealing only with local
variables. You're simply not handling len(sys.argv)==4!

> code:
> 
> def getarguments():
> 
>         box = ( 160, 120 )
>         newgifname = 'new.gif'
>         gifname = ''
> 
>         if (len(sys.argv) < 4 ):
>                 print 'usage: trans.py width height image.xxx newimage.gif'
>                 sys.exit(0)
> 
          if (len(sys.argv) >= 4):
>                 # program width, height, image.xxx, newimage.gif
>                 box = ( int(sys.argv[1]), int(sys.argv[2]) )
>                 gifname = sys.argv[3]
> 
          if (len(sys.argv) >= 5 ):
>                 newgifname = sys.argv[4]
> 
>         print box, gifname, newgifname
>         sys.exit(0)
>         return ( box, gifname, newgifname )
> 
> THE PROBLEM:  when I have more than 3 arguments, the global variables do not
> change to the correct variables, but instead print their default values.
> Help!

Just replace > with >=

HTH
  Siggy


-- 
Siggy Brentrup - bsb at winnegan.de - http://www.winnegan.de/
****** ceterum censeo javascriptum esse restrictam *******




From flognat at flognat.myip.org  Fri Mar 16 04:19:45 2001
From: flognat at flognat.myip.org (Andrew Markebo)
Date: 16 Mar 2001 10:19:45 +0100
Subject: Should I close after popen??
References: <XFMail.20010315142704.shaleh@valinux.com> <mailman.984697040.2458.python-list@python.org> <m3lmq6ppv0.fsf@master.athome> <98satl$1t8$0@216.39.151.169>
Message-ID: <m3lmq6nk9a.fsf@master.athome>

/ Donn Cave <donn at oz.net> wrote:
| [...] 
| For a moment there I was awfully confused, since Python won't
| even try to run that code, but I see you're running an earlier
| version, before 2.0.

Ahh hmm yep.. the scripts are running in 1.5.2, so what I do is to
remove the waits, (hmm are ther any possibility of doing something
similar??? am I supposed to use select/poll?)

Uhm how do I get the exit-code of stuff from popen? Looking at popen2
it forks and does waitpid.. 

Ahh well I suppse the right thing to do(TM) is to go popen2??

        /Andy


From tim.one at home.com  Sun Mar 11 22:59:08 2001
From: tim.one at home.com (Tim Peters)
Date: Sun, 11 Mar 2001 22:59:08 -0500
Subject: Einstein's Riddle
In-Reply-To: <UdSq6.2184$54.2166@www.newsranger.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCIEPHJEAA.tim.one@home.com>

[Gregory Jorgensen]
> The 2% that can solve it must be psychic or deranged. The riddle
> as posed is not solvable.
> ...
> But only four types of pet are mentioned: dogs, birds, cats, horse. We
> have no evidence that any of the five owns fish.
>
> More telling: how many people can't quickly determine that this puzzle is
> unsolvable?

It's a std feature of "advanced" IQ tests that the problem stmts are
intentionally somewhat vague:  part of what they're testing is whether you
can come up with the most reasonable assumptions necessary to make the
problem *interesting*.  Since the problem statement here asked "Who owns the
fish?", and the problem admits of a unique solution if you assume *someone*
owns a fish, that's a more interesting assumption to make than to complain
that the question is ill-defined; the ability to fill in gaps reasonably is
as much a kind of intelligence as the ability to follow chains of reasoning.

You could cover both cheeks by answering "if anyone owns a fish, it must be
the German".  Then again, you may prefer to complain that "on the left of"
doesn't necessarily imply adjacent to; or that the problem statement never
says that a person who "lives in" a house is also its "owner"; or etc etc.

reading-a-question-as-trivial-is-best-reserved-for-saints-ly y'rs  - tim




From fgeiger at datec.at  Wed Mar 28 01:57:47 2001
From: fgeiger at datec.at (Franz GEIGER)
Date: Wed, 28 Mar 2001 08:57:47 +0200
Subject: wxPython: How to get a child by name?
References: <99qpce$ljk$1@newsreaderg1.core.theplanet.net> <w88w6.971$NAC.9240731@news.randori.com>
Message-ID: <99s26k$uh3$1@newsreaderm1.core.theplanet.net>

Hi Robin,

thanks for your comprehensive response!

Yet these are my first steps with wxDesigner and what you
told e was definitly what I was looking for. Though, I saw this Get button
right in front of my eyes...
Anyway, all works fine now!

Thanks a lot and best regards
Franz


"Robin Dunn" <robin at stop.spam.alldunn.com> wrote in message
news:w88w6.971$NAC.9240731 at news.randori.com...
>
> "Franz GEIGER" <fgeiger at datec.at> wrote in message
> news:99qpce$ljk$1 at newsreaderg1.core.theplanet.net...
> > I currently try to polish some of my apps with a GUI and decided to go
for
> > wxPython.
> >
> > I use wxDesigner for the layout. As it generates code after every
change,
> I
> > named all the text controls and checkboxes etc. of the dialog hosting
all
> > those controls by applying SetName() to the controls. The Dialog is
small,
> > so I do not use validators yet.
> >
> > To get a text control of the dialog I thought I could write
> > theValue = self.FindWindowByName("tcSurname").GetValue().
> > This raises an exception, because the return value is of type
wxWindowPtr
> > which does not have such a method.
>
> ...
>
> > For a "quick" hack in a MSVC project I would downcast the returned
pointer
> > of GetDlgCtrl() to be able to call the subclassed object's methods. But
> how
> > do I proceed here?
>
>
> You're on the right track, you're looking for wxPyTypeCast which takes a
> shadow object and converts the swigified pointer within it to a different
> type and wraps a new shadow object around it of the new type.  There are
> some comments about it in wx.py.  (I'm making some changes soon that will
> hopefully let me finally turn this hack into a no-op...)
>
> But since you are using wxDesigner you could have saved yourself a bunch
of
> work and let it generate some "getters" for you.  Click on the Add Python
> Source button to add a new sourcefile to your project, then double click
the
> filename under Python sources in the tree view to open the file.  Once
there
> you can add classes that create and use the specific forms you've
designed,
> and also add to the class event handlers and getters for the fields on the
> form.  I usually end up changing a lot of the code that it generates for
the
> __init__ method but that's okay since it never regenerates this file, only
> adds to it.  For the getters it makes code like this:
>
>     def GetLoginBitmap(self):
>         return wxPyTypeCast( self.FindWindowById(ID_LOGINBITMAP),
> "wxStaticBitmap" )
>
>     def GetPasswd(self):
>         return wxPyTypeCast( self.FindWindowById(ID_PASSWD),
"wxTextCtrl" )
>
>     def GetUsername(self):
>         return wxPyTypeCast( self.FindWindowById(ID_USERNAME),
> "wxTextCtrl" )
>
>     def GetURL(self):
>         return wxPyTypeCast( self.FindWindowById(ID_URL), "wxComboBox" )
>
> You just need to be sure that you give each control a unique ID name.
Once
> you have the getters like this you can write code like:
>
>     self.GetUsername().GetValue()
>
>
> --
> Robin Dunn
> Software Craftsman
> robin at AllDunn.com       Java give you jitters?
> http://wxPython.org      Relax with wxPython!
>
>
>
>




From cfelling at iae.nl  Wed Mar 28 15:17:59 2001
From: cfelling at iae.nl (Carel Fellinger)
Date: 28 Mar 2001 22:17:59 +0200
Subject: diff lists
References: <3AC1DD4E.4BA7D421@aon.at>
Message-ID: <99tgtn$ght$1@animus.fel.iae.nl>

Oliver Vecernik <vecernik at aon.at> wrote:
> Hi,

> I've got following two lists:

> ['a', 'b', 'c', 'd', 'e', 'f']
> ['e', 'c', 'f']

> I'd like to have the result:

> ['a', 'b', 'd']

> The list need not to be ordered. ['d', 'a', 'b'] will also be ok. What
> is the most effective way to achive that result?

Like proposed in an other thread, dicts are nice to uniquize lists, like:

d = {}
for k in a:
    d[k] = 1

for k in b:
    try:
        del d[k]
    except:
        d[k] = 1

c = d.keys()

-- 
groetjes, carel


From dbrueck at edgix.com  Fri Mar 16 14:09:22 2001
From: dbrueck at edgix.com (Dave Brueck)
Date: Fri, 16 Mar 2001 12:09:22 -0700
Subject: Killing threads
In-Reply-To: <20010316192738.A5550@freedom.puma-ag.com>
Message-ID: <BLEOJGEMBNBNEAKMIFDHIEHCPMAA.dbrueck@edgix.com>


> -----Original Message-----
> From: python-list-admin at python.org
> [mailto:python-list-admin at python.org]On Behalf Of Steve Purcell
> > I have th program below. How does one go about kill a thread once
> > started? Why is there no method for killing threads? I thought del
> > would do it but obviously it doesn't.
>
> You should think about using a synchronisation mechanism
> (threading.Condition, threading.Semaphore, Queue.Queue) so that the worker
> thread can exit itself when told to do so by the main thread changing the
> synchronising object.

Or just a simple flag. Here's a class I use in designs that use work flows
(work objects pulled off one queue, processed, and moved onto the next
queue). If you just need an always-processing background thread you can
eliminate all the stuff with the semaphore.

Use it like this:

q = WorkQueue(someHandlerFunc)
q.Start()
...
q.AddWork(work)
q.AddWork(work)
...
q.Stop()

class WorkQueue:
    'Waits (on a background thread) for work to appear on a queue and then
dispatches it to a handler'

    def __init__(self, handler):
        self.handler = handler
        self.sem = threading.Semaphore(0)
        self.keepRunning = 0
        self.isRunning = 0
        self.workQ = []

    def Start(self):
        'Start processing'
        self.keepRunning = 1
        if not self.isRunning:
            threading.Thread(target=self.__thread).start()

    def Stop(self):
        'Shutdown'
        self.keepRunning = 0
        self.sem.release() # Wake up the thread
        while self.isRunning:
            time.sleep(0.01)

    def AddWork(self, work):
        self.workQ.append(work)
        self.sem.release() # Wake up the thread

    def __thread(self):
        self.isRunning = 1
        sem = self.sem
        workQ = self.workQ
        handler = self.handler
        try:
            while 1:
                sem.acquire() # Wait for work to appear
                if not self.keepRunning:
                    break # Someone called Stop so quit

                work = workQ.pop(0)
                handler(work)
        finally:
            self.isRunning = 0




From cyjair at remove_this_to_mail_me_hitel.net  Tue Mar  6 08:55:18 2001
From: cyjair at remove_this_to_mail_me_hitel.net (CY Jair)
Date: Tue, 6 Mar 2001 22:55:18 +0900
Subject: IEEE 754 conformance in Python
Message-ID: <AW5p6.142249$My5.5890439@news.hananet.net>

Python's floating-point numbers representations are done by C library, with
17 significant decimal digits preserved by the "%.17g" format in
floatobject.c file. It is advertised to be conformant to IEEE 754 double
precision standard.

Java is advertised as fully conformant to IEEE 754 by Sun. However, when I
compare the two, I'm getting a little bit doubtful.

-------------------
[from Python 2.0 on Intel win 32]
1) 0.40000000000000005 == 0.4000000000000008
(,which are internally 3FD999999999999B)

2) repr(0.1) == "0.1000000000000001"

[from Java and Jython on Intel win 32 and Sun Sparc]
1)0.40000000000000005 == 0.4000000000000001
(,which means internally 3FD999999999999B == 3FD999999999999A)

2)double to string or repr(0.1) == "0.1"
-------------------

One more thing that makes me wonder is the way Python does rounding on
floating-point numbers. In the IEEE standard, when rounding a "halfway"
result to the nearest floating-point number, it picks the one that is even.
But Python's way of "%.17g" doesn't seem to work that way.

Could anyone shed an insight on this?

cheers,
Jair.



From alex at shindich.com  Wed Mar  7 03:15:32 2001
From: alex at shindich.com (Alex Shindich)
Date: 7 Mar 2001 09:15:32 +0100
Subject: Static class methods
References: <9825od0omq@news1.newsguy.com>
Message-ID: <002201c0a6de$e00442a0$0300000a@acer>

Hmm... I agree with you Alex that static methods are seldom needed in Python. I have lived without them for a long time.
The need actually arose from using the local functions. Ever since the scoping behavior was fixed, the local functions became really useful.
The problem however is that an instance method can have a local function, where a class cannot. This seems weird, since the level above the class-- the modules, can have functions. So can the level below-- the instance methods. But the classes cannot.

If anything, the absence of static methods creates an inconsistency in the way functions are handled in different scopes.

Regards,

Alex Shindich
visit http://www.shindich.com/sources
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20010307/6d0f9b27/attachment.html>

From d98aron at dtek.chalmers.se  Mon Mar  5 15:36:12 2001
From: d98aron at dtek.chalmers.se (Fredrik Aronsson)
Date: 5 Mar 2001 20:36:12 GMT
Subject: Little problem with the "+" operator
References: <983802310.155838463@news.t-online.de>
    <mailman.983804468.16428.python-list@python.org>
    <980i6q$crs$1@news.mathworks.com>
    <mt1yscm7yu.fsf@astron.berkeley.edu>
Message-ID: <980tbs$gc$1@nyheter.chalmers.se>

In article <mt1yscm7yu.fsf at astron.berkeley.edu>,
	Johann Hibschman <johann at physics.berkeley.edu> writes:
> Joshua Marshall writes:
> 
>>> This is infinitely more sane than:
> 
>>> % perl -e 'print 500 + "cool"'
>>> 500
> 
>> Comparing Python to Perl here doesn't seem very fair.  They're very
>> different languages.  In general, I'm surprised by the urgency with
>> which people often jump to compare them.
> 
> Well, Perl's the only language that I know of that lets you add 500 to
> "cool" without an error.  Well, I suppose tcl might.  But it's still a
> fairly safe inference that the original poster was coming from Perl.
> 

Well, my first guess was Java.

public class Test {
    public static void main(String[] args) {
	int variableOne = 500;
	String myVariable = variableOne + "cool";
	System.out.println(myVariable);
    }
}

gere(1)> javac Test.java 
gere(2)> java Test
500cool

Since I havn't found a % operator or a printf function, I like it.

/Fredrik

(Well, I have gotten to like Haskell a lot more after learning Java,
wrote a complete parser+typechecker for a little language,
211 lines in Haskell
~2400 lines in Java (and I'm not counting the generated code.))


From syver at NOSPAMcyberwatcher.com  Tue Mar 27 13:53:21 2001
From: syver at NOSPAMcyberwatcher.com (Syver Enstad)
Date: Tue, 27 Mar 2001 20:53:21 +0200
Subject: Python scripts as Drop targets in Windows Explorer
Message-ID: <99qogo$o6g$1@troll.powertech.no>

Run the attached .reg file (or enter the contained information manually) to
make python scripts behave like .exe in the way that you can drop files on
them and the files will appear in the sys.argv list as command line
arguments.

Something to consider for the installer of the win32 port of python maybe?


begin 666 python filedrop.reg
M__Y7`&D`;@!D`&\`=P!S`" `4@!E`&<`:0!S`'0`<@!Y`" `10!D`&D`= !O
M`'(`( !6`&4`<@!S`&D`;P!N`" `-0`N`# `, `-``H`#0`*`%L`2 !+`$4`
M60!?`$,`3 !!`%,`4P!%`%,`7P!2`$\`3P!4`%P`4 !Y`'0`: !O`&X`+@!&
M`&D`; !E`%P`<P!H`&4`; !L`&4`> !<`$0`<@!O`' `2 !A`&X`9 !L`&4`
M<@!=``T`"@! `#T`(@![`#@`-@!#`#@`-@`W`#(`, `M`#0`,@!!`# `+0`Q
M`# `-@`Y`"T`00`R`$4`. `M`# `. `P`# `,@!"`#,`, `S`# `.0!$`'T`
*(@`-``H`#0`*````
`
end



From syver at NOSPAMcyberwatcher.com  Fri Mar 30 16:43:44 2001
From: syver at NOSPAMcyberwatcher.com (Syver Enstad)
Date: Fri, 30 Mar 2001 23:43:44 +0200
Subject: Just in time debugging idioms
Message-ID: <9a2uqk$oj5$1@troll.powertech.no>

I am used to VC++ on Windows giving you the option to debug a program if it
raises an assert. The closest I've come so far is like this:

try:
    ........
except AssertionError:
    import pywin.debugger
    pywin.debugger.post_mortem(sys.exc_info()[2])

Is putting the above at the top of the call stack the python way to invoke
the debugger on asserts, and other uncaught exceptions if you just place an
open except statement there instead.

Are there other debugging tecniques, idioms that people know of I'd be
interested to hear. Especially the possibility to recompile and reload
modules while stepping in the debugger would be nice. After all if you can
do it in C++ (in a way that is) it should be possible anywhere else.




From greg at cosc.canterbury.ac.nz  Thu Mar  1 22:12:09 2001
From: greg at cosc.canterbury.ac.nz (Greg Ewing)
Date: Fri, 02 Mar 2001 16:12:09 +1300
Subject: define a new func on the fly?
References: <3A9EE927.8744154E@troikanetworks.com> <o6Cn6.846$45.1115@www.newsranger.com> <3A9EF31B.4105F05C@troikanetworks.com>
Message-ID: <3A9F0F89.5A3C6DE2@cosc.canterbury.ac.nz>

Bruce Edge wrote:
> 
> Now, how can I get it to create this function in the global namespace?

  exec "def xxx():  print \"xxxxxxxx\"" in globals()

(By the way, you don't need the parentheses around the string
to be exec'ed, because exec is a statement, not a function.)

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


From aleaxit at yahoo.com  Sun Mar 11 15:01:08 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sun, 11 Mar 2001 21:01:08 +0100
Subject: Static typing (was Re: Java guy interested in Python)
References: <mailman.984334635.27345.python-list@python.org>
Message-ID: <98glhc02sga@news2.newsguy.com>

"Tim Peters" <tim.one at home.com> wrote in message
news:mailman.984334635.27345.python-list at python.org...
> [Rainer Deyke]
> > When I type 'int i;' in C++, what I really mean is "'i' behaves like an
> > integer variable.".
>
> Since that's all "int i;" *can* mean in C++, seems tautological <wink>.

Surely you're joking, Mister Peters?

    class int {
        public:
            int(const char*=0);
            ~int();
            const char* data() const;
    };

    int i;

Isn't this perfectly legal C++, meaning something very different from
"i behaves like an integer variable"...?  int is no "reserved word",
after all.


Back to "built-in" int...:

> > The limited range of type 'int' is part of its defined semantics,
> > as is its fast performance.
>
> I expect you'll search in vain for a promise in the std that ints are
"fast"
> <wink>.

3.9.1.2 says "plain _int_s have the natural size suggested by the
architecture of the execution environment", and it would take
quite a sea-lawyer to assert that the 'natural' size in this context
is any BUT the one providing the best performance on said
execution-environment's architecture.  But I think you are right
that this is not QUITE a "promise":-).


Alex





From matt at mondoinfo.com  Thu Mar  8 14:27:38 2001
From: matt at mondoinfo.com (Matthew Dixon Cowles)
Date: Thu, 08 Mar 2001 19:27:38 GMT
Subject: Another Tkinter question
References: <PUPp6.14231$tV.1448780@newsrump.sjc.telocity.net>
Message-ID: <slrn9afn9a.ps.matt@happy-hour.mondoinfo.com>

On Thu, 8 Mar 2001 12:08:47 -0600, dsavitsk <dsavitsk at e-coli.net>
wrote:

>I have a small tkinter app that contains a (PMW) ScrolledFrame which
>in turn has a grid of entry widgets (it sort of emulates a small
>spread sheet).  What I would like to do is add a new row of entry
>widgets while the thing is running, and I'm not sure how to go about
>this as I've already .packed them in.

Tkinter is very dynamic: that your entry widgets have already been
packed shouldn't make a difference. Try just adding more. If that
doesn't work, I'd need to see your code to figure out what's up.

>(Also, if anyone knows a better way to do this than tons of entry
>widgets that would be great.)

If the easy way works, I'd do it the easy way. Naturally, it's also
possible to move the data around in a bunch of static widgets. That a
bit annoying since you have to do your own scrolling. If you'd like to
see some code of mine that does that, I'd be glad to point you at it.
I've seen combinations of the two approaches done as well, but that
was ugly.

Regards,
Matt


From stepheb at comm.mot.com  Fri Mar  2 16:36:14 2001
From: stepheb at comm.mot.com (Stephen Boulet)
Date: Fri, 02 Mar 2001 15:36:14 -0600
Subject: cataloging words in text file
Message-ID: <3AA0124E.74A39783@comm.mot.com>

I remember this homework assignment for my data structures (c++)
class: read in a large file, and create a data structure containing
every word in the file and the number of times it appears.

I was wondering how to do this in python. In c++ we had to do it
with hash tables and b-trees.

Can you do it with dictionaries in python, with the key as the word
and the data the number of times it appears? If there's any
documentation in this problem domain that people could point out to
me I would appreciate it.

What about if you had a bunch of objects, like say stars, with
attributes like position (two coordinates), magnitude, color. If you
had to sort a bunch of them at a time by attributes (say at most
10^3 of them), would using dictionaries be a good idea, or should
you start looking at interfaces to, for example, mysql?

Thanks.

-- Stephen



From costas at meezon.com  Fri Mar 16 12:30:42 2001
From: costas at meezon.com (costas at meezon.com)
Date: Fri, 16 Mar 2001 17:30:42 GMT
Subject: Killing threads
Message-ID: <3ab23ea7.328575630@News.CIS.DFN.DE>

I have th program below. How does one go about kill a thread once
started? Why is there no method for killing threads? I thought del
would do it but obviously it doesn't.

Also the print in the run method does not print until a sys.exit() is
executed.

# Note: You cannot CTRL-BREAK out of this program in PythonWin.
# You have to kill it using the task manager.

import threading
import time
import sys

class wasteTime(threading.Thread):

    def run(self):    
        # run forever
        while 1:
            print time.time()
            time.sleep(1)

startTime=time.time()
th=wasteTime()
th.start()

while 1:
    time.sleep(0.5)
    if time.time()>startTime+5:
        # I want to kill the th thread right here.
        del th
        sys.exit()


TIA

Costas


From loewis at informatik.hu-berlin.de  Fri Mar  2 11:07:06 2001
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 02 Mar 2001 17:07:06 +0100
Subject: WDDX in Python.
References: <3a9faea1$1@news1>
Message-ID: <j47l28jgt1.fsf@informatik.hu-berlin.de>

"Digital Hyakugei" <jos at trapeze.removespam.com> writes:

> Hello again. I've scoured the python.org site, used Google and have only had
> hints that there _might_ be a WDDX parser/encoder for Python... If anyone
> has a more concrete example, source code, link, anything, i'd really
> appreciate it!

PyXML 0.6.4 (and earlier) has a wddx library, see pyxml.sourceforge.net.

Regards,
Martin


From gherman at darwin.in-berlin.de  Mon Mar 19 03:21:50 2001
From: gherman at darwin.in-berlin.de (Dinu Gherman)
Date: Mon, 19 Mar 2001 09:21:50 +0100
Subject: How to find out about free space on a given drive?
Message-ID: <3AB5C19E.FA40DDA9@darwin.in-berlin.de>

Hi,

is there a simple way (or any way at all) to find out for 
any given hard disk how much free space is left on that
device? I looked into the os module, but either not hard
enough or there is no such function. Of course, the ideal
solution would be platform-independant, too... :)

Regards,

Dinu


From objectway at divalsim.it  Thu Mar  8 08:24:48 2001
From: objectway at divalsim.it (Nicola Musatti)
Date: Thu, 08 Mar 2001 14:24:48 +0100
Subject: Can anyone recomend a good intoduction to C...
References: <3aa366c6_2@nnrp1.news.uk.psi.net> <8gNo6.440$y6.128357@ruti.visi.com> <3aa6551e.212870101@news.laplaza.org> <Taup6.1292$y6.235409@ruti.visi.com>
Message-ID: <3AA78820.67A56767@divalsim.it>


Grant Edwards wrote:
[...]
> You had 1's?  In my day we had to write everything using nothign but 0's!

I believe this condition is called "Turing impaired".

Best regards,
Nicola Musatti


From johngrayson at home.com  Fri Mar 23 08:39:47 2001
From: johngrayson at home.com (johngrayson at home.com)
Date: Fri, 23 Mar 2001 13:39:47 -0000
Subject: Tkinter - Widget Colours
Message-ID: <99fjn3+m14m@eGroups.com>

--- In python-list at y..., "O'Rourke Clodagh-corour01" <corour01 at m...>
wrote:

>
> I was wondering does anyone know how the bag/fag colours can be
changed after the widget is packed.


b = Button(...)
b.pack(...)

b.configure(background='red')
b.config(bg='blue')
b['bg'] = "#fea405"

   etc...

    John Grayson





From kalle at gnupung.net  Wed Mar 21 12:05:02 2001
From: kalle at gnupung.net (Kalle Svensson)
Date: Wed, 21 Mar 2001 18:05:02 +0100
Subject: help for newbie
In-Reply-To: <3AB8D031.2E54827E@rmsb.u-bordeaux2.fr>; from Henri.Valeins@rmsb.u-bordeaux2.fr on Wed, Mar 21, 2001 at 05:00:49PM +0100
References: <3AB8D031.2E54827E@rmsb.u-bordeaux2.fr>
Message-ID: <20010321180502.B24260@father>

Sez Henri Valeins:
> Hi,
> 
> I try to use python program named PLUCKER for retrieve web page on
> internet. When I try this program I have this message :
> 
> Retrieved failed: 404 -- [Errno url error] unknown url type: 'http'

I have never used (or heard about, I think) PLUCKER, but here's an idea:
How do you run the program?
If you run it like
> PLUCKER http://www.somesite.org/
try
> PLUCKER www.somesite.org
instead.

Peace,
  Kalle
-- 
Email: kalle at gnupung.net     | You can tune a filesystem, but you
Web: http://www.gnupung.net/ | can't tune a fish. -- man tunefs(8)
PGP fingerprint: 0C56 B171 8159 327F 1824 F5DE 74D7 80D7 BF3B B1DD
 [ Not signed due to lossage.  Blame Microsoft Outlook Express. ]



From myudkin at compuserve.com  Tue Mar 13 01:48:57 2001
From: myudkin at compuserve.com (Mark Yudkin)
Date: Tue, 13 Mar 2001 07:48:57 +0100
Subject: Einstein's Riddle
References: <oNPq6.396644$ge4.138479566@news2.rdc2.tx.home.com> <98glrd$1si1a$1@ID-75892.news.dfncis.de> <3AABE712.499CCF4C@home.com> <98gp5b$1u0oi$1@ID-75892.news.dfncis.de> <98i5880pl2@news2.newsguy.com>
Message-ID: <98kfrl$pug$1@sshuraaa-i-1.production.compuserve.com>

Rather than looking at America sites for information about emu's, consider
that it's an AUSTRALIAN flightless bird, and there's more to it than eating
it.

"Alex Martelli" <aleaxit at yahoo.com> wrote in message
news:98i5880pl2 at news2.newsguy.com...
> "Boris Popov" <no-spam-popov at bigserver.com> wrote in message
> news:98gp5b$1u0oi$1 at ID-75892.news.dfncis.de...
> > "Bruce Elrick" <belrick at home.com> wrote in message
> > news:3AABE712.499CCF4C at home.com...
> > > I swear there is no evidence that any of them own a fish.  The German
> > > owns an emu.
> > >
> > > Bruce
> >
> > I am missing something then. I thought my table satisfies all given
> > statements and it has fish right under the german...
> >
> > -Boris
> > P.S. What's emu ?
>
> A large flightless bird, whose meat is appreciated by more and
> more connaisseurs (it's red meat, reasonably tasty, quite soft,
> light in cholesterol & fats).  You can get a LOT of information
> about it from the American Emu Association, http://www.aea-emu.org/.
> (Particularly with the many recent scares about "mad cow's disease"
> aka BSE, emu meat has received a huge market boost -- it's now
> quite common for supermarkets around here to offer it, restaurants
> to feature it in several dishes, etc).
>
> Any association with Europe's "Economic and Monetary Union" is,
> one can hope, totally coincidental (some people WOULD argue that
> _that_ EMU _is_ 'owned' "by the German", but that's a base and
> unsubstantiated canard...).
>
>
> Alex
>
>
>




From mda at idatar.com  Sat Mar 17 13:45:31 2001
From: mda at idatar.com (David Allen)
Date: Sat, 17 Mar 2001 18:45:31 GMT
Subject: Should I close after popen??
References: <m38zm6r8ih.fsf@master.athome>
Message-ID: <fhOs6.510$Fb.777424@typhoon2.ba-dsg.net>

In article <m38zm6r8ih.fsf at master.athome>, "Andrew Markebo"
<flognat at flognat.myip.org> wrote:

> I am hacking away a small background program, using popen to call commands,
> should I close the file in some way when done or it is taken care of
> automagically??
> 
> I do something like this once every 10 mins..
> 
>         fd1=os.popen("command") fd2=os.popen("command") os.wait(fd1)
>         os.wait(fd2) values=fd1.read()+fd2.read()
> 
> And.. anything to worry about, nesting popen/wait like this??
> (Linux/Solaris)

IIRC when garbage collected, all file handles are 
closed automagically by python.

But that doesn't mean that it's OK to do it, that's
just for safety.  You should ALWAYS close your own
handles, so you can keep track of where they're good
and where they're no good.  You're not going to really
screw things up if you don't explicitly close them
but it's a better idea to do it yourself.  It's good
practice.

As for doing two popen()'s concurrently, that shouldn't
be a problem as long as you're in an environment where
forking is kosher, since popen() does a fork.  From
the C popen() call manpage:  (Python pretty much just
wraps the C interface, I think):

The  popen()  function opens a process by creating a pipe,
forking, and invoking the shell.

-- 
David Allen
http://opop.nols.com/
----------------------------------------
Be wary of strong drink. It can make you shoot at tax collectors and 
miss 
- Lazarus Long, "Time Enough for Love"


From tim.one at home.com  Fri Mar  2 13:46:59 2001
From: tim.one at home.com (Tim Peters)
Date: Fri, 2 Mar 2001 13:46:59 -0500
Subject: extended slices and buffer objects ????
In-Reply-To: <3a9f9581.5089898@news.laplaza.org>
Message-ID: <LNBBLJKPBEHFEDALKOLCMEHGJDAA.tim.one@home.com>

[Mats Wichmann]
> Perhaps someone can shed light on my confusion about these two Python
> features?
>
> (1) Extended slicing.  I've seen claims that it's useful for
> "multidimensional data structures such as matrices and arrays", but my
> brain just isn't making sense of this. Can anyone give me a practical
> example or two?

Core Python makes no use of these.  They were added for the NumPy extension,
and that's where they're used.

> [2] Buffer objects.  What are buffer objects for, anyway? I find this
> description from the Reference Manual somewhat less than enlightening:
>
> Buffers are not directly supported by Python syntax, but can be
> created by calling the builtin function buffer().

They're a not-ready-for-prime-time generalization under the covers, seeking
to unify some aspects common to, e.g., processing strings and memory-mapped
files.  Ignore them for now!  This is still wizard-level stuff, and almost
wholly undocumented even in the internals.

By the way, you forgot to ask about the undocumented file.readinto() method
too <wink>.

if-buffer()-were-ready-for-prime-time-readinto()-would-be-documented-ly
    y'rs  - tim




From donn at oz.net  Sun Mar 18 01:01:00 2001
From: donn at oz.net (Donn Cave)
Date: 18 Mar 2001 06:01:00 GMT
Subject: Should I close after popen??
References: <XFMail.20010315142704.shaleh@valinux.com> <mailman.984697040.2458.python-list@python.org> <m3lmq6ppv0.fsf@master.athome> <98satl$1t8$0@216.39.151.169> <m3lmq6nk9a.fsf@master.athome> <98uf48$552$1@216.39.151.169> <m3bsr0muvz.fsf@master.athome>
Message-ID: <991ius$8hm$0@216.39.151.169>

Quoth Andrew Markebo <flognat at flognat.myip.org>:
| / Donn Cave <donn at oz.net> wrote:
| Basically, run two or more popens in parallell, and when there is data
| on one of them read it and do some work.. 
|
| Basically what I do now is do two popen2, do a wait on the first, chew
| some data, wait for the second, not critical at the moment, just
| interested what to do if I had to chew the data from the second
| process first.

Given two or more inputs, select() is the way to find out which one
is readable.  "Readable" means os.read() will return right away, with
either data or an empty string indicating the pipe was closed.  Any
I/O on the file object will severely confuse the issue.

...
|| One unrelated advantage of popen2 is that if you supply a list of
|| arguments, it will use them directly instead of a shell command
|| line.  That is much safer.  A command line constructed at run
|
| You mean like popen2(["command", "param", "param"]...)?

Right.  (For the same difference vs. system(), use spawnv().)

	Donn Cave, donn at oz.net


From support at internetdiscovery.com  Fri Mar  9 15:54:37 2001
From: support at internetdiscovery.com (Mike Clarkson)
Date: Fri, 09 Mar 2001 20:54:37 GMT
Subject: FW: Need help with Tkinter bind statement
References: <mailman.983910277.5499.python-list@python.org>
Message-ID: <3aa92bd1.43265069@24.0.228.33>

On Tue, 6 Mar 2001 14:22:49 -0600, "Mellor, Mike"
<Mike.mellor at tbe.com> wrote:
>I have been working on an RPN calculator program for a while now, and 
>I am having problems with some bindings.  Here is a code snippet of 
>where I'm having trouble:
>
># Bindings
>xStack.bind('<+>', lambda event, fxn = 'y+x':Calc(fxn))
>xStack.bind('<->', lambda event, fxn = 'y-x':Calc(fxn)) # This is 
>causing an error

It's a Tcl subtlety/bug. 

<+> is really a shorthand for <Key-+>
But <-> doesn't seem to parse, so use <Key--> instead.

Mike.


From scarblac at pino.selwerd.nl  Wed Mar 21 07:38:15 2001
From: scarblac at pino.selwerd.nl (Remco Gerlich)
Date: 21 Mar 2001 12:38:15 GMT
Subject: Bug or feature?
References: <99a433$b8b$1@sun.rhrk.uni-kl.de>
Message-ID: <slrn9bh80k.dju.scarblac@pino.selwerd.nl>

Markus Reitz <Markus_Reitz at yahoo.com> wrote in comp.lang.python:
> Hi,
> 
> I'm using Python 2.0 and have found a, according to my opinion, strange
> behaviour.
> 
> a=[]
> a.extend('Test')
> 
> Printing a results in:
> 
> a=['T','e','s','t']
> 
> I expected a TypeError message, because a String is not a list and so it
> can't be used to extend an existing list.
> 
> On the other hand, a string can be seen as a kind of "list of characters"
> ...
> 
> Is this behaviour a bug or a feature?

Feature. Strings are sequences, just like lists, tuples and any class
instances with the right methods (__getitem__, mostly).

>>> for a in "whee": print a
w
h
e
e
>>> list("whee")
['w','h','e','e']

-- 
Remco Gerlich


From cribeiro at mail.inet.com.br  Sat Mar 31 20:41:14 2001
From: cribeiro at mail.inet.com.br (Carlos Alberto Reis Ribeiro)
Date: Sat, 31 Mar 2001 22:41:14 -0300
Subject: PythonLabs Team Moves to ActiveState
In-Reply-To: <E14jW8x-0001Rj-00@mail.python.org>
Message-ID: <5.0.2.1.0.20010331223927.021590d0@mail.inet.com.br>

It's 2 hours before April Fool's Day here (?!?!?)

At 19:54 31/03/01 -0500, Guido van Rossum wrote:
>PythonLabs Team Moves to ActiveState
>
>To all Python users and developers:
>
>    Less than half a year ago, I moved with my team to Digital Crations, 
> in the
>    hope of finding a new permanent home for Python development. At
>    Digital Creations we've done several good things for Python, such as 
> forming
>    the Python Software Foundation and the management of the Python 2.1 
> release.
>
>    Unfortunately, Digital Creations's original plans for PythonLabs didn't
>    work out as hoped, and we weren't able to reach mutual agreement on
>    workable alternative plans - despite trying for months.
>
>    I am proud to have found a new home for my entire team: starting
>    today, Tim Peters, Barry Warsaw, Jeremy Hylton, Fred Drake and myself
>    are working for ActiveState. We will be spending part of our
>    time on core Python development (including Jython and Mailman) and
>    part of our time on Python infrastructure improvements that also
>    benefit ActiveState's efforts such as Komodo and Python.NET.
>
>    Python will remain Open Source; ActiveState has no desire to
>    monetize or brand the Python language or specific Python
>    distributions. All future work we do on Python as ActiveState
>    employees will be owned by the PSF.
>
>    We're excited to be working for ActiveState: they are one of the companies
>    most committed to Python, and they have great products! Plus, we know
>    they have deep financial backing. We trust that ActiveState will
>    provide a stable home for Python for many years.
>
>    ActiveState has also offered to take over hosting of the
>    python.org and starship sites. On behalf of the Python community,
>    we're grateful for this support of the two prime community sites for
>    Python, and we expect to be implementing the transitions shortly.
>
>    These are exciting times for the PythonLabs team - and also for Python
>    and its community. Mainstream successes for Python are showing up
>    everywhere, and we're proud to be a part of such a smart and friendly
>    community. A great year lies ahead!
>
>    Note, however, that moving to ActiveState will have some minor effects
>    on the future of Python development. ActiveState is concerned about
>    the seeming rift between the Perl and Python communities and wants
>    to do its best to help the Open Source scripting community work together.
>    In order to make it easier for Perl programmers to transition to Python,
>    we will add optional curly braces for Python. While we are fully aware
>    of the effect this will have on Python usability, we have finally decided
>    to bite the bullet and accept that until this happens, Python will never
>    be popular.
>
>    Barry has already approved PEP 357, Adding Block Delimiters to Python.
>    Since we are heavily commited to backwards compatability, a transition
>    period of three months has been mandated. For those who want the changes
>    to start sooner, we added "from __future__ import block_delimiters" in
>    Python 2.1, which will be released about two weeks from now.
>
>    Larry Wall and I agreed that now it makes sense for Perl to have
>    the Perl compiler enforce correct indentation in Perl sources. This
>    will probably be implemented in Perl 5.7, and will certainly be in Perl 6.
>    Also, after some talks with Larry, I've decided to adopt the Perl
>    numerical model for Python. The Perl model is simple, predictable and
>    easy to explain. This is a small enough change that it does not need
>    a PEP, and Moshe will implement it before 2.1rc1 is out.
>
>    Now that the catalog-sig finally has a fully working product, ActiveState
>    decided they will base a CPAN competitor on it. This server will be able
>    to deal with both Perl and Python modules. While the client will remain
>    completely Open Source, it was decided that the server will remain
>    ActiveState proprietary. While it is true that the original server was
>    written by the Python community, ActiveState has put a lot of effort
>    into making it general enough, and we need to justify the costs somehow.
>
>    --Guido van Rossum (home page: http://www.python.org/~guido/)
>
>
>--
>http://mail.python.org/mailman/listinfo/python-list





From digitig at cix.co.uk  Sun Mar 18 17:35:00 2001
From: digitig at cix.co.uk (Tim Rowe)
Date: Sun, 18 Mar 2001 22:35 +0000 (GMT Standard Time)
Subject: Thoughts about extensions to the Python language
References: <mailman.984608060.6280.python-list@python.org>
Message-ID: <memo.20010318223553.39809A@digitig.compulink.co.uk>

In article <mailman.984608060.6280.python-list at python.org>, 
carribeiro at yahoo.com (Carlos Ribeiro) wrote:

> (flames off please ...)
> 
> First of all, some background information... I'm watching this list for 
> the past few days. Thats my first contribution to the list. I work on 
> computer related jobs for some time (since 1984), and I have saw a lot 
> of things in the past decade or so. I'm learning Python since mid-2000, 
> mainly for use as a tool to gather network management information.
> 
> What attracted me to Python was the simplicity and cleanliness of the 
> language. Code written in Python is as clean as code can be. I have 
> written systems using several other languages, including C, C++ and 
> Perl, and none of these languages can even come close of Python in 
> terms of readability. I think that most people on this list will agree 
> with me that making code readable is one of the most important (and 
> relegated) steps to improve programming skills.
> 
> I'm not saying that Python is perfect - as a "work in progress" we can 
> surely think about a lot of things that can be improved. However, I'm 
> seeing a lot of effort these days toward language extensions. Call it 
> "featuritis" or anything like it, but I dont like it, and something 
> inside tells me that this is the wrong approach for Python's evolution. 
> I can number several ways this approach is failed:

<snip>

[Applause].

This has been worrying me a lot, as /almost/ every suggested "improvement" 
I see:

a. Is an import from a language I have already rejected in favour of 
Python (at least when I'm programming Python) and

b. Muddies the clarity of the language.

If people /really/ want these features, could they put them into a new 
language /based/ on Python, please, so that those of us happy with Python 
can carry on undisturbed (just as most of us using Modula2 ignored 
Modula3)? I suggest the name "Monty", as I see it as a retrograde step!

I'm not against development of the language -- the set inclusion operators 
look fine to me, and I've been known to use them (admittedly in an 
unofficial obfusticated code competition!) but if a feature doesn't fit 
easily within the present language philosophy (dummy includes to 
change the semantics of the language, anyone?) then I venture to suggest 
that the feature probably doesn't belong in Python, and that the great and 
good should rethink what they're trying to achieve and come up with 
features that do it the /Python/ way!

<underwear material='asbestos'/>


From iddwb at imap1.asu.edu  Fri Mar  9 15:19:54 2001
From: iddwb at imap1.asu.edu (iddwb)
Date: Fri, 9 Mar 2001 20:19:54 +0000
Subject: Dictionary attributes and functions
Message-ID: <Pine.LNX.4.21.0103092014190.19640-100000@moroni.pp.asu.edu>

I'm looking to read a list of items where their may be duplicates.  I'd
like to use a dictionary to do this but am way to new to python to know
fully how to do it.  I'm hoping there is some set of fuctions or
attributes that can be reference regarding dictionaries... specifically if
key(a) is already in dictionsary x I'd like to know prior to
adding.  rather than change the value of key(a) I'd like to add the value
of key(`a) to the value of key(a) so I end up with

{keya: [123]}
on insert of keya againg have something like
{keya: [123, 234]}

Any pointers on what to look for in the docs?

btw, I've been going through Mark Lutz's book on "programming python" but
I'd like something with a little more langugage reference in it.

David Bear
College of Public Programs/ASU



From stadel at excite.com  Sun Mar  4 22:28:32 2001
From: stadel at excite.com (stadel)
Date: Sun, 04 Mar 2001 20:28:32 -0700
Subject: Name of the Script
References: <20010304.185354.2114937732.8277@localhost.localdomain> <NJCo6.6720$qv3.2834377@nnrp5-w.sbc.net>
Message-ID: <20010304.202832.950390868.8277@localhost.localdomain>

Thanks alex .  Now that you say that it seems rather obvious.  I feel
like a dolt :P

Thanks again
Stadel



In article <NJCo6.6720$qv3.2834377 at nnrp5-w.sbc.net>, "Alex Shindich"
<alex at shindich.com> wrote:

> try the following:
> 
> import sys print sys.argv[0]
> ....
> The first argument in the argument list is the name of the script that
> got invoked.
> 
> I hope this helps.
> 
> Regards,
> 
> Alex Shindich visit http://www.shindich.com/sources
> "stadel" <stadel at excite.com> wrote in message
> news:20010304.185354.2114937732.8277 at localhost.localdomain...
>> What is the method for returning the file name of the running script?
>>
>> Thanks in advance. Stadel
>>
> 
>


From jgraves3 at austin.rr.com  Fri Mar 16 06:22:02 2001
From: jgraves3 at austin.rr.com (jay graves)
Date: Fri, 16 Mar 2001 11:22:02 GMT
Subject: Basic information requested
References: <MHks6.144$lC5.52172@news2-win.server.ntlworld.com>
Message-ID: <3ab1f558.42256838@news-server.austin.rr.com>

On Fri, 16 Mar 2001 09:05:53 -0000, "Jamie MacIsaac"
<jamie at cctech.co.uk> wrote:

>Hi,
>I've just started looking at Python and would like to use it to perform some
>simple admin tasks on a Windows network. However, I'm having difficulty
>finding out how to accomplish elementary tasks.
>Can someone please point me to sample code that shows how to map a network
>share and copy some files? Is there a central resource of code samples?

This mapping question was asked in July of 2000 and I saved a couple
of the postings.  These allowed me to find the relevant thread on
Yahoo Groups.  (I really miss DejaNews.  I can't wait for Google to
bring the entrie archive back online)
Anyway here is the thread.

http://groups.yahoo.com/group/python-list/message/99872

As far as copying files, check out the shutil module in the standard
library.

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

Hope this helps.
...
jay



From loewis at informatik.hu-berlin.de  Wed Mar 21 17:31:22 2001
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 21 Mar 2001 23:31:22 +0100
Subject: urllib.urlopen doesn't accept Unicode strings.
References: <998tau$q6i$1@troll.powertech.no>
Message-ID: <j4n1aeycsl.fsf@informatik.hu-berlin.de>

"Syver Enstad" <syver at NOSPAMcyberwatcher.com> writes:

> The offending line is at 241 in urllib.py (python 2.0)
> 
>     def open_http(self, url, data=None):
>         """Use HTTP protocol."""
>         import httplib
>         user_passwd = None
>         # next line patched, original line was: if type(url) is type(""):
>         if type(url) is type("") or type(url) is type(u""):
>             host, selector = splithost(url)
>             if host:
> 
> Or maybe this has been fixed in Python 2.1?

No, it hasn't. Please submit a patch to
sourceforge.net/projects/python. I recommend you start from the
current code base, though - using types.UnicodeType is much better
than type(u""). I normally write this as

          if type(url) in [types.StringType, types.UnicodeType]

Regards,
Martin



From fredrik at pythonware.com  Thu Mar 22 02:55:32 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Thu, 22 Mar 2001 07:55:32 GMT
Subject: learning python...
References: <3AB8F4E1.184DD9C@noaa.gov> <Z97u6.16733$Im6.1761313@newsread1.prod.itd.earthlink.net> <3ab90559_6@goliath.newsfeeds.com> <99b44s$m5j$1@panix6.panix.com> <3AB9182A.8FEBCF3E@noaa.gov> <N9bu6.66$Hp.3347@typhoon.aracnet.com> <vMeu6.144753$__6.26762056@typhoon.southeast.rr.com>
Message-ID: <Udiu6.2957$4N4.402777@newsc.telia.net>

Don Tuttle wrote:
> Why not post a review yourself?  I intend to as soon as I've finished
> reading it.  (Which might be awhile given my penchant for reading 5
> or more books at once.  ;-)

at 1,200 pages, reading PP2e is pretty much like reading five
or more books at once (there's *lots* of stuff in there...)

Cheers /F




From frank at canyon-medical.com  Tue Mar  6 01:18:05 2001
From: frank at canyon-medical.com (Frank Sergeant)
Date: Tue, 6 Mar 2001 00:18:05 -0600
Subject: gah! I hate the new string syntax
References: <mailman.983615170.23313.python-list@python.org> <tN5o6.2746$eH3.14913@NewsReader> <Koco6.7518$1D5.244602@e420r-atl1.usenetserver.com>
Message-ID: <ij%o6.3934$vY1.860919@feed.centurytel.net>

"Steve Holden" <sholden at holdenweb.com> wrote in message
news:Koco6.7518$1D5.244602 at e420r-atl1.usenetserver.com...
> Ha! As an Englishman married to a Scot I'm not going to let you get
 ...

> Still, at least we can remind all that Scotland and England are different
> countries on c.l.py.

Yeah, sure, on c.l.py they are different countries, but that doesn't mean
they are different countries in real life.

> Lest people might think this is intended as criticism rather than
humo(u)r,
> I hasten to add that I lived in Scotland for three years before moving to
> the USA, and found the Scots to be extraordinarily warm and generous
people.

Yeah, sure, but what else can you say when you your wife is probably holding
one of those damn stocking daggers to your throat.


-- Frank
frank at canyon-medical.com
bag pipes are weapons of war





From jeremy at digicool.com  Tue Mar 27 15:57:04 2001
From: jeremy at digicool.com (Jeremy Hylton)
Date: Tue, 27 Mar 2001 15:57:04 -0500 (EST)
Subject: StandaloneZODB at cvs.python.org
Message-ID: <15040.65184.461098.646113@w221.z064000254.bwi-md.dsl.cnc.net>

I have created a new CVS project based on Andrew Kuchling's ZODB & ZEO
packages.  The new project is called StandaloneZODB and is available
from cvs.zope.org.  

This is not an official release of any sort.  The CVS repository
contains the current development versions of the code.  The distutils
setup scripts have been tested with Python 2.1, but not with earlier
versions.  (Python 2.0 should work; Python 1.5.2 probably does not.)

There is also a mailing list, zodb-checkins at zope.org, that receives
checkin messages for the components in StandaloneZODB.  If you already
subscribe to the zope-checkins at zope.org list, you will see two copies
of every checkin message for the components found in both projects.

The project is tied directly to the ZODB code used in the Zope2 CVS
repository.  When a change is made to Zope, it will also appear in
StandaloneZODB, and vice versa.  Specifically, we use cvs modules to
incorporate parts of the Zope2 and ZEO code bases.

The project contains the key components you need to run ZODB and ZEO,
but not all of the ancillary packages that Andrew's distribution
includes.  I expect to add more components over time, but only those
that are expected to be useful for Python programmers using ZODB.

You can check out from cvs.zope.org by following the instructions at
http://www.zope.org/Resources/CVS_info.

If you want to find out more about ZODB, a good place to start is the
ZODB Wiki: http://www.zope.org/Wikis/ZODB/FrontPage

Jeremy



From mbac at nyct.net  Thu Mar 22 10:39:54 2001
From: mbac at nyct.net (Michael Bacarella)
Date: Thu, 22 Mar 2001 10:39:54 -0500
Subject: Jython is cool?
Message-ID: <20010322103954.A18539@sync.nyct.net>

Howdy,

	When I first read about Jython, what immediately popped
into my head was that you get a Python interpreter written in Java
that you can say, embedd in a web page and pass a parameter which says
which python script to execute.

<fantasy>

<applet code="jython_interpreter">
	<param name="script" value="clock.py">
	<param name="use_precompiled" value=0>
</applet>

</fantasy>

	I find myself frustrated by Java which is why the thought of
doing web applets in Python was so exciting. Unfortunately, it seems
that in downloading Jython, a JDK is still required.

	Has someone already done this/is it already possible and I just
didn't look at the package well enough or am I out of luck until I
bite the bullet some day and implement it myself?

Thanks

-- 
Michael Bacarella <mbac at nyct.net>
Technical Staff / System Development,
New York Connect.Net, Ltd.



From fredrik at pythonware.com  Wed Mar 28 11:46:14 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 28 Mar 2001 16:46:14 GMT
Subject: Speed problems with Python vs. Perl
References: <99sv33$8hk$1@news.cineca.it> <1znw6.1969$sk3.665386@newsb.telia.net> <87y9tplvuu.fsf@elbereth.ccraig.org>
Message-ID: <qzow6.1978$sk3.668310@newsb.telia.net>

Christopher A. Craig wrote:
> with Python 2.0 and Perl 5.6.0 on a Celeron 300A using a 242000 line
> input file and I got the following results:
>
> python: 61.10s user 0.42s system 87% cpu 1:10.37 total
> perl:   11.31s user 0.10s system 76% cpu 14.960 total
>
> admittedly Python is being hurt by the slower 2.0 line I/O, but I
> would guess that regular expressions are still hurting performance.

s/regular expressions/re.split/g

note that re.split is written in Python, while I'm pretty
sure Perl's split is written in C.

to get a better idea of how fast the regular expression
engine itself is, run your benchmark with "findall" instead
of "split".

Cheers /F




From erno-news at erno.iki.fi  Sun Mar 11 17:59:32 2001
From: erno-news at erno.iki.fi (Erno Kuusela)
Date: 12 Mar 2001 00:59:32 +0200
Subject: Problems writting a wrapper around linux' ssh command
References: <31rd89.lki.ln@127.0.0.1>
Message-ID: <kuhf0zj4iz.fsf@lasipalatsi.fi>

In article <31rd89.lki.ln at 127.0.0.1>, gradha at iname.com writes:

| Hi.
| I am trying to write a python wrapper around ssh. The last and only problem
| I have is password verification. I try to popen2.popen3 the ssh command,
| but no matter what I write to the input file, the ssh command ignores it
| and keeps requesting *keyboard* input.

| Is there any way to avoid this?

do not embed the password in your python program. use a passphraseless
rsa key instead, it is safer and less pain.

  -- erno


From ws-news at gmx.at  Wed Mar 14 04:05:58 2001
From: ws-news at gmx.at (Werner Schiendl)
Date: Wed, 14 Mar 2001 10:05:58 +0100
Subject: This math scares me
References: <3aad4062.1290021@News.CIS.DFN.DE> <20010312170356.B3513@better.net> <mailman.984480856.18097.python-list@python.org> <984486395.407379@newsmaster-04.atnet.at> <slrn9asf1d.ja3.qrczak@qrnik.zagroda> <984518199.127140@newsmaster-04.atnet.at> <jwbnews-09D75F.13313313032001@corp.supernews.com> <984521930.27710@newsmaster-04.atnet.at> <Vyxr6.1238$Tg.168802@ruti.visi.com>
Message-ID: <984561044.512222@newsmaster-04.atnet.at>

> >
> >Does this make a difference, as soon as arithmetics takes place?
>
> Yes.  All FP calculations took place in BCD.  You got exact
> results as long as your values didn't exceed the
> representations limits.
>
Exactly, and this is for example the case for 1/3, 1/7 and so on...
Doesn't sound like an real improvement...

regards
werner




From loewis at informatik.hu-berlin.de  Sun Mar 11 15:26:34 2001
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 11 Mar 2001 21:26:34 +0100
Subject: segfault with PY file but ok with PYC
References: <mailman.984172812.12080.python-list@python.org>
Message-ID: <j4elw4m4qt.fsf@informatik.hu-berlin.de>

Niki Spahiev <spahievi at vega.bg> writes:

> Any suggestions?

Not a real suggestion, but I also get the impression that Python 2
sometimes crashes in relation to generating byte code file, and works
fine if the byte code file is there.

A really useful contribution to Python would be a detailed analysis of
the problem, accompanied by a patch.

Regards,
Martin


From sdm7g at Virginia.EDU  Thu Mar 29 20:09:35 2001
From: sdm7g at Virginia.EDU (Steven D. Majewski)
Date: Thu, 29 Mar 2001 20:09:35 -0500 (EST)
Subject: diff lists
In-Reply-To: <m3zoe4b2tu.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <Pine.NXT.4.21.0103291959040.228-100000@localhost.virginia.edu>


On 29 Mar 2001, Michael Hudson wrote:

> 
> Tangentially, does anyone know of any good algorithms for "edit
> distance" between two sequences?  E.g. if I have 
> 
> "abcdef"
> 
> and want to get to
> 
> "abQUACKcde"
> 
> I want to get the answer back "insert 'QUACK' at position 3 and delete
> a character at position 11".
> 
> "Good" means "pretty quick", here.
> 

I used to have a Levenshtein string edit distance done in Icon, and
I think I may have translated it to python once, but I can't find 
a copy. 

If you do a google search on that, you'll find a lot of stuff:

http://www-apparitions.ucsd.edu/~rmckinle/string/
http://www-igm.univ-mlv.fr/~lecroq/lec_en.html
http://www.ling.ohio-state.edu/~cbrew/795M/string-distance.html

However, the algorithm for the distance does not give location, but
maybe this info can be extracted from the intermediate matrix. 


-- Steve Majewski




From amk at mira.erols.com  Fri Mar  9 13:38:59 2001
From: amk at mira.erols.com (A.M. Kuchling)
Date: 9 Mar 2001 18:38:59 GMT
Subject: IPC9 quotes wanted
Message-ID: <slrn9ai9ln.10q.amk@mira.erols.com>

I intended to put a note on the message board about this, but
forgot...  If you were at IPC9 and heard someone say something
quotable, please let me know who it was and what they said, so that it
can be added to the quotation file.  Thanks!

(I reserve the right to not add quotations, so don't feel bad if your
submitted quotation doesn't get in.)

--amk


From mlh at idi.ntnu.no  Sun Mar 25 16:19:44 2001
From: mlh at idi.ntnu.no (Magnus Lie Hetland)
Date: Sun, 25 Mar 2001 23:19:44 +0200
Subject: Yet Another PEP:  Query Protocol Interface or __query__
References: <Pine.LNX.4.21.0103201837570.21343-100000@clarkevans.com> <mailman.985490195.2917.python-list@python.org>
Message-ID: <99lnc6$ck4$1@tyfon.itea.ntnu.no>

"Clark C. Evans" <cce at clarkevans.com> wrote in message
news:mailman.985490195.2917.python-list at python.org...
[snip]
>     This proposal puts forth a light-weight explicit mechanism
>     for the adaptation (including verification) of an object to
>     a context where a specific type, class, interface, or other
>     protocol is expected.

I like the generality here -- nice PEP overall, IMO :)

[snip]

>     To enable the first case listed in the requirements,
>     the adapt function first checks to see if the objects
>     type or the objects class are identical to the protocol.
>     If so, then the adapt function returns the object directly
>     without further ado.

Hm. So in some cases the protocol object is merely an example
(prototype) of an object satisfying the protocol, and in other
cases it represents the protocol itself (with __prodapt__ etc)?

That seems a bit ... impure to me. Maybe I've missed some
arguments pertaining to this. How about something like this:

   def adapt(obj, protocol=None, prototype=None, can_wrap=1):
       # ...

(A standard abbreviation for prototype is "proto", but I'm
afraid that could be confusing here... Perhaps "sample" could
be used? Or some other word?)

Then one could use adapt(obj,protocol) if a protocol was
available, and adapt(obj,prototype=someObject) if the protocol
was to be inferred from a prototype (i.e. the class or type).

In that case, one could perhaps use adapt(obj,SomeClass), and
the function would check if obj was an instance of SomeClass;
it would *not* check whether SomeClass and obj were instances
of the *same* class (which seems semantically weird in this
context to me...)

Oh, well. Perhaps worth 2 pennies...

>     Q:  The name __prodapt__ sucks.
>
>     A:  Suggest a better alternative; I suppose __check__ could
>         be used for the protocol side, however I dont want to
>         trample on Paul Prescods proposal.

Hm. This could, I guess, be named __adapt__ as well, except that
there is no way of marking the transitive mode :)

How about __adjust__? Sounds more transitive than __adapt__ to
me...

>From the good ol' Merriam-Webster:

> ADAPT implies a modification according to changing
> circumstances  (adapted themselves to the warmer
> climate)

And:

> ADJUST suggests bringing into a close and exact correspondence
> or harmony such as exists between parts of a mechanism
> (adjusted the budget to allow for inflation)

Hm...

*Or*, the intransitive version of __adapt__ could be
called __conform__, which is indeed intransitive. Then the
transitive version could be called __adjust__ or __adapt__ or
whatever. I think that gets my vote:

  __conform__ - method in an object which can conform to
                a protocol
  __adapt__   - method in a protocol used to adapt an
                object to it

Perhaps not perfect, but __prodapt__ **really**
sucks. <0.1 wink>

--

  Magnus Lie Hetland         http://www.hetland.org

 "Reality is that which, when you stop believing in
  it, doesn't go away."           -- Philip K. Dick





From sholden at holdenweb.com  Thu Mar 22 22:26:59 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 22 Mar 2001 22:26:59 -0500
Subject: Why Python/Jython?
References: <99cl9e$b2b$1@Merlin.et.bocholt.fh-gelsenkirchen.de> <mailman.985260089.28444.python-list@python.org>
Message-ID: <srzu6.12868$_46.475705@e420r-atl2.usenetserver.com>

"Steve Purcell" <stephen_purcell at yahoo.com> wrote in message
news:mailman.985260089.28444.python-list at python.org...
> Andreas Freier wrote:
> > In several articles I read that Python/Jython imporoves the programmers
> > performance many times. Why?
>
> Simple: Python code is usually many times shorter than the corresponding
> Java code.
>
> Programmer performance is usually gauged by coding time and
> number of bugs, both of which are vaguely proportional to the number of
> lines of code in most sane languages*.
>
> * Perl is a noteworthy exception
>
I saw no inconsistency here. Why would Perl be an exception to your rule?

regards
 Steve





From sholden at holdenweb.com  Mon Mar  5 21:24:14 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Mon, 5 Mar 2001 21:24:14 -0500
Subject: gah! I hate the new string syntax
References: <97ohh2$polt0$1@ID-77331.news.dfncis.de><XFMail.20010302085406.shaleh@valinux.com> <mailman.983843650.6177.python-list@python.org>
Message-ID: <MWXo6.19412$ML1.974917@e420r-atl2.usenetserver.com>

"Skip Montanaro" <skip at mojam.com> wrote in message
news:mailman.983843650.6177.python-list at python.org...
>
>     Sean> Turning a string literal (";") suddenly into a string object and
>     Sean> then using its methods is not an easy read.
>
> In Python, everything is an object.  Thus, a string literal is a string
> object - no "turning into" required.  Just because it wasn't created by
> instantiating a class doesn't make it any less an object.
>
> The presence of join methods for strings and why that's where they landed
> has been hashed and rehashed several times.  Please check out FAQ 4.96 for
> details.
>
This isn't really fair, since I wrote 4.96 in response to this thread.

>     Sean> ";" . join() # in perl, this means concat ";" with the result of
>     Sean> join()
>
> Nobody ever claimed that Python was trying to be Perl... ;-) Just because
> '";".join' is a fragment of valid expressions in both languages doesn't
mean
> they have to mean the same thing.
>
> (In reality, I actually do my best to make my Perl code look as much like
> Python to my Python-infected mind as I can.)
>
just-getting-the-hang-of-this-time-machine-ly y'rs  - steve





From jicman at cinops.xerox.com  Tue Mar 20 16:28:03 2001
From: jicman at cinops.xerox.com (Jose Isaias Cabrera)
Date: Tue, 20 Mar 2001 16:28:03 -0500
Subject: ASCII encoding error: ordinal not in range(128)
References: <91o5ih$2kv$1@news.wrc.xerox.com> <20001219172933.B29727@ActiveState.com>
Message-ID: <006201c0b184$a60b9430$5aa4af0d@dssc.mc.xerox.com>

Greetings all!

I have a python program that interfaces with Microsoft Outlook.  I am running into a problem when I have attachments and subject
titles with different characters that are not your normal everyday character.  For example:  The program is giving me a Traceback
because of the following attachment:

"Desenbanco_customiza??o docushare.doc"

Here is the Trackback...

C:\>qna0.py
Traceback (most recent call last):
  File "C:\myprograms\qna0.py", line 1570, in ?
    main()
  File "C:\myprograms\qna0.py", line 1556, in main
    ProcessAllItems(wf2,workpath,workpath,NSpace)
  File "C:\myprograms\qna0.py", line 1484, in ProcessAllItems
    FindIfAttachments(em,Rec['SPAR'],wID,fam)
  File "C:\myprograms\qna0.py", line 1156, in FindIfAttachments
    name = code.encode()
UnicodeError: ASCII encoding error: ordinal not in range(128)

I've search on python sites for this, and though there is something about Unicode, there is no python example code to use.

Does anyone have any idea how to fix this?

Thanks for your help.

jos?

>




From lsloan at yahoo.com  Fri Mar  9 14:12:40 2001
From: lsloan at yahoo.com (Lance E Sloan)
Date: Fri, 9 Mar 2001 11:12:40 -0800 (PST)
Subject: function access to caller's symbols?
Message-ID: <20010309191240.98744.qmail@web9506.mail.yahoo.com>

I'm a Python newbie, so if I'm approaching this problem
the wrong way, let me know.

I defined a function that takes a string and a dictionary
as arguments.  I wanted the function to use the global
symbol table as the dictionary if it was not specified.
I defined it this way:

def myFunc(theString = None, theDict = vars()):

This worked fine when the function was defined in the
same file as the code that called it.  However, when I
moved the function to a different file and used "import"
to bring it in, it wouldn't work.  theDict in the function
didn't contain any of the symbols that were defined in
code that called the function.

So, how can I make my function access its caller's
symbol table?

Thanks in advance!


=====
Lance E Sloan
lsloan at bigfoot.com


__________________________________________________
Do You Yahoo!?
Get email at your own domain with Yahoo! Mail. 
http://personal.mail.yahoo.com/



From NoSpam at NoSpam.com  Fri Mar  2 15:05:57 2001
From: NoSpam at NoSpam.com (Tom)
Date: Fri, 02 Mar 2001 20:05:57 GMT
Subject: how to get the last interactive command enetered
Message-ID: <F2Tn6.15639$UZ4.4310299@news4.rdc1.on.home.com>

Assuming that python is in interactive mode, how do get the text of the
command that the user has just entered?

For example, say that I set a trace function.  I find the frame's code
object, look at co_filename and find that it is <stdin>.  But I want to know
what the actual command was.  Is there anyway to do this (in C or Python)?

Thanks,
Tom.







From bawolk at ucdavis.edu  Sun Mar  4 12:21:14 2001
From: bawolk at ucdavis.edu (Bruce Wolk)
Date: Sun, 04 Mar 2001 09:21:14 -0800
Subject: PyQt Impressive
Message-ID: <3AA2798A.F46081B9@ucdavis.edu>

I recently downloaded the PyQt bindings to Qt and after a minor install
issue (they don't tell you that you have to run ldconfig) I was up and
running and going through the Qt tutorials.   The tutorials are a
wonderful introduction to Qt and fun to explore.  They even contain
suggestions for making minor modifications to the programs to test one's
grasp of the concepts.  I am really impressed with the quality of the
documentation and the power and versatility of the Qt classes.  The use
of "signals" and "slots" at first seemed a bit foreign, but I quickly
got the hang of it and I think I like them a lot better than the Tkinter
command bindings.

Bruce





From brangdon at cix.co.uk  Fri Mar  2 03:34:00 2001
From: brangdon at cix.co.uk (Dave Harris)
Date: Fri, 2 Mar 2001 08:34 +0000 (GMT Standard Time)
Subject: Stream interfaces
References: <MPG.1505ab2efd31b1639896b4@news.earthlink.net>
Message-ID: <memo.20010302083453.61911A@brangdon.madasafish.com>

topmind at technologist.com (Topmind) wrote (abridged):
> s = newStream(....)
> s.peekable = true
> s.positionable = false
> s.external = blah
> s.buffered = blah
> s.readAble = true
> s.writeAble = true
> [use the stream now.....]

This doesn't seem very extensible. Here are some things we might 
reasonably expect in a streams library:

   Formatting
       End-user (print, reporting)
       Programmer (display, debug info etc)
       HTML
       XML
       Legacy structured layout
       
   Underlying Data Type
       ASCII
       Unicode
       Multi-byte
       Byte
       Binary (big-endian integers etc)
       Object (Integers etc)
       
   Filtering
       Compression
       Encryption
       Message digests/digital signatures
       Buffering
   
   Sinks/Sources
       Collection
       File
       TCP/IP Socket
       Database

There is a lot going on here. Few applications will need all of it, 
and many shops will want to insert their proprietary or legacy formats 
and algorithms. We need something flexible and extensible.

I agree that a single inheritance hierarchy is not the answer. On the 
other hand, we can't hope to get XML formatting just by writing:

    s.inXml = true

either. I expect we need more delegation. The main thing is to 
separate out the formatting front-end, which needs a rich and probably 
varied API, from the data-sink back-end which can use a much smaller 
and more uniform API. Then we can do buffering etc by means of 
pluggable adaptors, and construct pipelines like:
    XmlStream -> CompressionAdaptor -> SocketStream

This is roughly the approach of the Java library. As far as I know it 
works reasonably well.

  Dave Harris, Nottingham, UK | "Weave a circle round him thrice,
      brangdon at cix.co.uk      |   And close your eyes with holy dread,
                              |  For he on honey dew hath fed
 http://www.bhresearch.co.uk/ |   And drunk the milk of Paradise."


From dg at pearl.tao.co.uk  Wed Mar 14 05:54:26 2001
From: dg at pearl.tao.co.uk (David Given)
Date: Wed, 14 Mar 2001 10:54:26 +0000
Subject: Atomic Event.{wait,clear}
References: <mailman.984557958.20327.python-list@python.org>
Message-ID: <2lin89.ndm.ln@127.0.0.1>

In article <mailman.984557958.20327.python-list at python.org>,
	"Tim Peters" <tim.one at home.com> writes:
[...]
> Hard to guess what you're doing, but it sounds like you're using an Event as
> a "hey, there's something to do!" signal, where there will be something to do
> more than once over the life of your job.

I had another look at the documentation, and it seems a Lock is easiest
for what I want to do (I have a thread that does complicted stuff, and a
message queue that tells it what to do; I use the Lock to block when the
queue is empty). Everything seems to work, and I don't have a race
condition any more. Ta.

-- 
+- David Given ---------------McQ-+ "USER'S MANUAL VERSION 1.0:  The
|  Work: dg at tao-group.com         | information presented in this publication
|  Play: dgiven at iname.com         | has been carefully for reliability." ---
+- http://wired.st-and.ac.uk/~dg -+ anonymous computer hardware manual         


From johann at physics.berkeley.edu  Thu Mar 15 13:13:32 2001
From: johann at physics.berkeley.edu (Johann Hibschman)
Date: 15 Mar 2001 10:13:32 -0800
Subject: gah! I hate the new string syntax
References: <mailman.983485923.1136.python-list@python.org> <97sudi01i26@news1.newsguy.com> <98elrk$li8$1@nntp.Stanford.EDU> <98fc070thh@news1.newsguy.com> <98iqo2$56o$1@nntp.Stanford.EDU> <98iv3s01sc2@news2.newsguy.com> <slrn9avqlm.8ee.quinn@retch.ugcs.caltech.edu> <x4Sr6.331$4d3.75760@newsb.telia.net>
Message-ID: <mtk85qzyr7.fsf@astron.berkeley.edu>

((I think I accidentally emailed this the first time around, so here
goes nothing.))

Fredrik Lundh writes:

> Quinn Dunkan wrote:
>> Note that in ruby, any time you write
>> 
>> class Foo
>> ...
>> end
>> 
>> you are adding to, not overriding, 'Foo'.  Thus,

> does all classes live in the same namespace?

No, you can define a 'module', which is simply a namespace, for your
own classes and variables.  The system classes, such as String, etc.,
do live in one namespace, however.

I actually like this behavior; it makes interactive development
easier.  If I make a change in a source file, I can simply load it in,
and all of the objects in the running system are effectively updated
to reflect those changes.  This is also how Common Lisp behaves, and
CL has similar advantages for interactive development.

But that's going far astray.  I like features like this because I do
very exploratory programming, and have been bitten before by problems
having all the modules and classes update correctly when I make a
change.

Cheers,

--J

-- 
Johann Hibschman                           johann at physics.berkeley.edu


From DanielK at jBASE.com  Wed Mar 21 14:35:53 2001
From: DanielK at jBASE.com (Daniel Klein)
Date: Wed, 21 Mar 2001 11:35:53 -0800
Subject: Substitute for join()
Message-ID: <Mo7u6.58$Hp.2896@typhoon.aracnet.com>

I've got to make a script (written in version 2.0) to work on a machine
running 1.5.2. The only incompatible line is one that uses the
"".join(mylist) method. To 'get out of the woods' quickly, I substituted it
with

s = ''
for element in mylist: s = s + element

Is this the most efficient way to code this?

Thanks,
Daniel Klein




From no-spam-popov at bigserver.com  Tue Mar 13 11:07:54 2001
From: no-spam-popov at bigserver.com (Boris Popov)
Date: Tue, 13 Mar 2001 08:07:54 -0800
Subject: Einstein's Riddle
References: <oNPq6.396644$ge4.138479566@news2.rdc2.tx.home.com> <UdSq6.2184$54.2166@www.newsranger.com> <984380185.768920@news.aei.ca> <uP%q6.2375$54.2207@www.newsranger.com> <4w6r6.193645$df5.5192565@news1.crdva1.bc.home.com> <3aae4114.63604234@nntp.interaccess.com>
Message-ID: <98lgle$2f04i$1@ID-75892.news.dfncis.de>

"Thaddeus L Olczyk" <olczyk at interaccess.com> wrote in message
news:3aae4114.63604234 at nntp.interaccess.com...
> On Mon, 12 Mar 2001 16:08:32 GMT, "David Murphy" <dwmurphy at home.com>
> wrote:
>
> >This reminds me of a test my class was given WAY back in grade 6.  It was
> >two sheets of paper long, and at the very top, it was clearly stated that
> >the student should 'Read the entire test carefully before beginning'.
> >
> >The point was, at the bottom of the second page, the student was told
> >something to the effect of: 'After reading through the test, place your
> >pencil on you desk and wait for further instructions'.
> >
> >Well, 90%+ of the class immediately went to work on this test.  There
were a
> >couple of questions where students had to get up and do things, and it
was
> >rather amusing to watch them run around.  About 30 minutes later, they
> >started to get to the end, and they realized that they had been running
> >around and scribbling down answers for nothing.  It was a rather profound
> >lesson, at least for me.
> >
> The version I saw the last question was sign your name and turn it in.
> There were ( AFAIK ) no questions about getting up and running around.
> We were given an hour told it was very important. I gave up after
> reading the test halfway in the first thirty minutes and started
> answering questions.
>
> It didn't help that we had a strange test two weeks.
> For the first time the HS I was in had set up a special schedual for
> midterms/finals. In this particular class we would meet for two
> seperate hours. THe teacher gave us an exam that should have lasted
> one hour ( by her estimation ). She was going to use the second hour
> for a class. It turned out that most fiished about 3/4 of the test in
> the two hours. So she had a reputation for giving long tests.
>
> >*IF* Einstein was just talking about people's ability to deal with just
the
> >facts, and not to make assumptions, I think 2% +/- 5% is probably a good
> >estimate.   Take a look at the first draft of the next project
specification
> >you get *grin*
> >
> The problem is typical of problems published in many magazines ( here
> in Dell Crosswords ). I suspect that 2%-5% will have the motivation to
> sit through solving the puzzle unless there were some other reward,
> but I figure half the population could actually solve it given the
> proper motivation.
>
> I also figure the fish thing was simply a typo, and that this guy (
> who hasn't resdponded ) was probably given this as a homework
> assignement, and is simply looking for a cheap way to get it done.

There are a lot of references to this puzzle on the internet mentioning
Einstein, 2% and non-existing "fish"... But again... You never know... :)

-Boris




From bdupire at seatech.fau.edu  Fri Mar 23 12:17:20 2001
From: bdupire at seatech.fau.edu (Benoit Dupire)
Date: Fri, 23 Mar 2001 12:17:20 -0500
Subject: I want to dynamically reload a module...
Message-ID: <3ABB8520.D42D5E66@seatech.fau.edu>

My goal is to dynamically import  a module called 'func' and to get the
references to the objects that 'func' created.
It works like this:
prog1.py  dynamically imports func.py.
func.py creates a setheading object, and prog1 gets its reference. All
works fine !
The result is:
>>>reload(prog1)
test
<module 'prog1' from 'C:\Python20\prog1.pyc'>


# prog1.py

class foo:
    def __init__(self, module):
        self.aa=__import__(module)          # dynamically import func
        chRef= getattr(self.aa, 'setheading')         # get reference to
the object that func created.
        chRef.run()

my_foo = foo('func')

# ----------------------------------
# func.py
class SetHeading:
    def run(self):
        print 'test'

setheading=SetHeading()


Now I change func.py, replacing ' setheading=SetHeading()' with
'toto=SetHeading()'
Normally, this should create an error when I run prog1.py, because the
setheading object does not exist anymore...
However:
>>> reload(prog1)
test
<module 'prog1' from 'C:\Python20\prog1.pyc'>


func is not reimported! this shouldn't work...

How can I force prog1.py to dynamically reload  module='func' ?
I did not see any __reload__ function in the Library Reference ....





From aleaxit at yahoo.com  Thu Mar 22 05:04:17 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 22 Mar 2001 11:04:17 +0100
Subject: Standalone executables
References: <998igr$4al$1@niagara.nonet> <iXWt6.156032$Z8.33024358@typhoon.southeast.rr.com> <99aldn$7dh$1@niagara.nonet> <99cf1s$hrsb$1@ID-59885.news.dfncis.de>
Message-ID: <99cin701b94@news2.newsguy.com>

"Thomas Heller" <thomas.heller at ion-tof.com> wrote in message
news:99cf1s$hrsb$1 at ID-59885.news.dfncis.de...
> > I wonder, for Linux, what if I wrote in Jython, compiled with jythonc,
> > then used a Java native-code compiler? Would that be too insane? (-:
> >
> With virtually every Linux coming with python, why is this needed at all?

...perhaps because each Linux installation has its own, not
fully compatible installation of Python...?


Alex





From emile at fenx.com  Tue Mar 27 09:36:25 2001
From: emile at fenx.com (Emile van Sebille)
Date: Tue, 27 Mar 2001 06:36:25 -0800
Subject: who is my caller?
References: <mailman.985700854.31956.python-list@python.org> <slrn9c18o0.ojm.scarblac@pino.selwerd.nl>
Message-ID: <99q8nv$2aibn$1@ID-11957.news.dfncis.de>

"Remco Gerlich" <scarblac at pino.selwerd.nl> wrote in message
news:slrn9c18o0.ojm.scarblac at pino.selwerd.nl...
<snip>
>
> Anyways, yes, it's more like black magic in Python. There is a call stack
> you can inspect, but only if you just caught an exception.
>

If I'm not mistaken, with 2.0 you can directly import traceback without
needing to generate an exception first.


--

Emile van Sebille
emile at fenx.com




From nayland at mindspring.com  Wed Mar 14 10:26:37 2001
From: nayland at mindspring.com (Mark Tokay)
Date: Wed, 14 Mar 2001 10:26:37 -0500
Subject: Can you please post this opening somewhere on your website?
Message-ID: <3AAF8DAD.724401A0@mindspring.com>

Python Developer needed (Arlington, Virginia)

Primary responsibilities will involve integrating Web Design
technology with various Website applications.

Desired Skill Set: Looking for well rounded individual
who has experience developing client side code
for client-server applications on various operating systems
including Windows (Win32 API) as well as web-based
clients.

Expert Knowledge (2+ years) of Python is required.
A stable work history is required.
No contractors or consultants.
Principals only.  Local candidates only.

For more information, please contact:

Mark Tokay
(828) 686-7777
Email: nayland at mindspring.com







From aleaxit at yahoo.com  Sun Mar 25 12:10:05 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sun, 25 Mar 2001 19:10:05 +0200
Subject: "object" vs "dynamic" programming styles
References: <3ABD117E.1E5492C4@erols.com> <3ABD2390.A2A28A12@letterror.com>
Message-ID: <99l8vh01fo3@news1.newsguy.com>

"Just van Rossum" <just at letterror.com> wrote in message
news:3ABD2390.A2A28A12 at letterror.com...
    [snip]
> My personal preference would definitely be version 3, the factory
function.
> Less magic, clearer code.

Seconded!!!  Indeed: thirded, fourthed, etc...


> When I first started writing OO code (which was with Python) I was often
> tempted to do lots of smart things in the __init__ method. But it only
> makes things harder in the long run. I recently had to extend a module I
> had written a few years back: all the smart things in the __init__ were
> of no use to me now: it assumed a quite different usage.
> So I changed the __init__ to be as simple as possible, and added a couple
> of factory functions. It's all nice and clean now...

An absolutely crucial consideration.  Temptation's even greater in
languages offering the ability to provide multiple overloaded
constructors -- that makes it even easier to yield to the temptation
to perform too much clever magic in the constructor[s].  Factory
functions rule...


Alex





From phd at phd.pp.ru  Wed Mar 21 04:41:08 2001
From: phd at phd.pp.ru (Oleg Broytmann)
Date: Wed, 21 Mar 2001 12:41:08 +0300 (MSK)
Subject: Regular expressions and non-standard character set
In-Reply-To: <0XMt6.2503$4N4.321668@newsc.telia.net>
Message-ID: <Pine.LNX.4.30.0103211234520.15322-100000@fep132.fep.ru>

Hi!

   What is the python version you are running? I have this problem with
python 2.0. Do you use 2.1b?

On Tue, 20 Mar 2001, Fredrik Lundh wrote:
> Petri Mikael Kuittinen wrote:
> > I want to match word boundaries using the special sequences \b and \B
> > of regular expressions. They work OK when using the "standard"
> > alphanumeric set [a-zA-Z0-9_]. But I would like them to work with
> > character set which also contains various "national characters"
> > e.g. ?, ?, ?, ?, ?, ?, ? etc. and their uppercase equivalents.
> >
> > Locale doesn't seem to be the proper way to do it
>
> are you sure?
>
> >>> import locale
> >>> locale.setlocale(locale.LC_ALL, "")
> 'Swedish_Sweden.1252'
> >>> import re
> >>> re.findall(r"\b...\b", "spam, egg, bacon, and ???")
> ['egg', 'and']
> >>> re.findall(r"(?L)\b...\b", "spam, egg, bacon, and ???")
> ['egg', 'and', '???']

Oleg.
----
     Oleg Broytmann            http://phd.pp.ru/            phd at phd.pp.ru
           Programmers don't die, they just GOSUB without RETURN.




From neelk at alum.mit.edu  Tue Mar 20 22:15:11 2001
From: neelk at alum.mit.edu (Neelakantan Krishnaswami)
Date: 21 Mar 2001 03:15:11 GMT
Subject: Yet Another PEP:  Query Protocol Interface or __query__
References: <Pine.LNX.4.21.0103201837570.21343-100000@clarkevans.com> <99917n$2jf$1@panix6.panix.com>
Message-ID: <slrn9bgb7f.kam.neelk@alum.mit.edu>

On 20 Mar 2001 17:47:35 -0800, Aahz Maruch <aahz at panix.com> wrote:
>[posted & e-mailed]
>
>In article <Pine.LNX.4.21.0103201837570.21343-100000 at clarkevans.com>,
>Clark C. Evans <cce at clarkevans.com> wrote:
>>
>>    A new built-in method, __query__ is proposed.  This method has
>>    a single argument, an interface protocol identifier, and either
>>    returns an object supporting the given protocol, or throws an
>>    unknown exception. [...]
>
> [...]
>
> Most importantly, how does one find out what interfaces an object *does*
> support?

Why is this important?

If you are writing some client code, presumably all you care about is
that the object you receive supports certain interfaces X, Y and Z
that you are using. It wouldn't seem to matter at all what -other-
interfaces it supports.

I can see heavy-duty reflective programs wanting this capability, but
it doesn't seem like something that would be used very often in
regular code. So perhaps it is better left out of the first draft.


Neel


From Dan.Rolander at marriott.com  Fri Mar 23 08:43:10 2001
From: Dan.Rolander at marriott.com (Rolander, Dan)
Date: Fri, 23 Mar 2001 08:43:10 -0500
Subject: Good GUI for Python
Message-ID: <6176E3D8E36FD111B58900805FA7E0F80CCF6471@mcnc-mdm1-ex01>

I think wxPython is *very* easy to use, if you can maneuver your way through
the C++ documentation.
 
Dan

-----Original Message-----
From: winson at mdsserv.mds.com.tw [mailto:winson at mdsserv.mds.com.tw]
Sent: Thursday, March 22, 2001 9:19 PM
To: Python List
Subject: Good GUI for Python


Hi, All
 
I surveyed 3 GUI framework for Python including wxPython, PythonWin's
included win32ui and Tkinter. My development environment is Mark Hammond's
PythonWin downloaded from ActiveStates.
 
These're my issues:
 
First, ActiveStates' Python tool cannot support Tkinter and I confirmed with
their stuff. So Tkinter maybe good, but not compatible with PythonWin.
 
Second, Mark's masterpiece win32ui and total MFC solution is excellent, but
I'd wanna bind on Windows platform.
 
Third, wxPython sounds great but not so easy as win32ui.
 
Is there any suggestions ? Change my develop environment or is there another
good gui for me ??

--
Best Regards,
 
Winson Chang
 
3F, 2, 150 Lane, Sec. 5, Hsin I Rd., Taipei, Taiwan, R.O.C.
 
Tel: +886-2-27225333 ext 376
Fax: +886-2-27222330
 
Email:  <mailto:winson at mdsserv.mds.com.tw> winson at mdsserv.mds.com.tw

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20010323/b8606a62/attachment.html>

From qrczak at knm.org.pl  Tue Mar 13 13:22:16 2001
From: qrczak at knm.org.pl (Marcin 'Qrczak' Kowalczyk)
Date: 13 Mar 2001 18:22:16 GMT
Subject: From SETL to Python translation
References: <mailman.984376575.3864.python-list@python.org>
Message-ID: <slrn9aspao.hcd.qrczak@qrnik.zagroda>

Mon, 12 Mar 2001 00:55:06 -0500, Tim Peters <tim.one at home.com> pisze:

> >         if #(odds := {x in nodes | odd(#G{x})}) > 2 then
> 
> is the Python:
> 
>     odds = []
>     for x in nodes:
>         successors = [y for (z, y) in G if z == x]
>         if len(successors) % 2:
>            odds.append(x)
>     if len(odds) > 2:

Or:
      def successors(G, x): return [y for (z, y) in G if z == x]
      odds = [x for x in nodes if len(successors(G, x)) % 2]
      if len(odds) > 2:

-- 
 __("<  Marcin Kowalczyk * qrczak at knm.org.pl http://qrczak.ids.net.pl/
 \__/
  ^^                      SYGNATURA ZAST?PCZA
QRCZAK


From ruediger.maehl at dzsh.de  Thu Mar 29 04:23:19 2001
From: ruediger.maehl at dzsh.de (Ruediger Maehl)
Date: Thu, 29 Mar 2001 11:23:19 +0200
Subject: i need help on builtin function math
References: <mailman.985822299.17200.python-list@python.org>
Message-ID: <newscache$ancyag$9so$1@www.dzsh.de>

Marcus Konermann schrieb in Nachricht ...
>i want to use the sqrt() function of the builtin funtion math in that
>kind:
>
>import math
>...
>a=0
>a=sqrt(whatever)
>

Try this:
>>> import math
>>> a = 0
>>> a = sqrt(34)
Traceback (innermost last):
  File "<interactive input>", line 1, in ?
NameError: There is no variable named 'sqrt'
>>> a = math.sqrt(34)
>>> a
5.8309518948453007

or this:
>>> from math import *
>>> a = sqrt(34)
>>> a
5.8309518948453007

HTH,
R?diger




From jim.vickroy at noaa.gov  Fri Mar  9 16:26:02 2001
From: jim.vickroy at noaa.gov (j vickroy)
Date: Fri, 09 Mar 2001 14:26:02 -0700
Subject: how to make a function thread-safe ?
Message-ID: <3AA94A6A.50138001@noaa.gov>

If a function may be applied (simultaneously) by multiple threads is the
safe (correct) procedure to wrap the function inside a class and have
each thread create its own instance of the class?

Thanks


From neelk at alum.mit.edu  Fri Mar 16 23:01:06 2001
From: neelk at alum.mit.edu (Neelakantan Krishnaswami)
Date: 17 Mar 2001 04:01:06 GMT
Subject: Who's minister of propaganda this week?
References: <mailman.984595938.11127.python-list@python.org> <98otuq0nnf@news1.newsguy.com> <slrn9b0cc2.gsa.neelk@alum.mit.edu> <98q6ro2uoa@news2.newsguy.com> <slrn9b2u0r.j74.neelk@alum.mit.edu> <98t8830nlb@news1.newsguy.com>
Message-ID: <slrn9b5sbv.lu3.neelk@alum.mit.edu>

On Fri, 16 Mar 2001 15:33:08 +0100, Alex Martelli <aleaxit at yahoo.com> wrote:
>"Neelakantan Krishnaswami" <neelk at alum.mit.edu> wrote in message
>news:slrn9b2u0r.j74.neelk at alum.mit.edu...
>    [snip]
>> > Keying a dict on class has the unfortunate side-effect of
>> > not allowing polymorphism by inheritance -- objects of classes
>> > _derived from_ the one used as the key will not match.
>> >
>> > You can preserve this basic idea by using as key, instead of the
>> > class, some ad-hoc attribute which will be inherited (if it IS
>    [snip]
>> I actually have a function that walks the class tree checking it
>> against the dictionary, which solves this problem.  You don't need to
>> add a special attribute in Python since __class__ gives you what you
>> need for free.
>
> ...except it doesn't, because it's _not_ inherited -- you have
> to 'manually' walk the directed acyclic graph (tree? what tree?
> we have _multiple_ inheritance:-) or iterate through the dictionary's
> keys (assumed to be types) using isinstance for each on your
> object.

Okay, I don't understand what you're saying at all.

If I call __class__ on an object, I get its class. Now, I can ask
each class what its superclasses are using the __bases__ attribute on
each class. This is okay to leave as a function, because Python
promises to walk the class heterarchy (happy? :) in a depth-first
left-to-right order.

So I have a dictionary like this

>>> d
{<class Bar at 80ea570>:   2,
 <class Glarp at 80e9bd8>: 3,
 <class Foo at 80dcff8>:   1}

and when I call lookup(d, Baz()) where Baz is a direct subclass of
Foo, it will return 1. (C types are easy to handle, because they don't
admit any subtyping.)

Am I off-base here?

>> I only added it to my toolkit recently, though, sine I tend to use
>> classes like algebraic data types in ML or Haskell -- as tagged
>> variants rather than designed for user-extension. I find
>> subclassing friendly designs are a lot harder to design in a way
>> that doesn't reveal too much internal structure.
>
> Inheritance (of implementation) is strong coupling, yes.  But,
> it IS a very handy implementation technique [...]

No disagreement here. I'm just hesitant to use it much because I don't
want to accidentally break some invariant I accidentally left out of
my unit tests 6 months ago. :)

> Here's a sketch of a general-by-type dispatcher module to ensure one
> can dispatch on either type OR class (and in the latter case gets
> inheritance by default, unless it's explicitly overridden):

This looks a lot like the one I have, except that I pass all
information in arguments, and you've factored into smaller
functions. (Good for you!)


Neel



From jmg at ecs.soton.ac.uk  Wed Mar 14 05:26:10 2001
From: jmg at ecs.soton.ac.uk (Jacek Generowicz)
Date: 14 Mar 2001 10:26:10 +0000
Subject: Extending Python: rewriting a single method in C
References: <g0wv9vxms0.fsf@scumbag.ecs.soton.ac.uk> <slrn9apn8f.382.jepler@potty.housenet> <g0ofv6xbj4.fsf@scumbag.ecs.soton.ac.uk> <98l1i901k4u@news1.newsguy.com> <g0lmq9xxz0.fsf@scumbag.ecs.soton.ac.uk> <98m6g20fgv@news1.newsguy.com>
Message-ID: <g0g0ggy7ct.fsf@scumbag.ecs.soton.ac.uk>

"Alex Martelli" <aleaxit at yahoo.com> writes:

> "Jacek Generowicz" <jmg at ecs.soton.ac.uk> wrote in message
> news:g0lmq9xxz0.fsf at scumbag.ecs.soton.ac.uk...
>     [snip]
> > As I understand it, I have to provide a __getattr__ method in my
> > python class which will get called when I use PyObject_GetAttr. (If
> > this is so then I don't see why the function isn't automatically
> > provided; any creative definition---ie differing from the blatantly
> > obvious (and hence tedious)---surely only opens the door to
> > horrors. Still, mine is not to reason why.)
> 
> You need only provide __getattr__ if you have to do something
> special -- and when you do need it, it's nice to have the ability.
> But attribute-getting works just fine for simple cases without
> you writing anything special!

That's what I thought until I tried something like the following:

>>> class foo:
...     def __init__ ( self, a ):
...             self.bar = a
...     def show ( self ):
...             print self.bar
... 
>>> a = foo( 4 )
>>> a.show()
4
>>> print a.__getattr__("bar")
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
AttributeError: 'foo' instance has no attribute '__getattr__'

So I had a look at some documentation. Let me quote Beazley, page 30:

  a = x.s     # Invokes __getattr__(x,"s")

You never know, strager things have happened, let's give it a go :

>>> z = __getattr__(foo,"bar")
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
NameError: There is no variable named '__getattr__'

Hmmmmm.

OK . . . I guess this means that the default implementation is not
accessible via __getattr__, but it is available via PyObject_GetAttr ?



  [Useful references to relevant documentation snipped]

> > (Would it be fair to say that the documentaion is, on the whole, less
> > that satisfactory ?)
> 
> I find it pretty decent, myself.

I guess that it is if you know what you are looking for and where to
find it :-)


Cheers,

Jacek


From syver at NOSPAMcyberwatcher.com  Fri Mar 16 01:27:48 2001
From: syver at NOSPAMcyberwatcher.com (Syver Enstad)
Date: Fri, 16 Mar 2001 07:27:48 +0100
Subject: Python and ADO
References: <98r9qn$tbi$1@troll.powertech.no> <mailman.984690501.20811.python-list@python.org> <98rk2k04jt@news2.newsguy.com>
Message-ID: <98sbtb$936$1@troll.powertech.no>

"Alex Martelli" <aleaxit at yahoo.com> wrote in message > > > Has anyone
considered making a DB API compliant wrapper around MS ADO?
> Or
> > > isn't this considered worthwhile?
>
> I _considered_ it often, but never got a round tuit... yet!

Ok, then I'll be considering it too, and we will see if I will ever get
round to it ;-)




From mhuening at zedat.fu-berlin.de  Wed Mar 28 10:44:53 2001
From: mhuening at zedat.fu-berlin.de (Matthias Huening)
Date: Wed, 28 Mar 2001 17:44:53 +0200
Subject: Speed problems with Python vs. Perl
References: <99sv33$8hk$1@news.cineca.it>
Message-ID: <99t0u9$2liv2$1@fu-berlin.de>

Did you try to avoid re for this simple task and to use the string
method 'split' instead (or the string module if you are using 1.5)? It
defaults to any whitespace string as separator and should be faster:
f = line.split()

Matthias








From support at internetdiscovery.com  Sun Mar 25 22:10:25 2001
From: support at internetdiscovery.com (Mike Clarkson)
Date: Mon, 26 Mar 2001 03:10:25 GMT
Subject: Jython and IDE/debugger
References: <9980fp$sm0$1@news.ost.eltele.no>
Message-ID: <3abeaf46.110041538@24.0.228.33>

On Tue, 20 Mar 2001 17:28:45 +0100, "Steinar Eriksen"
<steinar.eriksen at fikse.net> wrote:

>I would like to prototype my Java application in Jython, byt calling up
>Jython scripts from my Java classes. Whilst developing my Jython scripts; I
>am looking for some IDE or debugging tool to help me test my code. Anyone
>seen that available for Jython ?

Tkinter was ported to jPython using JNI
(http://sourceforge.net/projects/jtkinter)
Could it be brought up to date for jython 2?

I know it is not a "pure" Java solution, but there
are a lot of people like myself who need to use
some of the massive Java libraries, and are looking
to Jython to help provide a practical solution.

Moreover, it would help to extend the completeness
of CPython/Jython to include GUI work, wich is an added plus.

>I know the Python IDE's are excellent, but since the libraries differ I
>suppose they are not applicable,

Finn, how hard would it be to bring it up to date for jython 2?

Many thanks,

Mike.



From support at internetdiscovery.com  Sat Mar 17 16:20:19 2001
From: support at internetdiscovery.com (Mike Clarkson)
Date: Sat, 17 Mar 2001 21:20:19 GMT
Subject: How do I avoid the Tkinter 30 msec polling loop.
References: <98s781$7a8o$1@fido.engr.sgi.com>
Message-ID: <3ab3d191.6695740@24.0.228.33>

On 16 Mar 2001 05:10:25 GMT, pj at sgi.com (Paul Jackson) wrote:
>
> [Sorry if this is a repeat - I don't see my post from yesterday. -pj]
>
>I have a little 'biff' (BSD-speak for email watchdog) like
>program, written in Python, using Tkinter, that shows one of
>two text messages in a little window, depending on whether my
>email box has new mail in it.  The app polls the mailbox every
>minute, with an os.stat() call.
>
>Unfortunately, the app consumes cpu almost continuously,
>even when I wish it would sleep for the minute.
>
>The same busy loop can be see even on the smallest "Hello,
>world" example, such as:
>
>    import Tkinter
>    root = Tkinter.Tk()
>    label = Tkinter.Label(root, text = "Hello, world.")
>    label.pack()
>    root.mainloop()
>
>A system call trace of the above, on Irix 6.5.11, using
>
>    Python 2.1a1 (#23, Feb  2 2001, 17:25:50) [C] on irix6,
>
>shows the following 2 select() system calls, every 30 msec,
>non-stop:
>
> hello.py( 3693): select(4, [3], [<empty>], [<empty>], {sec=0, usec=0})
> hello.py( 3693): select(0, 0, 0, 0, {sec=0, usec=20000})
>
>This is on a system with a 10 msec TICK, and a timeout on
>the select of 20 msec, which ends up taking 30 msec each
>loop (timeouts are rounded up to the next TICK).
>
>===
>
>Anyhow, is there anyway to stop or radically slow this loop?

The short answer is no - you cannot stop or radically slow this loop.

It comes from  the way _tkinter.c is handling threads, tcl events
and Python signals. I spoke to Guido about it at Python9, and am
hoping we can do a non-looping solution using Tcl_CreateEventSource.

For the long answer, look at the attached email I sent to someone
who kindly offered to help take a look at this problem.

>                          I won't rest till it's the best ...
>                          Manager, Linux System Software
>                          Paul Jackson <pj at sgi.com> 1.650.933.1373

If you can't rest until it's the best, and if you know Tcl and Python,
send me email, and maybe you can help solve this misfeature.

Mike.

----------------------------------------------------------------------------------------
At 03:26 PM 1/29/01 +0100, you wrote:
Subject: Fundamental Tkinter inefficiency?

I'm looking at the implementation of Tk_MainLoop in _tkinter.c:

	quitMainLoop = 0;
	while (Tk_GetNumMainWindows() > threshold &&
	       !quitMainLoop &&
	       !errorInCmd)
	{
		int result;

#ifdef WITH_THREAD
		Py_BEGIN_ALLOW_THREADS
		PyThread_acquire_lock(tcl_lock, 1);
		tcl_tstate = tstate;
		result = Tcl_DoOneEvent(TCL_DONT_WAIT);
		tcl_tstate = NULL;
		PyThread_release_lock(tcl_lock);
		if (result == 0)
			Sleep(20);
		Py_END_ALLOW_THREADS
#else
		result = Tcl_DoOneEvent(0);
#endif

		if (PyErr_CheckSignals() != 0)
			return NULL;
		if (result < 0)
			break;
	}


and was wondering if it would be better to use a Tcl Event procedure
with a MaxBlock time to check the Python signals, instead of calling
Sleep.  Then it wouldn't be calling a hard sleep.  This means Python
burns a fair amount of CPU idling. You can see a significant
difference between Tk applications and the identical TkInter
application for CPU usage, I suspect because of this tight Sleep loop.
This even if you don't "use" threads - it's if you compile with
threads enabled.   

The idea would be to establish a Tcl Event procedure that would check
for Python signals every (say) 100 msec., and handle Tcl events all of
the rest of the time.

Does anyone know how to do this in Tcl - it shouldn't be more than a
few lines I suspect. Look at Tcl_CreateEventSource and
Tcl_SetMaxBlockTime in tcl/generic/tclNotify.c

Many thanks,

Mike.



From moshez at zadka.site.co.il  Tue Mar 20 04:24:49 2001
From: moshez at zadka.site.co.il (Moshe Zadka)
Date: Tue, 20 Mar 2001 11:24:49 +0200
Subject: [Very Long (11K)] Numeric PEPs, first public posts
In-Reply-To: <99669h$so4$1@panix6.panix.com>
References: <99669h$so4$1@panix6.panix.com>, <slrn9bcp3h.ns2.qrczak@qrnik.zagroda> <mailman.984748941.31248.python-list@python.org> <Q5Ns6.432845$ge4.152574719@news2.rdc2.tx.home.com> <mailman.985035504.26190.python-list@python.org>
Message-ID: <E14fINl-0007hq-00@darjeeling>

On 19 Mar 2001 15:55:29 -0800, aahz at panix.com (Aahz Maruch) wrote:
 
> Because a lot of people don't have time to subscribe to every mailing
> list that comes along.

I respectfully disagree. The numerics sig has managed to get many people
that probably don't read c.l.py subscribed to it. 

'''
The champion offers to take the discussion off-line
    and specifies a location (e.g. egroups, python.org, Roundup).
'''

> If I have the time, I'm going to be posting some
> comments about your proposal, and they're going to be here on c.l.py.

Comments posted here have much less of a chance to get into my PEPs
(as open issues or as changes), simply because here I have *less* of
a peer review (for example, I don't think Paul D. or Guido read c.l.py
regularily, and their opinions are very important to me.)
So, you're much better off CCing python-numerics. (I do try to read
c.l.py regularily, and I try to integrate things. But I am a finite
person, and python-numerics seems to me to strike a perfect balance
between mailing me personally (people accused me of being unfair
in the past -- on a public mailing list they have the feeling that
it doesn't go into a black hole) and being on .c.l.py (which I 
occasioanlly have to purge do to my thoroughly non-robotic nature)).

> Note carefully this quote from PEP 1:
> 
>     A PEP that has not been discussed on python-list and python-dev will
>     not be accepted for review.  However, wherever possible, long
>     open-ended discussions on public mailing lists should be avoided.

Of course. But the PEP won't be ''accepted for review'' anyway until all
the Open Issues are resolved, and that has much less of a chance happening
here. 

-- 
"I'll be ex-DPL soon anyway so I'm        |LUKE: Is Perl better than Python?
looking for someplace else to grab power."|YODA: No...no... no. Quicker,
   -- Wichert Akkerman (on debian-private)|      easier, more seductive.
For public key, finger moshez at debian.org  |http://www.{python,debian,gnu}.org



From syver at NOSPAMcyberwatcher.com  Tue Mar 27 14:08:22 2001
From: syver at NOSPAMcyberwatcher.com (Syver Enstad)
Date: Tue, 27 Mar 2001 21:08:22 +0200
Subject: Threading legality/morality
References: <fsDv6.1247$NR.100201@news3.oke.nextra.no> <99nnbv$phs$1@panix6.panix.com> <h0%v6.1842$NR.144463@news3.oke.nextra.no> <99qa3v$6e0$1@panix6.panix.com>
Message-ID: <99qoje$o6q$1@troll.powertech.no>

"Aahz Maruch" <aahz at panix.com> wrote
> There is a misfeature in urllib such that the standard FancyURLOpener
> class prompts for input if it gets an HTTP 401 (authorization needed).
> You probably pressed ENTER.
Think so yes, how can it be fixed? I am pretty well versed in python but a
pointer could be nice so I didn't use quite so much time on it.




From gtalvola at nameconnector.com  Tue Mar 20 11:10:02 2001
From: gtalvola at nameconnector.com (Geoff Talvola)
Date: Tue, 20 Mar 2001 11:10:02 -0500
Subject: Python and ADO
References: <98r9qn$tbi$1@troll.powertech.no> <3AB130B7.6A09C3F1@NameConnector.com>
Message-ID: <3AB780DA.70B5DD4@NameConnector.com>

Geoff Talvola wrote:

> Syver Enstad wrote:
>
> > Has anyone considered making a DB API compliant wrapper around MS ADO? Or
> > isn't this considered worthwhile?
> >
> > As ADO is the database api that I use, I could have a look at making a
> > wrapper for it if there was any interest in it and if making such a thing is
> > not too involved.
>
> Michel Orengo already wrote a preliminary version of a DB-API wrapper for ADO but I don't think he ever finished it.  I have a copy of it, but I have never actually tried it.  If you can't find it in the newsgroup archives, drop me an email and I'll send you a copy.
>
> And I do think this would be quite useful...

I've now sent a copy of this to at least a dozen people, so there's clearly a lot of interest for a DB-API ADO wrapper.  At least one of you is bound to take it and make improvements -- please share it with the list if and when you do; a lot of people will appreciate your work!

--


- Geoff Talvola
  Parlance Corporation
  gtalvola at NameConnector.com




From not.this at seebelow.org  Thu Mar 15 23:25:07 2001
From: not.this at seebelow.org (Grant Griffin)
Date: Thu, 15 Mar 2001 22:25:07 -0600
Subject: inverse of asctime
References: <98rfn6016rf@drn.newsguy.com> <m3d7bippf3.fsf@master.athome>
Message-ID: <3AB195A3.CFC511@seebelow.org>

Andrew Markebo wrote:
> 
> / Grant Griffin <not.this at seebelow.org> wrote:
> | Hi Gang,
> |
> | I'm in the market for the inverse of asctime().  I looked in the time module but
> | couldn't find it.  (Nope, strptime isn't available on my Windows installation.)
> | Seems like there oughtta be sucha thing somewhere.
> 
> Ehm.. I actually have a strptime.py, used by RogueWave ;-) *stretches
> his back* You can locate it at my mismaintained code-page, will have
> to make it come back to life some day.
> 
> http://www.fukt.hk-r.se/~flognat/hacks/
> http://www.fukt.hk-r.se/~flognat/hacks/strptime.py
> 
> *stretching-his-backly-yours*
> 
>         /Andy
> 
> BTW Have you actually tried searching the python site :-)
>

Nope--not a bad idea, thanks.  I'm embarassed that I hadn't thought of
it: after I had searched Google (Deja) and came up with nothing, the
very next thing I thought of was to ask you folks.  (That way, I'll be
able to find it in Google _next_ time. <wink>)

Your code looks great, but as it turned out, I only needed a conversion
for purposes of sorting in terms of day, so I just wrote a little
funtion that generates a bogus "day number", which worked OK for that. 
(So I guess I must be deliberately wasting your time <wink>.)

somebody-oughtta-should-write-a-pep-to-add-strptime.py-to-the
   -time-module-<hint, wink>-ly y'rs,

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation            http://www.iowegian.com


From max at alcyone.com  Tue Mar 13 11:59:42 2001
From: max at alcyone.com (Erik Max Francis)
Date: Tue, 13 Mar 2001 08:59:42 -0800
Subject: What's wrong with me/Python ?
References: <3AAE0652.D68F253E@daimlerchrysler.com>
Message-ID: <3AAE51FE.AAB1234C@alcyone.com>

Alexander Stirzel wrote:

> I just wrote a small script for killing some hanging netscape
> processes,
> the source is listed below.
> Now if I run this script I get this error message:
> 
> 8<----------- error message -----------------------------
> Traceback (most recent call last):
>   File "/usr/sbin/netscape-killer.py", line 11, in ?
>     splitted = string.split(line)
> NameError: There is no variable named 'string'
> 8<----------- error message -----------------------------
> 
> Now here's the script:
> 
> 8<------------ script -----------------------------------
> #!/usr/local/bin/python
> import os
> import time
	...

> What did I do wrong? And why does Python want a "variable named
> 'string'" here?

Because you're referencing `string' and it's never seen the variable
before.  You're missing an `import string' statement at the top.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ I am a gentlemen:  I live by robbing the poor.
\__/ George Bernard Shaw
    Kepler's laws / http://www.alcyone.com/max/physics/kepler/
 A proof of Kepler's laws.


From aleaxit at yahoo.com  Fri Mar 23 01:43:30 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 23 Mar 2001 07:43:30 +0100
Subject: extending the built in dictionary type
References: <umfKujAz2iu6EwHa@jessikat.demon.co.uk> <YJzu6.12928$_46.482847@e420r-atl2.usenetserver.com>
Message-ID: <99eren016ps@news2.newsguy.com>

"Steve Holden" <sholden at holdenweb.com> wrote in message
news:YJzu6.12928$_46.482847 at e420r-atl2.usenetserver.com...
> Look in the library and inherit from UserDict. A search on UserDict
revelas
> that this probably isn't quite as FAQtually recorded as it might be.

Or else, as for any other built-in object, use the "automatic delegation
in lieu of inheritance" pattern.  But, from context, it seems Robin is
trying to extend dictionaries _in a C-coded extension_, which is a
whole 'nother ball of wax...:

> > Seems that although it exists I can't chain the original mapp_methods of
> > dictobject as that is not visible outside pythonxx.dll. So is there an
> > easier way to get at these internal types?


Alex





From Eugene.Leitl at lrz.uni-muenchen.de  Wed Mar  7 04:41:57 2001
From: Eugene.Leitl at lrz.uni-muenchen.de (Eugene Leitl)
Date: Wed, 7 Mar 2001 10:41:57 +0100 (MET)
Subject: GadFly and gfplus.py working with Python 2.0?
Message-ID: <Pine.GSO.4.03.10103071040190.13047-100000@sun1.lrz-muenchen.de>

See subject.
Aye, nay, maybe?

TIA,
-- Eugene
______________________________________________________________
ICBMTO  : N48 10'07'' E011 33'53'' http://www.lrz.de/~ui22204 
57F9CFD3: ED90 0433 EB74 E4A9 537F CFF5 86E7 629B 57F9 CFD3




From bowman at montana.com  Sun Mar 18 10:21:57 2001
From: bowman at montana.com (bowman)
Date: Sun, 18 Mar 2001 08:21:57 -0700
Subject: windows script host and python
References: <%tKr6.2570$pX.26174@news1.onlynews.com> <98oq3r$1e0$1@troll.powertech.no> <8Aps6.104$pl.1515@newsfeed.slurp.net> <98tc4d0rmk@news1.newsguy.com> <lvf8bt4p6e6naq0vimgnkhs175eoqbajfr@4ax.com>
Message-ID: <cv5t6.126$M62.975@newsfeed.slurp.net>

Tim Roberts <timr at probo.com> wrote in message
> >
> >Here, for example, is the "list all running processes" WMI scripting
> >example translated into Python:
>
> Cool!  Is this supposed to be NT-only?  It doesn't fly on my Win98 box.

http://msdn.microsoft.com/downloads/default.asp?URL=/code/sample.asp?url=/ms
dn-files/027/001/576/msdncompositedoc.xml

don't know if that reference will work. if not, go to msdb.microsoft.com,
click on 'downloads' and
look for 'Windows Management Instrumentation'. ignore the pessimistic blurb
about Win98 SE.

no. there is a core WMI that installs on Win9.x and NT. I don't believe it
is on NT4.0 by default, but can be installed there. The SDK will install on
NT also, but i'm not sure it will install on Win9.x. The problem is the SDK
contains the .chm file that is required reading for playing around. it is
available on line at msdn.microsoft.com. It is a part of 2000/ME.

the WMI is ms's shot at WBEM. The long form to invoke it would be (sorry
'bout the Perl, but the Python calls are very similar):


$locator = Win32::OLE->new("WbemScripting.SWbemLocator");
if (!defined $locator) {
  die "couldn't get the locator object";
}

$server = $locator->ConnectServer();
if (!defined $server) {
  die "couldn't get the server object";
}

$processes = $server->InstancesOf("Win32_process");
if (!defined $processes) {
  die "couldn't get the processes";
}

The whole thing is pretty integral to the Win2000 world view. The idea is to
reduce TCO by allowing the sysadmin to query all systems in the net. WMI
probably gets into the AD implementation,too.
wbem is an industry initiative, so in theory one could use the locator to
connect to a linux, mac, etc server and access CIM data in a platform
independent way.

I recently did a unix style 'ps' for NT/2000 in C, and it was about 200
lines of pretty arcane calls into the HKEY_PERFORMANCE_DATA, psapi, ntdll
etc to get stuff like the parent pid, user name, ccommand line, etc, with
WMI, it take about 10 lines of Python/Perl/JScript/VBScript. the scripts are
a little slower, with the native WSH engines and Perl the fastest, and
Python lagging a bit. Still quite usable on a 233 mhz Win95 box.




From support at internetdiscovery.com  Sat Mar 24 07:08:21 2001
From: support at internetdiscovery.com (Mike Clarkson)
Date: Sat, 24 Mar 2001 12:08:21 GMT
Subject: Why Python/Jython?
References: <99cl9e$b2b$1@Merlin.et.bocholt.fh-gelsenkirchen.de> <mailman.985260089.28444.python-list@python.org> <srzu6.12868$_46.475705@e420r-atl2.usenetserver.com>
Message-ID: <3abc8d36.83155745@24.0.228.33>

On Thu, 22 Mar 2001 22:26:59 -0500, "Steve Holden"
<sholden at holdenweb.com> wrote:

>"Steve Purcell" <stephen_purcell at yahoo.com> wrote in message
>> Programmer performance is usually gauged by coding time and
>> number of bugs, both of which are vaguely proportional to the number of
>> lines of code in most sane languages*.
>>
>> * Perl is a noteworthy exception
>>
>I saw no inconsistency here. Why would Perl be an exception to your rule?

It's not a sane language :-)

It's like vicegrips: the wrong tool for every job.

Mike.



From owen at astrono.spamwashington.emu  Fri Mar  2 10:59:12 2001
From: owen at astrono.spamwashington.emu (Russell E. Owen)
Date: Fri, 02 Mar 2001 07:59:12 -0800
Subject: PEP 227 (was Re: Nested scopes resolution -- you can breathe again!)
References: <LNBBLJKPBEHFEDALKOLCMEEPJBAA.tim.one@home.com> <200102230259.VAA19238@cj20424-a.reston1.va.home.com> <mailman.983384008.21434.python-list@python.org> <97lqsr0311d@news2.newsguy.com> <mailman.983466367.2968.python-list@python.org> <KNSz0UAhfpn6EwZj@jessikat.fsnet.co.uk>
Message-ID: <owen-0203010759120001@cs206-69.fsmodem.washington.edu>

In article <KNSz0UAhfpn6EwZj at jessikat.fsnet.co.uk>, Robin Becker <robin at jessikat.fsnet.co.uk> wrote:

>In article <mailman.983466367.2968.python-list at python.org>, Jeremy
>Hylton <jeremy at alum.mit.edu> writes
>>Or: global NAME in FUNCTION
>>
>>def eggs():
>>    z = 2
>>    def spam():
>>        def inner():
>>            global x in spam, z in eggs
>>            x += 1
>>            z += x
>>        x = 23
>>        inner()
>>
>>Jeremy
>>
>I like that a lot better than having it default possibly wrongly, but
>how would this be doable/spelled in the lambda case?

As somebody else pointed out, a nice possible notation for that has been proposed. 

lambda arg1, arg2; externalVar1, externalVar2: expression

One interesting question is whether we could just implement that and throw out the rest of the nested scope proposal. If so, that gets my vote, followed by an explicit keyword for external variables, followed by chucking the whole thing and living with ugly lambdas. I loathe the idea of externally declared variables polluting a function's namespace. That works fine in languages that make you declare all variables, but I do not think it is safe or reasonable in Python.

-- Russell


From aleaxit at yahoo.com  Wed Mar 14 09:02:20 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Wed, 14 Mar 2001 15:02:20 +0100
Subject: Extending Python: rewriting a single method in C
References: <g0wv9vxms0.fsf@scumbag.ecs.soton.ac.uk> <slrn9apn8f.382.jepler@potty.housenet> <g0ofv6xbj4.fsf@scumbag.ecs.soton.ac.uk> <98l1i901k4u@news1.newsguy.com> <g0lmq9xxz0.fsf@scumbag.ecs.soton.ac.uk> <98m6g20fgv@news1.newsguy.com> <g0g0ggy7ct.fsf@scumbag.ecs.soton.ac.uk>
Message-ID: <98ntl80vrp@news2.newsguy.com>

"Jacek Generowicz" <jmg at ecs.soton.ac.uk> wrote in message
news:g0g0ggy7ct.fsf at scumbag.ecs.soton.ac.uk...
    [snip]
> > But attribute-getting works just fine for simple cases without
> > you writing anything special!
    [snip]
> OK . . . I guess this means that the default implementation is not
> accessible via __getattr__, but it is available via PyObject_GetAttr ?

Yes, and also available via builtin-function getattr.


> > > (Would it be fair to say that the documentaion is, on the whole, less
> > > that satisfactory ?)
> >
> > I find it pretty decent, myself.
>
> I guess that it is if you know what you are looking for and where to
> find it :-)

With the HTML-Help form of the docs (I forget the helpful person
from whose site I downloaded that from), searching throughout the
docs is instantaneous -- which roughly makes up for defects in
the index and/or organizational problems in the docs themselves.

It does help to read the sources (which are very well organized,
clear, and readable) -- except that sometimes one may be left in
doubt if something "just happens to work this way today" vs "is
_designed_ to work this way forever":-).


Alex





From root at [127.0.0.1]  Thu Mar 29 02:11:43 2001
From: root at [127.0.0.1] (deadmeat)
Date: Thu, 29 Mar 2001 07:11:43 GMT
Subject: list vs tuple
References: <000e01c0b7ac$25f9f510$b503a8c0@activestate.ca> <mailman.985801119.10859.python-list@python.org> <2595ctkf19m63bmkd62dq73cgvbl6d002j@4ax.com>
Message-ID: <PeBw6.816$p5.3450@news1.rivrw1.nsw.optushome.com.au>

> >>> dir_list = ('/path', '/foo/bar', '/baz/bat/files')
> >>> print dir_list
> ('/path', '/foo/bar', '/baz/bat/files')
> >>> dir_list = ('/usr/bin/env', '/usr/bin', '')
> >>> print dir_list
> ('/usr/bin/env', '/usr/bin', '')
> I don't see how dir_list is a constant?

The values within it are read-only, not the name 'dir_list'

dir_list[0] = 'this will fail'

convert to a list and run it again it'll work fine.






From vlouie at telusplanet.net  Sun Mar 25 16:44:06 2001
From: vlouie at telusplanet.net (Victor Louie)
Date: Sun, 25 Mar 2001 21:44:06 GMT
Subject: Is it possible to sort a dictionary?
Message-ID: <GEtv6.59583$Lm2.6378825@news0.telusplanet.net>

Hi

I was just wondering if anyone knows how to sort a dictionary by
value...since the dictionary is <key: value> and where my value is a list.

Thanks in advance

Victor




From thomas at xs4all.net  Wed Mar  7 17:30:07 2001
From: thomas at xs4all.net (Thomas Wouters)
Date: Wed, 7 Mar 2001 23:30:07 +0100
Subject: setjmp.h
In-Reply-To: <3AA52B80.507E23A7@packeteer.com>; from nospam@packeteer.com on Tue, Mar 06, 2001 at 10:25:04AM -0800
References: <3AA52B80.507E23A7@packeteer.com>
Message-ID: <20010307233007.E404@xs4all.nl>

On Tue, Mar 06, 2001 at 10:25:04AM -0800, Bob Purvy wrote:

> I'm trying to port Python 2.0 to pSOS (yes, that's just our little cross
> to bear here!).  I hit a roadblock with pypcre.c:  it seems to require
> long jumps, and there is no way to configure that out.  I'm still
> checking whether it's possible for us to enable them in our software,
> but we definitely don't have them right now.

> Is there any way around this requirement?  Maybe an earlier version of
> Python?

You can probably disable the pre module altogether, which would avoid the
use of pypcre.c. Python 2.0 comes with two regular expression engines, 'pre'
for the old, Perl Compatible Regular Expression Engine, and 'sre' for the
new, unicode aware, much more flexible Secret Labs Regular Expressions
Engine. 'sre' is the default, but 'pre' is still around for backward
compatiblity. 

-- 
Thomas Wouters <thomas at xs4all.net>

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



From ws-news at gmx.at  Thu Mar  8 04:54:33 2001
From: ws-news at gmx.at (Werner Schiendl)
Date: Thu, 8 Mar 2001 10:54:33 +0100
Subject: pipe problems
References: <3AA6D886.4DD8D3A4@iems.nwu.edu>
Message-ID: <984045555.82149@newsmaster-04.atnet.at>

If there is no data to read, readline it will block for such to become
available.
I don't think that cout.flush() is of any use, btw.
Maybe I'm wrong, but I think data can only be flushed out by the sender of a
pipe...
Further, what's if the program prints out something in between, or more than
1 line?

Maybe you can do the reading in a seperate thread.
Alternatively you can try to find out if there _is_ sth you can read.

hth
werner

Leonardo B Lopes <leo at iems.nwu.edu> wrote in message
news:3AA6D886.4DD8D3A4 at iems.nwu.edu...
> I use a program that doesn't have readline compiled in. So I figured, it
> would be easy to add a python wrapper around it. So I wrote the wrapper,
> using 'cat' as a test command. With 'cat' it works fine. But when I try
> using another program, such as 'python', the program just hangs at
> readline(). Anybody know how to accomplish this? Here is the code:
>
> #!/usr/local/bin/python
>
> import sys,os
> import cmd
>
> class Amplcmd(cmd.Cmd):
>
>         def __init__(self):
>                 cmd.Cmd.__init__(self)
>                 self.cin,self.cout = os.popen2('/usr/local/bin/python')
>
>         def default(self,line):
>                 self.cin.write(line+'\n')
>                 self.cin.flush()
>                 self.cout.flush()
>                 print self.cout.readline(),
>
>         def do_EOF(self,line):
>                 sys.exit()
>
> mycmd = Amplcmd()
> mycmd.cmdloop()
>
>
>
> --
> =======================================================================
> Leonardo B. Lopes                                      leo at iems.nwu.edu
> Ph.D. Student                                             (847)491-8470
> IEMS - Northwestern University             http://www.iems.nwu.edu/~leo




From deblNonOspammy at theworld.com  Wed Mar 14 23:14:51 2001
From: deblNonOspammy at theworld.com (David Lees)
Date: Thu, 15 Mar 2001 04:14:51 GMT
Subject: Help doing simple threading
References: <mailman.984574578.29018.python-list@python.org>
Message-ID: <3AB041C6.FDDB84D3@theworld.com>

Thanks, these comment were quite helpful.  The most serious error was
'a)'.  when I just put the main to sleep after starting up the threads
it worked fine.  The problem was that I was printing out the array
values, before all the threads finished executing.

David

Andreas Jung wrote:
> 
<snip>
> 
> Your code has several problems:
> 
> a) You don't synchronize your main thread with the threads you start.
>     That means your main thread continues running BEFORE the
>     started threads make any work.
> 
> b) Modifying shared resources within threads is dangerous. Think
>    about using locks to get rid of this problem.
> 
> c) Try the threading module. It gives you a better control and programmer
>     interface to work with thread in Python.
> 
> ----
> Andreas Jung (andreas at digicool.com)
> Digital Creations - www.digicool.com, www.zope.org


From rcameszREMOVETHIS at dds.removethistoo.nl  Sat Mar 31 17:29:20 2001
From: rcameszREMOVETHIS at dds.removethistoo.nl (Robert Amesz)
Date: Sat, 31 Mar 2001 22:29:20 GMT
Subject: Gaussian line profile using numeric python
References: <3ac31698@zfree.co.nz> <m3puf0re81.fsf@chinon.cnrs-orleans.fr> <Pine.SOL.4.30.0103291850410.13267-100000@primrose.csv.warwick.ac.uk> <m3lmpn7dxw.fsf@chinon.cnrs-orleans.fr>
Message-ID: <Xns90763FC63132rcamesz@127.0.0.1>

Konrad Hinsen wrote:

>"John J. Lee" <phrxy at csv.warwick.ac.uk> writes:
>
>> > 2) If you work with large data sets, you ought to use FFTs to
>> >    compute the convolution, that's O(N*log(N)) in the size of the
>> >    data set, instead of O(N**2) for the straightforward method.
>> >    Any good book on FFTs should explain how this works in detail.
>> 
>> but you probably don't need to know in detail, it boils down to
>> 
>> inv_FFT(FFT(data)*FFT(kernel))
>
>Except for zero-padding in case of non-periodic signals. That's why
>I always recommend to look at a detailed description before doing
>this; some of my colleagues keep telling me that FFT techniques "don't
>work", and I suspect this is the reason.

Well, why dont you/they try a different kind of integral transform, 
that might give a better result. Perhaps something like wavelets? Of 
course, a fast algorithm must be available to do the transform and its 
inverse but if there is it might be worth to do some experimenting.


Robert Amesz


From kalle at gnupung.net  Tue Mar 27 08:12:30 2001
From: kalle at gnupung.net (Kalle Svensson)
Date: Tue, 27 Mar 2001 15:12:30 +0200
Subject: Reading e-mail body (newbie)
In-Reply-To: <3AC08CBB.52061537@geo.co.za>; from jacobus@geo.co.za on Tue, Mar 27, 2001 at 02:51:08PM +0200
References: <3AC08CBB.52061537@geo.co.za>
Message-ID: <20010327151230.A17310@father>

Sez Jacobus van der Merwe:
> Hi,
> 
> I have a script to which e-mail to a certain address is relayed.
> 
> I need to write the body of the e-mail to a file.
> 
> I use
> 
> rfc822.Message(sys.stdin)
> 
> to read the e-mail, but I can't figure out how to access the body.
> It seems that rfc822 only have methods to access the headers. ????

Read from rfcobj.fp, I think:

rfcobj = rfc822.Message(sys.stdin)
body = rfcobj.fp.read()

Peace,
  Kalle
-- 
Email: kalle at gnupung.net     | You can tune a filesystem, but you
Web: http://www.gnupung.net/ | can't tune a fish. -- man tunefs(8)
PGP fingerprint: 0C56 B171 8159 327F 1824 F5DE 74D7 80D7 BF3B B1DD
 [ Not signed due to lossage.  Blame Microsoft Outlook Express. ]



From francois.granger at free.fr  Fri Mar 16 15:43:28 2001
From: francois.granger at free.fr (=?ISO-8859-1?Q?Fran=E7ois_Granger?=)
Date: Fri, 16 Mar 2001 20:43:28 GMT
Subject: Python style questions
References: <98tb91$9t9$1@saltmine.radix.net> <98tdu1$cfk$1@panix2.panix.com>
Message-ID: <1eqdo5u.1xxmzfi2kjk0qN%francois.granger@free.fr>

Aahz Maruch <aahz at panix.com> wrote:

> In article <98tb91$9t9$1 at saltmine.radix.net>,
> Cary O'Brien <cobrien at Radix.Net> wrote:

> >3. I really miss not having a "switch" or "case" statement.  Sniff.

> Use a dictionary.

Some code snippet as exemple ?
At this point, I can't imagine a good way, and I don't remember seeing
something on this.

-- 
Le "bon sens" ?tant la chose du monde la mieux partag?e, 
il est normal que chacun en ai si peu - Inconnu


From fredrik at pythonware.com  Thu Mar 29 03:53:32 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Thu, 29 Mar 2001 08:53:32 GMT
Subject: htmllib samples
References: <3AC2F48E.4EF77C71@aon.at>
Message-ID: <gKCw6.4130$4N4.837469@newsc.telia.net>

Oliver Vecernik wrote:
> has anyone got some samples or point me to where to find them on how to
> implement this module?

you mean "use", I assume?

http://www.oreilly.com/catalog/pythonsl/chapter/ch05.html#HTMLLIB-MODULE
http://www.oreilly.com/catalog/pythonsl/chapter/ch05.html#FORMATTER-MODULE

note that when you think you need the htmllib module, you
usually want the sgmllib module:

http://www.oreilly.com/catalog/pythonsl/chapter/ch05.html#SGMLLIB-MODULE

Cheers /F




From aleaxit at yahoo.com  Tue Mar 13 05:46:28 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Tue, 13 Mar 2001 11:46:28 +0100
Subject: Extending Python: rewriting a single method in C
References: <g0wv9vxms0.fsf@scumbag.ecs.soton.ac.uk> <slrn9apn8f.382.jepler@potty.housenet> <g0ofv6xbj4.fsf@scumbag.ecs.soton.ac.uk>
Message-ID: <98l1i901k4u@news1.newsguy.com>

"Jacek Generowicz" <jmg at ecs.soton.ac.uk> wrote in message
news:g0ofv6xbj4.fsf at scumbag.ecs.soton.ac.uk...
    [snip]
> AM> > Yes, but I was trying to avoid the function call overhead
> AM> > (particularly as one of the functions I need to rewrite is very
short
> AM> > indeed).
> AM>
> AM> No way to do that -- the function call overhead is there even
> AM> if you call directly from Python to the C extension.
>
> But isn't the function call overhead greater for calling python
> functions than it is for calling the C extensions ?

On my old box (Pentium-II 266) I measure about 1 microsecond
difference per call, yes: a loop of 1,000 'pass' takes about
1.1 msec, one of 1,000 calls to an empty built-in (which
ignores its args and returns None) about 2.2, one of 1,000
calls to a similarly-empty (takes *args, body is 'pass')
Python-coded function about 3.2 (that's in WinNT4 SP6, with
Python 2).  Much of that is overhead of global-name lookup,
of course (for uniformity I assigned either kind of function
to a global variable and called through that); moving it to
a local name is an important optimization:

def timecals(afun):
    start = time.clock()
    for i in range(1000):
        afun()
    stend = time.clock()
    print stend-start

Here, it's about 1.8 msec if the empty C-coded function is
passed as the argument, about 2.8 if the Python one is --
so, the global name lookup overhead was about 0.4 msec in
each case (and the rest of the empty-loop overhead another
0.7 msec or so, leaving about 0.7 vs 1.7 for C vs Python
coded function's _sheer_ call overhead from Python).

If microseconds (small fractions thereof on more modern
machines, actually) are so crucial, then, yes, you can
shave some further tiny fractions this way.


> AM> If the function/method is called from a few places, you might
> AM> manually inline in there, maybe.
>
> Not sure I understand what you mean exactly. Are you suggesting that I
> call an extension function wherever I would previously have called and
> instance method ?

Yes.

> How do I get at the members of a class in C if that class was defined
> in python ?

By passing the reference to the instance to the function, and
receiving it on the C side as a PyObject* -- you can then call
methods and access attributes on it from C, just as you could
from Python.  See the "O" format code in the documentation of
PyArg_ParseTuple (1.7 in the "Extending and Embedding" manual).


> JE> In fast_a and fast_b, you can get arguments by using the ParseTuple
string
> JE> "(Oii)" and then use PyObject_SetAttrString and PyObject_GetAttrString
> JE> to store or retrieve attributes (including, if necessary, instance
> JE> methods) from the instance by their name.
>
> Is there an organized reference where I can find the specifications of
> all these functions ?  All I have come across shows example code, but
> no full descriptions of what can/must be done.

It's somewhat disorganized.  I think Beazley's "Python Essential
Reference" (New Riders) is your best bet among current books (but
it only covers Python 1.5.2).


> Maybe some kind soul could show me what the correct code for a minimal
> extension type is ?
>
> For example, I tried to write an extension type equivalent to
>
> class foo:
>     def __init__ ( self, a ):
> self.a = a
>     def show ( self ):
> print a
>
> My various attempts, almost work, but there is always something not
> quite right.

Getting equivalence to this is not going to be easy -- the
type/class separation hits hard, and calling Python's "print"
statement from C is not quite trivial either.


> Incidentally, why does the xxobject.c file in the Objects directory of the
> Python distribution contain no initxxobject function ?

Hmmm, because it's meant for a "real builtin type" (in a hypothetical
modification of Python) rather than for an extension...?


Alex





From juergen0509 at web.de  Sun Mar  4 16:07:42 2001
From: juergen0509 at web.de (Juergen Bocklage)
Date: Sun, 04 Mar 2001 22:07:42 +0100
Subject: should i or shouldnt i?
References: <1103_983731947@breakdown>
Message-ID: <3AA2AE9E.91D351C@web.de>

Fist take a look at the tutorial which comes with the python
distribution.
When you still think that you need a book, I'm happy with the Quick
Python Book from Daryl Harms and Kenneth Mc Donald. 
Also the is a online book Dive into Python from Mark Pilgrim
(www.diveintopython.org), which makes a good impression.

- JB

matt wrote:
> 
> i'm doing a course which revolves around smalltalk, but the pace is a bit slow for me.
> i was thinking of tackling another language to fill the hours and wondered about python.
> am i mad or would it be a good idea? what can python offer me? if theres 1 book to buy (ive already downloaded eckels book), which should it be?
> i run win98 and linux.
> thanks

-- 
Juergen Bocklage
================================
mailto:juergen.bocklage at gmx.net


From eye at niksula.hut.fi  Tue Mar 20 12:16:46 2001
From: eye at niksula.hut.fi (Petri Mikael Kuittinen)
Date: 20 Mar 2001 19:16:46 +0200
Subject: Regular expressions and non-standard character set
Message-ID: <ka8y9u01htt.fsf@cheerful.cs.hut.fi>

I want to match word boundaries using the special sequences \b and \B
of regular expressions. They work OK when using the "standard"
alphanumeric set [a-zA-Z0-9_]. But I would like them to work with
character set which also contains various "national characters"
e.g. ?, ?, ?, ?, ?, ?, ? etc. and their uppercase equivalents.

Locale doesn't seem to be the proper way to do it and appending these
characters to string.letters, string.uppercase, string.lowercase,
before import re doesn't help at all.

What is the proper way of doing it?

Petri

-- 
<(O)> Petri Kuittinen, also known as Eye, Dj Eye or Peku               <(O)>
<(O)> ADDRESS: Postipuuntie 10 A 14, FIN-02600 Espoo, Finland          <(O)>
<(O)> EMAIL: eye at iki.fi WWW: http://www.iki.fi/~eye/ PHONE: 09-5472380 <(O)>
~If dandelions were hard to grow, they would be most welcome on any lawn.


From kca17 at dial.pipex.com  Thu Mar 29 08:22:10 2001
From: kca17 at dial.pipex.com (Eddie and Babs)
Date: Thu, 29 Mar 2001 14:22:10 +0100
Subject: Quoted identifiers in Python?
Message-ID: <3ac3387f$0$12248$cc9e4d1f@news.dial.pipex.com>

While reading an article about the Dylan language a few years ago, I was 
surprised to find that you could have quoted identifiers (ie, variable and
function names, etc). So you could have a function named "turn around" (with
quotes) and then call it using something like:

    "turn around"(xxx)

The idea is that you can put any characters within the quotes.

NOW... would this be a good thing for Python to have in the future? It would
make element access consistent with subscripting:-

        x = obj."Ivor the engine"

            ...would be the same as

        x = obj["Ivor the engine"]

..no?

But the particular advantage would be with __getattr__ (ie, defining
attribute access for a class). It seems that the most obvious use of
__getattr__ is to allow Python-esque access to a dynamic system external to
Python. For example, you could have a class which allowed access to the file
system via Python objects:-

    import filesys

    # MyHD is a hard disk which contains the path "MyHD/projects/web/index".

    fileObj = FileSys()

    textOfFile = fileObj.MyHD.projects.web.index.readAll()


...etc. This falls down (slightly) when the file is called "index.htm", as
the dot is interpreted incorrectly by Python. What you need to be able to
say is:-

    textOfFile = fileObj.MyHD.projects.web."index.htm".readAll()

You could of course use the subscripting notation
(MyHD.projects.web["index.htm"]) but this seems to disrupt the flow
somewhat.


Maybe this idea is flawed in ways I can't see, but I thought it was at least
worth mentioning.


Cheers,
&.



From tim.one at home.com  Mon Mar 12 23:41:37 2001
From: tim.one at home.com (Tim Peters)
Date: Mon, 12 Mar 2001 23:41:37 -0500
Subject: This math scares me
In-Reply-To: <yu993dcimjt8.fsf@europa.research.att.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCOEEFJFAA.tim.one@home.com>

[Andrew Koenig]
> I'm curious -- Does Python now follow either the IEEE or the more
> restrictive Scheme rules for floating-point conversion?
> ...

All aspects of Python fp are inherited from whatever the platform C compiler
and libraries happen to do.  repr() conversion to string uses the platform
sprintf's %.17g format; str() conversion %.12g; both are fiddled just a
little to ensure that an exponent or a decimal point (or both) appears.
string->float is a x-platform adventure, since some platforms can't read back
all the C doubles they print (thinking of infinities and NaNs here); Python
inherits that too, whenever it happens.

it's-all-x3j11's-fault<wink>-ly y'rs  - tim




From loewis at informatik.hu-berlin.de  Tue Mar 20 15:14:55 2001
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 20 Mar 2001 21:14:55 +0100
Subject: New PEP: The directive statement
References: <mailman.985117312.5806.python-list@python.org>
Message-ID: <j4u24oi4e8.fsf@informatik.hu-berlin.de>

"Clark C. Evans" <cce at clarkevans.com> writes:

> >     Introducing 'directive' as a new keyword might cause
> >     incompatibilities with existing code. Following the guideline in
> >     [1], in the initial implementation of this specification,
> >     directive is a new keyword only if it was used in a valid
> >     directive_statement (i.e. if it appeared as the first non-string
> >     token in a module).
> 
> I'm new to python, but why not have the module have a built-in
> set of directives as a list?  As this could then be an attribute
> of many different objects, including modules.
> 
> __directives__ = { 'transitional', 'nested-scopes' }

The main problem is that the intended use really is not for run-time
semantics, i.e. the interpreter won't dynamically look at
__directives__ and do magic things. Instead, it is meant as a
"compile-time" thing, i.e. with effects when the source code is
interpreted. In that sense, it is the second declaration in python
(after global); normally, Python statements are actions (including def
and class).

That is also why the PEP 236 proposal (__future__ imports) finds that
much resistance: it uses an action statement to declare things.

Regards,
Martin

P.S. If "directive transitional nested_scopes" is approved, an
interesting observation is that all Python declarations deal with
variable visibility and name binding.


From grante at visi.com  Wed Mar  7 14:17:22 2001
From: grante at visi.com (Grant Edwards)
Date: Wed, 07 Mar 2001 19:17:22 GMT
Subject: ANNOUNCE: 2.1b1 SRPM
References: <mailman.983944149.1622.python-list@python.org> <3aa861c3.1122387@news.newsguy.com>
Message-ID: <6Pvp6.1459$y6.239152@ruti.visi.com>

In article <3aa861c3.1122387 at news.newsguy.com>, Owen F. Ransen wrote:
>On Tue, 6 Mar 2001 22:48:35 -0700, Sean Reifschneider
><jafo-pythonlist at tummy.com> wrote:
>
>>The SRPM is available at:
>What is an SRPM?

RPM is Redhat Package Manager: an open-source project that
provides an infrastructure to install/uninstall packages under
Unix (with support for package dependancies).  Packages can be
distribute as binaries or sources.  SRPM is a Source RPM
package that contains package sources as well as configuration
build information.  A binary package is often referred to as an
"RPM".  (e.g.  I found a SPARC Python2.1 RPM at .....)

RPM is the "native" package format used by several Linux
distributions (RedHat, Mandrake, Caldera?).

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


From fpm at u.washington.edu  Thu Mar  8 17:26:10 2001
From: fpm at u.washington.edu (Frank Miles)
Date: 8 Mar 2001 22:26:10 GMT
Subject: Database with Python ??
References: <988vmn$t77$05$1@news.t-online.com>
Message-ID: <9890u2$uiq$1@nntp6.u.washington.edu>

In article <988vmn$t77$05$1 at news.t-online.com>,
Thomas Mundt <lux_mundt at t-online.de> wrote:
>Has Python a own database language and do someone know a link to a Tutorial
>in database-working with python?

pygresql is commonly available with postgres.  There's some sort of
python link to mysql as well.

HTH...

	-frank
-- 


From fredrik at pythonware.com  Thu Mar 22 10:45:02 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Thu, 22 Mar 2001 15:45:02 GMT
Subject: email address verification
References: <99d5qi$95l$1@solaria.cc.gatech.edu>
Message-ID: <26pu6.3007$4N4.413401@newsc.telia.net>

Joseph wrote:
> does python provide quick/simple way to verify that an email address exist?

not more than any other programming language -- there is no
way to do what you want.

for the full story, see:
http://www.perl.com/pub/doc/manual/html/pod/perlfaq9.html#How_do_I_check_a_valid_mail_addr

    Q: How do I check a valid mail address?
    A: You can't, at least, not in real time. Bummer, eh?

Cheers /F





From cce at clarkevans.com  Tue Mar 27 04:20:30 2001
From: cce at clarkevans.com (Clark C. Evans)
Date: Tue, 27 Mar 2001 04:20:30 -0500 (EST)
Subject: Yet Another PEP:  Query Protocol Interface or __query__
In-Reply-To: <5.0.2.1.0.20010326100519.02198bb0@mail.inet.com.br>
Message-ID: <Pine.LNX.4.21.0103270416080.17716-100000@clarkevans.com>

On Mon, 26 Mar 2001, Carlos Alberto Reis Ribeiro wrote:
> a) The proxy adapter is completely transparent to the object instance.
> b) The adapter maps only the methods that are mandatory by the protocol 
> specification. This leads to better encapsulation and abstraction (all 
> details of the object are hidden from the protocol user).
> c) Method names may be different in the object. The protocol does the name 
> translation.
> d) The protocol instance may point to methods in disparate objects, which 
> leads to a interesting type of 'composition'.

Right.  I think this complements what I was proposing nicely;
and could be built-in as a way for adapter/proxy to be built.

> Hummm... I see your point. My intention is different - avoid mixing 
> 'protocols' and 'class definitions'. One of the reasons was the potential 
> for name clashes, as shown in my example. A protocol does not need to be 
> represented *exclusively* by a class in the hierarchy. A single class may 
> implement multiple protocols, but some protocols may need a composition of 
> objects to do the work. For example, a composite object may contain lots of 
> objects that gets exposed through interfaces in the container.
> 
> Also I see no reason to avoid defining some kind of 'simple' interfaces in 
> this PEP - in the end, protocols are better mapped to 'interfaces' than to 
> 'classes'. It's much more flexible in this way.

Yep.  I like your mechansim for supporting interface proxy's.

> class ProtocolSequence:
>    """ Protocol to implement sequence-like objects """
>    def __len__ (self): pass
>    def __getitem__(self, key): pass
>    def __setitem__(self, key, value): pass
>    def __delitem__(self, key): pass
> 
> In this example, ProtocolSequence is the specification of the methods used 
> for sequence type objects. Sequence objects does not need to be derived 
> from ProtocolSequence; they may be written from the scratch, or be derived 
> from UserList, etc. Again, keep in mind the potential of composition as a 
> technique to build large systems. It is important to avoid the assumption 
> that inheritance is *the only way* to build OO systems.

In this respect, what you are proposing is very much aligned
to what the interface proposals out there try to do, i'd hazard
to say this is the "bridge" between the two proposals.  No?

Best,

;) Clark




From root at rainerdeyke.com  Sat Mar  3 13:22:18 2001
From: root at rainerdeyke.com (Rainer Deyke)
Date: Sat, 03 Mar 2001 18:22:18 GMT
Subject: Missing the functional mind set
References: <mlb2at8ptrr7sh5dgq8qlk2201kquh6tfl@4ax.com>
Message-ID: <uDao6.335972$ge4.119457077@news2.rdc2.tx.home.com>

"Daniel Klein" <danielk at aracnet.com> wrote in message
news:mlb2at8ptrr7sh5dgq8qlk2201kquh6tfl at 4ax.com...
> The purpose of this code is to return a list of dictionary values sorted
by
> key, resulting in 'templist'.
>
> >>> d = {3: 'd', 2: 'b', 1: 'a', 0: 'z'}
> >>> templist = []
> >>> for n in range(len(d)):
> templist.append(d[n])

This is equivalent to:

templist = [d[n] for n in range(len(d))]


But if you are indexing with integers from 0 to n, why are you using a
dictionary in the first place?


--
Rainer Deyke (root at rainerdeyke.com)
Shareware computer games           -           http://rainerdeyke.com
"In ihren Reihen zu stehen heisst unter Feinden zu kaempfen" - Abigor




From James_Althoff at i2.com  Fri Mar 16 14:40:15 2001
From: James_Althoff at i2.com (James_Althoff at i2.com)
Date: Fri, 16 Mar 2001 11:40:15 -0800
Subject: I come to praise .join, not to bury it...
Message-ID: <OF315120A1.C93DC823-ON88256A11.00690C6F@i2.com>

All joking aside ...

Although overloading __if__ is not so common, overloading the equivalent of
"__for__  __in__" is a very common, convenient, and powerful design pattern
in Smalltalk  (see "Smalltalk Best Practice Patterns", Kent Beck, 1997,
ISBN 0-13-476904-X, page 146).  The idea is that for a collection class you
define a "do:" method that takes a block of code as the argument (this is
very easy to do in Smalltalk).  When invoked, the "do:" method in the
collection class executes the code block for each item in the collection.
That way the collection class does not have to export artifacts like lists,
iterators, or indexed-access to contained items just so that the collection
can be used inside built-in looping control structures (such as "for in").
Rather, the collection class takes care of the iteration across its
contained items in its own internal implementation.  The Python equivalent
defining a method that takes as the argument a pre-defined function (too
inconvenient to have to create a function for a random block of code -- and
lambdas can't have statements) or a code-string to eval or exec (too slow
and unreadable) is not anywhere near as convenient or usable, so this
design pattern is rarely seen in Python programs (compared to just pulling
each item out of the collection in a "for in" or "while" loop).

Jim






                                                                                                                   
                    "Ken Seehof"                                                                                   
                    <kens at sightreader.        To:     <python-list at python.org>                                     
                    com>                      cc:                                                                  
                    Sent by:                  Subject:     Re: I come to praise .join, not to bury it...           
                    python-list-admin@                                                                             
                    python.org                                                                                     
                                                                                                                   
                                                                                                                   
                    03/15/01 09:47 AM                                                                              
                                                                                                                   
                                                                                                                   




Hmm ... I wonder what I'd do if I could overload __if__ ...

Hey, I know!  You could use it to implement fuzzy logic for quantum
computers.

The new __if__ would cause multiple superimposed quantum states based on
the
probability wave distribution of the quantum bit argument.

----- Original Message -----
From: "Darren New" <dnew at san.rr.com>
Newsgroups: comp.lang.python
To: <python-list at python.org>
Sent: Thursday, March 15, 2001 12:15 PM
Subject: Re: I come to praise .join, not to bury it...


> Alex Martelli wrote:
> > And then again, the _appearance_ that too much emphasis
> > was being placed on syntax-sugary issues kept me *AWAY*
> > from Python for longer than it should have
>
> Actually, what I found quite cool about Python is that the procedural
syntax
> maps to OO semantics. I.e., in Smalltalk, everything looks like a method
> call, including "if". In Eiffel, everything that isn't a method call
doesn't
> look like a method call, and vica versa. But in Python, the convenient
and
> familiar "x = y[5:10]" maps to a real live method call you can override.
>
> Well, *I* thought it was neat. It lets you be OO where you want, and
keeps
> it out of your face otherwise.
>
> --
> Darren New / Senior MTS & Free Radical / Invisible Worlds Inc.
> San Diego, CA, USA (PST).  Cryptokeys on demand.
> --
> http://mail.python.org/mailman/listinfo/python-list
>


--
http://mail.python.org/mailman/listinfo/python-list







From f8dy at yahoo.com  Fri Mar  2 09:59:57 2001
From: f8dy at yahoo.com (Mark Pilgrim)
Date: Fri, 2 Mar 2001 09:59:57 -0500
Subject: Has anyone been able to convince Bruce Eckel to write  "Thinking in Python"
References: <BD674CDC0A68C642B60038986FBB56DF015C47@mailsrv.ASPATECH.COM.BR> <mailman.983482202.19147.python-list@python.org>
Message-ID: <97oce8$pdrjf$1@ID-77331.news.dfncis.de>

"Bruce Eckel" <Bruce at EckelObjects.com> wrote in message
news:mailman.983482202.19147.python-list at python.org...
> See:
> http://www.mindview.net/Python/ThinkingInPython.html

<plug class="shameless">
Sounds very promising.  Until this project gets underway, may I humbly
suggest my free book, "Dive Into Python":
  http://diveintopython.org/

It is targeted at programmers who have some experience at least one other
language (Java, Perl, C++, Delphi, Powerbuilder, or even just Visual Basic)
who can build on that experience to learn Python quickly.  It is organized
around annotated code examples, and it has "futher reading" links in each
section to relevant resources elsewhere on the web.  3 chapters are already
complete; more are coming.
</plug>

-M
You're smart; why haven't you learned Python yet?
http://diveintopython.org/





From joconnor at cybermesa.com  Thu Mar  1 11:14:13 2001
From: joconnor at cybermesa.com (Jay O'Connor)
Date: Thu, 01 Mar 2001 09:14:13 -0700
Subject: str behaviour
References: <xjvn1b55vs4.fsf@skepnuskapur.ifi.uio.no>
Message-ID: <3A9E7555.E3E7CC39@cybermesa.com>


"Igor V. Rafienko" wrote:

> Hi,
>
> I was playing around with printing tuples, when I noticed how str
> behaves with "compound" types (such as tuples):
>
> >>> class A:
> ...     def __str__( self ):
> ...         return "I'm A"
> ...
> >>> a = A()
> >>> t = ( A, a )
> >>> str( t )
> '(<class __main__.A at 80fd498>, <__main__.A instance at 810f090>)'
> >>> str( a )
> "I'm A"
> >>>
>
> I expected str to be called recursively for the constituent types.
> But, alas, that did not happen. Am I misunderstanding something or is
> it the intended behaviour? If the latter is the case, could someone
> elaborate on why this behaviour was chosen?

Not looking inside the code for printing a code, experimation indicates
that __str__ is not being called, but __repr__ is...

>>> class ClassA:
    def __str__(self):
        return "I'm an A"
    def __repr__(self):
        return "I'm really an A"

>>> a = ClassA()
>>> a
I'm really an A
>>> t = (ClassA, a)
>>> t
(<class __main__.ClassA at 0096F9AC>, I'm really an A)
>>>

--
Jay O'Connor
joconnor at cybermesa.com
http://www.cybermesa.com/~joconnor

Python Language Forum -
http://pub1.ezboard.com/fobjectorienteddevelopmentpython




From lars at pixar.com  Thu Mar  8 01:54:05 2001
From: lars at pixar.com (Lars Damerow)
Date: Wed, 7 Mar 2001 22:54:05 -0800
Subject: class object access in class definition?
In-Reply-To: <CFCp6.30994$ML1.1565785@e420r-atl2.usenetserver.com>
References: <Pine.LNX.4.21.0103061556510.2015-100000@zero> <slrn9ac2sa.ah1.scarblac@pino.selwerd.nl> <Pine.LNX.4.21.0103071835560.6130-100000@zero> <CFCp6.30994$ML1.1565785@e420r-atl2.usenetserver.com>
Message-ID: <Pine.LNX.4.21.0103072252350.6784-100000@zero>

That's a good compromise. It would have been nice to have everything under the
scope of the class definition, but certainly not crucial.

Thanks!
-lars

On Wed, 7 Mar 2001, Steve Holden wrote:

| Well, if you're prepared to be pragmatic about it, what's wrong with:
| 
| derivedClasses = []
| 
| class BaseClass:
|     pass
| 
| class DerivedClass(BaseClass):
|     pass
| 
| derivedClasses.append(DerivedClass)
| 
| Not elegant, but practical.
| 
| regards
|  Steve
| 
| 
| 
| 

___________________________________________________________
lars damerow
button pusher
lars at pixar.com

"Nauseous.  Nauseated.  The first means 'sickening to contemplate'; the second
means 'sick at the stomach.'  Do not, therefore, say, "I feel nauseous," unless
you are sure you have that effect on others."

- William Strunk Jr., "The Elements of Style"



From fuess at att.net  Thu Mar 15 07:48:13 2001
From: fuess at att.net (David Fuess)
Date: Thu, 15 Mar 2001 12:48:13 GMT
Subject: Problem with string.translate
Message-ID: <s3e1bt8ujpmbrtdcsojffc2u9mt63jebip@4ax.com>

I am using Python 2.0 and attempting to perform a string.translate.
The translate string is statically allocated in the setup phase. I
perform a

result = string.translate(source, ttab)

I have also tried

result = source.translate(ttab)

The error I get is that an internal library function in the string
module is calling a builtin function translate with too few arguments!
Is this an implementation problem in the new string class? I have
tried changing the library function, but then it generates other
errors. Please advise.

Thanks,
Dave


From matt at mondoinfo.com  Tue Mar  6 19:34:15 2001
From: matt at mondoinfo.com (Matthew Dixon Cowles)
Date: Wed, 07 Mar 2001 00:34:15 GMT
Subject: from Tkinter import *
References: <mailman.983920449.5230.python-list@python.org>
Message-ID: <slrn9ab0g7.s6.matt@happy-hour.mondoinfo.com>

On Tue, 6 Mar 2001 15:16:31 -0800, Timothy Grant
<tjg at exceptionalminds.com> wrote:

> All of Fredrik's examples in his guide on Tkinter use

>from Tkinter import *

>Now, I'm aware that the effbot is never wrong about anything<wink>,
>but is it safe to assume that I should not be using the above idiom,
>but the import Tkinter idiom instead?

Rest assured, the effbot is right as usual.

There are really two things going on with "from foo import *": The
first is that under some circumstances doing it at other than module
scope (e.g. in a function or method) is eventually going to become
illegal. That's unlikely to be a problem for very many people since
"from Tkinter import *" is almost always at the left margin.

The second reason someone might not want to do "from foo import *" is
that it's possible to cause surprises by importing a name that shadows
something important. If my module foo defines foo.open and I do "from
foo import *", I may get surprising results the next time I try to
open a file. Tkinter's names are chosen to make that sort of problem
unlikely.

Everyone assumes that everyone does "from Tkinter import *" and I
don't think that anyone is going to make it a bad idea any time soon.

Regards,
Matt


From danielk at aracnet.com  Tue Mar  6 09:10:30 2001
From: danielk at aracnet.com (Daniel Klein)
Date: Tue, 06 Mar 2001 06:10:30 -0800
Subject: Static class methods
References: <rdzo6.154$gc4.80445@news.pacbell.net> <aqa7at0b6cbqjaq23mli0flmumlp8n6s9s@4ax.com> <7SZo6.7063$qv3.3125634@nnrp5-w.sbc.net>
Message-ID: <qbr9atklivqc9mphck04suggdsbeubopqi@4ax.com>

On Mon, 5 Mar 2001 20:40:29 -0800, "Alex Shindich" <alex at shindich.com> wrote:

>There are indeed many ways to simulate the behavior. But shouldn't such a
>strong OOP language like Python have support for static methods?

It's not really necessary  to be honest. The approach outlined in TQJB was just
a 'suggestion' cos you seemed 'distressed' you couldn't define a class method
associated to a specific class.  When I need such a method I simply define it
as a 'top level' function in the module and call if from the class instances.
The fact that it isn't 'bound' to a particular class does not bother me. That's
what doc strings are for. :-) On occasion there are situations where one of
these 'top level' methods becomes useful to more than one class in the module
and then I reap the benefits of reuse.

Daniel Klein
Portland OR USA


From Jaeho.Lee at brooks.com  Wed Mar 21 17:34:17 2001
From: Jaeho.Lee at brooks.com (Lee, Jaeho)
Date: Wed, 21 Mar 2001 17:34:17 -0500
Subject: learning python...
Message-ID: <857F15D7E3D8D3118D290008C7CF058602C9757F@mail-naeast1.brooks.com>

> I've also got "Core Python Programming" and if I had to review it, I would
> say that it tries to be all things to everyone but never quite hits the
> mark. The examples always seem to fall short of what you are really
looking
> for. The author seems to explain the fundamental stuff very well but just
as
> it starts to get interesting, the section/chapter ends. I think instead of
> using extra thick paper and large fonts to give the visual appearance of
an
> impressive body of work, the author should have provided a bit more detail
> in some of the more advanced areas like network programming.

I was looking a second Python book in addition to "Programming Python 1". I
could not wait PP2. When I scan "Core Python Programming" through, my
impression is that "too small contents compared with the thickness of the
book". It contains about half of the contents that PP1, I think. And the
explanation is not that great. So I did not buy it. I did not read it
through. But I don't think it is worth to read.
My case, I read PP1 and using "python reference manual" as reference (as the
name says ^^). PP1 taught me how python is different from other language.
But I agree that it is not good for reference. I am waiting PP2 now. So I
can not say about it.

/Jaeho



From vecernik at aon.at  Wed Mar 28 07:47:10 2001
From: vecernik at aon.at (Oliver Vecernik)
Date: Wed, 28 Mar 2001 14:47:10 +0200
Subject: diff lists
Message-ID: <3AC1DD4E.4BA7D421@aon.at>

Hi,

I've got following two lists:

['a', 'b', 'c', 'd', 'e', 'f']
['e', 'c', 'f']

I'd like to have the result:

['a', 'b', 'd']

The list need not to be ordered. ['d', 'a', 'b'] will also be ok. What
is the most effective way to achive that result?

Oliver


From Eugene.Leitl at lrz.uni-muenchen.de  Wed Mar 28 10:56:16 2001
From: Eugene.Leitl at lrz.uni-muenchen.de (Eugene Leitl)
Date: Wed, 28 Mar 2001 17:56:16 +0200 (MET DST)
Subject: help with passing arguments to published objects a la Zope
Message-ID: <Pine.GSO.4.03.10103281707500.16808-100000@sun1.lrz-muenchen.de>

...but without Zope.

I'm trying to build an application with an embedded webserver, so far with
limited success. Frozen CGIHTTPServer.py is ideal in terms of performance
and footprint, the more so, since I can trap access to fake document in
the URI to a python method, packing the bulk of the application into a
single binary.

However, is there a simple mechanism, allowing me to pass arguments
via the URI, as well as use the CGI module (i.e. to parse arguments into a
dictionary)? For instance, I've got a java applet passing on information
via a POST method. This seems to be done wia a JavaScript
document.submitform.action, which calls a cgi-bin script. Assuming, I'll
trap a call to that document tree, where will the POSTed stuff be
coming through, STDIN?

What about arguments passed via the standard 

http://blah.net:666/cgi/raowr.cgi?rhesus=positive#maybe

scheme. How can I get at them, preferably mapped to function parameters?

The all singing/dancing/juggling Zope is too fat for my needs (and I might
have to support a lot of weird platforms, starting with AIX, where it may
not even build), as is Medusa. If it's impossible to add this
functionality to CGIHTTPServer without jumping through a lot of hoops, I
guess I could use Medusa/Bobo. It's still plain Python, so I only need to
get Python to build.

If I could get Bobo, that is. It seems to have vanished into Zope's
entrails. I'm browsing through Zope's source tree right now, and can't
even find it. Gah.

So, what are my options? Some help would be really appreciated.

TIA,

Eugene




From emile at fenx.com  Sat Mar 31 08:03:20 2001
From: emile at fenx.com (Emile van Sebille)
Date: Sat, 31 Mar 2001 05:03:20 -0800
Subject: newbie variables question
References: <mailman.986005960.10805.python-list@python.org>
Message-ID: <9a4kol$3j756$1@ID-11957.news.dfncis.de>

This is the normal way to get variables into strings.

--

Emile van Sebille
emile at fenx.com

---------
"Michael Hall" <py.list at mulga.com.au> wrote in message
news:mailman.986005960.10805.python-list at python.org...
>
> I've been trying to do this:
>
> SYSPRODIR = /home/admin/systemprofile
> os.system('mv /tmp/install.log SYSPRODIR/install.log')
>
> No matter how I play around with single and double quotes, this won't
> work.
>
> But this does work:
>
> os.system('mv /tmp/install.log %s/install.log' %s SYSPRODIR)
>
> Is this the normal/best way to do this in Python ... seems a little
> convoluted me, like there must be a simpler way. Bash for example would be
> happy with something like:
>
> mv /tmp/install.log $SYSPRODIR/install.log
>
> while PHP is happy with things like:
>
> include ("$SYSPRODIR/install.log");
>
> Thanks for any advice. I haven't found any clear info about this in
> Learning Python yet.
>
> Mick
>
>




From slhath at home.com  Wed Mar  7 10:31:50 2001
From: slhath at home.com (Scott Hathaway)
Date: Wed, 07 Mar 2001 15:31:50 GMT
Subject: how to move a window with overrideredirect(1) in tkinter
Message-ID: <Gvsp6.144432$bb.1485275@news1.rdc1.tx.home.com>

Hello All,

Can anyone tell me how to move a window with overrideredirect(1) in tkinter
(with a drag event)?

Thanks,
Scott





From mcfletch at home.com  Sun Mar 18 23:32:02 2001
From: mcfletch at home.com (Mike C. Fletcher)
Date: Sun, 18 Mar 2001 23:32:02 -0500
Subject: Help Rewriting/Refactoring/Rethinking Parsing Algorythm
In-Reply-To: <3AB5675A.8363C9F2@mindspring.com>
Message-ID: <000501c0b02d$8c35b200$a9807018@cr706570a>

Not precisely what you're trying to do, but I do most of my programming with
voice dictation (Dragon NS 5), and I find that a few command phrases are
fairly useful and let me program in just about any editor:

	py-equal (pie-equal) ->  ==
	py-init              -> __init__
	py-main              -> __main__
	py-name              -> __name__ # would be redefined for you...
	py-def               -> def
	dublex (dooblex)     -> wx\Caps Next Word\No-Space After
	triple-quote         -> \No-Space ''' \No-Space
	py-len               -> len

Looking at your list, adding:
	py-caps  -> \No-Space-On\Caps-On
	py-norm  -> \No-Space-Off\Caps-Off
	py-name  -> <word> \No-Space-On\Caps-On [ not sure if you can do this
without a macro... ]

Would give you:
	py-name number customers py-norm
	py-caps Abstract Class py-norm
	if error type py-equal 5 then
	dublex-python equalsign 6
	py-name whiskey type py-norm equalsign 'peachy'
	print "what's wrong with %s?" % py-name first name py-norm

For the indicated phrases.  I find that the py-phrases give a decent
recognition rate.  With vocedit (for Dragon) you can setup the options for
the words easily (e.g. for dublex-, and making _ put no spaces around
itself).

I like the idea of adding "py-name" modes to the mix (would save lots of
mucking around with \Cap \No-space.  Will give it a try.  The dublex- thing
saves lots of headaches for me, I had a similar one for Fox when I was using
that library.

As for travelling the parsing-path:
	I think Aycock's parsing framework would fairly easily handle this kind of
work, it's got an extremely flexible algo (which is apparently fairly slow,
but should hold for interactive work I'd think).

Good luck, will be interested to see what you finally build,
Mike


-----Original Message-----
From: Boopy Bootles [mailto:aschneid at mindspring.com]
Sent: Sunday, March 18, 2001 8:57 PM
To: python-list at python.org
Subject: Help Rewriting/Refactoring/Rethinking Parsing Algorythm


I'm trying to write a simple piece of code to make programming by voice
recognition software easier to do.  I wrote a very simple function that
would, for example, convert "number customers equals 5" to
"numberCustomers = 5".  But once I started using it, I quickly
discovered several more cases I had to handle.  So far, this is the
list:

number customers            - >     numberCustomers
Abstract Class                 ->    AbstractClass
if error type == 5 then     ->    if errorType == 5:
whiskey x-ray python equals 6     ->     wxPython = 6
normal whiskey type = 'peachy'        ->    whiskeyType = 'peachy'
print "what's wrong with %s?" % first name     ->     print "what's
wrong with %s?" % firstName

(the whiskey x-ray stuff is the International Communications Alphabet,
which you use when spelling something out in Dragon NaturallySpeaking if
NaturallySpeaking is having trouble understanding you).

I've gotten all but the last case to work.  But once I started trying to
incorporate the last case--not messing up quoted strings--my already
overly messy code turned into a hideous snarl.  I know there's _got_ to
be a better way to parse this input, but I don't have a clue where to
start.  I'd like to avoid building a full-blown language parser, which
seems like overkill.

I've included the code below, which correctly translates all but the
last case.  Any thoughts would be greatly appreciated, esp. thoughts re:
a simple object-oriented approach (I'm pretty sure there is one, I just
don't have enough experience writing OO code to figure it out).

Thanks,
Anders Schneiderman

P.S. Once I've gotten this code in better shape, I'll post it
somewhere.  Even at this primitive stage, it makes a _huge_ difference
in writing code by voice (I use NaturallySpeaking plus Natlink, Joe
Gould's terrific system for writing NaturallySpeaking macros using
Python).

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

""" voicecode.py:  routines for translating and otherwise manipulating
voice input into code.
"""

from string import *

SpecialWords = {'equals': ' = ', '=': ' = ', '==': ' == ', '%': ' % ',
   'if': 'if ', 'then': ':', 'elsif': 'elif', 'dot': '.', '.':'.',
'open': ' = open(',
   'try': 'try:', 'init': '__init__ (self, ', 'define': 'def ',
'except':'except:',
   'finally':'finally:', 'tab':'   ', 'blank': ' '}
ICA = {'alpha': 1, 'bravo': 1, 'charlie':1, 'delta':1, 'echo':1,
       'foxtrot': 1, 'golf':1, 'hotel':1, 'india':1, 'juliet': 1,
       'kilo':1, 'lima': 1, 'mike':1, 'november':1, 'oscar':1,
       'papa':1, 'quebec':1, 'romeo':1, 'sierra':1, 'tango':1,
       'uniform':1, 'whiskey':1, 'x-ray':1, 'xray':1, 'yankee':1,
'zulu':1 }

quote = {"'":1, '"':1 }



def translate(words):
    """Given an array of words, translate into code.
The rules are:
 * In general, convert lists of words into wordWordWord
 * For certain words/symbols, aka "special words", convert 'em
 * For ICA words--alpha, bravo, etc.--convert to a single letter
 * If the word is "normal", use the next word exactly as is

NOTE:  Right now, this will NOT work on lines that are quoted:
    print 'this is a test' will translate to 'thisIsATest'.
    I need to find a cleaner way to solve this problem.
"""

    line = ''
    firstWord = 1
    normalWord = 0
    for word in words:
        if normalWord:
            # This word should be used exactly as is (prev word was
'normal')
            if firstWord:
                # Don't capitalize the first word of a variable or a
word in a quote
                line = line + word
                firstWord = 0
            else:
                line = line + capitalize(word)
            normalWord = 0
        elif lower(word) == 'normal':
            normalWord = 1            # next word should be used exactly
as is
        elif SpecialWords.has_key(word):
            line = line + SpecialWords[word]
            firstWord = 1
        elif ICA.has_key(lower(word)):
            # International Alphabet -- convert 'alpha' to 'a', etc.
            line = line + word[0]
            firstWord = 0
        elif firstWord:
            # Don't capitalize the first word of a variable or a word in
a quote
            line = line + word  # Don't capitalize the first word of a
variable
            firstWord = 0
        else:
            line = line + capitalize(word)
    return line




--
http://mail.python.org/mailman/listinfo/python-list




From mda at idatar.com  Fri Mar 30 16:12:10 2001
From: mda at idatar.com (David Allen)
Date: Fri, 30 Mar 2001 21:12:10 GMT
Subject: Know of Substantial Apps Written in Python?
References: <3ac4e4a8_1@news4.newsfeeds.com>
Message-ID: <KE6x6.415$SR6.235983@typhoon1.ba-dsg.net>

In article <3ac4e4a8_1 at news4.newsfeeds.com>, "Dry Ice" <nomail at nomail.com>
wrote:

> Know of substantial apps written in Python?
> 
> Preferably something one could download for examination.
> 
> Source code would be nice, but not essential.
> 
> I'm operating under the theory that whatever the POTENTIAL of a language, or
> the 'elegance' of a language's syntax and structure, how large apps actually
> behave in the real world will speak volumes.
> 
> I like, for example, much about Java.  But nearly every example I have seen
> running in the Windows environment has been slow, leaked resources and
> eventually contributed to system crashes.
> 
> (I don't care whether it's the language or the OS
> or the VM or even "bad programming."  What counts are the usual results
> obtained by normal application developers.)
> 
> Suggestions?
> 
> 

What's the definition of "substantial"?

There's Grail, a web browser
http://grail.sourceforge.net

There's ZOPE, an internet publishing thingy
(thingy is a technical term  :)

Check www.python.org - there are plenty of examples
of programs with thousands and thousands of lines of
code.  I'm not sure if you mean "large" when you
say substantial or if you mean well known and widely
used...
-- 
David Allen
http://opop.nols.com/
----------------------------------------
Anarchy may not be the best form of government, but it's better than no 
government at all.


From embed at geocities.com  Wed Mar 14 13:37:21 2001
From: embed at geocities.com (Warren Postma)
Date: Wed, 14 Mar 2001 13:37:21 -0500
Subject: Dynamic construction of a print format string
References: <3AAFB663.57FCCE7@troikanetworks.com>
Message-ID: <yVOr6.9485$TW.38415@tor-nn1.netcom.ca>

"Bruce Edge" quoth:
>   fmtstr = "%s%s%s" % ( "    %", namelen, "s - %s" )
> to get:
>   "    %19s - %s"

fmtstr = "  %"+`namelen`+" - %s"





From kens at sightreader.com  Thu Mar 15 16:29:15 2001
From: kens at sightreader.com (Ken Seehof)
Date: Thu, 15 Mar 2001 15:29:15 -0600
Subject: Magnitude of the wx* market (was: Python In A Nutshell - suggestions)
Message-ID: <005501c0ad96$fd34f5e0$04090a0a@upcast.com>

oops, sorry bout the accidental cross post.

----- Original Message -----
From: "Ken Seehof" <kens at sightreader.com>
To: "python-list" <python-list at python.org>
Sent: Thursday, March 15, 2001 3:27 PM
Subject: Re: Magnitude of the wx* market (was: Python In A Nutshell -
suggestions)


> I'll definitely buy a wxPython book when it comes out.  ...





From robin at jessikat.fsnet.co.uk  Wed Mar 28 11:44:46 2001
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Wed, 28 Mar 2001 17:44:46 +0100
Subject: StandaloneZODB at cvs.python.org
References: <mailman.985746941.27962.python-list@python.org>
Message-ID: <B1ix6IA+Thw6Ewia@jessikat.demon.co.uk>

In article <mailman.985746941.27962.python-list at python.org>, Jeremy
Hylton <jeremy at digicool.com> writes
...
>The project is tied directly to the ZODB code used in the Zope2 CVS
>repository.  When a change is made to Zope, it will also appear in
>StandaloneZODB, and vice versa.  Specifically, we use cvs modules to
>incorporate parts of the Zope2 and ZEO code bases.
>
>The project contains the key components you need to run ZODB and ZEO,
>but not all of the ancillary packages that Andrew's distribution
>includes.  I expect to add more components over time, but only those
>that are expected to be useful for Python programmers using ZODB.
...
so is there some advantage in maintaining two projects?
-- 
Robin Becker


From tuttledon at hotmail.com  Fri Mar  9 09:22:11 2001
From: tuttledon at hotmail.com (Don Tuttle)
Date: Fri, 09 Mar 2001 14:22:11 GMT
Subject: Python In A Nutshell
Message-ID: <nG5q6.54910$__6.11376601@typhoon.southeast.rr.com>

Hey Alex,  congrats on landing "Python In A Nutshell" book deal.  One
problem.  I went down to the book store and it's not there.  What gives?
<wink>

Don




From stephen_purcell at yahoo.com  Sun Mar 18 05:00:22 2001
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Sun, 18 Mar 2001 11:00:22 +0100
Subject: Help with 500 error on cgi script
In-Reply-To: <quv8bto7r8s6f1tckc7qpmuir0kvufp1nb@4ax.com>; from sheila@spamcop.net on Sun, Mar 18, 2001 at 09:33:57AM +0000
References: <quv8bto7r8s6f1tckc7qpmuir0kvufp1nb@4ax.com>
Message-ID: <20010318110022.C7546@freedom.puma-ag.com>

Sheila King wrote:
> Well, I was getting 500 internal server errors. I've checked everything. The
> path, etc...
> 
> I've compared my file to the other two files and just can't see what I'm doing
> wrong.
> 
> So, here is my script. I'm sure someone here will see it right off?


You need to write out a 'Content-type' header when form_ok is true, too!

It's the lack of that header that causes the 500 error. If you're pretty
sure your content is going to be text/html, it can be good practice to write
out the content header before any other code.

-Steve

P.S. What's with the 'spamcop' reply address? Is that your real address or
do unsuspecting helpful responders get their mail accounts terminated?

-- 
Steve Purcell, Pythangelist
Get testing at http://pyunit.sourceforge.net/
Any opinions expressed herein are my own and not necessarily those of Yahoo



From qrczak at knm.org.pl  Mon Mar  5 17:25:28 2001
From: qrczak at knm.org.pl (Marcin 'Qrczak' Kowalczyk)
Date: 5 Mar 2001 22:25:28 GMT
Subject: the demise of 'from foo import * and its implications?
References: <3kVBAfATrDo6EwEM@jessikat.fsnet.co.uk><mailman.983582584.8198.python-list@python.org><z8dAGJAyqLo6EwlX@jessikat.fsnet.co.uk> <mailman.983746566.27513.python-list@python.org> <suPo6.16824$ML1.841861@e420r-atl2.usenetserver.com>
Message-ID: <slrn9a84io.ogf.qrczak@qrnik.zagroda>

Mon, 5 Mar 2001 11:47:51 -0500, Steve Holden <sholden at holdenweb.com> pisze:

> I did begin to wonder whether it might be possible to specify a
> name prefix, along the lines of
> 
>     from modulewithaverylongname import * prefix sh_
> 
> to allow
> 
>     sh_i

import modulewithaverylongname as sh
for name in dir(sh): globals()['sh_'+name] = getattr(sh, name)
del sh

I don't see why do to this, and it's certainly not a feature to go
into the standard language.

An opposite trick can be used if a module is designed to be
star-imported and uses name prefixes, but we dislike star-import.
We can remove prefixes while moving objects to a separate namespace.

-- 
 __("<  Marcin Kowalczyk * qrczak at knm.org.pl http://qrczak.ids.net.pl/
 \__/
  ^^                      SYGNATURA ZAST?PCZA
QRCZAK


From spamfranke at bigfoot.de  Fri Mar  9 20:10:31 2001
From: spamfranke at bigfoot.de (Stefan Franke)
Date: Sat, 10 Mar 2001 01:10:31 GMT
Subject: A counter-counter-proposal for PEP 236: #pragma( nested_scopes)
References: <mailman.983656746.22190.python-list@python.org>
Message-ID: <3aaa72c4.9442134@news.btx.dtag.de>

On Sat, 3 Mar 2001 16:59:01 -0500, "Tim Peters" <tim.one at home.com> wrote:

>last-msg-from-me-on-this-until-there's-a-pep-ly y'rs  - tim
>

Hmm. Propably somewhat late to reply to this thread, but it's hard to keep up
with c.l.py these days...

I thought a little bit about the "from __future__ import nested_scopes" stmt, 
and what I found most disturbing is that it breaks the "explicit is better than 
implicit" rule. Wasn't it you who...?

IMO the fact that we have a real module __future__ in the filesystem that is 
*really* imported, but silently triggers some additional magic, is not a feature
but rather dangerous.

Consider a newbie who read one of the many pre-2.1 books or a Python
veteran who hasn't used the language for a while. What would they do
if they see the import statement?

They would propably see the __future__ module in the filesystem, notice 
that a tuple was imported and that's it! 
In the best case some comments around the definitions in the __future__ 
module are recognized that explain the additional magic.

But since the whole process is implicit, they wouldn't possibly ever dream up 
that there might be something special about it.

> "__future__" screams "Deep Magic" because of the double
> double-underscores;

I don't think so. We already have __init__.py modules all over. IMO they 
signal that the files are used by Python's runtime system, but not
necessarily "black magic".

 * * *

(Now for my own little ?0.02 proposal..)

Why not make __future__ an additional builtin function? Double underscore
names are reserved anyway, so we don't have the additional statement 
problem.
Ok, we still have the magic here, but the first thing I guess my imaginary 
newbies/veterans would do is to look up the behaviour of the 

  __future__("nested_scopes") 

call in the docs, since it is something unknown, and it must be built in (it
appears before any other statement or imports).

Maybe it's just my personal sense, but I find this explicit enough to make
someone looking into the docs, whereas the implicit import variant looks 
normal enough to be taken for a standard case.

Of course __future__ might not be the best name for the function(*). Note
that we get the other benefits too:

- Like before, the __future__("nested_scopes") call has to appear among 
  the first statements of a module, otherwise an exception is raised
- The version tuple could be the return value
- Additional parameters are possible
- There's no syntax change involved

from __future__ import *-ly y'rs
Stefan


[* Perhaps the name is everything: 

from __RTFM_immediately_when_you_see_this__ \
import stackless_behaviour_now

might be just as effective..]




From bsass at freenet.edmonton.ab.ca  Thu Mar  1 16:13:13 2001
From: bsass at freenet.edmonton.ab.ca (Bruce Sass)
Date: Thu, 1 Mar 2001 14:13:13 -0700 (MST)
Subject: [Distutils] Re: CPAN functionality for python - requirements
In-Reply-To: <20010228170725.B1781@tummy.com>
Message-ID: <Pine.LNX.4.33.0103011402420.13134-100000@bms>

> >Interestingly, developers probably don't "expect" to have to include
> >dependency information, and hence many don't - resulting in the problems you
>
> I think we can deal with that in an iterative manner.  First get them to build
> distutils packages, then when it fails to install on some user's machine
> because they don't have foo.py we can educate them on the joys of listing
> third-party module requirements.

Has anyone written a program that looks at a .py and tries to
determine the versions of Python it will work with?

Maybe have a module that grabs info about the system the code is being
developed on, then use some of that to generate default dependencies
which the developer can modify (be made less restrictive in most
cases, I imagine) before submitting the package to the archive.  Even
if the developer doesn't modify the dependencies, the user will at
least have some idea of what is required.


- Bruce




From jcm at bigskytel.com  Fri Mar 16 01:07:09 2001
From: jcm at bigskytel.com (David Porter)
Date: Thu, 15 Mar 2001 23:07:09 -0700
Subject: tempfile.tempdir
In-Reply-To: <3ab1a806$0$48743$6e49188b@news.goldengate.net>; from volucris@hotmail.com on Thu, Mar 15, 2001 at 11:43:41PM -0600
References: <3ab1a806$0$48743$6e49188b@news.goldengate.net>
Message-ID: <20010315230709.A7453@bigskytel.com>

* Voluciris <volucris at hotmail.com>:
> If I change the value of tempfile.tempdir is it necessary to change it back
> when I'm finished? 

no.

> Does the change just happen inside Python, without affecting any enviroment
> variables?

yes.




From tim.one at home.com  Thu Mar  1 00:09:25 2001
From: tim.one at home.com (Tim Peters)
Date: Thu, 1 Mar 2001 00:09:25 -0500
Subject: long int behaviour
In-Reply-To: <20010228172937.L14262@trufflehunter.avalongroup.net>
Message-ID: <LNBBLJKPBEHFEDALKOLCCEPDJCAA.tim.one@home.com>

[Timothy Grant]
> ...
> Given the following experimenting...
>
> >>> hex(4294965248L)
> '0xFFFFF800L'
> >>> hex(~4294965248L)
> '-0xFFFFF801L'
> >>>
>
> The first output is what I would expect, however, I would have
> expected the second output to be
>
> 0x000007FFL
>
> It appears to be something to do with two's complement storage,
> but I'm at a loss as to why my expectations were so incorrect.

Take heart!  Your expectations weren't born incorrect, they were made
incorrect, by a world of evil languages promulgating their blasphemous
treachery that integers are finite <wink>.

Python does store longs in 2's-comp form, but remember that Python longs are
unbounded:  a long less than 0 has a (conceptually) infinite string of
leading 1 bits, just as a long >= 0 has an infinite string of leading 0 bits.
It's easy to show you an infinite string of leading 0 bits:  we just ignore
them.  That infinite string of leading 1 bits isn't so easy to hide, though!

You've been further warped into believing that 32 bits is somehow a "natural"
size for an int.  If that's what you believe, you can force the issue:

>>> hex(~4294965248L & (2L**32 - 1))
'0x7FFL'
>>>

That is, by masking your input with a string of 32 low-order 1-bits, the
infinite string of leading 1 bits goes away, leaving the output you expected.
Or, if you're lucky enough to believe that ints have 101 bits, similarly:

>>> hex(~4294965248L & (2L**101 - 1))
'0x1FFFFFFFFFFFFFFFFF000007FFL'
>>>

So that's the trick:  to represent the infinite by the finite, you either
need a convention, or you need to throw an infinite amount of info away.  The
"-" in '-0xFFFFF801L' is Python's convention for compressing the unbounded
string of sign bits in such a way that no info is lost:

>>> eval(hex(~4294965248L)) == ~4294965248L
1
>>>

next-step:-infinite-memory-ly y'rs  - tim




From aahz at panix.com  Fri Mar 16 18:46:37 2001
From: aahz at panix.com (Aahz Maruch)
Date: 16 Mar 2001 15:46:37 -0800
Subject: list.sort(cmpfunc) question
References: <m1ae6lg5qt.fsf@pclabs188.labs.trema.com> <98t9dq$qvu$1@panix6.panix.com> <3ab2a132@si-nic.hrz.uni-siegen.de>
Message-ID: <98u8kt$35e$1@panix6.panix.com>

In article <3ab2a132 at si-nic.hrz.uni-siegen.de>,
Simon Budig  <Simon.Budig at unix-ag.org> wrote:
>
>As a related question since I did not find it in the documentation:
>
>Is the sort-function guaranteed to be stable? Does it keep the order,
>when cmpfunc returns 0 for a pair of items?

IIRC, the current implementation is stable but is not guaranteed to be
stable.
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

Three sins: BJ, B&J, B&J


From aleaxit at yahoo.com  Sun Mar  4 16:39:53 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sun, 4 Mar 2001 22:39:53 +0100
Subject: Missing the functional mind set
References: <mlb2at8ptrr7sh5dgq8qlk2201kquh6tfl@4ax.com> <uDao6.335972$ge4.119457077@news2.rdc2.tx.home.com> <t6f2at870pk82q96bnd5a2od4n8ns5kf83@4ax.com> <ntg2atcemm3hp9rb26hn7c5uin4dfn8lc9@4ax.com> <97rqgb$lt2$1@tyfon.itea.ntnu.no> <3AA298F5.7742810E@wwc.com>
Message-ID: <97ucs102v2k@news1.newsguy.com>

"Steve Williams" <stevewilliams at wwc.com> wrote in message
news:3AA298F5.7742810E at wwc.com...
    [snip]
> To be consistent with the "index == offset" mindset, counting should
> start at zero.
>
> As in:  "I have 2 children--0 boy and 1 girls".

Counting _does_ start at 0 -- you enumerate elements of a set
and check which ones satisfy a certain property, the count starts
and stays at zero until it's bumped up the first time, then it goes
to 1, and so on.  But 0+1=1, so you have an off-by-one error
somewhere in your family, I'm afraid...


Alex





From Dan.Rolander at marriott.com  Fri Mar  2 13:23:52 2001
From: Dan.Rolander at marriott.com (Rolander, Dan)
Date: Fri, 2 Mar 2001 13:23:52 -0500 
Subject: Unresolved symbols in MSVCRT.DLL when running a py2exe packag
	ed program.
Message-ID: <6176E3D8E36FD111B58900805FA7E0F80CCF63F3@mcnc-mdm1-ex01>

The problem with replacing individual DLLs is that they come in matched
sets. Unless you replace the whole set bad things can happen. It is often
easiest to install a service pack or product that'll upgrade the DLLs for
you. Or if you do use a professional installer like IPWI, load a merge
module to handle it for you.

This is an excellent article from Microsoft which explains the DLL Hell
pitfalls and how Win2K is attempting to handle it (along with a
troubleshooting tool for all Win32 systems): 
http://msdn.microsoft.com/library/default.asp?URL=/library/techart/DLLdanger
1.htm

Dan

-----Original Message-----
From: Tim Peters [mailto:tim.one at home.com]
Sent: Friday, March 02, 2001 1:10 PM
To: python-list at python.org
Subject: RE: Unresolved symbols in MSVCRT.DLL when running a py2exe
packaged program.


[Tom]
> I disagree.
>
> You can distribute system dll's.  You are expected to, and sometimes you
> have to.  You must insure that the version you load is at least
> equal to the version that you linked against.
>
> I have done this (distribute system DLL's) on all windows except WinME and
> the 16bit ones.  Unfortunately I can't say how I did it precisely, because
> I just used InstallShield.

And Python just uses the Wise installer.  They aren't that fancy under the
covers, though -- they just "follow the rules", checking version numbers,
bumping refcounts on refcounted DLLs, and arranging for a reboot if they
need
to replace something.  But that stuff is ineffective on the newer systems
with SFP.

> But I believe that, where SFP exists, you must use specific and new
> API's to replace DLL's, otherwise the system will back out your changes.

My understanding is that SFP isn't so easily fooled; e.g.,

    http://support.microsoft.com/support/kb/articles/Q253/5/71.ASP

    http://www.microsoft.com/hwdev/sfp/wfp.htm

SFP won't allow replacement of protected files via just using API tricks; it
also sucks in a digital signature scheme, and only MS can supply signatures
that work.

> One solution is to specify the changes in an .INF file and let the system
> worry about these details (but this has its limitations).
>
> I assume that Python itself updates the MSVCRT.dll.

The PythonLabs Windows installer tries to, yes.

> Might look at how it does it on Win2K.

My understanding is that it can't (but Win2K ships with a "good enough" set
of DLLs, so the version check shouldn't trigger there -- yet).

BTW, I'm *glad* MS is doing SFP!  Whatever Windows Life may be in theory,
DLLs in practice cannot be mixed and matched -- you need a self-consistent
set else Very Strange Things Happen.  And with about 1,000 DLLs under a
typical Windows system directory, that's just impossible for app developers
to keep straight.  May well be impossible for MS to keep straight too, but
at
least it's no longer impossible for them to *try* to keep 'em straight now.

something-to-ponder-each-time-your-win9x-crashes-ly y'rs  - tim


-- 
http://mail.python.org/mailman/listinfo/python-list



From sholden at holdenweb.com  Mon Mar  5 11:47:51 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Mon, 5 Mar 2001 11:47:51 -0500
Subject: the demise of 'from foo import * and its implications?
References: <3kVBAfATrDo6EwEM@jessikat.fsnet.co.uk><mailman.983582584.8198.python-list@python.org><z8dAGJAyqLo6EwlX@jessikat.fsnet.co.uk> <mailman.983746566.27513.python-list@python.org>
Message-ID: <suPo6.16824$ML1.841861@e420r-atl2.usenetserver.com>

"Jeremy Hylton" <jeremy at alum.mit.edu> wrote in message
news:mailman.983746566.27513.python-list at python.org...
> >>>>> "RB" == Robin Becker <robin at jessikat.fsnet.co.uk> writes:
>
>   RB> my preference would be that statements legal in one context
>   RB> should be legal in another where they make sense. I cannot see
>   RB> how using import * or eval etc in nested scopes is somehow
>   RB> obviously bad or different from the same usage in non-nested
>   RB> scopes. The fact that the prohibition comes from making things
>   RB> easier for the compiler adds to my confusion.
>
> Using import * in function's is asking for trouble, because it makes
> it impossible for someone reading the code to know what a variable
> refers to.  Consider this hypothetical bit of code:
>
[ ... ]

I would go further, and remove "in functions" from the above. While it is
true that wildcard imports are more manageable at module level there is
still always the chance that a badly-designed module will stamp on your
names.

Of course, the "import module as name" variant allows us to define shorter
names to be qualified by the names from the module. This can help a lot if
you wnat to do

    import modulewithaverylongname

but don't want to have to refer to

    modulewithaverylongname.i

In that case you can just do

    import modulewithaverylongname as sh

and refer to

    sh.i

I did begin to wonder whether it might be possible to specify a name prefix,
along the lines of

    from modulewithaverylongname import * prefix sh_

to allow

    sh_i

as a reference to the module's "i", but I'm not sure that the gain is worth
the pain. Even if such a proposal were to be PEP-ified and accepted it could
not be mandated because that would break humongous amounts of existing code.

Thoughts?

regards
 Steve




From shaleh at valinux.com  Wed Mar 21 15:09:05 2001
From: shaleh at valinux.com (Sean 'Shaleh' Perry)
Date: Wed, 21 Mar 2001 12:09:05 -0800 (PST)
Subject: Substitute for join()
In-Reply-To: <Mo7u6.58$Hp.2896@typhoon.aracnet.com>
Message-ID: <XFMail.20010321120905.shaleh@valinux.com>

On 21-Mar-2001 Daniel Klein wrote:
> I've got to make a script (written in version 2.0) to work on a machine
> running 1.5.2. The only incompatible line is one that uses the
> "".join(mylist) method. To 'get out of the woods' quickly, I substituted it
> with
> 
> s = ''
> for element in mylist: s = s + element
> 
> Is this the most efficient way to code this?
> 

or use the syntax that works in both:

import string # or from string import join

string.join(mylist, "")



From tjg at exceptionalminds.com  Wed Mar  7 12:15:14 2001
From: tjg at exceptionalminds.com (Timothy Grant)
Date: Wed, 7 Mar 2001 09:15:14 -0800
Subject: from Tkinter import *
In-Reply-To: <slrn9ab0g7.s6.matt@happy-hour.mondoinfo.com>; from matt@mondoinfo.com on Wed, Mar 07, 2001 at 12:34:15AM +0000
References: <mailman.983920449.5230.python-list@python.org> <slrn9ab0g7.s6.matt@happy-hour.mondoinfo.com>
Message-ID: <20010307091514.S3579@trufflehunter.avalongroup.net>

Thanks for the input. I realized that most of the issues come
from importing deeper than the module level, but the discussion
of from blah import * has been fascinating.

My primary concern was name clashes.

On Wed, Mar 07, 2001 at 12:34:15AM +0000, Matthew Dixon Cowles wrote:
> On Tue, 6 Mar 2001 15:16:31 -0800, Timothy Grant
> <tjg at exceptionalminds.com> wrote:
> 
> > All of Fredrik's examples in his guide on Tkinter use
> 
> >from Tkinter import *
> 
> >Now, I'm aware that the effbot is never wrong about anything<wink>,
> >but is it safe to assume that I should not be using the above idiom,
> >but the import Tkinter idiom instead?
> 
> Rest assured, the effbot is right as usual.

-- 
Stand Fast,
    tjg.

Timothy Grant                         tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.         <><       (503) 246-3630
>>>>>>>>>>>>>Linux, because rebooting is *NOT* normal<<<<<<<<<
>>>>This machine was last rebooted:  49 days 21:27 hours ago<<



From grante at visi.com  Fri Mar  2 17:29:47 2001
From: grante at visi.com (Grant Edwards)
Date: Fri, 02 Mar 2001 22:29:47 GMT
Subject: cataloging words in text file
References: <3AA0124E.74A39783@comm.mot.com>
Message-ID: <v9Vn6.396$z6.33544@ruti.visi.com>

In article <3AA0124E.74A39783 at comm.mot.com>, Stephen Boulet wrote:

>I remember this homework assignment for my data structures (c++)
>class: read in a large file, and create a data structure containing
>every word in the file and the number of times it appears.
>
>I was wondering how to do this in python. In c++ we had to do it
>with hash tables and b-trees.
>
>Can you do it with dictionaries in python, with the key as the word
>and the data the number of times it appears?

Sure:

#!/usr/local/bin/python2.1
import sys
d={}
for w in sys.stdin.read().split():
    if d.has_key(w):
        d[w] += 1
    else:
        d[w] = 1
print d    

You may want something a little more sophisticated than
split(), since this solution is case sensitive and treats
"foo" and "foo." as two different words.


-- 
Grant Edwards                   grante             Yow!  .. I don't understand
                                  at               the HUMOR of the THREE
                               visi.com            STOOGES!!


From tim.one at home.com  Sat Mar  3 23:44:19 2001
From: tim.one at home.com (Tim Peters)
Date: Sat, 3 Mar 2001 23:44:19 -0500
Subject: [Python-Dev] Re: Bug fix releases
In-Reply-To: <ILio6.29087$Z8.5197836@typhoon.southeast.rr.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCIEMFJDAA.tim.one@home.com>

[Don Tuttle]
> But there is the CVS. And there are those like myself  who are CVS
> impaired.  Right now there are Fredric's SRE

Which you already have.

> and _tkinter on steroids updates,

AFAICT, PythonWare makes uiToolkit for Tkinter available only to their
PythonWorks Pro customers, so we couldn't distribute that regardless.

> and Mark's improved COM support.

Also outside PythonLabs' cherubic little hands.

> I'd love to have these.

And while Guido may love to give them to you, they're not his to give.  It's
also curious that none of these have to do with bugfixes, they're all
"shinier feature" things -- exactly what Aahz is telling us users *don't*
want between major releases.  Good thing I don't believe anything you read
<wink>.

> (Alright I've got /F's. He compiled 'em.  But Mark's is still in the
> damn CVS.)  I've prodded ActiveState to issue Python updates and
> they've indicated they would--they just haven't decided when.
>
> I don't think it has to be as formal as an official  pep or separate
> development track.  Just whenever our fearless dictator sees that there's
> enough good stuff in the CVS, let him decree that a new build
> shall be made.

None of the stuff above is in our SourceForge CVS tree.  So you've made an
excellent case that there's nothing whatsoever we can do for you <wink>.

If we could, barking a decree is easy but isn't enough to make a release
*happen*.  I put in at least 80 hours during the week of a release, and I'm
only looking out for Windows (+ masses of general patches and bugfixes and
email panics, etc).  The other PythonLabs guys aren't on vacation either.

Somebody has to step up and do *work*.  But, historically, nobody volunteers.
If you look at Misc/HISTORY, you may be surprised to discover that the most
recent "pure bugfix release" was in November of 1994(!  and despite what it
was *called*, it snuck some new Tkinter features anyway).

> Of course we've gotta sell the idea to the dictator.

That's easy -- he's always been in favor of bugfix releases, but provided
somebody else volunteers the work to produce them.

> Nobody's lost their head trying to do that (gulp) have they?

I try not to think too much about that we don't hear much about Jim Hugunin
anywore <wink>.

OTOH, the PSU




From dougfort at downright.com  Thu Mar 22 18:26:27 2001
From: dougfort at downright.com (Doug Fort)
Date: Thu, 22 Mar 2001 18:26:27 -0500
Subject: PyUnit in Python 2.1 standard library
Message-ID: <3ABA8A23.85EA76CE@downright.com>

Just recieved word that Steve Purcell has checked PyUnit  into the
standard library.  Aside from being a useful tool to me PyUnit typifies
the kind of development that ESR advocates in 'The Cathedral and the
Bazaar'.   A project started to 'scratch an itch', then shared with the
communityto evolve and grow.  None of the stuff I  work on seems to come
out so clean.  Congratulations Steve!  PyUnit is an example to us all.

--
Doug Fort (dougfort at downright.com)
Senior Meat Manager
Downright Software LLC
http://www.dougfort.net





From bedge at troikanetworks.com  Wed Mar 14 18:19:52 2001
From: bedge at troikanetworks.com (Bruce Edge)
Date: Wed, 14 Mar 2001 15:19:52 -0800
Subject: Dynamic construction of a print format string
References: <3AAFB663.57FCCE7@troikanetworks.com> <7MOr6.281$4d3.61498@newsb.telia.net>
Message-ID: <3AAFFC98.8322CCB3@troikanetworks.com>

Ahh, %*s, the magic trick.

Thanks to all who responded.

-Bruce.


Fredrik Lundh wrote:
> 
> Bruce Edge wrote:
> > I'm using:
> >
> >   fmtstr = "%s%s%s" % ( "    %", namelen, "s - %s" )
> >
> > to get:
> >
> >   "    %19s - %s"
> 
> how about:
> 
>     format = "    %%%ds - %%s" % namelen
>     text = format % (arg1, arg2)
> 
> or
> 
>     text = "    %*s - %s" % (namelen, arg1, arg2)
> 
> Cheers /F


From winson at mdsserv.mds.com.tw  Thu Mar 29 00:36:50 2001
From: winson at mdsserv.mds.com.tw (=?big5?B?sWm4zq9x?=)
Date: Thu, 29 Mar 2001 13:36:50 +0800
Subject: Python GUI with Swing
Message-ID: <001e01c0b812$43f00760$340210ac@winson>

Hi All,

Is there any good solution for using Java Swing in Python ?? I surveyed Jython, PyJava, JPE and JPI. Jython sounds great but I only needs Swing as my GUI and the whole Jython is based on Java, speed will be a issue. Actually, I would like to find out a module can access Java's Swing package.

PyJava, JPE and JPI only for AWT package, no Swing.

--
Best Regards,

Winson Chang

3F, 2, 150 Lane, Sec. 5, Hsin I Rd., Taipei, Taiwan, R.O.C.

Tel: +886-2-27225333 ext 376
Fax: +886-2-27222330

Email: winson at mdsserv.mds.com.tw
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20010329/4f29b1c8/attachment.html>

From qrczak at knm.org.pl  Thu Mar 22 17:46:14 2001
From: qrczak at knm.org.pl (Marcin 'Qrczak' Kowalczyk)
Date: 22 Mar 2001 22:46:14 GMT
Subject: gah! I hate the new string syntax
References: <mailman.983485923.1136.python-list@python.org> <97sudi01i26@news1.newsguy.com> <98elrk$li8$1@nntp.Stanford.EDU> <98fc070thh@news1.newsguy.com> <98iqo2$56o$1@nntp.Stanford.EDU> <j4kr6.68739$ML1.3282814@e420r-atl2.usenetserver.com> <Pine.LNX.4.21.0103161339210.2809-100000@bast.jhb.equinox.co.za> <98t5bv0l8b@news1.newsguy.com> <Pine.LNX.4.21.0103161647280.2809-100000@bast.jhb.equinox.co.za> <98tet00v03@news1.newsguy.com>
Message-ID: <slrn9bl05l.3q9.qrczak@qrnik.zagroda>

Fri, 16 Mar 2001 17:27:10 +0100, Alex Martelli <aleaxit at yahoo.com> pisze:

> Non-invasive mods would be a better target -- and I think a
> runtime variation of Haskell typeclasses might be a nice way
> to present such mods.  E.g., suppose that, parallel to the
> existing builtin 'isinstance(object, [class or type object])',
> that just returns 0 or 1, we have in some future Python a
> new builtin:
>     getinstance(object, [typeclass object, or etc etc])
> which returns either None (meaning the required 'cast' is
> not feasible) OR a reference x such that the axiom:
>     x = getinstance(y, z)
>         =>
>     x is None or isinstance(x, z)
> is satisfied, and x 'represents y as a z'

I imagine it differently, although it's perhaps shifted too far from
current Python's practice towards Haskell.

Interfaces (corresponding to Haskell's typeclasses) are objects which
contain attributes (usually functions).  These functions examine
types/classes of arguments they are applied to, and dispatch the
implementation according to registered instances of that interface,
or raise an exception if the combination has not been registered.

An interface also offers querying whether a combination of
types/classes is valid.

This is an alternative to methods having magical names. Instead
of __hash__ method there would be a Hasbable interface. The builtin
function hash is bound to Hashable.hash. A class C can be made hashable
by executing the following:

    instance Hashable(C):
        def hash(obj):
            return obj.spam ^ obj.eggs

It smoothly extends to multiple dispatch:

    instance Collide(Ship, Rock):
        def collide(ship, rock):
            ...

-- 
 __("<  Marcin Kowalczyk * qrczak at knm.org.pl http://qrczak.ids.net.pl/
 \__/
  ^^                      SYGNATURA ZAST?PCZA
QRCZAK


From stephen_purcell at yahoo.com  Tue Mar 27 11:30:51 2001
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Tue, 27 Mar 2001 18:30:51 +0200
Subject: Uploading a file via POST ?
In-Reply-To: <99qdlf$a9r$1@news.rrz.Uni-Koeln.DE>; from a2619725@uni-koeln.de on Tue, Mar 27, 2001 at 04:03:59PM +0000
References: <99qdlf$a9r$1@news.rrz.Uni-Koeln.DE>
Message-ID: <20010327183051.D1260@freedom.puma-ag.com>

Oliver Hofmann wrote:
> Is there an easy way to upload a file and pass parameters to the
> server script?

If what you really want to do is programmatically upload a file from a python
script to a CGI on the server, you may be able to use or steal some code from

  http://webunit.sourceforge.net/

It's a side project I'm just starting to hack on, related to web testing;
the code currently there supports multipart form submissions.

-Steve

-- 
Steve Purcell, Pythangelist
Get testing at http://pyunit.sourceforge.net/
Any opinions expressed herein are my own and not necessarily those of Yahoo



From arinagour at yahoo.co.uk  Fri Mar  9 04:59:16 2001
From: arinagour at yahoo.co.uk (chris lamb)
Date: Fri, 9 Mar 2001 09:59:16 -0000
Subject: OODB question
Message-ID: <WM1q6.7817$Z%5.6737@NewsReader>

I am sure that I read somewhere of an OODB implementation developed
specifically with Python or having a Python interface.

I am beginning to wonder if I dreamt it up since I cannot remember where I
read about it!

Can anyone help me ? (I doubt it!)

Best wishes

Chris




From reinhold.koch at computer.org  Fri Mar 16 06:21:41 2001
From: reinhold.koch at computer.org (Reinhold Koch)
Date: Fri, 16 Mar 2001 12:21:41 +0100
Subject: os.spawnv problem
Message-ID: <3AB1F745.865F01A@computer.org>

Hi python experts!

As a newcomer I just hit a wall with the little one-liner:

os.spawnv('P_WAIT',r'C:\rei\bin\PFE32.EXE', ('xx','yy'))

resulting in:

Traceback (innermost last):
  File "<pyshell#21>", line 1, in ?
    os.spawnv('P_WAIT',r'C:\rei\bin\PFE32.EXE', ('xx','yy'))
TypeError: an integer is required

First question: where is which integer required? Trying to debug and
view stack did not yield anything reasonable to me.
Second question: is spawnv generally not (yet) stable?

My environment:
Win2000 SP1
Python 2.0 (#8, Oct 16 2000, 17:27:58) [MSC 32 bit (Intel)] on win32
IDLE 0.6 -- press F1 for help

TIA for any comments

Reinhold Koch
reinhold.koch at computer.org




From aleaxit at yahoo.com  Fri Mar 16 08:53:39 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 16 Mar 2001 14:53:39 +0100
Subject: I come to praise .join, not to bury it...
References: <mailman.983485923.1136.python-list@python.org> <97p7iq$kk2$1@nntp6.u.washington.edu> <97rka002k9@news1.newsguy.com> <3AAF7E47.294617E1@sage.att.com> <98qeco02hn7@news1.newsguy.com> <98srjf$fov$1@tyfon.itea.ntnu.no>
Message-ID: <98t5t40lle@news1.newsguy.com>

"Magnus Lie Hetland" <mlh at idi.ntnu.no> wrote in message
news:98srjf$fov$1 at tyfon.itea.ntnu.no...
    [snip]
> > Good thing (for me) that I eventually looked beyond the
> > pretty face, of course:-).
>
> Why? If you don't care about the pretty face, what are
> the advantages of Python as opposed to, say, Perl?

Transparent, predictable, repeatable, regular behavior,
WITHOUT a horde or three of 'convenient' ad-hoc cases.

> To me it would seem that syntax (sugar or not) is
> the main issue... Not that I've used Perl a lot, but
> it seems very much like Python with a lot of ugly bits
> thrown in...

If you HAD used Perl a lot (and I have) you might well
go beyond this "very surface" impression.  Basically,
Python goes a good way towards meeting the principle
of least astonishment.  A common and delightful experience
of the Python learner -- you learn about something new
but don't know the details yet -- you guess at those
details *and more often than not, you guess right*!

Perl seems to be designed by and for poets (not so far
off the mark...) who subscribe to Giovan Battista
Marino's 1619 manifesto of what poetry is all about:
    E` del poeta il fin la meraviglia,
    chi non sa far stupir vada alla striglia
("The purpose of the poet is astonishment", with a
strong recommendation to those who don't know how to
astonish that a preferable career path for them might
be as stable-lads).


Alex





From tjreedy at udel.edu  Tue Mar 20 12:05:08 2001
From: tjreedy at udel.edu (Terry Reedy)
Date: Tue, 20 Mar 2001 12:05:08 -0500
Subject: Thinking Outside the Box
References: <mailman.985083926.17260.python-list@python.org>
Message-ID: <9982ic$ei$1@news.udel.edu>

...
>
>    http://washingtonpost.com/wp-dyn/articles/A23937-2001Mar18.html
>       (found on /.)

I really like the idea of substituting a War on Bugs for the War on Drugs!

A fun read.  Thanks for posting something I would certainly have missed
otherwise.

 TJR





From doughellmann at bigfoot.com  Thu Mar 15 08:12:55 2001
From: doughellmann at bigfoot.com (Doug Hellmann)
Date: 15 Mar 2001 08:12:55 -0500
Subject: Who's minister of propaganda this week?
In-Reply-To: <3AB0176D.8070903@yahoo.com>
References: 	<B98041329020E007.7DA84CAD9D0562AB.817BBC572F480869@lp.airnews.net>
	<98nu1m$12e@dispatch.concentric.net>
	<660ACCA1E4816F7E.C359CDF4168EF3FB.3B6B05538D619770@lp.airnews.net>
	<3AAFA13D.76E32C3F@divalsim.it> <98oa4o01i4v@news2.newsguy.com> 
	<3AB0176D.8070903@yahoo.com>
Message-ID: <20010315131024.CPTY4364.femail4.sdc1.sfba.home.com@localhost>

On 14 Mar 2001 19:14:21 -0600, Michael Chermside wrote:
> Alex Martelli wrote:
>          ... [snip]...
> 
>   >
>   > So, all the compile-time checking is buying is catching (a small
>   > subset of) the errors that would be caught in testing anyway, a
>   > little bit earlier (thus, a little bit cheaper) -- it's never the
>   > case that one has to write a test which would not be needed at
>   > all if type-checking was static, since "the object has the right
>   > type" is a small subcase of "the object _behaves_ per the specs".
>   >
> I'm really not sure I see it this way. If the method  foo(x) is known to
> take a FancyDateObject
> there are three kinds of errors we could make. One is that foo() is
> written badly so it
> doesn't do what it's supposed to. The unit tests of foo() need to guard
> against this.
> Another possible error is that FancyDateObject isn't written properly.
> The unit tests
> of FancyDateObject need to guard against this. And the third type of
> error is that
> that somewhere where we CALL foo(), we might pass it a DateObject
> instead... or
> even a String... which will cause it to perform wrong. To guard against
> this in a
> dynamically typed language, we have to write unit tests for *every
> single place*
> that we call foo(). Of course, we'd be writing unit tests for those
> functions anyway,
> but we won't be able to assume that foo() works properly, and will need
> extra
> tests to ensure this.

We *can* assume foo() works correctly, because we have already written
unit tests for foo() above. So, when bletch() calls foo() I write tests
of bletch() to verify that *bletch()* works correctly.  If it does not,
it could be because it is using foo() incorrectly, but that is not a bug
in foo().

So we need 3 sets of unit tests ( for foo(), for FancyDateObject, and
for bletch() ) but *unit* tests focus on the unit they are testing.  :-)

Doug





From jmarshal at mathworks.com  Wed Mar 21 08:44:48 2001
From: jmarshal at mathworks.com (Joshua Marshall)
Date: 21 Mar 2001 13:44:48 GMT
Subject: New PEP: The directive statement
References: <mailman.985133305.6706.python-list@python.org>
Message-ID: <99ab8g$8q8$1@news.mathworks.com>

Delaney, Timothy <tdelaney at avaya.com> wrote:
> There are however, two things against this PEP.

> 1. It allows for one incompatible language change (directive as keyword can
> interfere with binding to name "directive" as the first line of a file).

Again (and parser-engineering aside), this does not need to cause any
backward-incompatibility.  The proposed syntax is not ambiguous.


From jepler at inetnebr.com  Mon Mar 26 18:31:09 2001
From: jepler at inetnebr.com (Jeff Epler)
Date: Mon, 26 Mar 2001 23:31:09 GMT
Subject: close(), exceptions and problems
References: <slrn9bm5u5.klj.kuntsi@myntti.helsinki.fi> <99i3j7$lqr$1@news.cybercity.dk> <slrn9bpiej.a63.qrczak@qrnik.zagroda> <99l8f4$1fob$1@news.cybercity.dk> <slrn9buioe.3ta.jepler@potty.housenet> <slrn9bveoq.r0r.kuntsi@myntti.helsinki.fi>
Message-ID: <slrn9bvk9t.42i.jepler@potty.housenet>

On 26 Mar 2001 21:56:43 GMT, Antti Kuntsi
 <kuntsi at cc.helsinki.fi.spam.no> wrote:
>Close, the problem is that the implementation in at least Python 1.5.2
>does not check the return code of the close() and thus does not raise an
>exception, at least not with sockets.

This appears to be true of Python 2.0 as well.

It looks like PySocketSock_{accept,close,dup,makefile,dealloc} and
PySocket_socket all potentially need to be modified to correctly deal with
-EINTR (or other) returns from close(socket).  (The users of the SOCKETCLOSE
macro)

Jeff


From alex at shindich.com  Sun Mar  4 22:17:52 2001
From: alex at shindich.com (Alex Shindich)
Date: Sun, 4 Mar 2001 19:17:52 -0800
Subject: Static class methods
References: <rdzo6.154$gc4.80445@news.pacbell.net> <0bDo6.16804$7Y1.1656672@newsread2.prod.itd.earthlink.net>
Message-ID: <EyDo6.6722$qv3.2848295@nnrp5-w.sbc.net>

That will indeed work. But it's plain ugly.
Thanks for sharing the trick though!
"Chris Tavares" <ctavares at develop.com> wrote in message
news:0bDo6.16804$7Y1.1656672 at newsread2.prod.itd.earthlink.net...
> Try this:
>
> >>> class Foo:
> ...     class FooMethods:
> ...             def foo1(self):
> ...                     print "FooMethods.foo1"
> ...     class_methods = FooMethods()
> ...
> >>> foo = Foo()
> >>> foo.class_methods.foo1()
> FooMethods.foo1
> >>> Foo.class_methods.foo1()
> FooMethods.foo1
> >>>
>
> -Chris
>
> "Alex Shindich" <alex at shindich.com> wrote in message
> news:rdzo6.154$gc4.80445 at news.pacbell.net...
> > Just out of my curiosity, have any of you ladies and gentlemen
experienced
> a
> > need for an equivalent of C++/Java static method?
> > Because I have, but it is impossible add them to Python classes -- all
> class
> > methods are treated as instance methods.
> > This is especially unfortunate given that local functions now work
> properly
> > thanks to PEP 227.
> > I was thinking that a new keyword "unbound" could be used to identify
> > functions that belong to a class but do not expect "self" as their first
> > attribute.
> > From outside the class "unbound" methods would be invoked by specifying
> the
> > class name -- Foo.my_static_method (). From within the class they would
> work
> > similar to local functions that are defined at the class level. The __
> trick
> > should work for the "unbound" methods as well to allow for creation of
> > private static methods.
> >
> > Examples:
> > >>>class Foo:
> >     """This is a sample class..."""
> >
> >     __myPrivateStaticVariable  = None
> >
> >     unbound getMyPrivateStaticVariable ():
> >         """getMyPrivateStaticVariable (obj) -> obj
> >
> >         This method returns the value stored in private static
variable...
> >         """
> >         return Foo.__myPrivateStaticVariable
> >
> >
> >     unbound modifyMyPrivateStaticVariable (obj):
> >         """modifyMyPrivateStaticVariable (obj) -> void
> >
> >         This method assigns obj to the private static variable...
> >         """
> >         Foo.__myPrivateStaticVariable = obj
> >
> > >>> type (Foo.modifyMyPrivateStaticVariable)
> > <type 'unbound method'>
> > >>> type (Foo.getMyPrivateStaticVariable ())
> > <type 'None'>
> > >>> Foo.modifyMyPrivateStaticVariable ("abc")
> > >>> Foo.getMyPrivateStaticVariable ()
> > "abc"
> > >>>
> >
> >
> >
> >
>
>
>




From tim.one at home.com  Fri Mar  2 16:58:38 2001
From: tim.one at home.com (Tim Peters)
Date: Fri, 2 Mar 2001 16:58:38 -0500
Subject: the demise of 'from foo import * and its implications?
In-Reply-To: <XFMail.20010302133625.shaleh@valinux.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCGEIKJDAA.tim.one@home.com>

[Sean 'Shaleh' Perry]
> ## as an example of what happens
> ## and sure, I could try to list every function I want to import,
> ## but man does that get ugly
> ...

What gave you the idea that "import *" is going away?




From grante at visi.com  Fri Mar 30 13:59:33 2001
From: grante at visi.com (Grant Edwards)
Date: Fri, 30 Mar 2001 18:59:33 GMT
Subject: Why not an __assign__ method?
References: <mailman.985969720.18484.python-list@python.org> <EF3x6.114663$Ch.15367147@typhoon.we.rr.com>
Message-ID: <pI4x6.9252$SB2.843637@ruti.visi.com>

In article <EF3x6.114663$Ch.15367147 at typhoon.we.rr.com>, Pete Shinners wrote:
>> The question is: Why not have a __assign__ method, that gets called on
>> assignment? It should be called on the *right side* of the expression (at
>> the left side it does not make sense - I leave the proof as an exercise for
>> the reader <wink>). The return value would be then assigned to the left
>> side of the expression. This is the default behavior:
>
>i'm not quite sure how this would work. assigning variables
>have nothing to do with the type a variable 'currently' refers to.

That's not a problem for his proposal.

>a = None
>a = huge_array1 + huge_array2
>
>the variable could point to anything before being assigned.

Exactly -- his proposal doesn't care what the left-side name
points to.

>the assignment in no way effects what the variable originally
>refers to, so it doesn't seem like whatever that was should get
>a say in what happens to the "new owner"

a = huge_array1 + huge_array2

What he wants (as I understand it) is to evaluate the RHS, call
the __assign__ method on the result of that evaluation, and
bind "a" to whatever the __assign__ method returns.

-- 
Grant Edwards                   grante             Yow!  Yow! I like my new
                                  at               DENTIST...
                               visi.com            


From francois.granger at free.fr  Sat Mar 17 13:40:21 2001
From: francois.granger at free.fr (=?ISO-8859-1?Q?Fran=E7ois_Granger?=)
Date: Sat, 17 Mar 2001 18:40:21 GMT
Subject: Python style questions
References: <98tb91$9t9$1@saltmine.radix.net> <98tdu1$cfk$1@panix2.panix.com> <1eqdo5u.1xxmzfi2kjk0qN%francois.granger@free.fr> <98u24p0a98@news2.newsguy.com> <1eqfb5l.10gfjcu1qi9xrwN%francois.granger@free.fr> <CxNs6.432931$ge4.152635172@news2.rdc2.tx.home.com>
Message-ID: <1eqfi2c.nllfyvqv0srvN%francois.granger@free.fr>

Rainer Deyke <root at rainerdeyke.com> wrote:

> Game(HumanPlayer(), NNPlayer()).run()

Wow greet!
It will do for many evening of coding !

-- 
"Le sol n'oublie rien, il transmet."
- Dani?le


From danielk at aracnet.com  Sun Mar  4 23:54:02 2001
From: danielk at aracnet.com (Daniel Klein)
Date: Sun, 04 Mar 2001 20:54:02 -0800
Subject: Overriding UserList methods
Message-ID: <b566atck0ruu4tae8n1srrkntbmme26gor@4ax.com>

I must be missing something, cos I can't figure out how to override a method
using the UserList module. Here is some code:

import UserList
class MyList(UserList.UserList):
      __def __init__(self):
        UserList.UserList.__init__(self)

    def append(self, object):
        self.append(object)

Understandably, this goes into a recursive frenzy until it hits the stack
limit. :-(

The only way I could get this to work was to give the method another name, ie

    def add(self,object):   ## This works but I would rather use 'append'
        self.append(object)   ## as the method name.

So how can I override the append() method so that I can add my specialized code
to it?

Daniel Klein


From aahz at panix.com  Sat Mar 10 10:18:03 2001
From: aahz at panix.com (Aahz Maruch)
Date: 10 Mar 2001 07:18:03 -0800
Subject: How to efficently build a nested dictionary
References: <3AAA2042.E69A53FC@gmx.de>
Message-ID: <98dgjb$iee$1@panix3.panix.com>

In article <3AAA2042.E69A53FC at gmx.de>,
Carsten Gaebler  <cgaebler at gmx.de> wrote:
>
>I wrote a script for FTP traffic accounting, i.e. for each "top level"
>FTP directory it counts the bytes transferred per day. The traffic is
>stored in a dictionary that has the following structure:
>
>traffic = {dir: {year: {month: {day: 42}}}}

I think Emile's suggestion to use a single compound key is a good one.
If that's not sufficient *and* you can use Python 2.0, take a look at
the dict.setdefault() method.
-- 
                      --- Aahz (Copyright 2001 by aahz at pobox.com)

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

"Perhaps God rewards martyrs, but life seldom does..." --Ulrika O'Brien


From ddb at crystal.uwa.edu.au  Sun Mar 18 23:32:31 2001
From: ddb at crystal.uwa.edu.au (Douglas du Boulay)
Date: Mon, 19 Mar 2001 12:32:31 +0800
Subject: Jython and jpythonc2
Message-ID: <3AB58BDF.56EA463C@crystal.uwa.edu.au>

Hi, I'm  using JPython1.1 and am in the process of trying 
to run the jpythonc2 compiler on some python sources.

The problem I'm having is that some component of the 
JVM/jpython/jpythonc2 compiler seems to be incredibly memory hungry.
Essentially, by the time around 15 source files (~2000 lines of code)
have
been processed, the compiler has consumed around 80 Megabytes of ram. 
At that point on a 128Meg(RAM) Compaq Alpha the process apparently 
gets stashed in the swap space. Thereafter, instead of running with 98% 
of the CPU time it runs at about 2%, with continuous disk thrashing.
At the moment, after 3 hours real time, the process has only 
had 8 minutes of actual CPU time and is using 180 megabytes of virtual
RAM. (There are about 80 source files involved here)
  

My questions are: Has any work been done on the jpythonc2 compiler 
regarding memory use optimization (i.e. would upgrading to 
Jython2.? help), and, are there programming constructs which 
are known to consume vast amounts of memory during a jpythonc2 compile
(i.e. are there coding rules that "power users" know in order to 
avoid such problems)? 

TIA
Doug


From aleaxit at yahoo.com  Thu Mar 29 07:25:09 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 29 Mar 2001 14:25:09 +0200
Subject: C++ and Python
References: <99q99l$g21$1@news.clinet.fi>
Message-ID: <99v9kc028u3@news2.newsguy.com>

"Ahto" <ahto at jussila.as> wrote in message
news:99q99l$g21$1 at news.clinet.fi...
> I'm very new to Python and haven't quite figured out how to use c++ and
> python together on windows...if someone could send me an example program
of
> tutorial webpage where this is explained?!?

One possibility is http://www.boost.org/libs/python/doc/index.html.


Alex





From thomas at xs4all.net  Mon Mar 12 03:27:37 2001
From: thomas at xs4all.net (Thomas Wouters)
Date: Mon, 12 Mar 2001 09:27:37 +0100
Subject: from Tkinter import *
In-Reply-To: <YjzUJnEqAHA.188@typhoon.i3s.net>; from kevin_cazabon@hotmail.nospamplease!.com on Thu, Mar 08, 2001 at 08:34:07PM -0700
References: <mailman.983920449.5230.python-list@python.org> <YjzUJnEqAHA.188@typhoon.i3s.net>
Message-ID: <20010312092737.O404@xs4all.nl>

On Thu, Mar 08, 2001 at 08:34:07PM -0700, Kevin Cazabon wrote:

> The screwey thing about Fred's document using "from Tkinter import *" is
> that there is an "Image" method in Tkinter that totally mucks up the PIL
> Image module if you did an "import Image" prior to this...

> Seeing as both are in Fred's court, it's kinda a bad idea...

FredRIK. Fred and Fredrik are two different people.

-- 
Thomas Wouters <thomas at xs4all.net>

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



From garry at sage.att.com  Mon Mar 26 14:37:21 2001
From: garry at sage.att.com (Garry Hodgson)
Date: Mon, 26 Mar 2001 14:37:21 -0500
Subject: warrior geeks
References: <3ABF72F1.98CB78F@sage.att.com> <3ABFA209.CF27BC20@optushome.com.au>
Message-ID: <3ABF9A71.28C2D8C4@sage.att.com>

Tim Churches wrote:

> I'll use our time zone advantage to jump in early on this thread and
> opine that it is way off-topic. Not that I am opposed to
> marital...whoops...martial arts (although they are not my cup of tea),
> but I am concerned that if this threads continues, sooner or later
> someone will mention the whole "Geeks with Guns" thing that Eric S.
> Raymond promotes (see http://www.tuxedo.org/~esr/geeks-with-guns/ or for
> even worse IMHO see http://www.tuxedo.org/~esr/guns/ ), especially given
> ESR's apparent fondness for Python. That sort of thing gives me a fit of
> the shudders (playing with guns, not ESR's fondness for Python). I would
> have thought that if male hackers were looking for a hobby which
> involves phallic-shaped objects and chemically-propelled projectiles,
> then model rocketry would be just the ticket.
> 
> There, that should put an immediate end to this very off-topic thread...

sounds to me like you just created one.


-- 
Garry Hodgson                   Once in a while
Senior Hacker                   you can get shown the light
Software Innovation Services    in the strangest of places
AT&T Labs                       if you look at it right



From ws-news at gmx.at  Fri Mar  9 05:41:07 2001
From: ws-news at gmx.at (Werner Schiendl)
Date: Fri, 9 Mar 2001 11:41:07 +0100
Subject: learning C
References: <mailman.983972829.23481.python-list@python.org> <983975751.18833@newsmaster-04.atnet.at> <3AA82422.53026D3D@cosc.canterbury.ac.nz> <mailman.984107772.3536.python-list@python.org> <kh%p6.1339$54.1462@www.newsranger.com>
Message-ID: <984134753.143458@newsmaster-04.atnet.at>

ROTFL

This one I like too... :-)

regards
werner

Gregory Jorgensen <gregj at pobox.com> wrote in message
news:kh%p6.1339$54.1462 at www.newsranger.com...
> Stroustrup himself said it best: "C makes it easy to shoot yourself in the
foot;
> C++ makes it harder, but when you do it blows your whole leg off."
>
> http://www.research.att.com/~bs/bs_faq.html#really-say-that
>
>
> In article <mailman.984107772.3536.python-list at python.org>, D-Man says...
> >
> >On Fri, Mar 09, 2001 at 01:30:26PM +1300, Greg Ewing wrote:
> >| Werner Schiendl wrote:
> >| >
> >| > The fine thing is: C++ still HAS the scalpel, so what?
> >|
> >| C++ is a Swiss army knife with both scissors and scalpel,
> >| and a nasty tendency for the scalpel to shoot out and
> >| stick you in the palm when all you were trying to do
> >| was use the scissors...
> >|
> >
> >Nice.  I like this description.
> >
> >yet-not-joining-the-war-ly y'rs  -D
> >
> >
>
> Greg Jorgensen
> Deschooling Society
> Portland, Oregon, USA
> gregj at pobox.com




From andymac at bullseye.apana.org.au  Wed Mar 28 06:19:49 2001
From: andymac at bullseye.apana.org.au (Andrew MacIntyre)
Date: Wed, 28 Mar 2001 21:19:49 +1000 (EST)
Subject: warrior geeks
In-Reply-To: <B43D149A9AB2D411971300B0D03D7E8B0FBFE0@natasha.auslabs.avaya.com>
Message-ID: <Pine.OS2.4.32.0103282117270.1228-100000@central>

On Tue, 27 Mar 2001, Delaney, Timothy wrote:

> Australians in general are not overly keen on guns. Except for certain
> security guards, non-military and non-police *cannot* get a license for a
> handgun (and military and police cannot carry a gun unless on duty).
> Likewise, it is almost impossible to buy ammunition for handguns.

I'm sorry, but the above is complete garbage.  But this isn't the forum
for debating the details.

--
Andrew I MacIntyre                     "These thoughts are mine alone..."
E-mail: andrew.macintyre at aba.gov.au    (work) | Snail: PO Box 370
        andymac at bullseye.apana.org.au  (play) |        Belconnen  ACT  2616
        andymac at pcug.org.au           (play2) |        Australia




From ws-news at gmx.at  Fri Mar  9 05:24:30 2001
From: ws-news at gmx.at (Werner Schiendl)
Date: Fri, 9 Mar 2001 11:24:30 +0100
Subject: "!=" is bad form. Re: sorry....never mind
References: <DZkp6.18811$pF4.3015652@news1.telusplanet.net> <fdlp6.16982$go.2157953@news0.telusplanet.net> <3AA6E308.5B63ECC6@cosc.canterbury.ac.nz> <3AA72253.FB021815@Lugoj.Com> <slrn9aehgr.33d.tim@vegeta.ath.cx> <984047188.296438@newsmaster-04.atnet.at> <slrn9agdkv.9ms.tim@vegeta.ath.cx>
Message-ID: <984133751.749323@newsmaster-04.atnet.at>

I'm happy not to be alone there liking C and C++ for some purposes :-)

Tim Hammerquist <tim at vegeta.ath.cx> wrote in message
news:slrn9agdkv.9ms.tim at vegeta.ath.cx...
> Werner Schiendl <ws-news at gmx.at> wrote:
> > > I hate '<>' because it's too much like VB (and so many other BASIC
> > > variants).
> >
> > I think there are at least the same number of programmers that hate the
> > other way because it looks to much like C.
> > Maybe we should invent some completely new operator, ending up with line
> > noise one day... :-]
>
> I realize I seem to be in the minority in this group when I say I like
> C/C++.  NB: I like them as _languages_, but would prefer NOT to do
> everyday programming in them, as they're just a pain in the ass most of
> the time.
>
> I started out programming in BASIC on the TRS-80 and remember the utter
> confusion upon seeing the '<>' operator ("It's greater than _and_ less
> than?")  I had no Pascal background.  I like C's '!=' because it just
> makes more sense to my logical, math-adept brain.
>
> I have no problem with anyone else using it, and if I saw it in
> someone's code, I would understand it.  I just prefer not to use it.
> And since I've not heard any mention of either inequality operator being
> deprecated, I have no cause to change.  ;)
>
> --
> -Tim Hammerquist <timmy at cpan.org>
>
> I don't have any solution, but I certainly admire the problem.
> -- Ashleigh Brilliant




From colinmeeks at home.com  Mon Mar  5 13:16:03 2001
From: colinmeeks at home.com (Colin Meeks)
Date: Mon, 05 Mar 2001 18:16:03 GMT
Subject: Python & mySQL
Message-ID: <DJQo6.27114$c7.7825529@news3.rdc1.on.home.com>

Does anybody have any pointers for integrating mySQL with Python.  I've been
using Python a while and am looking to develop a web backend using Python
with mySQL.  My site is hosted by a Linux based ISP who has both Python and
mySQL but they don't currently have a link between the two.  I'm using
Windows2000 as my development base, so I want something that is usuable on
both, so it's easier for me to track any problems.

Ideally I want something that doesn't involved recompiling Python if that's
possible, as i want the installation to be as straightforward for my ISP and
of course me :-)

Many thanks in advance

Colin




From aahz at panix.com  Thu Mar 29 00:36:47 2001
From: aahz at panix.com (Aahz Maruch)
Date: 28 Mar 2001 21:36:47 -0800
Subject: Various issues regarding thread
References: <mailman.985116565.4789.python-list@python.org>
Message-ID: <99uhlf$9kl$1@panix6.panix.com>

In article <mailman.985116565.4789.python-list at python.org>,
Gabriel Ambuehl  <gabriel_ambuehl at buz.ch> wrote:
>
>I've been developping a small resource monitoring syste, in Python in
>the last time and got some questions regarding threading.thread. I
>figured it would probably be the easiest way (forking of new processes
>seems very unwise as RAM to useage) to have a dedicated thread for
>every resource that has to be monitored so it would be possible to just
>have them sleep between the work and let Python do the scheduling. Now
>however, I thought it could be cool to have the thing using pickle to
>dump its state before it exits so it could continue its work right
>where it left last time (not yet sure if this makes much sense though
>;-). The only problem there is that pickle doesn't appear to be working
>on lock objects which the thread subclasses inherently got. So is there
>any possibility to serialize a class that has been inherited from
>threading.thread?

I haven't tested this, so I don't know whether pickle can handle lock
objects (though it makes sense to me that they can't).  What might make
sense for you would be to create a "worker" class that isn't directly
thread-aware (but is thread-safe), which you can then pickle, and have
the thread class do all the locking around method calls to the worker
class.

>Another thing I couldn't figure out is how one could stop a thread
>reliably. I first thought one could have a flag that gets checked by
>the run method every now and then but then found this to be to slow as
>it could well be that a thread sleeps for several minutes before it
>notices that the flag had been set and thus should have exited some
>time ago. What am I overlooking?

Have the resource threads wait on a Queue.Queue().  Have one timer
thread that puts work/die tokens on the queue.
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

"Boost the stock market -- fire someone"


From dnagata at creo.com  Fri Mar 23 22:52:40 2001
From: dnagata at creo.com (Dale Nagata)
Date: Sat, 24 Mar 2001 03:52:40 GMT
Subject: Python Script to dial a VPN connection, automate VPN routing table updates.
References: <FCNu6.66711$A6.13912607@news4.rdc1.on.home.com>
Message-ID: <3ABC1A08.247@creo.com>

Warren Postma wrote:
> 
> Sorry for the nearly-off-topic nature of this post, but ....
> 
> I work from home now, and there's something in Windows 2000 that's
> annoyingly manual:
> 
> I have a permanent internet connection via @home cable modem, and I "dial"
> through that to create a VPN connection to the office.  The problems are
> two:
> 
> (1) Windows won't accept or save my userid, password and the domain name, so
> I have to enter it manually.
> 
> (2) After connecting, I have two default gateways in my routing tables
> (ROUTE PRINT) and I need to update it so only the traffic that is destined
> to the office goes through that interface.  The IP addresses I get given are
> different each time, making a batch file insufficient, so voila, a perfect
> use for python.
> 
> So after solving #2 above in Python, I wondered, does anyone know how, in
> Python win32 extensions, one might dial an RAS connection, and if it's
> possible to do so  through the win32 API somewhere or through COM? Then I
> could solve #1 above as well.
> 
> Anyone else think this might make a nice little Python applet if I build in
> some more VPN/RAS stuff and maybe even Gui-fy it a bit in wxPython or
> tkInter.
> 

I have a similar home setup.

The win32ras module didn't have all the functions that I needed,
so I used a combination of shelling out to the GUI dialer and
monitoring the RAS connection state via some calls to RASAPI32.DLL
using the windll module, then I just shell out to the route command
to set up the routing table.

The attached script should work if you have set up a dialup networking
entry
named 'Office VPN', but you can change PHONEBOOK_ENTRY at the top of the
script to whatever you want (heck, you can change the whole script
to whatever you want :-)

Rename the script to vpnadmin.cmd or .bat - our mail system screens out
attachments that look like executable content

works for me


--
Dale Nagata         |  tel : +1 604.451.2700 ext. 2254 (UTC-0800)
Software Developer  |  fax : +1 604.437.9891 
Creo Products Inc.  |  pgr : +1 604.691.8279
Burnaby BC Canada   |  http://www.creo.com/
-------------- next part --------------
@python -x %~f0 %* && pause && goto :EOF
#
# vpnadmin.py
#
# 1. Dial RAS phonebook entry PHONEBOOK_ENTRY. (skip if already connected)
# 2. figure out what IP address was assigned to connection.
# 3. Modify routing table so that office network goes through this gateway address.
#
# This works with my ADSL connection.
#
# W2K: disable default prompting for username/password
# or wait in script while in this state???
#
# also remember to disable default gateway for PPTP connection entry
#

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

# standard Python modules
import os
import sys
import time
import string
import struct
import msvcrt

# Win32 Extension modules
import win32api
import win32con
import win32ras
import winerror

# DLL pseudo-modules, for directly callng raw Win32 APIs that aren't wrapped
import windll
user32 = windll.module( 'user32' )
rasapi32 = windll.module( 'rasapi32' )

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

# name of phonebook entry to dial, in default system phonebook
PHONEBOOK_ENTRY = 'Office VPN'

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

# other constants...
RESERVED = 0

# <ras.h>
RASCS_PAUSED = 0x1000
RASCS_DONE   = 0x2000

RASCS_OpenPort = 0
RASCS_PortOpened = 1
RASCS_ConnectDevice = 2
RASCS_DeviceConnected = 3
RASCS_AllDevicesConnected = 4
RASCS_Authenticate = 5
RASCS_AuthNotify = 6 # sending username and password...
RASCS_AuthRetry = 7
RASCS_AuthCallback = 8
RASCS_AuthChangePassword = 9
RASCS_AuthProject = 10
RASCS_AuthLinkSpeed = 11
RASCS_AuthAck = 12
RASCS_ReAuthenticate = 13
RASCS_Authenticated = 14
RASCS_PrepareForCallback = 15
RASCS_WaitForModemReset = 16
RASCS_WaitForCallback = 17
RASCS_Projected = 18

#if (WINVER >= 0x400)
RASCS_StartAuthentication = 19
RASCS_CallbackComplete = 20
RASCS_LogonNetwork = 21
#endif

RASCS_SubEntryConnected = 22
RASCS_SubEntryDisconnected = 23

RASCS_Interactive = RASCS_PAUSED
RASCS_RetryAuthentication = RASCS_PAUSED + 1
RASCS_CallbackSetByCaller = RASCS_PAUSED + 2
RASCS_PasswordExpired = RASCS_PAUSED + 3

RASCS_Connected = RASCS_DONE
RASCS_Disconnected = RASCS_DONE + 1

RASP_PppIp=0x8021

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

# return string formatted like default exception traceback
def format_traceback():
    import sys
    x = sys.exc_info()
    buf = 'Traceback (innermost last):\n'
    import traceback
    tblist = traceback.extract_tb( x[2] )
    for tb in tblist:
        buf = buf + '  File "%s", line %d, in %s\n    %s\n' % tb
    buf = buf + '%s: %s\n' % ( x[0].__name__, str( x[1] ) )
    return buf

# print traceback to stdout
def print_traceback( buf=None ):
    if not buf:
        buf = format_traceback()
    import sys
    sys.stderr.write( buf )

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

# get HWND of window matching given class name and title
def FindWindow( wndclassname, windowname ):
    hwnd = user32.FindWindow( windll.cstring( wndclassname ), windll.cstring( windowname ) )
    if hwnd == 0:
        errcode = win32api.GetLastError()
        errmsg = win32api.FormatMessage( errcode )
        raise win32api.error( errcode, 'FindWindow', errmsg )
    return hwnd

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

# dial a phonebook entry if it isn't already connected
# run the GUI tool so it looks nicer
# wait for the dialog to close
# check connection status afterwards
def DialPhoneBookEntry( phonebook_entry ):
    isconnected = 0
    conns = win32ras.EnumConnections()
    for conn in conns:
        #print conn
        if conn[1] == phonebook_entry:
            isconnected = 1

    if isconnected:
        print 'Connected to', phonebook_entry
    else:
        print 'Dialing %s . . .' % phonebook_entry
        win32api.WinExec( 'rasphone -d \"%s\"' % phonebook_entry )
        # TODO: handle Cancel within rasphone
        status = RASCS_Disconnected
        while not isconnected:
            win32api.Sleep( 1000 )
            conns = win32ras.EnumConnections()
            for conn in conns:
                if conn[1] == phonebook_entry:
                    hConn = conn[0]
                    status = win32ras.GetConnectStatus( hConn )
                    # intermediate states 5 = RASCS_Authenticate, 14=RASCS_Authenticated
                    if status[0] == RASCS_Authenticate:
                        if status != status[0]:
                            status = status[0]
                            print 'Authenticating...'
                    elif status[0] == RASCS_Authenticated:
                        if status != status[0]:
                            status = status[0]
                            print 'Authenticated.'
                    elif status[0] == RASCS_Connected:
                        print 'Connected.'
                        isconnected = 1
                        break
                    else:
                        print 'status:', status
            else:
                # *** this only works in NT4
                # *** need to figure out equiv for W2K
                winver = win32api.LOWORD( win32api.GetVersion() )
                if winver < 5:
                    try:
                        hwnd = FindWindow( '#32770', 'Connecting to %s...' % phonebook_entry )
                    except win32api.error, err:
                        if err[0] == winerror.ERROR_PROC_NOT_FOUND:
                            print 'Connection cancelled.'
                            time.sleep( 1 )
                            return
        #while not connected
    #else not connected

    return isconnected
#DialPhoneBookEntry

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

# return dotted decimal IP address for RAS connection
def GetRasIpAddress( phonebook_entry ):
    connlist = win32ras.EnumConnections()
    #[(655360, 'Office PPTP', 'VPN', 'RASPPTPM')]
    for conninfo in connlist:
        if conninfo[1] == phonebook_entry:
            break
    else:
        print "RAS connection '%s' not found" % phonebook_entry
        sys.exit(1)

    hrasconn = conninfo[0]
    structsize = struct.calcsize( 'L L 16s' )

    dw = windll.membuf( struct.calcsize( 'i' ) )
    dw.mb.write( struct.pack( 'i', structsize ) )

    buf = windll.membuf( structsize )
    buf.mb.write( struct.pack( 'L', structsize ) ) # MUST set struct member

    # on W2K/laptop, need a bit of a delay???
    maxretries = 10
    count = 0
    while count < maxretries:
        err = rasapi32.RasGetProjectionInfo( hrasconn, RASP_PppIp, buf, dw )
        if err == 0:
            break
        else:
            count = count + 1
            if count < maxretries:
                win32api.Sleep( 1000 )
            else:
                print 'RasGetProjectionInfo error %d' % err
                print 'go look up the error code in <ras.h>'
                sys.exit( 1 )

    dwSize, err, rawipaddr = struct.unpack( 'L L 16s', buf.mb.read() )
    ipaddr, junk = string.split( rawipaddr,'\0', 1 )
    #'192.168.91.4\000_ma'

    print 'IP address:', ipaddr
    return ipaddr
#GetRasIpAddress

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

# update routing table for office subnets given IP address to use as gateway
def SetupOfficeRouting( gateway ):
    os.system( 'route add 192.197.216.0 mask 255.255.252.0 %s' % gateway )
    os.system( 'route add 192.168.0.0 mask 255.255.128.0 %s' % gateway )
    os.system( 'route add 204.174.12.32 mask 255.255.255.240 %s' % gateway )
    os.system( 'route add 204.174.12.48 mask 255.255.255.240 %s' % gateway )
    os.system( 'route add 204.174.12.112 mask 255.255.255.240 %s' % gateway )
    print 'IP routing established for office network using PPTP connection.'

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

# hangup RAS connection, if connected
def HangUpConnection( phonebook_entry ):
    connlist = win32ras.EnumConnections()
    #[(655360, 'Office PPTP', 'VPN', 'RASPPTPM')]
    for conninfo in connlist:
        if conninfo[1] == phonebook_entry:
            hrasconn = conninfo[0]
            win32ras.HangUp( hrasconn )
            break
    else:
        print 'Not connected.'
        #print "RAS connection '%s' not found" % phonebook_entry
        sys.exit(1)

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

# manage Office PPTP connection
def vpnadmin():
    USAGE = 'Usage: vpnadmin { dial | hangup | print }'

    argc = len( sys.argv )
    if argc != 2:
        print USAGE
        sys.exit( 1 )

    cmd = sys.argv[1]
    if cmd == 'dial':
        try:
            isp_entry = os.environ['ISP_PHONEBOOK_ENTRY']
            DialPhoneBookEntry( isp_entry )
            pass
        except KeyError:
            pass
        DialPhoneBookEntry( PHONEBOOK_ENTRY )
        ipaddr = GetRasIpAddress( PHONEBOOK_ENTRY )
        SetupOfficeRouting( ipaddr )
    elif cmd == 'hangup':
        HangUpConnection( PHONEBOOK_ENTRY )
    elif cmd == 'print':
        os.system( 'ipconfig /all' )
        os.system( 'route print' )
    else:
        print 'Unrecognized command:', cmd
        print USAGE
        sys.exit( 1 )

def main():
    try:
        vpnadmin()
    except SystemExit:
        pass
    except:
        print_traceback()
        raw_input( 'Press ENTER to finish . . . ' )

if __name__ == '__main__':
    main()

From ws-news at gmx.at  Wed Mar  7 09:50:38 2001
From: ws-news at gmx.at (Werner Schiendl)
Date: Wed, 7 Mar 2001 15:50:38 +0100
Subject: Python script for deMSWordification?
References: <985h02$1pbm$1@swen.emba.uvm.edu>
Message-ID: <983976916.181372@newsmaster-04.atnet.at>

If you have it on your machine you can use itself with Automation to do the
job.

otherwise

.doc Files are basically OLE Compound Documents that represents an IStorage.
IStorage is an interface that models DCOM structured storage alongside
IStream.
Comapred to a file system IStorage is a directory, and IStream is a file.
There are APIs to cope with that kind of file.
With Visual Studio comes a tool called DocFile Viewer that shows the basic
content of the file.

But I guess you want to do that on non-MS platform.
This sounds to me like big fun, so enjoy it!

hth
werner


<kelvin.chu at uvm.edu> wrote in message
news:985h02$1pbm$1 at swen.emba.uvm.edu...
> Dear fellow Pythoneers;
>
> Numerous colleagues and administration types send memos, colloqium
> announcements and other things around in Micro$oft word format.  Has
anyone
> written a filter in Python that can convert these things into normal text?
>
> If not, is there a basic set of rules for decoding .doc format files
> anywhere that I can look at?
>
> Cheers,
>
> -k
>
> --
>
>   kelvin.chu at uvm.edu      (802) 656-0064
>   http://www.uvm.edu/~kchu/ FAX: (802) 656-0817
>




From danielk at aracnet.com  Mon Mar 12 09:52:24 2001
From: danielk at aracnet.com (Daniel Klein)
Date: Mon, 12 Mar 2001 06:52:24 -0800
Subject: Private class and instance methods?
References: <3AAA73BF.9CDCF3B2@alcyone.com> <B6D000FE.4D14%f8dy@diveintopython.org>
Message-ID: <saopatstmaqi4kuhfg261ehl5qfpkimll6@4ax.com>

I'm trying to get a grasp of the Python paradigm. Could you expound a bit on
the advantages of this over the seemingly simpler getX/setX methodology?

Seeing as there is virtually nothing that is really 'hidden' in Python (which
is a _good_ thing imo), why not simply use the _underscore rules (and good
documentation) to specify privacy?

Thanks,
Daniel Klein

On Sat, 10 Mar 2001 16:04:14 -0500, Mark Pilgrim <f8dy at diveintopython.org>
wrote:
>There is a better way to do this in Python, using the __getattr__ and
>__setattr__ special methods.
>
>class foo:
>    privatekeys = ('x',)
>    
>    def __init__(self, x):
>        self.x = x
>    def __getattr__(self, key):
>        print "getting private value %s" % key
>        return self.__dict__['__%s' % key]
>    def __setattr__(self, key, value):
>        if key in self.__class__.privatekeys:
>            print "setting private value %(key)s to %(value)s" % locals()
>            self.__dict__['__%s' % key] = value
>        else:
>            self.__dict__[key] = value
>
>if __name__=='__main__':
>    f = foo('abc')
>    print "f.x = %s" % f.x
>    f.y = []
>    print "f.y = %s" % f.y
>    f.x = 10
>    print "f.x = %s" % f.x



From ben at .  Wed Mar 21 13:37:21 2001
From: ben at . (Benjamin.Altman)
Date: Wed, 21 Mar 2001 13:37:21 -0500
Subject: learning python...
Message-ID: <3AB8F4E1.184DD9C@noaa.gov>

Hello.  From the perspective of someone who already has programming
experience, would anyone know if O'reillys "Learning Python" book would
be better than say Prentice Hall's "Core Python Programming"?  Or would
it be better to go straight into something like "Programming Python"?

Thanks,
Ben



From edwardt at trillium.com  Mon Mar 26 18:39:27 2001
From: edwardt at trillium.com (edwardt at trillium.com)
Date: 26 Mar 2001 23:39:27 GMT
Subject: how to write a text file search & replace script
Message-ID: <99ojvf$ufj$1@news.netmar.com>

Hi I am trying to write a python script to do the following.
Situation:
There are currently a bunch of files which need the following to be replaced:
1. Function names,
2. the function arguments and
3. some global variables

These are a bunch of private library functions that these files are calling.
Since all these library function names are changed. Some of the function
argument names need to be changed too, and there are some extra function
arguments!!
The files calling thest functions need to be changed too! But too many files,
so too many changes. I want to automate this with a python script.

For example: 
   def main():
       tstPrp = "Damn this thing"
       pixitObj.variant = "Python"
       PrnHdr(sid, a="OLD function name", b="those people should think before
they write the library functions", c="I am frustrated")
shall be replaced by
  def main():
       testPurpose = "Damn this thing"
       guiCb.pixitObj.variant = "Python"
       CtfPrintHeader(sid,a1="OLD function name",b1="those people should think
before they write the library functions",c="I am frustrated", d="Please
help")

Can anyone help? Also, can some kind of data structure store a list of the
function name and the associated function arguments that the script canr ead
from? So that if there are more to be changed, all I need to do just to go in
and edit the entry like a spread sheet?

Thanks 

 -----  Posted via NewsOne.Net: Free (anonymous) Usenet News via the Web  -----
  http://newsone.net/ -- Free reading and anonymous posting to 60,000+ groups
   NewsOne.Net prohibits users from posting spam.  If this or other posts
made through NewsOne.Net violate posting guidelines, email abuse at newsone.net


From not.this at seebelow.org  Thu Mar 15 17:28:54 2001
From: not.this at seebelow.org (Grant Griffin)
Date: 15 Mar 2001 14:28:54 -0800
Subject: inverse of asctime
Message-ID: <98rfn6016rf@drn.newsguy.com>

Hi Gang,

I'm in the market for the inverse of asctime().  I looked in the time module but
couldn't find it.  (Nope, strptime isn't available on my Windows installation.) 
Seems like there oughtta be sucha thing somewhere.

either-the-forest-is-missing-a-tree-or-it's-got-too-many
   -tress-that-i-don't-need-ly y'rs,

=g2

_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation            http://www.iowegian.com



From moshez at zadka.site.co.il  Thu Mar 22 04:02:32 2001
From: moshez at zadka.site.co.il (Moshe Zadka)
Date: Thu, 22 Mar 2001 11:02:32 +0200
Subject: New PEP: The directive statement
In-Reply-To: <20010321215413.B25810@glacier.fnational.com>
References: <20010321215413.B25810@glacier.fnational.com>, <mailman.985085006.24236.python-list@python.org> <slrn9bi7q3.2sa.hzhu@rocket.knowledgetrack.com> <Pine.LNX.4.33.0103212154070.6632-100000@bms>
Message-ID: <E14g0zI-0005o7-00@darjeeling>

On Wed, 21 Mar 2001 21:54:13 -0800, Neil Schemenauer <nas at python.ca> wrote:

> May I suggest reading the PEP:
> 
>     http://python.sourceforge.net/peps/pep-0236.html
> 
> Its not long and reading it will make you better in bed.
 
Like Python2C, I wouldn't expect great miracles: 30% is probably
the most one can expect.

python.ca-addresses-huh?-cool!-ly y'rs, Z.
-- 
"I'll be ex-DPL soon anyway so I'm        |LUKE: Is Perl better than Python?
looking for someplace else to grab power."|YODA: No...no... no. Quicker,
   -- Wichert Akkerman (on debian-private)|      easier, more seductive.
For public key, finger moshez at debian.org  |http://www.{python,debian,gnu}.org



From fig at monitor.net  Tue Mar 20 04:04:20 2001
From: fig at monitor.net (Stephen R. Figgins)
Date: Tue, 20 Mar 2001 01:04:20 -0800
Subject: HELP - beginner to programing/python
References: <Vpxt6.416511$Pm2.6478704@news20.bellglobal.com> <3AB6B6F7.5BF4BBD7@northwestern.edu>
Message-ID: <3AB71D14.DB007A1B@monitor.net>

> As proof of Stephen Figgins genuineness, even though he is writing for the
> O'Reilly web site, he does not recommend any of the O'Reilly books :-)

Thanks, Louis.  I do actually recommend O'Reilly books, just not for
beginning programmers.   I like Learning Python and find it very useful
when I need something wordier than Python Essential Reference.   

My top recommendation for beginning programmers, however, isn't a
particular book or tutorial at all.  The best thing a beginner could do
is find a good mentor.  Books are okay, but good mentors are
invaluable.  It is best if it is someone you know personally, someone
who is doing the kinds of things you want to be able to do.  If you
don't know anyone like that, find people who really want to see you
succeed and are willing to help you.   For that, both this list and the
Python Tutor list are great resources.  Pick any beginning tutorial and
dive in.  Try some things out and when you have questions, ask. 

Stephen


From parkw at better.net  Thu Mar  8 17:35:35 2001
From: parkw at better.net (William Park)
Date: Thu, 8 Mar 2001 17:35:35 -0500
Subject: Database with Python ??
In-Reply-To: <988vmn$t77$05$1@news.t-online.com>; from lux_mundt@t-online.de on Thu, Mar 08, 2001 at 11:13:04PM +0100
References: <988vmn$t77$05$1@news.t-online.com>
Message-ID: <20010308173535.A32695@better.net>

On Thu, Mar 08, 2001 at 11:13:04PM +0100, Thomas Mundt wrote:
> Has Python a own database language

No.

> and do someone know a link to a Tutorial in database-working with
> python?

Browse "special topic" at <www.python.org>.

---William Park, Open Geometry Consulting, Linux/Python, 8 CPUs.



From aleaxit at yahoo.com  Wed Mar 21 10:42:13 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Wed, 21 Mar 2001 16:42:13 +0100
Subject: Yet Another PEP:  Interface Adapter Mechanism __adapt__
References: <mailman.985177412.11344.python-list@python.org>
Message-ID: <99ai4i01t4a@news2.newsguy.com>

"Robin Thomas" <robin.thomas at starmedia.net> wrote in message
news:mailman.985177412.11344.python-list at python.org...
    [snip]
> could assimilate issubclass(). I support that work, but I see that it
gives
> some complexity to the meaning of the return value -- in adapter-like
> cases, you want to return self or an adapter of self; in test-like cases,
> you want to return a boolean. I also think that is very cool.

Yes, but, None/non-None generally works fine in a Boolean context
(just test with 'is None'/'is not None' if need be).  Not that I
think boolean use will be frequent, see later.


> A word that works nicely as an operator name, especially a simple word
that
> novices can grasp quickly, would be ideal as a name for this new feature.
>
> Suggestions:
>
> 1) "like"

I suggest 'as' instead -- if the same will-become-an-operator
considerations apply ("give me X ``as'' an Y" sounds more
natural to me than ``like'' does in this context); 'like'
would strongly suggest the boolean reading, which is less
often appropriate (all that boolean testing does is tell
you that, yes, you COULD use X as an Y if you redid the
same operation you just did but avoided throwing away the
result...?!-).


Alex





From tino at wildenhain.de  Fri Mar  2 04:34:51 2001
From: tino at wildenhain.de (Tino Wildenhain)
Date: Fri, 02 Mar 2001 10:34:51 +0100
Subject: catching a del
References: <j%zn6.322252$f36.11513719@news20.bellglobal.com>
Message-ID: <3A9F693B.BF42BE36@wildenhain.de>

Hi Dan,

Dan Parisien wrote:
> 
> class obj:
>         def __init__(self, name, parent):
>                 parent.children[name] = self
>                 self.parent = parent
>                 self.name = name
> 
>         def onDelete(self): #?!
>                 del self.parent[self.name]
> 
> I want to be able to catch a del of an instance of the above class so I can
> remove it from its parent dictionary. Is there a way? (note it cannot be in
> a destructor because there will always be a reference inside the parent
> dict)

perhaps its __del__() ?

Regards
Tino



From aahz at panix.com  Sat Mar 24 09:47:22 2001
From: aahz at panix.com (Aahz Maruch)
Date: 24 Mar 2001 06:47:22 -0800
Subject: [ANN] HappyDoc 1.4 released
References: <mailman.985440901.18241.clpa-moderators@python.org>
Message-ID: <99ic1q$e1t$1@panix6.panix.com>

In article <mailman.985440901.18241.clpa-moderators at python.org>,
Doug Hellmann  <doughellmann at bigfoot.com> wrote:
>
>Announcing the latest version of HappyDoc, a Python documentation
>extraction tool.

Do not taunt fun HappyDoc.
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

"I won't accept a model of the universe in which free will, omniscient
gods, and atheism are simultaneously true."  -- M


From qrczak at knm.org.pl  Wed Mar 14 17:30:40 2001
From: qrczak at knm.org.pl (Marcin 'Qrczak' Kowalczyk)
Date: 14 Mar 2001 22:30:40 GMT
Subject: Expect -- or popen2 -- or what?
References: <98ojsa$95n$1@tyfon.itea.ntnu.no>
Message-ID: <slrn9avs8g.vuo.qrczak@qrnik.zagroda>

Wed, 14 Mar 2001 21:22:30 +0100, Magnus Lie Hetland <mlh at idi.ntnu.no> pisze:

> I'm trying to communicate with an interactive program which seems
> to be a bit reluctant to flush its output...

import pty

I'm sorry to not give more details.

-- 
 __("<  Marcin Kowalczyk * qrczak at knm.org.pl http://qrczak.ids.net.pl/
 \__/
  ^^                      SYGNATURA ZAST?PCZA
QRCZAK


From LangaK at discoveryhealth.co.za  Fri Mar  2 06:22:40 2001
From: LangaK at discoveryhealth.co.za (Langa Kentane)
Date: Fri, 2 Mar 2001 13:22:40 +0200 
Subject: playing with IP's
Message-ID: <A69BC9CA4B39D411BF2200105A45B45B0B14ECE7@dhexchange.discoveryhealth.co.za>

Greetings.
Here is what I wanna do:
Given a network number and mask ie:
192.168.1.0/26
I want to create a list with all the valid hosts on that range.
Problem is I don't know how to play with binary

Where would I start.. any sample code?
Thanks

_________________________________________________________
Langa Kentane		|	Tel:	[011] 290 3218
Security Administrator	|	Cell:	082 606 1515
[CNA MCSE CCSA CCNA]	|	www.discoveryhealth.co.za
_________________________________________________________



From gerhard.nospam at bigfoot.de  Sat Mar  3 01:07:13 2001
From: gerhard.nospam at bigfoot.de (Gerhard =?iso-8859-1?Q?H=E4ring?=)
Date: Sat, 03 Mar 2001 07:07:13 +0100
Subject: PyRun_SimpleFile - impossible to call it under Windows?
References: <3aa07704.2672581@news.newsguy.com>
Message-ID: <3AA08A11.95BC7745@bigfoot.de>

"Owen F. Ransen" wrote:
> 
> I've seen that there is an odd bug in PyRun_SimpleFile
> probably because of version incompatibilities in the
> FILE structure and using DEBUG modes in VC6
> 
> I'd like to call PyRun_SimpleFile in the debug version
> of my VC6 program, is it impossible currently with Python 2?

I suggest that you use the debug version of the Python libraries,
available at

http://www.activestate.com/Products/ActivePython/Download.html

and elsewhere that I don't remember.

Gerhard
-- 
Sorry for the fake email, please use the real one below to reply.
contact: g e r h a r d @ b i g f o o t . d e
web:     http://highqualdev.com


From ransen_spam_me_not at nemo.it  Mon Mar 12 00:52:29 2001
From: ransen_spam_me_not at nemo.it (Owen F. Ransen)
Date: Mon, 12 Mar 2001 05:52:29 GMT
Subject: PyRun_SimpleFile - impossible to call it under Windows?
References: <3aa07704.2672581@news.newsguy.com> <3AA08A11.95BC7745@bigfoot.de> <3aa0bd3f.708696@news.newsguy.com> <3AAACE47.8000802@ActiveState.com>
Message-ID: <3aab28f3.1012326@news.newsguy.com>

On Sun, 11 Mar 2001 01:04:05 GMT, Mark Hammond <MarkH at ActiveState.com>
wrote:

>Owen F. Ransen wrote:
>
>> I am already linking to Python20_d.lib, is that not the correct
>> version?
>
>I hope I am not too late - but make sure your extension is compiled with 
>"/MD" for release builds and "/MDd" for debug builds.
>
>Mark.

Yes.

Thanks Mark. After help from this newsgroup I've managed to get
things going now. Apart from a few memory leaks at the very end
of the program instance things look pretty good    :)


--
Owen F. Ransen
http://www.ransen.com/
Home of Gliftic & Repligator Image Generators


From tim.one at home.com  Thu Mar  1 17:41:06 2001
From: tim.one at home.com (Tim Peters)
Date: Thu, 1 Mar 2001 17:41:06 -0500
Subject: A counter-proposal to __future__ in PEP 236
In-Reply-To: <97ll2u02qmp@news2.newsguy.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCGECLJDAA.tim.one@home.com>

[Alex Martelli]
> ...
> The _semantics_ of PEP 236 are a great idea; I do earnestly hope there
> is some way, ANY way, to avoid its syntax, and this 'directive' idea
> gives me a glimmer of hope...!

ANY way, you say?  OK, the new syntax is

    from __future__ >> nested_scopes

I thought you'd like that better than

    print __future__ << nested_scopes

although both work for me <wink>.




From jsoons at juilliard.edu  Wed Mar 21 11:32:39 2001
From: jsoons at juilliard.edu (Jonathan Soons)
Date: Wed, 21 Mar 2001 11:32:39 -0500
Subject: os.system output
Message-ID: <B7C74E1A6FBAD411BFC60002B309DDF4E42F@mailbox.juilliard.edu>

I am trying to save the output of

>>>dirs = os.system('ls')

into a variable but I do not need any output to
the screen (altho it doesn't hurt). How do I
go about this?
My fat book only explains how to get output to the screen.

I could do

>>>os.system('ls > file')
>>>f = open('file')
>>>dirs = f.readlines()

but it looks crude. Is there a better way?



From aleaxit at yahoo.com  Tue Mar 20 19:04:43 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Wed, 21 Mar 2001 01:04:43 +0100
Subject: Yet Another PEP:  Query Protocol Interface or __query__
References: <Pine.LNX.4.21.0103201837570.21343-100000@clarkevans.com>
Message-ID: <998r4j02vmb@news2.newsguy.com>

"Clark C. Evans" <cce at clarkevans.com> wrote in message
news:Pine.LNX.4.21.0103201837570.21343-100000 at clarkevans.com...
> Please have a look at the PEP below, it is intended to be a
> light-weight solution to "typing" based on programmer intent.

At least when read at 1 AM -- sounds cool.  Further, the
builtin query function could support _adding interfaces to
an object post-facto, non-invasively_, much like the
'typeclass'-based ideas that were floating in my mind
recently and surfaced in some of my recent posts.  I.e.,
the query() builtin would both try getting the interface
from the _object_, by passing that interface to the obj's
query method, AND also try doing it the other way
'round -- asking the interface if it knows how to wrap
the object so it (the interface) gets implemented.  That
would require interface-identifiers to be richer objects
than just string, or a registry which associates a string
identifying an interface with such a richer object.

I fully agree (as a COM expert) that COM's identity rules
are not appropriate as the semantics of query().  Indeed,
COM defines an alternate approach to getting an
interface providing a service (the IServiceProvider
interface) for the specific purpose of uncoupling the
identity of the object providing the service from that
of the object being queried for it -- it's OK if the same
object does both, but not a requirement.  Particularly if
we accept the idea that the query built-in can query
both the object AND the interface, this decoupling seems
the right semantics for this case.

Since 'interface' is so deucedly overloaded a word, maybe
'protocol' might be be better -- so we only get confused
with FTP, HTTP and the like:-).


Alex





From fredrik at pythonware.com  Wed Mar 21 10:29:36 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 21 Mar 2001 15:29:36 GMT
Subject: Html entities
References: <Sp2u6.3034$mh4.198592@news3.oke.nextra.no>
Message-ID: <AN3u6.892$sk3.291241@newsb.telia.net>

Syver Enstad wrote:
> Is there an easy way to convert ISO Latin-1 characters that are above 127
> ascii to their HTML, XML entity form?

something like this might work:

# htmlentitydefs-example-3.py
# from (the eff-bot guide to) the standard python library

import htmlentitydefs
import re, string

# this pattern matches substrings of reserved and non-ASCII characters
pattern = re.compile(r"[&<>\"\x80-\xff]+")

# create character map
entity_map = {}

for i in range(256):
    entity_map[chr(i)] = "&%d;" % i

for entity, char in htmlentitydefs.entitydefs.items():
    if entity_map.has_key(char):
        entity_map[char] = "&%s;" % entity

def escape_entity(m, get=entity_map.get):
    return string.join(map(get, m.group()), "")

def escape(string):
    return pattern.sub(escape_entity, string)

print escape("<spam&eggs>")
print escape("? i ?a ? e ?")

## prints:
## &lt;spam&amp;eggs&gt;
## &aring; i &aring;a &auml; e &ouml;

Cheers /F

<!-- (the eff-bot guide to) the standard python library:
http://www.pythonware.com/people/fredrik/librarybook.htm
-->




From ixokai-nf at myNOSPAM.org  Wed Mar 28 05:56:08 2001
From: ixokai-nf at myNOSPAM.org (Ixokai)
Date: Wed, 28 Mar 2001 02:56:08 -0800
Subject: types.AnyType - describing interfaces with sequences of types
References: <mailman.985756894.12348.python-list@python.org>
Message-ID: <3ac1c5b3_1@goliath.newsfeeds.com>

Try this:

class _AnyType:
    def __cmp__(self, other):
        return 0
    def __rcmp(self, other):
        return 0

AnyType = _AnyType()

>>> IntType == AnyType
1
>>> type("string") == AnyType
1

You get the idea. AnyType above will be eternally happy to be equivelent to
anything you compare it to. Note that you can't do 'IntType is AnyType',
that tells you actual object identity. Now, if you wanted it to only show up
as 'equivelent' to TypeTypes, and not say, strings, (e.g. "string" ==
AnyType) you could do a check to make sure what its being compared to is a
TypeType.....

Its a hack, yes.. :)

--Ixokai
( Replace 'NOSPAM with 'seraph' to respond)

Carlos Alberto Reis Ribeiro wrote in message ...
>My problem - short question:
>
>I would like to have a type description called AnyType, in such a way that
>it always return true when compared for equality or membership against any
>valid type. Do something like this exist? Does it make sense?
>
>   Example:
>   >>> IntType is AnyType
>   1
>   >>> FloatType == AnyType
>   1
>   >>> "" = AnyType
>   0
>
>Long question:
>
>I was tinkering with some of my proxy experiments, trying a lot of
>different ideas to describe interfaces using only plain Python. One of them
>is the following code snippet:
>
>from types import *
>
>class Interface:
>   """ base class for interface specification """
>   pass
>
>class InterfaceSequence(Interface):
>   """ interface to implement sequence-like objects """
>   __len__      = ((), (IntType,))
>   __getitem__  = ((IntType,), (NoneType,))
>   __setitem__  = ((IntType,NoneType), ())
>   __delitem__  = ((IntType,), ())
>
>The idea here is to define some member attributes inside the
>InterfaceSequence class. A helper function then scans
>InterfaceSequence.__dict__ to build the proxy object. Take this only as an
>example, please - the issue here is not the interface, but the types the
>are available in the the 'types' module.
>
>My problem now is: how to describe the method signature? I'm using the
>following data structure in my example:
>
><method-name>    = (<parameter list>, <return-type>)
><parameter-list> = sequence of type
><return-type>    = type
>
>For example, take the __getitem__ method:
>
>   __getitem__  = ((IntType,), (NoneType,))
>
>This means that __getitem__ have one parameter of IntType (for a sequence,
>remember!), and returns a value of... NoneType (?!?!). This would be much
>better:
>
>   __getitem__  = ((IntType,), (AnyType,))
>
>So that's the problem. I would like to use a value of AnyType for cases
>like this, however, I dont have AnyType as a valid type description. What
>is worse, there is no UserType module to extend type definitions.
>
>
>p.s. I have some approaches being explored for the interface definition.
>This is one of the alternatives. As for the method signature, there are
>some other alternatives that include more information, such as:
>
>a)   __getitem__  = ((('key', IntType), ), (AnyType,))
>b)   __getitem__  = ({'key':IntType}, (AnyType,))
>c)   __getitem__  = ({'key':IntType}, AnyType)
>
>The last one seems to be the cleanest. It limits the return value
>description to asingle type, but this seems to be enough for my intentions.
>However, the problem with 'AnyType' is still there.
>
>
>Carlos Ribeiro
>
>
>




-----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
-----==  Over 80,000 Newsgroups - 16 Different Servers! =-----


From deblNoNoSpammy at theworld.com  Tue Mar  6 12:14:23 2001
From: deblNoNoSpammy at theworld.com (David Lees)
Date: Tue, 06 Mar 2001 17:14:23 GMT
Subject: Trouble running a Grayson Example
References: <3AA3E573.6DBE6D2E@theworld.com> <3aa43d78.258996734@news1.on.sympatico.ca>
Message-ID: <3AA51AF7.336BA869@theworld.com>

Robert,

Thanks, you figured it out.  I made your suggested changes and the
example works fine.

David Lees


Robert Roy wrote:
> 
> On Mon, 05 Mar 2001 19:13:50 GMT, David Lees
> <deblNonOspammy at theworld.com> wrote:
> 
> >I am trying to run example 8.9 (Example_8_9.py) from Graysons book,
> >under Python 2.0 and get the following errors.  Can someone tell me what
> >is going on?  Perhaps this has to do with changes between 1.5.2 and 2.0
> >Python?
> >
> >Thanks in advance.
> >
> >david lees
> >
> >
> >Traceback (most recent call last):  File
> >"d:\Python20\Pythonwin\pywin\framework\scriptutils.py", line 301, in
> >RunScript
> >    exec codeObject in __main__.__dict__
> >  File "E:\Grayson1\Chapter08\Example_8_9.py", line 79, in ?
> >   ddnotebook = DDNotebook()
> >  File "E:\Grayson1\Chapter08\AppShell.py", line 55, in __init__
> >  self.__createInterface()
> >  File "E:\Grayson1\Chapter08\AppShell.py", line 282, in
> >__createInterface
> >   self.createInterface()
> >  File "E:\Grayson1\Chapter08\Example_8_9.py", line 75, in
> >createInterface
> >    self.createNotebook()
> >  File "E:\Grayson1\Chapter08\Example_8_9.py", line 26, in
> >createNotebook  Pmw.NoteBookR,(self.interior(),),)
> >  File "E:\Downloads\pmw\Pmw\Pmw_0_8_4\lib\PmwLoader.py", line 174, in
> >__getattr__
> >AttributeError: NoteBookR
> 
> The examples in the Grayson book were written using  an older version
> of Pmw probably 0_8_1. NoteBookR ceased to exist in 0_8_3 and you are
> using 0_8_4. There were some fundamentat changes to the way Notebook
> works.
> 
> The following code will work.
> 
>     def createNotebook(self):
> # changed Pmw.NoteBookR to Pmw.NoteBook
>         self.notebook = self.createcomponent('notebook', (), None,
>                                          Pmw.NoteBook,
> (self.interior(),),)
>         self.notebook.pack(side=TOP, expand=YES, fill=BOTH, padx=5,
> pady=5)
>         self.formwidth = self.root.winfo_width()
> 
>     def addPage(self, dictionary):
>         table, top, anchor, incr, fields, \
>             title, keylist = dataDict[dictionary]
> # there is no label attribute anymore but a tab_text attribute
> #(optional)
> #        self.notebook.add(table,label=title)
>         self.notebook.add(table, tab_text=title)
>         self.current = 0
>         ypos = top
>         idx = 0
>         for label, field, width, proc, valid, nonblank in fields:
>             pstr = 'Label(self.notebook.page(table),'\
>                    'text="%s").place(relx=%f,rely=%f, anchor=E)\n' % \
>                    (label, (anchor-0.02), ypos)
>             if idx == keylist[0]:
> # self.notebook.page(table).interior() to self.notebook.page(table)
> # did this in 3 places
>                 pstr = '%sself.%s=Entry(self.notebook.page(table),'\
>                        'text="",insertbackground="yellow",
> width=%d+1,'\
>                        'highlightthickness=1)\n' % (pstr,field,width)
>             else:
>                 pstr = '%sself.%s=Entry(self.notebook.page(table),'\
>                        'text="", insertbackground="yellow",'\
>                        'width=%d+1)\n' % (pstr,field,width)
>             pstr = '%sself.%s.place(relx=%f, rely=%f,'\
>                    'anchor=W)\n' % (pstr,field,(anchor+0.02),ypos)
>             exec '%sself.%sV=StringVar()\n'\
>                  'self.%s["textvariable"] = self.%sV' % \
>                                           (pstr,field,field,field)
>         ypos = ypos + incr
>         idx = idx + 1


From andy at reportlab.com  Fri Mar 23 11:57:37 2001
From: andy at reportlab.com (Andy Robinson)
Date: Fri, 23 Mar 2001 16:57:37 +0000
Subject: Reportlab Enterprise Reporting suite - downloads available
Message-ID: <ppvmbt8ekq4q7b1j2jk2ntkharlmsk351u@4ax.com>

ReportLab announces the availability of evaluation beta downloads for
its forthcoming enterprise reporting tools:

RML2PDF:        Translates high level XML document descriptions
                to PDF format.

        ** DOWNLOAD NOW: http://www.reportlab.com/rml **

PageCatcher:    Creates PDF documents which import and modify pages
                and graphics from existing PDF documents.

        ** DOWNLOAD NOW: http://www.reportlab.com/pageCatcher **

 ReportLab's Document Creation Solution
========================================
ReportLab's solution solves several central problems that ebusinesses
face in creating publishing caliber reports that are customized,
produced in real time, in volume, and platform independent. Existing
reporting tools are limited to database reports, are typically
Windows-based,
have problematic restrictions on layout and graphic design, and
go straight to a printer.  More complex publishing systems
involve pipelines of applications which are simply too unwieldy for
real-time use in large scale enviroments

ReportLab's product suite allows direct creation of rich PDF reports
on web or application servers in real time. The tools run on any
platform, can actively acquire data from any source (XML, flat files,
databases, COM/Corba/Java), place no limits on the output, and
facilitate electronic delivery and archival.  The ReportLab suite
lets you define your own business rules to automatically create custom
online reports, catalogs, business forms, and other documents.

Today's businesses must provide concise and timely information to
customers, potential customers, investors, partners, regulators and
other "information consumers," preferably customized to an audience of
one.  In many cases these documents can begenerated automatically, and
the ReportLab toolkit provides the best tool for generating these
documents.  The ReportLab suite can assemble complex documents from a
mix of pre-prepared and automatically constructed text, images,
tables, and graphics.

The integrated ReportLab document generation suite links the
flexibility of XML and HTML with precise control of PDF to deliver
targeted and personalized documents in the right format and with the
right impact.

Components:
===========

The ReportLab suite's three main components are the commercial RML2PDF
and PageCatcher products, together with the Open Source ReportLab core
API.

Our RML2PDF product is a translator which takes a document
specification written in RML and translates the specification into a
PDF document.  RML (Report Markup Language) is an XML dialect which is
as easy to use as HTML but which also provides a high degree of
control for page layouts and graphical presentations.  RML also
includes interfaces to PageCatcher (see below), and plug-in
functionality allowing custom charts and data displays to be added.
Companies can use the tools they know best to generate RML files much
as they would HTML. To create a PDF document, describe the document
using RML, possibly importing whole pages directly from other PDF
documents, and RML2PDF will construct the document for you.  No
programming required!
        ** DOWNLOAD NOW: http://www.reportlab.com/rml **

Our PageCatcher product  enables the reuse of any existing PDF files,
allowing firms to use their favorite design elements and tools.
PageCatcher extracts graphical elements fromPDF files for inclusion in
documents generated by RML2PDF or the ReportLab core API.  Since any
external program with the ability to print can produce PDF files via
Adobe Acrobat, this means that a ReportLab document can include
graphical elements created by virtually any program.
        ** DOWNLOAD NOW: http://www.reportlab.com/pageCatcher **

The ReportLab core API provides programming interfaces for PDF file
generation. Using the API programmers can build plug in components for
RML2PDF, or whole programs for creating PDF documents.
        ** DOWNLOAD NOW: http://www.reportlab.com/download.html **

Example Uses
=============

These are some of the many possible applications for Reportlab
technology:

Electronic Forms and Fulfilment:
--------------------------------
Invoices, purchase orders and custom contracts which would normally be
sent by post can be generated at the point of sale and delivered
immediately.  Application forms can be generated with pre-filled data
for signature by clients, increasing response rates and accuracy. The
same PDF document can be archived on the vendor's system, providing a
complete record; delivered through the web or email; and, if needed,
routed for subsequent printing and posting.

Realtime online securities statement.
-------------------------------------
A securities firm, fund manager, or bank could use the ReportLab
toolkit to build a downloadable customer statement in real time, in
response to a click on a web form. The statement would be
indistinguishable from a personalized investment analysis with charts
and tables reflecting the customer's investment portfolio, and perhaps
some automatic risk/return analysis together with hypothetical
projections and analysis. The attractively laid out statement could
provide several views of the customers portfolio data, performance
histories of the securities held, and various analytical reports such
as portfolio analysis indicating opportunities to rearrange the
holdings according to the customer's stated goals.

Management, scientific and engineering reporting.
-------------------------------------------------
Periodic reports are an essential tool of modern corporate life. These
reports frequently integrate text, tables, and charts, often in a
standard format. An automated script can use the ReportLab toolkit to
construct these reports by querying multiple databases and accounting
systems, acquiring text, and constructing charts to order. Other
programs could automatically send the reports out by electronic mail
or print them on appropriate printers, or otherwise archive and
distribute the reports electronically.

Personalized portal newsletter.
-------------------------------
Many web portals provide personalized spaces associated with
personalized advertising. A portal could use the ReportLab toolkit to
create downloadable personalized documents, and include personalized
advertising targeted to the user's profile and interests. PDF is much
better for framing personalized content and advertizing than HTML
since PDF provides complete control of page layout, rather than the
limited control in HTML. Content and advertising presented in PDF
could attain magazine quality presentation even when printed on a
user's low-priced color printer.



Bottom Line:
============
The modern world is awash in information and data.  To be successful,
the modern business must present the right information to the right
person in the right way. ReportLab's document creation suite is the
next generation in easy, fast, scalable presentation-caliber
reporting.

CONTACT:
        info at reportlab.com
        http://www.reportlab.com/





From john at underdog.nmt.edu  Sat Mar 24 12:16:44 2001
From: john at underdog.nmt.edu (John Shipman)
Date: Sat, 24 Mar 2001 17:16:44 +0000 (UTC)
Subject: New Tkinter and PIL documentation (free)
References: <99en42$2vl6$1@newshost.nmt.edu> <Y_Vu6.179074$Z8.37733056@typhoon.southeast.rr.com>
Message-ID: <99ikps$1qmo$1@newshost.nmt.edu>

First I wrote:

+--
| At this site you will find my newly written reference document
| for Tkinter:
|
|     http://www.nmt.edu/tcc/help/pubs/lang.html
+--

Then Don Tuttle <tuttledon at hotmail.com> replied:

+--
| I downloaded the Tkinter PDF, but apparently I don't have the same
| font you used.  (The text is very fuzzy)  Assuming this is the
| problem, would you consided remaking the PDFs with a more standard
| font? Thanks.
+--

The document was designed for print, not for screen use.  It
should print cleanly.  If you're reading it with Acrobat,
try increasing the magnification.




-- 
John Shipman (john at nmt.edu), Applications Specialist, NM Tech Computer Center,
Speare 128, Socorro, NM 87801, (505) 835-5950, http://www.nmt.edu/~john
  ``Let's go outside and commiserate with nature.''  --Dave Farber


From nas at python.ca  Fri Mar 30 23:39:46 2001
From: nas at python.ca (Neil Schemenauer)
Date: Fri, 30 Mar 2001 20:39:46 -0800
Subject: Is there a bug tracker program written in Python?
In-Reply-To: <20010330.110051.1782436840.1258@scoot.zenplex.com>; from scott@zenplex.com on Fri, Mar 30, 2001 at 11:00:51AM -0500
References: <20010330.110051.1782436840.1258@scoot.zenplex.com>
Message-ID: <20010330203946.A2623@glacier.fnational.com>

scott at zenplex.com wrote:
> I'm looking for a Bug Track system like bugzilla but written in Python.
> Does anybody know of such?  Just learned Python and would would like to
> use a system based on Python if at all possible.

Roundup is cool.  My college, Greg Ward, is also working on a
system called SPLAT! which uses Quixote and bsdbd3.  Its not done
yet though.

  Neil



From qrczak at knm.org.pl  Sat Mar 24 11:22:43 2001
From: qrczak at knm.org.pl (Marcin 'Qrczak' Kowalczyk)
Date: 24 Mar 2001 16:22:43 GMT
Subject: close(), exceptions and problems
References: <slrn9bm5u5.klj.kuntsi@myntti.helsinki.fi> <99i3j7$lqr$1@news.cybercity.dk>
Message-ID: <slrn9bpiej.a63.qrczak@qrnik.zagroda>

Sat, 24 Mar 2001 12:23:03 +0000 (UTC), Erwin S. Andreasen <erwin at andreasen.com> pisze:

> But if you call close(fd) you aren't supposed to do something like:
> 
> int fd, rv;
> while (((rv = close(fd)) < 0 && errno == EINTR))
> 	;
> if (rv < 0)
> 	perror("close");

Linux documentation says otherwise.

File: libc.info,  Node: Opening and Closing Files,  Next: I/O Primitives,  Up: Low-Level I/O

[...]

 - Function: int close (int FILEDES)
     The function `close' closes the file descriptor FILEDES.  Closing
     a file has the following consequences:

        * The file descriptor is deallocated.

        * Any record locks owned by the process on the file are
          unlocked.

        * When all file descriptors associated with a pipe or FIFO have
          been closed, any unread data is discarded.

     This function is a cancellation point in multi-threaded programs.
     This is a problem if the thread allocates some resources (like
     memory, file descriptors, semaphores or whatever) at the time
     `close' is called.  If the thread gets cancelled these resources
     stay allocated until the program ends.  To avoid this, calls to
     `close' should be protected using cancellation handlers.

     The normal return value from `close' is 0; a value of -1 is
     returned in case of failure.  The following `errno' error
     conditions are defined for this function:

    `EBADF'
          The FILEDES argument is not a valid file descriptor.

    `EINTR'
          The `close' call was interrupted by a signal.  *Note
          Interrupted Primitives::.  Here is an example of how to
          handle `EINTR' properly:

               TEMP_FAILURE_RETRY (close (desc));

    `ENOSPC'
    `EIO'
    `EDQUOT'
          When the file is accessed by NFS, these errors from `write'
          can sometimes not be detected until `close'.  *Note I/O
          Primitives::, for details on their meaning.

-- 
 __("<  Marcin Kowalczyk * qrczak at knm.org.pl http://qrczak.ids.net.pl/
 \__/
  ^^                      SYGNATURA ZAST?PCZA
QRCZAK


From kalle at gnupung.net  Thu Mar 29 11:24:18 2001
From: kalle at gnupung.net (Kalle Svensson)
Date: Thu, 29 Mar 2001 18:24:18 +0200
Subject: Printing  "%"
In-Reply-To: <99vm4k$22l$1@ctb-nnrp2.saix.net>; from mix77@usa.net on Thu, Mar 29, 2001 at 05:58:46PM +0200
References: <99vm4k$22l$1@ctb-nnrp2.saix.net>
Message-ID: <20010329182418.A5741@father>

Sez Mix:
> How do you print a percenrage (%)sign in Python?
> I tried :
>   return "select * from fax where id LIKE '%s\%'" %ident

Use double %-sign:
  return "select * from fax where id LIKE '%s%%'" % ident

Peace,
  Kalle
-- 
Email: kalle at gnupung.net     | You can tune a filesystem, but you
Web: http://www.gnupung.net/ | can't tune a fish. -- man tunefs(8)
PGP fingerprint: 0C56 B171 8159 327F 1824 F5DE 74D7 80D7 BF3B B1DD
 [ Not signed due to lossage.  Blame Microsoft Outlook Express. ]



From francois.granger at free.fr  Fri Mar  9 16:59:12 2001
From: francois.granger at free.fr (=?ISO-8859-1?Q?Fran=E7ois_Granger?=)
Date: Fri, 09 Mar 2001 21:59:12 GMT
Subject: gah! I hate the new string syntax
References: <mailman.983485923.1136.python-list@python.org> <B7In6.16281$Qb7.2658902@newsb.telia.net> <97ohh2$polt0$1@ID-77331.news.dfncis.de> <97opsb$nhu$1@nntp6.u.washington.edu> <mailman.983615170.23313.python-list@python.org> <tN5o6.2746$eH3.14913@NewsReader> <Koco6.7518$1D5.244602@e420r-atl1.usenetserver.com> <mailman.983997252.5628.python-list@python.org> <m2vgpkyfjo.fsf@mycroft.actrix.gen.nz> <mailman.984063199.18779.python-list@python.org> <98an7i01tgp@news2.newsguy.com> <mailman.984154996.23254.python-list@python.org>
Message-ID: <1eq0wri.16iujikxa9nloN%francois.granger@free.fr>

D-Man <dsh8290 at rit.edu> wrote:

> So we think, but no one has such records.  Actually, we are all
> "people" and came from the Garden of Eden

Sorry, I can't find it on a map ?

But I have heard of Lucy and some toher beeing our common grand
mother...

-- 
La realit?,  y compris la realit? ?comonique,
est toujours complexe. - Noelle Adam


From sholden at holdenweb.com  Fri Mar  2 22:31:01 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Fri, 2 Mar 2001 22:31:01 -0500
Subject: To the point of defeat.....IDLE
References: <3A9AE62B.5070402@yahoo.com> <slrn99s6tu.8rl.garrett@falcon.cs.colostate.edu> <%ftn6.12742$zL.491362@e420r-atl1.usenetserver.com> <slrn99ubbn.9ul.garrett@eagle.cs.colostate.edu>
Message-ID: <rDZn6.848$ML1.66922@e420r-atl2.usenetserver.com>

"Deon Garrett" <garrett at cs.colostate.edu> wrote in message
news:slrn99ubbn.9ul.garrett at eagle.cs.colostate.edu...
>
> Well, don't take this as the gospel truth -- I haven't actively tried to
> break anything since upgrading my Python to 2.0.  But, it has been a while
> and nothing has failed in my day-to-day use of the machine.  I'm actually
> using Mandrake 7.2, but I use some of the RedHat stuff for administration
> and it all seems to work.  Of course, I don't mind re-installing stuff
from
> scratch if I screw it up.  If you're talking about a mission critical
system,
> it would definitely be worth checking out before you bail right in to the
> upgrade.  Just my two cents...
>
> Deon
>
>
> >
> >Has anyone actually verified that the myriad pieces of RedHat Python
aren't
> >incompatible with 2.x? [...]

Well, actual experience is a good substitute for rumor and guess! THanks for
letting me know: it wouldn;t be critical if a reinstall was required.

regareds
 Steve





From root at rainerdeyke.com  Thu Mar 22 17:18:55 2001
From: root at rainerdeyke.com (Rainer Deyke)
Date: Thu, 22 Mar 2001 22:18:55 GMT
Subject: PEP 240 scares me
References: <Pine.LNX.4.21.0103222338530.29360-100000@kev.ioc.ee>
Message-ID: <jTuu6.473437$ge4.165137318@news2.rdc2.tx.home.com>

"Pearu Peterson" <pearu at cens.ioc.ee> wrote in message
news:Pine.LNX.4.21.0103222338530.29360-100000 at kev.ioc.ee...
> It is true that 1/2 == 0 surprise new (Python) programmers (but I quickly
> learned to write 1./2 or 1/2. or 1/float(n) etc).
> But making 1.2 to be a rational
>                                 1
>                                ---
>                                 2

That would be stupid.  '1.2' should be the rational '6/5'.  To write '1/2',
you would actually write '1/2' (possibly with some operator other than '/').
I don't think anybody was arguing differently.


--
Rainer Deyke (root at rainerdeyke.com)
Shareware computer games           -           http://rainerdeyke.com
"In ihren Reihen zu stehen heisst unter Feinden zu kaempfen" - Abigor




From Jerry.Spicklemire at IFLYATA.COM  Fri Mar 16 09:22:09 2001
From: Jerry.Spicklemire at IFLYATA.COM (Spicklemire, Jerry)
Date: Fri, 16 Mar 2001 09:22:09 -0500
Subject: Complete Docs for Python Programming (was several other things)
Message-ID: <977A39E65CFCD3119ABF00D0B741D849E0FFD6@innt-73.ata.com>

Don Tuttle wrote:

>Hmm... I guess I didn't make myself clear.  I'm not really interested in
>Python apps.  I'm interested in more specific topics on Python Development.
>In short, don't make me learn VB, Perl, or C++ just so I can THEN figure
out
>how to do it in Python.  

Amen. This sums up my exasperation with a wonderful product with great 
potential, that fell off the face of the planet. GFA Basic for Windows 
was a port of the popular GFA product, a staple of european developers. 
It could do "anything", and was so far ahead of VB that it was painful. 
However, the pitch was, "Do MS Windows development without having to 
use C, C++". Sound familiar? Really the same as the VB appeal, except 
GFA Basic actually delivered, at least for some. 

So what was the problem? All the documentation said something to the 
effect of, "See Programming Windows" (a C/C++ book) "for details".

Perfect, all I have to do to use this product, so I won't have to 
use C/C++, is to learn how to do everything in C/C++, and then 
translate it. Let's not leave Python in the same fix! 

There are so many amazing tools out there, but the examples in the 
docs are rarely adequate to tell a newby how to do anything real. 
It's difficult, and time consuming, but next time you are moved to 
send an example in answer to a question, think about whether a newby 
will have to learn three other things before it makes sense. A few 
basic comments, and pointers to further documentation can do wonders.

To all of the very generous Guru class Pythoneers that frequent this 
list, please don't take this the wrong way. I am so grateful for the 
effort and support. It's just that there are lot's of folks out here 
who need Python, and the best way for them to see that is with clear, 
complete instructions.

Sincerely,
Jerry S.





From achim.domma at syynx.de  Tue Mar 20 05:36:09 2001
From: achim.domma at syynx.de (Achim Domma)
Date: Tue, 20 Mar 2001 11:36:09 +0100
Subject: xml,sax,EntityResolver
References: <997bvq$2p7$07$1@news.t-online.com>
Message-ID: <997c1v$soi$03$1@news.t-online.com>

ups ... sorry for the pgp-post !


Hi,

I try to process a sgml-file via sax. The file contains 'undefined
entities' which I don't care about, so would like to ignore them. I
have tried to derive my Handler from EntityResolver and have
resolveEntity overloaded.

Here is some code :

class
MyHandler(handler.ContentHandler,handler.ErrorHandler,handler.EntityRe
solver):
 def __init__(self):
  handler.ContentHandler.__init__(self)

 def startDocument(self):
  print "start"

 def startElement(self,name,attrs):
  print name

 def error(self, exception):
  print exception

 def fatalError(self, exception):
  print exception

 def warning(self, exception):
  print exception

 def resolveEntity(self, publicId, systemId):
  print publicID
  print systemID

parser = make_parser()
h = MyHandler()
parser.setContentHandler(h)
parser.setErrorHandler(h)
parser.setEntityResolver(h)
parser.parse("main.sgm")

If the parser reaches &alpha; in main.sgm fatalError is called twice,
but it seams that resolveEntity is never called - why ? What's the
simplest way to ignore such entities ?

greetings
Achim







From winson at mdsserv.mds.com.tw  Fri Mar 30 02:21:45 2001
From: winson at mdsserv.mds.com.tw (???)
Date: Fri, 30 Mar 2001 15:21:45 +0800
Subject: Python GUI with Swing
References: <001e01c0b812$43f00760$340210ac@winson> <mailman.985889320.22860.python-list@python.org> <3AC3BE89.3F95C8DC@ou.edu>
Message-ID: <001e01c0b8ea$16a14660$340210ac@winson>

Let me explain more about it. I mean my customer need Swing, but I need
Python to build a pilot system. Swing maybe slow but my customer can accept.
If I use Swing thru Jython, it will be slower then directly using Java to
develop the system. So I need a solution to use Swing in Python, that's all.

Is there a module for Python called PySwing, only use Swing's look and feel
GUI ?? It maybe a big issue to rewrite Java's Swing library into C source
and transfer into Python' s module.

Thanks all guys here.

--
Best Regards,

Winson Chang

3F, 2, 150 Lane, Sec. 5, Hsin I Rd., Taipei, Taiwan, R.O.C.

Tel: +886-2-27225333 ext 376
Fax: +886-2-27222330

Email: winson at mdsserv.mds.com.tw
----- Original Message -----
From: "Luke L" <lrl at ou.edu>
Newsgroups: comp.lang.python
To: <python-list at python.org>
Sent: Friday, March 30, 2001 7:00 AM
Subject: Re: Python GUI with Swing


> > If speed is an issue, don't use Swing.  Swing is quite slow (from
> > experience) as it is implemented entirely in Java.  If you only need
> > to support a single platform you can use Jython+Swing+JNI+C using JNI
> > to connect Java to the speed critical stuff written in C.  Otherwise I
> > would recommend looking at other toolkits such as PyGTK and wxPython.
>
> Yeah tell me about it...lightweight components are the wrong way to go....
> Awt would be a step faster, but that's still using Java of course
>
> --
> http://mail.python.org/mailman/listinfo/python-list
>




From max at alcyone.com  Sat Mar 31 02:35:00 2001
From: max at alcyone.com (Erik Max Francis)
Date: Fri, 30 Mar 2001 23:35:00 -0800
Subject: very basic, how do you find square roots?
References: <9a408a$1mpp$1@raewyn.paradise.net.nz>
Message-ID: <3AC588A4.AF93BFAC@alcyone.com>

johnston112 wrote:

> Yes, that is my question :)
> I am trying to make a program that dose the quadratic equation.

Python 2.0 (#5, Nov 10 2000, 21:47:15) 
[GCC egcs-2.91.66 19990314/Linux (egcs-1.1.2 release)] on linux2
Type "copyright", "credits" or "license" for more information.
>>> import math
>>> print math.sqrt.__doc__
sqrt(x)

Return the square root of x.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ Money talks / But love is for real / And you know that
\__/ Neneh Cherry
    Physics reference / http://www.alcyone.com/max/reference/physics/
 A physics reference.


From aleaxit at yahoo.com  Thu Mar  1 10:37:29 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 1 Mar 2001 16:37:29 +0100
Subject: Combinations of n Lists
References: <NI8n6.2979$TW.15640@tor-nn1.netcom.ca> <89bn6.1121$DY.186465@news1.rdc1.az.home.com>
Message-ID: <97lqcm030au@news2.newsguy.com>

"Tim Hochberg" <tim.hochberg at ieee.org> wrote in message
news:89bn6.1121$DY.186465 at news1.rdc1.az.home.com...
    [snip]
> > def combinations(list1,list2):
> >     return [ (i,j) for i in list1 for j in list2 ]
> >
> > print combinations( [1,2,3], ['a','b','c'] )
> >
> > [(1, 'a'), (1, 'b'), (1, 'c'), (2, 'a'), (2, 'b'), (2, 'c'), (3, 'a'),
(3,
    [snip]
> zip([1,2,3], [4,5,6], [7,8,9], .... )

Nope -- this only gives [(1,4,7),(2,5,8),(3,6,9)], *not* all of the
other 24 requested combinations (1,4,8), (1,4,9), ..., (3,6,8).


Alex





From scarblac at pino.selwerd.nl  Mon Mar 19 12:27:27 2001
From: scarblac at pino.selwerd.nl (Remco Gerlich)
Date: 19 Mar 2001 17:27:27 GMT
Subject: Why "from __future__" stinks; a counter-offer
References: <mailman.984975803.9006.python-list@python.org> <jwbnews-3793B2.08394319032001@corp.supernews.com>
Message-ID: <slrn9bcg72.9q8.scarblac@pino.selwerd.nl>

John W. Baxter <jwbnews at scandaroon.com> wrote in comp.lang.python:
> For the future, the time to complain about PEP contents is when the PEP 
> is posted, not after it has been posted, discussed, implemented, and the 
> result distributed.

And the place to complain is the Python-DEV list, which is not public,
although everyone can post to it and everyone can read it using some
archives...

And counterproposals need their own PEP unless it's a minor change in the
proposal.

(both only 'as far as I know')
-- 
Remco Gerlich


From dg at pearl.tao.co.uk  Fri Mar  9 15:28:42 2001
From: dg at pearl.tao.co.uk (David Given)
Date: Fri, 9 Mar 2001 20:28:42 +0000
Subject: PIL & XPM saving?
Message-ID: <qdeb89.j5o.ln@127.0.0.1>

I notice that the PIL XPM plugin doesn't save XPMs. I really need this; is
there anywhere where I get code to do this?

(I'm using Python 1.5, Linux, Debian testing.)

-- 
+- David Given ---------------McQ-+ "I told you to make one longer than
|  Work: dg at tao-group.com         | another, and instead you have made one
|  Play: dgiven at iname.com         | shorter than the other -- the opposite."
+- http://wired.st-and.ac.uk/~dg -+ --- Sir Boyle Roche                        


From sholden at holdenweb.com  Thu Mar  8 05:39:37 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 8 Mar 2001 05:39:37 -0500
Subject: "!=" is bad form. Re: sorry....never mind
References: <DZkp6.18811$pF4.3015652@news1.telusplanet.net> <fdlp6.16982$go.2157953@news0.telusplanet.net> <3AA6E308.5B63ECC6@cosc.canterbury.ac.nz> <3AA72253.FB021815@Lugoj.Com> <slrn9aehgr.33d.tim@vegeta.ath.cx> <984047188.296438@newsmaster-04.atnet.at>
Message-ID: <8nJp6.33937$ML1.1663316@e420r-atl2.usenetserver.com>

"Werner Schiendl" <ws-news at gmx.at> wrote in message
news:984047188.296438 at newsmaster-04.atnet.at...
> > I hate '<>' because it's too much like VB (and so many other BASIC
> > variants).
> >
>
> I think there are at least the same number of programmers that hate the
> other way because it looks to much like C.
> Maybe we should invent some completely new operator, ending up with line
> noise one day... :-]
>
> regards
> werner
>
>
How about >< ? Actually, I have no problem with !=.

regards
 Steve




From scarblac at pino.selwerd.nl  Mon Mar 19 10:15:15 2001
From: scarblac at pino.selwerd.nl (Remco Gerlich)
Date: 19 Mar 2001 15:15:15 GMT
Subject: newbie - infinite loop
References: <985012513.18971.0.nnrp-12.c1c3e154@news.demon.co.uk>
Message-ID: <slrn9bc8et.95q.scarblac@pino.selwerd.nl>

Paul Brian <pbrian at demon.net> wrote in comp.lang.python:
> I have solved the opriginal problem but am left with a serious question.
> 
> Before I discovered the keywords in and not in, I tried to count unique
> instances of names in a file.
> I devcided to create a (neaarly) empty list, compare each name in the file
> with that list. If the name was not in, append to list, otherwise leave it
> alone.

Your code belows does the opposite: if it *is* already in the list, append
it once more. And since you change the list you're currently looping over
(which is a mortal sin, etc) you get an infinite loop.

> After the following was run (and crashed) a couple of times I started using
> my brain and realised there must be some "in" operator. However the problem
> still stands - a for condition seems completely mutable.

It's a list. It's mutable. It's not a "for condition", for just tries to get
new elements from the list until that gives an error, then the loop is
finished. If you append to the list during the loop, the new elements are
also visited. If you add or delete things in the middle of the list,
anything might happen.

Do not taunt happy fun for loops. Do not change lists you are looping over.

> Surely this should not happen.  Surely the line 3 condition "for every
> element in this list", should not constantly expand but should be set solid
> till the next pass?
> 
> What have I done wrong - is it the mutable list v the immutable tuple
> problem (but it should work..surely?) Or is this just the way things are -
> "live with it and do not try anything like this again?"
> 
> thanks in advance
> ==============
> uniqueList = ['q']               # create an nearl empyty list (seems to
> ignore line 3 if truly empty)
> line = 'q'                          # fake the retrieval of line from file
> 
> for listElement in uniqueList:             #3
>     if listElement == line:                    #4
>         uniqueList.append(line)            #5 at this point it seems to
> think that there is now another list
>                                                         #element to compare
> and so loops for ever...help!
> 
> print len(uniqueList)

This idea would work something like this (although the better method is to
use a dictionary, see below):

uniqueList = []   # No words seen yet
line = 'q'        # Or get some line some other way

for listElement in uniqueList:
   if line == listElement:
      break   # Do nothing, we have this word already, stop the for loop
else:
   # The else clause runs after the for loop finishes, if no break was hit
   uniqueList.append(line)
   

But this still has to go through the whole list for every new word, which is
very slow. Use a dictionary, set dict["word"] to 1 for every word, it
doesn't matter if the word was already seen once since it's just
overwritten, at the end you have all the unique words in dict.keys():

uniqueDict = {}

# Do the following for every line, of course
line = 'q'
uniqueDict[line] = 1

# This prints all the unique lines we've seen
print uniqueDict.keys()

-- 
Remco Gerlich


From costas at meezon.com  Fri Mar  2 20:57:20 2001
From: costas at meezon.com (Costas Menico)
Date: Sat, 03 Mar 2001 01:57:20 GMT
Subject: Crashed PythonWin. How can I fix it?
Message-ID: <3aa04cf5.34464140@News.CIS.DFN.DE>

I crashed PythonWin and cant get it backup and running. I was using
Tkinter when this happened.  Does anyone wknow what this Traceback
window means when I start PythinWin? I tried deleting.pyc files but no
luck... And there is no file called glob2.py that I could find.  .

  File "c:\python20\pythonwin\pywin\framework\intpyapp.py", line 163,
in InitInstance
    import interact
  File "c:\python20\pythonwin\pywin\framework\interact.py", line 26,
in ?
    import winout
  File "c:\python20\pythonwin\pywin\framework\winout.py", line 46, in
?
    from pywin.scintilla.view import is_platform_unicode
  File "c:\python20\pythonwin\pywin\scintilla\view.py", line 650, in ?
    LoadConfiguration()
  File "c:\python20\pythonwin\pywin\scintilla\view.py", line 634, in
LoadConfiguration
    from config import ConfigManager
  File "c:\python20\pythonwin\pywin\scintilla\config.py", line 20, in
?
    import glob
  File "glob.py", line 11, in ?
    glob2.what2()
  File "glob2.py", line 3, in what2
    print x
exceptions.NameError: There is no variable named 'x'


From chris.gonnerman at usa.net  Sat Mar 10 10:53:26 2001
From: chris.gonnerman at usa.net (Chris Gonnerman)
Date: Sat, 10 Mar 2001 09:53:26 -0600
Subject: Sucking in a text file
References: <53hkats5d5noic51fskjv0dm72f1rgc29g@4ax.com>
Message-ID: <003b01c0a97a$48271900$a100000a@local>

----- Original Message ----- 
From: "Simon Faulkner" <News at Titanic.co.uk>
Subject: Sucking in a text file


> I have a .py cgi script that writes a page to the web and it works!
> 
> To allow the users to customise the out put I would like the script to
> suck in the first part of the page from a text file and then to spit
> this out to the web server.
> 
> Is this easy or do I need to open the file and read each byte in turn?

hdrf = open("my_header.html", "r") # the "r" is not required actually
header = hdrf.read()
hdrf.close()
sys.stdout.write(header)

Hope this helps.





From barry at digicool.com  Sat Mar  3 03:39:57 2001
From: barry at digicool.com (Barry A. Warsaw)
Date: Sat, 3 Mar 2001 03:39:57 -0500
Subject: ANNOUNCE Mailman 2.0.2 (important privacy patch)
Message-ID: <15008.44509.146939.425219@anthem.wooz.org>

I'm please to announce the 2.0.2 release of Mailman, the GNU Mailing
List Manager.  Mailman is released under the GNU General Public
License (GPL).

Version 2.0.2 is an important bug fix release that also fixes a
potential privacy hole, which could allow a list administrator to get
access to user passwords.  Even with those passwords, I believe
there's little additional harm that a list admin could do, but still
they probably shouldn't have access to those passwords.

There are a few other important fixes in this release, so I recommend
that all sites running Mailman 2.0 or 2.0.1 should upgrade.  Mailman
2.0.2 is being released as both a gzip'd source tarball and as a
patchfile.

Mailman is software to help manage electronic mail discussion lists,
much like Majordomo or Smartmail.  Mailman gives each mailing list a
unique web page and allows users to subscribe, unsubscribe, and change
their account options over the web.  Even the list manager can
administer his or her list entirely via the web.  Mailman has most of
the features that people want in a mailing list management system,
including built-in archiving, mail-to-news gateways, spam filters,
bounce detection, digest delivery, and so on.

Mailman is compatible with most web servers, web browsers, and mail
servers.  It should run on any Unix-like operating system.  Mailman
2.0.2 requires Python 1.5.2 or newer.  To install Mailman from
source, you will need a C compiler.

For more information on Mailman, including links to file downloads,
please see the following mirrors:

    http://www.gnu.org/software/mailman
    http://mailman.sourceforge.net
    http://www.list.org

[Only the SourceForge site is up-to-date at the moment, the others
will be updated soon.]

There are email lists (managed by Mailman, of course!) for both
Mailman users and developers.  See the web sites above for details.

Enjoy,
-Barry

[From the NEWS file]

2.0.2 (03-Mar-2001)

    Security fix:

    - A fix for a potential privacy exploit where a clever list
      administrator could gain access to user passwords.  This doesn't
      allow them to do much more harm to the user then they normally
      could, but they still shouldn't have access to the passwords.

    Bug fixes:

    - In the admindb page, don't complain when approving a
      subscription of someone who's already on the list (SF bug
      #222409 - Thomas Wouters).

      Also, quote for HTML the Subject: text printed for held
      messages, otherwise messages with e.g. "Subject: </table>" could
      royally screw page formatting.

    - In Netscape.py bounce processor, don't bomb out on ill-formed
      messages (no semi-colon separating parameters), otherwise mail
      delivery could grind to a halt.  Bug reported by Kambiz
      Aghaiepour.

    - Docstring fix bin/newlist to remove mention of "immediate"
      argument (Thomas Wouters).

    - Fix for bin/update when PREFIX != VAR_PREFIX (SF bug #229794 --
      Thomas Wouters).



From garry at sage.att.com  Fri Mar 16 16:28:40 2001
From: garry at sage.att.com (Garry Hodgson)
Date: Fri, 16 Mar 2001 21:28:40 GMT
Subject: I come to praise .join, not to bury it...
References: <mailman.983485923.1136.python-list@python.org> <97p7iq$kk2$1@nntp6.u.washington.edu> <97rka002k9@news1.newsguy.com> <3AAF7E47.294617E1@sage.att.com> <98qeco02hn7@news1.newsguy.com> <98srjf$fov$1@tyfon.itea.ntnu.no> <98t5t40lle@news1.newsguy.com> <jwbnews-13D058.07293016032001@corp.supernews.com>
Message-ID: <3AB28588.35EC06FD@sage.att.com>

"John W. Baxter" wrote:

> I came to Python (and dragged my company along) by way of a flooded
> Tower Books store [Bellevue, WA].  They had a [slightly] water damaged
> "Programming Python" on the shelf at something like 60% off as part of
> their flood sale.  [The bottoms of most of the pages are still stained
> and oddly curled.]

you might want to arrange for another flood.  the 2nd edition
is out, effectively a whole new book, and it looks to be
*really* good.

-- 
Garry Hodgson                   Once in a while
Senior Hacker                   you can get shown the light
Software Innovation Services    in the strangest of places
AT&T Labs                       if you look at it right


From bob at staff.cs.usyd.edu.au  Mon Mar 26 06:19:51 2001
From: bob at staff.cs.usyd.edu.au (Bob Kummerfeld)
Date: 26 Mar 2001 21:19:51 +1000
Subject: python 2.0 rpm for redhat 7.0?
Message-ID: <99n8kn$kn@staff.cs.usyd.edu.au>

Can anyone point me to an rpm for Python 2.0 that will install on a 
stock standard redhat 7.0 system?

My daughter asked me to help install python 2.0 on her new system. I
told her it was trivial..... some hours later I'm advising her to
reformat the disk and install Mandrake 7.2!

The BeOpen rpm for py2.0 complains of missing libraries. I decided to
make py2.0 from source (normally a trivial exercise). It complains that
gcc appears to be broken. I install the many update rpms for gcc from
redhat. It's STILL broken....

help :-(

Bob Kummerfeld

-- 



From ejr at cs.berkeley.edu  Fri Mar 16 22:43:57 2001
From: ejr at cs.berkeley.edu (Edward Jason Riedy)
Date: 17 Mar 2001 03:43:57 GMT
Subject: More accurate summations [was Re: This math scares me]
References: <mailman.984726142.2994.python-list@python.org>
Message-ID: <98umht$7e0$1@agate.berkeley.edu>

And <jurgen.defurne at philips.com> writes:
 - 
 - Try summing all financial operations of one day together, or
 - better, all operations involving percentages and then adding them
 - together : cumulative errors.

More concretely, look at the Vancouver Stock Exchange index's 
problems of the late 70s-early 80s.  They had chosen to truncate 
floating point values rather than round them, causing a downward 
drift of around 20 points a month[1].  

You could say that's because of poor handling of floating-point 
arithmetic, and you'd be right.  However, just about every inaccuracy 
in floating-point financial calculations can be attributed to poor 
handling of fp arithmetic.

That doesn't mean floating point is a poor choice for all financial
programs.  Indeed, the HP financial calculators _all_ use floating
point, although decimal.  Those calculators were designed twenty-odd
years ago, and they're still selling for almost as much as a Pilot.

If you want a _very_ accurate sum for all the operations in one day, 
use the following algorithm.  It is often attributed to Kahan, but, 
if I recall his comments correctly, he's not entirely sure of the 
origins of the embodied idea.  Other names around this style of
algorithm are Moller (with a backwards slash through the o), Dekker,
Pichat, Knuth, Bohlender, and Linnainmaa in no particular order.
That's a surprising number of people for just adding numbers.  ;)

class DistilledSum:
  "Keep an accurate running sum."

  def __init__ (self):
    self.s = 0.0
    self.c = 0.0

  def add (self, x):
    "Incorporate x into the running sum."
    y = x - self.c
    t = self.s + y
    self.c = (t - self.s) - y
    self.s = t

  def value (self):
    "Return the sum itself."
    return self.s

  def asTuple (self):
    "Return both the sum and the error term."
    return (self.s, self.c)

The accumulated error in s from floating-point calculations is at 
most one unit in the last place of each summand x, and the value of
c gives you a tighter bound on that error.  With not entirely 
horrible amounts of extra work, you can have however much precision 
you need[2].  Unless your numbers are horribly scaled (perhaps the
debt and GNPs of the entire globe along with a person's), this in 
double plus rounding the stored answer to float works wonders for 
financial accumulations.

You can find an explanation of how this works either in [2] or [3].  
Essentially, the self.c term serves as a small correction.  The
trick with (t - self.s) - y first zeros the high-order bits that
are shared between the old and new sum, then zeros the bits shared
with the pre-corrected x, giving a new correction.  

Jason, unsure of how to create other languages' characters from
this particular keyboard...
  
[1] Q20 at http://catless.ncl.ac.uk/Risks/3.41.html#subj4.1
[2] http://citeseer.nj.nec.com/priest92propertie.html
[3] http://www.validlab.com/goldberg/paper.ps
-- 


From Eugene.Leitl at lrz.uni-muenchen.de  Fri Mar 23 08:33:22 2001
From: Eugene.Leitl at lrz.uni-muenchen.de (Eugene Leitl)
Date: Fri, 23 Mar 2001 14:33:22 +0100 (MET)
Subject: Python MPI
Message-ID: <Pine.GSO.4.03.10103231431120.132-100000@sun1.lrz-muenchen.de>

I've just came back from a 3 day MPI course, and promptly noticed that
I can use Python with MPI (yay!):

	http://www.fysik.dtu.dk/~schiotz/comp/PythonAndSwig/pythonMPI.html

Any MPI Python users here? I'll be interested in hearing your experiences.




From jafo at tummy.com  Thu Mar  1 22:30:06 2001
From: jafo at tummy.com (Sean Reifschneider)
Date: Thu, 1 Mar 2001 20:30:06 -0700
Subject: python irc channel?
In-Reply-To: <97n15e$9g3$1@nntp1.ba.best.com>; from mjbjr@best.com on Fri, Mar 02, 2001 at 02:39:42AM +0000
References: <97n15e$9g3$1@nntp1.ba.best.com>
Message-ID: <20010301203006.A1619@tummy.com>

On Fri, Mar 02, 2001 at 02:39:42AM +0000, Martin Brown wrote:
>So, is there a preferred irc channel where python people hangout?

I believe that the EFNet #python is gatewayed to the IRC server at
irc.openprojects.net #python.  I haven't been on EFNet in quite a
while.  It was more a social channel that occasionally broke out
into code, but usually was talking about computers and cars and
motorcycles.  ;-)

Good group of people.

Sean
-- 
 "Action...  Romance...  Welding...  That movie has everything."
                 -- Evelyn Mitchell speaking about _Subway_, 1998
Sean Reifschneider, Inimitably Superfluous <jafo at tummy.com>
tummy.com - Linux Consulting since 1995. Qmail, KRUD, Firewalls, Python



From mcherm_python at yahoo.com  Sat Mar 17 20:14:11 2001
From: mcherm_python at yahoo.com (Michael Chermside)
Date: Sat, 17 Mar 2001 19:14:11 -0600
Subject: Who's minister of propaganda this week?
References: <B98041329020E007.7DA84CAD9D0562AB.817BBC572F480869@lp.airnews.net> <98nu1m$12e@dispatch.concentric.net> <660ACCA1E4816F7E.C359CDF4168EF3FB.3B6B05538D619770@lp.airnews.net> <3AAFA13D.76E32C3F@divalsim.it> <98oa4o01i4v@news2.newsguy.com> <3AB0176D.8070903@yahoo.com> <3AB0C70A.C6A55DF5@ActiveState.com>
Message-ID: <3AB40BE3.3080001@yahoo.com>

Paul Prescod wrote:

> Michael Chermside wrote:
> 
>> ...
>> I'm really not sure I see it this way. If the method  foo(x) is known to
>> take a FancyDateObject there are three kinds of errors we could make. 
>> One is that foo() is written badly so it doesn't do what it's supposed 
>> to. The unit tests of foo() need to guard against this.
>> Another possible error is that FancyDateObject isn't written properly.
>> The unit tests of FancyDateObject need to guard against this. And 
>> the third type of error is that that somewhere where we CALL foo(), 
>> we might pass it a DateObject instead... or even a String... which 
>> will cause it to perform wrong. To guard against this in a
>> dynamically typed language, we have to write unit tests for *every
>> single place* that we call foo(). 
> 
> 
> Your analysis is incomplete in two ways:
> 
> 1. I don't think it makes sense to speak of writing a unit test for "one
> thing calling another". You test the first thing, you test the second
> thing and if the second thing depends upon the first then you've
> implicitly tested the connection between them.

If foo() is supposed to be passed a FancyDateObject, then the unit
tests for foo() will only pass it FancyDateObject's... it's basically
a precondition on foo(). If bar() calls foo(), then SOMEWHERE we
need to ensure that the value being passed to foo() is, indeed, a
FancyDateObject. (Or, in a dynamically typed language, an object
which behaves as a FancyDateObject is expected to behave.) Failure
to meet this condition might result in subtle behavior (eg, foo()
counts on the special rounding behavior in FancyDateObject, and
when passed a plain DateObject instead, certain invariants in foo()
get broken). It's possible to write unit tests for bar() which
test this, but that requires the author of unit tests for bar() to
know something about foo(). And bar2() and bar3() and bar4()... all
of which also call foo(). In this senario it's particularly convenient
to express the precondition "argument to foo must have special
rounding behavior" by declaring foo(FancyDateObject: x).

> 2. In object oriented programming languages, we can't just test one
> function against one input object class and be confident that the code
> is correct. There are an infinite number of potential subclasses of
> FancyDateObject. Therefore we should NOT depend on testing every single
> one of them with foo. We just need to make sure that foo is correct,
> that each FancyDateObject is correct and that the communication protocol
> between them is well-understood and tested by both.

Here I agree. Test FancyDateObject in its unit tests. When writing
foo(), test that it works when passed a FancyDateObject. When you
write FancyDateObjectWithBellsOnIt, test that it fulfills the contract
of FancyDateObject. Now you know that foo(x) works when x is a
FancyDateObjectWithBellsOnIt. But you could STILL mistakenly call 
foo(y) where y is NOT a DateObject... that's my case 3 above.

> 3. Statically typed languages are absolutely not immune to type errors
> for a variety of reasons:
> 
>  * Null pointer exceptions are type errors. You've passed the null
> object when the code expected some other object. 

Excellent point!

>  * Almost every statically typed language also has a set of features for
> working around the type system (dynamic casts). 

This I don't feel bad about. If a language protects you against a class 
of errors EXCEPT when the programmer specially asks to turn OFF the
protection, then the programmer cannot complain of the language that
it fails to protect you when you turn it off. If you want protection,
don't turn it off (ie, don't dynamic cast except in cases where you
don't MIND finding out at runtime).

>  * Even when the compiler thinks the "type" of your object is correct,
> there are a variety of type-related errors you can make such as passing
> a negative number where only positive ones are allowed, passing an
> gopher: URL where only an http: URL is allowed, passing an absolute
> pathname where only a relative pathname is allowed and so forth. If you
> try to catch all of these in the static type checking system you wlil
> spend more of your life doing conversions than actually coding. *PLUS*
> you will need to do the runtime conversion-compatibility checks
> YOURSELF!

True. Your point here is unassailable... static type checking will only
catch a subset of all possible type errors. Excellent examples!

> 4. Yes, if you unit test every unit in your program, and do good
> coverage testing you WILL catch ever occurrence of some code calling
> foo. That's what good testing is about. And you need it, not to catch
> the basic type errors but to catch the subtle semantic errors. Catching
> the basic type errors comes for free.

Yes, I can see that. But sometimes it's awefully nice to have the
compiler do it for you automatically. It certainly doesn't hurt. 
(But see Alex's response, where he argues that perhaps it DOES hurt,
for other reasons.)

-- Michael Chermside


_________________________________________________________
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com




From erno-news at erno.iki.fi  Sat Mar  3 17:11:55 2001
From: erno-news at erno.iki.fi (Erno Kuusela)
Date: 04 Mar 2001 00:11:55 +0200
Subject: Bug fix releases
References: <LNBBLJKPBEHFEDALKOLCMEEPJBAA.tim.one@home.com> <mailman.982897324.9109.python-list@python.org> <200103031821.NAA24060@panix3.panix.com> <mailman.983646726.27322.python-list@python.org>
Message-ID: <kur90elcyc.fsf@lasipalatsi.fi>

In article <mailman.983646726.27322.python-list at python.org>, Guido van
Rossum <guido at digicool.com> writes:

| But I understand that you're saying that the community has grown so
| conservative that it can't stand new features even if they *are* fully
| backwards compatible.

| I wonder, does that extend to new library modules?  Is there also
| resistance against the growth there?  I don't think so -- if anything,
| people are clamoring for more stuff to become standard (while at the
| same time I feel some pressure to cut dead wood, like the old SGI
| multimedia modules).

it is good if the library is big and plentiful and i don't even
mind if it's bloated. the core language on the other hand
should be simple, and it should not have spurious features.

  -- erno


From loewis at informatik.hu-berlin.de  Sat Mar  3 04:47:48 2001
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 03 Mar 2001 10:47:48 +0100
Subject: how to get the last interactive command enetered
References: <F2Tn6.15639$UZ4.4310299@news4.rdc1.on.home.com>
Message-ID: <j4bsrjxjy3.fsf@informatik.hu-berlin.de>

"Tom" <NoSpam at NoSpam.com> writes:

> For example, say that I set a trace function.  I find the frame's code
> object, look at co_filename and find that it is <stdin>.  But I want to know
> what the actual command was.  Is there anyway to do this (in C or Python)?

No. Once the code is executed, the source code is dropped. All you can
get at is the binary code.

Of course, "in C", you can rewrite the Python interpreter to behave
differently. It takes more than an extension module to do so, though.

Regards,
Martin



From n8gray at caltech.edu.is.my.e-mail.address  Thu Mar 22 22:30:49 2001
From: n8gray at caltech.edu.is.my.e-mail.address (Nathaniel Gray)
Date: Thu, 22 Mar 2001 19:30:49 -0800
Subject: Good Editor?
References: <3ab9a32b$1_1@news2.one.net>
Message-ID: <99egd0$mse@gap.cco.caltech.edu>

fjs wrote:

> Can someone recomend a good editor to write python in?  Im just getting
> started learning python and dont know which ones support python.
> 
> thanks

On *NIX check out NEdit (http://www.nedit.org).  It's got syntax 
highlighting support for a ton of languages, a simple and powerful 
scripting language, rock-solid stability, and it's light on resources.  
Plus, you don't have to time-warp backwards 30 years to understand the 
keyboard accellerators (unless you already think C-x C-f for "open file" is 
intuitive.  I'd bash the vi command too if I'd ever bothered to learn what 
it was...).

-n8

-- 
_.~'^`~._.~'^`~._.~'^`~._.~'^`~._.~'^`~._
             Nathaniel Gray
   California Institute of Technology
     Computation and Neural Systems
     n8gray <at> caltech <dot> edu
_.~'^`~._.~'^`~._.~'^`~._.~'^`~._.~'^`~._



From roy at panix.com  Sat Mar 17 11:39:54 2001
From: roy at panix.com (Roy Smith)
Date: Sat, 17 Mar 2001 11:39:54 -0500
Subject: Thoughts about extensions to the Python language
References: <mailman.984608060.6280.python-list@python.org> <3AB2FA0F.E61862B0@seebelow.org> <Pine.LNX.4.21.0103170930420.26659-100000@kev.ioc.ee> <3AB38DE8.788C86D6@earthlink.net>
Message-ID: <roy-49798F.11395417032001@news.panix.com>

Mitch Chapman <mitchchapman at earthlink.net> wrote:
> I think he meant "write Python code with the mindset of a
> Fortran programmer."  This would entail lots of global variables, 
> humongous function bodies and modules, etc.

Gee, sounds like the last perl program I worked on :-)


From bbollenbach at homenospam.com  Thu Mar 22 18:35:01 2001
From: bbollenbach at homenospam.com (Brad Bollenbach)
Date: Thu, 22 Mar 2001 23:35:01 GMT
Subject: Good Editor? 
References: <3ab9a32b$1_1@news2.one.net> <5Xju6.601501$U46.18300887@news1.sttls1.wa.home.com>
Message-ID: <F_vu6.13509$rB2.1250767@news1.rdc1.mb.home.com>

[snip]
> Most of my Python programs are so small that IDLE works for me. My
favorite
> editors are:
>
> Windows: EditPlus (www.editplus.com)
[snip]

I'll definitely second EditPlus when working on Win32. It's syntax
highlighting isn't nearly as nice as vim's (but this may have been due to
the fact that the syntax highlighting config files just weren't well written
enough, I haven't take the time to verify). One thing to be weary of though:
watch out for autocomplete config files that use tabs instead of spaces in
the config syntax. You'll especially love Editplus if you happen to also be
needing a text editor that's good for writing HTML.

I also think PythonWin makes a nice editor.

On Unix, I'd be using vim.




From sverker.is at home.se  Thu Mar  1 07:04:30 2001
From: sverker.is at home.se (Sverker Nilsson)
Date: Thu, 01 Mar 2001 13:04:30 +0100
Subject: Allow anything in identifiers?
References: <3A9DC13A.1647EE51@home.se> <mailman.983420883.30039.python-list@python.org>
Message-ID: <3A9E3ACE.60F8C639@home.se>

David Porter wrote:
> 
> * Sverker Nilsson <sverker.is at home.se>:
> > This would be a new kind of identifiers, I suppose one
> > can make them by writing for example id <string> like in:
> >
> > id 'a>b' = a > b
> > id 'x.b' = x.b
> > alu.id'+' = lambda x, y...
> >
> > I think it could make programs easier to read by allowing
> > more evident identifiers in many cases. The extended identifier
> > syntax should not break anything, and it seems it might be easy to
> > implement...
> 
> It would break any code using the name 'id', likely a lot of code.

How? Not id <string>, should likely be on the same line, may allow
space in between. That gives a syntax error currently, except when id
is some of a few special ones like r, u, etc. Adding id (or i?) to
that set of string prefixes shouldn't break anything, if I understand
it correctly and I'm pretty sure... to prove your statement you just
need to come up with an example that would be broken so I'll not try
to refute it (more) right now.

Well thanks for the comment and giving me the opportunity to come up
with some examples of where this could be useful:

* Equivalents to the operator module would need (less) new names
  invented.  eg. operator.id'+', operator.id'<' etc instead of
  operator.__add__, operator.__lt__ (though I am just guessing about
  the latter one and I  am sure it's wrong - that's a point of the
  argument :-)

* Same holds for the corresponding class methods __add__ etc.

* Command interpreters could dispatch to functions that had obvious
  names corresponding to the actual commands.

* Variables corresponding to files with fixed name, could have
  the obvious name.

* Temporary constants could be named after their initializing
   expression, i.e. if a variable is introduced to avoid
   recalculation of a[i] in an inner loop it could be clearer if
   it was called precisely that: id'a[i]'. (Note that the extended
   identifier shouldn't incur any extra overhead in the interpreter.)

* Variables could be named in any language, with national specific
  characters in Unicode for example.

One drawback I can think of myself right now is a kind of uglyness
by using quotes but at least they make the construct stand out so it
should not be so confusing to ordinary id's. I guess there are
alternative syntaxes but I couldn't think of any better myself yet.

Sverker


From ben.hutchings at roundpoint.com  Tue Mar  6 17:34:27 2001
From: ben.hutchings at roundpoint.com (Ben Hutchings)
Date: 06 Mar 2001 14:34:27 -0800
Subject: package/namespace confusion
References: <3AA3EDD2.60393490@troikanetworks.com>
Message-ID: <usnkqle6k.fsf@roundpoint.com>

Bruce Edge <bedge at troikanetworks.com> writes:

> Hi,
> I'm sure this is a real rookie issue here, but I can't seem to define a
> global then access it in a package.
> I'm using __init__.py to define glob_var, then load util.py.
> 
> I would have expected util/ufunc() to have access to glob_var.
>
> test/__init__.py:
> 
> glob_var=0        <- is this where I should stick package globals?

Yes.

> __all__=["util"]
> 
> 
> test/util.py:
> 
> def ufunc():
> 	global glob_var
<snip>

The global namespace for this function is the namespace of the module
(test.util) not that of the package (test).

If you want package globals to be module globals too, then use
'from test import *' at the top of your modules.  Alternatively you
can 'import test' and refer to 'test.glob_var'.

-- 
Any opinions expressed are my own and not necessarily those of Roundpoint.


From syver at NOSPAMcyberwatcher.com  Wed Mar  7 06:40:58 2001
From: syver at NOSPAMcyberwatcher.com (Syver Enstad)
Date: Wed, 7 Mar 2001 12:40:58 +0100
Subject: How to Xml escaping of characters above 7 bit ascii.
Message-ID: <N6pp6.711$tR1.9978@news1.oke.nextra.no>

Hi,

Is there a method in Python that takes a unicode string and escapes it
according to the xml file format? Or will I have to do it myself?




From howardbgolden at earthlink.net  Fri Mar 16 17:08:14 2001
From: howardbgolden at earthlink.net (Howard B. Golden)
Date: Fri, 16 Mar 2001 22:08:14 GMT
Subject: Using INRIA's Thot toolkit from Python?
Message-ID: <i9ws6.2114$Im6.295124@newsread1.prod.itd.earthlink.net>

Has anyone used INRIA's Thot toolkit from Python?  (Thot is the basis for
the W3C Amaya editor.)  Is a Python wrapper available?

I would appreciate any information or suggestions.  Thanks in advance.




From paulp at ActiveState.com  Sun Mar 11 14:29:42 2001
From: paulp at ActiveState.com (Paul Prescod)
Date: Sun, 11 Mar 2001 11:29:42 -0800
Subject: any interest in type-scoped static constants?
References: <3AAB9578.4C8754E4@fxtech.com>
Message-ID: <3AABD226.25D30254@ActiveState.com>

Paul Miller wrote:
> 
> ...
> 
> If I create an extension type for Point, I can't do this:
> 
>         p = Point.ZERO
> 
> But I'd like to.

Yes you can! 

[out of order]
> .... 
> My solution would be to add a dictionary to the extension type for
> constants, that the extension code can initialize, and then add a
> scoping test for Type.ATTR to add a check in the Type's attribute
> dictionary.

Right, you can do this in any extension type.

> I'm seriously considering working on a patch to add type-specific
> constants to provide OO access to extension type constants, rather than
> putting all constants in the module scope. I wonder if anyone has
> started this, and if this is considered a good idea. Do I need to write
> a PEP?

Why do you need a PEP when you can implement it for any extension you
need. Just put a dictionary in the type and using type.getattr to look
stuff up in your dictionary.

-- 
Python:
    Programming the way
    Guido
    indented it.
       - (originated with Skip Montanaro?)



From stephen_purcell at yahoo.com  Fri Mar 16 06:41:00 2001
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Fri, 16 Mar 2001 12:41:00 +0100
Subject: os.spawnv problem
In-Reply-To: <3AB1F745.865F01A@computer.org>; from reinhold.koch@computer.org on Fri, Mar 16, 2001 at 12:21:41PM +0100
References: <3AB1F745.865F01A@computer.org>
Message-ID: <20010316124100.A2303@freedom.puma-ag.com>

Reinhold Koch wrote:
> Hi python experts!
> 
> As a newcomer I just hit a wall with the little one-liner:
> 
> os.spawnv('P_WAIT',r'C:\rei\bin\PFE32.EXE', ('xx','yy'))

'P_WAIT' should not be quoted. Try the following:

   os.spawnv(os.P_WAIT,r'C:\rei\bin\PFE32.EXE', ('xx','yy'))

-Steve

-- 
Steve Purcell, Pythangelist
Get testing at http://pyunit.sourceforge.net/
Any opinions expressed herein are my own and not necessarily those of Yahoo



From lloyd at venix.com  Thu Mar  1 09:11:27 2001
From: lloyd at venix.com (Lloyd Kvam)
Date: Thu, 01 Mar 2001 09:11:27 -0500
Subject: Collection interfaces
References: <3a90fa97.0@news1.mweb.co.za> <3A933830.7F98D118@ix.netcom.com> <CgJk6.515$v77.66568@newsread2.prod.itd.earthlink.net> <220220010813598042%cmh@bDistributed.com> <ud7cakvi8.fsf@roundpoint.com> <MPG.150312ceb68f1ada98969f@news.earthlink.net> <8Gem6.473353$U46.14105062@news1.sttls1.wa.home.com> <MPG.1503548620e632919896a4@news.earthlink.net> <MCjm6.474652$U46.14163288@news1.sttls1.wa.home.com> <MPG.150394195f7f04a19896a7@news.earthlink.net> <3A9A86B1.9A8C69A2@ix.netcom.com> <MPG.1504d14563dce6499896af@news.earthlink.net> <97fjla128ir@news2.newsguy.com> <MPG.1505a76831c34b749896b3@news.earthlink.net> <97h5s40vhc@news2.newsguy.com> <MPG.1506007591cf818b9896bc@news.earthlink.net> <slrn99opop.gr.amk@mira.erols.com> <MPG.1506265ebf6d934d9896be@news.earthlink.net> <97ibmc0k47@news1.newsguy.com> <Nh3n6.15937$Qb7.2602111@newsb.telia.net>
Message-ID: <3A9E588F.406A7C3E@venix.com>

One of the beauties of Python is the ease with which you can combine "table
lookup" style logic with "OOP inheritence" style logic within the same program.

Fredrik Lundh wrote:
> 
> > Before you knock it, TRY it
> 
> the martelli battlebot vs. the "galileo of
> programming" crackpot?
> 
> now this could be really entertaining ;-)
> 
> Cheers /F

-- 
Lloyd Kvam
Venix Corp.
1 Court Street, Suite 378
Lebanon, NH 03766-1358

voice:	603-443-6155
fax:	801-459-9582


From sholden at holdenweb.com  Fri Mar  9 18:53:46 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Fri, 9 Mar 2001 18:53:46 -0500
Subject: gah! I hate the new string syntax
References: <mailman.983615170.23313.python-list@python.org> <mailman.984014353.5058.python-list@python.org> <BcCp6.37817$1D5.1582482@e420r-atl1.usenetserver.com> <98amlj01sp6@news2.newsguy.com> <98asrq$645$1@panix6.panix.com>
Message-ID: <G5eq6.47941$ML1.2142034@e420r-atl2.usenetserver.com>

"Aahz Maruch" <aahz at panix.com> wrote in message
news:98asrq$645$1 at panix6.panix.com...
> In article <98amlj01sp6 at news2.newsguy.com>,
> Alex Martelli <aleaxit at yahoo.com> wrote:
> >"Steve Holden" <sholden at holdenweb.com> wrote in message
> >news:BcCp6.37817$1D5.1582482 at e420r-atl1.usenetserver.com...
> >    [snip]
> >> > | > What about one who is both English and Scots, yet never set foot
on
> >> > | > that island ;-)?
> >    [snip]
> >> Either you were born in England, or you were born in Scotland, or you
are
> >> neither English nor a Scot. Part-either-or-both, maybe, but not fully
> >
> >Wrong!  You're talking about people whose seafaring skills and
> >traditions are (deservedly) legendary.  It's perfectly acceptable
> >for either an English or a Scot to be born aboard a ship...!!!
>
> Since when are *women* allowed aboard ships?

Women were allowed aboard British ships in the eighteenth and nineteenth
centuries, I believe. The phrase "shake a leg" (meaning to stop lazing about
and start work) is held to have been used by ships' mates to get the seamen
out of their bunks. A woman would show her leg to let the mate know she
should not be rousted out to work.

regard
 Steve





From tim.one at home.com  Thu Mar  1 14:30:25 2001
From: tim.one at home.com (Tim Peters)
Date: Thu, 1 Mar 2001 14:30:25 -0500
Subject: A counter-proposal to __future__ in PEP 236
In-Reply-To: <3A9E9B23.23FFCEDF@home.se>
Message-ID: <LNBBLJKPBEHFEDALKOLCMEBKJDAA.tim.one@home.com>

[Sverker Nilsson]
> I think the directive directive is better too. When I read about
> the future import I was worried over what the new semantics
> of import would be, for example what would happen to the
> identifier that would be imported: would it be visible,
> probably not but that's a new semantics to understand & learn.

No, I think PEP 236 is quite clear that, at runtime, future_statements *are*
utterly vanilla import statements.  They do import module __future__.py, and
they do import the feature names (which are bound to detailed release info,
which is important metadata for people who use this; the "directive" proposal
loses that).




From leo at iems.nwu.edu  Thu Mar  8 14:27:18 2001
From: leo at iems.nwu.edu (Leonardo B Lopes)
Date: Thu, 08 Mar 2001 13:27:18 -0600
Subject: pipe problems
References: <3AA6D886.4DD8D3A4@iems.nwu.edu> <984045555.82149@newsmaster-04.atnet.at>
Message-ID: <3AA7DD16.93C0730D@iems.nwu.edu>

Werner Schiendl wrote:
> 
> If there is no data to read, readline it will block for such to become
> available.
> I don't think that cout.flush() is of any use, btw.

When I tried the program with 'cat', it only worked if I used
cout.flush(). But maybe it doesn't work in general. My guess is that
this call is the crux of the problem. 

> Maybe I'm wrong, but I think data can only be flushed out by the sender of a
> pipe...
> Further, what's if the program prints out something in between, or more than
> 1 line?

That is something I'll deal with at a later point, when I get the basic
stuff working. I have the impression there is something about pipes that
I don't understand correctly.

Thanks,
Leo.

> 
> Maybe you can do the reading in a seperate thread.
> Alternatively you can try to find out if there _is_ sth you can read.
> 
> hth
> werner
> 
> Leonardo B Lopes <leo at iems.nwu.edu> wrote in message
> news:3AA6D886.4DD8D3A4 at iems.nwu.edu...
> > I use a program that doesn't have readline compiled in. So I figured, it
> > would be easy to add a python wrapper around it. So I wrote the wrapper,
> > using 'cat' as a test command. With 'cat' it works fine. But when I try
> > using another program, such as 'python', the program just hangs at
> > readline(). Anybody know how to accomplish this? Here is the code:
> >
> > #!/usr/local/bin/python
> >
> > import sys,os
> > import cmd
> >
> > class Amplcmd(cmd.Cmd):
> >
> >         def __init__(self):
> >                 cmd.Cmd.__init__(self)
> >                 self.cin,self.cout = os.popen2('/usr/local/bin/python')
> >
> >         def default(self,line):
> >                 self.cin.write(line+'\n')
> >                 self.cin.flush()
> >                 self.cout.flush()
> >                 print self.cout.readline(),
> >
> >         def do_EOF(self,line):
> >                 sys.exit()
> >
> > mycmd = Amplcmd()
> > mycmd.cmdloop()
> >
> >
> >
> > --
> > =======================================================================
> > Leonardo B. Lopes                                      leo at iems.nwu.edu
> > Ph.D. Student                                             (847)491-8470
> > IEMS - Northwestern University             http://www.iems.nwu.edu/~leo

-- 
=======================================================================
Leonardo B. Lopes                                      leo at iems.nwu.edu 
Ph.D. Student                                             (847)491-8470
IEMS - Northwestern University             http://www.iems.nwu.edu/~leo


From gzeljko at sezampro.yu  Fri Mar 16 08:19:55 2001
From: gzeljko at sezampro.yu (gzeljko)
Date: Fri, 16 Mar 2001 14:19:55 +0100
Subject: How do I get Win32 to locate DLLs elsewhere (not really a Python thing)?
References: <mailman.984506597.12469.python-list@python.org> <3AB1F697.485A776D@yahoo.fr>
Message-ID: <98t3sn$8t9$1@neptun.beotel.net>

Francois Forest <francois_forest at yahoo.fr> wrote in message
news:3AB1F697.485A776D at yahoo.fr...
> I got the same kind of problem on WinNT and Tcl80.dll,
> I did the same tinkering you did and saw no improvement either.
> Finally, I cleaned up the registry of all useless path (a poor install
> program had set some C:\RECYCLER\123456xxx\... in many paths).
> Do not forget to reboot (actually I logged off).
> I suspect one PATH variable to have reached a maximum length.
> Anyway, the PATH variables on NT are totally unclear to me
>

Your have system provaided interface:
ControlPanel/System/Environment  (No need for reboot or 'relog')

search algoritam for dll's: current dir, exe's dir, path dir's

Note: User environment is just appendix to system environment
    (except in overriding case)
    and user path is just appendix to system path.

ly-y'rs-gzeljko




From frohne at gci.net  Thu Mar  1 19:55:37 2001
From: frohne at gci.net (Ivan Frohne)
Date: Thu, 1 Mar 2001 15:55:37 -0900
Subject: Hypergeometric distribution ?
References: <y9vzof58mn0.fsf@genome.cbs.dtu.dk>
Message-ID: <t9trs9735tj45d@corp.supernews.com>

----- Original Message -----
From: "Thomas Sicheritz-Ponten" <thomas at cbs.dtu.dk>

> Does anybody have a python function/class for calculating the
> hypergeometric distribution ?

The reference here is to the discrete hypergeometric
probability distribution, not the mathematical special function.

Calculating hypergeometric distribution probabilities
requires some utility functions -- factorial(n) and
log-factorial(n), n a non-negative integer.  Some care
needs to be taken to avoid impossible combinations of
parameters, and there are symmetries which can
be exploited.  Then, for large parameters, you need
to approximate using the normal, binomial or Poisson
distributions.  Your particular situation may allow
cutting some corners here.

You will probably end up translating Fortran or C
code, and you will end up with a package, not just
a function.

--Ivan Frohne




From grante at visi.com  Wed Mar  7 14:02:58 2001
From: grante at visi.com (Grant Edwards)
Date: Wed, 07 Mar 2001 19:02:58 GMT
Subject: Can anyone recomend a good intoduction to C...
References: <mailman.983894288.19965.python-list@python.org> <983896344.701456@newsmaster-04.atnet.at> <m3lmqibw0a.fsf@atrus.jesus.cam.ac.uk> <983911749.217622@newsmaster-04.atnet.at> <UScp6.1097$y6.196347@ruti.visi.com> <983916572.286918@newsmaster-04.atnet.at> <9852b8015jm@news1.newsguy.com> <983967284.246700@newsmaster-04.atnet.at> <o9up6.1291$y6.235469@ruti.visi.com> <983989153.204220@newsmaster-04.atnet.at>
Message-ID: <CBvp6.1318$y6.238504@ruti.visi.com>

In article <983989153.204220 at newsmaster-04.atnet.at>, Werner Schiendl wrote:

>> >What use would it be to check the language?
>>
>> The point is that the more complex the language, the more
>> difficult it is to inspect the application.  You can look at a
>> few lines of C and have a pretty good chance at guessing what
>> they do.  In C++ it's much harder to look at snippet of code
>> and figure out what it's going to do.
>>
>
>At least if the design is bad.
>
>Usually an object oriented program should be easier to read.
>Just because the syntax does more closely resemble the way
>humans think.

It _should_ be easier to read.  But in my experience it isn't.

The problem is that C++ is a very complex language.  It's
difficult to determine by inspection what a piece of code does
because of all the overloading/templating/whatnot that C++
allows.  There are too many times when a line of code doesn't
do what it appears to do to a reader.

Just the problem of deciding whether "foo" is a local, global,
or instance variable can be difficult for a human inspector.
Throw in some cpp tricks and it's hopeless.  I know that C++
was supposed to eliminate the need for cpp, but it still gets
used.

>(But I saw enough examples that demonstrated just the
>opposite...)

In my experience, C++ is too complex a language to allow for
practical, thorough inspection of source code.  And since C++
still allows you to generate dangling pointers, memory leaks,
indexes out of range and whatnot, you _need_ to inspect source
code to detect many of those types of problems.

C also allows you to generate those problems, but the language
is simple enough that inspection is a practical exercise.

OTOH, Python (or Modula-3 or Smalltalk or ....) won't _allow_
you to generate a pointer to nothing or index off the end of an
array.  Additionally, they're far simpler languages than C++, so
other types of errors are easier to spot.

-- 
Grant Edwards                   grante             Yow!  Either CONFESS now or
                                  at               we go to "PEOPLE'S COURT"!!
                               visi.com            


From dallen at wi.net  Sat Mar  3 19:27:21 2001
From: dallen at wi.net (mark)
Date: Sat, 3 Mar 2001 18:27:21 -0600
Subject: Beginer
Message-ID: <3aa18bf8$0$65279$38c90ffe@news1.twtelecom.net>

can anyone tell me where i can find a real beginer tutorial for python? The
one that comes with the program is still to complicated.thanks>




From aleaxit at yahoo.com  Sun Mar 25 16:59:00 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sun, 25 Mar 2001 23:59:00 +0200
Subject: Yet Another PEP:  Query Protocol Interface or __query__
References: <Pine.LNX.4.21.0103201837570.21343-100000@clarkevans.com> <mailman.985490195.2917.python-list@python.org> <99lnc6$ck4$1@tyfon.itea.ntnu.no>
Message-ID: <99lpkt0235c@news1.newsguy.com>

"Magnus Lie Hetland" <mlh at idi.ntnu.no> wrote in message
news:99lnc6$ck4$1 at tyfon.itea.ntnu.no...
    [snip]
> >     To enable the first case listed in the requirements,
> >     the adapt function first checks to see if the objects
> >     type or the objects class are identical to the protocol.
> >     If so, then the adapt function returns the object directly
> >     without further ado.
>
> Hm. So in some cases the protocol object is merely an example
> (prototype) of an object satisfying the protocol, and in other

I suspect a mis-reading here.  What I read is 'type or class
identical to the protocol', i.e. the protocol can be a type object
or class object; NOT 'identical to *that of* the protocol'.

Maybe the wording needs clarification (make more explicit).


> How about __adjust__? Sounds more transitive than __adapt__ to
> me...

Or __accept__.

>   __conform__ - method in an object which can conform to
>                 a protocol
>   __adapt__   - method in a protocol used to adapt an
>                 object to it

Agree, probably best.


Alex





From kevin_cazabon at hotmail.nospamplease!.com  Sat Mar 17 13:49:19 2001
From: kevin_cazabon at hotmail.nospamplease!.com (Kevin Cazabon)
Date: Sat, 17 Mar 2001 11:49:19 -0700
Subject: image handling
References: <mailman.984795862.9586.python-list@python.org>
Message-ID: <scIN4LxrAHA.187@typhoon.i3s.net>

Check out the Python Imaging Library... join the Image-SIG at Python.com.

You can download source and binaries at:  www.pythonware.com... they've
recently renamed the library the "Pythonware Image Library", still PIL...
don't like the commercialism of the new name though myself... They've done
most of it, but a LOT of people have contributed (including myself).

Kevin.


"Michael Hall" <admin at mulga.com.au> wrote in message
news:mailman.984795862.9586.python-list at python.org...
|
| G'day people:
|
| I'm brand new on this list and am looking for resources about image
| processing in a web environment in Python. I'm wondering if there exists a
| module that does things like the PHP image handling functions
| (creating, copying, resizing, etc).
|
| I've had a look at the Python Functions Library but there doesn't appear
| to be much in there. I've started to use Python a bit in a web environment
| and like what I've seen so far, and from what I've heard I'm sure there
| must be some pretty flash image handling capabilities hanging around
| somewhere.
|
| Thanks
| Mick
|
| ***********************************************
| Michael Hall - mick at mulga.com.au - mulga.com.au
| ***********************************************
|
|
|




From barry at digicool.com  Mon Mar 12 17:42:41 2001
From: barry at digicool.com (Barry A. Warsaw)
Date: Mon, 12 Mar 2001 17:42:41 -0500
Subject: ANNOUNCE Mailman 2.0.3
Message-ID: <15021.20705.958300.336520@anthem.wooz.org>

[I think I forgot to announce the Mailman 2.0.2 release on these
mailing lists.  Apologies.  -Barry]

I've just announced the 2.0.3 release of Mailman, the GNU Mailing List
Manager.  Mailman is released under the GNU General Public License
(GPL).

I highly recommend that all sites running earlier versions of Mailman
upgrade to 2.0.3.  This release includes the important privacy patch
from Mailman 2.0.2.  Sadly, Mailman 2.0.2 contained a small but nasty
typo requiring this new version.  Mailman 2.0.3 is being released as
both a gzip'd source tarball and as a patchfile.

Mailman is software to help manage electronic mail discussion lists,
much like Majordomo or Smartmail.  Mailman gives each mailing list a
unique web page and allows users to subscribe, unsubscribe, and change
their account options over the web.  Even the list manager can
administer his or her list entirely via the web.  Mailman has most of
the features that people want in a mailing list management system,
including built-in archiving, mail-to-news gateways, spam filters,
bounce detection, digest delivery, and so on.

Mailman is compatible with most web servers, web browsers, and mail
servers.  It runs on GNU/Linux and should run on any other Unix-like
operating system.  Mailman 2.0.3 requires Python 1.5.2 or newer.  To
install Mailman from source, you will need a C compiler.

For more information on Mailman, including links to file downloads,
please see the Mailman WWW page: http://www.gnu.org/software/mailman

And its mirrors at:

    http://mailman.sourceforge.net
    http://www.list.org

[Only the SourceForge and list.org sites are up-to-date at the moment,
I expect the gnu.org site to be updated shortly.]

If you want to download just the 2.0.2 -> 2.0.3 patch, please see:

    http://sourceforge.net/project/showfiles.php?group_id=103

There are email lists (managed by Mailman, of course!) for both
Mailman users and developers.  See the web sites above for details.

Enjoy,
-Barry



From max at alcyone.com  Mon Mar  5 12:09:03 2001
From: max at alcyone.com (Erik Max Francis)
Date: Mon, 05 Mar 2001 09:09:03 -0800
Subject: the demise of 'from foo import * and its implications?
References: <3kVBAfATrDo6EwEM@jessikat.fsnet.co.uk><mailman.983582584.8198.python-list@python.org><z8dAGJAyqLo6EwlX@jessikat.fsnet.co.uk> <mailman.983746566.27513.python-list@python.org> <suPo6.16824$ML1.841861@e420r-atl2.usenetserver.com>
Message-ID: <3AA3C82F.E2F6CA7F@alcyone.com>

Steve Holden wrote:

> I did begin to wonder whether it might be possible to specify a name 
> prefix,
> along the lines of
>
>     from modulewithaverylongname import * prefix sh_
>
> to allow
>
>     sh_i
>
> as a reference to the module's "i", but I'm not sure that the gain is
> worth
> the pain. Even if such a proposal were to be PEP-ified and accepted it
> could
> not be mandated because that would break humongous amounts of existing
> code.
> 
> Thoughts?

I don't see how changing a . to an _ really helps anything, and all it
does is make the language more complex, and add a new keyword which is
_very_ likely to be used in existing code.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ Even paranoids have real enemies.
\__/ Delmore Schwartz
    Crank Dot Net / http://www.crank.net/
 Cranks, crackpots, kooks, & loons on the Net.


From aahz at panix.com  Fri Mar 16 22:40:17 2001
From: aahz at panix.com (Aahz Maruch)
Date: 16 Mar 2001 19:40:17 -0800
Subject: This math scares me
References: <20010312170356.B3513@better.net> <mailman.984664520.26463.python-list@python.org> <HR4s6.877$lz1.73512@ruti.visi.com> <98qnq0$qc2$1@panix3.panix.com>
Message-ID: <98umb1$f2i$1@panix3.panix.com>

In article <98qnq0$qc2$1 at panix3.panix.com>, Aahz Maruch <aahz at panix.com> wrote:
>In article <HR4s6.877$lz1.73512 at ruti.visi.com>,
>Grant Edwards <grante at visi.com> wrote:
>>
>>One step further is base-60, which is still used for many
>>things.  Time and angular measurement for example is partially
>>a base-60 system.  I believe that one or more of the ancient
>>Aztecs/Incans/Mayans used a base-60 number system.
>
>Paging Ivan Van Laningham.....

Ivan isn't reading c.l.py these days, but he sent me an e-mail response
that he said I could post:

We get base 60 from the Babylonians, who got it from the Sumerians. 
Base 60 is *much* more useful for time/date/astronomical calculations
than any other.  One of the projects in my datebook is a base60 package,
but if anyone else writes one I'll be happy to use it;-)

Aztecs and Mayans used base 20, expanded from an earlier base 10 system,
which appears to have grown from an earlier base 5.

Counting in Mayan civilization was, um, weird.  Check Lounsbury, Floyd
G., ``Maya Numeration, Computation, and Calendrical Astronomy,'' in
Dictionary of Scientific Biography, ed., Charles Coulston Gillespie,
Vol. 15, Supplement 1 (1978), Scribners, New York, 1978.  As a small
example, instead of saying "twenty-three," Mayans would say "three
towards forty."  <g>

The Incans used base ten exclusively, but recorded the numbers in knot
form (on _quipus_).

Many other radices were popular in pre-Conquest America.  Base 3, bases
4, 5, 6, 8, 10 and 20.  Check out Closs, Michael P., ed., Native
American Mathematics, University of Texas Press, Austin, 1986.  Still in
print, in trade paper edition.

<one-towards-the-third-thousand>-ly y'rs,
Ivan;-)
----------------------------------------------
Ivan Van Laningham
Symantec
http://www.pauahtun.org/
http://www.foretec.com/python/workshops/1998-11/proceedings.html
Army Signal Corps:  Cu Chi, Class of '70
Author:  Teach Yourself Python in 24 Hours


From tim.one at home.com  Thu Mar  1 15:45:17 2001
From: tim.one at home.com (Tim Peters)
Date: Thu, 1 Mar 2001 15:45:17 -0500
Subject: A counter-proposal to __future__ in PEP 236
In-Reply-To: <j47l29lpo8.fsf@informatik.hu-berlin.de>
Message-ID: <LNBBLJKPBEHFEDALKOLCMEBOJDAA.tim.one@home.com>

[Martin von Loewis]
> To selectively activate nested scopes in Python 2.1, PEP 236 proposes
> the "future" statement. That means that you have to write
>
> from __future__ import nested_scopes
>
> Even though this looks like an import statement, it is not:

Not so:  it is an import statement.  Read the PEP <0.9 wink>.

> instead, it changes the meaning of variable binding in the presence
> of nested functions.

Which (the compile-time meaning) is the more important part, but not the only
part (at run-time it's an ordinary import statement).

> I personally consider this a confusing notation,

I don't.  "__future__" screams "Deep Magic" because of the double
double-underscores; it *is* a perfectly vanilla import statement at runtime;
the word "future" helpfully suggests temporality; it works with all existing
tools today without change (so it's not confusing even the stupidest of them
<0.1 wink>); and the imported nested_scopes name gets bound to release data
that the importing code can examine; e.g.,

from __future__ import nested_scopes
assert nested_scopes[1] > sys.version_info, "don't need this anymore"

> and propose a different syntax, which would read
>
> directive nested_scopes

Which loses all of that.  Incompatible language changes are a serious special
case, and I don't want to see them swallowed up by some perfectly general
mechanism that makes them impossible to distinguish visually from some
innocuous request to, e.g., optimize for space instead of time.

> The directive statement currently only supports activating nested
> scopes,

I really dislike that

    directive nested_scopes

*looks* exactly the same as

    directive favor_space

or

    directive favor_time

or

    directive alabaster_weenoblobs

That will add to confusion over time, not reduce it.  Incompatible language
changes have to kick you in the teeth.

> but it may get further use later, e.g. to activate case-insensitivity
> or set the source file character set.
>
> A patch implementing the directive statement is at
>
> http://sourceforge.net/tracker/index.php?func=detail&aid=404997&gro
> up_id=5470&atid=305470
>
> In this implementation, directive is only considered as a keyword if
> it appears at the beginning of the module.

In that case it's too feeble for a pragma facility; e.g., perhaps I want to
optimize one method for space but another for time.  Or perhaps I want to
specify that one class in a module uses traditional dicts for its namespace,
but another freezes its attributes into integer-indexable vectors at
compile-time.  Etc.  A usefully general pragma facility can't be restricted
to the top of the module, because not all useful pragmatic choices apply to
an entire module.

> Therefore, code that happens to use directive as an identifier will
> not break (unless it assigns to directive as the first thing in the
> module).

Make it general enough to be useful, and it will get hairier than that.  Bite
the bullet:  if you want a new statement, make directive a reserved word.
Then we can introduce it via

    from __future__ import directive_is_reserved

in the release before it breaks code <0.7 wink>.

Note that it's also not enough to allow just one or two optional tokens
following "directive".  One of the primary reasons no pragma facility ever
got into Python is that in previous attempts nobody could agree on syntax
that *was* sufficiently general (as a hint of the horrors you're signing up
for, note that "optional static typing" fits-- because of the "optional"
part --into a pragma framework; and once people realize that, they're going
to hijack your proposal into an overly-generalized mess too).

> I'd appreciate any comments as to whether this directive statement is
> better or worse for the purpose of indicating that nested scopes are
> used in a module.

IMO it would be *much* better for everything *except* incompatible language
changes, for which latter purpose I dislike it.  The future_statement is
specifically aimed at the latter, and I believe will support it very well.
And I want not to confuse that purpose with other stuff.

> Please note that making nested scopes the default in a major release
> after 2.1 is an unrelated issue:

It is related:  the "__future__" in a future_statement helpfully implies that
a future_statement is temporary (lost via "directive"); the detailed release
info in __future__.py supplies a programatic means for automating tasks
related to when a future_statement becomes obsolete (lost via "directive");
and ditto for the feature name imported (lost via "directive").

> it can happen with either future statements or directives;

Except that directives lack all of the special support built into PEP 236 for
supporting incompatible language changes specifically.

> when it happens, both the future statement and the directive can continue
> to work (although they would be unnecessary); tools could automatically
> remove unnessary directives if desired.

But much more easily for future_statements, and especially if you generalize
the directive stmt enough to cover all the easily predictable uses for it,
and don't make "directive" a reserved word.

Note that I'm not opposed to a general pragma facility.  It could be a nice
hammer.  But that doesn't mean everything is a nail.

sell-it-on-its-real-merits-ly y'rs  - tim




From chega_ at yahoo.com  Sun Mar 18 23:05:20 2001
From: chega_ at yahoo.com (Vadim Chugunov)
Date: Mon, 19 Mar 2001 04:05:20 GMT
Subject: Why "from __future__" stinks; a counter-offer
References: <mailman.984949523.1896.python-list@python.org>
Message-ID: <4Aft6.2171$pP2.372173@paloalto-snr1.gtei.net>

"Tim Peters" <tim_one at email.msn.com> wrote in message
news:mailman.984949523.1896.python-list at python.org...
> [Jason Orendorff]
> > Okay, nested scopes are a good idea.  Having the feature
> > "disabled by default" in 2.1 is a good idea.  But this
> > "from __future__" stuff isn't so hot.
> Did you read Martin's messages on this?  He picked "directive", and
submitted
> a patch to implement it (or some part of it).  But he didn't write a PEP,
so
> you'll have to search c.l.py and SourceForge for the relevant bits.  He
had
> in mind a more general facility, and in particular wants to specify
> source-file character encodings someday; I actively want not to conflate
> future_statements with "general directives" (although "directive future
xxx"
> might be OK -- alas, without a PEP, it's a non-starter).

But, surely, the PEP can be updated!   This "from __future__" syntax is
ugly, ugly, ugly!!!





From aahz at panix.com  Fri Mar 16 19:16:25 2001
From: aahz at panix.com (Aahz Maruch)
Date: 16 Mar 2001 16:16:25 -0800
Subject: Newbie confusion about 'return'
References: <3AB2A2BE.A467DE00@hotmail.com>
Message-ID: <98uacp$ads$1@panix2.panix.com>

In article <3AB2A2BE.A467DE00 at hotmail.com>, scott  <smarsh at hotmail.com> wrote:
>I'm mystified. The same code in IDLE (or at the interactive prompt)
>gives a different result compared to entering it into a file and running
>it. I'm using Python 2.0 from python .org on NT4. I checked on Solaris 8
>with python 2.0 (interactive vs running the file) and got the same
>result. The code is from "How to think like a computer scientist Chapter
>7"
>-----------------------------------
>IDLE:
>
>>>> def find(str, ch):
>              index = 0
>              while index < len(str):
>                      if str[index] == ch:
>                             return index
>                      index = index + 1
>              return -1
>
>>>> find ("abcde", "d")
>3 (in blue indicating stdout)
>
>The above is what I expected.

That's because IDLE itself is a Python program that captures the output
of return.  Try entering *any* Python expression, and its output will be
printed.  You don't want that in a regular program, though, so you need
to be explicit:

>
>Running python on the file (find.py):
>
>def find(str, ch):
>      index = 0
>      while index < len(str):
>             if str[index] == ch:
>                    return index
>             index = index + 1
>      return -1
>
>find ("abcde", "d")

Try

print find ("abcde", "d")
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

Three sins: BJ, B&J, B&J


From NoSpam at NoSpam.com  Tue Mar 27 12:46:36 2001
From: NoSpam at NoSpam.com (Tom)
Date: Tue, 27 Mar 2001 17:46:36 GMT
Subject: macro file (containing C def's) for source navigator?
Message-ID: <0m4w6.91100$A6.20818962@news4.rdc1.on.home.com>

Does anyone have a Source Navigator macro file for the Python source code?
(This is a file that lists the preprocessor symbol definitions for a
project.)

Do the python internals hackers use Source Navigator?  What do they use?

Is this information available somewhere in the source distribution, so I can
make my own macro file?

Thanks,

Tom.




From bbollenbach at homenospam.com  Wed Mar  7 21:10:22 2001
From: bbollenbach at homenospam.com (Brad Bollenbach)
Date: Thu, 08 Mar 2001 02:10:22 GMT
Subject: Can Python replace Visual Basic? Should it?
References: <5DZo6.11126$hn5.1578410@news1.rdc1.mb.home.com> <mailman.984010931.26257.python-list@python.org>
Message-ID: <iSBp6.14005$hn5.2035325@news1.rdc1.mb.home.com>

[snip]

> I'd say go for it -- get the Python on Windows book and convert one of
your

[snip]

"I'd say go for it" too...I'm looking for people that ARE going for it
though. :)

> distribution).  I converted one of our ASP pages to Python from VBScript
to
> see if it would be faster (it wasn't), and was pleasantly surprised at how
> easy it was.

Yes, and that's WEB programming, not end-user GUI. In CGI or ASP, using
Python isn't something that would surprise me, and if I knew more about it
right this minute, I might not use Perl to do the odd consulting jobs I do.
But I'm not at that stage just yet.

> I should say, though, that we haven't used any of the GUI toolkits --
maybe
> someone else can recommend the best one to use when migrating from VB
Forms.

[snip]

Yes, and those were the people I was hoping (as in Mr. Dukarm's case) would
be answering my post, not theoretical "give it a try!"'s...I coulda said
that. :)

Anyways, do I sound bitter? :) No, but maybe the fact that only one person
who saw my original post has actually migrated from VB to Python is an
answer in itself.




From f8dy at diveintopython.org  Fri Mar 16 19:00:00 2001
From: f8dy at diveintopython.org (Mark Pilgrim)
Date: Fri, 16 Mar 2001 19:00:00 -0500
Subject: HTMLparsing abnormal html pages
References: <98pvp1$15t$1@news.netmar.com>
Message-ID: <B6D81330.4FDA%f8dy@diveintopython.org>

in article 98pvp1$15t$1 at news.netmar.com, asle at spam.com at asle at spam.com
wrote on 3/15/01 3:50 AM:

> Considering the small program below. Running it will show that the
> HTMLparser
> is truncating urls in the HTML page.
> [...]
> import htmllib
> [...]
> One solution is of course to preprosess the whole HTML page and
> replacing invalid URL's whith valid URL's (using regex??), however I have

Don't use htmllib, use sgmllib.  It does exactly what you want: uses regular
expressions to pull out the tags and attributes of potentially messy HTML,
then calls methods on itself based on the tags.  You can subclass it and
provide methods for each tag.

from sgmllib import SGMLParser

class MessyURLParser(SGMLParser):
  def reset(self):
    SGMLParser.reset(self)
    self.urls = []
  def start_a(self, attrs):
    # at this point, attrs is a list of tuples (attrname, attrvalue)
    # attrname is converted to lowercase by SGMLParser
    # so for a tag <a HREF=index.html>, attrs would be
    # [('href', 'index.html')]
    hrefpair = [v for k, v in attrs if k=='href']
    if hrefpair:
      self.urls.append(hrefpair[0])

This is such a useful technique, I'm writing an entire chapter on it in my
book, "Dive Into Python".  The code example in the book is more complicated
than what you're looking for (it's designed to both consume and produce
HTML, whereas you want to consume HTML and produce a list), and the chapter
is barely started, but all the code is there along with a more complete
explanation of how sgmllib works:
  http://diveintopython.org/dialect_divein.html

Hope this helps.

-M
You're smart; why haven't you learned Python yet?
http://diveintopython.org/





From shaleh at valinux.com  Wed Mar 21 18:33:19 2001
From: shaleh at valinux.com (Sean 'Shaleh' Perry)
Date: Wed, 21 Mar 2001 15:33:19 -0800 (PST)
Subject: Simple regex that has me banging my head against a wall
In-Reply-To: <20010321150205.N24341@trufflehunter.avalongroup.net>
Message-ID: <XFMail.20010321153319.shaleh@valinux.com>

> 
>>>> re.sub(r'\$', '\$', x)
> '\\$1\\$asdfjke\\$lkjasdkjfsdasdfj'
> 
> That's why I'm a bit confused.
> 

The double \ is printed by python.  If you do s = re.sub() and then print s, it
looks just fine.  I was confused by this when I tried it out myself.



From ws-news at gmx.at  Wed Mar 14 09:18:08 2001
From: ws-news at gmx.at (Werner Schiendl)
Date: Wed, 14 Mar 2001 15:18:08 +0100
Subject: How can I get the logged in users username on Win32?
References: <3aaf7ab4.3152221@news.is.co.za>
Message-ID: <984579774.633036@newsmaster-04.atnet.at>

> As a side note, how can I enable Windows Scripting Host support for
> python - there is mention of python in the original WSH docs - and if
> I can use python scripts in WSH how would I call the built-in
> Wscript.<methods>
>

I installed the ActivePython distribution from ActiveState and this works by
default.
Just give your file the extension ".pys" and it works.

Samples are found in the installation as well (Python 2.0)
<Install-Dir>\win32comext\axscript\demos\client\wsh

hth
werner

> TIA
>
> nomad




From tanzer at swing.co.at  Wed Mar 14 01:54:51 2001
From: tanzer at swing.co.at (Christian Tanzer)
Date: Wed, 14 Mar 2001 07:54:51 +0100
Subject: Yet another Python vs. Perl speed issue/question
In-Reply-To: Your message of "Tue, 13 Mar 2001 22:17:49 GMT."
             <98m6ad+i25s@eGroups.com>
Message-ID: <m14d5BL-000wcBC@swing.co.at>

nanotech at europa.com wrote:

> Python 2.0 on an HP (nice and big) takes ~30 seconds to lowercase all
> the lines in a 3meg file. Perl takes less than a second. What am I
> doing wrong?!?
>
> Python
> ------
> import sys
> import string
>
> stdin=sys.stdin
> lower=string.lower
>
> print map(lower,stdin.readlines())
>
> Perl
> ----
>
> print map(lc,<STDIN>);

Your Python code is doing unnecessary work (BTW, it doesn't do what
you expect). Just for comparison, the timings for a 3MB file on my
aging Pentium Pro 200 MHz machine:

    $ time python /tmp/test1.py < /tmp/test.txt > /tmp/test.out 
    real    0m10.246s
    user    0m4.810s
    sys     0m1.190s
    $ time python /tmp/test2.py < /tmp/test.txt > /tmp/test.out 
    real    0m0.781s
    user    0m0.200s
    sys     0m0.270s

    #### test1.py                       #### test2.py
    import sys                          import sys
    import string                       import string
    stdin=sys.stdin                     stdin=sys.stdin
    lower=string.lower                  lower=string.lower
    print map(lower,stdin.readlines())  sys.stdout.write(lower(stdin.read()))

Your perl code gives a compilation error:

    $ time perl /tmp/test1.pl < /tmp/test.txt > /tmp/test.out
    Not enough arguments for lower case at /tmp/test1.pl line 1, near "lc,"
    Execution of /tmp/test1.pl aborted due to compilation errors.

Slightly changed, it lies halfway between the two python versions:

    $ time perl /tmp/test2.pl < /tmp/test.txt > /tmp/test.out
    real    0m3.297s
    user    0m2.830s
    sys     0m0.360s

    #### test2.pl
    print map {lc ($_)} <STDIN>;

-- 
Christian Tanzer                                         tanzer at swing.co.at
Glasauergasse 32                                       Tel: +43 1 876 62 36
A-1130 Vienna, Austria                                 Fax: +43 1 877 66 92




From hinsen at cnrs-orleans.fr  Fri Mar 30 05:27:18 2001
From: hinsen at cnrs-orleans.fr (Konrad Hinsen)
Date: 30 Mar 2001 12:27:18 +0200
Subject: NumPy loop efficiency
References: <3ac44e27$0$16692$73bec57d@news.be.uu.net>
Message-ID: <m3itkr7drt.fsf@chinon.cnrs-orleans.fr>

"S?bastien de Menten" <sdementen at hotmail.com> writes:

> I was wondering if writing the following statement with NumPy arrays z,a,b
> and c:
> z = a + b + c

Two loops plus creation of a temporary array.

> And if
> a[:]=cos(a)+sin(a)
> creates temporary variables or assigns directly the result (like
> a[i]=cos(a[i])+sin(a[i])

It does create a temporary array. Getting rid of this would be a lot
of work, array expressions would have to use lazy evaluation.
-- 
-------------------------------------------------------------------------------
Konrad Hinsen                            | E-Mail: hinsen at cnrs-orleans.fr
Centre de Biophysique Moleculaire (CNRS) | Tel.: +33-2.38.25.56.24
Rue Charles Sadron                       | Fax:  +33-2.38.63.15.17
45071 Orleans Cedex 2                    | Deutsch/Esperanto/English/
France                                   | Nederlands/Francais
-------------------------------------------------------------------------------


From loewis at informatik.hu-berlin.de  Sun Mar  4 05:51:04 2001
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 04 Mar 2001 11:51:04 +0100
Subject: how to get the last interactive command enetered
References: <F2Tn6.15639$UZ4.4310299@news4.rdc1.on.home.com> <j4bsrjxjy3.fsf@informatik.hu-berlin.de> <I59o6.19418$UZ4.5669336@news4.rdc1.on.home.com>
Message-ID: <j4zof17qp3.fsf@informatik.hu-berlin.de>

"Tom" <NoSpam at NoSpam.com> writes:

> > No. Once the code is executed, the source code is dropped. All you can
> > get at is the binary code.
> 
> But when I catch a 'call' trace event the code has not been executed.  I
> believe that it has been parsed and compiled but not executed (till I get
> the 'return' trace event).

Yes, it has been parsed and compiled, the result of which was a code
object (which in turn was put into a function object). After that, the
source code, and the abstract syntax tree, is discarded (actually the
source code is discarded in the process of parsing).

The code object has a file name in it and line numbers, that's all
that's left from the source.

Regards,
Martin


From erno-news at erno.iki.fi  Mon Mar 12 16:52:17 2001
From: erno-news at erno.iki.fi (Erno Kuusela)
Date: 12 Mar 2001 23:52:17 +0200
Subject: Problems writting a wrapper around linux' ssh command
References: <31rd89.lki.ln@127.0.0.1> <kuhf0zj4iz.fsf@lasipalatsi.fi> <u26i89.0bm.ln@127.0.0.1>
Message-ID: <kubsr6irji.fsf@lasipalatsi.fi>

In article <u26i89.0bm.ln at 127.0.0.1>, gradha at iname.com writes:

| Erno Kuusela <erno-news at erno.iki.fi> wrote:
|| do not embed the password in your python program. use a passphraseless
|| rsa key instead, it is safer and less pain.

| At the moment I am using the getpass module to retrieve the pass from the
| user and feed it directly to ssh and scp. Is that secure enough?
 
ah, if you are getting the password from the user, there is no
security issue. i had assumed you were storing the password
embedded in your python program.

rsa keys are more convenient though. ssh-agent + rsa key with
passphrase = safe yet no need to type in passwords all the time.

  -- erno


From grante at visi.com  Fri Mar 23 08:58:23 2001
From: grante at visi.com (Grant Edwards)
Date: Fri, 23 Mar 2001 13:58:23 GMT
Subject: Tkinter - Widget Colours
References: <mailman.985346222.22848.python-list@python.org>
Message-ID: <3EIu6.1971$SB2.254481@ruti.visi.com>

In article <mailman.985346222.22848.python-list at python.org>, O'Rourke Clodagh-corour01 wrote:

>I was wondering does anyone know how the bag/fag colours can
>be changed after the widget is packed.

widgetObject.configure(foreground="#000")  # black?
widgetObject.configure(background="#fff")  # white?

-- 
Grant Edwards                   grante             Yow!  It's OKAY --- I'm an
                                  at               INTELLECTUAL, too.
                               visi.com            


From chris.gonnerman at usa.net  Sat Mar 31 23:37:13 2001
From: chris.gonnerman at usa.net (Chris Gonnerman)
Date: Sat, 31 Mar 2001 22:37:13 -0600
Subject: Know of Substantial Apps Written in Python?
References: <LNBBLJKPBEHFEDALKOLCKEIHJJAA.tim.one@home.com>
Message-ID: <004101c0ba65$7098cfe0$a100000a@local>

----- Original Message -----
From: "Tim Peters" <tim.one at home.com>
Subject: RE: Know of Substantial Apps Written in Python?


> [Moshe Zadka]
> > ...
> > Python would never fit on a floppy.
>
> Sure it can!  Python + the entire base std library (top level of Lib/)
still
> fits on a floppy with ease, if you use any common compression scheme.
Just
> for fun, I just built a zip file with that stuff for the Windows 2.1b2
> distribution, and it came in at just over a tad above 1Mb (so about 70% of
> one floppy).  What that leaves out from the *std* Windows distribution is,
> primarily, 7Mb (unpacked) of HTML docs, Tcl/Tk, the C header files and
link
> libraries, and the Tools/ directory (incl. IDLE).

Have you seen Marc-Andr? Lemburg's mxCGIPython project?  Most of the
all-in-one-file Python interpreters there fit on a floppy (although of
course supporting libraries/DLL's may not).  I use CGIPython for WinNT
on CD-ROM's for Win95/98 with good success, and am using the Linux version
for CGI's on a Linux-based free hosting service which does not otherwise
support Python.  Don't let the CGI prefix fool you, you can do non-CGI work
(basically anything not needing the interactive mode) with this package.

Most of the binary builds are 1.5.2 but that doesn't bother me.





From topmind at technologist.com  Sat Mar  3 20:00:30 2001
From: topmind at technologist.com (Topmind)
Date: Sun, 04 Mar 2001 01:00:30 GMT
Subject: Collection interfaces
References: <3a90fa97.0@news1.mweb.co.za> <3A933830.7F98D118@ix.netcom.com> <CgJk6.515$v77.66568@newsread2.prod.itd.earthlink.net> <220220010813598042%cmh@bDistributed.com> <ud7cakvi8.fsf@roundpoint.com> <MPG.150312ceb68f1ada98969f@news.earthlink.net> <8Gem6.473353$U46.14105062@news1.sttls1.wa.home.com> <MPG.1503548620e632919896a4@news.earthlink.net> <MCjm6.474652$U46.14163288@news1.sttls1.wa.home.com> <MPG.150394195f7f04a19896a7@news.earthlink.net> <3A9A86B1.9A8C69A2@ix.netcom.com> <MPG.1504d14563dce6499896af@news.earthlink.net> <97fjla128ir@news2.newsguy.com> <MPG.1505a76831c34b749896b3@news.earthlink.net> <97h5s40vhc@news2.newsguy.com> <MPG.1506007591cf818b9896bc@news.earthlink.net> <slrn99opop.gr.amk@mira.erols.com> <MPG.1506265ebf6d934d9896be@news.earthlink.net> <97ibmc0k47@news1.newsguy.com> <MPG.1506cf3293b094b49896c1@news.earthlink.net> <97m0c7080c@news2.newsguy.com> <MPG.15085517debf4b649896ca@news.earthlink.net> <Bj7o6.2298$yv4.130639@bgtnsc07-news.ops.worldnet.att.net>
Message-ID: <MPG.150b35e9da7231029896ce@news.earthlink.net>

> "Topmind" <topmind at technologist.com> wrote in message
> news:MPG.15085517debf4b649896ca at news.earthlink.net...
> >
> > Well, my experience tells me that the "task axis"
> > is the most "business friendly" axis to build
> > along (being that multi-axis paradigms still need
> > a lot of work and thus we are forced to emphasize
> > one at the expense of others for now.)
> >
> We'll make an OOer of you yet.
> Psst. Less structural taxonomy and more task-oriented
> thought. You might even think of a little "agent" doing
> that task for you. Then, you'll dress that little agent up
> in some nice clothes, and presto, an object.
> 


If you can show a benefit of such wrapping.


> Or component. Perhaps you should just skip OO and
> go right to components. Just don't suggest components
> should be wrapped in tables.
> 

Sometimes the interfaces can.

> I vote YES for comp.top
> 

Only because you want to get rid of me.

-tmind-
 


From fredp at mygale.org.nospam  Wed Mar 21 12:11:29 2001
From: fredp at mygale.org.nospam (Fred Pacquier)
Date: 21 Mar 2001 17:11:29 GMT
Subject: Accessing Oracle from Python remotely ?
References: <Xns906AB9C901807PaCmAnRDLM@161.48.76.20> <eE3u6.776$9%.203510@newsr1.maine.rr.com> <F92141C233DA318B.A84C85D98997C694.A9951520D4A22E7C@lp.airnews.net>
Message-ID: <Xns906BB938FA21FPaCmAnRDLM@161.48.76.20>

claird at starbase.neosoft.com (Cameron Laird) said :

>In article <eE3u6.776$9%.203510 at newsr1.maine.rr.com>,
>gkiffney <gkiffney at my-deja.com> wrote:
>>There's http://www.zope.org/Products/DCOracle which should get you
>>started. There was an article by Uche Ogbuji in Linuxworld that helped
>>me get going with this, but it's apparently fallen off the Internet map
>>into the bit bucket.  Using the dcoracle module (which is a bit of a
>>pain to get going, 
>               .
>Do you mean
><URL: http://www.linuxworld.com/lw-2000-04/lw-04-oracle8i.html >?

Excellent ! Thanks ! Truly scary though... :-))

-- 
YAFAP : http://www.multimania.com/fredp/


From donn at u.washington.edu  Tue Mar 20 14:11:29 2001
From: donn at u.washington.edu (Donn Cave)
Date: 20 Mar 2001 19:11:29 GMT
Subject: Functionalism, aesthetics Was:(RE: I come to praise .join,  not
References: <mailman.985052844.17993.python-list@python.org> <996vss013mp@news2.newsguy.com>
Message-ID: <998a11$q3e$1@nntp6.u.washington.edu>

Quoth "Alex Martelli" <aleaxit at yahoo.com>:
...
| Buildings as such (and more generally dwelling-places) are old
| enough that it's quite possible that we all share some basic
| biological basis for sound judgment -- I'm not pronouncing
| on that, but that's what Alexander is mostly writing about.

I don't know if that means he was looking for a habitat evaluation
gene, though.  The way I read it, he was talking about a kind
of thinking.  It works because we have a biological basis for
sound judgement, but not specifically about buildings, we just
have a deep knowledge of the world we live in and don't do well
to defer to architects to tell us what will work for us, based
on their diagrams or their trained esthetics.

I certainly would not want to argue with an architect who is talking
about what will hold up to an earthquake, but an architect's ideas
about form and function are no substitute for the shared sense of an
evolved cultural landscape.

I don't know about his pattern language.  Like the other ideas
Alexander had along those lines - user on-site design, some kind of
weird construction techniques - he was looking for a route from
here to there, from modern western building and land use to the
kind of vernacular tradition that can use that shared sense.
It has been too long since I read the book, but I don't think
he meant the patterns to be adopted as that vernacular tradition,
they were more like crutches until we could walk on our own.

Anyway, as long as we're talking about things that we spend our
days doing, we certainly do have the capacity for sound judgement.
It doesn't mean our judgement is therefore infallibly sound, but
the collective judgement of an established culture brings a kind
of sense to bear that can dwarf a brilliant appeal to abstract
principles.

	Donn Cave, donn at u.washington.edu


From daniel.dittmar at sap.com  Wed Mar 28 03:30:52 2001
From: daniel.dittmar at sap.com (Daniel Dittmar)
Date: Wed, 28 Mar 2001 10:30:52 +0200
Subject: Where is the DBI module?
References: <MPG.152ab2bb88e6482d9896a4@news.pacbell.net>
Message-ID: <99s7ft$hjt$1@news1.wdf.sap-ag.de>

> I can "import DBI" but when I try to find the actual module in order to
> examine it I can't. Am I missing something obvious?

General solution:
[0] d:\ >python
Python 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on win32
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import dbi
>>> dbi
<module 'dbi' from 'D:\SAPDevelop\Devtool\Python\win32\dbi.dll'>
>>>

--
Daniel Dittmar
daniel.dittmar at sap.com
SAP DB, SAP Labs Berlin
http://www.sapdb.org/





From tuttledon at hotmail.com  Sat Mar  3 22:37:12 2001
From: tuttledon at hotmail.com (Don Tuttle)
Date: Sun, 04 Mar 2001 03:37:12 GMT
Subject: [Python-Dev] Re: Bug fix releases
References: <mailman.983649366.2475.python-list@python.org>
Message-ID: <ILio6.29087$Z8.5197836@typhoon.southeast.rr.com>

"Tim Peters
> Note there was never a bugfix release for 1.5.2, despite that 1.5.2 had
some
> serious bugs, and that 1.5.2 was current for an unprecedentedly long time.
> Guido put out a call for volunteers to produce a 1.5.2 bugfix release, but
> nobody responded.  Past is prelude ...
>
> everyone-is-generous-with-everyone-else's-time-ly y'rs  - tim

But there is the CVS. And there are those like myself  who are CVS impaired.
Right now there are Fredric's SRE and _tkinter on steroids updates, and
Mark's improved COM support.  I'd love to have these. (Alright I've got
/F's. He compiled 'em.  But Mark's is still in the damn CVS.)  I've prodded
ActiveState to issue Python updates and they've indicated they would--they
just haven't decided when.

I don't think it has to be as formal as an official  pep or separate
development track.  Just whenever our fearless dictator sees that there's
enough good stuff in the CVS, let him decree that a new build shall be made.

Of course we've gotta sell the idea to the dictator.  Nobody's lost their
head trying to do that (gulp) have they?

Don





From cognoscente at rediffmail.com  Thu Mar 22 23:58:09 2001
From: cognoscente at rediffmail.com (prashant  patel)
Date: 23 Mar 2001 04:58:09 -0000
Subject: Ruby will overtake Python within four years.
Message-ID: <20010323045809.9369.qmail@mailweb22.rediffmail.com>

No I am not a ruby programmer not even using any scripting language other once used vbscript in ASP.
my major programming language has been java. But with the experience in java, I want learn a scripting language which increases my productivity, java is simply to redundant.
for example : doing an IO required to use so many classes.
for the past one month i have been find a alternative, 
and found three scripting language python,perl,and ruby.
for now python is my choice, perl is surelty a no for its syntax.but how about python and ruby, as i would be 
allocating a lot of time learing language the particular language, could some one give out facts about ruby.
I would be unjustified actually to ask python lover's the good and bad of ruby compared to python, but these discusions will help us also evaluate python.
And yes the subject line and be taken from. 
http://www.amazon.com/exec/obidos/tg/feature/-/99126/102-2233024-4095364
But my inclination are towards python.



_____________________________________________________
Chat with your friends as soon as they come online. Get Rediff Bol at
http://bol.rediff.com







From aleaxit at yahoo.com  Wed Mar  7 05:36:03 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Wed, 7 Mar 2001 11:36:03 +0100
Subject: ScriptControl (was Re: Can Python replace Visual Basic? Should it?...and generalities : ))
References: <mailman.983905329.22236.python-list@python.org> <2Ybp6.12845$hn5.1818206@news1.rdc1.mb.home.com>
Message-ID: <9852ui0166a@news1.newsguy.com>

"Brad Bollenbach" <bbollenbach at homenospam.com> wrote in message
news:2Ybp6.12845$hn5.1818206 at news1.rdc1.mb.home.com...
> This reply is interesting, but still contains a lot of the "I've started
> invistigating Python and these are the things I like about it" comments,
> which I was hoping to avoid.
>
> [snip]
>
> >   A Python "OCX" control needs to be written.  This control
> >   would simply encapsulate Python for us Visual Basic idiots.
> >   This way we could 'pop in the control in our VB project and
> >   start using Python for parts of the program.
>
> That's a neat idea. I hope someone smarter than me decides to do it. :)

Microsoft has done it!-)

They cleverly call it the ScriptControl, so it won't be immediately
obvious that its real purpose is to let you embed Python in your VB
programs without breaking a sweat; and, accordingly, you have to
explicitly set its Language property to "Python" (but that, no
doubt, is in pursuance of Python's "explicit is better than implicit"
philosophy -- bet you never expected THAT one from MS, hm?-).

You also have to have the ActiveState build of Python installed
(or, install the standard build, then go to ActiveState anyway
to pick up the win32all extensions and install those on top of
the standard Python distribution).


> As, you mentioned in a reply to the other thread of this thread (heh),
> coding really is the most insignificant part of a large programming
project
> in a coporate environment. In theory at least, you should have a clear
idea
> of what you're going to do before you do it, in which case the coding
> (almost irregardless of the language) is often times just grunt work.

The *maintenance* part of the program's lifetime costs is paramount,
and that involves LOTS of fiddling with code too -- therefore, the
claim that "coding ... is insignificant" (while not absurd if only
applied to the INITIAL coding) is quite misleading.  That code will
have to be read/understood/fixed/modified a thousand times...!


Alex





From tjg at exceptionalminds.com  Fri Mar  9 12:59:33 2001
From: tjg at exceptionalminds.com (Timothy Grant)
Date: Fri, 9 Mar 2001 09:59:33 -0800
Subject: Divorcing Command Line and GUI
Message-ID: <20010309095933.T3579@trufflehunter.avalongroup.net>

Hello Everyone.

This is something of a strange question, but please bear with
me.

I have a series of tools that I have created that work from the
command line. I have now been tasked with creating GUIs for
those tools. The problem I'm faced with is that the command
line tools and the GUI tools will probably be licensed
differently (e.g, command line will be Open Source, GUI will be
something undetermined as of yet.) This means that I can't put
my GUI code in with my CL code.

My first cut at this demarcation sort of worked, but was also
troublesome. So while I have some new ideas, I'm looking to see
if anyone else in the python community has ever had to do
something like this, and if there are any easily maintainable
solutions to this problem.



-- 
Stand Fast,
    tjg.

Timothy Grant                         tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.         <><       (503) 246-3630
>>>>>>>>>>>>>Linux, because rebooting is *NOT* normal<<<<<<<<<
>>>>This machine was last rebooted:  51 days 22:09 hours ago<<



From dave at see.sig.for.address  Wed Mar 28 13:55:48 2001
From: dave at see.sig.for.address (Dave)
Date: Wed, 28 Mar 2001 18:55:48 GMT
Subject: ANNOUNCE: Python Mock Objects
Message-ID: <5oc4cto5ec8iqvkdflaivv52805dn9rfdd@4ax.com>

[This is a repost, since the original seems to have got lost between
me and my ISP. My appologies if you receive it twice.]


I have released to the Python and XP communities a small Python module
for creating mock objects.  A mock object can be passed to a class or
method being tested to replace objects that are outside the scope of
the test. 

The file has been uploaded to the Extreme Programming mailing list on
yahoo groups, and may be downloaded from:
http://groups.yahoo.com/group/extremeprogramming/files/PythonMock.zip 

The idea is simple.  A mock object created with this class will allow
any method to be invoked on it, with any parameters (including keyword
arguments).  The method calls are stored and can be examined later to
see what parameters were passed.  By default all methods return None,
but the class constructor can take a dictionary of method names and
values to return for those methods.

The zip file contains documentation and a test suite using OmPyUnit,
which should easily be convertible to the other versions of PyUnit.
OmPyUnit can be downloaded from
http://www.xprogramming.com/software.htm or from
http://www.objectmentor.com/freeware/downloads.html.

If anyone finds this useful or converts it to other languages then
please let me know.  I believe Smalltalk or Ruby should be easy, but
if
anyone figures out how to do this in Java or C++ then please let
*everyone* know.  8^)


Share and Enjoy

        Dave Kirby

--------------------------------------------------
All great ideas start as heresy and end as dogma.

dkirby@   <-figure this out, spambots!
bigfoot.   My opinions are my own,
com        but I'm willing to share.


From tim.one at home.com  Sun Mar 11 23:35:31 2001
From: tim.one at home.com (Tim Peters)
Date: Sun, 11 Mar 2001 23:35:31 -0500
Subject: Static typing (was Re: Java guy interested in Python)
In-Reply-To: <98glhc02sga@news2.newsguy.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCCEPKJEAA.tim.one@home.com>

[Rainer Deyke]
> When I type 'int i;' in C++, what I really mean is "'i' behaves like an
> integer variable.".

[Tim]
> Since that's all "int i;" *can* mean in C++, seems tautological <wink>.

[Alex Martelli]
> Surely you're joking, Mister Peters?
>
>     class int {
>         public:
>             int(const char*=0);
>             ~int();
>             const char* data() const;
>     };
>
>     int i;
>
> Isn't this perfectly legal C++, meaning something very different from
> "i behaves like an integer variable"...?  int is no "reserved word",
> after all.

As in the "Einstein's Riddle" thread, I was doing Rainer the favor of
assuming he was trying to say something <ahem>.  If I don't extend you the
same favor, then before I can answer your question I have to be assured of a
great many things, such as that e.g. this fragment isn't in the scope of

#define char const

or

#define const (

or etc <wink>.

there's-more-than-meets-the-eye-when-there's-more-than-meets-the-eye-ly
     y'rs  - tim




From menscher+python at uiuc.edu  Thu Mar  8 01:20:43 2001
From: menscher+python at uiuc.edu (Damian Menscher)
Date: Thu, 08 Mar 2001 06:20:43 GMT
Subject: grabbing return codes from os.system() call
Message-ID: <%wFp6.9481$dL4.126404@vixen.cso.uiuc.edu>

I'm new to Python (as in, my experience is essentially the tutorial),
but I've already come up with an interesting question:

How do I get the return code from an os.system call?  I would have
expected I could do something like

---returncode---
#/bin/csh
echo do stuff
exit 3

and then in my python program I could do

print os.system('./returncode')

But it prints out 768.  Not particularly useful, even after I recognize
the trick of dividing by 256 (byte-swapping going on?  No, because a
return code of 768 reports as 0).  Given that my real return codes
will be (possibly large) integers, this limitation will likely cause
some serious problems down the line.

Even better would be a way of returning a string (the script I run
can be something other than csh, but it has to be a separate script).

Ideas?  I'm trying to avoid writing the string out to a file and then
read the file back in to the python program....

Damian Menscher
-- 
--==## Grad. student & Sys. Admin. @ U. Illinois at Urbana-Champaign ##==--
--==## <menscher at uiuc.edu> www.uiuc.edu/~menscher/ Ofc:(217)333-0038 ##==--
--==## Physics Dept, 1110 W Green, Urbana IL 61801 Fax:(217)333-9819 ##==--


From andreas at andreas-jung.com  Thu Mar 15 08:25:00 2001
From: andreas at andreas-jung.com (Andreas Jung)
Date: Thu, 15 Mar 2001 08:25:00 -0500
Subject: Help doing simple threading
References: <mailman.984574578.29018.python-list@python.org> <3AB041C6.FDDB84D3@theworld.com>
Message-ID: <98qg4q$f4c$1@slb2.atl.mindspring.net>

"David Lees" <deblNonOspammy at theworld.com> schrieb im Newsbeitrag
news:3AB041C6.FDDB84D3 at theworld.com...
> Thanks, these comment were quite helpful.  The most serious error was
> 'a)'.  when I just put the main to sleep after starting up the threads
> it worked fine.  The problem was that I was printing out the array
> values, before all the threads finished executing.

Using sleep() is also dangerous. This is not the way to synchronize threads.
The threading module offers you several possibilites (e.g. join() method)
for
a clean, working solution.

Andreas




From park997 at aol.comnotospam  Tue Mar 13 09:53:39 2001
From: park997 at aol.comnotospam (Park997)
Date: 13 Mar 2001 14:53:39 GMT
Subject: main in mac?
Message-ID: <20010313095339.19264.00000243@ng-dh1.aol.com>

Hi,

I am using Python 2.0 on a PPC mac. I have a module
that is set up as follows:

from Numeric import *

class Epower:

	3 class functions defined

if __name__ == '__main__':

	test class and print output

This code does nothing if the run all button is pressed
or if it is imported into interactive python.

when the above if statement is replaced by:

if 1:

the program runs properly with output. I thought that
if __name__ == '__main__':
is standard syntax. Does this need to be changed for mac
Python?

Thanks,

Wendell Cropper


From stadt at cs.utwente.nl  Sat Mar  3 21:01:11 2001
From: stadt at cs.utwente.nl (Richard van de Stadt)
Date: Sun, 04 Mar 2001 03:01:11 +0100
Subject: multipart/form-data submission with httplib + MimeWriter?
References: <mailman.983531043.10150.python-list@python.org>
Message-ID: <3AA1A1E7.1360139A@cs.utwente.nl>

Steve Purcell wrote:
> 
> Wise ones,
> 
> Can anyone point me towards some example code for performing
> 'multipart/form-data' submissions of form values and file contents
> using httplib and MimeWriter?
> 

I'm not sure what you mean.

Are you asking about 
 - how to *fill in* such webforms automatically, or
 - how to create a form with which a file can be uploaded, or
 - how to deal with the contents of such a form when submitted?

Richard.


From bbhaydon at bigpond.com  Sat Mar 10 11:17:09 2001
From: bbhaydon at bigpond.com (Brett Haydon)
Date: Sun, 11 Mar 2001 03:17:09 +1100
Subject: cursor.execute fails under mxODBC
References: <mailman.983993842.26448.python-list@python.org> <tUzp6.29925$ML1.1507874@e420r-atl2.usenetserver.com> <3AA7FF0E.50E6D70F@northwestern.edu>
Message-ID: <Oesq6.12063$v5.52124@newsfeeds.bigpond.com>

Hmmm looks like the new version is broken.

I am running mxODBC 1.1.1 no problems with both MySQL and Access
for the last 6 months on W2k.

I just tried the new version also with exactly the same problems.

The old one still works fine though..

regards,

bbhaydon
"Louis Luangkesorn" <lluang at northwestern.edu> wrote in message
news:3AA7FF0E.50E6D70F at northwestern.edu...
> I have the same problem.  Below is my output.  I will first use the odbc
> that comes included with ActivePython, then the mxODBC module  (by the
way,
> I've tried to rename the odbc.pyc file, that did not help, except break
> odbc)
>
> PythonWin 2.0 (#8, Oct 19 2000, 11:30:05) [MSC 32 bit (Intel)] on win32.
> Portions Copyright 1994-2000 Mark Hammond (MarkH at ActiveState.com) - see
> 'Help/About PythonWin' for further copyright information.
> >>>
> >>> import dbi,odbc
> >>> import pprint
> >>> channel = odbc.odbc('sqltutor')
> >>> ch1 = channel.cursor()
> >>> ch1.execute('SELECT FirstName, LastName, Address, City, State FROM
> EmployeeAddressTable')
> 0
> >>> test1 = ch1.fetchall()
> >>> pprint.pprint(test1)
> [('Joe', 'Smith', '83 First Street', 'Howard', 'Ohio'),
>  ('Mary', 'Scott ', '842 Vine Ave.', 'Losantiville', 'Ohio '),
>  ('Sam', 'Jones ', '33 Elm St.', 'Paris', 'New York '),
>  ('Sarah', 'Ackerman ', '440 U.S. 110', 'Upton', 'Michigan ')]
> >>> ch1.close()
> >>> channel.close()
> >>> import mx.ODBC
> >>> channel = mx.ODBC.Windows.connect('sqltutor')
> >>> ch1 = channel.cursor()
> >>> ch1.execute('SELECT FirstName, LastName, Address, City, State FROM
> EmployeeAddressTable')
>
> at this point the whole thing dies.  No error messages.  I didn't see any
> log files anywhere either.
>
> Louis
>
> --
> K Louis Luangkesorn
> lluang at northwestern.edu  http://pubweb.nwu.edu/~kll560  PGP:0xF3E2D362
> Whatsoever things are true, ... honest, ... just, ... pure, ... lovely,
...
> of good report; if there be any virtue, and if there be any praise, think
on
> these things.- motto - Northwestern University
>
>




From hamish_lawson at yahoo.co.uk  Tue Mar 20 12:31:01 2001
From: hamish_lawson at yahoo.co.uk (=?iso-8859-1?q?Hamish=20Lawson?=)
Date: Tue, 20 Mar 2001 17:31:01 +0000 (GMT)
Subject: Should __future__ statements ever be phased out?
Message-ID: <20010320173101.21650.qmail@web217.mail.yahoo.com>

The __future__ proposal envisages that future statements in a piece of
code will be phased out once the language change becomes standard.
(Many of the alternative proposals also envisage this.) Does this not
pose possible problems with running post-change-aware code on an
pre-change version of the interpreter?

Let me take a fanciful (and hopefully highly improbable!) example.
Suppose in Python 4.0 it is decided to introduce a change in the
semantics of the print statement so that it outputs its arguments in
uppercase. This change is inrended to become mandatory in Python 3.1
and those wanting the old behaviour after that will have to use write
statements. During the 4.0-to-4.1 transition I use a future statement
in my code to specify whether I want the old or the forthcoming
behaviour. After 4.1 comes out and I upgrade my interpreter, I phase
out all the print-related future statements from my code. However one
day I give my program to a friend who (unknown to me) is still running
Python 2.1. When he runs my program any print statements in it will
behave differently from what I intended, but without any warning.

Hamish Lawson


____________________________________________________________
Do You Yahoo!?
Get your free @yahoo.co.uk address at http://mail.yahoo.co.uk
or your free @yahoo.ie address at http://mail.yahoo.ie



From erno-news at erno.iki.fi  Fri Mar 30 18:55:30 2001
From: erno-news at erno.iki.fi (Erno Kuusela)
Date: 31 Mar 2001 02:55:30 +0300
Subject: best practice? null/None vs 'None' vs '' (in web/sql)
References: <mailman.985968951.17045.python-list@python.org>
Message-ID: <kuvgoqu80d.fsf@lasipalatsi.fi>

In article <mailman.985968951.17045.python-list at python.org>,
fluxent at yahoo.com writes:

| I can see that the ODBC result set distinguishes them. But of course 
| when things get rendered out to HTML you don't want to see single 
| quotes around every string value, so 'None' and None become the same 
| thing.

if you are dynamically generating the form, you can omit the fields
that you want to correspond to null. that is conceptually the same
thing as null. (if you are using the cgi module, reember to pass
keep_empty_fields=1 - it defaults to 0 for some reason)

  -- erno


From erwin at andreasen.com  Sat Mar 24 07:23:03 2001
From: erwin at andreasen.com (Erwin S. Andreasen)
Date: Sat, 24 Mar 2001 12:23:03 +0000 (UTC)
Subject: close(), exceptions and problems
References: <slrn9bm5u5.klj.kuntsi@myntti.helsinki.fi>
Message-ID: <99i3j7$lqr$1@news.cybercity.dk>

On 23 Mar 2001 09:30:46 GMT, Antti Kuntsi wrote:

>I have run into the following problem with python and close()s. When the
>system is under a heavy load the close() system call can be interrupted,
>but the Python never detects this. This leads into a lot of problems,
>including at least the following:

What exactly do you mean by "interrupted" ?

close(), the system call, usually cannot fail (unless you try to close a fd
that is not open).

If a system call is interrupted by a signal, it is restarted automatically
(unless the signal handler was setup with SA_NORESTART *or* the system call in
question is select(), which is never automatically restarted).

But if you call close(fd) you aren't supposed to do something like:

int fd, rv;
while (((rv = close(fd)) < 0 && errno == EINTR))
	;
if (rv < 0)
	perror("close");


the system call will be automatically restarted for you.


-- 
=======================================================================
<erwin at andreasen.com>           Herlev, Denmark       Software Designer
<URL:http://www.andreasen.org/>       <*>         LASAT^WEicon Networks
=======================================================================


From paulp at ActiveState.com  Wed Mar 14 09:55:38 2001
From: paulp at ActiveState.com (Paul Prescod)
Date: Wed, 14 Mar 2001 06:55:38 -0800
Subject: [Python-Dev] Revive the types sig?
References: <jeremy@alum.mit.edu> <15020.9404.557943.164934@w221.z064000254.bwi-md.dsl.cnc.net> <200103120711.AAA09711@localhost.localdomain> <mailman.984439040.24419.python-list@python.org> <s44tatolisnipdfm4r4p1aqt2e09e458hk@4ax.com>
Message-ID: <3AAF866A.968411B8@ActiveState.com>

Paul Moore wrote:
> 
>...
> 
> My view on all this is that type annotations are likely to be counterproductive,
> in the face of Python's high degree of polymorphism. For example, what is the
> type of
> 
> def fac(n):
>     if n <= 1: return 1
>     return n * fac(n-1)
> 
> ?
> 
> Obvious answer, integer -> integer. And that's likely what the programmer would
> say.
> 
> But why can't this function be used with *any* type which supports comparison
> with the number 1, multiplication, and subtraction of 1?

Sure it can. But as a programmer, I probably haven't really thought
through the meaning of this code for users who pass in strings or lists
of SocketObjects. Therefore I should not be blamed when someone passes
in a random type and gets strange behavior.

Anyhow, this kind of totally polymorphic code is exceedingly rare. If
you had done division in there, floats would work radically differently
than integers. It isn't safe to say that just because the right
operators are defined, they mean the right thing.

Type annotations of any type are a known trade-off between raw
flexibility, safety and error reporting. C++ and ML are way on one end
of the spectrum. Python is way on the other end. There is a middle
ground.

-- 
Python:
    Programming the way
    Guido
    indented it.
       - (originated with Skip Montanaro?)



From hzhu at users.sourceforge.net  Mon Mar 19 22:40:01 2001
From: hzhu at users.sourceforge.net (Huaiyu Zhu)
Date: Tue, 20 Mar 2001 03:40:01 GMT
Subject: So what's wrong with __future__?  (Was Re: Why "from __future__" stinks ...)
References: <AD70ED184AFED311ADFD009027AC917F83617A@dmc1nt.zoho.com> <mailman.984966923.14984.python-list@python.org>
Message-ID: <slrn9bdka2.3j6.hzhu@rocket.knowledgetrack.com>

I don't quite see what the complaint about __future__ is really about.  So
let me try to analyze a little bit.  Maybe this could encourage someone to
write a PEP to clarify.

What's wrong with import __future__?

1.  It is ugly.

Does this apply to syntax or semantics?

Syntax. 

But how?  It looks to me to be good on two accounts:
1. It stands out clearly as saying "magic".
2. It does not introduce new syntax or keyword.  Just a new module name.
How much better can you get if you want both?

Semantics.  It does not SHOUT "magic". A new keyword should be cleaner.

Not so.  Python __magic__ words have (at least) two effects:
1. automatic, like __init__.
2. side effect, like __setattr__, __builtin__.
The only magic of __future__ is side effects.

Does its semantics go beyond a regular module?  Well, yes.  But so what?  If
you redefine __setattr__, is its effect confined to itself?  If you redefine
operator.__add__, is its effect only felt when you call it explicitly?
Clearly you can complain that __setattr__ is not like any other method.  But
what would it look like if you introduce a new keyword in its place?


2. Future is the wrong word, because it concerns time.  A keyword like
__incompatible__ would convey meaning better.

Well, I can see both ways of thinking:

If you consider any given moment in time, its effect is indeed incompatible
change.  It's whether you want or don't want this change.

But if you consider the process of change, its effect is merely when a
change happens.  Your choice is only whether you want it now or in the
future.  (Maybe from __preview__ import *?)

I think Tim's intend is that anything that goes into __future__ will be
mandatory in some future version.  


3. It will introduce confusion about versions. 

This complaint will hold true no matter how you spell it.  The version info
in the __future__ module seems to minimize this problem.  In fact, if you
adopt the view that it is not an optional feature mechanism, but a time
travel mechanism, then the mechanism built in the PEP should be sufficient.

Now that's my rant about the complaints about __future__.  I'm even less
certain about what the proposed alternatives look like, and how they address
these problems.  Hopefully some proponents of the alternatives could write a
PEP along these lines?  I believe you will find the truth in Tim's comment
that any concrete proposal will attract lots of criticism.

If you are saying, __future__ is not good, but I do not have anything
substantially better, well then, what can be done?

Huaiyu


From deblNo at spammytheworld.com  Fri Mar 30 00:27:36 2001
From: deblNo at spammytheworld.com (David Lees)
Date: Fri, 30 Mar 2001 05:27:36 GMT
Subject: Looking for Python equivalent to a Perl'ism
References: <3AC375A7.A4BA87F1@spammraqia.com> <3d3dbwwbk1.fsf@ute.cnri.reston.va.us>
Message-ID: <3AC41952.549DCA85@spammytheworld.com>

Thanks for the solution.  Also, this is a big help to my understanding
the 'sub' section of the documentation at: 
http://www.python.org/doc/current/lib/Contents_of_Module_re.html

david lees

Andrew Kuchling wrote:
<snip>
> 
> To simply substitute the contents of a single group, you can
> include \N, or \g<N>:
> 
> p=re.compile(r'zz(\w+)zz')
> s='this is and so is zzxxxzz abc that an zzyyyzz abc zzxxxzz or else'
> print p.sub('\g<1>', s)
> 
> This produces "this is and so is xxx abc that an yyy abc xxx or else"
> You want to do a dictionary lookup, though; instead of passing a
> replacement string, you can pass a function which will be handed the
> match object, and can return whatever string it likes:
> 
> def sub_func(match):
>     return dict[ match.group(1) ]
> 
> print p.sub(sub_func, s)
> 
<snip>


From f8dy at diveintopython.org  Fri Mar 16 19:30:53 2001
From: f8dy at diveintopython.org (Mark Pilgrim)
Date: Fri, 16 Mar 2001 19:30:53 -0500
Subject: HTMLparsing abnormal html pages
References: <98pvp1$15t$1@news.netmar.com> <98ua8j$a6h$1@panix2.panix.com>
Message-ID: <B6D81A6D.4FE6%f8dy@diveintopython.org>

in article 98ua8j$a6h$1 at panix2.panix.com, Aahz Maruch at aahz at panix.com
wrote on 3/16/01 7:14 PM:

> In article <98pvp1$15t$1 at news.netmar.com>,  <asle at spam.com> wrote:
>> 
>> Considering the small program below. Running it will show that the
>> HTMLparser
>> is truncating urls in the HTML page. Now, most of you will probably say that
>> the page and in particular the URL's of this page are not valid according to
>> the RFC1738 protocol --bad luck. But there must be a work-around for this?
> 
> For this specific case, Mark's solution may well work (haven't tested it
> myself).  But you cannot easily find a generic solution because of all
> the different ways to mangle HTML.

My solution does work, (only) because the page the original poster was
trying to parse had unquoted attribute values, like <a href=index.html>.
sgmllib works fine in these cases.

In fact, the BaseHTMLProcessor class I define in my book can be used to
properly quote all attribute values, since it works by breaking down the
entire HTML (via sgmllib) and building up equivalent HTML with proper quotes
around attribute values.  This is decidedly *not* why it was written; I just
happened to notice it one day when I was testing the code for other reasons.
  http://diveintopython.org/dialect_divein.html

-M
You're smart; why haven't you learned Python yet?
http://diveintopython.org/





From kens at sightreader.com  Tue Mar 20 14:24:36 2001
From: kens at sightreader.com (Ken Seehof)
Date: Tue, 20 Mar 2001 11:24:36 -0800
Subject: Insertin **keywords into a class
References: <mailman.984971990.29857.python-list@python.org> <9960i1$tp0$1@nntp6.u.washington.edu> <99661m$scc$1@panix6.panix.com> <99877b$uss$1@nntp6.u.washington.edu> <Pine.GHP.4.21.0103201151090.4991-100000@don.et.byu.edu>
Message-ID: <006701c0b173$6bb5dd10$04090a0a@upcast.com>

Use setattr().

You are confusing an attribute with an attribute name (a string).

Each time through the loop you are assigning to the attribute 'kw' rather
than
the attribute whose name is the value of kw.

def __init__(self, *arguments, **keywords):
      self.radius = 0.0
      for kw in keywords.keys():
         print kw, ":", keywords[kw]
         setattr(self, kw, keywords[kw])
      print "self.radius =", self.radius

or simply...

def __init__(self, *arguments, **keywords):
      self.__dict__.update(keywords)

----- Original Message -----
From: "C. Porter Bassett" <porter at et.byu.edu>
Newsgroups: comp.lang.python
To: <kens at sightreader.com>
Cc: <python-list at python.org>
Sent: Tuesday, March 20, 2001 10:53 AM
Subject: Insertin **keywords into a class


> cCan you please tell me why the following code snippet doesn't update the
> value of radius?
>
>
> class myClass:
>    def __init__(self, *arguments, **keywords):
>       self.radius = 0.0
>       for kw in keywords.keys():
>          print kw, ":", keywords[kw]
>          self.kw = keywords[kw]
>       print "self.radius =", self.radius
>
>
> b = myClass(radius = 1.0)
>
>
> --------------------------------------------------------------------------
> "Pretend like this is a really witty saying." - Anonymous
> --------------------------------------------------------------------------
>
>
>
> --
> http://mail.python.org/mailman/listinfo/python-list
>




From fredrik at pythonware.com  Thu Mar  1 01:26:58 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Thu, 01 Mar 2001 06:26:58 GMT
Subject: Intersesting Questions involving extending and embedding
References: <3a9c6f89.9613214@news.digital.net> <0D1n6.15927$Qb7.2600092@newsb.telia.net> <3a9f0af7.37387355@news.digital.net> <508n6.15966$Qb7.2607706@newsb.telia.net> <3a9dc27e.1325053@news.digital.net>
Message-ID: <SYln6.16080$Qb7.2629142@newsb.telia.net>

"Dev_NuLL" wrote:
> > > The writers of Blender modules release the modules compiled to protect
> > > the source code, and that is what I am also interested in doing.
> >
> >compiled as in DLL/SO, or compiled as in PYC?
>
> from the original message:
>
>  How would I go about creating a Windows DLL or Linux
> .so that would do this?

from your original message:

> [a piece of python code]

> I would like to create a primitives module to create 3D solids I would
> like to wrap the above code (or similar) into a module and call it
> 'Primitives' with a 'plane' function [...]

to which the answer is to use a PY or PYC file, like everyone
else.

if you insist on wrapping the PYCs in DLL/SO files (as if that
would make them harder to reverse engineer), search the net
for python2c [1], or study the freeze sources [2]

Cheers /F

1) last seen here: http://lima.mudlib.org/~rassilon/p2c
2) in the source distribution




From mwh21 at cam.ac.uk  Thu Mar 29 17:02:46 2001
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 29 Mar 2001 23:02:46 +0100
Subject: Python assembler
References: <mailman.985799199.7319.python-list@python.org>
Message-ID: <m3d7b0cjy1.fsf@atrus.jesus.cam.ac.uk>

"Mike C. Fletcher" <mcfletch at home.com> writes:

> I _think_ (not really sure, as never used it) that bytecodehacks
> provides an "assembler" language for Python.  Project is at:
> http://bytecodehacks.sourceforge.net/

It's a bit of a joke, and the bytecodehacks are somewhat out of date
(sorta work with Python 2.0, probably die horribly with 2.1).  I
haven't really had any motivation to hack on the bytecodehacks lately.
I might get another rush of enthusiasm at some point, but don't hold
your breath.

Cheers,
M.

-- 
  I'm sorry, was my bias showing again? :-)
                                      -- William Tanksley, 13 May 2000


From cce at clarkevans.com  Mon Mar 26 05:18:47 2001
From: cce at clarkevans.com (Clark C. Evans)
Date: Mon, 26 Mar 2001 05:18:47 -0500 (EST)
Subject: win32com:MakePy -- Early Binding without the size jitters?
In-Reply-To: <Pine.LNX.4.21.0103260458020.19297-100000@clarkevans.com>
Message-ID: <Pine.LNX.4.21.0103260515040.19297-100000@clarkevans.com>

On Mon, 26 Mar 2001, Clark C. Evans wrote:
> 
>       doc = msxml.DOMDocument.default_interface(
>                 pythoncom.CoCreateInstance( 
>                     "MSXML2.DOMDocument",None, 
>                      pythoncom.CLSCTX_INPROC,
>                      pythoncom.IID_IDispatch))
> 
...

>    import msxml3
>    doc = msxml3.DOMDocument()

*blush*   I was working too hard, as the following is all I need:

  msxml.DOMDocument.default_interface(msxml.DOMDocument())

I was wondering if the wrapper generator could be fixed
to do this automatically?   For some reason, the __init__
below isn't doing it... go figure.

   class CoClassBaseClass:
       def __init__(self, oobj=None):
          if oobj is None: oobj = pythoncom.new(self.CLSID)
          self.__dict__["_dispobj_"] = self.default_interface(oobj)

...

Anyway, thank you for such an Excelent product!  Wow!  I can 
finally *DITCH* Visual Basic for "object scripting" ... I think.

Kind Regards,

;) Clark






From careye at spamcop.net  Wed Mar 28 02:16:33 2001
From: careye at spamcop.net (Carey Evans)
Date: 28 Mar 2001 19:16:33 +1200
Subject: large arbitrary binary data in python
References: <slrn9c1hf7.m5q.wcuddy@lserv.ja10629.home>
Message-ID: <87wv9ajrce.fsf@psyche.dnsalias.org>

[...]

>     3) pipe the data to a shell command like this "gunzip -dc | dd of=device"

As well as using the md5 module, as Steve has pointed out, you can do
all this in Python too, using the gzip module to decompress the data.

Instead if dd, you can use just a standard file object to output it.
If there is some need for blocks of a certain size, and the file
write() method doesn't cut it, use the os.open() and os.write()
functions.

>     How will python handle the large amount of binary data?  Is there a limit to
>     the size of a binary string in python?

The various string object functions in Python take the length as an
int, so you can fit however much that can measure on your platform.
If int is 32 bits, that's two gigabytes of data.

-- 
	 Carey Evans  http://home.clear.net.nz/pages/c.evans/

	    "Quiet, you'll miss the humorous conclusion."


From aahz at panix.com  Fri Mar  9 21:49:56 2001
From: aahz at panix.com (Aahz Maruch)
Date: 9 Mar 2001 18:49:56 -0800
Subject: Python In A Nutshell
References: <nG5q6.54910$__6.11376601@typhoon.southeast.rr.com> <Ufcq6.140786$47.2078615@news.bc.tac.net> <Blfq6.60716$__6.11886688@typhoon.southeast.rr.com>
Message-ID: <98c4ok$fhd$1@panix2.panix.com>

In article <Blfq6.60716$__6.11886688 at typhoon.southeast.rr.com>,
Don Tuttle <tuttledon at hotmail.com> wrote:
>
>Planning stages?  We've all seen his posts (a.k.a. treatises) here.  I'd say
>he should have the final draft to you in 3 weeks tops. ;-)

ITYM "first draft".  It will take another three-to-six months for Alex
to edit it down.  ;-)
-- 
                      --- Aahz (Copyright 2001 by aahz at pobox.com)

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

"..., and some of you may regard all women as evil traps that exist only
to tease, torture, and suck out your very soul."  --DrMax


From stephen_purcell at yahoo.com  Sun Mar 18 04:52:58 2001
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Sun, 18 Mar 2001 10:52:58 +0100
Subject: HTMLparsing abnormal html pages
In-Reply-To: <6ej8btssj4tf5t14gca2ge4dt6bdbajbtc@4ax.com>; from timr@probo.com on Sat, Mar 17, 2001 at 09:58:10PM -0800
References: <98pvp1$15t$1@news.netmar.com> <98ua8j$a6h$1@panix2.panix.com> <B6D81A6D.4FE6%f8dy@diveintopython.org> <6ej8btssj4tf5t14gca2ge4dt6bdbajbtc@4ax.com>
Message-ID: <20010318105258.A7546@freedom.puma-ag.com>

Tim Roberts wrote:
> I have been searching for an HTML pretty-printer; something where I can
> feed an arbitrary page and get a more structured, indented view.  I wrote a
> simple one myself, based on sgmllib; it does a fair job, but it is easily
> confused by such common offenses as omitted </p> tags.  It sounds like your
> BaseHTMLProcessor might be such a thing.  Is it available yet?
> 
> If not, is anybody aware of a fair HTML cleaner-upper?


You could use the python-xml code to slurp the HTML into a DOM, and then
format it using HtmlLineariser:

>>> from xml.dom.writer import HtmlLineariser
>>> from xml.dom.html_builder import HtmlBuilder
>>> builder = HtmlBuilder()
>>> builder.ignore_mismatched_end_tags = 1   # make less fussy 
>>> html_text = open('public_html/index.html').read()
>>> builder.feed(html_text)
>>> pretty_printed = HtmlLineariser().linearise(builder.document)

pretty_printed is now a nicely indented version of html_text.

It's not the fastest thing in the world, but it might help you.

-Steve

-- 
Steve Purcell, Pythangelist
Get testing at http://pyunit.sourceforge.net/
Any opinions expressed herein are my own and not necessarily those of Yahoo



From gregj at pobox.com  Thu Mar 22 04:52:01 2001
From: gregj at pobox.com (Greg Jorgensen)
Date: Thu, 22 Mar 2001 09:52:01 GMT
Subject: Good Editor? 
References: <3ab9a32b$1_1@news2.one.net>
Message-ID: <5Xju6.601501$U46.18300887@news1.sttls1.wa.home.com>

Most of my Python programs are so small that IDLE works for me. My favorite
editors are:

Windows: EditPlus (www.editplus.com)
Mac: BBEdit
*nix: vi, the One True Editor

"fjs" <fjs6 at usa.net> wrote in message news:3ab9a32b$1_1 at news2.one.net...
> Can someone recomend a good editor to write python in?  Im just getting
> started learning python and dont know which ones support python.





From just at letterror.com  Tue Mar 13 10:18:28 2001
From: just at letterror.com (Just van Rossum)
Date: Tue, 13 Mar 2001 16:18:28 +0100
Subject: main in mac?
References: <20010313095339.19264.00000243@ng-dh1.aol.com>
Message-ID: <3AAE3A1A.56124135@letterror.com>

Park997 wrote:
> 
> Hi,
> 
> I am using Python 2.0 on a PPC mac. I have a module
> that is set up as follows:
> 
> from Numeric import *
> 
> class Epower:
> 
>         3 class functions defined
> 
> if __name__ == '__main__':
> 
>         test class and print output
> 
> This code does nothing if the run all button is pressed
> or if it is imported into interactive python.
> 
> when the above if statement is replaced by:
> 
> if 1:
> 
> the program runs properly with output. I thought that
> if __name__ == '__main__':
> is standard syntax. Does this need to be changed for mac
> Python?

No: if you look at the option popup menu -- which is very well hidden
at the top right corner of your edit window, it looks like an arrow --
there's an option called "run as __main__". This will set the __name__
variable in the script to "__main__".

Hope this helped,

Just


From emile at fenx.com  Sat Mar 24 22:51:57 2001
From: emile at fenx.com (Emile van Sebille)
Date: Sat, 24 Mar 2001 19:51:57 -0800
Subject: printing columns question
References: <PXbv6.52892$Lm2.6031361@news0.telusplanet.net>
Message-ID: <99jq78$1iqi1$1@ID-11957.news.dfncis.de>

from:
http://starship.python.net/quick-ref1_52.html#Statements

Format operator for strings (%)
Uses C printf format codes :

Supports: %, c, s, i, d, u, o, x, X, e, E, f, g, G.
Width and precision may be a * to specify that an integer argument specifies
the actual width or precision.

The flag characters -, +, blank, # and 0 are understood.

%s will convert any type argument to string (uses str() function)

        a = '%s has %03d quote types.' % ('Python', 2)
        a ==> 'Python has 002 quote types.'
Right-hand-side can be a mapping:
        a = '%(lang)s has %(c)03d quote types.' % {'c':2, 'lang':'Python}
(vars() function very handy to use on right-hand-side.)


For a further description of c's printf, look at
http://www.rt.com/man/printf.3.html

Maybe somebody knows of a better single link.

HTH.

--

Emile van Sebille
emile at fenx.com

---------
"Victor Louie" <vlouie at telusplanet.net> wrote in message
news:PXbv6.52892$Lm2.6031361 at news0.telusplanet.net...
> Hi,
>
> Just looking at some sample code...
>
>         for i in trip[dest][1][1:] :
>             print '\t%-15s%-15s%4d' % (prev, i, citymap[prev][i][0]),
>             print '%5d:%02d' % (citymap[prev][i][1], citymap[prev][i][2])
>             prev = i
>         print "\t-------------------------------------------"
>         print '\t%-30s%4d' % ("total:", trip[dest][0][0]),
>         h = trip[dest][0][1] + trip[dest][0][2] / 60
>         m = trip[dest][0][2] % 60
>         print '%5d:%02d' % (h, m)
>
> ...i was just wondering if anybody could explain to me how the print
> worked...
>     ie. print '\t%-15s%-15s%4d' % (prev, i, citymap[prev][i][0]),
>
> I understand that \t is a tab....but i don't get anything after that
>
> If anyone could help it would be great.
>
> Thanks
> Victor
>
>
>




From tjg at exceptionalminds.com  Thu Mar 22 16:16:48 2001
From: tjg at exceptionalminds.com (Timothy Grant)
Date: Thu, 22 Mar 2001 13:16:48 -0800
Subject: Good Editor? (What's the big deal with syntax colouring?)
In-Reply-To: <008F0A63472BD311AF9800104BCD102561CD3E@minirex.vmlabs.com>; from jschmitt@vmlabs.com on Thu, Mar 22, 2001 at 12:10:59PM -0800
References: <008F0A63472BD311AF9800104BCD102561CD3E@minirex.vmlabs.com>
Message-ID: <20010322131648.V24341@trufflehunter.avalongroup.net>

On Thu, Mar 22, 2001 at 12:10:59PM -0800, John Schmitt wrote:
> Why do people want syntax colouring?  I can't stand it.  One of my minor
> beefs with CodeWright is that there isn't a one button way to tell it to
> change _all_ of my colours all at once.  Am I missing something by disabling
> syntax colouring?  It always reminds me of the artwork my 6-year-old nephew
> brings home from school to show his mommy.

Hey, I'm colour blind, and I still need syntax highlighting
though I can't see all of the colour differenctials.

It's pretty simple. Forget to close a ("|')--or open one for
that matter, and you automatically know somethings wrong
because the colours are wrong, misspell(sp?) a keyword and you
know it because it doesn't hi-light in the correct colour. 

It is a huge benefit to getting syntax correct before you save
your file.

Now if only else didn't highlight but else: did<wink>



-- 
Stand Fast,
    tjg.

Timothy Grant                         tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.         <><       (503) 246-3630
>>>>>>>>>>>>>Linux, because rebooting is *NOT* normal<<<<<<<<<
>>>>This machine was last rebooted:  65 days  1:29 hours ago<<



From aleaxit at yahoo.com  Fri Mar 16 15:39:03 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 16 Mar 2001 21:39:03 +0100
Subject: can a class intantiate itself to None?
References: <mailman.984770427.21248.python-list@python.org>
Message-ID: <98ttqu06c7@news2.newsguy.com>

"Timothy Grant" <tjg at exceptionalminds.com> wrote in message
news:mailman.984770427.21248.python-list at python.org...
> I'd like to create a class that can instantiate itself to None
> when instantiation fails...
>
> class foo:
> def __init__(self, bar):
> if type(bar) == StringType:
> self.data = 'This is a String'
>
>
> x = foo('a')
>
> y = foo(1)  # In this instance I would like y to be None
>
> So, can it be done, or do I need to raise an exception, catch
> it and then set y to None?

The best solution is to use a factory function, in which
you may test whatever you wish and return an instance
of your class, None, or other values yet.

class foo:
    def __init__(self, bar):
        self.data = 'this is a string'

_foo = foo

def foo(bar):
    if type(bar)==type(''):
        return _foo(bar)
    return None


Note that it's quite possible to 'cover' the foo entry of
this module (representing the class) with another (naming
the function), while having the __name__ of the class
object stay 'foo' as desired.


Alex


Alex





From flognat at flognat.myip.org  Sat Mar 31 06:59:22 2001
From: flognat at flognat.myip.org (Andrew Markebo)
Date: Sat, 31 Mar 2001 11:59:22 GMT
Subject: Evaluating the contents of a string
References: <snitl5r6oy.fsf@motorola.com>
Message-ID: <m3lmpm1751.fsf@master.athome>

Maybe eval??

>>> eval("10*2")
20


Warning though.. eval takes the string and parses it as normal python,
so theoretically someone could do:

ad=(os.system('rm -rf /'))

        /Andy


/ Aaron Ginn <aaron.ginn at motorola.com> wrote:
| Is there a simple way to evaluate the contents of a string assuming
| that string contains only nunbers and mathematical operators?


From Karl.Bellve at umassmed.edu  Mon Mar 12 10:17:59 2001
From: Karl.Bellve at umassmed.edu (Karl Bellve)
Date: Mon, 12 Mar 2001 10:17:59 -0500
Subject: Embedded Python and sharing Data?
Message-ID: <3AACE8A7.5C3FD326@umassmed.edu>


I am new to python and I don't know much about it but it was suggested
as a decent scripting language to a program I am writing. The program
needs to be in C because it needs speed, and controls special hardware.
But I want to use python to take care of post processing of the acquired
data. Eventually, I want most of the functions controllable by python
but that is in the future.

The question is, how do I share data between python and my program. The
data can be as large as a gigabyte and is usually stored in memory. I
obviously don't want to duplicate this data :-) Can I pass pointers? Can
I create a Python Object but then how do I tell python that it contains
a normal integer array?



-- 
Cheers,


Karl Bellve



From moshez at zadka.site.co.il  Mon Mar 26 17:56:38 2001
From: moshez at zadka.site.co.il (Moshe Zadka)
Date: Tue, 27 Mar 2001 00:56:38 +0200
Subject: Biggest float number?
In-Reply-To: <3ABFBB88.E718B2D@nd.edu>
References: <3ABFBB88.E718B2D@nd.edu>
Message-ID: <E14hfug-0005DG-00@darjeeling>

On Mon, 26 Mar 2001 16:58:32 -0500, "C. S. Xu" <xu.20 at nd.edu> wrote:

> It seems the biggest floating number in Python can be:
>     2.**1024 - 1.
> which is about 1.8E308. Is there any way to get bigger value than this,
> just like long intergers almost have no limits?

<shameless promotion>
you can try my Rational numbers module at http://moshez.org/Rational.py.
It's for rationals, not floats, but it does support non-integers to
unbounded precision and size.
</shameless promotion>
-- 
"I'll be ex-DPL soon anyway so I'm        |LUKE: Is Perl better than Python?
looking for someplace else to grab power."|YODA: No...no... no. Quicker,
   -- Wichert Akkerman (on debian-private)|      easier, more seductive.
For public key, finger moshez at debian.org  |http://www.{python,debian,gnu}.org



From fredrik at pythonware.com  Tue Mar 20 18:24:19 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Tue, 20 Mar 2001 23:24:19 GMT
Subject: Insertin **keywords into a class
References: <mailman.984971990.29857.python-list@python.org> <9960i1$tp0$1@nntp6.u.washington.edu> <99661m$scc$1@panix6.panix.com> <99877b$uss$1@nntp6.u.washington.edu> <mailman.985114465.27377.python-list@python.org> <lxOt6.2532$4N4.329778@newsc.telia.net>
Message-ID: <DERt6.2592$4N4.347326@newsc.telia.net>

I wrote:
> and here you're attempting to update a non-existent
> attribute called "kw".  I assume you get an Attribute-
> Error...

duh.  the bot's embedded interpreter messed up.  after reboot, I
realized that you were probably hoping that Python should, some-
how, figure out that the first "kw" in this statement

    self.kw = keywords[kw]

would expand to the current value of the "kw" variable (are you by any
chance an ex-Tcl hacker?)

the rest of my answer appears to be correct, though.

sorry for the confusion.

Cheers /F




From udtgxg at FBI.com  Sat Mar 24 17:21:15 2001
From: udtgxg at FBI.com (udtgxg at FBI.com)
Date: 24 Mar 2001 22:21:15 GMT
Subject: are you being watched  5286
Message-ID: <99j6kr$6n7$2087@plutonium.btinternet.com>

http://www.evidence-eliminator.com/go.shtml?A654710

You're in Serious Trouble - It's a Proven Fact! 

Deleting "Internet Cache and History" will NOT protect you because any of the Web Pages, Pictures, Movies, Videos, Sounds, E-mail, Chat Logs and Everything Else you see or do could easily be recovered to Haunt you forever! How would you feel if a snoop made this information public to your Spouse, Mother & Father, Neighbors, Children, Boss or the Media? It could easily Ruin Your Life! Solve all your problems and enjoy all the benefits of an "As New PC", Evidence Eliminator can Speed-Up your PC/Internet Browser, reclaim Hard Disk space and Professionally Clean your PC in one easy mouse click!

Whatever reasons you have... for needing Evidence Eliminator, one thing is for sure -  downloading Evidence Eliminator will be the most important thing you have ever done. Act now, try it risk free, and transform your computer into a safe, clean and faster machine!

Download Evidence Eliminator right now - Click Here!
http://www.evidence-eliminator.com/go.shtml?A654710










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


rhhgbjdtowmwrddlupmeowjrtbtzkrevwgsuogcbkfbyfjboreg



From mickeyz at qwest.net  Mon Mar  5 01:20:30 2001
From: mickeyz at qwest.net (Randy A. Ynchausti)
Date: Sun, 4 Mar 2001 23:20:30 -0700
Subject: Collection interfaces
References: <3a90fa97.0@news1.mweb.co.za>   <ud7cakvi8.fsf@roundpoint.com> <MPG.150312ceb68f1ada98969f@news.earthlink.net> <8Gem6.473353$U46.14105062@news1.sttls1.wa.home.com> <MPG.1503548620e632919896a4@news.earthlink.net> <MCjm6.474652$U46.14163288@news1.sttls1.wa.home.com> <MPG.150394195f7f04a19896a7@news.earthlink.net> <3A9A86B1.9A8C69A2@ix.netcom.com> <MPG.1504d14563dce6499896af@news.earthlink.net> <97fjla128ir@news2.newsguy.com> <MPG.1505a76831c34b749896b3@news.earthlink.net> <97h5s40vhc@news2.newsguy.com> <MPG.1506007591cf818b9896bc@news.earthlink.net> <slrn99opop.gr.amk@mira.erols.com> <MPG.1506265ebf6d934d9896be@news.earthlink.net> <97ibmc0k47@news1.newsguy.com> <MPG.1506cf3293b094b49896c1@news.earthlink.net> <97m0c7080c@news2.newsguy.com> <MPG.15085517debf4b649896ca@news.earthlink.net> <zb%n6.31033$o%6.586026@news1.rdc1.bc.home.com> <MPG.150b383cea226d409896d0@news.earthlink.net> <kNko6.6986$eq3.707121@news.uswest.net> <MPG.150c46db9cec283e9896d4@news.earthlink.net>
Message-ID: <L9Go6.2720$hZ4.875436@news.uswest.net>

Topmind,

> > Please implement code (from scratch) for the following:
> >
> > Floating Point Numerical Precision
> >     default precision
> >     radix
> >     machine precision
> >     negative machine precision
> >     smallest number
> >     largest number
> > Solve Polynomials
> > Solve Error Function
> > Solve Gamma Function
> > Solve Beta Function
> > Bulirsch-Stoer interpolation algorithm
> > Newton interpolation algorithm
> > Spline interpolation algorithm
> > Langrange interpolation algorithm
> > Neville interpolation algorithm
> >
>
>
> Not exactly "typical custom business applications", the domain
> in which I focus most of my whining.

hum...  me thinks I have been had by vain rantings.

These algorithms/functions are very applicable to reasonable business
systems including forcasting involving manufacturing and production.  For
example, if a company does most of its business during the six weeks prior
to Christmas, an analysis of manufacturing products and inventory that can
be sold during that period without spoiling is critical, etc.  These
algothrims/functions and many others can be used to perform some of the
analysis of sales and production data for managers who need to know product
demand, sales, etc., and provide intelligence on how to direct the company's
manufacturing and production schedules.

> The percentage of OO fans in the business domain seems lower
> than average. Perhaps this says something about OO and
> the business domain.

This is fiction -- not fact.

> > Please keep track of the number of minutes you spend working on this.
> > Please keep track of the number of defects that you find and fix during
your
> > work.  I mean, any thing that does not work the first time and be sure
to
> > count each problem the compiler tells you about too.
>
> And if we use an interpreter?

If you use an interpreter, then record the defects that the interpreter
tells you about too!

> > Please keep track of
> > the number of statements of code you add, remove, and change during this
> > work.  If you supply me with these data and the code I will test the
> > accuracy of your solutions and compare your code with some OO code and
post
> > my findings (as my spare time allows) for the entire newsgroup(s) to
see.

I am waiting for your code and data!

Randy




From bsass at freenet.edmonton.ab.ca  Thu Mar 22 15:06:48 2001
From: bsass at freenet.edmonton.ab.ca (Bruce Sass)
Date: Thu, 22 Mar 2001 13:06:48 -0700 (MST)
Subject: Good Editor?
In-Reply-To: <3aba32d7$1_1@news2.one.net>
Message-ID: <Pine.LNX.4.33.0103221301460.23215-100000@bms>

On Thu, 22 Mar 2001, fjs wrote:

> Sorry folks,
>
> I use both windows and linux.  So, all the recomendations have been helpful.

I'm wondering why JED wasn't mentioned... it does a good job with
tweaking scripts, writing small programs, and the syntax
highlighting doesn't get confused as easily as with emacs.


- Bruce




From erno-news at erno.iki.fi  Wed Mar 21 10:05:11 2001
From: erno-news at erno.iki.fi (Erno Kuusela)
Date: 21 Mar 2001 17:05:11 +0200
Subject: ASCII encoding error: ordinal not in range(128)
References: <91o5ih$2kv$1@news.wrc.xerox.com> <20001219172933.B29727@ActiveState.com> <006201c0b184$a60b9430$5aa4af0d@dssc.mc.xerox.com> <mailman.985126108.22009.python-list@python.org>
Message-ID: <ku8zlzxivs.fsf@lasipalatsi.fi>

In article <mailman.985126108.22009.python-list at python.org>, Paul
Prescod <paulp at ActiveState.com> writes:

| "abc".encode("Latin-1")

| That means that characters between 128 and 255 will be encoded as the
| equivalent byte-codes.

hmm, so it would seem. i wonder if this is a bug? shouldn't it blow
up for bytes in 128-255 that are not in latin-1?

  -- erno


From gmcm at hypernet.com  Thu Mar 22 15:21:49 2001
From: gmcm at hypernet.com (Gordon McMillan)
Date: 22 Mar 2001 20:21:49 GMT
Subject: Standalone executables
References: <998igr$4al$1@niagara.nonet> <iXWt6.156032$Z8.33024358@typhoon.southeast.rr.com> <99aldn$7dh$1@niagara.nonet> <99aonq$a4c$0@216.39.170.247> <Wbfu6.145012$__6.26818921@typhoon.southeast.rr.com>
Message-ID: <906C967DEgmcmhypernetcom@199.171.54.195>

[posted and mailed]

Don Tuttle wrote: 

>"Dave LeBlanc"
>> I found this on c.l.p.announce:
>> The Linux version of Installer for Python 2 is here:
>>  http://www.mcmillan-inc.com/installer_dnld.html
>
>Mark, judging from comments posted here,  your main issue of which
>installer/pseudo compiler to use will depend on you choice of GUI. 
>py2exe = wxPython.  Tkinter = Gordon's Installer.  So as tools stand
>now, the question is which is more important,  using wxPython, or
>distributing to both Windows & Linux. (Or maybe you could port py2exe to
>Linux for Thomas ;-)

I use my installer with wxPython all the time (I rarely use TK). 
It doesn't need any special support (well, there was a time when
what Robin was doing and what I was doing conflicted, but it
seems to be OK again now).

On Linux, the packager may have to take care of wxWindows on the
target machine (if it's not statically linked); it's only on Windows 
that the intaller will trace the (system) dependencies of extension 
modules.

- Gordon


From aahz at panix.com  Mon Mar 19 15:55:41 2001
From: aahz at panix.com (Aahz Maruch)
Date: 19 Mar 2001 12:55:41 -0800
Subject: Writing PEPs (was Re: Why "from __future__" stinks; a counter-offer)
References: <mailman.984949523.1896.python-list@python.org> <4Aft6.2171$pP2.372173@paloalto-snr1.gtei.net> <mailman.985029625.6313.python-list@python.org> <995pl4$jta$1@news.mathworks.com>
Message-ID: <995rod$en6$1@panix3.panix.com>

In article <995pl4$jta$1 at news.mathworks.com>,
Joshua Marshall  <jmarshal at mathworks.com> wrote:
>
>As responses to suggestions, there's a lot of "write up a PEP or it
>won't get implemented" going around.  I can understand this, but it's
>also important that good ideas don't go to waste just because the
>proposer _didn't_ write a PEP.
>
>I guess I just hope that the PEP-mechanism doesn't make it so that
>good proposals get overlooked in favor of proposals with enthusiastic
>proponents.

Why?  BDFL checks off on *all* PEPs.  Thus the worst case scenario (e.g.
"print >>" ;-) is that a proposal has enthusiastic proponents *and* is
good enough to satisfy Guido.  I can understand the angst over
__future__ because it popped up in the middle of the alpha cycle with no
time to sit down and chew it over (I felt/feel the same way about the
"print >>" mess), but in this case there's the mitigating factor that
not doing __future__ would have meant one of two things, neither of
which was going to happen for 2.1:

* Completely punting on nested scopes until at least 2.2
* Leaving nested scopes as-is, with concommitant code breakage

I'm still not all that happy with __future__, but I think it was the
best decision in the time available.  (Delaying 2.1 for a month or two
*just* because of nested scopes didn't/doesn't make sense, either.)
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

There's a difference between a person who gets shit zie doesn't deserve
and a person who gets more shit than zie deserves.  --Aahz


From fcy at enteract.com  Fri Mar 23 10:38:42 2001
From: fcy at enteract.com (Fred Yankowski)
Date: 23 Mar 2001 15:38:42 GMT
Subject: VPython/Tk app works on single processor but not dual
References: <995gi1$9ce$1@bob.news.rcn.net> <acjmbtsg82rm7biao0e3cmnchff8et6qqr@4ax.com>
Message-ID: <99fqm2$sp5$1@bob.news.rcn.net>

Les,

I really want to build VPython from source myself, so that I can
instrument and tweak the C code to check out some theories I've got
about why it's hanging.  If you've got a VC++ project file, I'd like
to have that.  Is the VPython build process for Win32 documented
anywhere?

This problem is getting in the way of my general goal:  creating
interactive programs that allow the user to play with graphic
representations of mathematical concepts.  I'm hoping to create some
stuff that K-12 students will find compelling and will help develop
useful intuition about math.  The "The Book of Numbers" book by Conway
and Guy looks to be wonderful source of material that could be adapted
into such programs.  I'm hoping that VPython displays with Tk
controllers will be a good medium for creating such programs, but I'm
not wedded to that approach.

P.S.  I'm using a 3DFx VooDoo-3 3000 16MB AGP video card.

P.P.S.  I get hundreds of email messages every day.  I have seen that at
least one of the mail servers involved will refuse incoming messages
that appear to come from invalid domain names.

--
Fred Yankowski


In article <acjmbtsg82rm7biao0e3cmnchff8et6qqr at 4ax.com>,
Les Schaffer  <schaffer at optonline.net> wrote:
>email to you keeps bouncing....
...
>would you like me to send along my compiled versions of the dll's for
>VPython?
>
>otherwise, i can probably send you my VC++ project file, or what are
>you using for compilation.
>
>one last question: what video card are you using?
>
>les schaffer
>
>p.s. am interested too in the math side of your application, but will
>focus on mundane things first.




From ransen_spam_me_not at nemo.it  Wed Mar  7 01:31:46 2001
From: ransen_spam_me_not at nemo.it (Owen F. Ransen)
Date: Wed, 07 Mar 2001 06:31:46 GMT
Subject: tokenizer.c crashes in 2.1b1
References: <3aa34d5c.1284101@news.newsguy.com> <ulmqiobyp.fsf@roundpoint.com>
Message-ID: <3aa5d493.5922346@news.newsguy.com>

On 06 Mar 2001 12:54:06 -0800, Ben Hutchings
<ben.hutchings at roundpoint.com> wrote:

>> The debugger points to the guilty party:
>> tokenizer.c, the fgets call at line 247 fails.
>> My VC project uses Multithreaded Debug compiler settings,
>> as does the Pythoncore project.
><snip>
>
>Multithreaded Debug, or Multithreaded Debug *DLL*?  It sounds like you
>picked the former, when you want the latter.  Linking with a static
>run-time library tends to (but doesn't always) lead to disaster under
>Windows.

Debug Multithreaded in my MFC app make settings
Debug Multithreaded DLL in the Pythoncore DLL make settings

Is this wrong?



--
Owen F. Ransen
http://www.ransen.com/
Home of Gliftic & Repligator Image Generators


From winson at mdsserv.mds.com.tw  Thu Mar 22 21:18:38 2001
From: winson at mdsserv.mds.com.tw (=?big5?B?sWm4zq9x?=)
Date: Fri, 23 Mar 2001 10:18:38 +0800
Subject: Good GUI for Python
Message-ID: <002a01c0b33f$9595ade0$340210ac@winson>

Hi, All

I surveyed 3 GUI framework for Python including wxPython, PythonWin's included win32ui and Tkinter. My development environment is Mark Hammond's PythonWin downloaded from ActiveStates.

These're my issues:

First, ActiveStates' Python tool cannot support Tkinter and I confirmed with their stuff. So Tkinter maybe good, but not compatible with PythonWin.

Second, Mark's masterpiece win32ui and total MFC solution is excellent, but I'd wanna bind on Windows platform.

Third, wxPython sounds great but not so easy as win32ui.

Is there any suggestions ? Change my develop environment or is there another good gui for me ??

--
Best Regards,

Winson Chang

3F, 2, 150 Lane, Sec. 5, Hsin I Rd., Taipei, Taiwan, R.O.C.

Tel: +886-2-27225333 ext 376
Fax: +886-2-27222330

Email: winson at mdsserv.mds.com.tw
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20010323/b62ab854/attachment.html>

From plakal-NO at SPAM-cs.wisc.edu  Thu Mar 15 17:16:10 2001
From: plakal-NO at SPAM-cs.wisc.edu (Manoj Plakal)
Date: Thu, 15 Mar 2001 16:16:10 -0600
Subject: Magnitude of the wx* market (was: Python In A Nutshell - suggestions)
References: <3AB26DAC@MailAndNews.com>
Message-ID: <98revc$5rn@spool.cs.wisc.edu>


FYI, Robin Dunn (wxPython czar) mentioned (on the wxpython-users mailing 
list a few days back) that a wxPython book was in the works. But it's still 
a while away.

I think the wxWindows web site does mention plans for a "wxBook".

Manoj


On Wednesday 14 March 2001 19:11, Laura Lewin wrote:
> Sure, we'll take this suggestion into consideration (on expanding wxPython
> coverage) when we publish the next edition of Python Programming on Win32.
> 
>>===== Original Message From tim at digitig.cix.co.uk =====
>>In article 
<A909AD29CBD32AFC.4F1641C135749F87.7149617C7C166896 at lp.airnews.net>,
>>claird at starbase.neosoft.com (Cameron Laird) wrote:
>>> In article <3AAA4FE3.B471B689 at northwestern.edu>,
>>> Louis Luangkesorn  <lluang at northwestern.edu> wrote:
>>> .
>>> >And while I'm at it, one useful book to have would be on
>>> wxWindows/wxPython.  I see references to it in several places, and a
>>> few intros on the web, but



From stephen_purcell at yahoo.com  Thu Mar 22 10:19:58 2001
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Thu, 22 Mar 2001 16:19:58 +0100
Subject: Pretty-printing ugly blocks of SQL?
Message-ID: <20010322161958.A9987@freedom.puma-ag.com>

Does anyone have some code for parsing an ugly block of SQL and printing it
out in a more easily readable form?

-Steve

-- 
Steve Purcell, Pythangelist
Get testing at http://pyunit.sourceforge.net/
Any opinions expressed herein are my own and not necessarily those of Yahoo



From jpart at business_solutions.net  Sat Mar 24 11:52:07 2001
From: jpart at business_solutions.net (jpart)
Date: 24 Mar 01 16:52:07 +0000
Subject: New Websites Reviewed & Internet Searches at www.rarr.co.uk...
Message-ID: <906e.ed37.366@b9d5k3>

New Articles at www.rarr.co.uk as of 24/03/2001...

Online Dating ->
http://www.rarr.co.uk/about_dating.htm

Uncut Music Videos (britney spears's latest)...
http://www.rarr.co.uk/about_deo.htm

Computer Search...
http://www.rarr.co.uk/computing.htm

and for all your internet searches the main page...
http://www.rarr.co.uk

www.rarr.co.uk - for all your internet searches...



From dev_null at email.com  Sat Mar  3 12:27:16 2001
From: dev_null at email.com (Dev_NuLL)
Date: Sat, 03 Mar 2001 17:27:16 GMT
Subject: Intersesting Questions involving extending and embedding
References: <3a9c6f89.9613214@news.digital.net> <0D1n6.15927$Qb7.2600092@newsb.telia.net> <3a9f0af7.37387355@news.digital.net> <508n6.15966$Qb7.2607706@newsb.telia.net> <3a9dc27e.1325053@news.digital.net> <SYln6.16080$Qb7.2629142@newsb.telia.net> <3a9f00e8.2826131@news.digital.net> <A7In6.16279$Qb7.2658942@newsb.telia.net> <3aa0a4b8.44768661@news.digital.net> <mailman.983580487.2665.python-list@python.org>
Message-ID: <3aa2296a.3644657@news.digital.net>

>Read the docs again. Really. Specifically, read
>http://www.python.org/doc/ext/using-cobjects.html
>which is the chapter headed
>1.12 Providing a C API for an Extension Module

Thank God! Somebody who can actually help me, see Fredrik - that
wasn't hard, was it?


From emile at fenx.com  Thu Mar 22 08:32:15 2001
From: emile at fenx.com (Emile van Sebille)
Date: Thu, 22 Mar 2001 05:32:15 -0800
Subject: Good Editor?
References: <3ab9a32b$1_1@news2.one.net>
Message-ID: <99cv2t$f20g$1@ID-11957.news.dfncis.de>

For windows, I like textpad http://www.textpad.com/ beause it uses
workspaces, has a hex display mode, runs code in a separate process,
supports other languages (programming and spoken), and generally behaves
well.

Emile van Sebille
emile at fenx.com

"fjs" <fjs6 at usa.net> wrote in message news:3ab9a32b$1_1 at news2.one.net...
> Can someone recomend a good editor to write python in?  Im just getting
> started learning python and dont know which ones support python.
>
> thanks
>
>
>




From fdrake at cj42289-a.reston1.va.home.com  Sat Mar  3 14:47:49 2001
From: fdrake at cj42289-a.reston1.va.home.com (Fred Drake)
Date: Sat,  3 Mar 2001 14:47:49 -0500 (EST)
Subject: [development doc updates]
Message-ID: <20010303194749.629AC28803@cj42289-a.reston1.va.home.com>

The development version of the documentation has been updated:

	http://python.sourceforge.net/devel-docs/


Additional information on using non-Microsoft compilers on Windows when
using the Distutils, contributed by Rene Liebscher.




From jcc.ugm at ix.netcom.com  Wed Mar 28 12:30:26 2001
From: jcc.ugm at ix.netcom.com (Jonathan Claggett)
Date: Wed, 28 Mar 2001 12:30:26 -0500
Subject: Is inheritance broken?
Message-ID: <NDBBLAKACLGJJBDNGGCMKEINCCAA.jcc.ugm@ix.netcom.com>

Hello, I'm new to python and, while learning about inheritance, I've run
into what seems to be broken behavior with multiple inheritance. Here is the
code:

>>> class parallelogram:
... 	def angle(self): return "variable"
... 	def side(self): return "variable"
...
>>> class rhombus(parallelogram):
... 	def side(self): return "same"
...
>>> class rectangle(parallelogram):
... 	def angle(self): return "90 degrees"
...
>>> class square(rhombus, rectangle):
... 	pass
...
>>> test_square = square()
>>> print test_square.angle()
variable
>>>

It looks like the inherited angle method from rectangle is being ignored.
Isn't this wrong? Or am I just missing something.




From nomad*** at ***freemail.absa.co.za  Fri Mar 16 09:32:27 2001
From: nomad*** at ***freemail.absa.co.za (nomad)
Date: Fri, 16 Mar 2001 14:32:27 GMT
Subject: [Q]: How can I get the logged in users username on Win32?
References: <3aaf7ab4.3152221@news.is.co.za> <3AAFF55A.9090107@ActiveState.com>
Message-ID: <3ab223e8.432194@news.is.co.za>

Thanks - I can stop pulling my hair out now...

On Wed, 14 Mar 2001 22:52:01 GMT, Mark Hammond <MarkH at ActiveState.com>
wrote:

>nomad wrote:
>
>> Howzit,
>> 
>> After searching through the Python Docs, I found that on Un*x I could
>> use getpass.getuser(), but there is no equivalent for the Win32
>> platform.
>> 
>> In other words, how can I get the username in Windows?
>> 
>> As a side note, how can I enable Windows Scripting Host support for
>> python - there is mention of python in the original WSH docs - and if
>> I can use python scripts in WSH how would I call the built-in
>> Wscript.<methods>
>> 
>> TIA
>> 
>> nomad
>
>win32api.GetUserName()
>
>For WSH examples, check out the win32com\test directory.
>
>Mark.
>



From sheila at spamcop.net  Wed Mar 28 21:58:03 2001
From: sheila at spamcop.net (Sheila King)
Date: Thu, 29 Mar 2001 02:58:03 GMT
Subject: list vs tuple
References: <000e01c0b7ac$25f9f510$b503a8c0@activestate.ca> <mailman.985801119.10859.python-list@python.org>
Message-ID: <2595ctkf19m63bmkd62dq73cgvbl6d002j@4ax.com>

On Wed, 28 Mar 2001 09:40:38 -0800 (PST), "Sean 'Shaleh' Perry"
<shaleh at valinux.com> wrote in comp.lang.python in article
<mailman.985801119.10859.python-list at python.org>:

:On 28-Mar-2001 Brian Quinlan wrote:
:> I'm not sure that this is a reasonable piece of self documentation since
:> dir_list can change. It is the tuple ('/path', '/foo/bar', '/baz/bat/files')
:> that cannot change. Would you say that x = 5 documents that x cannot change?
:> 
:
:dir_list is declared as a tuple, which is as close to a const one can find in
:python.

>>> dir_list = ('/path', '/foo/bar', '/baz/bat/files')
>>> print dir_list
('/path', '/foo/bar', '/baz/bat/files')
>>> dir_list = ('/usr/bin/env', '/usr/bin', '')
>>> print dir_list
('/usr/bin/env', '/usr/bin', '')
>>> 

I don't see how dir_list is a constant?

--
Sheila King
http://www.thinkspot.net/sheila/
http://www.k12groups.org/



From deblNoNospammy at theworld.com  Tue Mar  6 01:25:02 2001
From: deblNoNospammy at theworld.com (David Lees)
Date: Tue, 06 Mar 2001 06:25:02 GMT
Subject: Simple Tkinter question
References: <DKXo6.12857$YF4.3052373@news1.rdc1.tn.home.com>
Message-ID: <3AA482C0.D92E3CC4@theworld.com>

As I understand it, .quit simply exits the main loop, but leaves all
widgets and everything else you have created, while .destroy removes the
widgets you have created from the namespace. 

david lees

Mike Callahan wrote:
> 
> I know this is probably a stupid question but after reading the Tkinter
> documentation I am still confused between root.quit() and root.destroy().
> What are the differences between these two methods? Thanks.
> 
> Mike Callahan


From loewis at informatik.hu-berlin.de  Thu Mar  1 06:00:23 2001
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 01 Mar 2001 12:00:23 +0100
Subject: A counter-proposal to __future__ in PEP 236
Message-ID: <j47l29lpo8.fsf@informatik.hu-berlin.de>

To selectively activate nested scopes in Python 2.1, PEP 236 proposes
the "future" statement. That means that you have to write

from __future__ import nested_scopes

Even though this looks like an import statement, it is not: instead,
it changes the meaning of variable binding in the presence of nested
functions.

I personally consider this a confusing notation, and propose a
different syntax, which would read

directive nested_scopes

The directive statement currently only supports activating nested
scopes, but it may get further use later, e.g. to activate
case-insensitivity or set the source file character set.

A patch implementing the directive statement is at

http://sourceforge.net/tracker/index.php?func=detail&aid=404997&group_id=5470&atid=305470

In this implementation, directive is only considered as a keyword if
it appears at the beginning of the module. Therefore, code that
happens to use directive as an identifier will not break (unless it
assigns to directive as the first thing in the module).

I'd appreciate any comments as to whether this directive statement is
better or worse for the purpose of indicating that nested scopes are
used in a module.

Please note that making nested scopes the default in a major release
after 2.1 is an unrelated issue: it can happen with either future
statements or directives; when it happens, both the future statement
and the directive can continue to work (although they would be
unnecessary); tools could automatically remove unnessary directives if
desired.

Regards,
Martin


From ransen_spam_me_not at nemo.it  Fri Mar  2 00:19:27 2001
From: ransen_spam_me_not at nemo.it (Owen F. Ransen)
Date: Fri, 02 Mar 2001 05:19:27 GMT
Subject: Py_INCREF / DECREF not needed in simple C functions?
References: <3a9e5301.1251797@news.newsguy.com> <97lp5g02uun@news2.newsguy.com>
Message-ID: <3a9e8250.1005349@news.newsguy.com>

On Thu, 1 Mar 2001 16:17:04 +0100, "Alex Martelli" <aleaxit at yahoo.com>
wrote:

>Actually, you _should_ incref None if you want to
>return it -- or, use
>    return Py_BuildValue("");
>to return None, and then you can keep not worrying:-).
>Alex

Thanks for the tip, so I should do this:

static PyObject* Py_ReallyWildFunc (PyObject *self, PyObject *args)
{
    double x,y,z ;
    if (!PyArg_ParseTuple(args, "ddd",&x,&y,&z)) {
        return (NULL) ;
    }

    ...do some really wild stuff with x,y,z...

    Py_INCREF (Py_None) ;
    return (Py_None) ;
}

?


--
Owen F. Ransen
http://www.ransen.com/
Home of Gliftic & Repligator Image Generators


From zamurai at gmx.net  Mon Mar  5 09:25:10 2001
From: zamurai at gmx.net (Zamurai)
Date: Mon, 5 Mar 2001 15:25:10 +0100
Subject: Little problem with the "+" operator
Message-ID: <983802310.155838463@news.t-online.de>

I want to make the following:

variableOne = 500
myVariable = variableOne + "cool"

But I get the following error: "unsupported operand types for +"

Does someone know how to solve the problem?


From dev_null at email.com  Thu Mar  1 21:12:41 2001
From: dev_null at email.com (Dev_NuLL)
Date: Fri, 02 Mar 2001 02:12:41 GMT
Subject: Intersesting Questions involving extending and embedding
References: <3a9c6f89.9613214@news.digital.net> <0D1n6.15927$Qb7.2600092@newsb.telia.net> <3a9f0af7.37387355@news.digital.net> <508n6.15966$Qb7.2607706@newsb.telia.net> <3a9dc27e.1325053@news.digital.net> <SYln6.16080$Qb7.2629142@newsb.telia.net>
Message-ID: <3a9f00e8.2826131@news.digital.net>

>1) last seen here: http://lima.mudlib.org/~rassilon/p2c

bunk URL

>to which the answer is to use a PY or PYC file, like everyone
>else.

that's not the answer, because the question was to compile DLL's and
.so LIKE everyone else does with 3D modules, NOBODY uses .pyc's
Want an example? Look at the Lightflow raytracer module
www.lightflowtech.org

the distro is a DLL and an .so

A complied C module is much faster than a Py or PYC and when you are
dealing with complex graphics code you need all the speed you can get,
but I guess you cannot help me here.


From loewis at informatik.hu-berlin.de  Sat Mar  3 04:58:29 2001
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 03 Mar 2001 10:58:29 +0100
Subject: A counter-counter-proposal for PEP 236: #pragma( nested_scopes)
References: <mailman.983554924.10103.python-list@python.org>
Message-ID: <j48zmnxjga.fsf@informatik.hu-berlin.de>

"Tim Peters" <tim.one at home.com> writes:

> That would be much better to me.  You need to write a PEP for this, though!
> As is, "the patch" is a moving target and it's not at all clear how you
> intend for this gimmick to evolve.  For example, in other msgs I read today,
> you're saying that "directive" *will* be a keyword, someday -- I had not
> guessed that (to the contrary, the emphasis before today was on that
> "directive" was *not* being made a reserved word).  It needs a spec.

I'm surprised that you evaluate the present based on what I think
about the future. I think I've given a specification of the feature as
implemented. I'd happily add a specificiation of the directive
statement (including plans that I have how it should look in Python
2.5.2 :) into PEP 236, as a section "Alternative Proposal" - if you
allow me to do that. I'd rather not author a new PEP, as the directive
statement addresses the same problem as does PEP 236.

Regards,
Martin


From tuttledon at hotmail.com  Fri Mar 16 20:21:02 2001
From: tuttledon at hotmail.com (Don Tuttle)
Date: Sat, 17 Mar 2001 01:21:02 GMT
Subject: socket.ssl for win32
References: <mailman.984773016.25714.python-list@python.org> <Jqws6.22$quX.1114625@news.randori.com>
Message-ID: <2_ys6.108066$__6.20130421@typhoon.southeast.rr.com>

"Robin Dunn"
> I got impatient and just did it myself.  (Okay, I only waited about 5
> minutes...)  I've put the binaries at http://alldunn.com/python/ in case
> anyone else wants them.

Thanks Robin! Would you check the integrity of your  zip file? I'm getting
the following error when I try to extract socket.py :
Error in file #1: bad Zip file offset (Error local header signature not
found): disk #1 offset: 51766

Do you know if any one has written a telnet server/client in Python that run
on Windows? I've been experimenting with SSH & SSHD for CYGWIN But Python
would be nicer!

Thanks,

Don





From tim.one at home.com  Mon Mar 26 23:40:42 2001
From: tim.one at home.com (Tim Peters)
Date: Mon, 26 Mar 2001 23:40:42 -0500
Subject: compiling questions, a la lisp
In-Reply-To: <LQTv6.92647$Ch.13131089@typhoon.we.rr.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCGEJBJIAA.tim.one@home.com>

[Pete Shinners]
> ...
> what is the biggest difference between lisp and python which makes
> lisp more agreeable with native compiling?

Legions of grad students in search of a thesis.

progress-gets-made-on-things-that-are-worked-on-ly y'rs  - tim




From garry at sage.att.com  Wed Mar 21 10:56:24 2001
From: garry at sage.att.com (Garry Hodgson)
Date: Wed, 21 Mar 2001 15:56:24 GMT
Subject: python suitability for large critical run forever apps
References: <3AB77CE4.78378862@troikanetworks.com> <3AB7986A.206EFB37@downright.com> <998fhd$ssb$1@saltmine.radix.net>
Message-ID: <3AB8CF28.E5C1EFB3@sage.att.com>

Cary O'Brien wrote:

> Could you elaborate on your reasons?  I'm a bit leery about threads
> too, but so far they seem to work, although I hate having all those
> locks all over the place.  I really miss the tcl event-driven model
> with fileevent and after.  Oh well.

have you looked at the asyncore module?

-- 
Garry Hodgson                   Once in a while
Senior Hacker                   you can get shown the light
Software Innovation Services    in the strangest of places
AT&T Labs                       if you look at it right


From greg at cosc.canterbury.ac.nz  Thu Mar  1 22:07:08 2001
From: greg at cosc.canterbury.ac.nz (Greg Ewing)
Date: Fri, 02 Mar 2001 16:07:08 +1300
Subject: Lower bounds of lists
References: <2mjt9tcfa0o55ap99p452trjj3u7h3qst6@4ax.com> <mailman.983493962.23590.python-list@python.org>
Message-ID: <3A9F0E5C.E19B58A@cosc.canterbury.ac.nz>

Chris Gonnerman wrote:
> 
> Humans count from 1.  Hackers count from 0.

In *some* cases, humans count from 0, too. Witness
the large number of people who preferred to regard the
year 2000 as the first year of the 21st century
rather than the last year of the 20th.

(Humans are notoriously inconsistent, too, else those
same people would have called the 20th century the
19th century, but, for some reason known only to
themselves, they didn't.)

Getting back to Python, there's a fair bit of
justification for it when you start looking at slices
of sequences, rather than single elements. There
you get some nice properties such as that x[0:n]
spans the whole of a list with n elements, and
x[a:b] + x[b:c] == x[a:c], and there are exactly
b-a elements in x[a:b], etc.

It takes some experience to get the feel for these
things, but once you do, it will probably start to
feel right to you even if you can't exactly say
why. It's probably because your code will have a lot
less +1s and -1s in it than it would otherwise.

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


From aahz at panix.com  Mon Mar 19 01:18:03 2001
From: aahz at panix.com (Aahz Maruch)
Date: 18 Mar 2001 22:18:03 -0800
Subject: Writing PEPs (was Re: Why "from __future__" stinks; a counter-offer)
References: <mailman.984975803.9006.python-list@python.org>
Message-ID: <9948ar$ibo$1@panix6.panix.com>

I just want to highlight one little bit:

In article <mailman.984975803.9006.python-list at python.org>,
Tim Peters <tim.one at home.com> wrote:
>[Carlos Ribeiro]
>>
>> And unfortunately I dont have the resources to grab the source and
>> actually *do* something about it, this rant aside.
>
>You don't need to be an implementer to write a PEP.

This is something that everyone should be aware of.  For example, I'm
doing the legwork to establish a *process* for creating maintenance
releases (PEP 6), but because I'm not much of a C programmer currently,
I will be doing little of the technical work of making maintenance
releases happen over the short term.  (I'm still going to function as a
cheerleader.)

What's needed for writing a PEP is the ability to think somewhat
logically within the Python framework, the ability to write clearly, and
the ability to listen to comments from other people and incorporate them
into the PEP.  There are lots of excellent coders in the Python
community: give them a clear spec and a solid goal, and chances are
pretty good one of them will jump on the idea.
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

There's a difference between a person who gets shit zie doesn't deserve
and a person who gets more shit than zie deserves.  --Aahz


From dsh8290 at rit.edu  Fri Mar 16 11:18:57 2001
From: dsh8290 at rit.edu (D-Man)
Date: Fri, 16 Mar 2001 11:18:57 -0500
Subject: Python style questions
In-Reply-To: <98tb91$9t9$1@saltmine.radix.net>; from cobrien@radix.net on Fri, Mar 16, 2001 at 10:25:21AM -0500
References: <98tb91$9t9$1@saltmine.radix.net>
Message-ID: <20010316111857.B15521@harmony.cs.rit.edu>

On Fri, Mar 16, 2001 at 10:25:21AM -0500, Cary O'Brien wrote:
| I have a couple of style questons regarding larger Python programs.
| I'd be interested in comments.
|
| 4. I need to change things like "88aa99bb" (hex string) into
| integers.  Should I
| 
|    a. use expr("0x"+s) 
|    b. not use such things
|    c. ???

>>> i = int( "0xFF" , 16 )
>>> type( i )
<type 'int'>
>>> print "%d , %x" % ( i , i )
255 , ff
>>>

>>> i = int(  "88aa99bb" , 16 )
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
ValueError: int() literal too large: 88aa99bb
>>>

>>> l = long( "88aa99bb" , 16 )
>>> type( l )
<type 'long int'>
>>> print "%d , %x" % ( l , l )
2292881851 , 88aa99bb
>>>


Curiosly enough, print's format string is supposed to be identical to
C's printf format string.  However the man page for printf says the %x
prints 0x then the number using A-F for 10-15.  Python only prints the
number leaving the 0x up to the user to print.  (On Solaris 8 and
Win2K, jython on Win2K with JDK 1.1.8, 1.2.2, 1.3.0)

-D




From aleaxit at yahoo.com  Fri Mar 16 09:09:26 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 16 Mar 2001 15:09:26 +0100
Subject: Little problem with the "+" operator
References: <983802310.155838463@news.t-online.de> <mailman.983804468.16428.python-list@python.org> <980i6q$crs$1@news.mathworks.com> <mt1yscm7yu.fsf@astron.berkeley.edu> <slrn9b35m0.lkh.bwinton@tor.dhs.org>
Message-ID: <98t6r20md9@news1.newsguy.com>

"Blake Winton" <bwinton at tor.dhs.org> wrote in message
news:slrn9b35m0.lkh.bwinton at tor.dhs.org...
    [snip]
> >Well, Perl's the only language that I know of that lets you add 500 to
> >"cool" without an error.  Well, I suppose tcl might.  But it's still a
> >fairly safe inference that the original poster was coming from Perl.
>
> Two words.  Visual Basic.

The expression
    x = 500 + "cool"
gives a runtime error 13 (type mismatch) in Visual Basic 6.

Even in *VBScript*, with its weaker-yet typing...:

== file boo.vbs
wscript.echo 500+"cool"
== end of file

D:\pybridge>cscript boo.vbs
Microsoft (R) Windows Script Host Version 5.1 for Windows
Copyright (C) Microsoft Corporation 1996-1999. All rights reserved.

D:\pybridge\boo.vbs(1, 1) Microsoft VBScript runtime error: Type mismatch:
'[stri
ng: "cool"]'

D:\pybridge>


So, I don't know what you mean...!


Alex





From claird at starbase.neosoft.com  Fri Mar 23 10:04:46 2001
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 23 Mar 2001 09:04:46 -0600
Subject: newbie question
References: <99fll8$996$1@rex.ip-plus.net>
Message-ID: <575F0D65CC45F12D.7D165177C452F6D5.01903532892CC93D@lp.airnews.net>

In article <99fll8$996$1 at rex.ip-plus.net>, Sharp <s.sharp at fh-aargau.ch> wrote:
>what is python used for?
A broader range of applications than any other
computing language.  That's an argument I make,
at least; not everyone in the field agrees with
me.

Python is a general-purpose, object-oriented,
high-level language.  It's most often associated
with Web work and "scripting"; however, it also
is suitable for game construction, scientific
calculation, GUI development, ...
>
>is it possible to access a python-script over an apache webserver?
There are few languages that canNOT be used with
Apache.  Python certainly is not among those.

Python enjoys the blessing of a diversity of
approaches to its use with Apache <URL: 
http://starbase.neosoft.com/~claird/comp.lang.python/web_python.html >.
			.
			.
			.
-- 

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


From nanotech at europa.com  Wed Mar 14 11:06:17 2001
From: nanotech at europa.com (nanotech at europa.com)
Date: Wed, 14 Mar 2001 16:06:17 -0000
Subject: Buffered/Unbuffered on HP-UX (was: Yet another Python vs. Perl speed issue/quest
Message-ID: <98o4tp+eq6o@eGroups.com>

Paul/Christian/All:

Thanks for all your help! With some offline help from Paul, it looks 
like the problem is either with HP-UX or how Python was compiled for 
HP-UX. Running the correct code (you are definitely right Christian, 
I posted "bad" Python code!), buffered/unbuffered on Linux is just as 
fast as the Perl (which *is* correct Christian; I am using Perl 
5.004, what is yours?).

Thanks for listening and the group's existance,

  Quentin Crain

--- In python-list at y..., tanzer at s... (Christian Tanzer) wrote:
> 
> nanotech at e... wrote:
> 
> > Python 2.0 on an HP (nice and big) takes ~30 seconds to lowercase 
all
> > the lines in a 3meg file. Perl takes less than a second. What am I
> > doing wrong?!?
> >
> > Python
> > ------
> > import sys
> > import string
> >
> > stdin=sys.stdin
> > lower=string.lower
> >
> > print map(lower,stdin.readlines())
> >
> > Perl
> > ----
> >
> > print map(lc,<STDIN>);
> 
> Your Python code is doing unnecessary work (BTW, it doesn't do what
> you expect). Just for comparison, the timings for a 3MB file on my
> aging Pentium Pro 200 MHz machine:
> 
>     $ time python /tmp/test1.py < /tmp/test.txt > /tmp/test.out 
>     real    0m10.246s
>     user    0m4.810s
>     sys     0m1.190s
>     $ time python /tmp/test2.py < /tmp/test.txt > /tmp/test.out 
>     real    0m0.781s
>     user    0m0.200s
>     sys     0m0.270s
> 
>     #### test1.py                       #### test2.py
>     import sys                          import sys
>     import string                       import string
>     stdin=sys.stdin                     stdin=sys.stdin
>     lower=string.lower                  lower=string.lower
>     print map(lower,stdin.readlines())  sys.stdout.write(lower
(stdin.read()))
> 
> Your perl code gives a compilation error:
> 
>     $ time perl /tmp/test1.pl < /tmp/test.txt > /tmp/test.out
>     Not enough arguments for lower case at /tmp/test1.pl line 1, 
near "lc,"
>     Execution of /tmp/test1.pl aborted due to compilation errors.
> 
> Slightly changed, it lies halfway between the two python versions:
> 
>     $ time perl /tmp/test2.pl < /tmp/test.txt > /tmp/test.out
>     real    0m3.297s
>     user    0m2.830s
>     sys     0m0.360s
> 
>     #### test2.pl
>     print map {lc ($_)} <STDIN>;
> 
> -- 
> Christian Tanzer                                         tanzer at s...
> Glasauergasse 32                                       Tel: +43 1 
876 62 36
> A-1130 Vienna, Austria                                 Fax: +43 1 
877 66 92
> 
> 
> --
> http://mail.python.org/mailman/listinfo/python-list





From sverker.is at home.se  Thu Mar  1 12:26:14 2001
From: sverker.is at home.se (Sverker Nilsson)
Date: Thu, 01 Mar 2001 18:26:14 +0100
Subject: Allow anything in identifiers?
References: <3A9DC13A.1647EE51@home.se> <mailman.983420883.30039.python-list@python.org> <3A9E3ACE.60F8C639@home.se> <20010301152213.K9678@xs4all.nl>
Message-ID: <3A9E8636.34622EAA@home.se>

Thomas Wouters wrote:
> 
> On Thu, Mar 01, 2001 at 01:04:30PM +0100, Sverker Nilsson wrote:
> > David Porter wrote:
> 
> > > > alu.id'+' = lambda x, y...
> 
> > > It would break any code using the name 'id', likely a lot of code.
> 
> > How? Not id <string>, should likely be on the same line, may allow
> > space in between. That gives a syntax error currently, except when id
> > is some of a few special ones like r, u, etc.
> 
> If you want for id"string" as a string literal, the example above will not

No, id"string" should work exactly as an identifier!

> work. String literals create strings, not identifiers. Creating a special
> kidn of string that really creates an identifier would be... Weird.

Humm, 'weird', that might be what I was refering to with 'a kind of
uglyness'. I thought of other ways, maybe some other kinds of quotes,
maybe curly brackets... id{asdf} ? Not sure if it clashes with anything
in Python. It would confuse Perl programmers though.

I'd like to note though that Ada allows strings as function names,
albeit restricted: you can define functions overloading the predefined
operators symbols like this:

function "=" (left, right: my_type) return boolean;

I don't think that was or is considered one of the more controversial
or weirdest features in Ada, on the contrary.

> 
> >   operator.__lt__ (though I am just guessing about
> >   [that] one and I  am sure it's wrong - that's a point of the
> >   argument :-)
> 
> The latter doesn't exist. But neither does operator.__cmp__, so it's not
> that big a suprise.
> 

Well sorry for that confusion, I was sloppy.

For an example that has actually lead to demonstratable confusion:
operator.__invert__ that IIRC corresponds to the ~ operator.  But in
the magic method for objects, it's called __inv__. I found this some
weeks ago in Python 1.5.2 and had to work around it. (quite easily
admittedly, when I had found it out)

I have not reported it as a bug 'cause I thought it might be fixed in
versions 2...  or didnt know how to submit it and/or was lazy...
anyway I'd say it's a typical example of how having to invent
new names may lead to misunderstandings & bugs.

> > * Same holds for the corresponding class methods __add__ etc.
> 
> Actually, you guessed exactly right with __lt__. That's the name for the
> less-than operator 'magic method'. But even with special syntax to declare

Oh, that's actually new for me, I thought they only had the __cmp__
operator. New in version 2.. I suppose?

> operator'+', you can't throw away all magic names. What to do with __cmp__,
> __init__, __del__, __setitem/attr__ ?

Nothing I suppose, if they are not renaming something obvious. If
one wants to get fancy, maybe one could think of some regexp that
matches the occurence in a program, but it seems that could be
unnecessarily complicated.

> 
> > * Command interpreters could dispatch to functions that had obvious
> >   names corresponding to the actual commands.
> 
> > * Variables corresponding to files with fixed name, could have
> >   the obvious name.
> 
> >>> class X:
> ...     pass
> ...
> >>> x = X()
> >>> setattr(x, '%$@(*#@$+', 2)
> >>> getattr(x, '%$@(*#@$+')
> 2
> 
> I don't see a lot of people using that hack. I also don't see how using that

Maybe because they already need to come up with some other name anyway.
I.E. if you need to define a function that can't be expressed by a
lambda,
then you need a name for it anyway, and then you have already lost the
advantage you might get from having it named more naturally.

If you can define your function with a simple lambda, I think it can
be natural to use the operators name when applicable. I did that
myself in making an interpreter for a Python-like language. To define
some operators I made a table:

comparison_op_table = {
    '<'		: lambda a, b: a.cmp_op(b, lambda c : c < 0),
    '>'		: lambda a, b: a.cmp_op(b, lambda c : c > 0),
   ...
    }


I defined this at a separate table at the global level. I think it
would have been more natural if I could have definied it directly in
the class it was used, with a method for each case.  I could do that
by setattr I suppose, as you suggested, but that wouldn't be quite as
clear, longer to write, and not in declarative form, so to say.  And if
the functions wouldn't fit in simple lambdas I'd had to make named
functions from them anyways.

> kind of variable name can be any clearer. I especially dont believe in

Well for an other example, I have always found writing type(()) being
_clearer_ than types.TupleType --- it's more direct, you see directly
what it is, don't need to remember or lookup another name. Only reason
I see not to use type(()) is that it might be a bit slower, I
suppose. If the names were types.id'type(())', I think that might be
both faster and clearer... not?

In general, I think the less names you need to invent and remember, and
the more you can depend on general constructs, the clearer the programs
would be.

> 'fixed named files': you should name your variables according to what

Well you are right about not tying general programs to file names that
can be parameters-in-some-sense to the program.

> function they perform, not what they contain -- what they contain is already
> obvious ;) I might call a template file
> '/usr/local/lib/<application group>/<application>.tmpl', but the variable
> would simply be called 'tmpl' or 'TEMPLATE' or some such.

Right, I suppose. But if I make a special short script to handle a
hard-coded file 'mother's recipes', why do I need to invent a new name
to hold the file object? I don't usually do that with imported modules
- despite they are file names in the same way, only difference is that
their creators know they'r going to be used as Python
identifiers. Mother doesn't necessarily know that.

Well I'm hesitating about file names my self. Maybe if the file system
was more integrated with the variables it would be more obvious.

Repeating a quote taking it in another context:

> function they perform, not what they contain -- what they contain is already
> obvious ;)

I'd agree at least for a variable that can contain different things
that perform similar things - one would want to name it according to
what is similar. Actually I think these new identifiers could be
mostly applicable to constants. Functions and other constants. 

Example: From the math constants: Looking in C's math.h you have
examples of how having more general identifiers would have
helped. There are constants for simple constants like pi, e, but also
2/sqrt(pi), sqrt(2), 1/sqrt(2) with non-obvious names like M_2_SQRTPI,
M_SQRT2, M_SQRT1_2. (They may follow a pattern but it's not THAT
obivous and not general enough to allow for more complicated
expressions, lacking parenthesises for example.)

>From Python's math.h I recall only having constants e and pi last
time I looked. I suppose more constants would be useful and maybe
it would help if we could define many of them with the 'obvious'
names so we don't have to think longer about what to call them...
and also not be tied to C's more or less arbitrarily subset of
useful constants.

> 
> > One drawback I can think of myself right now is a kind of uglyness
> > by using quotes but at least they make the construct stand out so it
> > should not be so confusing to ordinary id's.
> 
> The main drawback I see is your 'strong point': I don't believe allowing
> more characters in identifiers is going to make programs easier to read or
> follow, or going to make constructs more selfexplanatory. I just see added
> complexity with no obvious benifit.

Humm. Well, I am trying to understand why you are beliving and seeing
those things but I haven't managed yet. If you'd like to explain it
more it would be helpful, I would appreciate it but of course it's
about your time and effort.

I'am interested in this not only as a possible addition to Python but
also more in general, I am thinking about whether to add or not to add
this feature to a small language that I am making that is based on
Python. (Part of a bigger project.)

Thanks for your comments so far!

Sverker Nilsson



From nas at arctrix.com  Wed Mar 21 19:59:43 2001
From: nas at arctrix.com (Neil Schemenauer)
Date: Wed, 21 Mar 2001 16:59:43 -0800
Subject: New PEP: The directive statement
In-Reply-To: <ulmpzjrpb.fsf@cs.uu.nl>; from piet@cs.uu.nl on Wed, Mar 21, 2001 at 12:18:24PM +0100
References: <200103201042.LAA20248@pandora.informatik.hu-berlin.de> <mailman.985107140.4586.python-list@python.org> <9988ie02k7u@drn.newsguy.com> <ulmpzjrpb.fsf@cs.uu.nl>
Message-ID: <20010321165943.A25383@glacier.fnational.com>

On Wed, Mar 21, 2001 at 12:18:24PM +0100, piet at cs.uu.nl wrote:
> It would be much less of a joke when __future__ would be replaced by
> __features__, I think.

__features__ is a terrible name.  Its pretty obvious that you
(and a lot of other people on this list) don't understand what
__future__ does.

A small hint: pretend you are using 2.2 and want to use nested
scopes.  Will you be using "from __future__ import
nested_scopes"?  What if you are using 2.2 and don't want nested
scopes?

  Neil



From jmarshal at mathworks.com  Mon Mar 26 17:19:22 2001
From: jmarshal at mathworks.com (Joshua Marshall)
Date: 26 Mar 2001 22:19:22 GMT
Subject: Biggest float number?
References: <3ABFBB88.E718B2D@nd.edu>
Message-ID: <99of9a$q0c$1@news.mathworks.com>

C. S. Xu <xu.20 at nd.edu> wrote:
> Hi all,

> It seems the biggest floating number in Python can be:
>     2.**1024 - 1.
> which is about 1.8E308. Is there any way to get bigger value than this,
> just like long intergers almost have no limits?

One possible workaround is to use a pair of long ints as numerator and
denominator of a rational number.


From bedge at troikanetworks.com  Thu Mar 22 16:18:58 2001
From: bedge at troikanetworks.com (Bruce Edge)
Date: Thu, 22 Mar 2001 13:18:58 -0800
Subject: InteractiveConsole locals parameter, solved, but UGLY!
References: <3AB7FE07.5F7584F5@troikanetworks.com>
Message-ID: <20010322.131857.352118606.18970@mead.troikanetworks.com>

I found a way to do it, but, tell me there's a cleaner way:

change the run_python() as follows

def run_python():
  banner = "Use Ctrl-D to exit python and return to CLI"	
  import code
  import __main__
  locals = __main__.__dict__
  x = code.InteractiveConsole(locals)
  x.locals.update( { "run_python": run_python } ) <-- add this line !!!! 
  x.interact( banner )


In short, you have to manually add items to the namespace. At least this
is true for items which the application creates.

In article <3AB7FE07.5F7584F5 at troikanetworks.com>, "Bruce Edge"
<bedge at troikanetworks.com> wrote:

> I can't get my __main__.__dict__ passed to an InteractiveConsole:
> 
> def run_python():
> 	import code import __main__ locals = __main__.__dict__ x =
> 	code.InteractiveConsole(locals) x.interact("")
> 
> Shouldn't this have the same namespace as my main interpreter?
> 
> If not, how can I get my main interp namespace propagated down to my
> InteractiveConsole?
> 
> Are there any detailed docs on this?  I've read all the ones I can find.
> 
> Thanks, Bruce.
> 
> 
> Here's my test:
> 
> Python 2.0 (#42, Mar 15 2001, 13:24:46) 
> [GCC 2.95.3 20010219 (prerelease)] on linux2
> Type "copyright", "credits" or "license" for more information.
> 
>>>> from CLI import *	<--run my CLI app
> 
> k->  k-> python 		<-- command to start subshell, calls run_python, above
>>>> dir()		<-- this is all that's defined
> ['__builtins__', '__doc__', '__name__', 'atexit', 'histfile', 'os',
> 'readline', 'rlcompleter', 'savehist', 'sys']
>>>> 
>>>> 			<-- exit subshell
> k-> 			<-- exit app
>>>> dir()		<-- top level, this is what I want in my subshell:
> ['CommandException', 'KCompleter', 'Kcmd', 'Port', 'Zone',
> '__builtins__', '__doc__', '__name__', 'atexit', 'basedir', 'cNode',
> 'cOpt', 'cType', 'cli', 'cmd', 'cmds', 'copy', 'ctree', 'dprintf',
> 'dprintf_exception', 'dprintf_init', 'dprintf_level', 'histfile',
> 'histname', 'kcmd', 'kcompl', 'kcomplMode', 'kcompleter', 'kinterp',
> 'kuser', 'mType', 'main', 'modules', 'nType', 'os', 'port', 're',
> 'readline', 'register_new_command', 'rights', 'rlcompleter', 'savehist',
> 'sl', 'string', 'sys', 'tokens', 'topcmds', 'traceback', 'types',
> 'util']


From neilh at scintilla.org  Sun Mar 18 02:19:49 2001
From: neilh at scintilla.org (Neil Hodgson)
Date: Sun, 18 Mar 2001 07:19:49 GMT
Subject: Cameron Laird Quote on www.unixinsider.com -- Probably old news to y'all, but new to me.
References: <bJOr6.9481$TW.38047@tor-nn1.netcom.ca>
Message-ID: <pkZs6.8451$992.49916@news-server.bigpond.net.au>

Warren Postma:
> I found this quote on www.unixinsider.com's main page when I went there
> today:
>
> "I think we are seeing an end to the emphasis on statically typed
languages
> like C++ and Java. As this decade progresses, I expect to see an ever
> increasing use of dynamically typed languages, such as Python, Ruby, and
> even Smalltalk.
> ...
> They are much easier to refactor and have virtually zero
> compile time."
> --Robert C. Martin, talking to Cameron Laird in the Interviews forum

   C compilation time is also close to zero for much code now. For
'scripting size' problems where you have up to a few hundred lines in a
single file and fairly simple linkage you can press the go button in an
editor and see results starting immediately. For larger programs,
precompiled headers and incremental compilation and linking can reduce
edit-to-go time greatly.

   With Java, using a fast compiler such as IBM's will produce similar
development speed.

   Modern C++ that uses the standard C++ library and so requires a lot of
template expansion may slow down compilation to half a second but that is
still quite competitive with Python.

   This is one benefit of the 'scripting languages' that is rapidly
disappearing as machines get faster and C++ compilers are optimised.

   Neil




From bawolk at ucdavis.edu  Mon Mar 19 00:15:56 2001
From: bawolk at ucdavis.edu (Bruce Wolk)
Date: Sun, 18 Mar 2001 21:15:56 -0800
Subject: stripping punctuation from tuples
References: <3AB57F0C.594CF3E9@northwestern.edu>
Message-ID: <3AB5960C.C65EE8B6@ucdavis.edu>

How about this?
>>> a=[('Alan Tsang',),('George Bush',)]
>>> b = [i[0].split() for i in a]
>>> b
[['Alan', 'Tsang'], ['George', 'Bush']]

Cheers, Bruce

Louis Luangkesorn wrote:

> How do you strip the punctuation from tuples?  In particular, say I have
> a tuple a[10] (which happened to come from a not very well designed
> database)
>
> >>> a[10]
> ('Alan Tsang',)
>
> How do I get something to say 'Alan Tsang'  with out the parens or the
> ','
>
> Or, what I'm really trying to do is make this so
> >>>b = breakname(a[10])
> >>>b[0], b[1]
>
> 'Alan','Tsang'
>
> I've tried to do:
> >>> d = split(str(a[0]))
> >>> d
> ["('Aaron", "Fan',)"]
> >>> d[0]
> "('Aaron"
> >>> d[1]
> "Fan',)"
>
> But that is not quite what I want.
> Many thanks.
> Louis
> --
> K Louis Luangkesorn
> lluang at northwestern.edu  http://pubweb.nwu.edu/~kll560  PGP:0xF3E2D362
>
> Whatsoever things are true, ... honest, ... just, ... pure, ... lovely,
> ... of good report; if there be any virtue, and if there be any praise,
> think on these things.
>      - motto - Northwestern University (Phil 4:8)



From pnwlhc at cleaning-your-pc.com  Sat Mar 24 04:41:47 2001
From: pnwlhc at cleaning-your-pc.com (pnwlhc at cleaning-your-pc.com)
Date: 24 Mar 2001 09:41:47 GMT
Subject: KEEP YOUR PC CLEAN --- CLICK HERE  9879
Message-ID: <99hq4r$oi6$835@uranium.btinternet.com>

http://www.evidence-eliminator.com/go.shtml?A654710

You're in Serious Trouble - It's a Proven Fact! 

Deleting "Internet Cache and History" will NOT protect you because any of the Web Pages, Pictures, Movies, Videos, Sounds, E-mail, Chat Logs and Everything Else you see or do could easily be recovered to Haunt you forever! How would you feel if a snoop made this information public to your Spouse, Mother & Father, Neighbors, Children, Boss or the Media? It could easily Ruin Your Life! Solve all your problems and enjoy all the benefits of an "As New PC", Evidence Eliminator can Speed-Up your PC/Internet Browser, reclaim Hard Disk space and Professionally Clean your PC in one easy mouse click!

http://www.evidence-eliminator.com/go.shtml?A654710
mrkkjflwpwnlpxjdgkekseryyytiddxddbfgdhipgrdsntcdmqlkdubuigzheedjvnynsyuwkbftmuoebfbqylqhyu



From paulp at ActiveState.com  Sat Mar 10 06:43:15 2001
From: paulp at ActiveState.com (Paul Prescod)
Date: Sat, 10 Mar 2001 03:43:15 -0800
Subject: Urgent Question about Python licensing
References: <98cpfe$5oj$0@216.39.170.247>
Message-ID: <3AAA1353.7B7AA72D@ActiveState.com>

Dave LeBlanc wrote:
> 
>...
> My questions are these:
> 1. What prevents Python and GPL code from being distributed together?
> If I write some Python code that drives a GPL'd app (but  which does
> not depend on that app), what prevents me from offering both my python
> code and the GPL'd app on the same medium (disk/CD/web)?

Nothing. But many people want to build apps that combine GPL'ed code
with the Python code base.

> 2. Is it currently so, or will Python end up with, some license that
> requires that python scripts be covered by the GPL?

I'll go out on a limb and say that will never, ever happen. There is
very low interest in the Python community in promoting the philosophy
behind the GPL. Very little software in the Python world is licensed
under GPL-like terms.

And even if Guido were to jump on the FSF bandwagon, it would require a
whole new level of religion to require that *SCRIPTS* also use the GPL.
Even Richard Stallman would not suggest that Guido should be that
restrictive. Gnu compiler tools are not that restrictive.

Scripts are not at issue. The interpreter is at issue.

-- 
Python:
    Programming the way
    Guido
    indented it.



From aleaxit at yahoo.com  Wed Mar  7 05:11:17 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Wed, 7 Mar 2001 11:11:17 +0100
Subject: polymorphjsm &c (was Re: I come to praise .join, not to bury it...)
References: <mailman.983485923.1136.python-list@python.org> <97p7iq$kk2$1@nntp6.u.washington.edu> <97rka002k9@news1.newsguy.com> <9811a8$k5k$1@nntp6.u.washington.edu> <982cja0v5c@news1.newsguy.com> <983u1l$6mk$1@animus.fel.iae.nl>
Message-ID: <9851gf014vq@news1.newsguy.com>

"Carel Fellinger" <cfelling at iae.nl> wrote in message
news:983u1l$6mk$1 at animus.fel.iae.nl...
    [snip]
> >>> class MixedSplitter:
> ...     def __init__(self, splitter1, splitter2):
> ...         self.splitter1, self.splitter2 = splitter1, splitter2
> ...     def split(self, sequence):
> ...         s = self.splitter1(sequence)
> ...         return s[:-1] + self.splitter(s[-1])

self.splitter will give an AttributeError here (you mean .splitter2).
But you won't get here unless self.splitter1 is callable, and who
ever said it was?  Guess you mean self.splitter1.split(sequence) on
the first line, and similarly self.splitter2.split on the second.

With these changes, you get no errors any more out of your code:

> >>> ms = MixedSplitter(', ', ' and ')
> >>> ms.split('1, 2, 3 and 4')
> some nasty errormessage snipped

but neither do you get what you want, because a.split(b) works
irregularly...:

    -- if a is a re object, or if it's module-object os.path,
        then string b is split using splitter-object a
    -- if a is a string object (including a Unicode string),
        then string a is split using splitter-object b (which
        must also be a string)

I.e., the semantics are backwards in the two cases, alas.


> But the real evil is that no code out there will allow us to easily
> use the polimorphic nature of this splitter.

Actually, code written to split by a regular-expression object
(which it takes as an argument and only uses for splitting)
will work fine, polymorphically, with your MixedSplitter (when
the latter is amended to call the split methods of its splitter
attribute objects!-) -- as long as MixedSplitter is correctly
primed with two splitter objects, and string objects are not
splitter-objects in Python's architecture; they have a .split
method, but it doesn't work at all like those of regular expression
objects and the os.path module object do -- it's backwards.


> The moral being that we shouldn't argue about the uglines joiner.join
> but of the lack of a tru polymorphic splitter.split!

It's no doubt too late to remedy the discrepancy between
the semantics of a.split(b) depending on a's type (string
versus re) and the resulting irregularity (which we no
doubt will have to live with anyway) is unpleasant, to say
the least.


Alex





From loewis at informatik.hu-berlin.de  Tue Mar 20 15:32:53 2001
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 20 Mar 2001 21:32:53 +0100
Subject: re.match
References: <997skf$qoc$1@bmerhc5e.ca.nortel.com> <TrMt6.2490$4N4.318602@newsc.telia.net> <mailman.985114106.26896.python-list@python.org> <998bke$88i$1@bmerhc5e.ca.nortel.com>
Message-ID: <j4ofuwi3ka.fsf@informatik.hu-berlin.de>

msoulier at nortelnetworks.com (Michael P. Soulier) writes:

>   No biggy, but surprising compared to every other implementation
> of regexps that I've seen.

What implementations of regexps have you seen? /usr/include/regex.h
defines both re_search and re_match; I believe this is the POSIX
interface to regular expressions.

Regards,
Martin



From aahz at panix.com  Tue Mar 20 16:27:53 2001
From: aahz at panix.com (Aahz Maruch)
Date: 20 Mar 2001 13:27:53 -0800
Subject: Functionalism, aesthetics Was:(RE: I come to praise .join,  not
References: <mailman.985052844.17993.python-list@python.org> <997v08$s4o$1@panix3.panix.com> <qDMt6.2497$4N4.319619@newsc.telia.net>
Message-ID: <998i0p$621$1@panix3.panix.com>

In article <qDMt6.2497$4N4.319619 at newsc.telia.net>,
Fredrik Lundh <fredrik at effbot.org> wrote:
>Aahz Maruch wrote:
>> Carlos Ribeiro  <carribeiro at yahoo.com> wrote:
>>>
>>>Its unfortunate that in many cases the non-expert user keeps quiet,
>>>assuming that whatever opinion the "master" haves, its going to be better
>>>than theirs - after all they are the so-called experts.
>>
>> There's some truth to that -- but rarely on Usenet newsgroups.
>
>I doubt most experts bother to post to Usenet newsgroups
>(or read them, for that matter).

Fair enough...

>and yes, this is true for Python too...

...although I think it's less true for Python than some others, given
people like you in this fine newsgroup.  ;-)
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

"I won't accept a model of the universe in which free will, omniscient
gods, and atheism are simultaneously true."  -- M


From mcalla at home.com  Mon Mar 19 21:46:44 2001
From: mcalla at home.com (Mike Callahan)
Date: Tue, 20 Mar 2001 02:46:44 GMT
Subject: Catching exceptions in Tkinter's mainloop
References: <%dws6.9947$Q47.2723350@news1.rdc1.tn.home.com> <drMs6.32482$HR6.3694604@nnrp4.clara.net>
Message-ID: <owzt6.21061$Q47.5935326@news1.rdc1.tn.home.com>

I tried the following with no luck:

import Tkinter as T
import sys

class Gui:
    def __init__(self, master):
        T.Button(master, text='Crash2', command=self.crash).pack()

    def crash(self):
        z = zz

class Catcher:
    def __init__(self, func, subst, widget):
        self.func = func
        self.subst = subst
        self.widget = widget

    def __call__(self, *args):
        try:
            if self.subst:
                args = apply(self.subst, args)
            return apply(self.func, args)
        except SystemExit, msg:
            raise SystemExit, msg
        except:
            import traceback
            traceback.print_exc()
            self.widget.quit()

root = T.Tk()
f = open('test.log', 'w')
sys.stderr = f
T.CallWrapper = Catcher
app = Gui(root)
root.mainloop()

What am I doing wrong?

"Richard Townsend" <NOstarfighterSPAM at freeuk.com> wrote in message
news:drMs6.32482$HR6.3694604 at nnrp4.clara.net...
> "Mike Callahan" <mcalla at home.com> wrote in message
> news:%dws6.9947$Q47.2723350 at news1.rdc1.tn.home.com...
> > I want to make sure that if any exception pops up inside a mainloop,
that
> > exception is printed to a log file and the mainloop quits. I tried to
> > simulate this with the following code but failed:
> >
> > from Tkinter import *
> >
> > class Gui:
> >     def __init__(self, master):
> >         self.master = master
> >         self.button = Button(master, text='Crash', command=self.crash)
> >         self.button.pack()
> >
> >     def crash(self):
> >         z = zz  # trigger exception
> >
> > root = Tk()
> > app = Gui(root)
> > try:
> >     root.mainloop()
> > except:
> >     print 'caught'
> >     root.quit()
> >
> > It doesn't work. The exception is printed out by standard traceback and
> the
> > mainloop continues. What am I doing wrong?
> >
> > Mike Callahan
> >
>
> Tkinter catches exceptions raised in callbacks and Tk by using the class
> 'CallWrapper' (defined in Tkinter.py). I copied the technique used by Pmw,
> and created my own CallWrapper class (based on Tkinter.CallWrapper) and
then
> after initialising Tk, I use the following statement:
>
>     Tkinter.CallWrapper = MyCallWrapper
>
> MyCallWrapper's __call__ method invokes my error logging function in the
> final except clause, you could get yours to print info to your log file
and
> quit.
>
> Note, you must always have an except clause for SystemExit (like
> Tkinter.CallWrapper does) otherwise calling sys.exit() in a callback won't
> work!
>
> e.g.
>
> class MyCallWrapper:
>     """Internal class. Stores function to call when some user
>     defined Tcl function is called e.g. after an event occurred."""
>     def __init__(self, func, subst, widget):
>         """Store FUNC, SUBST and WIDGET as members."""
>         self.func = func
>         self.subst = subst
>         self.widget = widget
>     def __call__(self, *args):
>         """Apply first function SUBST to arguments, than FUNC."""
>         try:
>             if self.subst:
>                 args = apply(self.subst, args)
>             return apply(self.func, args)
>         except SystemExit, msg:
>             raise SystemExit, msg
>         except:
>             # substitute your functionality here...
>
> Richard
>
>
>
>




From Arthur_Siegel at rsmi.com  Fri Mar 23 15:02:53 2001
From: Arthur_Siegel at rsmi.com (Arthur_Siegel at rsmi.com)
Date: Fri, 23 Mar 2001 14:02:53 -0600
Subject: Subject: Re: VPython/Tk app works on single processor but not dual
Message-ID: <005B2EE9.N22121@rsmi.com>

Fred wrote -
>If you've got a VC++ project file, I'd like
>to have that.  Is the VPython build process for Win32 documented
>anywhere?

The source distro at:

http://virtualphoton.pc.cc.cmu.edu/projects/visual/developers.htm

has a VC++ project file.

Build was very straightforward.  Redirect to my Python20 include and
libs and shoot.

>This problem is getting in the way of my general goal:  creating
>interactive programs that allow the user to play with graphic
>representations of mathematical concepts.  

As you might know from my edu-sig ramblings, I am using VPython
to similar end - concentrating on pure geometric rather than number based
math concepts. 

You will have something to share sometime soon?  






From mud at mud...  Tue Mar 13 10:57:50 2001
From: mud at mud... (mudshark)
Date: 13 Mar 2001 15:57:50 GMT
Subject: Einstein's Riddle
References: <oNPq6.396644$ge4.138479566@news2.rdc2.tx.home.com> <98glrd$1si1a$1@ID-75892.news.dfncis.de> <3AABE712.499CCF4C@home.com> <98gp5b$1u0oi$1@ID-75892.news.dfncis.de> <98lf4g$2dqj2$1@ID-63952.news.dfncis.de>
Message-ID: <98lg1u$iuh$1@porthos.nl.uu.net>

"J. A. Durieux \(reply from http://www.biep.org/\)" <bogus at ddress.com> 
wrote in <98lf4g$2dqj2$1 at ID-63952.news.dfncis.de>:

>"Boris Popov" <no-spam-popov at bigserver.com> wrote in message
>news:98gp5b$1u0oi$1 at ID-75892.news.dfncis.de...
>> P.S. What's emu ?
>
>A European currency unit, now replaced by the Euro..
>

That's ecu (echo? echo echo echo ....)



From tim.one at home.com  Sat Mar 10 14:21:18 2001
From: tim.one at home.com (Tim Peters)
Date: Sat, 10 Mar 2001 14:21:18 -0500
Subject: "!=" is bad form. Re: sorry....never mind
In-Reply-To: <15018.30583.852611.836468@anthem.wooz.org>
Message-ID: <LNBBLJKPBEHFEDALKOLCCELJJEAA.tim.one@home.com>

[Barry A. Warsaw]
> Personally, I agree.  I hate "!="; to me <> looks much better and I
> use it all the time.  Guido has been known to disagree with me so
> violently that he'll make a patch to my code just to change that. :)

I'm afraid it's again time to quote the Reference Manual:

    The forms <> and != are equivalent; for consistency with C, != is
    preferred; where != is mentioned below <> is also accepted. The
    <> spelling is considered obsolescent.

Now I can channel many things in Guido's absence, but not even by asking
outright have I been able to divine the differences among "obsolescent",
"obsolete" and "deprecated" when applied to snake scales.  All we can be sure
of is that they each refer to one of the modes in which a python seeks to
shed old skin.

If you look at the CVS history of Grammar/Grammar, you'll discover that In
The Beginning, Python allowed only the <> spelling.  != wasn't added until
revision 1.5 (of Grammar, not Python).  So this is a curious case where Guido
disowned his first impression (from only <>, to <> is obsolescent); but since
this is the same process by which your print>>file got into the language, you
had better not protest too loudly <wink>.

print>>-got-in-for-"consistency-with-c"-too-ly y'rs  - tim




From syver at NOSPAMcyberwatcher.com  Tue Mar 27 14:06:56 2001
From: syver at NOSPAMcyberwatcher.com (Syver Enstad)
Date: Tue, 27 Mar 2001 21:06:56 +0200
Subject: Threading legality/morality
References: <fsDv6.1247$NR.100201@news3.oke.nextra.no> <99nnbv$phs$1@panix6.panix.com> <h0%v6.1842$NR.144463@news3.oke.nextra.no> <99qa3v$6e0$1@panix6.panix.com> <Pine.SOL.4.30.0103271851370.7646-100000@mimosa.csv.warwick.ac.uk>
Message-ID: <99qogp$o6g$2@troll.powertech.no>

"John J. Lee" <phrxy at csv.warwick.ac.uk> wrote in message > Perhaps you're
right, but I've seen the same behaiviour on win95, without
> multi-threading.  When you run command-line (do we mean the same thing by
> console and command-line?) apps it freezes until a key gets pressed.  Not
> using urllib either.
>
> It is possible (likely, even) that I was using cygwin, so maybe that's
> where the blame lies.  I'll see if I can reproduce it with the standard
> DOS shell.

I am using console program/command line whatever you like, and I am not
using Cygwin. I've experienced this both on the BeOpen and ActiveState
distros.




From nomad*** at ***freemail.absa.co.za  Fri Mar  2 02:56:06 2001
From: nomad*** at ***freemail.absa.co.za (nomad)
Date: Fri, 02 Mar 2001 07:56:06 GMT
Subject: [Q]: How can I get the logged in users username on Win32?
Message-ID: <3a9f50fd.58862977@news.is.co.za>

Howzit,

After searching through the Python Docs, I found that on Un*x I could
use getpass.getuser(), but there is no equivalent for the Win32
platform.

In other words, how can I get the username in Windows?

As a side note, how can I enable Windows Scripting Host support for
python - there is mention of python in the original WSH docs - and if
I can use python scripts in WSH how would I call the built-in
Wscript.<methods>

TIA

nomad


From bob_cannard at mentor.com  Tue Mar 27 13:46:40 2001
From: bob_cannard at mentor.com (Bob Cannard)
Date: Tue, 27 Mar 2001 10:46:40 -0800
Subject: R: unique items in lists
References: <_nZv6.2878$yU4.468236@news.infostrada.it> <slrn9c0ph1.o5h.scarblac@pino.selwerd.nl> <t5_v6.3466$yU4.477532@news.infostrada.it> <slrn9c1424.o5h.scarblac@pino.selwerd.nl>
Message-ID: <3AC0E010.3CDC94FC@mentor.com>


Remco Gerlich wrote:
> 
> comp.lang.python <micampe at f2s.com> wrote in comp.lang.python:
> > Here is what I'm doing now:
> >
> >         for w in words:
> >             while words.count(w) > 1:
> >                 words.remove(w)
> >
> > comments?
> 
> First, you modify a list that you are looping over. That can give
> unpredictable results, but in this case it seems to work.

The list [1,1,2,2,3,3,2,1] breaks this algorithm, which produces
[3,3,2,1]. The reason it fails is that the algorithm preserves the
last element of each value, not the first, so the 3's are moving
to the left while the for loop's internal pointer is moving to the
right; the two cross each other without ever meeting. Heed that
warning in the language reference; iterating over a changing list
can seriously damage your mental health, not to mention your
algorithm!

> It will take a long time. 'words.remove(w)' takes time proportional to the
> length of the list; words.count(w) takes time proportional to the length of
> the list itself, and you repeat that for every duplicate of a word, and the
> total is repeated for every unique word. In total it takes time proportional
> to N^3 (for a list of length N), I think.

That's right. It's a horrific run time for any reasonably large N:
a billion operations for N = 1000.

> The dictionary method does one dictionary action for each word, and then it
> does a dict.keys() which takes time proportional to each unique word.

I don't know how dictionaries are implemented, but hopefully they're
a lot faster than that - one would expect a hash or tree implementation,
giving a logarithmic run time for each lookup. A linear implementation
would be an obvious cause of poor run time performance for Python
generally.

> In total
> it takes time proportional to N.

More than that; it would be order N for a list such as [1,1,1,1,1...]. I
would estimate something between N and N log N depending on the input
data and the implementation of dictionaries. In any case, it's vastly
better than the previous method: only several thousand operationsfor
N = 1000. In the worst case (a linear implementation of dictionaries
plus pathological data) the run time is still only N^2, or a million
operations for N = 1000.

Cheers,

        Bob.


From robin at jessikat.fsnet.co.uk  Thu Mar 22 11:52:03 2001
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Thu, 22 Mar 2001 16:52:03 +0000
Subject: extending the built in dictionary type
Message-ID: <umfKujAz2iu6EwHa@jessikat.demon.co.uk>

I'm interested in extending the built in dictionary. It's fairly simple
to wrap, but seems incredibly hard to chain the methods. Is there any
obvious way to use most of the built in methods whilst modifying a few.

Seems that although it exists I can't chain the original mapp_methods of
dictobject as that is not visible outside pythonxx.dll. So is there an
easier way to get at these internal types?
-- 
Robin Becker


From steven at haryan.to  Sun Mar 11 10:28:52 2001
From: steven at haryan.to (Steven Haryanto)
Date: Sun, 11 Mar 2001 22:28:52 +0700
Subject: Going ZODB
Message-ID: <4.3.2.7.2.20010311222003.02c2aad0@202.53.255.51>

I am curious as to how many people in this list actually
use ZODB in production environment and what is the typical
uses of it (especially ZODB with relational/berkeleydb
storage). Aside from along with Zope, of course.

Also, there are one issue I am considering before going
ZODB. If later on the application grows beyond the scale
that ZODB can provide at the moment, would it be
awkwardly hard to convert the code to use other backend.
For example in the relational world, if I keep the SQL
statements portable enough, I can just, theoretically,
switch the backend with another DB.

Thanks,
Steve




From fredrik at pythonware.com  Wed Mar 14 02:44:27 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 14 Mar 2001 07:44:27 GMT
Subject: Unpickling from strings
References: <qHxr6.19658$pR3.3830844@news2-win.server.ntlworld.com> <3gyr6.159$4d3.34762@newsb.telia.net> <3AAEB597.4010007@ntlworld.com>
Message-ID: <vjFr6.180$4d3.41807@newsb.telia.net>

Marcin Chady wrote:
> Both pickling and unpickling is done by the same script, and the classes
> are defined in a separate module imported by that script.

should work.  better post some code.

Cheers /F




From claird at starbase.neosoft.com  Fri Mar 23 17:36:13 2001
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 23 Mar 2001 16:36:13 -0600
Subject: Install Python on Apache
References: <99fph5$a2n$1@rex.ip-plus.net>
Message-ID: <EDC008F28CADC736.9881C22ECC137D33.5745C372F32CC002@lp.airnews.net>

In article <99fph5$a2n$1 at rex.ip-plus.net>, Sharp <s.sharp at fh-aargau.ch> wrote:
>Hi Everybody
>
>I sucked down Python from python.org and started to install the program
>on a win2000 apache system. that has worked perfectly. but then I tried
>to set up the apache httpd.conf file with the following options
>
>AddType application/x-httpd-py .py
>Action application/x-httpd-py /cgi-bin/Python/python.exe
>
>(cgi.bin is set to the right place because php and perl do work)
>
>But then it did not work. I hava also tried several variations but always
>with the same comands.
>The Error Message in the Apache error.log was
>
>[Fri Mar 23 15:54:50 2001] [error] [client xxx.xxx.xxx.xxx] Premature end of
>script headers: c:/web/cgi-bin/python/python.exe
>
>the script is looking like this
>
>#!c:/web/cgi-bin/Python/python.exe
>print "<html>";
			.
			.
			.
I confess I've never used Apache on Win*.  I'm
willing to bet, though, that starting this with
  #!c:/web/cgi-bin/Python/python.exe
  print "Content-type: text/html\n"
  print "<html>";
     ...
instead will get you closer to what you want.
-- 

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


From petasis at iit.demokritos.gr  Mon Mar 12 01:58:48 2001
From: petasis at iit.demokritos.gr (Petasis George)
Date: Mon, 12 Mar 2001 08:58:48 +0200
Subject: Einstein's Riddle
References: <oNPq6.396644$ge4.138479566@news2.rdc2.tx.home.com> <UdSq6.2184$54.2166@www.newsranger.com> <984380185.768920@news.aei.ca>
Message-ID: <3AAC73A8.F12BC8B3@iit.demokritos.gr>

sulfugor wrote:
> 
> Gregory Jorgensen wrote in message ...
> >The 2% that can solve it must be psychic or deranged. The riddle as posed
> is not
> >solvable.
> >
> >We are told of five houses anf five people. The hints mention five colors
> (red,
> >green, white, yellow, blue), five drinks (tea, coffee, milk, beer, water),
> five
> >smokes (Pall Mall, Dunhill, Blends, Bluemaster, Prince), and five
> nationalities
> >(Brit, Swede, Dane, Norwegian, German). But only four types of pet are
> >mentioned: dogs, birds, cats, horse. We have no evidence that any of the
> five
> >owns fish.
> 
> But the fish is mentionned in the question before the clues are given. We
> are able to find who owns each of the 4 pets mentionned in the hint list. is
> it so unreasonable to assume then that the last pet is a fish, since we were
> asked about it ? The question IMplied that one of the 5 people owned a fish.
> 
> sulfugor
> 
> ps = by the way, how long is this supposed to take ( solving the riddle) ?
> 
Yes, that the problem: it implies, it does not states :-)
It took me about 20 minutes to solve it
(with paper+pencil of-course :-))...

George


From aleaxit at yahoo.com  Fri Mar 16 10:10:19 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 16 Mar 2001 16:10:19 +0100
Subject: I come to praise .join, not to bury it...
References: <mailman.983485923.1136.python-list@python.org> <97p7iq$kk2$1@nntp6.u.washington.edu> <97rka002k9@news1.newsguy.com> <3AAF7E47.294617E1@sage.att.com> <98qeco02hn7@news1.newsguy.com> <98rmod$c8p$1@animus.fel.iae.nl>
Message-ID: <98tacr0q5s@news1.newsguy.com>

"Carel Fellinger" <cfelling at iae.nl> wrote in message
news:98rmod$c8p$1 at animus.fel.iae.nl...
    [snip]
> > some people (just as it is going to attract others), so,
> > it's not a clear "win" either way, even if you just frame
> > it as a popularity-race.  Focus on *substance*, on the
>
> but if you get too blinded by this substance thing, you might easily
> overlook that in the end you have to weight things and there again the
> seasoned sense of aesthetics of the gifted ones is what makes the
> difference.

I appreciate "form follows function" as an aesthetic principle.

Clearly, given such masterpieces of functional spareness as
Bauhaus and other artistic schools of this century have been
able to produce, I'm not alone in this (and the spare functional
beauty of Medieval Francescan churches, etc, stands to prove
that this is not by any means an idea new to this century; also,
of course, it's NOT by any means an idea limited to Western
culture, as for example a certain Zen-inspired style of Japanese
furniture shows).

Just as clearly, given such masterpieces of rich ornamentation
as great Baroque architects (etc, etc -- again, this is not
limited to one century, or one continent!) have produced, it
is anything but a _universally accepted_ guiding principle for
the fashioning of beauty in human artefacts.

You seem to imply some moderation (have to weight things) is
key to YOUR aesthetics -- fine, Aristotle will agree, as will
no doubt others, but much breathtaking beauty is instead in
works which pushed their guiding principles to absolute
extremes (Bach's "Art of the Fugue", Picasso's "Guernica",
Olbrich's "Sezession Haus", Leopardi's "Infinito" -- where
is any 'moderation' or 'weighting' in these and many other
masterpieces across all arts and centuries and continents?!).

In my own profession, I make certain choices which I perceive
as guided by 'aesthetics' when I cannot afford to articulate
in full detail the path back from the practical example up to
the guiding-principle; my experience (and 'gift', if any) may
help this synthesis (a productivity plus) by helping me
perceive as "elegant" an architecture which will prove helpful
to the design-task at hand.  If and when challenged, though, I
can and will fall back to the time-consuming, perhaps boring,
but often enlightening task of detailed analysis of the
constituent elements -- there is nothing 'inherently ineffable'
in this.  I see architects of manufacts with more psychological
interactions, such as GUI's or buildings, perform similarly --
a sense of beauty to guide their everyday work (as reasoning
back to first principles would take forever and slow things
down unacceptably), but solid, explainable reasons they can
trace back if and when needed... *IF* they're worth a dime as
building-architects and/or GUI-architects.  And for the only
'art per se' I _do_ know a little bit about, cinema, I see a
very similar pattern again -- yet farther removed, OK, but a
great movie can be fully analyzed (and IS, over and over again,
in all sorts of debates and critical publications).

That doesn't mean it's possible to make somebody _without_
appropriate gifts into a great movie director, GUI designer,
building architect, OR software architect, of course -- but,
it DOES mean each of these professionals need not fall back
on vague generalities and empty hand-waving if and when their
design choices get challenged.

Somebody who's unable to do anything but talk of aesthetics
when discussing a choice in the architecture of a software
component is simply not contributing to the discussion.  It
looks good to you, it looks bad to him, it looks indifferent
to her, yawn, great debate, now let's get beyond this to the
REAL meat of the issue, please -- the technical implications
of this, that, or the other choice in a specific matter.

If you _can't_ get down to concrete specifics, then either
you're not investing enough time & energy to contribute
(in which case, staying out of a debate WOULD be a nice
matter of courtesy), or you lack some specific (experience,
culture, professional background, whatever) to be able to
analyze and express your surface impressions in the issue --
in which case, it's quite possible that they're misfounded,
and, IF the huge investment of time and energy doesn't
faze you, in-depth analysis to see if they ARE might well
be warranted (and that is basically what I've been doing
throughout this thread -- striving to help others articulate
their objections, in great detail, by first laying out in
just as much details my reason for NOT objecting at all to
the architecture Guido has chosen for .join; it won't of
course have any direct repercussion on how Python behaves
in the matter, but it MAY have positive implications on
technical growth of the people involved, and thus, very
indirectly, on future efforts on their part).


> > other hand, is going to be a win for all who _are_ involved
> > with Python -- not just those who happen to be on the
> > "I like it" side vs the "I like it not" one (and I've
> > never seen any argument about _aesthetics_ [as opposed to
>
> I hope you'll see the argument in the above:)
>
> > _usability_ -- human-factors studies DO have a substantial
> > side too:-)] that doesn't boil down to such trifles]).

Nope: it _still_ boils down to aesthetics being AT BEST
a mental and verbal shorthand to avoid the considerable
investment required to articulate the concrete underlying
realities in detail.  AT WORST (and, I suspect, more often,
particularly in Usenet debates) a way for somebody to keep
spewing words on a subject on which they actually have
nothing concrete, constructive, and useful to say.


Alex





From benjamin.altman at noaa.gov  Wed Mar 21 16:13:59 2001
From: benjamin.altman at noaa.gov (Benjamin.Altman)
Date: Wed, 21 Mar 2001 16:13:59 -0500
Subject: learning python...
References: <3AB8F4E1.184DD9C@noaa.gov> <Z97u6.16733$Im6.1761313@newsread1.prod.itd.earthlink.net> <3AB91006.6D7E35F2@seatech.fau.edu>
Message-ID: <3AB91997.3ECD003C@noaa.gov>

Do you know how it compares to the first edition in terms of presentation and
indexing?  I have heard from a few people that it is difficult to look up things in
the 1st edition.  The only review on amazon of the 2nd ed. claims it is a complete
rewrite despite his admitance that he has never read the book...

Thanks,
Ben

Benoit Dupire wrote:

> Simon Kesenci wrote:
>
> > If you know programming, forget _Learning_ _Python_.  I tried _Programming_
> > _Python_ last year, but it covered version 1.3.  A good book if they
> > updated it since then.  As a programmer, you should find the online
> > tutorial easy and helpful <http://python.org/doc/current/tut/tut.html>.
>
> Programming Python was updated and now covers Python 2.0
> I am reading it, actually, and, till now, I find it great.. (but I am only page
> 35! ;o) ).
>
> This edition is intended to be the 'sequel' to " Learning Python"... so yes,
> you can try to read the Python tutorial and this book...



From bkc at Murkworks.com  Thu Mar  1 09:01:58 2001
From: bkc at Murkworks.com (Brad Clements)
Date: Thu, 1 Mar 2001 09:01:58 -0500
Subject: [ANNOUNCE] pyAda, first public release
References: <3A9DB9E1.8DD0E36D@bigfoot.de>
Message-ID: <Xvsn6.7$R4.940@news.clarkson.edu>

I thought one of the big points of using Ada was its verifiability -- making
it more suitable for complex, "large-scale" (i.e, expensive) projects, like
missile defense.

Plugging in a "non-verifiable" script plug-in kinda defeats this purpose..
doesn't it?


--
Brad Clements,
bkc at murkworks.com
"Gerhard H?ring" <gerhard.nospam at bigfoot.de> wrote in message
news:3A9DB9E1.8DD0E36D at bigfoot.de...

> PS: This is my first major Ada project, so I am open to any suggestions
> for improvement.





From aleaxit at yahoo.com  Sat Mar 17 03:13:26 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sat, 17 Mar 2001 09:13:26 +0100
Subject: Pep 238 Why new // operator (Non-integer Division)
References: <mailman.984811161.25962.python-list@python.org>
Message-ID: <98v77t01qcg@news2.newsguy.com>

"Tim Peters" <tim.one at home.com> wrote in message
news:mailman.984811161.25962.python-list at python.org...
    [snip]
> > So maybe some other symbol should be used.
>
> Why bother?  Python's # looks an awful lot like C++'s preprocessor
directive
> flag too.  Guido would probably prefer infix "div" (a la Pascal), but a
new
> keyword is a very hard sell.

So what's so terrible about non-infix notation, a builtin div(a,b) [and
mod(a,b)
too since we're at it...?].  Sure, infix can be handy -- but then this might
be
the occasion to steal a leaf from Haskell's book... let any 2-operand
function
be used as an infix operator (possibly with some stropping if that is needed
to disambiguate syntax).


Alex





From danielk at aracnet.com  Sat Mar 24 13:19:47 2001
From: danielk at aracnet.com (Daniel Klein)
Date: Sat, 24 Mar 2001 10:19:47 -0800
Subject: Komodo
Message-ID: <i0ppbt0iqse4shuofjr4dfmf6eqt85u12l@4ax.com>

I just spent that last couple of hours getting comfortable with the ActiveState
Komodo IDE. Let me say "Nice job ActiveState!". I especially like the project
management and the ability to expand/collapse the classes/functions in-place.

Good work!

Daniel Klein
Portland OR USA


From emile at fenx.com  Sun Mar 25 12:29:38 2001
From: emile at fenx.com (Emile van Sebille)
Date: Sun, 25 Mar 2001 09:29:38 -0800
Subject: Newbie needs help to find bug in his first Perl program
References: <3ABCC757.6C6F17CA@earthlink.net>
Message-ID: <99la4f$1pl7q$1@ID-11957.news.dfncis.de>

>>> print "%y" % 1
Traceback (innermost last):
  File "<pyshell#0>", line 1, in ?
    print "%y" % 1
ValueError: unsupported format character 'y' (0x79)
>>>

HTH,

--

Emile van Sebille
emile at fenx.com

---------
"Ron Stephens" <rdsteph at earthlink.net> wrote in message
news:3ABCC757.6C6F17CA at earthlink.net...
> Ok, I am a total newbie. I took some programming course in college, did
> some relatively simple programs in Fortran, Basic, PL/1 and even one in
> assembler for an old IBM "mainframe" using punched cards. But that was
> along time ago.
>
> OK, I have just fallen head over heels in love with Python. Wrote my
> first little "program", about 200 lines of a thing to do simple
> "Decision Analysis" in a general sort of way, to take input from an
> "expert" in any given (simple) field, and set up criteria based on the
> experts experience, then calculate a weighted average to decide which
> option to choose.
>
> I find the debugging to be extraordinarily good, but I have one bug I
> cant seem to figure out yet. Been up all night its noon of the next day
> ;-))) Need some sleep
>
> The offending line of code is
>
> sixtwo = raw_input ("where does %s rank for %y ?" % (option6,
> criteria2))
>
> Can anyone see what is wrong with this? I have several lines like this
> and they all give me error messages.
>
> Sorry for the trivial newbie nature of my question and first post ;-)))
>
> But I do luv Python....
>




From jarober at mail.com  Sun Mar  4 15:37:26 2001
From: jarober at mail.com (James A. Robertson)
Date: Sun, 04 Mar 2001 20:37:26 GMT
Subject: Collection interfaces
References: <3a90fa97.0@news1.mweb.co.za> <ud7cakvi8.fsf@roundpoint.com> <MPG.150312ceb68f1ada98969f@news.earthlink.net> <8Gem6.473353$U46.14105062@news1.sttls1.wa.home.com> <MPG.1503548620e632919896a4@news.earthlink.net> <MCjm6.474652$U46.14163288@news1.sttls1.wa.home.com> <MPG.150394195f7f04a19896a7@news.earthlink.net> <3A9A86B1.9A8C69A2@ix.netcom.com> <MPG.1504d14563dce6499896af@news.earthlink.net> <97fjla128ir@news2.newsguy.com> <MPG.1505a76831c34b749896b3@news.earthlink.net> <97h5s40vhc@news2.newsguy.com> <MPG.1506007591cf818b9896bc@news.earthlink.net> <slrn99opop.gr.amk@mira.erols.com> <MPG.1506265ebf6d934d9896be@news.earthlink.net> <97ibmc0k47@news1.newsguy.com> <MPG.1506cf3293b094b49896c1@news.earthlink.net> <97m0c7080c@news2.newsguy.com> <MPG.15085517debf4b649896ca@news.earthlink.net> <zb%n6.31033$o%6.586026@news1.rdc1.bc.home.com> <MPG.150b383cea226d409896d0@news.earthlink.net>
	 <kNko6.6986$eq3.707121@news.uswest.net> <MPG.150c46db9cec283e9896d4@news.earthlink.net>
Message-ID: <3AA2A898.450B659E@mail.com>

Topmind wrote:
> 

<snip>

> 
> Not exactly "typical custom business applications", the domain
> in which I focus most of my whining.
> 
> The percentage of OO fans in the business domain seems lower
> than average. Perhaps this says something about OO and
> the business domain.
> 

90% + of the customers using VisualWorks are writing 'typical' business
apps.  

> > Please keep track of the number of minutes you spend working on this.
> > Please keep track of the number of defects that you find and fix during your
> > work.  I mean, any thing that does not work the first time and be sure to
> > count each problem the compiler tells you about too.
> 
> And if we use an interpreter?
> 
> > Please keep track of
> > the number of statements of code you add, remove, and change during this
> > work.  If you supply me with these data and the code I will test the
> > accuracy of your solutions and compare your code with some OO code and post
> > my findings (as my spare time allows) for the entire newsgroup(s) to see.
> >
> > <TongueInCheek>Oh... and a... no pair programming please!</TongueInCheek>
> >
> > Randy
> >
> >
> 
> -tmind-

-- 
James A. Robertson
Product Manager (Smalltalk), Cincom
jarober at mail.com
<Talk Small and Carry a Big Class Library>


From NoSpam at NoSpam.com  Thu Mar  1 19:35:25 2001
From: NoSpam at NoSpam.com (Tom)
Date: Fri, 02 Mar 2001 00:35:25 GMT
Subject: Unresolved symbols in MSVCRT.DLL when running a py2exe packaged program.
References: <RKln6.6630$hn5.853203@news1.rdc1.mb.home.com>
Message-ID: <hVBn6.7888$UZ4.3012833@news4.rdc1.on.home.com>

One difference (between the Win2K & WinNT machines) is the new system file
protection 'feature' in Win2K.  It is supposed to disallow inappropriate
updates to system DLL's (of which MSVCRT.DLL is one).  Another aspect of
this feature is that apps can load their own versions of system DLL's, so
multiple different versions MSVCRT.DLL can be memroy resident at the same
time (but not on NT, just 2K).

Someone suggested deleting the DLL to see if it is in use.  I'm not sure
about DLL's, but with drivers (.SYS) you can delete it even while it is in
use (wierd).  There are utilities around that will tell you what DLL's your
app has loaded.  Maybe even the sysinfo utility, included in Win2K, might
tell you this?

When you replace the MSVCRT.DLL in system32, you will have to reboot before
it takes effect.

Tom.

"Parzival Herzog" <parz at home.com> wrote in message
news:RKln6.6630$hn5.853203 at news1.rdc1.mb.home.com...
> I have a Python application (which uses the wxPython extension) that I
have
> packaged
> for distribution with py2exe.  The dfistributed app works on my own
> computer,
> and on another machine I have, which has never seen any Python software,
nor
> any
> development software (i.e. no dll's not a part of a standard Win2000
> installation are present).
>
> I delivered my app to my client, who is expected to modify some
> configuration code, then
> use py2exe to package the app, and run the packaged app on a server
machine.
> My client
> has a notebook computer, running Win NT (Sp 6), plus MsVc and assorted
> development tools.
> My app worked on that noebook, and py2exe again packaged the app and the
> package was
> copied to the server machine, a Win NT sp 6 desktop computer. When we
> attempted to
> start the app, it failed immediatly, with the message that the symbol
> ??_U at YAPAXI@Z could
> not be found in MSVCRT.DLL.
>
> That symbol is apperrently defined in the MSVCRT.DLL file in the py2exe
> package, but the
> MSVCRT.DLL in the package is not the same as MSVCRT.DLL in the windows
> system32
> directory, and that version of the dll appears to have the symbol
> ??2 at YAPAXCI@Z defined
> (what horrible identifiers!). So we assumed that instead of finding the
> py2exe supplied
> dll, the packaged python app is finding the other dll, which is much older
> (dated 1997).
> Of course, we don't have any way of determining the dll that is actually
> being searched).
> So, we attempted to replace the old MSVCRT, by installing a recent copy of
> MSVC++ 6.0.
> This indeed did replace the dll, replacing it with a version dated July 7,
> 2000. However,
> the problemwith the python app persisted exactly as before.
>
> What I can tell, with the computers I have available, is that the the
Python
> app works
> on three different machines, 2 running Windows 2000, with the latest
> MSVCRT.DLL
> and one running Win NT (sp 6), with an MSVCRT.DLL dated 1999. What I can't
> tell
> is whether or not the py2exe packeged MSVCRT.DLL is being used by the
python
> app,
> or the system32 dll that happens to be on the machine where the app is to
> run.
>
> Can anyone help me towards resolving this problem? Truly, I program in
> Python, because I
> really am far too stupid to cope with ??2 at YAPAXCI@Z missing symbol
problems,
> and I
> fancy that I shouldn't have to either. But I appear to be misguided, and I
> need some help.
>
> - Parzival
>
>
>
>




From scarblac at pino.selwerd.nl  Thu Mar 22 10:49:59 2001
From: scarblac at pino.selwerd.nl (Remco Gerlich)
Date: 22 Mar 2001 15:49:59 GMT
Subject: New PEP: The directive statement
References: <mailman.985085006.24236.python-list@python.org> <slrn9bi7q3.2sa.hzhu@rocket.knowledgetrack.com> <j4snk6yxiv.fsf@informatik.hu-berlin.de>
Message-ID: <slrn9bk7k0.fnm.scarblac@pino.selwerd.nl>

Martin von Loewis <loewis at informatik.hu-berlin.de> wrote in comp.lang.python:
> Generalization is a dangerous thing: you can too easily cross the
> border to over-generalization.

This is exactly what 'directive' is trying to do, lumping transitional
language changes in with options.

-- 
Remco Gerlich


From flognat at flognat.myip.org  Thu Mar 15 18:41:04 2001
From: flognat at flognat.myip.org (Andrew Markebo)
Date: 16 Mar 2001 00:41:04 +0100
Subject: Should I close after popen??
References: <XFMail.20010315142704.shaleh@valinux.com> <mailman.984697040.2458.python-list@python.org>
Message-ID: <m3hf0uppm7.fsf@master.athome>

/ "Ken Seehof" <kens at sightreader.com> wrote:
| [...]
| Your usage of wait is fine, (assuming that the fd1 process is never
| dependent on the fd2 process).

Hmm is it possible to do some kind of wait (fd1, fd2) to continue when
either of them are finished, or thats what select and poll are for?? 

Hmm poll seems to be a choice..  

        /Andy


From cribeiro at mail.inet.com.br  Sat Mar 31 22:01:43 2001
From: cribeiro at mail.inet.com.br (Carlos Alberto Reis Ribeiro)
Date: Sun, 01 Apr 2001 00:01:43 -0300
Subject: PythonLabs Team Moves to ActiveState
Message-ID: <5.0.2.1.0.20010401000139.0222ab10@mail.inet.com.br>

Guido,

This is the best news that I could ever get from you. Now that we have 
peace with the Perl camp, why not rename the language? I propose the name 
Pyrl, and the motto: TINWTMIWNMHHYT, for: "There is no way to make it work 
no matter how hard you try".

As for the numeric PEP, I have a sugestion of my own: implement a 
IRRATIONAL number type. With irrationals, we will be able to represent pi 
with more precision than any other language on earth! We will be also 
giving a great contribution to all the students of the world, who would not 
need to worry *anymore* about such numbers as sqrt(2).

Best regards, and keep with the good work.

At 19:54 31/03/01 -0500, you wrote:
>PythonLabs Team Moves to ActiveState
>
>To all Python users and developers:
>
>    Less than half a year ago, I moved with my team to Digital Crations, 
> in the
>    hope of finding a new permanent home for Python development. At
>    Digital Creations we've done several good things for Python, such as 
> forming
>    the Python Software Foundation and the management of the Python 2.1 
> release.
>
>    Unfortunately, Digital Creations's original plans for PythonLabs didn't
>    work out as hoped, and we weren't able to reach mutual agreement on
>    workable alternative plans - despite trying for months.
>
>    I am proud to have found a new home for my entire team: starting
>    today, Tim Peters, Barry Warsaw, Jeremy Hylton, Fred Drake and myself
>    are working for ActiveState. We will be spending part of our
>    time on core Python development (including Jython and Mailman) and
>    part of our time on Python infrastructure improvements that also
>    benefit ActiveState's efforts such as Komodo and Python.NET.
>
>    Python will remain Open Source; ActiveState has no desire to
>    monetize or brand the Python language or specific Python
>    distributions. All future work we do on Python as ActiveState
>    employees will be owned by the PSF.
>
>    We're excited to be working for ActiveState: they are one of the companies
>    most committed to Python, and they have great products! Plus, we know
>    they have deep financial backing. We trust that ActiveState will
>    provide a stable home for Python for many years.
>
>    ActiveState has also offered to take over hosting of the
>    python.org and starship sites. On behalf of the Python community,
>    we're grateful for this support of the two prime community sites for
>    Python, and we expect to be implementing the transitions shortly.
>
>    These are exciting times for the PythonLabs team - and also for Python
>    and its community. Mainstream successes for Python are showing up
>    everywhere, and we're proud to be a part of such a smart and friendly
>    community. A great year lies ahead!
>
>    Note, however, that moving to ActiveState will have some minor effects
>    on the future of Python development. ActiveState is concerned about
>    the seeming rift between the Perl and Python communities and wants
>    to do its best to help the Open Source scripting community work together.
>    In order to make it easier for Perl programmers to transition to Python,
>    we will add optional curly braces for Python. While we are fully aware
>    of the effect this will have on Python usability, we have finally decided
>    to bite the bullet and accept that until this happens, Python will never
>    be popular.
>
>    Barry has already approved PEP 357, Adding Block Delimiters to Python.
>    Since we are heavily commited to backwards compatability, a transition
>    period of three months has been mandated. For those who want the changes
>    to start sooner, we added "from __future__ import block_delimiters" in
>    Python 2.1, which will be released about two weeks from now.
>
>    Larry Wall and I agreed that now it makes sense for Perl to have
>    the Perl compiler enforce correct indentation in Perl sources. This
>    will probably be implemented in Perl 5.7, and will certainly be in Perl 6.
>    Also, after some talks with Larry, I've decided to adopt the Perl
>    numerical model for Python. The Perl model is simple, predictable and
>    easy to explain. This is a small enough change that it does not need
>    a PEP, and Moshe will implement it before 2.1rc1 is out.
>
>    Now that the catalog-sig finally has a fully working product, ActiveState
>    decided they will base a CPAN competitor on it. This server will be able
>    to deal with both Perl and Python modules. While the client will remain
>    completely Open Source, it was decided that the server will remain
>    ActiveState proprietary. While it is true that the original server was
>    written by the Python community, ActiveState has put a lot of effort
>    into making it general enough, and we need to justify the costs somehow.
>
>    --Guido van Rossum (home page: http://www.python.org/~guido/)
>
>
>--
>http://mail.python.org/mailman/listinfo/python-list





From johann at physics.berkeley.edu  Mon Mar 26 14:56:11 2001
From: johann at physics.berkeley.edu (Johann Hibschman)
Date: 26 Mar 2001 11:56:11 -0800
Subject: python, OS X, and fcntl
Message-ID: <mthf0g719g.fsf@astron.berkeley.edu>

Hi folks,

I'm having trouble with the fcntl module under Mac OS X.  It
sounds like this is an old problem that has resurfaced, since I can
only find references to it in 1999 archives about OS X Server.
Is there a known workaround?

I get, while running test_fcntl:

Status from fnctl with O_NONBLOCK:  0
struct.pack:  '\x00\x03\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00'
Traceback (most recent call last):
  File "test_fcntl.py", line 31, in ?
    rv = fcntl.fcntl(f.fileno(), FCNTL.F_SETLKW, lockdata)

Any hints?

-- 
Johann Hibschman                           johann at physics.berkeley.edu


From jhg at galdon.com  Wed Mar 21 06:36:57 2001
From: jhg at galdon.com (Juan Huertas)
Date: Wed, 21 Mar 2001 12:36:57 +0100
Subject: Pick Systems D3 Database
References: <rG%t6.555$45.2416@newsfeeds.bigpond.com>
Message-ID: <99a3ce$hma$1@talia.mad.ttd.net>

You can try to obtain information on www.picksys.com, (Raining   Data). I
use D3 very years, it's good. I think the best is to use ODBC, but D3 has te
possibility to import/export data to ASCII format using the copy instruction
to logical devices.

Into Aix, the normal solution is to use Telnet to work with the Server D3
(D3Vme).

Un saludo.

Juan Huertas.


Ben <bencr at bigpond.com> escribi? en el mensaje de noticias
rG%t6.555$45.2416 at newsfeeds.bigpond.com...
> Hi,
>
> I have not been able to find any direct information about this so I
thought
> I would ask here in the slim chance someone may have had to do something
> similar ... what would be the best way to push/pull information from a D3
> (PickSystems) database running on AIX using Python? Is ODBC my only option
> or are there others? If you have done something similar how succesful was
> your solution?
>
> Thanks in advance
>
>




From carribeiro at yahoo.com  Mon Mar 19 20:49:22 2001
From: carribeiro at yahoo.com (Carlos Ribeiro)
Date: Mon, 19 Mar 2001 22:49:22 -0300
Subject: Functionalism, aesthetics Was:(RE: I come to praise .join,
  not 
Message-ID: <5.0.2.1.0.20010319224910.021134d0@mail.inet.com.br>

I'm lost for words... This summarizes a lot of concerns that sometimes I 
find so hard to explain. Anyone who has actually *read* Christopher 
Alexander's books on design should know that the best solutions are the 
ones that take into account the inner sense of satisfaction of their end 
users. As Alexander puts it itself, it is a rewarding experience born out 
of the "completeness" of the solution. The best solutions also share 
another common trait: they look *and feel* good to the untrained eye. 
Beware of works that can only be appreciated by experts - they are way too 
trained to see the obvious.
Mike C. Fletcher wrote:
>         The reason I felt it necessary to jump in here again is the implicit
>creation of a "inner circle" of the "initiated" who are considered "worthy"
>to contribute to a discussion about the evolution of the language.  The
>users of a system, whether it is a building, a language, a piece of
>software, or a chair, need to be heard.  A user complaining that something
>is "ugly", "bad" or "wrong" needs the nature of their discomfort
>ascertained, and that knowledge integrated into further design decisions.
>         Designers who ignore their users, assuming that their own 
> understanding of
>the system is correct, and that anyone who is unable to speaking the dialect
>of designers cannot possibly contribute to a design, will create
>monstrosities which are next to impossible to inhabit.  Functionalism only
>works when the understanding of "function" is broadened to include the
>entire process of inhabitation, not just the parti-sketch with which the
>architect is fascinated.  The modernist movement proved this time and again
>with awe-inspiring buildings which failed to be good places to live.  The
>postmodernists continue to prove it with each building they inflict upon the
>world.
>Your friendly neighbourhood designer signing off...
>Mike

To summarize it: a  *technically* optimal solution for a problem is not 
always the *best* solution. If it does not look intuitively good for most 
users, its probably a bad solution.

Its unfortunate that in many cases the non-expert user keeps quiet, 
assuming that whatever opinion the "master" haves, its going to be better 
than theirs - after all they are the so-called experts.

Carlos Ribeiro



_________________________________________________________
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com




From senum at singnet.com.sg  Sun Mar 11 22:08:07 2001
From: senum at singnet.com.sg (dr0)
Date: Mon, 12 Mar 2001 11:08:07 +0800
Subject: looking for old python books
Message-ID: <eafoat4r8rs21v1h50239iqf9sktaetnjv@4ax.com>

Hi all,
 I was wondering if any of you have any old books on python you can
sell to me?
I'm a student and the prices for online stores's a bit high.
Email me if you can help.

Any help is deeply appreciated,
thanks.
Kenny


From thomas at cbs.dtu.dk  Thu Mar  1 11:44:19 2001
From: thomas at cbs.dtu.dk (Thomas Sicheritz-Ponten)
Date: 01 Mar 2001 17:44:19 +0100
Subject: Hypergeometric distribution ?
Message-ID: <y9vzof58mn0.fsf@genome.cbs.dtu.dk>

Does anybody have a python function/class for calculating the
hypergeometric distribution ?

thx
-thomas
-- 
Sicheritz-Ponten Thomas, Ph.D  CBS, Department of Biotechnology
thomas at biopython.org           The Technical University of Denmark
CBS:  +45 45 252489            Building 208, DK-2800 Lyngby
Fax   +45 45 931585            http://www.cbs.dtu.dk/thomas

	De Chelonian Mobile ... The Turtle Moves ...



From jkraska1 at san.rr.com  Mon Mar 26 00:59:48 2001
From: jkraska1 at san.rr.com (Courageous)
Date: Mon, 26 Mar 2001 05:59:48 GMT
Subject: So what's wrong with __future__?  (Was Re: Why "from __future__" stinks ...)
References: <slrn9bdka2.3j6.hzhu@rocket.knowledgetrack.com> <LNBBLJKPBEHFEDALKOLCAEDKJIAA.tim.one@home.com> <mailman.985568441.15878.python-list@python.org>
Message-ID: <vvmtbto161c3n1m3n7ub4vii2nta490alb@4ax.com>

>I'm glad that it appears that even the Python gods are
>implementing this hideous travesty under protest. It is truly
>the ugliest hack ever--far worse than lambda, which to some
>extent could have been expected.

While I may be guilty of not paying attention, I just don't understand
why it's needed at all? Those who don't want to play with new features
should keep to an older Python revision. This is the standard way of
doing things in the language business. This import from __future__
thing seems not on like a glorious hack, but terribly unnecessary.
What am I missing?

C//



From magnus.heino at rivermen.se  Thu Mar  8 09:20:04 2001
From: magnus.heino at rivermen.se (Magnus Heino)
Date: Thu, 8 Mar 2001 15:20:04 +0100
Subject: m2crypto
Message-ID: <9880lf$rgu$1@Merlin.et.bocholt.fh-gelsenkirchen.de>

Has anyone successfully compiled M2Crypto 0.06 snapshot #2 on a linux box?

This version seems to be depending on OpenSSL 0.9.6, but if I use that 
version I get this errormessage after installation:

>>> import M2Crypto
Traceback (innermost last):
  File "<stdin>", line 1, in ?
  File "M2Crypto/__init__.py", line 7, in ?
    import _m2crypto
  File "M2Crypto/_m2crypto.py", line 2, in ?
    import _m2cryptoc
ImportError: M2Crypto/_m2cryptoc.so: undefined symbol: sk_new_null

I have tried different versions of SWIG, but that doesn't seem to be the 
problem.

Any clues?

--

/ Magnus


From chris.gonnerman at usa.net  Sat Mar 31 01:50:15 2001
From: chris.gonnerman at usa.net (Chris Gonnerman)
Date: Sat, 31 Mar 2001 00:50:15 -0600
Subject: newbie variables question
References: <Pine.LNX.4.10.10103311016540.31580-100000@localhost.localdomain>
Message-ID: <00af01c0b9ae$e20377a0$a100000a@local>

----- Original Message -----
From: "Michael Hall" <py.list at mulga.com.au>
Subject: newbie variables question>

> I've been trying to do this:
>
> SYSPRODIR = /home/admin/systemprofile
> os.system('mv /tmp/install.log SYSPRODIR/install.log')
>
> No matter how I play around with single and double quotes, this won't
> work.

You're thinking /bin/sh.  Quoting has little to do with it.

> But this does work:
>
> os.system('mv /tmp/install.log %s/install.log' %s SYSPRODIR)

I hope it actually says:

    os.system('mv /tmp/install.log %s/install.log' % SYSPRODIR)

This is one correct way to do what you want.

> Is this the normal/best way to do this in Python ... seems a little
> convoluted me, like there must be a simpler way. Bash for example would be
> happy with something like:
>
> mv /tmp/install.log $SYSPRODIR/install.log
>
> while PHP is happy with things like:
>
> include ("$SYSPRODIR/install.log");

The other way in Python is this:

    os.system('mv /tmp/install.log ' + SYSPRODIR + '/install.log')

which works fine so long as SYSPRODIR is a string.

Substitution into a string is almost always done with the % operator, as
in

    "format string" % value

or for multiple values:

    "format string" % (v1, v2, v3)

where the second operand is obviously a tuple.

> Thanks for any advice. I haven't found any clear info about this in
> Learning Python yet.
>
> Mick

Happy to help.





From thomas at xs4all.net  Mon Mar 12 21:51:10 2001
From: thomas at xs4all.net (Thomas Wouters)
Date: Tue, 13 Mar 2001 03:51:10 +0100
Subject: ANNOUNCE Mailman 2.0.3
In-Reply-To: <CCaB3KAyQWr6Ewo5@jessikat.fsnet.co.uk>; from robin@jessikat.fsnet.co.uk on Tue, Mar 13, 2001 at 12:05:06AM +0000
References: <mailman.984437116.20063.python-list@python.org> <CCaB3KAyQWr6Ewo5@jessikat.fsnet.co.uk>
Message-ID: <20010313035110.Z404@xs4all.nl>

On Tue, Mar 13, 2001 at 12:05:06AM +0000, Robin Becker wrote:
> In article <mailman.984437116.20063.python-list at python.org>, Barry A.
> Warsaw <barry at digicool.com> writes

> >I've just announced the 2.0.3 release of Mailman, the GNU Mailing List
> >Manager.  Mailman is released under the GNU General Public License
> >(GPL).
> >
> >I highly recommend that all sites running earlier versions of Mailman
> >upgrade to 2.0.3.  This release includes the important privacy patch
> >from Mailman 2.0.2.  Sadly, Mailman 2.0.2 contained a small but nasty
> >typo requiring this new version.  Mailman 2.0.3 is being released as
> >both a gzip'd source tarball and as a patchfile.

> thanks, I've just installed 2.0.2 and set up my first list. It's not a
> problem for me now, but is it easy to install over an existing
> installation?

Dead easy. You can even hand-patch your 2.0.2 installation if you wish to do
so. Still, patching your source tree and then running 'make install' does
all that is required, and even if something other than a few modules were
changed (like the format of the config.db), 'make install' would handle it.

Just patch, 'make', 'make install' and all should be fine. Making a backup
first is always a good idea, of course, but shouldn't be necessary <cross
fingers> :)

Crossing-fingers-is-what-caused-the-typo-in-the-first-place-ly y'rs,
-- 
Thomas Wouters <thomas at xs4all.net>

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



From f8dy at yahoo.com  Fri Mar  2 13:35:09 2001
From: f8dy at yahoo.com (Mark Pilgrim)
Date: Fri, 2 Mar 2001 13:35:09 -0500
Subject: gah! I hate the new string syntax
References: <mailman.983552044.31198.python-list@python.org>
Message-ID: <97op1m$plqrt$1@ID-77331.news.dfncis.de>

"Sean 'Shaleh' Perry" <shaleh at valinux.com> wrote in message
news:mailman.983552044.31198.python-list at python.org...
> > 3. I *like* string methods, even .join()
>
> ";" . join() # in perl, this means concat ";" with the result of join()
>
> Yes, I know this is an issue you learn to deal with.  However, the nice,
pretty
> string module at least makes python look like python.

I can't help you deal with the Perl issues of your childhood, but I can
recommend a good therapist. ;)

> Mark, I like a lot of what I read in your book.  The apihelper is a nifty
idea
> (if only it worked on the re module )-:).  However, the coding style and

Heh heh, it does work on the re module, it's just not very helpful because
the module doesn't include docstrings for the top-level functions.  The
authors are actually aware of this; see sre.py, line 41:
# FIXME: add docstrings

> decisions seem counter to the python way.  But maybe that is just me and
my
> hyper sensitivity to hard to read code after way too many lines of perl.

I am always open to suggestions.  I have already made some code rewrites
after discussions with readers, including several clarifications in chapter
1 and a substantial rewrite in chapter 2 (using getattr instead of eval).

A better place to discuss "Dive Into Python"-specific issues is Andamooka,
which is hosting a mirror of the book with an integrated discussion board on
each page.  Unfortunately, the version of the book they have is a month out
of date (and doesn't include the getattr rewrite), but it's better than
cluttering up the newsgroup.
  http://www.andamooka.org/index.pl?section=divein

-M
You're smart; why haven't you learned Python yet?
http://diveintopython.org/





From nkauer at users.sourceforge.net  Mon Mar 12 19:09:10 2001
From: nkauer at users.sourceforge.net (Nikolas Kauer)
Date: Mon, 12 Mar 2001 18:09:10 -0600 (CST)
Subject: syntax exploration tool
In-Reply-To: <15021.22659.616556.298360@anthem.wooz.org>
Message-ID: <Pine.LNX.4.10.10103121801530.7351-100000@falcon.physics.wisc.edu>

I'd volunteer to put in time and help create such a tool.  If someone 
sufficiently knowledgeable decides to go ahead with such a project 
please let me know.

---
Nikolas Kauer <nkauer at users.sourceforge.net>

> Second, if type annotations are to be seriously considered for
> inclusion in Python, I think we as a community need considerable
> experience with a working implementation.  Yes, we need PEPs and specs
> and such, but we need something real and complete that we can play
> with, /without/ having to commit to its acceptance in mainstream
> Python.  Therefore, I think it'll be very important for type
> annotation proponents to figure out a way to allow people to see and
> play with an implementation in an experimental way.




From duncan at rcp.co.uk  Mon Mar 26 09:44:38 2001
From: duncan at rcp.co.uk (Duncan Booth)
Date: Mon, 26 Mar 2001 14:44:38 +0000 (UTC)
Subject: Lists and Files
References: <99njdo$oa3$1@news.lth.se>
Message-ID: <Xns90709FB7E5A4duncanrcpcouk@127.0.0.1>

"David Andreas Alderud" <aaldv97 at student.removethispart.vxu.se> wrote in
<99njdo$oa3$1 at news.lth.se>: 

>I'm new to Python, started yesterday, is there a good way to handle
>lists and files?
>

See the documentation for the pickle and cPickle modules:
>>> stooges
['val1', 'val2', 1, [['val2_1', 'val2_2', 0], ['val2_1', 'val2_2', 0]]]
>>> fds = open('stooges', 'w')
>>> import cPickle
>>> p = cPickle.Pickler(fds)
>>> p.dump(stooges)
<Pickler object at 00827C20>
>>> fds.close()

>>> fds = open('stooges', 'r')
>>> u = cPickle.Unpickler(fds)
>>> new_stooges = u.load()
>>> print new_stooges
['val1', 'val2', 1, [['val2_1', 'val2_2', 0], ['val2_1', 'val2_2', 0]]]
>>>

Most common object types are picklable.


From loewis at informatik.hu-berlin.de  Thu Mar 22 04:15:52 2001
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 22 Mar 2001 10:15:52 +0100
Subject: New PEP: The directive statement
References: <mailman.985085006.24236.python-list@python.org> <slrn9bi7q3.2sa.hzhu@rocket.knowledgetrack.com>
Message-ID: <j4snk6yxiv.fsf@informatik.hu-berlin.de>

hzhu at users.sourceforge.net (Huaiyu Zhu) writes:

> So with this directive PEP, the parallel would be
> 
> directive transitional nested_scope
> directive experimentatal stackless
> directive optional static_type
> directive mandatory string_methods
> 
> Is this correspondence correct?  (You may not agree on the need for other
> directives, but I want to know if the keyword translations are correct.)

That is a possible application. The syntax allows other choices as well,
e.g.

  directive static_type
  directive stackless 0

> 1. It introduces a new syntax while an existing syntax could be reused
>    without much cost.

You mean, import can be reused without much cost? The cost of reusing
import is quite high; it causes irritation among users.

> 2. There is no place for version and documentation info.

That is true.

> So far as I can see, the impetus for directive keyword is that import is a
> run time action, while the __future__ is a compile time declaration.  But is
> it so far a stretch?  All Python __magic__ words go beyond normal meaning.
> Defining __setattr__ will affect more than this method alone.  Having
> __init__.py will cause other modules to be imported.  And so on.

That is what I dislike in the __ convention. Syntactically, they are
identifiers. Semantically, they are reserved words.

> If we look more broadly, in *nix, devices, processes, pipe and others are
> also represented as files.  Would it better to assign special syntax to all
> of them?

Or, taking the analogy further: sockets should be represented as files
as well, to open a socket, you should really open some file, not
invoke a special function. Yet, history shows that the version with
the special API was the successful one, not the version with the files.

Generalization is a dangerous thing: you can too easily cross the
border to over-generalization.

> IMHO, reusing existing syntax is closer to Python than Perl, which
> introduces new syntax for many things that are only slightly different.

Are you saying that activating nested scopes is only slightly
different from importing modules?

Regards,
Martin



From petr at ai.mit.edu  Fri Mar 23 23:48:37 2001
From: petr at ai.mit.edu (Petr Swedock)
Date: Fri, 23 Mar 2001 23:48:37 -0500
Subject: python, perl, lisp, tcl / got a decision to make, can you give
 me a view point please ? and a joke for python users :)
In-Reply-To: <slrn9bo25q.omf.tim@vegeta.ath.cx>
References: <3aba3076.14395287@news.vif.com> <slrn9bo25q.omf.tim@vegeta.ath.cx>
Message-ID: <Pine.GSO.4.21.0103232340400.29232-100000@life>


On Sat, 24 Mar 2001, Tim Hammerquist wrote:

> Yvon Boulianne <mysticsguy at yahoo.ca> wrote:
> > i know a bit of perl and tcl, (i find
> > perl too much cryptics, difficult to understand other people scripts)
> 
> Damn!  Not even a mystic can make sense of Perl.  =)

Normally, a mystic wouldn't have to, they'd just divine intent. This too
is impossible to do with Perl =-)

And while I don't have a 'joke for python users', I do often tell others
that if Guido van Rossum were a true Monty Python fan he'd have named 
the language "Raymond Luxury-Yacht" AND require that people pronounce it
"Throat Warbler Mangrove."  =-) Does that count?

Peace,

Petr

Systems, Networks and Gadgets, done with Artful Intelligence
                  -<>-<_>-<__>-<_>-<>-
Policy: ASCII/text attchmnts alway read. PDF maybe read. Others, by 
neccesity, may be ignored. Don't take it personally, it's a time issue.





From topmind at technologist.com  Thu Mar  1 22:11:54 2001
From: topmind at technologist.com (Topmind)
Date: Fri, 02 Mar 2001 03:11:54 GMT
Subject: Stream interfaces
References: <3A9E54F9.7AF90BCA@ix.netcom.com> <memo.20010301200321.16971A@brangdon.madasafish.com>
Message-ID: <MPG.1508b1b4438159e99896cc@news.earthlink.net>

> tgagne at ix.netcom.com (Thomas Gagne) wrote (abridged):
> > Your correct.  People are confusing inheritence and usefulness.  I 
> > don't have enough spare cycles to worry about a class'
> > inheritence unless I've discovered missing functionality.  Since
> > I haven't, I haven't.
> 
> Well, I have. In the streams hierarchy, too. Albeit not that vendor and 
> in a beta product so I probably shouldn't say too much about it, but the 
> hierarchy was very similar. The problem was analogous to methods added to 
> WriteStream not magically appearing in ExternalWriteStream.
> 
> The vendor's fix was to cut&paste methods from one class to the other. 
> But this is just sticking plaster. The class hierarchy is deeply wrong, 
> violates "once and only once" all over the place and needs refactoring. 
> It's not about missing functionality so much as how we can add new kinds 
> of stream without more error-prone cut&paste.
> 
> Whether this can be fixed without breaking existing clients, or fixed at 
> all without multiple inheritance, I don't know because I've not tried it. 
> I hope it can be and the problem is not with Smalltalk generally but just 
> this particular bit of library design. But for whatever reason, this bit 
> of library design is not ideal.
> 


Where did the deep hierarchy kind of library design trend start? 
Can anybody pinpoint an event or "leader" who appears to 
have started that trend/fad?

And, why did the other vendors follow?

"Technical sociology" would make a fascnating topic or
study area IMO.

>   Dave Harris, Nottingham, UK 

-tmind-



From emile at fenx.com  Wed Mar 21 17:21:00 2001
From: emile at fenx.com (Emile van Sebille)
Date: Wed, 21 Mar 2001 14:21:00 -0800
Subject: I'm sure there's a good reason!
References: <memo.20010321220225.57471D@digitig.compulink.co.uk>
Message-ID: <99b9m8$gq44$1@ID-11957.news.dfncis.de>

You aren't allowed to end a raw string with an odd number of \'s.

Emile van Sebille
emile at fenx.com

"Tim Rowe" <digitig at cix.co.uk> wrote in message
news:memo.20010321220225.57471D at digitig.compulink.co.uk...
> Can any of the assembled experts explain to me why:
> print r"\hello"
> is fine, but
> print r"hello\"
> isn't? In other words, why does the backslash escape quotes in a raw
> string? At one level I'm going to guess (without reading the spec) that
> it's because "that's what the spec says", but why? I thought the idea of
> raw strings was that they /didn't/ handle escapes!
>
> TIA,
>
> Tim
>




From rav at smo3k.z.pl  Tue Mar  6 06:39:11 2001
From: rav at smo3k.z.pl (Rafal Smotrzyk)
Date: Tue, 6 Mar 2001 12:39:11 +0100
Subject: python cgi syntax highlighter
References: <Pine.BSF.4.32.0103032234260.88338-100000@localhost>
Message-ID: <982i66$rrh$1@news.tpi.pl>

# Rafal Smotrzyk
# 24/11/2000
# mailto:rsmotrzyk at mikroplan.com.pl

import tokenize
import token
import keyword
import string
import sys
import os

d_builtins = {
   '__debug__' : 1,
   '__doc__' : 1,
   '__import__' : 1,
   '__name__' : 1,
   'abs' : 1,
   'apply' : 1,
   'buffer' : 1,
   'callable' : 1,
   'chr' : 1,
   'cmp' : 1,
   'coerce' : 1,
   'compile' : 1,
   'complex' : 1,
   'delattr' : 1,
   'dir' : 1,
   'divmod' : 1,
   'eval' : 1,
   'execfile' : 1,
   'exit' : 1,
   'filter' : 1,
   'float' : 1,
   'getattr' : 1,
   'globals' : 1,
   'hasattr' : 1,
   'hash' : 1,
   'hex' : 1,
   'id' : 1,
   'input' : 1,
   'int' : 1,
   'intern' : 1,
   'isinstance' : 1,
   'issubclass' : 1,
   'len' : 1,
   'list' : 1,
   'locals' : 1,
   'long' : 1,
   'map' : 1,
   'max' : 1,
   'min' : 1,
   'oct' : 1,
   'open' : 1,
   'ord' : 1,
   'pow' : 1,
   'quit' : 1,
   'range' : 1,
   'raw_input' : 1,
   'reduce' : 1,
   'reload' : 1,
   'repr' : 1,
   'round' : 1,
   'setattr' : 1,
   'slice' : 1,
   'str' : 1,
   'tuple' : 1,
   'type' : 1,
   'unload' : 1,
   'vars' : 1,
   'xrange' : 1,
}

class PrettyPyPrinter:
   def __init__(self,aspage=1):
      self.AsPage=aspage
   def
TokenEater(self,atokentype,atokenstring,abeginrowcolumntuple,aendrowcolumntu
ple,aline):
      brow,bcol=abeginrowcolumntuple
      erow,ecol=aendrowcolumntuple
      self.token_list.append([brow,bcol,erow,ecol,atokentype,atokenstring])
   def Process(self,fnamein,fnameout):
      fin=open(fnamein,'r')
      try:
         self.Tokenize(fin)
      finally:
         fin.close()
      fout=open(fnameout,'w')
      try:
         self.Generate(fout)
      finally:
         fout.close()
   def InsertTag(self,arow,acol,s):

self.lines[arow-1]=self.lines[arow-1][:acol]+s+self.lines[arow-1][acol:]
   def Tokenize(self,fin):
      self.token_list=[]
      tokenize.tokenize(fin.readline,self.TokenEater)
      fin.seek(0)
      self.lines=fin.readlines()
      self.token_list.sort()
      self.token_list.reverse()
   def Generate(self,fout):
      for i in range(len(self.token_list)):
         brow,bcol,erow,ecol,atokentype,atokenstring=self.token_list[i]
         processed=' '
         if atokentype==token.OP:
            self.InsertTag(erow,ecol,'</font>')
            self.InsertTag(brow,bcol,'<font color="PURPLE">')
            processed='*'
         elif atokentype==token.NAME:
            if keyword.iskeyword(atokenstring):
               self.InsertTag(erow,ecol,'</b></font>')
               self.InsertTag(brow,bcol,'<font color="NAVY"><b>')
            elif d_builtins.has_key(atokenstring) and
self.token_list[i+1][5]!='.':
               self.InsertTag(erow,ecol,'</font>')
               self.InsertTag(brow,bcol,'<font color="NAVY">')
            elif i>0 and self.token_list[i+1][5] in ['def','class']:
               self.InsertTag(erow,ecol,'</B></font>')
               self.InsertTag(brow,bcol,'<font color="BLUE"><B>')
            elif i>0 and self.token_list[i-1][5]=='(':
               self.InsertTag(erow,ecol,'</font>')
               self.InsertTag(brow,bcol,'<font color="PURPLE">')
            elif atokenstring=='None':
               self.InsertTag(erow,ecol,'</font>')
               self.InsertTag(brow,bcol,'<font color="RED">')
            else:
               self.InsertTag(erow,ecol,'</font>')
               self.InsertTag(brow,bcol,'<font color="BLACK">')
            processed='*'
         elif atokentype==token.NUMBER:
            self.InsertTag(erow,ecol,'</font>')
            self.InsertTag(brow,bcol,'<font color="RED">')
            processed='*'
         elif atokentype==token.STRING:
            self.InsertTag(erow,ecol,'</font>')
            self.InsertTag(brow,bcol,'<font color="MAGENTA">')
            processed='*'
         elif atokentype==token.N_TOKENS:
            self.InsertTag(erow,ecol,'</I></font>')
            self.InsertTag(brow,bcol,'<font color="NAVY"
style="background=YELLOW;"><I>')
            processed='*'
         elif atokentype in
[token.NEWLINE,token.INDENT,token.DEDENT,token.ENDMARKER,40]: #40==NL
            processed='*'
         if processed==' ':
            print '%s %5d%5d%5d%5d | %d%20s
>%s<'%(processed,brow,bcol,erow,ecol,atokentype,token.tok_name[atokentype],r
epr(atokenstring))
         i=i+1
      if self.AsPage:
         fout.write('<html><body>\n')
      fout.write('<pre>')
      for aline in self.lines:
         s=string.replace(aline,'\n','<br>\n')
         s=string.replace(aline,chr(255),'&nbsp;')
         fout.write(s)
      fout.write('</pre>')
      if self.AsPage:
         fout.write('\n</body></html>')

if __name__=='__main__':
   if len(sys.argv)==3:
      fname1=sys.argv[1]
      fname2=sys.argv[2]
   elif len(sys.argv)==2:
      fname1=sys.argv[1]
      fname2=os.path.splitext(fname1)[0]+'.html'
   else:
      print 'py2html input.py [output.html]'
      sys.exit()
   aprinter=PrettyPyPrinter()
   aprinter.Process(fname1,fname2)





From not.this at seebelow.org  Fri Mar 30 22:44:29 2001
From: not.this at seebelow.org (Grant Griffin)
Date: Fri, 30 Mar 2001 21:44:29 -0600
Subject: Know of Substantial Apps Written in Python?
References: <3ac4e4a8_1@news4.newsfeeds.com> <KE6x6.415$SR6.235983@typhoon1.ba-dsg.net> <dj7x6.9561$SB2.850896@ruti.visi.com>
Message-ID: <3AC5529D.8D7BCC63@seebelow.org>

Grant Edwards wrote:
> 
> In article <KE6x6.415$SR6.235983 at typhoon1.ba-dsg.net>, David Allen wrote:
> 
> >What's the definition of "substantial"?
> >
> >There's Grail, a web browser
> >http://grail.sourceforge.net
> 
> Anybody get Grail to work lately?  Last time I tried it
> wouldn't run.
> 
> >There's ZOPE, an internet publishing thingy
> >(thingy is a technical term  :)
> 
> Sketch is one of my favorites:
> 
>   http://sketch.sourceforge.net/

Hi Grant,

Hey, I looked at that, and it looks cool.  But I was somewhat puzzled by
the following statement:

   Sketch is an interactive vector drawing program for Linux
   and other UNIX compatible systems. <snip>
   
   A somewhat unique(for a drawing program) feature of Sketch is
   that it is implemented almost completely in a very high-level,
   interpreted language, Python.

If it's implemented almost completely in Python, why does it only run on
"UNIX-compatible" systems?  (BTW, does it delete hyphens? <wink>)  Is
there something about its "almost" part that can't be done in Windows.

free-software-isn't-just-for-unix-ly y'rs,

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation            http://www.iowegian.com


From db3l at fitlinxx.com  Fri Mar 30 13:17:33 2001
From: db3l at fitlinxx.com (David Bolen)
Date: 30 Mar 2001 13:17:33 -0500
Subject: mxODBC chokes on dates! Any Help?
References: <MPG.152e4c579fa0baef9896a5@news.pacbell.net>
Message-ID: <ulmpnnmte.fsf@ctwd0143.fitlinxx.com>

Owen Ap Owen <no at thank.you> writes:

> I'm using mxODBC from Python 2.0 under Win32. I have a functional ODBC 
> connection but when I try to select data that includes date fields mxODBC 
> chokes.

I ran into the same problem with some date fields in an MS SQL Server
database, and Marc-Andre was able to revert back to an earlier way of
handling date fields (he preferred a different C data type if defined
in 2.0.0 than in earlier versions) to resolve it.  I was traveling and
was a little tardy getting back to verify that the fix worked, but I
expect it will show up in the next release of mxODBC.

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


From djc at object-craft.com.au  Thu Mar  1 20:00:13 2001
From: djc at object-craft.com.au (Dave Cole)
Date: 02 Mar 2001 12:00:13 +1100
Subject: Help with finding how much memory a variable is using
References: <JAwl6.17037$5n4.338201@news6-win.server.ntlworld.com> <mailman.982950492.15473.python-list@python.org> <v3pm6.27666$5n4.596175@news6-win.server.ntlworld.com> <97dhh2$p9ul0$1@ID-11957.news.dfncis.de> <m33dd0juwy.fsf@vole.object-craft.com.au> <yw3n6.1025$sT1.130520@news2-win.server.ntlworld.com>
Message-ID: <m33dcxq92a.fsf@vole.object-craft.com.au>

>>>>> "Rich" == Rich Somerfield <rich_somerfield at tertio.com> writes:

Rich> That code would be gratefully received.
Rich> rich_somerfield at nO-sPaM.tertio.com

Oops - I somehow missed your response.

I wrote the code for someone else quite a while ago to demonstrate a
point.  They wanted to be able to load massive electronic schematic
files and make them available to Python.  The problem was similar to
yours, due the amount of the data they were loading, Python was just
too slow and too resource hungry.  They had a yacc grammar which they
were using and wanted to make the parse tree available to their Python
code.  I did not have any schematic files I could load, so I picked
the largest text file I could find on my computer; the Debian Packages
file.

Since I was only trying to demonstrate a technique, I took a few
shortcuts.  I did not bother trying to write a yacc grammar, I just
wrote a simple parser in C.  Instead of building my own collections, I
used those which come in GLIB.

There is too much code to post here, so I have placed it on our web
server.

        http://www.object-craft.com.au/projects/lazy/

There is a benchmark program which compares the speed of the raw
parser and the parser accessed through a Python module.  On my
machine, the Packages file is 121,111 lines and 4,471,834 bytes.  The
benchmark program prints the following:

Raw parser loads in 0.70 seconds
aalib1 1.2-18
aalib1 1.2
libc6 2.2.1
libgpmg1 1.14-16
slang1 1.3.0-0
xlibs 4.0.1-11
Wrapped parser loaded and used in 0.75 seconds

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
#!/usr/bin/env python

import os
import time

file = '/var/state/apt/lists/ftp.us.debian.org_debian_dists_unstable_main_binary-i386_Packages'

# Prime the buffer cache
os.system('./pkg_test %s' % file)

t0 = time.time()
os.system('./pkg_test %s' % file)
t1 = time.time()
print 'Raw parser loads in %.2f seconds' % (t1 - t0)

import dpkg
main = dpkg.load_file(file)
deps = main.packages['aalib-bin'].Depends
for dep in deps:
    print dep.Package, dep.Version
t2 = time.time()
print 'Wrapped parser loaded and used in %.2f seconds' % (t2 - t1)
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

The files are:

Makefile.pre.in -- used to build module
Setup.in        -- used to build module
benchmark.py    -- the above benchmark program
dpkg.c          -- the Python wrapper of the parser
pkg.c           -- simple (and incomplete) Packages file parser
pkg.h           -- parser interface
pkg_test.c      -- drives the parser from C

The key to the technique is that Python proxies for the parse tree
nodes are created on demand.  The Python attributes of each parse tree
node are created on demand.

dpkg.c is the interesting file as it demonstrates the technique.

The dpkg.c file should have probably been two files as it contains two
things; but what the hell, I wrote this for someone to demonstrate a
point.  It contains the building blocks for constructing the lazy
evaluation proxy objects, and it also contains the proxies for the
pkg.c parser.

If enough people are interested, I will do some more work on splitting
the code up to make it more general.

Oh, to build and test the code do this:

% make -f Makefile.pre.in boot
% make
% cc -g -Wall -o pkg_test pkg_test.c pkg.c `glib-config --cflags --libs`
% ./benchmark.py 

Traps for young players:

1- The data is read-only.

2- If you drop the reference to the top level parse tree object,
   things will go bad very quickly (main in the benchmark program).

- Dave

-- 
http://www.object-craft.com.au


From digitig at cix.co.uk  Wed Mar 14 19:42:00 2001
From: digitig at cix.co.uk (Tim Rowe)
Date: Thu, 15 Mar 2001 00:42 +0000 (GMT Standard Time)
Subject: DOS python?
References: <20010311104606.07469.00000559@ng-cj1.aol.com>
Message-ID: <memo.20010315004232.1527C@digitig.compulink.co.uk>

In article <20010311104606.07469.00000559 at ng-cj1.aol.com>, 
park997 at aol.comnotospam (Park997) wrote:

> Hi,
> 
> I would appreciate any pointers on obtaining a version of
> python that operates on an HP LX200 (DOS 5, 80186 CPU; 
> essentially 8 bit IBM XT equivalent). I found a python16 that
> runs on Windows, but not on the LX200.

I'd like that too, but I doubt you'd get much of the library in there even 
if you had the interpreter!


From ransen_spam_me_not at nemo.it  Sun Mar  4 11:15:39 2001
From: ransen_spam_me_not at nemo.it (Owen F. Ransen)
Date: Sun, 04 Mar 2001 16:15:39 GMT
Subject: Bug in Py_Finalize() ?
References: <3a9f636d.2908314@news.newsguy.com>
Message-ID: <3aa20ae3.3454319@news.newsguy.com>

On Fri, 02 Mar 2001 09:32:12 GMT, ransen_spam_me_not at nemo.it (Owen F.
Ransen) wrote:

With Python 2.1b1 this now works

    for (UINT i = 0 ; i < 100 ; i++) {
        Py_Initialize () ;
        Py_Finalize () ;
    }

This is all the Python I am using in my VC6
MFC based program in the InitInstance function
and which previously crashed Python.


--
Owen F. Ransen
http://www.ransen.com/
Home of Gliftic & Repligator Image Generators


From aleaxit at yahoo.com  Tue Mar 20 04:36:01 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Tue, 20 Mar 2001 10:36:01 +0100
Subject: Sequence-spreading
References: <3AB7167F.2816AA21@student.liu.se>
Message-ID: <99790u01c73@news2.newsguy.com>

"Rikard Bosnjakovic" <rikbo716 at student.liu.se> wrote in message
news:3AB7167F.2816AA21 at student.liu.se...
> Hi.
>
> I made a small spreading-sequence with the following code:
>
> ------
> arg = "some string"
> pm = 3
> offsets = []
> for i in xrange(pm):
>     everyX = filter(lambda y: not divmod(y, pm)[1], range(len(arg)))
>     added = map(lambda x: x+i, everyX)
>     offsets.append(filter(lambda x: x<len(arg), added))
>
> print offsets
> ------
>
> which gives the correct result:
>
> >>> ## working on region in file /usr/tmp/python-3115DCp...
> [[0, 3, 6, 9], [1, 4, 7, 10], [2, 5, 8]]
>
> "pm" is the number of sequences, and then the sequences are increasing.
> 0 in first, 1 in second, 2 in third, 3 in first and so on, until the
> length of the string 'arg' is reached.  However, the loop-code is pretty
> unreadable and awkward.
>
> Anyone got a better solution for it?

The following is simpler:

offsets = [ [] for i in range(pm) ]
for i in range(len(arg)):
    offsets[i%pm].append(i)


Alex





From mcfletch at home.com  Mon Mar 19 00:22:16 2001
From: mcfletch at home.com (Mike C. Fletcher)
Date: Mon, 19 Mar 2001 00:22:16 -0500
Subject: Dragon Macros (was: RE: Help Rewriting/Refactoring/Rethinking Parsing Algorythm)
Message-ID: <000601c0b034$90c79200$a9807018@cr706570a>

For those who are curious, here's a set of macros that seem to work nicely
for the names:

In VocEdit, define the following, then train them in vocabulary editor (the
items with nothing before the slash are supposed to have nothing there,
these are non-visible elements)...
	\pyname, LowerCase next word, Turn Capitalisation On, Turn off Spacing
between words
	\pycaps, Turn Capitalisation On, Turn off Spacing between words
	\pynorm, Restore Normal Capitalisation, Restore Normal Spacing
	\pynormal, Restore Normal Capitalisation, Restore Normal Spacing
	if __name__ == "__main__":\pymainline, Follow with one newline character

The last entry eliminates the need for a __name__ and __main__ entry, which
frees up the pyname entry for use as a lowerUpperUpper macro (and yes, I did
dictate that using the macro :) ).  pynormal is more natural feeling than
pynorm when you're first getting used to it.

I'm considering adding "add an extra space following this word" for pynorm
(without it you get some less-than-attractive effects, but with it I'd need
another phrase for "stop the caps and spaces but don't space" for when doing
a dotted name.  Oh, pyconstant might be introduced to give an ALL-CAPS + No
Space macro.

Making a script that auto-created all those + the def, equals, triple-quote,
etceteras and asked the user to train each a couple of times would be cool.
Maybe when I'm out of a job I'll look into it.

Anyway, enjoy all,
Mike

-----Original Message-----
From: Mike C. Fletcher [mailto:mcfletch at home.com]
Sent: Sunday, March 18, 2001 11:32 PM
To: 'Boopy Bootles'; python-list at python.org
Subject: RE: Help Rewriting/Refactoring/Rethinking Parsing Algorythm
...
Looking at your list, adding:
	py-caps  -> \No-Space-On\Caps-On
	py-norm  -> \No-Space-Off\Caps-Off
	py-name  -> <word> \No-Space-On\Caps-On [ not sure if you can do this
without a macro... ]
...
-----Original Message-----
From: Boopy Bootles [mailto:aschneid at mindspring.com]
Sent: Sunday, March 18, 2001 8:57 PM
To: python-list at python.org
Subject: Help Rewriting/Refactoring/Rethinking Parsing Algorythm


I'm trying to write a simple piece of code to make programming by voice
recognition software easier to do.  I wrote a very simple function that
would, for example, convert "number customers equals 5" to
"numberCustomers = 5".  But once I started using it, I quickly
discovered several more cases I had to handle.  So far, this is the
list:
...




From me at nospam.net  Sun Mar 25 02:09:34 2001
From: me at nospam.net (Scottie)
Date: Sat, 24 Mar 2001 23:09:34 -0800
Subject: "object" vs "dynamic" programming styles
References: <3ABD117E.1E5492C4@erols.com> <3ABD2390.A2A28A12@letterror.com>
Message-ID: <tbr6dkmnuodh0b@corp.supernews.com>

"Just van Rossum" <just at letterror.com> wrote in message
news:3ABD2390.A2A28A12 at letterror.com...
> "Edward C. Jones" wrote:
> > Here are three methods for doing the same thing. When is it better to
use
> > one of the approaches rather than another?
>
> > # Program 3
> >
> > # The function New_X returns an instance of either class A or class B.
> > def New_X(c):
> >     if c == 'a':
> >         return A()
> >     if c == 'b':
> >         return B()
>
> My personal preference would definitely be version 3, the factory
function.
> Less magic, clearer code.

I concur, but would cause an exception when your cases don't match.
This inspires a variation as follows:

    generators = {'a':A, 'b':B, 'c':C}

    def New_X(c):
         return generators[c]()

-Scott David Daniels






From cribeiro at mail.inet.com.br  Sat Mar 31 09:35:37 2001
From: cribeiro at mail.inet.com.br (Carlos Alberto Reis Ribeiro)
Date: Sat, 31 Mar 2001 11:35:37 -0300
Subject: list vs tuple
In-Reply-To: <MVhx6.1021$p5.4157@news1.rivrw1.nsw.optushome.com.au>
References: <hGlw6.782$p5.3210@news1.rivrw1.nsw.optushome.com.au>
 <3AC20CD3.265A2D94@alcyone.com>
 <ru69cto61jtmrqsf809e38hb3pbm2lhi7g@4ax.com>
 <bC1x6.4319$4N4.905946@newsc.telia.net>
 <71q9ct4le63anefmiajj5cbh3mg04qu1ga@4ax.com>
 <mailman.985991202.5578.python-list@python.org>
 <7Bfx6.989$p5.3791@news1.rivrw1.nsw.optushome.com.au>
 <Ukhx6.4421$4N4.950603@newsc.telia.net>
Message-ID: <5.0.2.1.0.20010331111632.021e92e0@mail.inet.com.br>

At 10:01 31/03/01 +0000, deadmeat wrote:
>What python DOES is irrelevant - the way it's written (and therefore viewed
>by programmers coming from other languages) makes it inconsistant.  The same
>statement (a = b) behaves entirely differently in each case.

I also came from a Pascal background (I used UCSB Pascal 15 years ago, and 
Turbo Pascal 1.0 (!) :-P), and I can tell you that this is not exactly true 
for Pascal either.

Using "pure" Pascal, an assignment is expected to copy the value from the 
right side to the left side. There is a number of catches, however:

- The copy operation can be expensive if the variable is a large object.
- Stack space is a concern. Copying large parameters take up a lot of stack 
space. There were some limitations on the return types accepted in older 
version of Pascal because of this also.
- There is an difference between "copy" and "deep copy". What happens if 
you make copy a structure that contains pointers in it? The pointers 
themselves are copied, but the "pointed" copy is still the same. As you 
don't have garbage collection or ref count, it's up to you to manage the 
situation to avoid freeing memory space while some copy of your structure 
is still pointing to the original memory area.

In Delphi, and also in other variants of Object Pascal, you have classes, 
and that changes everything. Class instances are always references, in a 
way very similar to Python classes. So if you do the following in Delphi, 
what do you get?

   a := TStringList.Create;
   b := a;
   a.Add('One line');
   /* b now contains the same line added to A */

As you may see, reference copy is everywhere - even inside Pascal :-)


Carlos Ribeiro








From lloyd at venix.com  Mon Mar 26 19:42:30 2001
From: lloyd at venix.com (Lloyd Kvam)
Date: Mon, 26 Mar 2001 19:42:30 -0500
Subject: boolean values in COM / VB / Python
References: <Pine.SOL.4.30.0103252251210.18627-100000@mimosa.csv.warwick.ac.uk>
Message-ID: <3ABFE1F6.4232DE4E@venix.com>

My guess is that you don't have a COM wrapper in place for the Document.  In
other words:

import win32com.client
ie = win32com.client.Dispatch("InternetExplorer.Application")
ie.Visible = 1	# toggles window visible
ie.Visible = 0	# toggles window invisible
ie.GetHome()	# loads the home page

Now when you refer to
ie.Document	
you are actually referring to a new object.  I believe that you will need to do
some work to provide the wrapper so that Python values get translated to the
equivalent COM values.

"John J. Lee" wrote:
> 
> ie.Document.execCommand("SaveAs", false, pathname)
> 
> where ie is an Internet Explorer object, pops up a dialogue box, no matter
> what value for false I try, from (1==0, 0, "0", None). According to the MS
> docs at
> 
> http://msdn.microsoft.com/workshop/author/dhtml/reference/constants/saveas.asp
> 
> and
> 
> http://msdn.microsoft.com/workshop/author/dhtml/reference/methods/execCommand.asp
> 
> this isn't supposed to happen, because when the second argument to
> execCommand is 'false', the command 'Does not display a user interface'.
> According to Mark Hammond just last month, 0 should do fine as a way of
> representing 'false':
> 
> http://groups.google.com/groups?hl=en&lr=&safe=off&ic=1&th=1cd0ed3943dead44&thn=Setting+boolean+COM+properties&seekd=926499672
> 
> Any clues as to what is happening?
> 
> Tips on the differences between C++ / VB and Python in general would also
> be appreciated -- presumably there must be some reliable and mechanical
> way of doing this?
> 
> John

-- 
Lloyd Kvam
Venix Corp.
1 Court Street, Suite 378
Lebanon, NH 03766-1358

voice:	603-443-6155
fax:	801-459-9582


From phlip_cpp at my-deja.com  Sat Mar 31 14:27:06 2001
From: phlip_cpp at my-deja.com (Phlip)
Date: 31 Mar 2001 19:27:06 GMT
Subject: Billy vs. Idle
Message-ID: <9a5b2a$fh9@freepress.concentric.net>

Hy on pt:

I done noticed that Bill Gate's program VC++ has this irrelevant little 
feature that nobody hardly notices: If you compile and your source has an 
error, you can tap <F4> and the editor will raise the offending file and 
highlight the offending line.

Why on earth doesn't IDLE have this feature?

(I know I know - because I haven't added it yet...)

-- 
  Phlip                          phlip_cpp at my-deja.com
============== http://phlip.webjump.com ==============
  --  Because I'm the sysadmin. That's why.  --


From me at nospam.net  Sat Mar 31 21:27:49 2001
From: me at nospam.net (Scottie)
Date: Sat, 31 Mar 2001 18:27:49 -0800
Subject: PythonLabs Team Moves to ActiveState
References: <mailman.986086501.5576.python-list@python.org> <v4wx6.121194$Ch.16134973@typhoon.we.rr.com>
Message-ID: <tcd4hd17gg2vbb@corp.supernews.com>

"Pete Shinners" <shredwheat at mediaone.net> wrote in message
news:v4wx6.121194$Ch.16134973 at typhoon.we.rr.com...
> i've got a knot in my stomach. i can't tell if this is a terrific
> april fools joke or just about the worst news i've heard in a long
> time.

I agree completely.
Imagine having to indent Perl!
the horror, the horror....




From johann at physics.berkeley.edu  Wed Mar 28 17:13:08 2001
From: johann at physics.berkeley.edu (Johann Hibschman)
Date: 28 Mar 2001 14:13:08 -0800
Subject: Is inheritance broken?
References: <mailman.985811257.29267.python-list@python.org>
Message-ID: <mtr8zha6ff.fsf@astron.berkeley.edu>

Jonathan Claggett writes:

> Interesting. Are there conventions for mixin classes which distinguish them
> from normal classes? For example, are mixins not suppposed to use
> inheritance (or at least not use mulitple inheritance)? Thanks for the help,

When I use mixins, I never use inheritance.  The mix-in just provides
methods, to be included into other classes.  There is no enforced
protocol, but I find that useful.

Otherwise, I have to start worrying about how python will search the mix-
ins, and that gets ugly quickly.

I also often define classes which are simply different combinations
of mix-ins, such as

class SpecPlotStepper(SpecMixin, PlotMixin, Stepper):
    pass

I hope that helps.  In theory, I like the way Common Lisp implemented
multiple inheritance, but that would require too much searching given
python's class system.

--J


-- 
Johann Hibschman                           johann at physics.berkeley.edu


From aahz at panix.com  Fri Mar  2 12:01:45 2001
From: aahz at panix.com (Aahz Maruch)
Date: 2 Mar 2001 09:01:45 -0800
Subject: gah! I hate the new string syntax
References: <mailman.983485923.1136.python-list@python.org> <B7In6.16281$Qb7.2658902@newsb.telia.net> <97ohh2$polt0$1@ID-77331.news.dfncis.de>
Message-ID: <97ojlp$g53$1@panix3.panix.com>

In article <97ohh2$polt0$1 at ID-77331.news.dfncis.de>,
Mark Pilgrim <f8dy at yahoo.com> wrote:
>
>-M feeling-old-and-crotchety-already-ly

<pat> <pat>  There, there, sonny, I know just how you feel.
-- 
                      --- Aahz (Copyright 2001 by aahz at pobox.com)

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

Nostalgia just ain't what it used to be


From nomail at nomail.com  Fri Mar 30 15:02:14 2001
From: nomail at nomail.com (Dry Ice)
Date: Fri, 30 Mar 2001 12:02:14 -0800
Subject: Know of Substantial Apps Written in Python?
Message-ID: <3ac4e4a8_1@news4.newsfeeds.com>

Know of substantial apps written in Python?

Preferably something one could download for
examination.

Source code would be nice, but not essential.

I'm operating under the theory that whatever
the POTENTIAL of a language, or the 'elegance'
of a language's syntax and structure, how large
apps actually behave in the real world will speak
volumes.

I like, for example, much about Java.  But nearly
every example I have seen running in the Windows
environment has been slow, leaked resources and
eventually contributed to system crashes.

(I don't care whether it's the language or the OS
or the VM or even "bad programming."  What counts
are the usual results obtained by normal application
developers.)

Suggestions?




From vincent_a_primavera at netzero.net  Tue Mar 27 10:57:03 2001
From: vincent_a_primavera at netzero.net (Vincent A. Primavera)
Date: Tue, 27 Mar 2001 10:57:03 -0500
Subject: Curses...
Message-ID: <01032710570303.00574@tos1>

Hello,
	I recently downloaded the "Curses Programming with Python" HOWTO and it 
suggests to begin by importing the curses module.  I tried to do this with: 
'import curses' to no effect.  I realize that there is a module wrapper.py in 
the /usr/lib/python2.0/curses directory but I am unsure of how to use this 
module to access the curses functions directly.  Any suggestions would be 
much appreciated.

	Thank You,
	Vincent A. Primavera



From phd at phd.pp.ru  Wed Mar 21 07:05:10 2001
From: phd at phd.pp.ru (Oleg Broytmann)
Date: Wed, 21 Mar 2001 15:05:10 +0300 (MSK)
Subject: Checking the type
In-Reply-To: <99a493$bhl$1@sun.rhrk.uni-kl.de>
Message-ID: <Pine.LNX.4.30.0103211504200.15322-100000@fep132.fep.ru>

On Wed, 21 Mar 2001, Markus Reitz wrote:
> is there a possibilty to check if a variable refers to an object of type x?

   Yes, builtin fuction type(x):

if type(x) <> type([]):
   print "not list!"

Oleg.
----
     Oleg Broytmann            http://phd.pp.ru/            phd at phd.pp.ru
           Programmers don't die, they just GOSUB without RETURN.




From amused at webamused.com  Thu Mar 22 17:23:03 2001
From: amused at webamused.com (Joshua Macy)
Date: Thu, 22 Mar 2001 22:23:03 GMT
Subject: Why Python/Jython?
References: <99cl9e$b2b$1@Merlin.et.bocholt.fh-gelsenkirchen.de> <mailman.985260089.28444.python-list@python.org> <1901D50DC30D30E4.16551C1AB0FB4363.C0D72E208759F8A7@lp.airnews.net> <99dkg301894@news1.newsguy.com>
Message-ID: <3ABAB2F7.127D3872@webamused.com>

Alex Martelli wrote:
> 
> "Cameron Laird" <claird at starbase.neosoft.com> wrote in message
> news:1901D50DC30D30E4.16551C1AB0FB4363.C0D72E208759F8A7 at lp.airnews.net...
>     [snip]
> >     meaningful.  People do NOT start learning
> >     Java lightly; it often takes hours just to
> >     install and configure a minimal JDK, let
> 
> *blink* -- I'm hardly a Java-ite, myself, but
> "install and configure a minimal JDK" I would
> estimate as typically a 10-minute job (on, say,
> NT -- is it an order of magnitude harder on,
> say, Linux?!).  Maybe I misunderstand what
> you mean...?
>

  It only takes about 10 minutes to download a Win32 installer and run
it--but if by configure one means understand and implement workable
solutions to the cryptic and bizarre interaction between Java package
namespaces, classpaths, and the WinNT filesystem, I think an hour might
be an underestimate.  Most programmers I know who art trying to "get
into" Java end up either putting every jar in the world into their
CLASSPATH environment variable, or downloading and installing something
like SmartJ...but it takes them more than an hour to arrive at one of
those solutions unless they have someone to help them.

  Joshua


From gradha at iname.com  Sat Mar 10 13:16:03 2001
From: gradha at iname.com (gradha at iname.com)
Date: Sat, 10 Mar 2001 19:16:03 +0100
Subject: Problems writting a wrapper around linux' ssh command
Message-ID: <31rd89.lki.ln@127.0.0.1>

Hi.

I am trying to write a python wrapper around ssh. The last and only problem
I have is password verification. I try to popen2.popen3 the ssh command,
but no matter what I write to the input file, the ssh command ignores it
and keeps requesting *keyboard* input.

Is there any way to avoid this?

-- 

 Grzegorz Adam Hankiewicz   gradha at iname.com - http://gradha.infierno.org
 Other web pages:  http://glub.ehu.es/  -  http://welcome.to/gogosoftware/


From grante at visi.com  Fri Mar 30 10:45:11 2001
From: grante at visi.com (Grant Edwards)
Date: Fri, 30 Mar 2001 15:45:11 GMT
Subject: Creation d'un NG francophone sur Python
References: <1eqzyk6.gxd4jr1tkjykzN%francois.granger@free.fr> <9a1e6f011es@enews3.newsguy.com>
Message-ID: <bS1x6.9225$SB2.838111@ruti.visi.com>

In article <9a1e6f011es at enews3.newsguy.com>, Michael Spencer wrote:

>Pardon, mais pourquoi un NG pour Python en fran?ais?
>
>Python a cr?? par un nederlandais qui a ?crit en anglais - pas
>sa langue natif. Il a ?crit en anglais ? assurer la plus grande
>distribution de ses id?es. Pourquoi on marcherait en arri?re
>aux temps quand chaque groupe a maintenu sa <<independance>>
>par s'isoler des autres?

I certainly would never vote against a French Python newsgroup.
But, I want to let you know that French language postings are
quite welcome in this group -- as are threads in any other
language.

You've even got a fair chance of getting replies to questions
posted in Esperanto.

-- 
Grant Edwards                   grante             Yow!  A GRAM?? A BRAM... A
                                  at               GROOM... A BROOM... Oh,
                               visi.com            Yeh!! Wash the ROOM!!


From greg at cosc.canterbury.ac.nz  Thu Mar  8 19:25:18 2001
From: greg at cosc.canterbury.ac.nz (Greg Ewing)
Date: Fri, 09 Mar 2001 13:25:18 +1300
Subject: "!=" is bad form. Re: sorry....never mind
References: <DZkp6.18811$pF4.3015652@news1.telusplanet.net> <fdlp6.16982$go.2157953@news0.telusplanet.net> <3AA6E308.5B63ECC6@cosc.canterbury.ac.nz> <3AA72253.FB021815@Lugoj.Com> <slrn9afcb5.lou.qrczak@qrnik.zagroda>
Message-ID: <3AA822EE.F77D956F@cosc.canterbury.ac.nz>

Marcin 'Qrczak' Kowalczyk wrote:
> 
> Icon uses "~=" for numbers and "~==" for strings. 

And I seem to recall it also uses ~=== for something else
as well (lists? dictionaries?).

By the time I got to the part about the corresponding 
assignment operators (~===:= et al) my brain was
laughing itself silly and thinking "This has *got* to
be a joke!"

Unfortunately, it wasn't.

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


From mwh21 at cam.ac.uk  Thu Mar 29 17:52:18 2001
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 29 Mar 2001 23:52:18 +0100
Subject: macro file (containing C def's) for source navigator?
References: <0m4w6.91100$A6.20818962@news4.rdc1.on.home.com>
Message-ID: <m33dbwchnh.fsf@atrus.jesus.cam.ac.uk>

"Tom" <NoSpam at NoSpam.com> writes:

> Does anyone have a Source Navigator macro file for the Python source
> code?  (This is a file that lists the preprocessor symbol
> definitions for a project.)

Well, I don't.

> Do the python internals hackers use Source Navigator?  What do they use?

Not a fully fledged python internals hacker, but I use emacs,
occasionally etags, cscope.  And gcc, of course.

> Is this information available somewhere in the source distribution,
> so I can make my own macro file?

Aren't there tools analagous etags & ctags to generate this sort of
info?

Cheers,
M.

-- 
  ARTHUR:  Why should a rock hum?
    FORD:  Maybe it feels good about being a rock.
                    -- The Hitch-Hikers Guide to the Galaxy, Episode 8


From ct at gocept.com  Thu Mar 22 11:56:22 2001
From: ct at gocept.com (Christian Theune)
Date: Thu, 22 Mar 2001 17:56:22 +0100
Subject: Curses under python 2.0
Message-ID: <99darm$o3o$00$1@news.t-online.com>

hmm ... it seems as the python-curses mod won't work:

Python 2.0 (#3, Feb 10 2001, 16:51:03) 
[GCC 2.95.2 19991024 (release)] on linux2
Type "copyright", "credits" or "license" for more information.
>>> import curses
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
  File "/usr/local/lib/python2.0/curses/__init__.py", line 15, in ?
    from _curses import *
ImportError: No module named _curses
>>> 

any idea? it is the installation from the python2.0 source, and I tested
the curses lib with python 1.5.2 - and it is working.

Thanks for replying to my mail address too.

best regards

Christian Theune


From danielk at aracnet.com  Fri Mar  9 11:03:35 2001
From: danielk at aracnet.com (Daniel Klein)
Date: Fri, 09 Mar 2001 08:03:35 -0800
Subject: Python RMI?
Message-ID: <ihvhatcm5v9rn943etearsp15k4q9425ia@4ax.com>

Does Python have an equivalent or substitute for Java's RMI?

Daniel Klein


From aahz at panix.com  Fri Mar 23 14:58:31 2001
From: aahz at panix.com (Aahz Maruch)
Date: 23 Mar 2001 11:58:31 -0800
Subject: New PEP: The directive statement
References: <200103201042.LAA20248@pandora.informatik.hu-berlin.de> <wzr8zpwn4e.fsf@sunshine.cs.uu.nl> <3d1yrp1osi.fsf@ute.cnri.reston.va.us> <3ABBA837.DEB8AB1B@san.rr.com>
Message-ID: <99g9t7$r9j$1@panix6.panix.com>

In article <3ABBA837.DEB8AB1B at san.rr.com>, Darren New  <dnew at san.rr.com> wrote:
>Andrew Kuchling wrote:
>>
>> It's really a documentation issue; people who release software should
>> say "I've tested this with Python 2.0", and people with 1.5.2 or 2.1
>> will have to look out, and perhaps provide patches to fix problems.
>
>What is the Python mechanism for ensuring a particular (or minimal) version
>of Python is being used? Is there one?  I'm talking about something like
>Tcl's [package require Tcl 8.3] or some such.

assert (sys.version_info[0] >= 2) and (sys.version_info[1] > 0)

(If you want this to work on Python 1.5.2 and earlier, use sys.version.)
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

"I won't accept a model of the universe in which free will, omniscient
gods, and atheism are simultaneously true."  -- M


From dnew at san.rr.com  Thu Mar 22 11:25:40 2001
From: dnew at san.rr.com (Darren New)
Date: Thu, 22 Mar 2001 16:25:40 GMT
Subject: email address verification
References: <99d5qi$95l$1@solaria.cc.gatech.edu>
Message-ID: <3ABA2785.3CF81953@san.rr.com>

> does python provide quick/simple way to verify that an email address exist?

It depends on what you mean by "exist" to a great extent.

> i looked at the smtp library and the verify method but that does not seem
> to be very standard anymore. 

It is standard, except spammers have caused most ISPs to turn off this
useful functionality, yes.  (Pendatically yours...)

> is there anything else that is simalar?

The best way to do this is to take the email address (fred at example.com) and
look up the MX records for the domain (example.com), or the A record if no
MX records. Find the MX record with the lowest priority number. Open an SMTP
connection to that host. Do a MAIL FROM:<> and then a RCPT
TO:<fred at example.com> then a RSET and a QUIT. If the RCPT TO and everything
before it succeeds, you have an "existing" email address. It might not be
the address that gets delivered to fred, but the mail will get delivered.
I.e., it might generate a bounce return, but there's no way to distinguish
between "I delivered it successfully to a program from which Fred will fetch
it" and "I delivered it successfully to a program which will automatically
generate a message saying Fred doesn't live here."

Hope this helps.

-- 
Darren New / Senior MTS & Free Radical / Invisible Worlds Inc.
San Diego, CA, USA (PST).  Cryptokeys on demand.
A million monkeys in a room with a million typewriters 
              will only yield half a million pregnant monkeys.


From sholden at holdenweb.com  Mon Mar  5 20:58:53 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Mon, 5 Mar 2001 20:58:53 -0500
Subject: Can anyone recomend a good intoduction to C...
References: <3aa366c6_2@nnrp1.news.uk.psi.net> <3aa3a3f0.36440038@news.laplaza.org> <mailman.983829368.26447.python-list@python.org>
Message-ID: <0zXo6.19377$ML1.968149@e420r-atl2.usenetserver.com>

"Timothy Grant" <tjg at exceptionalminds.com> wrote in message
news:mailman.983829368.26447.python-list at python.org...
> Hear, Hear.
>
> Another vote for K&R. It's ancient, but I still use it when I
> have to do C stuff.
>
> I used it to learn C programming 14 years ago. When I first
> picked it up, I said "I can't learn how to programme from this
> book!" So I bought one of those 1000 page doorstops referenced
> below. I quickly returned to K&R.
>
> On Mon, Mar 05, 2001 at 02:42:16PM +0000, Mats Wichmann wrote:
> > On Mon, 5 Mar 2001 10:14:35 -0000, "Simon Brunning"
> > <sbrunning at trisystems.co.uk> wrote:
> >
> > >... for a Python programmer?
> > >
> > >I'm going to have to bite the bullet and learn a bit of C. I'd at
> > >least like to learn the *right* way to do it.
> >
> > My advice is similar to the one that gets given to prospective Python
> > programmers:  start with the source - K&R (Kernighan & Ritchie). It's
> > kind of similar to the Python tutorial in ways: concise but completely
> > to the point, from people who know why the language was written that
> > way. This is what everyone learned the language from in the early
> > days. Stay away from those 1000-page doorstops with CDroms included.
>
> --
However, if you want to *read* C or C++ code rather than write it, Harbison
and Steele might be a better choice.

regards
 Steve





From mhuening at zedat.fu-berlin.de  Wed Mar 21 10:59:02 2001
From: mhuening at zedat.fu-berlin.de (Matthias Huening)
Date: Wed, 21 Mar 2001 16:59:02 +0100
Subject: python and mysqldb
References: <mjia99.b14.ln@braveheart.dyndns.org>
Message-ID: <99aj4a$digd$1@fu-berlin.de>

How did you try to insert the variable?
The following should work:
statement = "INSERT INTO test_table VALUES ('%s', 'test2', 'test3')" % temp


Matthias




From tjg at exceptionalminds.com  Wed Mar 21 13:26:51 2001
From: tjg at exceptionalminds.com (Timothy Grant)
Date: Wed, 21 Mar 2001 10:26:51 -0800
Subject: pwd behaviour when new users are added
In-Reply-To: <slrn9bg8hj.8m9.jepler@potty.housenet>; from jepler@inetnebr.com on Wed, Mar 21, 2001 at 03:38:28AM +0000
References: <mailman.985033765.15773.python-list@python.org> <slrn9bg8hj.8m9.jepler@potty.housenet>
Message-ID: <20010321102651.H16442@trufflehunter.avalongroup.net>

On Wed, Mar 21, 2001 at 03:38:28AM +0000, Jeff Epler wrote:
> On Mon, 19 Mar 2001 12:31:23 -0800, Timothy Grant
> >
> >if I 'reload(pwd)' it still doesn't work if I 'del pwd' and
> >then 'import pwd' it doesn't work.
> >
> >What am I doing wrong?
> 
> Calls to the underlying C library functions don't appear to re-open
> the password file.
> 
Thanks for the confirmation Jeff. As stated, I have written
Python version of all the functions in pwd and they work quite
well, with the added benefit that they return shadow
information as well. If anyone is interested, let me know.

-- 
Stand Fast,
    tjg.

Timothy Grant                         tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.         <><       (503) 246-3630
>>>>>>>>>>>>>Linux, because rebooting is *NOT* normal<<<<<<<<<
>>>>This machine was last rebooted:  63 days 22:40 hours ago<<



From shaleh at valinux.com  Fri Mar  2 17:43:31 2001
From: shaleh at valinux.com (Sean 'Shaleh' Perry)
Date: Fri, 02 Mar 2001 14:43:31 -0800 (PST)
Subject: cataloging words in text file
In-Reply-To: <3AA0124E.74A39783@comm.mot.com>
Message-ID: <XFMail.20010302144331.shaleh@valinux.com>

On 02-Mar-2001 Stephen Boulet wrote:
> I remember this homework assignment for my data structures (c++)
> class: read in a large file, and create a data structure containing
> every word in the file and the number of times it appears.
> 
> I was wondering how to do this in python. In c++ we had to do it
> with hash tables and b-trees.
> 
> Can you do it with dictionaries in python, with the key as the word
> and the data the number of times it appears? If there's any
> documentation in this problem domain that people could point out to
> me I would appreciate it.
> 


the algorithm is basically just like the C++ version (and same in any language
that supports hash/dictionaries).

while input:
   pieces = split input
   for each piece in pieces
       if hash.has_key(piece):
           hash[piece] = hash[piece] + 1
       else:
           hash[piece] = 0

the python advantage is the easy string manip.

> What about if you had a bunch of objects, like say stars, with
> attributes like position (two coordinates), magnitude, color. If you
> had to sort a bunch of them at a time by attributes (say at most
> 10^3 of them), would using dictionaries be a good idea, or should
> you start looking at interfaces to, for example, mysql?
> 

depends on the application, speed, storage ability, etc.  If this start list
will continue to grow a db sounds right.



From aleaxit at yahoo.com  Fri Mar  2 04:43:07 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 2 Mar 2001 10:43:07 +0100
Subject: catching a del
References: <j%zn6.322252$f36.11513719@news20.bellglobal.com>
Message-ID: <97npvd01b43@news1.newsguy.com>

Dan Parisien" <dan at eevolved.com> wrote in message
news:j%zn6.322252$f36.11513719 at news20.bellglobal.com...
> class obj:
>         def __init__(self, name, parent):
>                 parent.children[name] = self
>                 self.parent = parent
>                 self.name = name
>
>         def onDelete(self): #?!
>                 del self.parent[self.name]
>
> I want to be able to catch a del of an instance of the above class so I
can
> remove it from its parent dictionary. Is there a way? (note it cannot be
in
> a destructor because there will always be a reference inside the parent
> dict)

You need a "weak reference" in the 'parent' -- one that, by itself,
will not suffice to keep the referred-to entity 'alive' if no other
references exist.  See PEP 205, "Weak References" -- its URL is
http://python.sourceforge.net/peps/pep-0205.html, and the text
includes a good general explanation and pointers to previous
work on the subject.  If you need it right now, your best bet is
probably mxProxy, http://www.lemburg.com/files/python/mxProxy.html,
but you should also explore the other references in the PEP text.


Alex





From will at nospam.misconception.org.uk  Mon Mar 19 13:04:56 2001
From: will at nospam.misconception.org.uk (Will Newton)
Date: Mon, 19 Mar 2001 18:04:56 +0000
Subject: Subsetting a dictionary
References: <993dkm$ttr$1@news6.svr.pol.co.uk> <993qq3$4bajg$1@ID-11957.news.dfncis.de>
Message-ID: <995hj6$dk6$1@newsg3.svr.pol.co.uk>

Emile van Sebille wrote:

> Rather than copying and then deleting from the copy, try building a new
> dictionary while adding only those that qualify.

On average this makes the code ~10x slower. I haven't done an in depth 
analysis of how the routine is called, but my "backwards" way of doing 
things seems much faster for my task.



From tim.one at home.com  Wed Mar 21 19:59:59 2001
From: tim.one at home.com (Tim Peters)
Date: Wed, 21 Mar 2001 19:59:59 -0500
Subject: What is being compared?
In-Reply-To: <cfbu6.67$Hp.3494@typhoon.aracnet.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCKEGIJHAA.tim.one@home.com>

[Daniel Klein]
> Comparing a string to an integer...
>
> >>> 'a' < 1
> 0
>
> ...what is it actually comparing? Is it 'casting' anything?

No cast, but what is being compared isn't defined, just that cross-type
comparisons (except for cross-type all-numeric comparisons) are "consistent".
As a matter of implementation accident, CPython actually compares the string
names of the *types* in such cases, so the accident du jour is that
cmp(some_string, some_int) == cmp("string", "int"), which is greater than 0
(i.e., any string is "bigger than" any int -- today).

not-the-whole-truth-but-there's-not-much-future-seeking-truth-
    in-accidents-anyway-ly y'rs  - tim




From aleaxit at yahoo.com  Sat Mar 10 17:21:07 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sat, 10 Mar 2001 23:21:07 +0100
Subject: Setting an attribute on a Python object from C
References: <mailman.984151691.9715.python-list@python.org> <8Adq6.46418$1D5.2158622@e420r-atl1.usenetserver.com>
Message-ID: <98ea2a02ttb@news1.newsguy.com>

"Steve Holden" <sholden at holdenweb.com> wrote in message
news:8Adq6.46418$1D5.2158622 at e420r-atl1.usenetserver.com...
> "G. Sumner Hayes" <SumnerH at fool.com> wrote in message
> news:mailman.984151691.9715.python-list at python.org...
    [snip]
> > Here's what I'm doing (implementing a Python DB 2.0 driver for Microsoft
> > SQL server):
    [snip]
> Are you aware of mxODBC? Without wishing to steal your thunder, it's a
> pretty good ODBC for (SQL Server, Access, ...)

Yes, but ODBC is not the be-all, end-all of database interfacing -- it
is a pretty old technology, which its originator (Microsoft) is busy
trying to bury, and it does have its problems.  It may be a good thing
to have alternate, API-DB compliant ways to access RDBMS's...!!!


Alex





From jcm at bigskytel.com  Tue Mar  6 19:05:56 2001
From: jcm at bigskytel.com (David Porter)
Date: Tue, 6 Mar 2001 17:05:56 -0700
Subject: from Tkinter import *
In-Reply-To: <20010306151631.M3579@trufflehunter.avalongroup.net>; from tjg@exceptionalminds.com on Tue, Mar 06, 2001 at 03:16:31PM -0800
References: <20010306151631.M3579@trufflehunter.avalongroup.net>
Message-ID: <20010306170556.A6401@bigskytel.com>

* Timothy Grant <tjg at exceptionalminds.com>:

> With all the discussion around here recently about the
> 
> from foo import *
> 
> idiom, I thought I'd ask about Tkinter. All of Fredrik's
> examples in his guide on Tkinter use 
> 
> from Tkinter import *
> 
> Now, I'm aware that the effbot is never wrong about
> anything<wink>, but is it safe to assume that I should not be
> using the above idiom, but the import Tkinter idiom instead?

I don't recall reading any statements to that effect. Have you seen him use
it outside of the module scope (such as in a function)? The Language Ref
forbids doing that, and the addition of nested scopes further complicates
things. None of this would indicate that you shouldn't use the above idiom
at the module scope though (which is how I see it used).


David




From timo at alum.mit.edu  Wed Mar 21 00:46:05 2001
From: timo at alum.mit.edu (Timothy O'Malley)
Date: Wed, 21 Mar 2001 00:46:05 -0500
Subject: Cookies
References: <mailman.985050745.11337.python-list@python.org>
Message-ID: <210320010046053593%timo@alum.mit.edu>

[[ This message was both posted and mailed: see
   the "To," "Cc," and "Newsgroups" headers for details. ]]

hola.

> Thanks for your help and persistence.  I have come to an adequate
> compromise solution. . . .

Here's a wacky idea.

You could generate the id on the server or client side.  If the
client generates it in JavaScript, then he sends it to the server.  If
the server generates it, then he already has it.  In either case, the
the server who responds with an web page that includes an HTTP link
that points to the following URL
 
         http://localhost:8080/id=1234

On this page, you instruct the user to
    1) start up your Python application
    2) click on the link

The Python application should include a small (and stupid) HTTP sever
that listens on port 8080 of the localhost.  When it detects a
connection, it reads the ID out of the requested URL.  (In the above
example, the ID is the string "1234".)

When the user clicks on the link, the client browser makes a request
to the Python application running on his machine.  By this mechanism
is the unique key passed to the application.  (To be nice, the Python
application should respond with a nice web page.)


From shaleh at valinux.com  Fri Mar  2 16:36:25 2001
From: shaleh at valinux.com (Sean 'Shaleh' Perry)
Date: Fri, 02 Mar 2001 13:36:25 -0800 (PST)
Subject: the demise of 'from foo import * and its implications?
In-Reply-To: <XFMail.20010302130656.shaleh@valinux.com>
Message-ID: <XFMail.20010302133625.shaleh@valinux.com>

## as an example of what happens
## and sure, I could try to list every function I want to import,
## but man does that get ugly

import _gtk
import GTK

def hello(*args):
    print "Hello World"
    _gtk.gtk_widget_destroy(window)

def destroy(*args):
    _gtk.gtk_widget_hide(window)
    _gtk.gtk_main_quit()    

_gtk.gtk_init()

window = _gtk.gtk_window_new(GTK.WINDOW_TOPLEVEL)
_gtk.gtk_signal_connect(window, "destroy", destroy)
_gtk.gtk_container_set_border_width(window, 10)

button = _gtk.gtk_button_new_with_label("Hello World")
_gtk.gtk_signal_connect(button, "clicked", hello)
_gtk.gtk_container_add(window, button)
_gtk.gtk_widget_show(button)

_gtk.gtk_widget_show(window)

_gtk.gtk_main()




From t*b*r*u*m*l*e*y at f*u*s*e.n*e*t  Thu Mar 22 07:41:01 2001
From: t*b*r*u*m*l*e*y at f*u*s*e.n*e*t (Troy Brumley)
Date: Thu, 22 Mar 2001 07:41:01 -0500
Subject: Collection interfaces
References: <3a90fa97.0@news1.mweb.co.za>   <MPG.150394195f7f04a19896a7@news.earthlink.net> <3A9A86B1.9A8C69A2@ix.netcom.com> <MPG.1504d14563dce6499896af@news.earthlink.net> <MPG.1505a76831c34b749896b3@news.earthlink.net> <MPG.1506007591cf818b9896bc@news.earthlink.net> <slrn99opop.gr.amk@mira.erols.com> <MPG.1506265ebf6d934d9896be@news.earthlink.net> <97ibmc0k47@news1.newsguy.com> <MPG.1506cf3293b094b49896c1@news.earthlink.net> <97m0c7080c@news2.newsguy.com> s.earthlink.net> <zb%n6.31033$o%6.586026@news1.rdc1.bc.home.com> <MPG.150b383cea226d409896d0@news.earthlink.net> <kNko6.6986$eq3.707121@news.uswest.net> <MPG.150c46db9cec283e9896d4@news.earthlink.net> <3AA2A898.450B659E@mail.com> <gRCt6.12224$Im6.1308560@newsread1.prod.itd.earthlink.net> <3AB755F8.67E84126@mail.com> <%a%t6.17245$227.1601569@newsread2.prod.itd.earthlink.net> <3AB8A730.4BE13D22@mail.com> <B6DE17D2.D6B0%t*b*r*u*m*l*e*y@f*u*s*e.n*e*t> <anhu6.18497$Im6.1938203@newsread1.prod.itd.earthlink.net>
Message-ID: <B6DF5D0D.D7AE%t*b*r*u*m*l*e*y@f*u*s*e.n*e*t>

in article anhu6.18497$Im6.1938203 at newsread1.prod.itd.earthlink.net, topmind
at topmind at technologist.com wrote on 3/22/01 1:57 AM:

> 
> "Troy Brumley" <t*b*r*u*m*l*e*y at f*u*s*e.n*e*t> wrote in message
> news:B6DE17D2.D6B0%t*b*r*u*m*l*e*y at f*u*s*e.n*e*t...
>> in article 3AB8A730.4BE13D22 at mail.com, James A. Robertson at
>> jarober at mail.com wrote on 3/21/01 8:06 AM:
>> 
>>> topmind wrote:
>>>> 
>>> 
>>>>> http://www.cincom.com/smalltalk
>>>>> 
>>>> 
>>>> I only see a front page, not a parsing example.
>>>> 
>>> 
>>> Follow the tutorial link - it's right there on that page.
>>> 
>>> 
>>>> Besides, what are you comparing it to so say
>>>> it is better than a procedural/relational approach?
>>> 
>>> It's vastly easier than similar code done in C or basic, for example
>> 
>> User testimonial here ... last week I had to rip thorugh a pile of IIS
> logs
>> to get some statistics for some people, and I used the original (non-gui)
>> tutorial code as a starting point. I was once again impressed with how
> much
>> easier it was to do several common data processing tasks in Smalltalk than
>> it was in Assembly language or C or Pascal.
>> 
> 
> Give me specific requirements and I bet I can match it on code size.
> 
> C, Pascal, and Assembly are hardly what I would use for code-size
> contests.
> 
> Of course, there is more to comparing than code size, but those often
> get very subjective or fuzzy.

I rarely worry about code size except how it relates to code clarity. Even
back in my assembly oriented youth I tried to keep functions under one page
on a listing for the sake of clarity.

I mention C, Pascal, and Assembly because they are my usual tool set, or
were until I discovered Smalltalk. I've done a lot of business programming
over the years in procedural environments and used more languages than those
(even COBOL and RPG) but for the heavy lifting I would always stick with
them.

Again, until I discovered Smalltalk :)

I've seen Perl and run away screaming. Python and Eiffel are on the to learn
list, but Smalltalk seems to have the power I need for the tasks I want to
do these days. Objects are becoming more and more natural as I use them, and
my code improves contantly.

> 
>> I'm a long time procedural type, learning to really think in and use
>> Smalltalk has been an interesting experience.
>> 
>> I've also done a lot of database "fixing" and Smalltalk has usually beat
> out
>> raw SQL for anything but the most trivial tasks.
>> 
> 
> Fixing what? SQL is not good at string parsing.

Usually sweeping changes, or dangling referential integrity problems.

>> All that said, I think it's the environment (interactive) that improves
> the
>> experience as much as the language itself. If I had a reasonable
>> interpretive BASIC environment available, I could probably do many of the
>> things I did in Smalltalk almost as easily ... certainly more easily than
>> with C.
>> 
> 
> I find interpretive environments potentially more compact code-wise than
> static ones.

No argument here.

> 
>> 
>>> 
>>>> 
>>>>>> 
>>>>>>> [snip]
>>>>>> 
>>>>>> -tmind-
>>>>>> oop.ismad.com
>>>>> 
>>>>> --
>>>>> James A. Robertson
>>>>> Product Manager (Smalltalk), Cincom
>>>>> jarober at mail.com
>>>>> <Talk Small and Carry a Big Class Library>
>>>>> 
>> 
> 
> 



From dbrueck at edgix.com  Fri Mar  9 14:57:27 2001
From: dbrueck at edgix.com (Dave Brueck)
Date: Fri, 9 Mar 2001 12:57:27 -0700
Subject: "!=" is bad form. Re: sorry....never mind
In-Reply-To: <3AA928BC.1DAB4284@schneider-kamp.de>
Message-ID: <BLEOJGEMBNBNEAKMIFDHAELOPLAA.dbrueck@edgix.com>

> [mailto:python-list-admin at python.org]On Behalf Of Peter Schneider-Kamp
> Sent: Friday, March 09, 2001 12:02 PM
> To: Dave Brueck
> Cc: python-list at python.org
> Subject: Re: "!=" is bad form. Re: sorry....never mind

> Dave Brueck wrote:
> >
> > I don't really care if Python keeps <>, but the reason I don't
> like it is
> > because of the example listed above: to me seeing "<>" implies something
> > about the ordering of a and b, namely (a < b) or (a > b), even
> when such an
>
> Well, Python *does* define an ordering of a and b, no matter what
> a and b may be. E.g.:
>
> >>> 1 > (1,2)
> 0
> >>> 1 < (1,2)
> 1
> >>> {"a":1} < u"hello world"
> 1
> >>> {"a":1} > u"hello world"
> 0
>
> So, yes, interpreting "a <> b" as "(a < b) or (a > b)" makes
> perfect sense in Python.

Works? Yes. Makes sense? Definitely not.

My point wasn't about what is or isn't legal in Python but the intuitiveness
of '!=' vs '<>': at first glance the former says to me, "a is not the same
as b" while the latter says "a is less than b or a is greater than b". The
former intuitively makes sense for any "thing" while the latter makes sense
only on objects that can be ordered in some meaningful way.

Put another way, it doesn't really make any difference to me personally, but
'!=' would be easier to explain to a non-programmer than '<>'.

-Dave




From fredrik at pythonware.com  Thu Mar 22 18:23:31 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Thu, 22 Mar 2001 23:23:31 GMT
Subject: python, perl, lisp, tcl / got a decision to make, can you give me a  view point please ? and a joke for python users : )
References: <3aba3076.14395287@news.vif.com> <mailman.985284892.29036.python-list@python.org>
Message-ID: <TPvu6.3114$4N4.450174@newsc.telia.net>

Tim Churches wrote:
>
> from __future__ import *
>
> Crystal-ball-gazingly yours,

doesn't work; it imports both "warnings_are_errors" and
"from_import_star_is_evil", and as I've pointed out before,
catching ParadoxErrors is really hard.

(I'm sure Dave Beazley will fix WAD any day soon, though)

Cheers /F




From dusty at strike.wu-wien.ac.at  Thu Mar  8 12:48:31 2001
From: dusty at strike.wu-wien.ac.at (Hermann Himmelbauer)
Date: Thu, 8 Mar 2001 18:48:31 +0100
Subject: "Wrap" a python-cgi
Message-ID: <3aa7c5f9$0$21482@SSP1NO25.highway.telekom.at>

Hi,
I am currently using python for several cgi programs. One problem is that 
if my cgi-script has a bug it exits with an exception and the user gets an 
ugly error message done by the webserver.

I would like to put some really simple "bugfree" (if something like this 
can exist :-) "wrapper" around my cgi-scripts, let it trap Exceptions, log 
the error and let it output a nice error page to the user.

My first approach would perhaps be to wrap the whole script into a "try: 
except" clause but at first I don't know if nested try/except clauses are 
allowed and moreover the whole script looks somehow ugly as every line is 
indented.

Perhaps you know a better solution?

                Best Regards,
                Hermann

-- 
 ,_,
(O,O)     "There is more to life than increasing its speed."
(   )     -- Gandhi
-"-"--------------------------------------------------------------


From n8gray at caltech.edu.is.my.e-mail.address  Thu Mar  1 21:29:08 2001
From: n8gray at caltech.edu.is.my.e-mail.address (Nathan Gray)
Date: Thu, 01 Mar 2001 18:29:08 -0800
Subject: A counter-proposal to __future__ in PEP 236
References: <mailman.983479564.11788.python-list@python.org>
Message-ID: <3A9F0574.4040805@caltech.edu.is.my.e-mail.address>

Tim Peters wrote:

> [Martin von Loewis]
>> I personally consider this a confusing notation,
> 
> 
> I don't.  "__future__" screams "Deep Magic" because of the double
> double-underscores; it *is* a perfectly vanilla import statement at runtime;
> the word "future" helpfully suggests temporality; it works with all existing
> tools today without change (so it's not confusing even the stupidest of them
> <0.1 wink>); and the imported nested_scopes name gets bound to release data
> that the importing code can examine; e.g.,
> 
> from __future__ import nested_scopes
> assert nested_scopes[1] > sys.version_info, "don't need this anymore"
> 
> 
>> and propose a different syntax, which would read
>> 
>> directive nested_scopes
> 
> 
> Which loses all of that.  Incompatible language changes are a serious special
> case, and I don't want to see them swallowed up by some perfectly general
> mechanism that makes them impossible to distinguish visually from some
> innocuous request to, e.g., optimize for space instead of time.

I don't understand.  "import" is a pretty general mechanism and the proposed 
syntax is visually indistinguishable from a normal import except for the double-
double underbars.  There's no reason Martin's proposal couldn't be tweaked:
	import_directive __future__.nested_scopes
	import sys
	assert nested_scopes.mandatory_release > sys.version_info \
		"don't need this anymore"
with the same semantics (including importing the nested_scopes module) as your 
proposal.  Now you've got *both* the underbars *and* a new, scary keyword to warn 
you that voodoo is being done.  Call it "import_directive" to make the 
"importish" nature more explicit (and reduce possibilities of code-breakage 
because of "directive" being used as an identifier).  And yes, definitely make 
the new thing a keyword!  No sense beating around the bush!  If we're going to add
"magic imports" to the language, why not make them explicit?

As for the tool-breakage argument, the smart tools are going to have to be rewritten
to understand nested scopes anyhow.  How hard is adding a new keyword relative to
that?  And with "import_directive", the only change that needs to be made to the 
stupid tools is a cut-and-paste of the "import" code.

> 
> 
>> The directive statement currently only supports activating nested
>> scopes,
> 
> 
> I really dislike that
> 
>     directive nested_scopes
> 
> *looks* exactly the same as
> 
>     directive favor_space
>[snip]
> That will add to confusion over time, not reduce it.  Incompatible language
> changes have to kick you in the teeth.

But 
	from __future__ import nested_scopes

has the insidious effect of fooling the naive user into thinking he understands
what's going on when he actually doesn't!  At least with import_directive or
directive the naive user is forced to look at the documentation.  Once he does,
it's immediately clear that deep, dark things can be happening at compile time.

> 
> Note that it's also not enough to allow just one or two optional tokens
> following "directive".  One of the primary reasons no pragma facility ever
> got into Python is that in previous attempts nobody could agree on syntax
> that *was* sufficiently general (as a hint of the horrors you're signing up
> for, note that "optional static typing" fits-- because of the "optional"
> part --into a pragma framework; and once people realize that, they're going
> to hijack your proposal into an overly-generalized mess too).

Guilty as charged.
	import_directive static_typing
seems pretty clear to me.

If people are dead-opposed this becoming a general-purpose mechanism, call it
"import__future__" and only use it for the case PEP 236 addresses.  I think our 
objections are syntactic, not semantic.  I can understand the advantages
that import semantics bring to the situation, but I hate to see a keyword with 
clear, focussed meaning being extended for a single special situation.

-n8

-- 
_.~'^`~._.~'^`~._.~'^`~._.~'^`~._.~'^`~._
               Nathaniel Gray
     California Institute of Technology
      Computation and Neural Systems
       n8gray <at> caltech <dot> edu
_.~'^`~._.~'^`~._.~'^`~._.~'^`~._.~'^`~._



From tuttledon at hotmail.com  Thu Mar 15 13:50:35 2001
From: tuttledon at hotmail.com (Don Tuttle)
Date: Thu, 15 Mar 2001 18:50:35 GMT
Subject: Python Programming on Win32 (was: Magnitude of the wx* market (was: Python In A Nutshell - suggestions))
References: <3AB26DAC@MailAndNews.com>
Message-ID: <%98s6.116559$Z8.23732829@typhoon.southeast.rr.com>

"Laura Lewin" <Laura_01 at MailAndNews.com> wrote in message
news:3AB26DAC at MailAndNews.com...
> Hi,
> Sure, we'll take this suggestion into consideration (on expanding wxPython
> coverage) when we publish the next edition of "Python Programming on
Win32".

When do you plan to do that?

And please greatly expand the chapter on system administration!  Better yet,
create an entire book along the lines of Dave Roth's excellent "Win32 Perl
Scripting: The Administrators Handbook". In fact many sections of  "Python
Programming on Win32"  need to be entire books, not just quick intros.

This leads me to a mini rant that has been building for some time now.  I'll
try to contain myself ;-)

Over the last year I've spent several hundred dollars on NON Python books in
order to learn Python.  Or more precisely, what to do with Python.  It's not
that Python can't do it, it's that I don't know how to do it, or for that
matter, that I can do "it".  (I'm finding new "its" everyday.)  And most of
the "How To" books are written for other languages. So I have many books on
VBScript, JScript, WSH, Perl and Batch files. And this is just for Admin
work.  I haven't even started trying to use Python with databases.- Yet- But
I will.  (And then there's that network programming stuff...)

So, all you authors and Publishers out there...Put out more topic specific
Python "How-To" books.  Let's liberate Python from the hands of highly
skilled developers and give it to your average power user! Power to the
Python! <wink>

Don

PS  And don't forget all those pissed off VB programmers that are ripe for
the picking!







From igorr at ifi.uio.no  Fri Mar  2 13:17:39 2001
From: igorr at ifi.uio.no (Igor V. Rafienko)
Date: 02 Mar 2001 19:17:39 +0100
Subject: str behaviour
References: <mailman.983482383.19568.python-list@python.org>
Message-ID: <xjvitls3uik.fsf@hel.ifi.uio.no>

* Tim Peters

[snip]

Thanks to everyone who replied.

> You're on the edge of a very deep pit. The builtin container types
> (lists, tuples, dicts) "pass repr down" regardless of whether str()
> or repr() is applied at the top level. I've been whining about that
> for years myself <wink>.


Argggh! This is silly -- why won't repr call repr and str call str
recursively? It seems like a very logical solution (consistent, imvho,
at least).

 
> The primary reason is maddeningly simple: it's for the benefit of
> contained strings. If str() "got passed down", then, e.g.,
> 
>     d = {"colon": ":", "comma": ",", "' greeting": "I'm A"}
>     print d  # note that print implicitly applies str()
> 
> would print
> 
>    {' greeting: I'm A, comma: ,, colon: :}
> 
> instead of today's
> 
>    {"' greeting": "I'm A", 'comma': ',', 'colon': ':'}
> 
> so it would be an unreadable mess.


Suits me fine :)

Jokes aside, if that were the case, I'd be more than happy to write

>>> print repr( d )

rather than 

>>> print d

and be very happy with the fact that repr did in fact call repr and
str did in fact call str. But maybe I am a very strange person...

BTW, since things are the way they are, could someone provide a
cleaner alternative to this snippet:


try:
    # some code that might raise something nasty
except:
    return MyError( "Something went wrong: (%s, %s)" %
	 	    ( sys.exc_info()[0], sys.exc_info()[1] ) )
# yrt

(yes, I can teach emacs to do that for me, but this is very painful on
the eyes of a reader).

All I wanted to do was to extract the value and type of the exception
raised in an elegant way:

return MyError( "Something went wrong: %s" % sys.exc_info()[:2] )
			
(which, obviously (gaah!), does not work as intended). A possible
solution is to use repr, of course (and define a repr for the class
that can be raise'd).

[snip]


> Like most other issues involving repr-vs-str, it doesn't often bite
> so long as you stick to the builtin types. But once you write
> classes that implement their own notions of __str__ and __repr__, it
> seems that more often than not *neither* of them does what you want
> when instances end up in builtin containers.


I'd rather have str/repr call str/repr, than not. Am I missing
something obvious (apart from ugliness of strings, as you've
exemplified above)?





ivr
-- 
Besides, meat tends to run away when possible, or fights. Either
response presents behavioral challenges too complex for any existing
robot.
		-- Stuart Wilkinson, inventor of the "gastrobot"


From fig at monitor.net  Sat Mar 31 21:41:22 2001
From: fig at monitor.net (Stephen R. Figgins)
Date: Sat, 31 Mar 2001 18:41:22 -0800
Subject: PythonLabs Team Moves to ActiveState
References: <mailman.986089142.10423.python-list@python.org> <9a62cg$3it52$1@ID-11957.news.dfncis.de>
Message-ID: <3AC69552.7D6FAD6A@monitor.net>

This was a good one!  I was halfway through it before I realized it was
almost word for word the same as the DC announcement.  When I got to the
part about braces I about fell out of seat laughing. 

Check out the X-Apology header.

-Stephen


From ben.hutchings at roundpoint.com  Wed Mar 28 17:15:48 2001
From: ben.hutchings at roundpoint.com (Ben Hutchings)
Date: 28 Mar 2001 14:15:48 -0800
Subject: daemon app in win32
References: <99nk49$a2q$02$1@news.t-online.com> <mailman.985619314.10299.python-list@python.org> <99nsn3$q39$06$1@news.t-online.com> <99o2ge$9on$0@216.39.170.247> <uu24ebw63.fsf@roundpoint.com> <99rhsp$i4$06$1@news.t-online.com>
Message-ID: <uy9tp7d63.fsf@roundpoint.com>

Michael Bauer <Mike_B at T-Online.de> writes:

> Ben Hutchings wrote:
<snip>
> > Console applications can register to receive such notifications in the
> > same way they receive control-C notifications under *some* versions of
> > Windows.
> 
> Do you have any more details about this? Because i experienced that my app 
> does actually *not* receive the CTRL-Cs or similar i am pressing when i try 
> to cancel it "by hand"... (btw, this is win98)
<snip>

CPython is supposed to generate a KeyboardInterrupt exception in this
case.  If this doesn't work, Ctrl-Break might.  If your program has
stuck in a system function then the exception won't be thrown until
that returns.

-- 
Any opinions expressed are my own and not necessarily those of Roundpoint.


From tjreedy at udel.edu  Wed Mar 14 14:55:02 2001
From: tjreedy at udel.edu (Terry Reedy)
Date: Wed, 14 Mar 2001 14:55:02 -0500
Subject: This math scares me
References: <3aad4062.1290021@News.CIS.DFN.DE> <20010312170356.B3513@better.net> <mailman.984480856.18097.python-list@python.org> <984486395.407379@newsmaster-04.atnet.at> <slrn9asf1d.ja3.qrczak@qrnik.zagroda> <984518199.127140@newsmaster-04.atnet.at> <jwbnews-09D75F.13313313032001@corp.supernews.com>
Message-ID: <98ohuc$du9$1@news.udel.edu>

> Bill Gates got this right in the early days, delivering two Basic
> interpreters:  binary floating point for speed, and decimal floating
> point for money and such

If one counts pennies instead of dollars, then 32 bit ints (+- $20 million)
will serve for many purposes.   Exact, faster, and more widely available
than BCD




From tim.one at home.com  Tue Mar 20 22:18:25 2001
From: tim.one at home.com (Tim Peters)
Date: Tue, 20 Mar 2001 22:18:25 -0500
Subject: ASCII encoding error: ordinal not in range(128)
In-Reply-To: <3d4rwo10h8.fsf@ute.cnri.reston.va.us>
Message-ID: <LNBBLJKPBEHFEDALKOLCIEBAJHAA.tim.one@home.com>

[Fredrik Lundh]
>> UnicodeError: ASCII encoding error: ordinal not in range(128)
>
> translated to english, this overly cute error message attempts
> to say something like "cannot convert this Unicode string to
> ASCII as it contains non-ASCII characters"

[Andrew Kuchling]
> Hmm... hasn't changing this message before been discussed someplace?

Yes, /F griped about this on Python-Dev, Sat, 26 Aug 2000 17:43:38 +0200, and
suggested

    ASCII decoding error: character not in range 0-127

instead.  I counter-proposed

    ASCII decoding error: ord(character) > 127

since characters aren't integers in Python.  AFAICT, that's the grand total
of the debate.

> Was there some reason it wasn't changed?

None apparent.

> How about: "UnicodeError:  ASCII encoding error: character not in
> ASCII range 0-127" ?  (I vaguely recall Marc-Andre arguing for
> "ordinal", since that's used in the Unicode documents, but if it's
> less clear...)

Anything is better than what it says now!

even-a-ham-sandwich-ly y'rs  - tim




From lrl at ou.edu  Wed Mar 28 17:14:37 2001
From: lrl at ou.edu (Luke L)
Date: Wed, 28 Mar 2001 16:14:37 -0600
Subject: Recommendations for Python Books
References: <3ac2582c$1@news.starhub.net.sg>
Message-ID: <3AC2624D.F8704C44@ou.edu>

"Learn to Program using Python"
http://www.amazon.com/exec/obidos/ASIN/0201709384/qid=985817618/sr=1-10/ref=sc_b_11/102-6408521-3462514



From dougfort at downright.com  Mon Mar  5 09:47:01 2001
From: dougfort at downright.com (Doug Fort)
Date: Mon, 5 Mar 2001 09:47:01 -0500
Subject: iis and python
References: <F5622749FED6DA5D.AF4594816A4865C9.16CD98F1C1FAC3F6@lp.airnews.net>
Message-ID: <FFNo6.20259$Sl.896754@iad-read.news.verio.net>

My boss (David Cohen) has coined the acronym 'WIMP' (Windows, IIS, MySQL,
Python) corresponding to 'LAMP' (Linux, Apache, MySQL, Python).  Our  CGI
stuff will run under either Apache or IIS.

"tech" <im_cne at hotmail.com> wrote in message
news:F5622749FED6DA5D.AF4594816A4865C9.16CD98F1C1FAC3F6 at lp.airnews.net...
> I am going to do a project building a web front end  while my friend is
> going to build the back end.
>
> The backend will be MS SQL. The front end will be a web interface, IIS 5.0
> or whatever comes with win2k .
>
> I have a choice of doing php or python.
>
> In order to use python it would have to entail cgi or ASP. Is this
correct.
> Or is there an alternative. I dont know if Zope is the right thing. I hate
> DTML.
>
> Is using python in ASP a pain in the but? Am I going to have to learn a
> whole technology just to implelment python well into it?
>
> I hear troubleshooting ASP pages is not fun.
>
> The only time I have used iis is with php and it was a total snap.
Painless.
>
>




From tchur at optushome.com.au  Thu Mar  8 14:53:12 2001
From: tchur at optushome.com.au (Tim Churches)
Date: Fri, 09 Mar 2001 06:53:12 +1100
Subject: Who am I: can a class instance determine its own name?
References: <mailman.984034150.9635.python-list@python.org> <slrn9af71n.d6p.scarblac@pino.selwerd.nl>
Message-ID: <3AA7E328.3A28AF26@optushome.com.au>

Thanks to Gregory Jorgensen, Jeremy Reed, Steve Purcell, Steve Holden,
Matthias Oberlander, Daniel Klein and Remco Gerlich for responding to my
question: can a class instance determine the name of the variable which
refers to it? The answer appears to be "no, it can't". A number of
respondents pointed out that multiple variables may all refer to the
same class instance, so my question should have been: "can a class
instance determine the name(s) of the variable(s) which refer(s) to
it?". The answer is still no, and that the solution is to explicitly
assign each instance a name attribute at instantiation.

A number of people asked why I needed to know. Consider the problem of
composition. You define a container class Foo which can contain many
instances of the Bar class, which are added to Foo with the addBar
method.

######################################
Foo_One = Foo(name="Foo_One")
Foo_Two = Foo(name="Foo_Two")

Bar_A = Bar(name="Bar_A")
Bar_B = Bar(Name="Bar_B")

Foo_One.addBar(Bar_A)
Foo_One.addBar(Bar_B)

Foo_Two.addBar(Bar_B)

Foo_One.listBars()
...returns a sequence of Bar instance names: ['Bar_A','Bar_B'] 
Foo_Two.listBars()
...returns a sequence of Bar instance names: ['Bar_B'] 
######################################

I was just trying to avoid the inelegant redundancy of having to name
each Bar instance twice - once when specifying the name of the variable
which refers to the Bar instance, and once as an attribute of that
instance (the use of a keyword parameter to __init__ to specify the name
is just for clarity).

What I was aiming at was this sort of functionality:

######################################
Foo_One = Foo(name="Foo_One")
Foo_Two = Foo(name="Foo_Two")

Bar_A = Bar()
Bar_B = Bar()

Foo_One.addBar(Bar_A)
Foo_One.addBar(Bar_B)

Foo_Two.addBar(Bar_B)

Foo_One.listBars()
...returns a sequence of Bar instance names: ['Bar_A','Bar_B'] 
Foo_Two.listBars()
...returns a sequence of Bar instance names: ['Bar_B'] 
######################################

One solution would be:

######################################
Foo_One = Foo(name="Foo_One")
Foo_Two = Foo(name="Foo_Two")

Foo_One.addBar(Bar(name="Bar_A"))
Foo_One.addBar(Bar(name="Bar_B"))

Foo_Two.addBar(Bar(name="Bar_B"))

Foo_One.listBars()
...returns a sequence of Bar instance names: ['Bar_A','Bar_B'] 
Foo_Two.listBars()
...returns a sequence of Bar instance names: ['Bar_B'] 
######################################

...but that just creates a different form of redundancy...

The solution I have settled on is to accept the minor inelegance and
move on to more important matters...

Cheers,

Tim C



From jmarshal at mathworks.com  Thu Mar  1 16:21:40 2001
From: jmarshal at mathworks.com (Joshua Marshall)
Date: 1 Mar 2001 21:21:40 GMT
Subject: A counter-proposal to __future__ in PEP 236
References: <mailman.983479743.12289.python-list@python.org>
Message-ID: <97meh4$c0g$1@news.mathworks.com>

Tim Peters <tim.one at home.com> wrote:
> [Joshua Marshall]
>> ...
>> Note that it doesn't need to be implemented this way.  Currently,
>>
>>   directive foo
>>
>> is a syntax error.  Meaning can be assigned to this syntax without
>> breaking any existing code--it is unambiguous.

> You should study Python's parser before getting too optimistic about that
> <wink>.

So anybody care to post an example of how this can be ambiguous (in
context) so I don't need to go study Python's parser...?


From sholden at holdenweb.com  Wed Mar 28 16:50:58 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Wed, 28 Mar 2001 16:50:58 -0500
Subject: list vs tuple
References: <hGlw6.782$p5.3210@news1.rivrw1.nsw.optushome.com.au> <3AC1EB93.7BC87292@hack.org> <wEmw6.785$p5.3426@news1.rivrw1.nsw.optushome.com.au> <3AC1F93A.BAC9E34B@hack.org> <99teuu02p5f@news1.newsguy.com>
Message-ID: <O4tw6.6798$m04.497167@e420r-atl1.usenetserver.com>

"Alex Martelli" <aleaxit at yahoo.com> wrote in message
news:99teuu02p5f at news1.newsguy.com...
> "Rikard Bosnjakovic" <bos at hack.org> wrote in message
> news:3AC1F93A.BAC9E34B at hack.org...
> > deadmeat wrote:
> >
> > > so is that the only difference?
> >
> > Both are sequences, but tuples are "read-only", that's the only
> > difference (except for the in-place-methods which requires mutability).
>
> Actually, tuples have NO methods -- including those that would
> require no mutability at all:
>
> D:\pybridge\pybr>python
> Python 2.0 (#8, Oct 16 2000, 17:27:58) [MSC 32 bit (Intel)] on win32
> Type "copyright", "credits" or "license" for more information.
> >>> dir([])
> ['append', 'count', 'extend', 'index', 'insert', 'pop', 'remove',
'reverse',
> 'sort']
> >>> dir(())
> []
> >>>
>
> Note the lack of 'count' and 'index', which do not mutate a list
> when applied to it, but are still missing from tuples.
>
So, tuples are not only immutable, they are also uncooperative.

regards
 Steve





From ulf at light.mobiliant.com  Thu Mar 15 08:13:26 2001
From: ulf at light.mobiliant.com (=?iso-8859-1?Q?Ulf_Engstr=F6m?=)
Date: Thu, 15 Mar 2001 14:13:26 +0100
Subject: Problem with string.translate
References: <s3e1bt8ujpmbrtdcsojffc2u9mt63jebip@4ax.com>
Message-ID: <00b001c0ad51$bc0d0520$968647c1@asker>

What version and what platform are you using?
Works fine for me with 2.0final on a Windows2000 and on a SuSE Linux.
Include traceback (with orginial version of the library function) and I'll
see if I can help :)
Ulf


"David Fuess" <fuess at att.net>:
> I am using Python 2.0 and attempting to perform a string.translate.
> The translate string is statically allocated in the setup phase. I
> perform a
>
> result = string.translate(source, ttab)
>
> I have also tried
>
> result = source.translate(ttab)
>
> The error I get is that an internal library function in the string
> module is calling a builtin function translate with too few arguments!
> Is this an implementation problem in the new string class? I have
> tried changing the library function, but then it generates other
> errors. Please advise.





From phrxy at csv.warwick.ac.uk  Tue Mar 27 12:54:57 2001
From: phrxy at csv.warwick.ac.uk (John J. Lee)
Date: Tue, 27 Mar 2001 18:54:57 +0100
Subject: Threading legality/morality
In-Reply-To: <99qa3v$6e0$1@panix6.panix.com>
References: <fsDv6.1247$NR.100201@news3.oke.nextra.no> <99nnbv$phs$1@panix6.panix.com> <h0%v6.1842$NR.144463@news3.oke.nextra.no> <99qa3v$6e0$1@panix6.panix.com>
Message-ID: <Pine.SOL.4.30.0103271851370.7646-100000@mimosa.csv.warwick.ac.uk>

On 27 Mar 2001, Aahz Maruch wrote:

> In article <h0%v6.1842$NR.144463 at news3.oke.nextra.no>,
> Syver Enstad <syver at NOSPAMcyberwatcher.com> wrote:
> >
> >I have seen some strange behaviour in some multi threaded programs I have
> >made in python, where they seem to lockup, but a push on the the return key
> >(or another key, I don't know) keeps them going. These are console programs,
> >running under Win2000 server and professional, Python 2.0.
[...]
> There is a misfeature in urllib such that the standard FancyURLOpener
> class prompts for input if it gets an HTTP 401 (authorization needed).
> You probably pressed ENTER.
[...]

Perhaps you're right, but I've seen the same behaiviour on win95, without
multi-threading.  When you run command-line (do we mean the same thing by
console and command-line?) apps it freezes until a key gets pressed.  Not
using urllib either.

It is possible (likely, even) that I was using cygwin, so maybe that's
where the blame lies.  I'll see if I can reproduce it with the standard
DOS shell.

(using standard (BeOpen) Python 2.0 distribution)


John



From emile at fenx.com  Wed Mar 21 07:26:49 2001
From: emile at fenx.com (Emile van Sebille)
Date: Wed, 21 Mar 2001 04:26:49 -0800
Subject: Checking the type
References: <99a493$bhl$1@sun.rhrk.uni-kl.de> <99a539$buco$1@ID-59026.news.dfncis.de>
Message-ID: <99a6s1$b9no$1@ID-11957.news.dfncis.de>

There is "type" as well:

if type(a) == type(1)
if type(b) == type({})

etc

Emile van Sebille
emile at fenx.com

"Nils Hensel" <nils at hhk.de> wrote in message
news:99a539$buco$1 at ID-59026.news.dfncis.de...
> Look for the types module:
>
> from types import *
> if type(a) is ListType:
>   # do list stuff
>
> HTH,
> Nils
>
> "Markus Reitz" <Markus_Reitz at yahoo.com> schrieb im Newsbeitrag
> news:99a493$bhl$1 at sun.rhrk.uni-kl.de...
> > Hi,
> >
> > is there a possibilty to check if a variable refers to an object of type
> x?
> > Does a typeof-command exist in Python:
> >
> >   if a typeof list:
> >      #Commands only useable with list
> >
> > I read the Python Tutorial, but have not found a hint to this topic.
> >
> > Thanks
> >
> > Markus
> >
> >
>
>




From topmind at technologist.com  Thu Mar  1 21:57:53 2001
From: topmind at technologist.com (Topmind)
Date: Fri, 02 Mar 2001 02:57:53 GMT
Subject: Collection interfaces
References: <3a90fa97.0@news1.mweb.co.za> <3A933830.7F98D118@ix.netcom.com> <CgJk6.515$v77.66568@newsread2.prod.itd.earthlink.net> <220220010813598042%cmh@bDistributed.com> <ud7cakvi8.fsf@roundpoint.com> <MPG.150312ceb68f1ada98969f@news.earthlink.net> <8Gem6.473353$U46.14105062@news1.sttls1.wa.home.com> <MPG.1503548620e632919896a4@news.earthlink.net> <MCjm6.474652$U46.14163288@news1.sttls1.wa.home.com> <MPG.150394195f7f04a19896a7@news.earthlink.net> <3A9A86B1.9A8C69A2@ix.netcom.com> <MPG.1504d14563dce6499896af@news.earthlink.net> <97fjla128ir@news2.newsguy.com> <MPG.1505a76831c34b749896b3@news.earthlink.net> <97h5s40vhc@news2.newsguy.com> <MPG.1506007591cf818b9896bc@news.earthlink.net> <slrn99opop.gr.amk@mira.erols.com> <MPG.1506265ebf6d934d9896be@news.earthlink.net> <97ibmc0k47@news1.newsguy.com> <MPG.1506cf3293b094b49896c1@news.earthlink.net> <97m0c7080c@news2.newsguy.com>
Message-ID: <MPG.15085517debf4b649896ca@news.earthlink.net>

>
>I've often noticed that demanding metrics is the last
>line of defense of the conservatives in our field against
>innovation 

So science be damned? At least it is official now.

I am *not* against innovation. I am against the overhyping
of fads without proof.

Relational DB's are newer than OO, does that mean that
OO defenders are "holdouts"?

If OO is an "experiment", then fans of it should admit
it as such. The road is littered with paradigms and
ideas that were over-hyped, and then shrank back 
to where they belong. Heavy top-down is an example
that ran about a decade in the limelight. Expert
systems, CASE, etc, are others.

I hope Table Oriented Programming hits hard and does
the same. All the OO celebs will squirm. Call me
cruel, but they deserve it. It deserves as much
of a chance for experimentation in the mainstream
as anything out there now.


> -- since nobody's really funded to run serious
>(and hugely costly) controlled experiments, such demanders
>can rest reasonably tranquil that they won't be faced with
>such metrics.  Unfortunately, the world has this habit of
>passing them by anyway.

No, just learning by trial and error. 50 years from
now we will have a better perspective on this OO
experiment.


> Been there, done that, more or less whatever you mean by
> "table-oriented programming"

It may be purely subjective. I had an affinity for grids
and tables even in 2nd grade. Born to Table if you will.


> ...then your resulting sources will never be as good,
> clear, clean, maintainable, and high-quality, as those
> of a better programmer, who DOES refactor every time
> the occasion affords.

My favorite approaches are *not* orthogonal to refactoring,
but simply *less* fragile or messed up if you refactor
regularly.


> And, depending on the axis of change, any paradigm
> can result in code that is hard to change to meet
> the new/changed needs at equally high quality

Well, my experience tells me that the "task axis"
is the most "business friendly" axis to build
along (being that multi-axis paradigms still need
a lot of work and thus we are forced to emphasize
one at the expense of others for now.)


>You still need design ability, flair, and a decent
>amount of domain-specific experience somewhere in
>the team, to enjoy a substantial chance of delivering
>LARGE software-systems that present all the desired
>quality attributes

The SIZE thing again?

1. What is good for large size may not be optimal for
other sizes.

2. OO fans disagree about at what size OO shines in.

3. The task-centric approach is size friendly because
the task is mostly impervious to the rest of the
system (unless the task changes it's requirements, which
is an issue in any paradigm).

OO's noun-centricity seems more scale-poor in that
regard because schemas change when growth appears.
The 3-graph approach puts a layer between schema
changes and the task more or less.


> Fortunately, given the current market for IT professionals (and
> this comment applies to the last 25 years), no competent worker
> in the field ever needs to be stuck for long under management
> that is totally impervious to common sense and reason

For one, I don't buy the labor shortage crap. For example, try
to get a telecommuting job. If there was an IT labor shortage, then
bosses would fold on their demand to see a physical butt in a
physical chair.

Second, the percentage of PHB's is too great to avoid. 
The supply of non-PHB's is too low to satisfy the
desire of tech workers to escape them.


>> I have shown in great detail why procedural seems better able
>> to deal with non-boundaried differences, and I have yet to
>
> You have done no such thing (at least in your posts in
> this thread, and the URL you mentioned here).

I don't know what you have not seen yet, but my collection
of examples is far greater than what the other side
offers as far as comparative examples. All I get from
them are the typical shape, animal, and stack examples
(and a few other "classics").

Regarding an internal* component view, I will believe it when I
actually see it work. Like much in OO lore, it sounds
wonderful on paper, but rubs rough with the nitty
gritty of reality. I guess that is what happens when
you design an entire paradigm around idealized shape,
animal, and stack examples.

* External components are quite successful, but nobody
has figured out how to apply it to internal systems
very well. BTW, OO is not a prerequisite for external
components.


> maybe because you lack the experience to do it well.

Well, youses all seem to lack the articulation skills/experience
to describe why granularity is not a problem. You still
have not found an easy way to override 1/3 of a method.
You just hand-wave that "experience will magically fix
it all". I seems like you just "get used to" refactoring
when it happens. IOW, tolerate bad weather because you
lived their for so long.

> But -- without calling it that way, without the productivity
> that reusable patterns or good tools can give me

Patterns are not native to OO and "good tools" can be built
to lessen the shortcomings of *any* paradigm. Name a p/r
fault and I could describe a potential tool to help.

-tmind-



From cfelling at iae.nl  Mon Mar 12 10:24:45 2001
From: cfelling at iae.nl (Carel Fellinger)
Date: 12 Mar 2001 16:24:45 +0100
Subject: I have no idea
References: <g13r6.201$qt5.10920@news3.mia> <98io58$462$1@animus.fel.iae.nl>
Message-ID: <98ipnt$4jf$1@animus.fel.iae.nl>

Carel Fellinger <cfelling at iae.nl> wrote:
...
> or add a ',' to the print statement, like:

>     >>> while 1:
>     ...     print "Hello World"

I said add a ',', so I should, shouldn't I.

      ...     print "Hello World",

-- 
groetjes, carel


From sholden at holdenweb.com  Thu Mar 22 23:37:09 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 22 Mar 2001 23:37:09 -0500
Subject: Good Editor?
References: <3ab9a32b$1_1@news2.one.net> <99cv2t$f20g$1@ID-11957.news.dfncis.de> <sqzu6.12865$_46.475246@e420r-atl2.usenetserver.com> <99ej5k$t8im$1@ID-11957.news.dfncis.de>
Message-ID: <AtAu6.9171$G72.246663@e420r-atl1.usenetserver.com>

That's impressively well-though-out. Although really I meant Unicode
literals in Python sources, which might be a horse of a different color.

regards
 Steve

"Emile van Sebille" <emile at fenx.com> wrote in message
news:99ej5k$t8im$1 at ID-11957.news.dfncis.de...
> <OstrichMode>
>
> What's unicode look like in hex?  How would I test it?
>
> </OstrichMode>
>
>
> Oh wait, from the Help page:
>
> How to Work with Unicode Files
> Overview:
> TextPad automatically detects 16-bit Unicode and UTF-8 encoded characters,
> when opening files. Unicode characters may be in "little endian" (Intel)
or
> "big endian" (RISC) order, and the order is preserved when a file is
saved.
>
> Internally, these files are converted to single or double byte characters
> (DBCS), using the locale corresponding to the font script selected for the
> document class. For example, if the screen font for the Text document
class
> is MS Mincho, with the script set to Japanese, Unicode characters in *.TXT
> files will be converted to the corresponding DBCS characters in code page
> 932.
>
>
>
> Emile van Sebille
> emile at fenx.com
>
>
> "Steve Holden" <sholden at holdenweb.com> wrote in message
> news:sqzu6.12865$_46.475246 at e420r-atl2.usenetserver.com...
> > "Emile van Sebille" <emile at fenx.com> wrote in message
> > news:99cv2t$f20g$1 at ID-11957.news.dfncis.de...
> > >
> > > For windows, I like textpad http://www.textpad.com/ beause it uses
> > > workspaces, has a hex display mode, runs code in a separate process,
> > > supports other languages (programming and spoken), and generally
behaves
> > > well.
> > >
> > > Emile van Sebille
> > > emile at fenx.com
> > >
> > Does it show Unicode string constants correctly in hex?
> >
> > regards
> >  Steve
> >
> >
> >
>
>




From paulp at ActiveState.com  Fri Mar 16 01:40:40 2001
From: paulp at ActiveState.com (Paul Prescod)
Date: Thu, 15 Mar 2001 22:40:40 -0800
Subject: Python & XSLT?
References: <CDgs6.46717$tP3.760880@news1.rdc1.bc.home.com>
Message-ID: <3AB1B568.7B98C58F@ActiveState.com>

"John E. Barham" wrote:
> 
>...
> 
> So I want to use XSLT.  Short of wrapping a C++ XSLT engine like Apache's
> Xalan, is there any way to do XSLT from within Python?  As far as I can
> tell, the existing Python XML tools don't go much above DOM/SAX style
> processing...

I think you're looking for this:

http://4suite.org/index.epy

-- 
Take a recipe. Leave a recipe.  
Python Cookbook!  http://www.activestate.com/pythoncookbook



From juergen.erhard at gmx.net  Sun Mar 11 07:51:17 2001
From: juergen.erhard at gmx.net (=?ISO-8859-1?Q?=22J=FCrgen_A=2E_Erhard=22?=)
Date: Sun, 11 Mar 2001 13:51:17 +0100
Subject: "!=" is bad form. Re: sorry....never mind
In-Reply-To: <LNBBLJKPBEHFEDALKOLCCELJJEAA.tim.one@home.com>
References: <LNBBLJKPBEHFEDALKOLCCELJJEAA.tim.one@home.com>
Message-ID: <11032001.1@wanderer.local.jae.ddns.org>

>>>>> "Tim" == Tim Peters <tim.one at home.com> writes:

    Tim> I'm afraid it's again time to quote the Reference Manual:

    Tim>     [...] for consistency with C, != is preferred; [...]

    Tim> print>>-got-in-for-"consistency-with-c"-too-ly y'rs  - tim

<curious mode="slightly confused">
So when can we expect `||' and `&&' to replace `or' and `and'?
Shouldn't we `Beware of the slippery slope'?
</curious>

Bye, J

-- 
J?rgen A. Erhard    juergen.erhard at gmx.net   phone: (GERMANY) 0721 27326
     MARS: http://members.tripod.com/Juergen_Erhard/mars_index.html
        The Python Programming Language (http://www.python.org)
           C> (A)bort (R)etry (I)nfluence with a large hammer
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 248 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/20010311/b0385aa8/attachment.sig>

From sheila at spamcop.net  Sun Mar 18 05:29:08 2001
From: sheila at spamcop.net (Sheila King)
Date: Sun, 18 Mar 2001 10:29:08 GMT
Subject: Help with 500 error on cgi script
References: <quv8bto7r8s6f1tckc7qpmuir0kvufp1nb@4ax.com> <mailman.984909623.8814.python-list@python.org>
Message-ID: <5739bt04e50ku01rn3dptgqch3l07p2cnf@4ax.com>

On Sun, 18 Mar 2001 11:00:22 +0100, Steve Purcell <stephen_purcell at yahoo.com>
wrote in comp.lang.python in article
<mailman.984909623.8814.python-list at python.org>:

:You need to write out a 'Content-type' header when form_ok is true, too!
:
:It's the lack of that header that causes the 500 error. If you're pretty
:sure your content is going to be text/html, it can be good practice to write
:out the content header before any other code.

That is a good point, although I don't think that that is what the problem
was, because now I've changed the script to this, where I print the Content
header right away, and still I'm having problems. More below:


#!/usr/bin/env python


import cgi

print "Content-type: text/html\n"
print "<HTML>\n<Head>\n<Title>Result Page</Title></Head><Body>"

form = cgi.FieldStorage()
form_ok = 0
    
if form.has_key("subscriberEmail") and form.has_key("command"):
    form_ok = 1
if not form_ok:
	output = '''<H1>Error</H1>
You have not filled out all necessary information to complete this request
You <b>MUST</b> fill in your subscriber e-mail in the From: box
and you <b>MUST</b> select some action to modify your delivery options.
Please use the back button on your browser to return to the form and fill
in the required information
</body>\n</html>
'''
	print output
else:
	print "<H1>Good Result!</H1>
	print "</body>\n</html>"
"""	for k in form.keys():
    	print k, " : ", form[k]
	print "All done!"""




And I'm still getting errors. The Apache error log says, on the last run:

[Sun Mar 18 02:18:42 2001] [error] [client 4.3.200.226] Premature end of
script headers: e:/apache/cgi-bin/testform.py
[Sun Mar 18 02:18:42 2001] [error] [client 4.3.200.226]   File
"e:/apache/cgi-bin/testform.py", line 25

[Sun Mar 18 02:18:42 2001] [error] [client 4.3.200.226]     print "<H1>Good
Result!</H1>

[Sun Mar 18 02:18:42 2001] [error] [client 4.3.200.226]
^

[Sun Mar 18 02:18:42 2001] [error] [client 4.3.200.226] SyntaxError: invalid
token


But I'm at a loss to figure out what is wrong. :-(

:-Steve
:
:P.S. What's with the 'spamcop' reply address? Is that your real address or
:do unsuspecting helpful responders get their mail accounts terminated?

http://spamcop.net is a service that I subscribe to. Nothing untoward happens
to helpful responders. It is a protection for me, so that I do not have to
reveal my "real" e-mail address in the newsgroups. But, if you reply, the
replies will be delivered to me. If spammers reply, though, they will be
easily and promptly reported.

--
Sheila King
http://www.thinkspot.net/sheila/
http://www.k12groups.org/




From sholden at holdenweb.com  Tue Mar 20 02:31:48 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Tue, 20 Mar 2001 02:31:48 -0500
Subject: Why "from __future__" stinks; a counter-offer
References: <mailman.985051704.15160.python-list@python.org> <3xDt6.2394$4N4.290991@newsc.telia.net>
Message-ID: <ZKDt6.121953$ML1.5692363@e420r-atl2.usenetserver.com>

"Fredrik Lundh" <fredrik at pythonware.com> wrote in message
news:3xDt6.2394$4N4.290991 at newsc.telia.net...
> Tim Peters wrote:
> > Fine, so I'm an idiot
>
> Me too!
>
> Cheers /F
>
With idiots like you two guys, who needs gurus?

regards
 Steve




From hinsen at cnrs-orleans.fr  Thu Mar  1 04:41:41 2001
From: hinsen at cnrs-orleans.fr (Konrad Hinsen)
Date: 01 Mar 2001 10:41:41 +0100
Subject: d = {}; d[0:1] = 1; d[0:1] = 2; print d[0:1]
References: <mailman.983419142.24027.python-list@python.org> <mailman.983429941.26766.python-list@python.org>
Message-ID: <m3ae75rfl6.fsf@chinon.cnrs-orleans.fr>

Robin Thomas <robin.thomas at starmedia.net> writes:

> The question remains: should slice objects be hashable/comparable so that
> 
>  >>> slice(0,1) == slice(0,1)
> 1
> 
> I say YES YES YES. What do you say?

Another YES YES YES. That's what anyone would expect in my opinion.
-- 
-------------------------------------------------------------------------------
Konrad Hinsen                            | E-Mail: hinsen at cnrs-orleans.fr
Centre de Biophysique Moleculaire (CNRS) | Tel.: +33-2.38.25.56.24
Rue Charles Sadron                       | Fax:  +33-2.38.63.15.17
45071 Orleans Cedex 2                    | Deutsch/Esperanto/English/
France                                   | Nederlands/Francais
-------------------------------------------------------------------------------


From fgeiger at datec.at  Sat Mar 10 03:00:38 2001
From: fgeiger at datec.at (Franz GEIGER)
Date: Sat, 10 Mar 2001 09:00:38 +0100
Subject: Python In A Nutshell
References: <3AB1700C@MailAndNews.com>
Message-ID: <98cn2d$oad$1@newsreaderg1.core.theplanet.net>

> We all know Alex--hi Alex!--uses lots and lots of words to say things.

That's one of his qualities and should be greatly appreciated. Alex EXPLAINS
things and allways figures out the PYTHONIC way to do things. For me it
always is a pleasure to read Alex' answers. It' s high time that Alex writes
a book, isn't it? Consider this: A book has to save you time. If one writes
too much you can pick out what you need for your problem at hand. If one
writes too little, the book is a waste of time.

Regards
Franz

"Laura Lewin" <Laura_01 at MailAndNews.com> wrote in message
news:3AB1700C at MailAndNews.com...
> Hi,
> We all know Alex--hi Alex!--uses lots and lots of words to say things.
> <grin>
> But he certainly knows how to make a point....(wait and see!)  It will be
a
> nutshell not a treatise.  No schedule yet.
> Laura
> LLewin at oreilly.com
>
> >===== Original Message From "Don Tuttle" <tuttledon at hotmail.com> =====
> >> [Don Tuttle]
> >> >Planning stages?  We've all seen his posts (a.k.a. treatises) here.
I'd
> >say
> >> >he should have the final draft to you in 3 weeks tops. ;-)
> >>
> >>[Aahz ]
> >> ITYM "first draft".  It will take another three-to-six months for Alex
> >> to edit it down.  ;-)
> >
> >Good point Aahz. And of course another six months for O'reilly to cut it
> >down to a mere 1,200 pages!  All in all, it MIGHT be out just in time for
> >Christmas 2002  ;-)
> >
> >Don
> >
> >
>




From not.this at seebelow.org  Fri Mar 30 23:27:52 2001
From: not.this at seebelow.org (Grant Griffin)
Date: Fri, 30 Mar 2001 22:27:52 -0600
Subject: Know of Substantial Apps Written in Python?
References: <3ac4e4a8_1@news4.newsfeeds.com> <KE6x6.415$SR6.235983@typhoon1.ba-dsg.net> <3ac4ff12$1_2@news4.newsfeeds.com>
Message-ID: <3AC55CC8.44850ECF@seebelow.org>

Dry Ice wrote:
> 
...
> What "substantial" is might be defined in terms
> of project length.  Perhaps something which would
> take a good programmer several months.

If you measure "substantial" in terms of months, the answer is a
no-brainer.  Yup, legions of folks have "been there, done that".

For example, I did a Python project recently which grew "organically",
in spurts, over a period of about a year; it probably took several
months of total work.  The project was one of those things that nobody
really planned to do, but just needed doing.  It started out small and
grew bigger with each feature request.  In terms of Python code,
features grew into functions, functions grew into classes, classes grew
into modules, and modules grew into libraries.

Even though the project had not been planned whatsoever in advance,
virtually no code got thrown away: it was always a forward process. 
Python code wears well.

> Something as complex as a web browser might do,
> if the language in question does not merely act as
> glue for modules written in C or whatever.

Rumor has it that Google is implemented in Python.  (OK, maybe not _all_
of it. <wink>)

> In the end, this is like evaluating people.  We can
> listen to what others say about a person, or what a
> person says about his or her self- or we can simply
> see how they BEHAVE.

Then you probably won't be much intersted in what I'm about to say.
<wink>

A related question that is often asked about a programming language is,
"does it scale well"?  To that, I think the answer for Python is
definitely "yes", because:

- It is extremly readable--more so than any other programming language
I've seen.  (I can't think of anything that might constrain complexity
more than Humans not being able to understand something.  Take Perl, for
example. <wink>)
- It supports namespaces: you can write large pieces of code that don't
conflict with each other.
- It's extensible and embed-able: if you need more speed or whatever,
you can get it.

In fact, if we turn your question around to consider what might _limit_
complexity (in the extreme), the only thing I can think of is Python's
lack of a formal interface mechanism.  (For more about that, see PEP
245, at http://python.sourceforge.net/peps/pep-0245.html.)  It's no
coincidence that this issue is of interest to the Zope folks.

complex-is-better-than-complicated-ly y'rs,

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation            http://www.iowegian.com


From shaleh at valinux.com  Mon Mar 26 20:43:39 2001
From: shaleh at valinux.com (Sean 'Shaleh' Perry)
Date: Mon, 26 Mar 2001 17:43:39 -0800 (PST)
Subject: math.choice attribute error
In-Reply-To: <Pine.LNX.4.21.0103270124290.3010-100000@moroni.pp.asu.edu>
Message-ID: <XFMail.20010326174339.shaleh@valinux.com>

On 27-Mar-2001 iddwb wrote:
> 
> from the ref man math.choice says it takes a sequence object and returns a
> random choice. perfect for my needs. the follow happens:
> 
>>>> import math
>>>> list = ['apple', 'orange', 'pear']
>>>> print math.choice(list)
> Traceback ....
> attributerror: choice
> 

I know random does:

>>> import random
>>> l = [1, 2, 3]
>>> random.choice(l)
2




From kens at sightreader.com  Wed Mar 21 18:38:34 2001
From: kens at sightreader.com (Ken Seehof)
Date: Wed, 21 Mar 2001 15:38:34 -0800
Subject: Dynamically reloading extension modules
Message-ID: <00ff01c0b260$0bd225a0$04090a0a@upcast.com>

Not exactly.  I didn't state the problem very clearly.  The
problem is that while an extension module is loaded, the
module file (pyd on windows) is read-only.

At least this is true on windows.  On unix, I think maybe you
can rename the module file, rebuild it, and do a reload(module).
Is this true?  Windows doesn't allow you to rename or rewrite
a dll while it is in use, so something like FreeLibrary would be
necessary.

"Timothy Grant" <tjg at exceptionalminds.com> says:
> On Wed, Mar 21, 2001 at 02:01:44PM -0800, Ken Seehof wrote:
> > Has anyone found a technique for dynamically unloading and
> > reloading an extension module?  I want to be able to modify,
> > build, and reload an extension module from within my program.
> > 
> > Unfortunately python apparently doesn't use FreeLibrary when
> > a module object is released.  Maybe I can do some kind of
> > hack to call FreeLibrary explicitly.  Is there a way to do this that
> > is portable?
> 
> I'm not sure I know what your asking, but does
> 
> reload(modulename) fit the bill?

----------------------------------------------------
Copyright (c) 2001 by Ken Seehof
This document may not be distributed, copied,
duplicated, or replicated in any form without
express written permission by Ken Seehof.
Express permission is hereby granted.
kens at neuralintegrator.com
----------------------------------------------------


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20010321/934ddb4d/attachment.html>

From ws-news at gmx.at  Thu Mar 22 10:50:04 2001
From: ws-news at gmx.at (Werner Schiendl)
Date: Thu, 22 Mar 2001 16:50:04 +0100
Subject: Text printing under Windows
References: <slrn9bk130.10l.ego@macondo.visconti.home>
Message-ID: <985276499.676502@newsmaster-04.atnet.at>

I think you can just open the file 'LPT1' and write your data there, but not
sure about that - do not have a printer connected which I could use to try.
You probably need to open the device with Windows' CreateFile() API
function, provided by the win32file module in Python.

Another possibility is to dump the data to a text file and print it with
notepad.exe (it has an option /p to just print and exit...).

One more possibility is to use the ShellExecute(Ex) APIs, using 'print' as
the verb for that file (it must be a textfile (*.txt) for that to work)

hth
werner

"Gabriele Favalessa" <ego at macondo.visconti.home> wrote in message
news:slrn9bk130.10l.ego at macondo.visconti.home...
> Hi,
>
> how do I print using a dot matrix printer in text mode under Windows
> 98?  I have both a local and a remote printer.
>
> I use python 2.0 and wxPython 2.2.5.  The latter seems to support
> printing only in bitmap mode.
>
> My knowledge of win32 APIs is poor, I'm looking for a simple
> open/write/close API.  Something similar to:
>
>    printer = os.popen ("lpr -Pprinter1")
>    printer.write ("Hello world!\n")
>    printer.close ()
>
> that works fine under unix.
>
> Thanks in advance.
>
>    Gabriele




From TCHUR at doh.health.nsw.gov.au  Wed Mar 21 01:43:43 2001
From: TCHUR at doh.health.nsw.gov.au (Tim CHURCHES)
Date: Wed, 21 Mar 2001 17:43:43 +1100
Subject: Base 2 to long integers and back
Message-ID: <sab8e856.060@dohprimo.ccf.health.nsw.gov.au>

Brian,

Thanks. I just discovered that the string.atol() function (or the long() built-in in 2.x) takes an optional radix argument, thusly:

>>> a = '0000101010100000111010101001010000111110100101010100101010010100101010100101010100000111110101010101000101001001001010'>>> print string.atol(a,2)13796684847850186040196945077883466L
but there is no built-in or library function which does the reverse, it would seem. However your code does the trick nicely and quite quickly, although a small C extension to do the same would be nifty. Now where did I put my copy of "Teach Yourself C in 23 Minutes or More"?

Cheers,

Tim C

>>> Brian Quinlan <brian at sweetapp.com> 21/03/2001 17:19:11 >>>
I can't think of a library function that does this, but it should be pretty
easy to do it yourself.

String to int:

>>> test = '10110'
>>> val = 0L
>>> for i in test:
...     val <<= 1
...     if i == '1':
...             val += 1
>>> val
22

OR, for the anal and efficiency oriented:

>>> test = '10110'
>>> val = 0L
>>> for i in test:
...     val = ( val << 1 ) + ( ord( i ) & 1 )
>>> val
22

Int to string:

>>> out = ''
>>> val = 100
>>> while val:
...     out = str( val & 1 ) + out
...     val >>=1
>>> out
'1100100'

-----Original Message-----
From: python-list-admin at python.org 
[mailto:python-list-admin at python.org]On Behalf Of Tim CHURCHES
Sent: Tuesday, March 20, 2001 7:23 PM
To: python-list at python.org 
Subject: Base 2 to long integers and back


Can anyone suggest a fast method of converting base2 strings (e.g.
'0000101110') to Python long integers, and back? I couldn't locate any
functions in the standard library to do this, but no doubt I am overlooking
something obvious.

Cheers,

Tim Churches
Sydney, Australia



--
http://mail.python.org/mailman/listinfo/python-list 





From sebastien.libert at comexis.com  Tue Mar 13 05:11:30 2001
From: sebastien.libert at comexis.com (Sébastien Libert)
Date: Tue, 13 Mar 2001 11:11:30 +0100
Subject: file - codecs - unicode ???
References: <3aace9da$0$382$446d705c@news.skynet.be> <slrn9aqka1.3vk.qrczak@qrnik.zagroda>
Message-ID: <3aadf0bf$0$384$5a62924d@news.skynet.be>

Hi,

Thx for your advice !!!!!
It work ( ?? sometimes !!!!)
But i have a lot of problem with the /012 :
If i keep it :

>>> fline=unicode(fline,'UTF-16').encode('LATIN-1')
Traceback (innermost last):
  File "<interactive input>", line 1, in ?
UnicodeError: UTF-16 decoding error: truncated data

if i delete it :

>>> fline=fline[:-1]
>>> fline=unicode(fline,'UTF-16').encode('LATIN-1')
Traceback (innermost last):
  File "<interactive input>", line 1, in ?
UnicodeError: Latin-1 encoding error: ordinal not in range(256)

Another clue :
I have converted the file in UTF-8 ( with notepad ) :

gline
'0.00\011info\011Log_Version\0111.2\012'
>>> gline=unicode(gline,'UTF-8').encode('LATIN-1')
>>> gline
'0.00\011info\011Log_Version\0111.2\012'
;-)

So the file was in UTF-16 and is now in UTF-8 : The result is good for me !
When i do that with python :

f=open("c:\\mylog.log")
>>> fline=f.readline()
>>> fline=unicode(fline,'UTF-16').encode('utf-8')
Traceback (innermost last):
  File "<interactive input>", line 1, in ?
UnicodeError: UTF-16 decoding error: truncated data


If you have another advise for me !! Fell free to answer !
thx
"Marcin 'Qrczak' Kowalczyk" <qrczak at knm.org.pl> wrote in message
news:slrn9aqka1.3vk.qrczak at qrnik.zagroda...
> Mon, 12 Mar 2001 16:29:42 +0100, S?bastien Libert
<sebastien.libert at comexis.com> pisze:
>
> > >>> line
> >
'\377\3760\000.\0000\0000\000\011\000i\000n\000f\000o\000\011\000L\000o\000g
> >
\000_\000S\000t\000a\000n\000d\000a\000r\000d\000\011\000n\000g\000L\000o\00
> > 0g\000\015\000\012'
> >
> > What can i do with this kind of thing ?????
>
> It's encoded in UTF-16, so:
>     unicode(line, 'UTF-16').encode('ASCII')
> except that the '\012' at the end is bogus. It shouldn't be there.
>
> You may want to use a different encoding than ASCII, because ASCII
> is only able to encode Latin letters without accents. If the string
> contains other character, you will get an exception.
>
> --
>  __("<  Marcin Kowalczyk * qrczak at knm.org.pl http://qrczak.ids.net.pl/
>  \__/
>   ^^                      SYGNATURA ZAST?PCZA
> QRCZAK




From mda at idatar.com  Sat Mar 17 13:41:52 2001
From: mda at idatar.com (David Allen)
Date: Sat, 17 Mar 2001 18:41:52 GMT
Subject: How do I avoid the Tkinter 30 msec polling loop.
References: <98toci$ktjh$1@fido.engr.sgi.com>
Message-ID: <QdOs6.506$Fb.774749@typhoon2.ba-dsg.net>

In article <98toci$ktjh$1 at fido.engr.sgi.com>, "Paul Jackson" <pj at sgi.com>
wrote:

<snip>

> Anyhow, is there anyway to stop or radically slow this loop?
> 
> The little window responds to no key or mouse input.  It just changes the
> text message when Label.configure (text="you have mail") or some such is
> executed after the once a minute thread wakes up and notices a change in the
> mail box.

Check out:
http://www.pythonware.com/library/tkinter/introduction/x9129-alarm-handlers-and-other.htm

It is a page on Tkinter non-event callbacks.  It has
a set of methods described there for Tkinter where you
can call a certain function once every X milliseconds
where you specify X.  If you read that page, I think
it will answer your questions.

--
David Allen
http://opop.nols.com/
----------------------------------------
Be wary of strong drink. It can make you shoot at tax collectors and 
miss 
- Lazarus Long, "Time Enough for Love"


From aleaxit at yahoo.com  Sun Mar 25 11:50:49 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sun, 25 Mar 2001 18:50:49 +0200
Subject: static methods
References: <mailman.985474834.2256.python-list@python.org> <3ABD7673.9D447A32@home.com>
Message-ID: <99l7mu01ed6@news1.newsguy.com>

"Rick Lee" <rwklee at home.com> wrote in message
news:3ABD7673.9D447A32 at home.com...
> Thanks, and thanks to Alex.  This is very useful, also very telling that
it
> is pretty obscure.  For me, static, or class-wide methods are very useful.
> With this, you only have to pass the class reference to the caller.
> Otherwise, you have to pass both the class reference and the module
> reference to the caller.

Actually, the module reference will suffice; it will contain suitably
named callables to:
    -- generate instances of the class (the class-object itself and/or
        suitable factory functions),
    -- do whatever else you need to be done class-wise.

You can package up those callables any way you want; there is
really no special reason they should be attributes of the class
object as opposed to that of any other object, and a module
object is typically handiest.

The desire to use the class-object specifically, rather than any
other, typically seems to have more to do with familiarity with
other languages where classes must play certain special roles
(C++, Java, Smalltalk, Eiffel) than with any actual programming
needs.


Alex





From syver at NOSPAMcyberwatcher.com  Thu Mar 29 01:02:05 2001
From: syver at NOSPAMcyberwatcher.com (Syver Enstad)
Date: Thu, 29 Mar 2001 08:02:05 +0200
Subject: Recommendations for Python Books
References: <3ac2582c$1@news.starhub.net.sg>
Message-ID: <BcAw6.779$R6.22088@news1.oke.nextra.no>

If you're using Win32 buy Andy Robinson and Mark Hammond's Python
Programming on Win32. Also be sure to check out www.python.org and the
documentation that comes with Python. There are also much to learn from
studying the library and samples that come with python, check out the
Tool/scripts directory in the Python directory for lots of small and
interesting python programs.

"Wong Da Qiang" <noen at none.com> wrote in message
news:3ac2582c$1 at news.starhub.net.sg...
> Just a newbie to programming and python, need some recommendations on
books
> to buy
>
>




From loewis at informatik.hu-berlin.de  Wed Mar 14 05:50:58 2001
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 14 Mar 2001 11:50:58 +0100
Subject: Embedded Python and sharing Data?
References: <mailman.984410416.11454.python-list@python.org>
Message-ID: <j44rwwmxnx.fsf@informatik.hu-berlin.de>

Karl Bellve <Karl.Bellve at umassmed.edu> writes:

> The question is, how do I share data between python and my
> program. The data can be as large as a gigabyte and is usually
> stored in memory. I obviously don't want to duplicate this data :-)
> Can I pass pointers? Can I create a Python Object but then how do I
> tell python that it contains a normal integer array?

The best way would be to define your own data type, which should
implement the buffer protocol (see
http://www.python.org/doc/current/api/buffer-structs.html for the
definition of the buffer protocol, and xxmodule.c for a minimum
example of a C type). Then you get an object for which you can do
indexed access to bytes, slicing, etc. You can even use
PyBuffer_FromMemory if it is *always* stored in memory
(i.e. accessible through a pointer); in that case you don't have to
implement your own type.

If you want to expose the data in a more structured way (e.g. as an
array of ints), you again have to define your own type, and provide it
with accessor functions. You'ld implement the sequence protocol, and
construct integer objects when given an integer index.

Regards,
Martin


From nanotech at europa.com  Wed Mar 28 16:47:56 2001
From: nanotech at europa.com (nanotech at europa.com)
Date: Wed, 28 Mar 2001 21:47:56 -0000
Subject: Getting all the *files* from a directory -- A better way??
Message-ID: <99tm6c+bsk6@eGroups.com>

Thanks to everyone who answered!

Notice that:

  dir="/a/b/c"

  files1=[f for f in os.listdir(dir)
            if os.path.isfile(f)]
  files2=[os.path.basename(f) for f in os.listdir(dir)
                              if os.path.isfile(f)]

would be much "prettier" than:

  files1=[os.path.join(dir,f) for f in os.listdir(dir)
                              if os.path.isfile(os.path.join(dir,f))]
  files2=[f for f in os.listdir(dir)
            if os.path.isfile(os.path.join(dir,f))]

if os.listdir returned the given path prepended -- but oh well!!

Quentin Crain





From moshez at zadka.site.co.il  Fri Mar 30 15:24:19 2001
From: moshez at zadka.site.co.il (Moshe Zadka)
Date: Fri, 30 Mar 2001 22:24:19 +0200
Subject: Know of Substantial Apps Written in Python?
In-Reply-To: <3ac4e4a8_1@news4.newsfeeds.com>
References: <3ac4e4a8_1@news4.newsfeeds.com>
Message-ID: <E14j5RT-0008Bo-00@darjeeling>

On Fri, 30 Mar 2001 12:02:14 -0800, "Dry Ice" <nomail at nomail.com> wrote:

> Know of substantial apps written in Python?

Two big ones come to mind: Mailman (www.list.org) and Zope (www.zope.org)
Oh, and Sketch (sketch.sourceforge.net)
-- 
"I'll be ex-DPL soon anyway so I'm        |LUKE: Is Perl better than Python?
looking for someplace else to grab power."|YODA: No...no... no. Quicker,
   -- Wichert Akkerman (on debian-private)|      easier, more seductive.
For public key, finger moshez at debian.org  |http://www.{python,debian,gnu}.org



From francois.granger at free.fr  Fri Mar  9 16:59:07 2001
From: francois.granger at free.fr (=?ISO-8859-1?Q?Fran=E7ois_Granger?=)
Date: Fri, 09 Mar 2001 21:59:07 GMT
Subject: gah! I hate the new string syntax
References: <mailman.983485923.1136.python-list@python.org> <B7In6.16281$Qb7.2658902@newsb.telia.net> <97ohh2$polt0$1@ID-77331.news.dfncis.de> <97opsb$nhu$1@nntp6.u.washington.edu> <mailman.983615170.23313.python-list@python.org> <tN5o6.2746$eH3.14913@NewsReader> <Koco6.7518$1D5.244602@e420r-atl1.usenetserver.com> <mailman.983997252.5628.python-list@python.org> <m2vgpkyfjo.fsf@mycroft.actrix.gen.nz> <mailman.984063199.18779.python-list@python.org> <98an7i01tgp@news2.newsguy.com>
Message-ID: <1eq0wih.1s0r5163x0lfkN%francois.granger@free.fr>

Alex Martelli <aleaxit at yahoo.com> wrote:

> (Of course, intentions to
> return are not always carried out

There are a lot of french citizen wich are descendant from italian. At
some point of time we were really welcoming of political refugees. We
are not any more, and I am sorry for this.

We also needed a lot of workers for house building, and italian made a
perfect fit between the world wars. Later we "used" spanish, then north
african...

> -- few are aware, for example,
> that such an icon of French popular culture as Yves Montand was
> Italian -- real name Ivo Livi, born of just-as-Italian parents
> in a village near Florence; his father, a socialist, had to flee
> [and ended up in France] when the Fascists took power, and Ivo,
> as well as the rest of the family, followed him a while later;
> I'm not sure when he changed his citizenship over to France).

And lots of french people are really proud of Yves Montant, knowing or
not his origin as Ivo Livi.

-- 
"La connaissance est le chemin de la tol?rance, c'est valable pour 
tous, en toutes saisons."
- Raymond Page


From loewis at informatik.hu-berlin.de  Fri Mar  2 11:10:11 2001
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 02 Mar 2001 17:10:11 +0100
Subject: Editing Python source as XML?
References: <mailman.983546704.10789.python-list@python.org>
Message-ID: <j44rxcjgnw.fsf@informatik.hu-berlin.de>

Steven Haryanto <steven at haryan.to> writes:

> A recent experience of having to modify a long C source file has
> led me to wonder whether there are any tool/effort towards editing
> Python source code as XML or other form of structured document.

Not sure *why* you'd want to do that, but it is certainly possible to
convert any script automatically to an XML document with the help of
the parser module.

> What I would like is an editor that presents the source file as a
> tree outline, where I can collapse selected class or function
> definition body, certain long if's/for's, some section of code,
> view/hide the asserts, etc. 

PythonWin, and the Emacs Python mode, both can do that today.

Regards,
Martin


From shaleh at valinux.com  Thu Mar 22 15:59:02 2001
From: shaleh at valinux.com (Sean 'Shaleh' Perry)
Date: Thu, 22 Mar 2001 12:59:02 -0800 (PST)
Subject: Python and emacs
In-Reply-To: <99dkg711894@news1.newsguy.com>
Message-ID: <XFMail.20010322125902.shaleh@valinux.com>

On 22-Mar-2001 Alex Martelli wrote:
> "Sean 'Shaleh' Perry" <shaleh at valinux.com> wrote in message
> news:mailman.985287273.17896.python-list at python.org...
>>
>> On 22-Mar-2001 Don Dwiggins wrote:
>> > A semi-idle question: has anyone looked into embedding Python as an
>> > extension language in Emacs, alongside Elisp?  I'm toying with the idea
> of
>> > doing some Emacs enhancements/modifications, and given the choice, I'd
> much
>> > rather do them in Python.  (One such enhancement would be to add
>> > Pythonwin-style "intellisense" popups to python-mode.)
>>
>> An editor that spoke python would not be a bad thing.  However, trying to
> get
> 
> Whaddya mean, 'would'?  VIM does that right now, and has for years
> (see www.vim.org)...
> 

rephrase that -- 'something like emacs, but based on python'.  I grew up on vi,
but now I am an adult and have put away my toys (to paraphrase a famous poem).



From aleaxit at yahoo.com  Mon Mar 19 07:19:22 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Mon, 19 Mar 2001 13:19:22 +0100
Subject: Who's minister of propaganda this week?
References: <B98041329020E007.7DA84CAD9D0562AB.817BBC572F480869@lp.airnews.net> <98nu1m$12e@dispatch.concentric.net> <660ACCA1E4816F7E.C359CDF4168EF3FB.3B6B05538D619770@lp.airnews.net> <3AAFA13D.76E32C3F@divalsim.it> <98oa4o01i4v@news2.newsguy.com> <mailman.984619039.7856.python-list@python.org> <98q9h60122v@news2.newsguy.com> <mailman.984937946.5318.python-list@python.org> <9937en0bge@news2.newsguy.com> <slrn9bagbm.toq.qrczak@qrnik.zagroda> <994ab801h5o@news2.newsguy.com> <slrn9bbin2.11r.qrczak@qrnik.zagroda>
Message-ID: <994tg9026qa@news1.newsguy.com>

"Marcin 'Qrczak' Kowalczyk" <qrczak at knm.org.pl> wrote in message
news:slrn9bbin2.11r.qrczak at qrnik.zagroda...
    [snip]
> Design patterns designed for C++ and Java are not applicable to
> languages which use algebraic types and don't use subtyping. C++
> and Java don't have a convenient way to express "either this or that
> or these" (wrapping each case in a class is not convenient enough).
> OTOH Haskell and OCaml don't work by implicit coercion of elements
> to a common supertype.

Isn't that what the "O" in O'Caml adds to plain good old Caml?  (And
the "O" in O'Haskell too, but you're not addressing that one).


> Note that in these languages it's meaningless to mix all types of
> objects together in a container. If they are unrelated, you can't
> apply any operation to an unspecified member of them.

Not sure if by "these" you mean, here, "the former" (C++, Java, ...)
or "the latter" (ML, Haskell, ...).  In Java, objects can never be
truly unrelated, as they all derive from a common ancestor (Object);
thus, you can always hold anything through a reference to Object, and
try casting (which will fail with a predictable, catchable exception
if need be) to obtain usable interfaces.  The exact C++ translation
is not standard, as dynamic_cast<> with a void* argument need not be
implemented by all standard C++ compilers, so, you _do_ need to
explicitly provide some 'remote relation' on which dynamic_cast<>
can be portably applied (that 'relation' need not DO anything, just
as the interface used in 'Acyclic Visitor' doesn't -- as long as it
has SOME virtual, typically a virtual destructor, dynamic_cast<>
will work).


> Tell what do you really want to archieve. The whole thing looks like
> a poor fit to the style of these languages. You don't want to have a
> net of containers of everything. For example if they are registered

I'm not going to retype the whole relevant section of "Design
Patterns" -- please look it up for yourself.  Again, there is
pronominal ambiguity in the "these".  Assuming now that by
'these' you mean 'Haskell' (or ML or Caml without the 'O'),
it's not true that I don't _want_ to built heterogeneous,
structured containers -- it's that I _can't_ do it conveniently,
because of the language limitations imposed by the desire to
allow static type-checking.

> to let somebody do something with them, you would rather register
> "do something with them" actions instead of objects themselves.
> Objects are rarely mutable so even this looks very suspicious.

Objects are rarely mutable _in functional programming_ -- they
are commonly mutable _in object oriented programming_, and this
makes a very substantial difference.

Python allows and strongly encourages object-oriented programming,
as do other languages which implement compile-time static type
checking.  Haskell (as opposed to O'Haskell) doesn't; O'Caml, to
the extent that it does, also introduces ref/mutability and
other complexities.

Debating the need/opportunity for static/dynamic typing in
languages that focus on FP, not O-O, may no doubt be quite
interesting (I keep thinking that Erlang's real-world success,
and the huge boost it gives to programmer productivity, may
have something to do with its being dynamically typed albeit
in an FP context) but has precious little to do with this
thread, or with any language lacking such an FP focus.


Alex





From aleaxit at yahoo.com  Mon Mar 26 05:52:52 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Mon, 26 Mar 2001 12:52:52 +0200
Subject: static methods
References: <mailman.985586915.11526.python-list@python.org>
Message-ID: <99n72706g0@news2.newsguy.com>

"Christian Tanzer" <tanzer at swing.co.at> wrote in message
news:mailman.985586915.11526.python-list at python.org...
    [snip]
> > Otherwise, you have to pass both the class reference and the module
> > reference to the caller.
>
> Actually, the module reference will suffice; it will contain suitably
> named callables to:
>     -- generate instances of the class (the class-object itself and/or
>         suitable factory functions),
>     -- do whatever else you need to be done class-wise.
>
> You can package up those callables any way you want; there is
> really no special reason they should be attributes of the class
> object as opposed to that of any other object, and a module
> object is typically handiest.

: This seems to be the standard answer on c.l.py. I'm not convinced.
: There are two cases:
: - Modules defining a single class.
:  In this case, most people would use the same name for both module
:  and class. If you need the module object for scoping purposes, you
:  end up with code using lots of `Bletch.Bletch' which I find atrocious.

You don't often need to obtain *the class-object* from your caller --
it's generally only needed when you're defining a class of yours
*that _inherits_ from a caller-supplied class-object*; a pretty rare,
although undoubtedly elegant, highly-dynamic idiom.  For those pretty
rare needs, _when you also further need a few other functions that
are not-instance specific_ (also caller-supplied ones), I find that
just about any way to package things up works -- sure, you and your
caller may need to agree on attribute-names or whatnot, unless you
want to pass several objects, but then, you'd anyway need to agree on
the N names of the various non-instance-specific functions' names, and
one more agreed-upon attribute name (for the class-object itself), for
a total of N+1 instead of N, is hardly a hardship such as to justify
language-changes!

More often, what you need to get from your caller is just *a bunch
of callables* -- one or more to generate objects of the appropriate
class[es], zero or more to operate 'generically' (not as methods of
some instance).  Here, the class object _might_ be one of the
callables in the first subset (factory-callables), but there is
really no advantage in singling it out for special treatment as
against other factory-callables, and why should the *other* (if
any) generic functions be preferably packaged up as attributes
of one of the factory-callables?!  It seems a peculiar arrangement
indeed.  I see nothing wrong with needing a trivial wrapper to
setup this packaging arrangement if you're really keen on it.
Note that, in any case, you'll STILL need naming-conventions for
(at best) all but one of the callables (unless you want to pass
them as separate objects, of course).

The handiest packaging helper is the usual suspect:

class ABunchOfStuff:
    def __init__(self, **kwds):
        self.__dict__.merge(kwds)

whenever you need to package up, and pass as a function
argument, a bunch of stuff with some set of agreed names,
you do so by passing a suitably initialized ABunchOfStuff
instance.  There are no restrictions to the kinds of
callables (or other objects) that can live as attributes
of this bunch-o'-stuff, nor on their names.


:  Rick also talked about passing around a reference to the class. If
:  you pass a reference to a module instead, you need a convention how
:  to refer to the class defined by the module.

If you don't need other callables, but just one factory callable,
it's fine to pass just that callable object, of course -- it may,
as a special case, be the class-object, just as of-course.  When
you DO need to pass N+M callables (M of them being factories, the
other N non-factory, non-instance-directed generics) you do need
to pass them separately, or to have some naming convention if you
want them to pass them as attributes of some object, or other kind
of similar conventions (e.g. you could pass one or two tuples or
other sequences, but then you'll need conventions on the order in
which various callables occupy item slots in those sequences --
quite analogous to a naming-convention:-).

There's nothing intrinsically superior about the packaging
convention of passing one object and having the other N+M-1
be attributes of that distinguished object (and that still
needs a naming convention anyway).  You can get that sort of
arrangement if you want, but I don't see why a language should
go out of its way to make it handier for you wrt other kinds
of arrangements.

: Modules defining multiple classes. In this case, passing a module
: reference instead of a class reference won't fly, anyway.

It's still an issue of how you arrange/package things up,
naming (or other organizational) conventions, etc.  If I
have multiple similar classes in a module, I'm _particularly_
likely to want to pass from it, as the factory callable,
something OTHER than one of those class-objects -- a
factory function that will choose _which_ exact class to
instantiate depending on arguments, environment, etc.  If
the other generic functions need to know what specific _class_
has been instantiated, i.e. they're not really as generic
as all that, then having them as attributes of the actual
_instance_ (not necessarily the _class_ thereof) may work.

The module-object may be the appropriate object to pass if
you rely on naming conventions for the factory function[s]
and semi-generics.  Alternatively, repackaging whatever
appropriate subset of callables into a BunchOStuff instance
works just fine; as do other somewhat more complex approaches,
but I tend to prefer simplicity whenever feasible.

: You could dismiss modules with multiple classes out of hand, but
: IMNSHO this would be rash (and inconsistent with your approval of
: Lakos' ideas about large scale design).

Why should modules with multiple classes be problematic
here?  A given factory-function may instantiate the
proper class, whether the candidate classes are all in
one module, or spread among several modules, just as
easily.

> The desire to use the class-object specifically, rather than any
> other, typically seems to have more to do with familiarity with
> other languages where classes must play certain special roles
> (C++, Java, Smalltalk, Eiffel) than with any actual programming
> needs.

: I used languages with strong emphasis on the module as primary
: construct (like Modula-2 and Ada) and languages favoring the
: class as the primary construct (like Eiffel and C++). I found the
: latter much better suited to my programming needs. That doesn't mean

But does Modula-2 (or the Ada version you used) offer classes
*at all*?  Modules per se don't offer enough _polymorphism_ (if
any), which is a rather different issue; classes (or at-least
equivalent constructs, such as 'prototype objects' in languages
oriented to their use) are a real need because of that, while
not necessarily usurping modules' role for subsystem packaging.

: that I shun modules, though. Fortunately, Python doesn't force either
: direction (with some tricks to get class functions).

The well-known tricks are one possibility, sure, but different
organizations of the needed callables often shine brighter.


Alex






From hzhu at users.sourceforge.net  Wed Mar 21 16:37:16 2001
From: hzhu at users.sourceforge.net (Huaiyu Zhu)
Date: Wed, 21 Mar 2001 21:37:16 GMT
Subject: New PEP: The directive statement
References: <mailman.985085006.24236.python-list@python.org>
Message-ID: <slrn9bi7q3.2sa.hzhu@rocket.knowledgetrack.com>

With the __future__ PEP, I envision there could also be things like

from __future__ import nested_scope
from __experimentatal__ import stackless
from __optional__ import static_type
from __mandatory__ import string_methods

New feature would migrate from __experimental__, to __future__, then to
either __optional__ or __mandatory__.  Things in __experimental__ are not
guaranteed to be there in later versions.  Things in __future__ will
definitely be there in a later version.  The __mandatory__ could be used to
make the intention about version requirement more explicit than __version__
would allow. (Someone asked about whether __future__ should be phased out in
later code.) The __optional__ could be used for permament optional features
that affect syntax.

So with this directive PEP, the parallel would be

directive transitional nested_scope
directive experimentatal stackless
directive optional static_type
directive mandatory string_methods

Is this correspondence correct?  (You may not agree on the need for other
directives, but I want to know if the keyword translations are correct.)

If so, I see two problems.
1. It introduces a new syntax while an existing syntax could be reused
   without much cost.
2. There is no place for version and documentation info.

So far as I can see, the impetus for directive keyword is that import is a
run time action, while the __future__ is a compile time declaration.  But is
it so far a stretch?  All Python __magic__ words go beyond normal meaning.
Defining __setattr__ will affect more than this method alone.  Having
__init__.py will cause other modules to be imported.  And so on.

If we look more broadly, in *nix, devices, processes, pipe and others are
also represented as files.  Would it better to assign special syntax to all
of them?

IMHO, reusing existing syntax is closer to Python than Perl, which
introduces new syntax for many things that are only slightly different.

Huaiyu


From klxsyi at e.eliminate.com  Tue Mar 27 07:48:30 2001
From: klxsyi at e.eliminate.com (klxsyi at e.eliminate.com)
Date: Tue, 27 Mar 2001 12:48:30 GMT
Subject: Warning.............please read....................................................................                  .  1857
Message-ID: <y_%v6.148540$KL5.127784@NewsReader>

Do you want your family or the authorities to find out what sites you surfed?

When you access the Internet, your computer keeps permanent hidden records of your activities!

Files I shredded months earlier were clearly visible using this software.
I advise you to check it out...........
http://www.evidence-eliminator.com/go.shtml?A653704
qkjbtbodsvrsknpcwolfcrezvxyeewlrtbndsnbvscdtbhjwhwrqbrccykvnyuui



From bob_cannard at mentor.com  Wed Mar 21 12:37:35 2001
From: bob_cannard at mentor.com (Bob Cannard)
Date: Wed, 21 Mar 2001 09:37:35 -0800
Subject: re.match
References: <997skf$qoc$1@bmerhc5e.ca.nortel.com>
Message-ID: <3AB8E6DF.A2C32AD8@mentor.com>


"Michael P. Soulier" wrote:
> 
>     Hey people. Could someone explain the purpose of the match function in the
> re module? If it's only difference from search is to match at the beginning of
> the line, why does it exist? That's what the ^ in the regexp is for.

I'd like to add my own recent experience as a Python newbie getting to
grips with its implementation of regexps. Once upon a time there was
a regular expression which matched a - well, let's call it a Thing.
There were 3 questions that needed to be asked:

   - Does this string contain any Things? (re.search, obviously)

   - Find all the Things in this string. (re.findall)

   - Is this string a Thing? (re.match - er, no, that doesn't work,
     it only asks if this string begins with a Thing.)

Life would have been perfect if re.match, or some similar function,
required a match of the entire string against the regular expression.
But it doesn't, and there doesn't seem to be a function that does,
so the regular expression spawned another with a dollar sign on the
end. (Curious assymetry.) Alternatively the regexp could have been
left uncompiled so it could be manipulated, but that would have
caused a loss of speed and generality in my lower-level functions.
Another possibility would have been to combine compiled regular
expressions, but there doesn't seem to be a way to do this. Does
anyone know of a function or piece of magic that I've missed?

Anyway, my point is that there does seem to be a case, from my
experience, for having different functions that do different things
with the same regular expression. If anything I'd like to see more
of them.

Cheers,

       Bob.


From mwh21 at cam.ac.uk  Fri Mar  2 18:51:36 2001
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 02 Mar 2001 23:51:36 +0000
Subject: python-dev summary 2001-02-15 - 2001-03-01
References: <mailman.983559723.29891.python-list@python.org>
Message-ID: <m3d7bzd913.fsf@atrus.jesus.cam.ac.uk>

"Tim Peters" <tim.one at home.com> writes:

> [Michael Hudson]
> >  ...
> >  The iteraators discussion seems to have disappeared.  At least, your
> >  author can't find it!
> 
> Greg Wilson set up an egroups (now Yahoo) discussion group for it, where
> debate has continued:
> 
>     http://groups.yahoo.com/group/python-iter
> 
> Marc-Andre Lemburg set up another (or agitated toward this end) mailing list
> for it on SourceForge, for reasons that escape me at the moment.  I never got
> around to subscribing to that too; the link on SF:
> 
>     http://www.geocrawler.com/redir-sf.php3?list=python-iterators
> 
> appears broken, and the archive link:
> 
>     http://lists.sourceforge.net/archives//python-iterators/
> 
> leads to a page that doesn't make sense.

Ah, I'm subscribed to the sf one, which seems to be silent.  I shall
have to check out the Yahho!(bletch)groups one.

Cheers,
M.

-- 
  It's actually a corruption of "starling".  They used to be carried.
  Since they weighed a full pound (hence the name), they had to be
  carried by two starlings in tandem, with a line between them.
                 -- Alan J Rosenthal explains "Pounds Sterling" on asr


From ullrich at math.okstate.edu  Sat Mar  3 13:07:35 2001
From: ullrich at math.okstate.edu (David C. Ullrich)
Date: Sat, 03 Mar 2001 18:07:35 GMT
Subject: Hitting two targets:  OO + group theory (pedagogy)
References: <u2k0atkgkiatdp9ukj3mpd1ej0b648ldsm@4ax.com> <3aa1077a.1110803@nntp.sprynet.com> <0932atgt9v8d8aboq90evoi5ipj5c0sk8s@4ax.com> <3aa11a91.5998558@nntp.sprynet.com> <ko72atgf5hmg66suefbenp8lvn6vhblajv@4ax.com>
Message-ID: <3aa13152.173635120@news.okstate.edu>

On Sat, 03 Mar 2001 09:16:14 -0800, Kirby Urner
<urner at alumni.princeton.edu> wrote:

>ullrich at math.okstate.edu (David C. Ullrich) wrote:
>
>>Some people would say that the reference to Genesis
>>is out of place here... probably you know that some
>>people would say that. Might be regarded as just
>>generally inappropriate - some other people might
>>think that the bit about Abel is misleading, since
>>"Abelian" _does_ refer to a man named Abel, but
>>not the one in Genesis.
>
>True, but it's a useful mnemonic connected to a 
>well-known story.  Making allusions to Biblical 
>characters need be no more or less churchy than 
>alluding to Greek mythology or to cartoon characters
>like Bugs Bunny (or Coyote in Native American lore)
>-- just taking advantage of some shared cultural 
>heritage.  

Right. I didn't say it was inappropriate. Although in fact
something about the bit about "If you've studied Genesis"
or however you put it seems a little off-putting; if
you were referring to Bugs Bunny or Zeus you'd
just refer to them.

Never mind - could be I've been in Okllahoma
too long.

>Anyway, I learned this mnemonic from a very strait-
>laced group theory book (which I've since returned 
>to the library), plus saw it mentioned on another 
>list frequented by math heads -- sort of assumed 
>this clever CAIN/Abel thing was cited routinely 
>by those more familiar with the literature.  Agreed 
>though: we should make clear who the Abel in Abelian 
>was.
>
>>But why not start with a class Group instead
>>of special cases? (Would be a class with a
>>__mul__ method that does nothing but raise
>>an exception "must implement __mul__ in
>>a subclass", an Id attribute, etc. I can't think
>>of anything offhand that Group would actually
>>_do_, but if we're bent on this OO-math thing
>>it seems like groups _should_ be descendants
>>of a class Group...)
>>
>
>In my example, it's the elements of a group which
>get instantiated as separate objects, with each 
>element inheriting the same meaning for *, whatever
>that might be. 

Right. That's actually what I meant, a class Group
instances of which would represent elements
of a generic group - of course there's no such
thing so you'd never create instances of Group.
Then subclasses of Group represent actual
groups; instances of the subclasses being
elements of actual groups.

Not important, just seems to me like what it
"should" be.



From Laura_01 at MailAndNews.com  Wed Mar 14 20:11:09 2001
From: Laura_01 at MailAndNews.com (Laura Lewin)
Date: Wed, 14 Mar 2001 20:11:09 -0500
Subject: Magnitude of the wx* market (was: Python In A Nutshell - suggestions)
Message-ID: <3AB26DAC@MailAndNews.com>

Hi,
Sure, we'll take this suggestion into consideration (on expanding wxPython 
coverage) when we publish the next edition of Python Programming on Win32.
Thanks!
Laura
LLewin at oreilly.com

>===== Original Message From tim at digitig.cix.co.uk =====
>In article
><A909AD29CBD32AFC.4F1641C135749F87.7149617C7C166896 at lp.airnews.net>,
>claird at starbase.neosoft.com (Cameron Laird) wrote:
>
>> In article <3AAA4FE3.B471B689 at northwestern.edu>,
>> Louis Luangkesorn  <lluang at northwestern.edu> wrote:
>> 			.
>> 			.
>> 			.
>> >And while I'm at it, one useful book to have would be on
>> wxWindows/wxPython.  I see references to it in several places, and a
>> few intros on the web, but
>> 			.
>> 			.
>> 			.
>> This is an interesting point.  It's typical in computer
>> book publishing that publishers want more than authors
>> can produce; roughly, someone like ORA would happily
>> print many more titles than they can find authors willing
>> and able to write.
>>
>> I happen to know, though, that several publishers have
>> rejected wxWindows proposals.
>>
>> I doubt, though, that there'd be much trouble making the
>> case that the subject deserves a chapter in a larger work.
>
>And the larger work in question is "Python Programming on Win32" by Mark
>Hammond & Andy Robinson, already published by ORA, so I doubt they'd be
>particularly interested in another! It doesn't have full documentation of
>the wxPython classes, though, just a tutorial introduction. Maybe the way
>to go is to get them to extend it in the next edition?



From kens at sightreader.com  Wed Mar 21 18:57:01 2001
From: kens at sightreader.com (Ken Seehof)
Date: Wed, 21 Mar 2001 15:57:01 -0800
Subject: Dynamically reloading extension modules
References: <007f01c0b252$852bb370$04090a0a@upcast.com> <20010321145901.M24341@trufflehunter.avalongroup.net> <00e201c0b25f$b95d4a70$04090a0a@upcast.com> <20010321154600.Q24341@trufflehunter.avalongroup.net>
Message-ID: <012301c0b262$9fab59c0$04090a0a@upcast.com>

> On Wed, Mar 21, 2001 at 03:36:15PM -0800, Ken Seehof wrote:
> > Not exactly.  I didn't state the problem very clearly.  The
> > problem is that while an extension module is loaded, the
> > module file (pyd on windows) is read-only.
> > 
> > At least this is true on windows.  On unix, I think maybe you
> > can rename the module file, rebuild it, and do a reload(module).
> > Is this true?  Windows doesn't allow you to rename or rewrite
> > a dll while it is in use, so something like FreeLibrary would be
> > necessary.
> 
> Sorry, I can't answer Windows questions. I haven't had to run
> Windows in years.
> 
> But I would still think that reload() should work as long as
> you are using python modules, *not* c/c++ modules.

I -am- talking about c/c++ modules (which I have been referring
to as "extension" modules).  Hence there's a bit more to it than
reload().

I want to recompile and reload a c/c++ extension module from
within a continuously running python program.

> Beazley has no disclaimers about reload working only on
> specific platforms.
> 
> There is a caveat, objects created with the old module won't be
> updated.





From Marc.Poinot at onera.fr  Mon Mar 12 10:51:34 2001
From: Marc.Poinot at onera.fr (Marc Poinot)
Date: Mon, 12 Mar 2001 16:51:34 +0100
Subject: [ANNOUNCE] European Python Meeting
Message-ID: <3AACF086.31D78DBC@onera.fr>

*** Another Python Meeting... in Europe ;) ***

------------------------------------------------------------
Hallo Python-Freunde,

zwei Jahre nach JPF001 findet erstmals eine europaeische
Python-Konferenz statt, im Rahmen des Treffens "Freie Software"
in Bordeaux (Frankreich). Mehr Informationen gibt es bei

   http://lsm.abul.org

Die Python-Konferenz hat die Nummer 19.

------------------------------------------------------------
Hi you all Pythoneers.

Two years after JPF001, I'm proud to announce
the first attempt to organise an European Python event

This event will take place into the Libre Software Meeting
in Bordeaux, France. For more information see:

  http://lsm.abul.org

The Python topic is number 19.

------------------------------------------------------------
Saluton, amikoj de Pitono,

du jarojn post JPF001, okazos la unua euxropa kongreso pri Pitono,
kadre de la Renkontigxo pri Libera Programaro en Bordozo (Francio).
Pli da informoj (en la angla) haveblas cxe

   http://lsm.abul.org

La sesio pri Pitono havas la numeron 19.

------------------------------
Salut a toi, ami(e) de Python

Deux ans apres JPF001, une nouvelle conference Python
vous est proposee dans le cadre des Journees du Logiciel Libre
de Bordeaux:

 http://lsm.abul.org

La session Python est la 19.

Ceci *est* l'occasion pour se rencontrer, Pythoniens(niennes) 
de France et d'Europe. 

Pour le groupe de traduction en Francais, 
des ateliers seront mis en place, sous la houlette hop-la youpi
la-houlette de Benoit Lacherez...


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

Marcvs [alias Translations by Konrad Hinsen ;]



From jeremy at alum.mit.edu  Wed Mar 28 12:06:17 2001
From: jeremy at alum.mit.edu (Jeremy Hylton)
Date: Wed, 28 Mar 2001 12:06:17 -0500 (EST)
Subject: StandaloneZODB at cvs.zope.org
In-Reply-To: <B1ix6IA+Thw6Ewia@jessikat.demon.co.uk>
References: <mailman.985746941.27962.python-list@python.org>
	<B1ix6IA+Thw6Ewia@jessikat.demon.co.uk>
Message-ID: <15042.6665.731184.238118@w221.z064000254.bwi-md.dsl.cnc.net>

>>>>> "RB" == Robin Becker <robin at jessikat.fsnet.co.uk> writes:

  RB> ...  so is there some advantage in maintaining two projects?  --

That remains to be seen.  The administration of the project is
different, but the end goal is the same.  In the long term, I imagine
we'd have a single project, but it's hard to be sure.

The administrative different is that StandaloneZODB comes out of our
CVS repository (cvs.zope.org) will always have the latest changes made
by Zope developers at Digital Creations.  The project at SF needs to
be manually updated by someone.

Jeremy





From eferro at inicia.es  Mon Mar 19 09:26:14 2001
From: eferro at inicia.es (Eduardo Ferro)
Date: Mon, 19 Mar 2001 15:26:14 +0100
Subject: I lost characters at read
Message-ID: <9955b4$i7b$1@diana.bcn.ttd.net>

Hi!

I am trying to create a window to log the contents of a file, to do this
i first read all the file and later i periodically read the file if it
grow up

The problem is that when my periodic callback read from file i lost one
or two characters per line and I don`t  understand why
Me code is (the problem is at the checkFichero fuction):

class LogWindow:
	'''Clase para mostrar a modo de log lo leido de un fichero de texto'''

	def __init__(self,gtkContainer,fich,maxLineas):
		self.maxLineas = maxLineas
		self.numLineas = 0
		self.intervalo = 1000
		self.fich = fich
		self.gtkScrolledWindow = GtkScrolledWindow()
		gtkContainer.add(self.gtkScrolledWindow)
		self.gtkText = GtkText()
		self.gtkScrolledWindow.add_with_viewport(self.gtkText)
		self.gtkScrolledWindow.set_policy(POLICY_AUTOMATIC, POLICY_AUTOMATIC)
		
		self.gtkScrolledWindow.show_all()
		self.lstLineas = []
		self.timer = timeout_add(self.intervalo,self.checkFichero)

	def checkFichero(self):
		self.gtkText.freeze()
		linea = self.fich.read()
		while linea != '':
			print '"%s"', % linea
			self.lstLineas.append(linea)
			self.gtkText.insert_defaults(linea)
			linea = self.fich.read()

		self.gtkText.thaw()
		print 'checkFichero'
		return 'true'
		

#-------------------------------------
def main():
	gtkWindowMain = GtkWindow(type='toplevel')
	gtkWindowMain.connect('destroy',mainquit)
	fich = open('menu.txt','r')
	logWin = LogWindow(gtkWindowMain,fich,50)
	gtkWindowMain.show_all()
	mainloop()
#-------------------------------------

if __name__ == '__main__':
	main()


From ubmurzyn at cyf-kr.edu.pl  Tue Mar 27 06:29:10 2001
From: ubmurzyn at cyf-kr.edu.pl (Krzysztof Murzyn)
Date: Tue, 27 Mar 2001 13:29:10 +0200
Subject: Building Python with Alpha Compaq compiler
Message-ID: <3AC07985.2938D3AA@cyf-kr.edu.pl>

I was wondering whether anyone managed to compile Python with Compaq
compiler on Alpha running Linux RedHat. Any hints will be appreciated...

Thanks,
Krzysztof





From fredrik at pythonware.com  Mon Mar 12 18:12:50 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Mon, 12 Mar 2001 23:12:50 GMT
Subject: tkinter entry index
References: <iHcr6.27192$tV.5424761@newsrump.sjc.telocity.net>
Message-ID: <SJcr6.17936$Qb7.3120517@newsb.telia.net>

"dsavitsk" wrote:
> how can i get the position of the cursor in an entry widget? .index and
> .icursor don't seem to it. That is, if "foo" is in an entry, and the cursor
> is between the "o" and the other "o", I could get a return of 2.

    position = w.index(INSERT)

Cheers /F

<!-- pythonworks 1.2 coming soon:
http://www.pythonware.com/products/works/index.htm
-->




From emile at fenx.com  Thu Mar 22 08:54:55 2001
From: emile at fenx.com (Emile van Sebille)
Date: Thu, 22 Mar 2001 05:54:55 -0800
Subject: tempfile problem on win98
References: <g3nu6.369$hb.2141@news1.belrs1.nsw.optushome.com.au>
Message-ID: <99d0dg$os92$1@ID-11957.news.dfncis.de>

The mktemp routine guarantees to return a name unique in the target
directory.  Are you sure you're using fully qualified path names throughout
when subsequently discovering duplicate files?

Emile van Sebille
emile at fenx.com

"Steven Adams" <quickdry at users.sourceforge.net> wrote in message
news:g3nu6.369$hb.2141 at news1.belrs1.nsw.optushome.com.au...
> Hi,
>
> I'm using the tempfile module in Python 1.5.2 and it keeps giving me
> duplicate filenames.
>
> It's being called only once each time the script runs, (small cgi
> messageboard) with this is the code related to the tempfile:
>
>  import tempfile
>  tempfile.tempdir=SOME_DIRECTORY
>  tempfile.template=None
>  id=tempfile.mktemp()
>
> shouldn't this give me a unique value for id every time? I initially
> had it running without the tempfile.template being assigned None, but
> when I noticed the duplicate filenames, so I thought I'd try that.
>
> Currently I'm making it recompute tempfile names until it has one that
> does not already exist, but why does this happen?once I get a lot of
> messages in the system I don't want to be possibly iterating through
> the whole lot checking their names (unfortunately I don't have
> permissions or the space to use a proper database)
>
> thanks for any help,
>
> Steven
>
>




From s.sharp at fh-aargau.ch  Fri Mar 23 10:21:02 2001
From: s.sharp at fh-aargau.ch (Sharp)
Date: Fri, 23 Mar 2001 16:21:02 +0100
Subject: Install Python on Apache
Message-ID: <99fph5$a2n$1@rex.ip-plus.net>

Hi Everybody

I sucked down Python from python.org and started to install the program
on a win2000 apache system. that has worked perfectly. but then I tried
to set up the apache httpd.conf file with the following options

AddType application/x-httpd-py .py
Action application/x-httpd-py /cgi-bin/Python/python.exe

(cgi.bin is set to the right place because php and perl do work)

But then it did not work. I hava also tried several variations but always
with the same comands.
The Error Message in the Apache error.log was

[Fri Mar 23 15:54:50 2001] [error] [client xxx.xxx.xxx.xxx] Premature end of
script headers: c:/web/cgi-bin/python/python.exe

the script is looking like this

#!c:/web/cgi-bin/Python/python.exe
print "<html>";
print " <head>";
print " <title>Test Page for Apache Test Suite</title>";
print " </head>";
print " <body>";
print " <h1>It Worked!</h1>";
print " </body>";
print "</html>";

have you got me a solution? that would be great.

yours
Sharp




From glenfant at equod.com.nospam  Wed Mar  7 06:30:09 2001
From: glenfant at equod.com.nospam (Gillou)
Date: Wed, 7 Mar 2001 12:30:09 +0100
Subject: Remote script execution with Python ...
References: <7Ybp6.7729$r4.7336@news.indigo.ie> <ulmqil8qu.fsf@roundpoint.com>
Message-ID: <9855om$aa5$1@reader1.imaginet.fr>

rsh is in cygwin distribution (www.cygwin.com) so... this script should work
with any python version on any Win32 system if the client script is launched
from cygwin bash.
"Ben Hutchings" <ben.hutchings at roundpoint.com> a ?crit dans le message news:
ulmqil8qu.fsf at roundpoint.com...
> "Robert Gahan" <gahanr at gofree.indigo.ie> writes:
>
> > Hi there,
> >
> > I have installed python on a Windows PC and I am looking to
> > remotely execute a script on a UNIX box from a python program.
> > I need to be able to pipe a "y" to stdin also. Is there a windows
> > equivalent of the unix "rexec" that could help me ?. What is the
> > most appropriate way of doing this ?.
>
> You might be able to do this with something like:
>
>     script_in, script_out = os.popen2("rsh otherhost -c myscript")
>     print >>script_in 'y'
>     result = script_out.readlines()
>
> Windows 2000 includes an rsh client but I'm not sure about other
> versions of Windows.  You might want to use ssh instead for greater
> security.
>
> --
> Any opinions expressed are my own and not necessarily those of Roundpoint.



From erwin at andreasen.com  Sun Mar 25 12:04:36 2001
From: erwin at andreasen.com (Erwin S. Andreasen)
Date: Sun, 25 Mar 2001 17:04:36 +0000 (UTC)
Subject: close(), exceptions and problems
References: <slrn9bm5u5.klj.kuntsi@myntti.helsinki.fi> <99i3j7$lqr$1@news.cybercity.dk> <slrn9bpiej.a63.qrczak@qrnik.zagroda>
Message-ID: <99l8f4$1fob$1@news.cybercity.dk>

On 24 Mar 2001 16:22:43 GMT, Marcin 'Qrczak' Kowalczyk wrote:
>Sat, 24 Mar 2001 12:23:03 +0000 (UTC), Erwin S. Andreasen <erwin at andreasen.com> pisze:
>
>> But if you call close(fd) you aren't supposed to do something like:
>> 
>> int fd, rv;
>> while (((rv = close(fd)) < 0 && errno == EINTR))
>> 	;
>> if (rv < 0)
>> 	perror("close");
>
>Linux documentation says otherwise.

That's GNU libc documentation actually.

>    `EINTR'
>          The `close' call was interrupted by a signal.  *Note
>          Interrupted Primitives::.  Here is an example of how to
>          handle `EINTR' properly:
>
>               TEMP_FAILURE_RETRY (close (desc));

But this only happens when SA_RESTART is not set (OK, I thought it was
SA_NORESTART to *disable* restarting but it's SA_RESTART to enable
restarting).

Read the "Interrupted Primitives" node for more information.

Maybe the threading library on the platform the original poster uses does not
set the SA_RESTART flag (if its threads are userspace, switched by receiving
SIUGUSR1) ? But that's hard to believe.


-- 
=======================================================================
<erwin at andreasen.com>           Herlev, Denmark       Software Designer
<URL:http://www.andreasen.org/>       <*>         LASAT^WEicon Networks
=======================================================================


From nomail at nomail.com  Fri Mar 30 16:55:04 2001
From: nomail at nomail.com (Dry Ice)
Date: Fri, 30 Mar 2001 13:55:04 -0800
Subject: Know of Substantial Apps Written in Python?
References: <3ac4e4a8_1@news4.newsfeeds.com> <KE6x6.415$SR6.235983@typhoon1.ba-dsg.net>
Message-ID: <3ac4ff12$1_2@news4.newsfeeds.com>


> > Know of substantial apps written in Python?


[...]


> What's the definition of "substantial"?

Good question.  To sharpen the point of my inquiry:

As programs grow in size and complexity, both
programmers and the results begin to show fatigue.
Slick as Java is, for example, I have found the
real-world results of which I'm aware so far to
be troubling in the Windows environment.  Still
evaluating, but it does not look good.

What "substantial" is might be defined in terms
of project length.  Perhaps something which would
take a good programmer several months.

Something as complex as a web browser might do,
if the language in question does not merely act as
glue for modules written in C or whatever.

In the end, this is like evaluating people.  We can
listen to what others say about a person, or what a
person says about his or her self- or we can simply
see how they BEHAVE.


> There's Grail, a web browser
> http://grail.sourceforge.net
>
> There's ZOPE, an internet publishing thingy
> (thingy is a technical term  :)
>
> Check www.python.org - there are plenty of examples
> of programs with thousands and thousands of lines of
> code.  I'm not sure if you mean "large" when you
> say substantial or if you mean well known and widely
> used...

Yes, those are what I mean.  :-)





From digitig at cix.co.uk  Wed Mar  7 17:24:00 2001
From: digitig at cix.co.uk (Tim Rowe)
Date: Wed, 7 Mar 2001 22:24 +0000 (GMT Standard Time)
Subject: Can Python replace Visual Basic? Should it?
References: <5DZo6.11126$hn5.1578410@news1.rdc1.mb.home.com>
Message-ID: <memo.20010307222426.51645A@digitig.compulink.co.uk>

In article <5DZo6.11126$hn5.1578410 at news1.rdc1.mb.home.com>, 
bbollenbach at homenospam.com (Brad Bollenbach) wrote:

> Okay, sure this has been asked again and again, but I've never really 
> seen a
> clear answer to this question:
> 
>     Can Python replace Visual Basic? Should it?

I've seen a yes/no and a yes/yes answer so far to this. I'll add a 
no/maybe answer. That "no" is a /technical/ yes, but Python is never 
likely to have Microsoft's marketing muscle behind it (I hope!) and so 
will never replace Microsoft's language of choice. I work in a corporate 
environment that mandates VB (& VBA) for everything because it mandates 
Microsoft for everything and the people who do the mandating aren't 
interested in the technical issues -- we deliver using VB, so why should 
they have the maintenance overhead of multiple languages (even multiple 
MS languages)? Python isn't going to overcome that sort of corporate 
mindset.

But when I'm given a choice, I now use Python for everything. There is 
stuff I wouldn't use it for (safety critical, for one) but I don't do 
that stuff. I keep C++, Delphi, Eiffel and Ada in a back pocket; I have 
Perl, Java, Fortran, Prolog and even Forth on a mental back-shelf, and I 
would pull any of them out if needed. But at the moment I don't need them. 
I don't know if that's because of the nature of what I do in my own time 
(mainly scientific and data manipulation, no serious GUI stuff) or if it's 
across the board advantages of the language. I suspect the former; I'm 
reminded of Michael Jackson's comment that if something helps you with 
every possible problem then it's not going to help you very /much/ on any 
/particular/ problem. I'm going to try to keep a varied set of programming 
tools clean and sharp.

So that's another reply in vague generalities that's not going to be much 
use to you :-)




From tdelaney at avaya.com  Tue Mar 20 22:06:07 2001
From: tdelaney at avaya.com (Delaney, Timothy)
Date: Wed, 21 Mar 2001 14:06:07 +1100
Subject: Yet Another PEP:  Query Protocol Interface or __query__ 
Message-ID: <B43D149A9AB2D411971300B0D03D7E8B0FBFBA@natasha.auslabs.avaya.com>

So far as I can see, this PEP does not solve one basic problem - interfaces
which the class writer doesn't know about.

For example, I have an interface, mapped to "org.mydomain.something". Then I
have a slightly different interface mapped to "org.mydomain.somethingelse".

I have a class which implements org.mydomain.something. However, whilst it
can perfectly implement org.mydomain.somethingelse, it would be rejected as
it doesn't actually know about it.

I think it would be better if an interface was actually a class instance
which specified exactly what it required. This could not only be methods (or
other attributes), but also the (minimum) parameters required (and possibly
their types) and even any pre- or post- conditions required by the
interface. It could also mark "optional" features which if available will be
used. FOr example, an interface could specify an optional readlines()
method, but if it is not available the object would still meet the
requirements for the interface if it provided the read() method.

This instance then gets passed to the __query__() method. The __query__()
method then inspects the requirements, and returns either a class instance
or None. The class instance could either be self or a proxy to self or
something completely separate. A None of course means that it cannot meet
the requirements for the interface.

If no __query__() method is defined on the class, the query() function would
then inspect the methods and other attributes of the class to determine if
it meets the requirements of the interface. Naturally, interfaces with
post-conditions would have to fail immediately in this case, and any
pre-conditions which could not be readily determined would also have to fail
the interface test immediately. A "readily-determined" pre-condition would
be something like "attribute x == 3" (not a good example, but a simple one).

This seems to me a much more general and useful way of specifying
interfaces, whilst covering all the things an interface may need. It allows
the __query__() method to return different things on different invocations
for the same interface, depending on the state of the class. For example, a
file-like object may be unable to meet the "file-like" interface until it
has all its data in memory. Until this is the case, it would return None
from its __query__() method for an interface which required the read()
method.

I may well just try implementing this some time ... most of it shouldn't be
too hard. Unfortunately, I've got a nasty deadline coming up ... :(

Tim Delaney
Avaya Australia
+61 2 9352 9079

All recipient(s) of this email have permission to forward or reply to this
email, quoting this email in full or in part.

> Let me see, a string goes in, and an object 
> comes out. Sure.  Looks good to me!



From root at [127.0.0.1]  Sat Mar 31 16:43:18 2001
From: root at [127.0.0.1] (deadmeat)
Date: Sat, 31 Mar 2001 21:43:18 GMT
Subject: list vs tuple
References: <hGlw6.782$p5.3210@news1.rivrw1.nsw.optushome.com.au><3AC20CD3.265A2D94@alcyone.com><ru69cto61jtmrqsf809e38hb3pbm2lhi7g@4ax.com><bC1x6.4319$4N4.905946@newsc.telia.net><71q9ct4le63anefmiajj5cbh3mg04qu1ga@4ax.com><mailman.985991202.5578.python-list@python.org><7Bfx6.989$p5.3791@news1.rivrw1.nsw.optushome.com.au><Ukhx6.4421$4N4.950603@newsc.telia.net> <mailman.986049281.32365.python-list@python.org> <uGqx6.1052$p5.3997@news1.rivrw1.nsw.optushome.com.au> <tPrx6.18597$Uw.1388730@newsread1.prod.itd.earthlink.net>
Message-ID: <Wbsx6.1058$p5.4247@news1.rivrw1.nsw.optushome.com.au>

> Here's the root of your problem. a = b does NOT, I repeat, does NOT do any
> data copying!!!!!

That is what happens in Pascal, and that is what it _looks_ like is
happening in the first python example (integers only).  But then substitute
a list for the integer, and they behave like pointers.

> Did this help?

I'm not looking for assistance. I'm making an observation.




From mwh21 at cam.ac.uk  Thu Mar  8 17:47:46 2001
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 08 Mar 2001 22:47:46 +0000
Subject: I got 2.1b to dump core.
References: <mailman.984067513.4736.python-list@python.org>
Message-ID: <m33dcnc1yl.fsf@atrus.jesus.cam.ac.uk>

Laura Creighton <lac at cd.chalmers.se> writes:

> Here is a traceback.  Should I send this some place other than here?  

  http://sourceforge.net/tracker/?group_id=5470&atid=105470

What did you *do*?

Cheers,
M.

-- 
  The "of course, while I have no problem with this at all, it's
  surely too much for a lesser being" flavor of argument always 
  rings hollow to me.                       -- Tim Peters, 29 Apr 1998


From mcherm_python at yahoo.com  Sun Mar 18 18:17:24 2001
From: mcherm_python at yahoo.com (Michael Chermside)
Date: Sun, 18 Mar 2001 17:17:24 -0600
Subject: Why "from __future__" stinks; a counter-offer
References: <AD70ED184AFED311ADFD009027AC917F83617A@dmc1nt.zoho.com>
Message-ID: <3AB54204.1030808@yahoo.com>

Hear hear!! I agree!

"from __future__" is ugly. It is non-intuitive. It
    suggests that something having to do with modules
    is going on... which it isn't. And it is very
    difficult to explain to people who don't REALLY
    REALLY understand what's going on.

I don't really expect to get any vote in the matter,
but if we're just looking for opinions to be expressed,
then I'll agree with Jason

-- Michael Chermside

> 


_________________________________________________________
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com




From bos at hack.org  Tue Mar 27 04:51:49 2001
From: bos at hack.org (Rikard Bosnjakovic)
Date: Tue, 27 Mar 2001 11:51:49 +0200
Subject: PIL-resize, aspect ratio
Message-ID: <3AC062B5.D5D0D471@hack.org>

Image.resize() takes a tuple with x,y when I resize, but how do I keep
the aspect ratio? Do I need to calculate it on my own?


-- 
Rikard Bosnjakovic - http://a214.ryd.student.liu.se/cv/ - ICQ: 1158217

Anyone sending unwanted advertising e-mail to my address will be
charged $250 for network traffic and computing time. By extracting my
address from this message or its header, you agree to these terms.


From dsavitsk at e-coli.net  Sun Mar  4 01:38:39 2001
From: dsavitsk at e-coli.net (dsavitsk)
Date: Sun, 4 Mar 2001 00:38:39 -0600
Subject: python cgi syntax highlighter
References: <Pine.BSF.4.32.0103032234260.88338-100000@localhost>
Message-ID: <Uplo6.153501$Ch.31010678@newsrump.sjc.telocity.net>

i just uploaded it to
http://home.uchicago.edu/~dsavitsk/
if this is a problem, that is, if i shouldn't leave it there, please let me
know.

"Gregory (Grisha) Trubetskoy" <grisha at ispol.com> wrote in message
news:Pine.BSF.4.32.0103032234260.88338-100000 at localhost...
>
> Hello -
>
> I remember coming across a script that displayed python code in HTML with
> color syntax highlighting - does anyone know where I can find one? My
> searches don't turn up anything.
>
> Thanks,
>
> Grisha
>




From tjg at exceptionalminds.com  Mon Mar 26 12:38:20 2001
From: tjg at exceptionalminds.com (Timothy Grant)
Date: Mon, 26 Mar 2001 09:38:20 -0800
Subject: So what's wrong with __future__?  (Was Re: Why "from __future__" stinks ...)
In-Reply-To: <99nn3o$p0v$1@panix6.panix.com>; from aahz@panix.com on Mon, Mar 26, 2001 at 07:26:48AM -0800
References: <slrn9bdka2.3j6.hzhu@rocket.knowledgetrack.com> <LNBBLJKPBEHFEDALKOLCAEDKJIAA.tim.one@home.com> <mailman.985568441.15878.python-list@python.org> <99nn3o$p0v$1@panix6.panix.com>
Message-ID: <20010326093820.Q11655@trufflehunter.avalongroup.net>

On Mon, Mar 26, 2001 at 07:26:48AM -0800, Aahz Maruch wrote:
> 
> Overall, this is an ugly solution, but given the axioms that Tim was
> working with, probably the best possible solution in a short period of
> time.  Change the axioms and you get a different solution....

Then I would vote to change the axioms. I think we are seeing
long term ugliness to resolve a short term goal.

Lets have some short term ugliness and continue with the goal
of keeping Python Pythonic.

-- 
Stand Fast,
    tjg.

Timothy Grant                         tjg at exceptionalminds.com
Chief Technology Officer              www.exceptionalminds.com
HyperLINq Technologies, Inc.          <><       (503) 246-3630
>>>>>>>>>>>>>Linux, because rebooting is *NOT* normal<<<<<<<<<
>>>>This machine was last rebooted:  68 days 21:53 hours ago<<



From phrxy at csv.warwick.ac.uk  Tue Mar 20 17:51:03 2001
From: phrxy at csv.warwick.ac.uk (John J. Lee)
Date: Tue, 20 Mar 2001 22:51:03 +0000
Subject: Insertin **keywords into a class
In-Reply-To: <Pine.GHP.4.21.0103201151090.4991-100000@don.et.byu.edu>
References: <mailman.984971990.29857.python-list@python.org> <9960i1$tp0$1@nntp6.u.washington.edu> <99661m$scc$1@panix6.panix.com> <99877b$uss$1@nntp6.u.washington.edu> <Pine.GHP.4.21.0103201151090.4991-100000@don.et.byu.edu>
Message-ID: <Pine.SOL.4.30.0103202245130.10992-100000@mimosa.csv.warwick.ac.uk>

On Tue, 20 Mar 2001, C. Porter Bassett wrote:

> cCan you please tell me why the following code snippet doesn't update the
> value of radius?
>
>
> class myClass:
>    def __init__(self, *arguments, **keywords):
>       self.radius = 0.0
>       for kw in keywords.keys():
>          print kw, ":", keywords[kw]
>          self.kw = keywords[kw]
>       print "self.radius =", self.radius
>
>
> b = myClass(radius = 1.0)

You're setting self.kw to 1.0 when you do

self.kw = keywords[kw]

You want to set self.radius to 1.0, and you can do that with

self.__dict__[kw] = keywords[kw]


John



From ct at gocept.com  Wed Mar  7 18:47:54 2001
From: ct at gocept.com (Christian Theune)
Date: Thu, 08 Mar 2001 00:47:54 +0100
Subject: Copying objects
References: <986h3g$m3s$04$1@news.t-online.com>
Message-ID: <986hba$pi6$03$1@news.t-online.com>

ok ... like everytime ... i found the matching hint 2 minutes after
posting


import copy .... excuse me for the traffic


From s.sharp at fh-aargau.ch  Fri Mar 23 09:14:57 2001
From: s.sharp at fh-aargau.ch (Sharp)
Date: Fri, 23 Mar 2001 15:14:57 +0100
Subject: newbie question
Message-ID: <99fll8$996$1@rex.ip-plus.net>

what is python used for?

is it possible to access a python-script over an apache webserver?

thanx for your answer
Sharp




From cgaebler at gmx.de  Sat Mar 10 07:38:26 2001
From: cgaebler at gmx.de (Carsten Gaebler)
Date: Sat, 10 Mar 2001 13:38:26 +0100
Subject: How to efficently build a nested dictionary
Message-ID: <3AAA2042.E69A53FC@gmx.de>

Hi there!

I wrote a script for FTP traffic accounting, i.e. for each "top level"
FTP directory it counts the bytes transferred per day. The traffic is
stored in a dictionary that has the following structure:

traffic = {dir: {year: {month: {day: 42}}}}

for several dirs, years, months, days, of course.

While parsing the FTP log files I am building up the dictionary in this
way:

if not traffic.has_key(dir):
    traffic[dir] = {year: {month: {day: bytes}}}
elif not traffic[dir].has_key(year):
    traffic[dir][year] = {month: {day: bytes}}
elif not traffic[dir][year].has_key(month):
    traffic[dir][year][month] = {day: bytes}
elif not traffic[dir][year][month].has_key(day):
    traffic[dir][year][month][day] = bytes
else:
    traffic[dir][year][month][day] += bytes


Now, this looks a bit clumsy to me, especially if I consider
you-know-which-language-I-mean which would simply allow to write
$traffic{$year}{$month}{$day} += $bytes;

Any ideas how to do this more elegantly?

Regards
Carsten.


From shredwheat at mediaone.net  Fri Mar 30 12:48:20 2001
From: shredwheat at mediaone.net (Pete Shinners)
Date: Fri, 30 Mar 2001 17:48:20 GMT
Subject: Why not an __assign__ method?
References: <mailman.985969720.18484.python-list@python.org>
Message-ID: <EF3x6.114663$Ch.15367147@typhoon.we.rr.com>

> The question is: Why not have a __assign__ method, that gets called on
> assignment? It should be called on the *right side* of the expression (at
> the left side it does not make sense - I leave the proof as an exercise for
> the reader <wink>). The return value would be then assigned to the left
> side of the expression. This is the default behavior:

i'm not quite sure how this would work. assigning variables
have nothing to do with the type a variable 'currently' refers to.

a = None
a = huge_array1 + huge_array2

the variable could point to anything before being assigned. the
assignment in no way effects what the variable originally refers
to, so it doesn't seem like whatever that was should get a say in
what happens to the "new owner"





From m.wilson at bibliocraft.com  Mon Mar  5 00:29:56 2001
From: m.wilson at bibliocraft.com (Mark Wilson)
Date: Mon, 5 Mar 2001 05:29:56 -0000
Subject: Overriding UserList methods
Message-ID: <004b01c0a535$79cb7300$4a824ac3@mwdell1>

Daniel

try this:

    def append(self, object):
        UserList.UserList.append(object)

that will call your superclass method.

Mark





From gregj at pobox.com  Fri Mar 16 01:42:46 2001
From: gregj at pobox.com (Greg Jorgensen)
Date: Fri, 16 Mar 2001 06:42:46 GMT
Subject: tiny extension crashes
References: <Pine.SOL.4.30.0103141957240.10526-100000@mimosa.csv.warwick.ac.uk> <zUXr6.571195$U46.16976599@news1.sttls1.wa.home.com> <Pine.SOL.4.30.0103152349300.26347-100000@mimosa.csv.warwick.ac.uk>
Message-ID: <GBis6.576490$U46.17184977@news1.sttls1.wa.home.com>

"John J. Lee" <phrxy at csv.warwick.ac.uk> wrote in message
news:Pine.SOL.4.30.0103152349300.26347-100000 at mimosa.csv.warwick.ac.uk...

> What I still don't get, though, is why it was crashing even when only the
> hcf()  function was being imported and called by my (pure) Python code.
> Surely if it doesn't get called, the uninitialised pointer can't scribble
> over anything?

I don't see anything else as obvious as that dangling pointer. Can you give
any more debugging info? I would simplify the extensions bit by bit until
the code didn't crash, and then figure out what I had removed that was
causing the crash.

I haven't used the Numeric module's C API, but I did glance through the
docs. In both your array_max and hcf functions you pull long integers out of
the array with the expression *(long *)(array->data + i*array->strides[0]).
But the data elements aren't necessarily C long integers; your hcf function
will accept an array of short or long ints, and array_max doesn't check the
type at all. By casting the array pointer to long* and then fetching the
value pointed to, you are always fetching sizeof(long) bytes rather than the
size of the actual array elements. This may cause a crash if the OS is
watching your memory references, because at the end of the array you will be
fetching memory that isn't allocated to the array.

You can either test for the type like this:

    long elem;
    char *p;
    ...
    /* extract array[i] into elem, converting it to a long integer */
    p = array->data + i*array->strides[0];
    switch (array->descr->type_num ) {
        case PyArray_SHORT:
            elem = *(short *)p;
        case PyArray_INT:
            elem = *(int *)p;
        case PyArray_LONG:
            elem = *(long *)p;
        default:
            /* error */
    }
    ...

The Numeric module and C API may have better/easier ways to handle this.

> In fact I'll probably scrap all of these functions eventually -- they're
> only there because they were the simplest functions I had lying around,
> and I knew I'd certainly get tangled up in C if I picked something more
> complicated for my first extension.)

Here's a Python GCD implementation. It would of course be faster in C. The
algorithm is a little better than the one you are using. ;-)

def gcd(a,b)
    " return greatest common denominator of a and b"
    g = b
    while a > 0:
        g = a
        a = b % a
        y = g
    return g

def gcdlist(a)
    " return gcd of all numbers in list"
    g = a[0]
    for v in a:
        g = gcd(g,v)
    return g

Argument checking and error handling omitted for brevity.

--
Greg Jorgensen
PDXperts
Portland, Oregon, USA
gregj at pobox.com




From BrianQ at ActiveState.com  Mon Mar 26 22:20:50 2001
From: BrianQ at ActiveState.com (Brian Quinlan)
Date: Mon, 26 Mar 2001 19:20:50 -0800
Subject: math.choice attribute error
In-Reply-To: <XFMail.20010326174339.shaleh@valinux.com>
Message-ID: <004401c0b66c$ed0d02f0$b503a8c0@activestate.ca>

To be more explicit, the choice method does not live in the math module, it
lives in the random module. i.e.:

>>> import random
>>> list = ['apple', 'orange', 'pear']
>>> random.choice( list )
'apple'

Results should vary :-)

-----Original Message-----
From: python-list-admin at python.org
[mailto:python-list-admin at python.org]On Behalf Of Sean 'Shaleh' Perry
Sent: Monday, March 26, 2001 5:44 PM
To: iddwb
Cc: python-list at python.org
Subject: RE: math.choice attribute error



On 27-Mar-2001 iddwb wrote:
>
> from the ref man math.choice says it takes a sequence object and returns a
> random choice. perfect for my needs. the follow happens:
>
>>>> import math
>>>> list = ['apple', 'orange', 'pear']
>>>> print math.choice(list)
> Traceback ....
> attributerror: choice
>

I know random does:

>>> import random
>>> l = [1, 2, 3]
>>> random.choice(l)
2

--
Brian Quinlan
Python and XSLT Developer
BrianQ at ActiveState.com

Use the power of the dragon for your programming needs.
Download Komodo, our Mozilla-based, cross-platform IDE.
      http://www.ActiveState.com/Komodo




From wayne.izatt at _nospam_myself.com  Fri Mar  9 01:00:23 2001
From: wayne.izatt at _nospam_myself.com (Wayne Izatt)
Date: Fri, 09 Mar 2001 06:00:23 GMT
Subject: is it possible to use Tkinter on linux
References: <98714n$3p8$1@news2-2.kornet.net>
Message-ID: <Xj_p6.189678$df5.5158755@news1.crdva1.bc.home.com>

as i recall (this was with 1.5.2 anyway) tkinter could be enabled as part of
the build process (by default it wasn't). it was quite well documented in
the source code, so start there.

the whole thing may have changed with 2.0, but i would start by checking the
source.

cheers



"rambo" <rambo at linuxsecurity.co.kr> wrote in message
news:98714n$3p8$1 at news2-2.kornet.net...
> Hello, there.
> someone knows that it is possible to use Tkinter on linux?
> if it is, what program I should need to use Tkinter on linux,
> how to install , which environment variable should I set.
>
> thanks for your time.
>
>
>
>
>




From vecernik at aon.at  Thu Mar 29 03:38:38 2001
From: vecernik at aon.at (Oliver Vecernik)
Date: Thu, 29 Mar 2001 10:38:38 +0200
Subject: htmllib samples
Message-ID: <3AC2F48E.4EF77C71@aon.at>

Hi,

has anyone got some samples or point me to where to find them on how to
implement this module?

Oliver


From djc at object-craft.com.au  Thu Mar  1 07:29:31 2001
From: djc at object-craft.com.au (Dave Cole)
Date: 01 Mar 2001 23:29:31 +1100
Subject: writing a C function to manipulate python lists
References: <jllmqpnclz.fsf@endor.bbb.caltech.edu>
Message-ID: <m3snkxk6z8.fsf@vole.object-craft.com.au>

>>>>> "Michael" == Michael Vanier <mvanier at endor.bbb.caltech.edu> writes:

Michael> I've been programming in python for many years but have never
Michael> had to solve this particular problem.  I would like to write
Michael> a short C function that will be part of a python-callable
Michael> module.  The function will receive two python lists as
Michael> arguments.  Currently I do this in python, but my code uses
Michael> this function very heavily and it's the major speed
Michael> bottleneck in the code.  The hope is that the C function will
Michael> be much faster than iterating through the lists in python
Michael> (which may not be true, but I'd like to see for myself).

Since I have been doing a few of these for people on this newsgroup
lately, I suppose I should be fair and extend the same offer to you
:-)

If you post the code and it is simple enough, I might be able to whip
up the C equivalent in an hour or so.

- Dave

-- 
http://www.object-craft.com.au


From aleaxit at yahoo.com  Wed Mar 14 18:30:41 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 15 Mar 2001 00:30:41 +0100
Subject: gah! I hate the new string syntax
References: <mailman.983485923.1136.python-list@python.org> <97sudi01i26@news1.newsguy.com> <98elrk$li8$1@nntp.Stanford.EDU> <98fc070thh@news1.newsguy.com> <98iqo2$56o$1@nntp.Stanford.EDU> <98iv3s01sc2@news2.newsguy.com> <slrn9avqlm.8ee.quinn@retch.ugcs.caltech.edu>
Message-ID: <98out40osr@news1.newsguy.com>

"Quinn Dunkan" <quinn at retch.ugcs.caltech.edu> wrote in message
news:slrn9avqlm.8ee.quinn at retch.ugcs.caltech.edu...
> On Mon, 12 Mar 2001 17:56:33 +0100, Alex Martelli <aleaxit at yahoo.com>
wrote:
    [snip]
> >a typeswitch -- a very unadvisable construct.  Something like Haskell's
> >typeclasses, on the other hand, would provide the best of both worlds
> >(multimethods would be an even more powerful alternative) -- let the
> >compiler/interpreter/runtime do the dispatching, and just code in
    [snip]
> Note that in ruby, any time you write
>
> class Foo
>     ...
> end
>
> you are adding to, not overriding, 'Foo'.  Thus,

In Python, you can similarly add whatever Foo attribute you desire at
any time (even add a whole mixin class, &c); the tradeoffs are similar
(for classes -- Python has the added issue of the type/class split).

> Naturally, in a statically-typed functional language like haskell, these
> issues don't come up.  But in an imperative language, it's like frobbing
> __builtins__ ---probably not a very good idea.

Right.  But typeclasses need not be 'invasive' -- I think they could
be implemented dynamically, e.g. in some future Python version.

Now that might be a really powerful use for 'typechecking'.  At
any point I can assert that object X belongs to typeclass Y and
get back a Y-reference -- same as X if isinstance(X,Y) in the
normal Pythonic sense, or else some suitable wrapper when
I've registered with typeclass Y the way to make some existing
type X conform to it.  Now, I can call whatever methods of Y
I desire on that reference.  Hmmm, that might not be _too_
hard to implement in Python, syntactic sugar apart, now that
I think of it.

> In haskell, I could have simply defined a 'Taggable' class and instances
> implementing 'tagof' in the Display module... no problem.  But
"traditional"
> OO seems to make the solution much less clear...

Interesting train of thought...


Alex





From thomas.heller at ion-tof.com  Thu Mar  1 07:28:09 2001
From: thomas.heller at ion-tof.com (Thomas Heller)
Date: Thu, 1 Mar 2001 13:28:09 +0100
Subject: [Distutils] Re: CPAN functionality for python - requirements
References: <Pine.SOL.4.30.0103011156380.1448-100000@mimosa.csv.warwick.ac.uk>
Message-ID: <032e01c0a24b$13d30fa0$e000a8c0@thomasnotebook>

> [...] But *only* if there are actually some proper docs and good
> howto examples for Distutils itself!!  What happened to the effort a while
> back on that?  What remains to be done -- a todo list would be useful to
> spread the load, wouldn't it?
> 
I wrote some docs. They are online already in Fred Drake's development
version:
http://python.sourceforge.net/devel-docs/

Thomas




From victor at prodigy.net  Fri Mar 23 12:35:36 2001
From: victor at prodigy.net (Victor Muslin)
Date: Fri, 23 Mar 2001 17:35:36 GMT
Subject: Commercial Products in Python?
Message-ID: <3abb86ee.142862695@localhost>

A friend of mine is considering developing a shrink-wrapped commercial
product. Python seems like a possible choice, but it is not clear
whether there are any precedents (though probably similar concerns
apply to a Perl-based product).

I would very much appreciate an advice in this area and possibly
answers to some of the following questions:

1. Are there any existing shirnk-wrap commercial products written in
Python?

2. If so, what are they? 

3. What is the best way to go about it? Is it better to have customers
install Python and give them bytecode files or use one of the programs
that creates an executable?

4. There is an easy shrink-wrap install for Python (at least on
Windows NT). Are there such on Unix/Linux platforms? (Typically
customers don't like having to compile software from source).

5. If the answer to question #1 is "no", why not?

6. Are there any licensing issues to be aware of?

Anything else that comes to mind in this direction would be very
helpful.

TIA


From francois.granger at free.fr  Sun Mar 25 09:28:47 2001
From: francois.granger at free.fr (=?ISO-8859-1?Q?Fran=E7ois_Granger?=)
Date: Sun, 25 Mar 2001 14:28:47 GMT
Subject: Good Editor?
References: <3ab9a32b$1_1@news2.one.net> <5Xju6.601501$U46.18300887@news1.sttls1.wa.home.com>
Message-ID: <1eqsula.zv2tmfz7njr5N%francois.granger@free.fr>

Greg Jorgensen <gregj at pobox.com> wrote:

> Most of my Python programs are so small that IDLE works for me. My favorite
> editors are:
> 
> Windows: EditPlus (www.editplus.com)
> Mac: BBEdit

On the Mac, have a look at Pepper
http://www.hekkelman.com

-- 
"De quelque mani?re que l'on divise le monde, il y a toujours la m?me
proportion d'imb?ciles et de c.." - Propos de O.L.Barenton, Confiseur
A. Detoeuf


From smnordby at yahoo.com  Tue Mar  6 08:22:05 2001
From: smnordby at yahoo.com (Steve)
Date: Tue, 06 Mar 2001 05:22:05 -0800
Subject: Can anyone recomend a good intoduction to C...
References: <3aa366c6_2@nnrp1.news.uk.psi.net> <Xx0p6.183448$df5.5139683@news1.crdva1.bc.home.com>
Message-ID: <3AA4E47D.736CBD84@yahoo.com>

I like that K&R is concise, but when just starting out it's nice to have
more complete example code.  I found this Web tutorial which complements
K&R nicely: http://www.howstuffworks.com/c1.htm

-SteveN-

Wayne Izatt wrote:
> 
> Kerninghan and Ritchie - "The C Programming Language" is a beautiful text.
> Everything you want , nothing you don't. Plus, it weighs about as much as a
> magazine. Not many languages can teach you all you need to know in that kind
> of space. The Python Tutorial, of course, is in the same league!
> 
> cheers
> 
> "Simon Brunning" <sbrunning at trisystems.co.uk> wrote in message
> news:3aa366c6_2 at nnrp1.news.uk.psi.net...
> > ... for a Python programmer?
> >
> > I'm going to have to bite the bullet and learn a bit of C. I'd at
> > least like to learn the *right* way to do it.
> >
> > --
> > Cheers,
> > Simon Brunning
> > sbrunning at bigfoot.com
> >
> >
> >


From cfelling at iae.nl  Thu Mar 29 17:46:02 2001
From: cfelling at iae.nl (Carel Fellinger)
Date: 30 Mar 2001 00:46:02 +0200
Subject: Is it possible to sort a dictionary?
References: <GEtv6.59583$Lm2.6378825@news0.telusplanet.net> <3ABE771B.7B9E1631@student.liu.se> <99o33k$aq4$0@216.39.170.247> <99o51301qgb@news1.newsguy.com> <mailman.985660594.13367.python-list@python.org> <99q70i$1le$1@animus.fel.iae.nl> <99vap202a1b@news2.newsguy.com>
Message-ID: <9a0dva$ekv$1@animus.fel.iae.nl>

Alex Martelli <aleaxit at yahoo.com> wrote:
> "Carel Fellinger" <cfelling at iae.nl> wrote in message
> news:99q70i$1le$1 at animus.fel.iae.nl...
>     [snip]
>> I just wonder why the items() method is slooow; values() and keys()
>> are both much faster, going over the whole dict indexing from keys()
>> is even faster.  Is tuple making/unpacking so expensive?  Or is a dict

> I can't reproduce this observation... here's a datapoint:

> sortWithItems: 0.0438600314118
...
> sortWithKeys: 0.0460969072614

> i.e., items() seems a bit _faster_ than keys() then

hm, so it seems.

> going over the dict -- with the script b.py being:

> def sortWithKeys(adict):
>     keys = adict.keys()
>     keys.sort()
>     return [ (key, adict[key]) for key in keys ]

Ah, but besides going over the dict indexing from keys() you're
building an items list here too:) If you make a list of only the
values instead of key,value tuples, it will be the faster one.

And that's what surprised me.  I expected items() to be faster then
the combination of dict.keys(); [dict[key] for key in dict.keys] as
items is all done in C and the alternative is mostly done in Python.
E.g., in items there is no need to check that dict is still the same
for each key, but in the Python loop, dict has to be looked-up for
each key.  Somehow I thought that such things would make up for the
building of tuples.

-- 
groetjes, carel


From bvdpoel at uniserve.com  Thu Mar 29 13:36:30 2001
From: bvdpoel at uniserve.com (Bob van der Poel)
Date: Thu, 29 Mar 2001 11:36:30 -0700
Subject: forcing dictionary order
Message-ID: <3AC380AE.B225C07F@uniserve.com>

I'd like (not a must!) to walk a dictionary in the same order in which
it was created. Example:

	foo={ key1: (), key2: (), key3: () }

If I do:
	
	for a in foo.keys(): ...

the order is not "key1, key2, key3", which is what I want. Note, I can't
do something simple like sort() on the keys since the are not in any
alphanumeric order. I'm just using this to generate a largish table for
another program, and the actual order doesn't matter for it to work
properly, but it would nicer if it were in the correct logical order.
So, I'm not going to spend a lot of time on this, but if there is a
simple way????

-- 
Bob van der Poel ** Wynndel, British Columbia, CANADA **
EMAIL: bvdpoel at uniserve.com
WWW:   http://users.uniserve.com/~bvdpoel



From guijarro at wanadoo.fr  Thu Mar 22 12:29:21 2001
From: guijarro at wanadoo.fr (guijarro)
Date: Thu, 22 Mar 2001 18:29:21 +0100
Subject: Sockets in Python
Message-ID: <99dcpm$am1$1@wanadoo.fr>

Hello guys,

here is my problem : I'd like to create a simple network card game with
Python,
but I've got some troubles with the sockets...

In fact I'd like to find some kind of tutorial about this subject, because
although
my first-time code quite works, I don't think it is really good.

When I try all the samples I've found on the 'net about sockets, the "main"
loop for communicating, generally implemented with a While 1: ..., makes my
computer frozen up (under Windows, not under Linux).
That's why I thought it would be better to use threads, for the listening
function
for example, but it doesn't work well under Windows (sometimes it hangs up,
I can't figure out why !).

I'm sure it's possible creating a robust and nice app which uses the
sockets encapsulation provided with Python under Windows, but how ?
What are the "good" things to do ?
What is the best module to use between socket or asyncore ?

Well, if someone can help me with the sockets in general and the sockets
with Python, it would be greatly appreciated...
(I've already read the "classic" tutorials about sockets, I followed the
links
on the Python website, etc..)

Thanks in advance.









From corour01 at motorola.com  Tue Mar 20 12:47:56 2001
From: corour01 at motorola.com (O'Rourke Clodagh-corour01)
Date: Tue, 20 Mar 2001 17:47:56 -0000
Subject: Tkinter Problem
Message-ID: <13EE655665F4D311AB4D0008C789498A01EE2C85@zei02exm02.cork.cig.mot.com>

Hi all,

I hope someone can help me because this problem is driving me mad.....

I am creating a GUI system in Python.  

Each window is a class and launches ok on it's own.  

Classes:	AuditResync.py*   GuiSelection.py*  System.py* Cell.py*  InitialWin.py*   EnvVar.py*       

There is one Initial selection window (InitialWin.py) which contains several button each of which destroys the current window and launches a new one whcih is an instance of one of the other 4 classes.  

	 	self.destroy()
		<CLASSNAME>(self.parent)

I am getting a NameError (below) even though the classes does exist and launch perfectly as a stand-alone.   I have imported imported thef iles I am trying to access too.  The first instance created of a window launches fine and when I replace that with another window everything is still fine.  However if I go back to a window that I had already visited then this is where the error arises.

Exception in Tkinter callback
Traceback (innermost last):
  File "/usr/vob/omc_reference/ThirdParty/scriptSources/lib/python1.5/lib-tk/Tkinter.py", line 764, in __call__
    return apply(self.func, args)
  File "AuditResync.py", line 54, in newGui
    GuiSelection(self.parent)
NameError: GuiSelection

Has anyone else had any similar errors or knows of any possible causes?

Thanks a lot,
Clodagh O' Rourke



From nelson at crynwr.com  Tue Mar  6 01:56:24 2001
From: nelson at crynwr.com (Russell Nelson)
Date: Tue, 06 Mar 2001 01:56:24 -0500
Subject: Can Python replace Visual Basic? Should it?
References: <5DZo6.11126$hn5.1578410@news1.rdc1.mb.home.com>
Message-ID: <3AA48A18.FEEDA8B9@crynwr.com>

Brad Bollenbach wrote:
> Again, I'm much more interested in the real world experience of those who
> used to use the big visual development tools in Windows like VB, Delphi or
> Powerbuilder rather than theoretical answers.

I've built a little program in VB, and a program in
glade/pygtk/libglade/Python.  I can say for sure that the latter is
nearly as good as the former.  It needs to cooperate better with an
external text editor.  But there's no reason why it couldn't do that.

The really neat thing I've discovered so far is that when you use
libglade, your user interface layout has very little to do with the user
interface that drives it.  Your code knows to look for certain widgets,
and manipulate them as you have specified.  The user interface designer
knows to name certain widgets certain things.  Beyond that, though,
there's a nearly complete separation of functionality.  You can make the
same application run on wildly different size screens with NO code
changes.  Or, users can customize their screens.

-- 
-russ nelson <sig at russnelson.com>  http://russnelson.com
Crynwr sells support for free software  | PGPok | Watch out!  He's got
an
521 Pleasant Valley Rd. | +1 315 268 1925 voice | opinion, and he's not
Potsdam, NY 13676-3213  | +1 315 268 9201 FAX   | afraid to share it!


From aahz at panix.com  Wed Mar 14 00:14:29 2001
From: aahz at panix.com (Aahz Maruch)
Date: 13 Mar 2001 21:14:29 -0800
Subject: [Python-Dev] Re: Bug fix releases
References: <mailman.983681106.14778.python-list@python.org> <3aa2e9c3.210139114@news.laplaza.org>
Message-ID: <98munl$fup$1@panix3.panix.com>

[I'm writing up a PEP for this, but wanted to respond to one bit before
this post scrolls off my spool.]

In article <3aa2e9c3.210139114 at news.laplaza.org>,
Mats Wichmann <xyzmats at laplaza.org> wrote:
>
>I can't see that this will actually happen until there's a champion who
>will handle the "boring stuff" - that is, babysit the older released
>line (we don't all think that's boring work, but what the heck).  If
>I can point at the Linux kernel development (don't anybody gack now),
>Linus handles the "interesting stuff" - the latest branch; when it
>reaches a certain degree of stability he hands over to Alan Cox.  Did
>Aahz just volunteer to be our Alan Cox?

Maybe.  ;-)  Depends on what Alan Cox actually does; if he decides which
patches go in, I can't be Alan, not by myself.  I just don't have enough
knowledge of the Python kernel and I'm not all that good of a C
programmer (that's why I use Python!).

Can someone write a bit more about the actual Linux maintenance process?
-- 
                      --- Aahz <*>  (Copyright 2001 by aahz at pobox.com)

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

Python:
    Programming the way
    Guido indented it


From f8dy at diveintopython.org  Sat Mar  3 14:04:46 2001
From: f8dy at diveintopython.org (Mark Pilgrim)
Date: Sat, 03 Mar 2001 14:04:46 -0500
Subject: Missing the functional mind set
In-Reply-To: <mlb2at8ptrr7sh5dgq8qlk2201kquh6tfl@4ax.com>
References: <mlb2at8ptrr7sh5dgq8qlk2201kquh6tfl@4ax.com>
Message-ID: <B6C6AA7D.4907%f8dy@diveintopython.org>

in article mlb2at8ptrr7sh5dgq8qlk2201kquh6tfl at 4ax.com, Daniel Klein at
danielk at aracnet.com wrote on 3/3/01 1:08 PM:

> The purpose of this code is to return a list of dictionary values sorted by
> key, resulting in 'templist'.
> 
>>>> d = {3: 'd', 2: 'b', 1: 'a', 0: 'z'}
>>>> templist = []
>>>> for n in range(len(d)):
> templist.append(d[n])
> 
>>>> templist
> ['z', 'a', 'b', 'd']
> 
> My question is, is there a 'functional' (or list comprehension) solution to
> this?

Since you explicitly asked for hints, not answers, here are some hints:
- AFAIK, the fastest way for me to sort things in Python is with the sort()
method of a list, because it's written in C and has been optimized by people
smarter than me.
- sort() modifies an existing list in place; it does not return a new list.
- sort() will sort a list of any type of object, not just simple datatypes.
For instance, if you have a list of tuples, sort() will perform a multi-key
sort, treating the first element of each tuple as the primary key, the
second element as the secondary key (to break ties where the first elements
are the same), and so forth.
- Dictionaries have an items() method that returns a list of tuples [(key1,
value1), (key2, value2), ...]
- Once you have a list of sorted key/value pairs, you can use a list
comprehension to pull out just the value from each pair.

You can do it in 3 lines using a list comprehension.  Or 5 lines using a for
loop.  Or 1 heinous-looking UNIX command line.  Your choice.

-M
You're smart; why haven't you learned Python yet?
http://diveintopython.org/



From aleaxit at yahoo.com  Wed Mar 21 10:32:10 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Wed, 21 Mar 2001 16:32:10 +0100
Subject: Html entities
References: <Sp2u6.3034$mh4.198592@news3.oke.nextra.no> <3AB8BA56.9E41A333@downright.com>
Message-ID: <99ahho01sj1@news2.newsguy.com>

"Doug Fort" <dougfort at downright.com> wrote in message
news:3AB8BA56.9E41A333 at downright.com...
> Syver Enstad wrote:
>
> > Is there an easy way to convert ISO Latin-1 characters that are above
127
> > ascii to their HTML, XML entity form?
>
> See 'htmlentitydefs' in the Python library

This maps entitydef->ISO-Latin-1-character -- you'll need to
'reverse the dictionary' for Syver's purposes.


Alex





From corour01 at motorola.com  Wed Mar 21 08:48:14 2001
From: corour01 at motorola.com (O'Rourke Clodagh-corour01)
Date: Wed, 21 Mar 2001 13:48:14 -0000
Subject: Access Issues
Message-ID: <13EE655665F4D311AB4D0008C789498A01EE2C93@zei02exm02.cork.cig.mot.com>

Hi all,

I wonder if anyone can help.

I am trying call the constructor of the EnvVar class from the newEnvWin function of the Cell class.
But I am getting the following error

Exception in Tkinter callback
Traceback (innermost last):
  File "/usr/vob/omc_reference/ThirdParty/scriptSources/lib/python1.5/lib-tk/Tkinter.py", line 764, in __call__
    return apply(self.func, args)
  File "Cell.py", line 63, in newCellMgt
    self.newEnvWin('cellmgt')
  File "Cell.py", line 60, in newEnvWin
    instance = EnvVar.__init__(g, self.parent)
AttributeError: __init__

I have include some extracts of the code.
Thank you,
Clodagh.

**********************************************
import EnvVar, GuiSelection, AuditResync, System
from shellgui import EnvVarSite
import Pmw
from Tkinter import *

class Cell(EnvVarSite):
    def __init__(self, parent=None, **kw):      
	  ......

    def newCellMgt(self):
        self.newEnvWin('cellmgt')
 
    def newEnvWin(self, g):
      newParent = Toplevel()
      instance = EnvVar.__init__(g, self.parent)

********************************************
	
import informixdb, os, sys, string, Pmw
from Tkinter import *
from shellgui import EnvVarSite
import GuiSelection, AuditResync, Cell, System

class EnvVar(EnvVarSite):

    def __init__(self, g, parent=None):
    	self.sels = None 
      self.parent = parent
	self.group = g
	self.edit = 1
	self.GetDBValues()
	EnvVarSite.__init__(self,parent)

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



From mwh21 at cam.ac.uk  Wed Mar 14 15:32:02 2001
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 14 Mar 2001 20:32:02 +0000
Subject: Extending Python: rewriting a single method in C
References: <g0wv9vxms0.fsf@scumbag.ecs.soton.ac.uk> <slrn9apn8f.382.jepler@potty.housenet> <g0ofv6xbj4.fsf@scumbag.ecs.soton.ac.uk> <98l1i901k4u@news1.newsguy.com> <g0lmq9xxz0.fsf@scumbag.ecs.soton.ac.uk> <gYur6.119$4d3.23079@newsb.telia.net> <g0hf0wy7cw.fsf@scumbag.ecs.soton.ac.uk> <XZMr6.261$4d3.54975@newsb.telia.net>
Message-ID: <m3pufk6qil.fsf@atrus.jesus.cam.ac.uk>

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

> Jacek Generowicz wrote:
> > No, I'm quite sure that I didn't . . . for some reason I was buried
> > in http://www.python.org/doc/current/ext
> >
> > (Oops! Blush)
> >
> > > http://www.python.org/doc/current/api/
> >
> > This is much more like it.
> 
> maybe it's time to merge these two documents into one?
> 
> (or at least make them refer/link to each other)

They could both do with being rewritten from scratch (especially ext;
api just needs completing).

If I had a spare month...

Cheers,
M.

-- 
  Programming languages should be designed not by piling feature on
  top of  feature, but by removing the  weaknesses and restrictions
  that make the additional features appear necessary.
               -- Revised(5) Report on the Algorithmic Language Scheme


From paulp at ActiveState.com  Sun Mar 11 10:36:29 2001
From: paulp at ActiveState.com (Paul Prescod)
Date: Sun, 11 Mar 2001 07:36:29 -0800
Subject: "!=" is bad form. Re: sorry....never mind
References: <LNBBLJKPBEHFEDALKOLCCELJJEAA.tim.one@home.com> <11032001.1@wanderer.local.jae.ddns.org>
Message-ID: <3AAB9B7D.E7D892B6@ActiveState.com>

"J?rgen A. Erhard" wrote:
> 
> ...
> So when can we expect `||' and `&&' to replace `or' and `and'?
> Shouldn't we `Beware of the slippery slope'?

Slippery slope arguments are not helpful in language design or for that
matter most kinds of design. Each decision is distinct. Guido must weigh
a host of costs and benefits every time. If a "benefit" is compatibility
with C and the costs are low or non-existent then he will choose to be
compatible with C.

-- 
Python:
    Programming the way
    Guido
    indented it.



From whisper at oz.net  Mon Mar 26 15:40:36 2001
From: whisper at oz.net (Dave LeBlanc)
Date: 26 Mar 2001 20:40:36 GMT
Subject: PEP 245: Python interfaces
References: <Pine.LNX.4.32.0103222253510.26302-100000@localhost.localdomain>
Message-ID: <99o9g4$nh3$0@216.39.170.247>

Having read the PEP briefly, I have one question:

Is there a good reason to add the "implements" keyword?

Current syntax is (from Language Reference 7.6):
classdef:       "class" classname [inheritance] ":" suite
inheritance:    "(" [expression_list] ")"
classname:      identifier
(i'm adding this based on the text of 7.6:)
expression_list: Expression(s) which evaluate to a class object.

Any reason why expression_list couldn't include an interface? Seems to
me it's easier to just extend the semantics of expression_list to
include interface_name(s) instead of adding a variation like '"class"
classname [inheritance] ["implements"   interfaces] ":" suite'.
(Besides, two optional elements in a row makes for messy parsing :-).)
The end of a class definition might be slightly more complex to check
that all "pure" methods have implementations, but (almost certianly)
not hopelessly so... and you get to detect the "not implemented" error
early with a "missing definition" error rather then later with a
"can't instantiate" error.

Surely a python interpreter could be smart enough to realize that some
name in the list of inherited objects is actually an interface?
(OTOH, it doesn't seem smart enough to infer that the first argument
of a method within a class is "self" <g>).

Of course, one could do '"class" classname [" inherits" inheritance]
["implements" interfaces] ":" suite'  if verbosity is the goal. (This
would also be easier to parse :-).)

Just my $.02 worth.

Dave LeBlanc

On Thu, 22 Mar 2001 23:02:35 -0800, Michel Pelletier
<michel at digicool.com> wrote:

>
>Hi!
>
>I'm the author of PEP 245, a framework and syntax for interface objects in
>Python.
>
>http://python.sourceforge.net/peps/pep-0245.html
>
>It has just recently been made a PEP, and I'd like to solicite some more
>input from the community (you).
>
>For those of you who don't know the formal defintion of "interface", don't
>worry (there isn't one!).  There are, however, a lot of agreed and
>disagreed upon definitions out there in the world, and PEP 245 is one
>particular take on the situation, based on pioneering work done by the
>Type-sig and Jim Fulton over the last couple of years.
>
>There is a Wiki (referenced in the PEP) for formal discussion and, of
>course, this newsgroup, which I will hanging out in for the next few days
>answering any questions about the proposal.
>
>Thanks!
>
>-Michel
>



From mlh at idi.ntnu.no  Fri Mar 16 05:58:43 2001
From: mlh at idi.ntnu.no (Magnus Lie Hetland)
Date: Fri, 16 Mar 2001 11:58:43 +0100
Subject: I come to praise .join, not to bury it...
References: <mailman.983485923.1136.python-list@python.org> <97p7iq$kk2$1@nntp6.u.washington.edu> <97rka002k9@news1.newsguy.com> <3AAF7E47.294617E1@sage.att.com> <98qeco02hn7@news1.newsguy.com>
Message-ID: <98srjf$fov$1@tyfon.itea.ntnu.no>

"Alex Martelli" <aleaxit at yahoo.com> wrote in message
news:98qeco02hn7 at news1.newsguy.com...
[...]
> And then again, the _appearance_ that too much emphasis
> was being placed on syntax-sugary issues kept me *AWAY*
> from Python for longer than it should have -- sure, sure,
> I could save some punctuation, but, *WHAT DID I CARE*?!
>
> Good thing (for me) that I eventually looked beyond the
> pretty face, of course:-).

Why? If you don't care about the pretty face, what are
the advantages of Python as opposed to, say, Perl?
To me it would seem that syntax (sugar or not) is
the main issue... Not that I've used Perl a lot, but
it seems very much like Python with a lot of ugly bits
thrown in...

--

  Magnus Lie Hetland          http://www.hetland.org

 "Reality is what refuses to disappear when you stop
  believing in it"                 -- Philip K. Dick





From mwh21 at cam.ac.uk  Thu Mar 15 13:29:23 2001
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: Thu, 15 Mar 2001 18:29:23 +0000 (GMT)
Subject: python-dev summary, 2001-03-01 - 2001-03-15
Message-ID: <Pine.LNX.4.10.10103151828520.24973-100000@localhost.localdomain>

 This is a summary of traffic on the python-dev mailing list between
 Mar 1 and Mar 14 (inclusive) 2001.  It is intended to inform the
 wider Python community of ongoing developments.  To comment, just
 post to python-list at python.org or comp.lang.python in the usual
 way. Give your posting a meaningful subject line, and if it's about a
 PEP, include the PEP number (e.g. Subject: PEP 201 - Lockstep
 iteration) All python-dev members are interested in seeing ideas
 discussed by the community, so don't hesitate to take a stance on a
 PEP if you have an opinion.

 This is the third python-dev summary written by Michael Hudson.
 Previous summaries were written by Andrew Kuchling and can be found
 at:

   <http://www.amk.ca/python/dev/>

 New summaries will appear at:

  <http://starship.python.net/crew/mwh/summaries/>

 and will continue to be archived at Andrew's site.

   Posting distribution (with apologies to mbm)

   Number of articles in summary: 264

    50 |                                             ]|[        
       |                                             ]|[        
       |                                             ]|[        
       |                                             ]|[        
    40 | ]|[                                         ]|[        
       | ]|[                                         ]|[        
       | ]|[                                         ]|[        
       | ]|[                                         ]|[ ]|[    
    30 | ]|[                                         ]|[ ]|[    
       | ]|[                                         ]|[ ]|[    
       | ]|[                                         ]|[ ]|[ ]|[
       | ]|[                                         ]|[ ]|[ ]|[
    20 | ]|[                                         ]|[ ]|[ ]|[
       | ]|[ ]|[                                     ]|[ ]|[ ]|[
       | ]|[ ]|[                                     ]|[ ]|[ ]|[
       | ]|[ ]|[                                 ]|[ ]|[ ]|[ ]|[
    10 | ]|[ ]|[ ]|[                             ]|[ ]|[ ]|[ ]|[
       | ]|[ ]|[ ]|[     ]|[                     ]|[ ]|[ ]|[ ]|[
       | ]|[ ]|[ ]|[     ]|[ ]|[                 ]|[ ]|[ ]|[ ]|[
       | ]|[ ]|[ ]|[ ]|[ ]|[ ]|[ ]|[     ]|[ ]|[ ]|[ ]|[ ]|[ ]|[
     0 +-050-022-012-004-009-006-003-002-003-005-017-059-041-031
        Thu 01| Sat 03| Mon 05| Wed 07| Fri 09| Sun 11| Tue 13|
            Fri 02  Sun 04  Tue 06  Thu 08  Sat 10  Mon 12  Wed 14

 A quiet fortnight on python-dev; the conference a week ago is
 responsible for some of that, but also discussion has been springing
 up on other mailing lists (including the types-sig, doc-sig,
 python-iter and stackless lists, and those are just the ones your
 author is subscribed to).


   * Bug Fix Releases *

 Aahz posted a proposal for a 2.0.1 release, fixing the bugs that have
 been found in 2.0 but not adding the new features.

  <http://mail.python.org/pipermail/python-dev/2001-March/013389.html>

 Guido's response was, essentially, "Good idea, but I don't have the
 time to put into it", and that the wider community would have to put
 in some of the donkey work if this is going to happen.  Signs so far
 are encouraging.


    * Numerics *

 Moshe Zadka posted three new PEP-drafts:

  <http://mail.python.org/pipermail/python-dev/2001-March/013435.html>

 which on discussion became four new PEPs, which are not yet online
 (hint, hint).

 The four titles are

    Unifying Long Integers and Integers
    Non-integer Division
    Adding a Rational Type to Python
    Adding a Rational Literal to Python

 and they will appear fairly soon at

  <http://python.sourceforge.net/peps/pep-0237.html>
  <http://python.sourceforge.net/peps/pep-0238.html>
  <http://python.sourceforge.net/peps/pep-0239.html>
  <http://python.sourceforge.net/peps/pep-0240.html>

 respectively.

 Although pedantically falling slightly out of the remit of this
 summary, I should mention Guido's partial BDFL pronouncement:

  <http://mail.python.org/pipermail/python-dev/2001-March/013587.html>

 A new mailing list had been setup to discuss these issues:

  <http://lists.sourceforge.net/lists/listinfo/python-numerics>


    * Revive the types-sig? *

 Paul Prescod has single-handedly kicked the types-sig into life
 again.

  <http://mail.python.org/sigs/types-sig/>

 The discussion this time seems to be centered on interfaces and how to
 use them effectively.  You never know, we might get somewhere this
 time!

    * stackless *

 Jeremy Hylton posted some comments on Gordon McMillan's new draft of
 the stackless PEP (PEP 219) and the stackless dev day discussion at
 Spam 9.

  <http://mail.python.org/pipermail/python-dev/2001-March/013494.html>

 The discussion has mostly focussed on technical issues; there has
 been no comment on if or when the core Python will become stackless.


    * miscellanea *

 There was some discussion on nested scopes, but mainly on
 implementation issues.  Thomas Wouters promised <wink> to sort out
 the "continue in finally: clause" wart.

Cheers,
M.




From jasonic at nomadicsltd.com  Sat Mar 31 01:11:13 2001
From: jasonic at nomadicsltd.com (Jason Cunliffe)
Date: Sat, 31 Mar 2001 01:11:13 -0500
Subject: Know of Substantial Apps Written in Python?
References: <3ac4e4a8_1@news4.newsfeeds.com> <KE6x6.415$SR6.235983@typhoon1.ba-dsg.net> <dj7x6.9561$SB2.850896@ruti.visi.com> <3AC5529D.8D7BCC63@seebelow.org>
Message-ID: <tcassvke9huhd9@corp.supernews.com>

> Hey, I looked at that, and it looks cool.  But I was somewhat puzzled by
> the following statement:
>
>    Sketch is an interactive vector drawing program for Linux
>    and other UNIX compatible systems. <snip>
>
>    A somewhat unique(for a drawing program) feature of Sketch is
>    that it is implemented almost completely in a very high-level,
>    interpreted language, Python.
>
> If it's implemented almost completely in Python, why does it only run on
> "UNIX-compatible" systems?  (BTW, does it delete hyphens? <wink>)  Is
> there something about its "almost" part that can't be done in Windows.

The logic Python arhitecture in 'Sketch' is likely very portable, but I
_imagine_ the UI toolkit chosen may be the {for the time being} the reason
it is restricted to *nixen

A major issue for cross platform 'substantial' apps is the GUI part of the
code. And typically a lot of work goes into cool widgets. Python is more
prtable than GUIs

wxPython is a very cool interface to wxWindows.. For a substantial python
app take a look at it first, and then BOAConstructor
http://boa-constructor.sourceforge.net/

[I think the Mac versions of wxWindows/wxPython still lag behind..not
Python's fault. Hopefully MacOSX will improve much of this scenario. Within,
say, a year the Mac could have many more interesting tools migrated from
Linux/Unix developers]

The major alternatives to wxWindows appear to be Tkinter and Qt.
Check on the Vaults of Parnassus for latest on these.
http://www.vex.net/parnassus/

.. check under User Interfaces section.
for example
http://www.vex.net/parnassus/apyllo.py?i=247038364

For substantial Python apps thre is also the section titled Applications
under the 'tree' menu .. http://www.vex.net/parnassus/apyllo.py?i=77689886
In general checking the 'Latest' page will provide a wnoderful insight into
the substantial work onoing by dedicated python creative teams

Another little discussed substantial is 'OpenEV', writen by some very smart
dedicated GIS people in Python:

http://openev.sourceforge.net/


enjoy :-)

- Jason
___________________________________________________________
Jason CUNLIFFE = NOMADICS['Interactive Art and Technology']




From aahz at panix.com  Tue Mar 20 11:06:01 2001
From: aahz at panix.com (Aahz Maruch)
Date: 20 Mar 2001 08:06:01 -0800
Subject: Functionalism, aesthetics Was:(RE: I come to praise .join,  not
References: <mailman.985066344.32125.python-list@python.org>
Message-ID: <997v59$sbf$1@panix3.panix.com>

In article <mailman.985066344.32125.python-list at python.org>,
Clark C. Evans <cce at clarkevans.com> wrote:
>
>Well, this is a problem with open source in general.  Most 
>coders have the attitude: "If you can't code, get out of
>the kitchen".  Sad, but true.

There's some of this in the Python community, but more often around here
it's simply a demand to be clear about what's wanted.  As Tim is fond of
pointing out, every time he asks people how to do floating point "the
right way", he gets *at* *least* as many answers as people responding.
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

There's a difference between a person who gets shit zie doesn't deserve
and a person who gets more shit than zie deserves.  --Aahz


From scarblac at pino.selwerd.nl  Wed Mar  7 05:22:52 2001
From: scarblac at pino.selwerd.nl (Remco Gerlich)
Date: 7 Mar 2001 10:22:52 GMT
Subject: class object access in class definition?
References: <Pine.LNX.4.21.0103061556510.2015-100000@zero>
Message-ID: <slrn9ac2sa.ah1.scarblac@pino.selwerd.nl>

Lars Damerow <lars at pixar.com> wrote in comp.lang.python:
> Is it possible to reference a class from within its own definition? Something
> along the lines of:
> 
> class Foo:
>     print Foo
> 
> This works for functions--that is,
> 
> >>> def foo():
> ...     print foo
> ...
> >>> foo()
> <function foo at 0x8111024>
> 
> works as expected, but the class definition above raises a NameError:
> 
> >>> class Foo:
> ...     print Foo
> ...
> Traceback (most recent call last):
>   File "<stdin>", line 1, in ?
>   File "<stdin>", line 2, in Foo
> NameError: There is no variable named 'Foo'

The difference is that the function is executed *after* you defined it, so
that by then the name exists. In the case of the class, it is executed
during its definition, and at that time the name doesn't exist yet.

I have no idea what you're trying to do, but maybe a workaround is to create
an empty class of another name first, and inherit from that. You can then
use references to that class, and be able to influence the current class.

But well, no idea what you're trying to do :)

-- 
Remco Gerlich


From cribeiro at mail.inet.com.br  Mon Mar 26 21:37:29 2001
From: cribeiro at mail.inet.com.br (Carlos Alberto Reis Ribeiro)
Date: Mon, 26 Mar 2001 23:37:29 -0300
Subject: Is it possible to sort a dictionary?
In-Reply-To: <99op90$2tr$1@animus.fel.iae.nl>
References: <GEtv6.59583$Lm2.6378825@news0.telusplanet.net>
 <3ABE771B.7B9E1631@student.liu.se>
 <99o33k$aq4$0@216.39.170.247>
 <99o51301qgb@news1.newsguy.com>
Message-ID: <5.0.2.1.0.20010326232409.021e4930@mail.inet.com.br>

At 03:09 27/03/01 +0200, you wrote:
>Alex Martelli <aleaxit at yahoo.com> wrote:
>
> > Fine!  Even faster & simpler:
>
>Simpler yes, but faster? Are you sure?  Lately I tested something
>similar and was surprised when I found the dict.items function to be
>soo slow that going through the whole dict indexing each entry
>one by one was faster.

I was thinking about something similar today. It would be useful to have a 
"xitems()" method, behaving exactly like items(), but *without* the need to 
build the tuple in advance. I see two advantages:

- it would use less memory (no need to store the items sequence elsewhere). 
Python don't copy all the data, only the references, but even this can be 
big for a large dictionary.

- it would run faster for most cases, and possibly *much* faster if the 
loop have a "break" for some condition. The startup time would also be 
greatly reduced for a large dictionary.

As part of my recent work with adapters (as discussed on the list), I was 
thinking about implementing a 'proxy' object to take the dictionary and 
make it behave like a sequence of items. For the xitems() version to work, 
I need to find a way to walk sequentially through the hash table directly, 
and I haven't figured out to make it - I was far away from the computer 
when the idea ocurred to me. If someone know I would be glad to learn <wink>


Carlos Ribeiro





From gregj at pobox.com  Mon Mar 12 03:31:22 2001
From: gregj at pobox.com (Gregory Jorgensen)
Date: Mon, 12 Mar 2001 08:31:22 GMT
Subject: Einstein's Riddle
References: <oNPq6.396644$ge4.138479566@news2.rdc2.tx.home.com> <UdSq6.2184$54.2166@www.newsranger.com> <984380185.768920@news.aei.ca>
Message-ID: <uP%q6.2375$54.2207@www.newsranger.com>

Assuming the unknown pet is the fish, solving the puzzle doesn't take any great
skill with logic or math; certainly more than 2% of the population can figure it
out. I think the riddle demonstrates the fallacy of unwarranted assumptions.
It's a trick question; you work out the solution only to be told "I never said
anyone had fish." The more literal-minded the reader, the more likely they will
be taken in.


In article <984380185.768920 at news.aei.ca>, sulfugor says...
>But the fish is mentionned in the question before the clues are given. We
>are able to find who owns each of the 4 pets mentionned in the hint list. is
>it so unreasonable to assume then that the last pet is a fish, since we were
>asked about it ? The question IMplied that one of the 5 people owned a fish.
>
>sulfugor
>
>ps = by the way, how long is this supposed to take ( solving the riddle) ?

Greg Jorgensen
Deschooling Society
Portland, Oregon, USA
gregj at pobox.com


From ullrich at math.okstate.edu  Sat Mar  3 11:32:27 2001
From: ullrich at math.okstate.edu (David C. Ullrich)
Date: Sat, 03 Mar 2001 16:32:27 GMT
Subject: Hitting two targets:  OO + group theory (pedagogy)
References: <u2k0atkgkiatdp9ukj3mpd1ej0b648ldsm@4ax.com> <3aa1077a.1110803@nntp.sprynet.com> <0932atgt9v8d8aboq90evoi5ipj5c0sk8s@4ax.com>
Message-ID: <3aa11a91.5998558@nntp.sprynet.com>

On Sat, 03 Mar 2001 07:54:43 -0800, Kirby Urner
<urner at alumni.princeton.edu> wrote:

>ullrich at math.okstate.edu (David C. Ullrich) wrote:
>
[...]
>>Seems more interesting to me to start with a class
>>Group that includes stuff common to all groups, so
>>that a subclass can represent any group at all. If we're
>>going to implement what's above as part of a
>>"curriculum" then at least we should make it clear
>>that our sandbox is not "groups", it's a tiny part of
>>the category of all groups.
>
>Yes, just an example of a group.  In another essay, I 
>go with permutations, showing how these might be associated
>with "clubhouse codes" (exchanging letters for other 
>letters as in a Julius Caesar code).  I'd also like 
>to build on my OO + Polyhedra approach to get into 
>symmetry groups -- getting topics to connect around to
>other topics as much as feasible is what I mean by a
>"curriculum" (stuff connecting around in all 
>circumferential directions, to form a system -- the
>notion of circling-back is built right into the 
>etymology).
>
>I try to be fairly general in my approach to the 
>group of relatively prime integers under multiplication
>vis-a-vis modulus n in my little essay for 8th graders:
>
>   When we study group theory, we have a mnemonic 
>   that's helpful if you've studied the book of 
>   Genesis at least enough to know the story of 
>   Cain and Abel (use a search engine maybe).  Use 
>   the letters CAIN to stand for "Closure, Associative, 
>   Inverse, Neutral" -- and there you have the 
>   properties you need to call your set & operation 
>   a "group".
>
>   These properties are defined with respect to a set 
>   of elements, and an operation for relating them 
>   (we generally call the operation "multiplication", 
>   no matter what it actually does). If multiplication 
>   is not just associative, but commutative as well, 
>   then that's where Abel comes in -- we call it an 
>   Abelian Group.
>
>Probably there's some syntax that could be tightened
>up here too.

Some people would say that the reference to Genesis
is out of place here... probably you know that some
people would say that. Might be regarded as just
generally inappropriate - some other people might
think that the bit about Abel is misleading, since
"Abelian" _does_ refer to a man named Abel, but
not the one in Genesis.

But what the heck, now that W's in charge
we're not worrying about technicalities like
separation of church and state anymore.

But why not start with a class Group instead
of special cases? (Would be a class with a
__mul__ method that does nothing but raise
an exception "must implement __mul__ in
a subclass", an Id attribute, etc. I can't think
of anything offhand that Group would actually
_do_, but if we're bent on this OO-math thing
it seems like groups _should_ be descendants
of a class Group...)

>Thanks for your useful feedback.  Much appreciated.
>
>Kirby
>



From dev1.gemodek at t-online.de  Wed Mar 14 04:37:19 2001
From: dev1.gemodek at t-online.de (Stephane)
Date: Wed, 14 Mar 2001 10:37:19 +0100
Subject: zipfile: how to find out if a zipinfo object is a directory or file
Message-ID: <3AAF3BCF.580F3ECC@t-online.de>

Hi,

does somebody know how to find out in a zipinfo object represents a file
or directory.
I have zipfiles created on Linux and in Win32 filesystem.
Or is this a planed feature for Py 2.1?

Thanks

 Stephane


From tim.hochberg at ieee.org  Thu Mar 15 10:57:02 2001
From: tim.hochberg at ieee.org (Tim Hochberg)
Date: Thu, 15 Mar 2001 15:57:02 GMT
Subject: reversing a dictionary (newbie)
References: <mailman.984650604.21153.python-list@python.org>
Message-ID: <iD5s6.28262$Xt3.4593144@news1.rdc1.az.home.com>

"Christopher Brewster" <C.Brewster at dcs.shef.ac.uk> wrote in message :

> The processing of the files is quite quick (700 odd files (@ 100k mean
size)
> of directory 'A' in about 20 minutes)
> but the reversing of the dictionary took hours and hours. I have run the
> program successfully on a PC (500 MHz, 256 RAM, running Windows NT) and I
am
> still waiting for results from a Sun (1 Giga RAM) since yesterday. Here is
> the code - what have I done wrong?

[Reversal function that Outlook mangled anyway]

It's not obvious (to me anyway) why this so slow. There is some extraneous
list munging going on, but I'd be suprised if that slowed things down that
much. In any event here's are two version that might be somewhat faster
(untested):

def revdict1(diction):
   "The long form"
   revdict = {}
   for wordpair in diction.keys():
      freq = diction[wordpair]
      if revdict.has_key(freq):
         revdict[freq].append(wordpair)
      else:
         revdict[freq] = [wordpair]
   return revdict

def revdict2(dict):
   "The short form (assumes python 2.0)"
   rev = {}
   for wp, freq in dict.items():
      rev.setdefault(freq, []).append(wp)
   return rev

Perhaps that'll help. Who knows?

-tim




From grante at visi.com  Wed Mar  7 12:24:36 2001
From: grante at visi.com (Grant Edwards)
Date: Wed, 07 Mar 2001 17:24:36 GMT
Subject: Can anyone recomend a good intoduction to C...
References: <mailman.983894288.19965.python-list@python.org> <983896344.701456@newsmaster-04.atnet.at> <m3lmqibw0a.fsf@atrus.jesus.cam.ac.uk> <983911749.217622@newsmaster-04.atnet.at> <UScp6.1097$y6.196347@ruti.visi.com> <983916572.286918@newsmaster-04.atnet.at> <9852b8015jm@news1.newsguy.com> <983967284.246700@newsmaster-04.atnet.at>
Message-ID: <o9up6.1291$y6.235469@ruti.visi.com>

In article <983967284.246700 at newsmaster-04.atnet.at>, Werner Schiendl wrote:

>> > That is of course a point, if you need to let the program
>> > inspect you will try to keep it as simple as possible. But
>> > given the same functionality, I think a C++ program will not be
>> > more complex.
>>
>> The program will not be, the language is.  It's a hard-to-judge
>> tradeoff.
>
>As to my knowledge, the application itself is inspected for
>safety critical appliances. And you must not change anything
>afterwards in the software or you need to re-inspect the
>software.
>
>What use would it be to check the language?

The point is that the more complex the language, the more
difficult it is to inspect the application.  You can look at a
few lines of C and have a pretty good chance at guessing what
they do.  In C++ it's much harder to look at snippet of code
and figure out what it's going to do.

-- 
Grant Edwards                   grante             Yow!  Somewhere in DOWNTOWN
                                  at               BURBANK a prostitute is
                               visi.com            OVERCOOKING a LAMB CHOP!!


From quickdry at users.sourceforge.net  Fri Mar 23 09:43:39 2001
From: quickdry at users.sourceforge.net (Steven Adams)
Date: Fri, 23 Mar 2001 14:43:39 GMT
Subject: tempfile problem on win98
References: <g3nu6.369$hb.2141@news1.belrs1.nsw.optushome.com.au> <99d0dg$os92$1@ID-11957.news.dfncis.de>
Message-ID: <viJu6.473$hb.2279@news1.belrs1.nsw.optushome.com.au>

"Emile van Sebille" <emile at fenx.com> wrote in message
news:99d0dg$os92$1 at ID-11957.news.dfncis.de...
> The mktemp routine guarantees to return a name unique in the target
> directory.  Are you sure you're using fully qualified path names
throughout
> when subsequently discovering duplicate files?
>
> Emile van Sebille
> emile at fenx.com

ugh, I feel dumb, thats exactly the problem Emile, the file was
getting an extension added and so it wasn't actually unique, is there
a way to have a suffix included in the calculation of the filename
instead of the prefix?

thanks for opening my eyes to the glaringly obvious :-)

Steven





From bernhard at abnoba.intevation.de  Wed Mar 28 10:04:24 2001
From: bernhard at abnoba.intevation.de (Bernhard Reiter)
Date: 28 Mar 2001 15:04:24 GMT
Subject: Linux GPS program
References: <3AC01637.4EF3A6F2@crynwr.com>
    <slrn9c17k4.cfc.jra@dorothy.msas.net> <3AC0D915.83C2C0C3@crynwr.com>
    <slrn9c1p9t.d3g.jra@dorothy.msas.net>
Message-ID: <99suho$m7a$2@newsserver.rrzn.uni-hannover.de>

In article <slrn9c1p9t.d3g.jra at dorothy.msas.net>,
	jra at dorothy.msas.net (Jay R. Ashworth) writes:
> This one time, in band camp,
>   Russell Nelson <nelson at crynwr.com> wrote:
>> "Jay R. Ashworth" wrote:

>> > Any plans to do vector map data as well?  
>> 
>> I'm pretty ignorant here.  What sources of vector information are there,
>> how big is the dataset, and how do I render it so I can use lat/lon or
>> easting/northing to get my location?
> 
> There are *stunning* amounts of vector data, all in different formats. 

I agree, though most proprietory.

> I'm trying to (get time cleared up to) sort some of that all out even
> as I type.

The shapefile format is used for a couple of things.
Check www.freegis.org for several libraries dealing with it
(and other gps packages, btw).
There is a python wrapper for the shapereading library.

	Bernhard
-- 
Professional Service around Free Software                (intevation.net)  
The FreeGIS Project                                         (freegis.org)
Association for a Free Informational Infrastructure            (ffii.org)
FSF Europe                                            	  (fsfeurope.org)


From tim.one at home.com  Sun Mar 25 04:10:47 2001
From: tim.one at home.com (Tim Peters)
Date: Sun, 25 Mar 2001 04:10:47 -0500
Subject: So what's wrong with __future__?  (Was Re: Why "from __future__" stinks ...)
In-Reply-To: <slrn9bdka2.3j6.hzhu@rocket.knowledgetrack.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCAEDKJIAA.tim.one@home.com>

[Huaiyu Zhu]
> I don't quite see what the complaint about __future__ is really about.
> So let me try to analyze a little bit.  Maybe this could encourage
> someone to write a PEP to clarify.
>
> What's wrong with import __future__?
> ...

Oh, I don't think it's hard to grasp the primary complaint:  it's a hack!
Nobody in their right mind would *expect* an import statement to have
profound effects on the syntax and/or semantics of the module containing the
import, and it's un-Pythonically surprising to abuse import statements this
way.

OTOH, it was the best hack we could dream up under the constraints that (a)
we had to dream up *something*; (b) it had to be 100% backward compatible, no
exceptions no way no how; and, (c) since it was late in the release cycle,
any form of new syntax was a non-starter (there was no conceivable way to
make time to adjust all the tools we ship, let alone pull that trick  with
such scant warning on other people supplying Python tools).

It will probably go away by the time 2.2 ships.  In the meantime, it should
do the job it was designed to do, for the three highly vocal people who are
afraid of nested_scopes damage they won't actually suffer <0.9 wink>.

you've-seen-__future__-on-c.l.py-more-than-you'll-ever-see-it-in-
    real-life-ly y'rs  - tim




From fluxent at yahoo.com  Fri Mar 30 11:12:47 2001
From: fluxent at yahoo.com (fluxent at yahoo.com)
Date: Fri, 30 Mar 2001 16:12:47 -0000
Subject: best practice? null/None vs 'None' vs '' (in web/sql)
Message-ID: <9a2b9v+lrgp@eGroups.com>

(Python 2.x, Win2K, IIS5)

I've got some simple pages for viewing and editing SQL records via 
Python in ASP.

I would really like to distinguish between fields with null values 
and fields that might actually include the text string 'None'. Right 
now both cases come out from a SELECT as showing up as 'None', and 
then when I submit and edit they definitely get changed to 'None', so 
I end up with no nulls.

I can see that the ODBC result set distinguishes them. But of course 
when things get rendered out to HTML you don't want to see single 
quotes around every string value, so 'None' and None become the same 
thing.

So, I guess this isn't a platform bug, it's a nasty area calling for 
some sort of interface standard. Is there a best practice I should 
adopt?

Maybe (a) transform the outputs of each SELECT to a different 
representation of None (either '' or 'null'?), then (b) transform all 
UPDATE query strings before executing them to change any ='' (or 
='null') to =NULL?





From robin at alldunn.com  Tue Mar 13 00:44:19 2001
From: robin at alldunn.com (Robin Dunn)
Date: Mon, 12 Mar 2001 21:44:19 -0800
Subject: Easiest way to get Berkley DB 3.x running on Windows
References: <SQ3r6.1961$NY6.31072@news3.oke.nextra.no>
Message-ID: <_sir6.57$XlD.26870131@news.randori.com>

"Thomas Weholt" <thomas at cintra.no> wrote in message
news:SQ3r6.1961$NY6.31072 at news3.oke.nextra.no...
> Hi,
>
> Do I have to compile Berkley DB to get an updated bsddb-module for Python
on
> win32?
> I want to try the latest version available with python-bindings, cuz there
> seems to be a memory leak
> and a problem with deadloc(k)s in the one in the current Python lib.
>
> Thanks for any pointers.
>

http://pybsddb.sourceforge.net/

Binaries for Windows are available.


--
Robin Dunn
Software Craftsman
robin at AllDunn.com       Java give you jitters?
http://wxPython.org      Relax with wxPython!





From nas at arctrix.com  Thu Mar 22 10:54:41 2001
From: nas at arctrix.com (Neil Schemenauer)
Date: Thu, 22 Mar 2001 07:54:41 -0800
Subject: Jython is cool?
In-Reply-To: <20010322103954.A18539@sync.nyct.net>; from mbac@nyct.net on Thu, Mar 22, 2001 at 10:39:54AM -0500
References: <20010322103954.A18539@sync.nyct.net>
Message-ID: <20010322075441.A26566@glacier.fnational.com>

On Thu, Mar 22, 2001 at 10:39:54AM -0500, Michael Bacarella wrote:
> 	Has someone already done this/is it already possible and I just
> didn't look at the package well enough or am I out of luck until I
> bite the bullet some day and implement it myself?

AFAIK, its possible to make Jython applets.  The problem is that
the whole Jython runtime has to be downloaded to the client.

I'm hoping that the Mozilla guys catch the Python bug based on
what Active State has been doing with Komoto and adds "<script
language=python>" to Mozilla.

  Neil



From gerhard.nospam at bigfoot.de  Sat Mar  3 01:22:07 2001
From: gerhard.nospam at bigfoot.de (Gerhard =?iso-8859-1?Q?H=E4ring?=)
Date: Sat, 03 Mar 2001 07:22:07 +0100
Subject: cataloging words in text file
References: <3AA0124E.74A39783@comm.mot.com>
Message-ID: <3AA08D8F.255748A0@bigfoot.de>

Stephen Boulet wrote:
> 
> I remember this homework assignment for my data structures (c++)
> class: read in a large file, and create a data structure containing
> every word in the file and the number of times it appears.

Funny. This was the task I had to code in Java for a recent job
application.

Gerhard
-- 
Sorry for the fake email, please use the real one below to reply.
contact: g e r h a r d @ b i g f o o t . d e
web:     http://highqualdev.com


From kevin_cazabon at hotmail.nospamplease!.com  Thu Mar  8 22:34:07 2001
From: kevin_cazabon at hotmail.nospamplease!.com (Kevin Cazabon)
Date: Thu, 8 Mar 2001 20:34:07 -0700
Subject: from Tkinter import *
References: <mailman.983920449.5230.python-list@python.org>
Message-ID: <YjzUJnEqAHA.188@typhoon.i3s.net>

The screwey thing about Fred's document using "from Tkinter import *" is
that there is an "Image" method in Tkinter that totally mucks up the PIL
Image module if you did an "import Image" prior to this...

Seeing as both are in Fred's court, it's kinda a bad idea...

I've battled the problem a couple times (but have learned my lesson) when
getting method errors for Image... really messed me up the first time.

Kevin Cazabon.



"Timothy Grant" <tjg at exceptionalminds.com> wrote in message
news:mailman.983920449.5230.python-list at python.org...
| Hi all,
|
| With all the discussion around here recently about the
|
| from foo import *
|
| idiom, I thought I'd ask about Tkinter. All of Fredrik's
| examples in his guide on Tkinter use
|
| from Tkinter import *
|
| Now, I'm aware that the effbot is never wrong about
| anything<wink>, but is it safe to assume that I should not be
| using the above idiom, but the import Tkinter idiom instead?
|
|
| --
| Stand Fast,
|     tjg.
|
| Timothy Grant                         tjg at exceptionalminds.com
| Red Hat Certified Engineer            www.exceptionalminds.com
| Avalon Technology Group, Inc.         <><       (503) 246-3630
| >>>>>>>>>>>>>Linux, because rebooting is *NOT* normal<<<<<<<<<
| >>>>This machine was last rebooted:  49 days  3:27 hours ago<<
|




From spam at spam.com  Sat Mar 31 01:09:36 2001
From: spam at spam.com (Penfold)
Date: Sat, 31 Mar 2001 07:09:36 +0100
Subject: Fetch object in imaplib
References: <3ac47845$0$30795$456d72a3@news.skynet.be>
Message-ID: <tcatb0lri88n95@corp.supernews.com>

"alphatrooper" <alphatrooper at skynet.be> wrote in message
news:3ac47845$0$30795$456d72a3 at news.skynet.be...
> I'm a bit stuck with the fecth object from the iamplib .
> in the lib ref they mention :
(snipped)
> Now my question is : what are all the possibilities of "message part
> names", I looked in the rfc's and on the net but can't find anything
> about it.

If you did really click through on the RFC link in the documentation
(http://www.ietf.org/rfc/rfc2060.txt)
you would find a large document describing the IMAP4 protocol.  It defines
explicitly for you the very many options you can pass to an IMAP FETCH
command.

So somehow I either dont believe you could find nothing on it, or I dont
believe
you tried particularly hard ;-)






From dsh8290 at rit.edu  Wed Mar  7 17:46:46 2001
From: dsh8290 at rit.edu (D-Man)
Date: Wed, 7 Mar 2001 17:46:46 -0500
Subject: I come to praise .join, not to bury it...
In-Reply-To: <Ntyp6.1479$y6.244056@ruti.visi.com>; from grante@visi.com on Wed, Mar 07, 2001 at 10:19:25PM +0000
References: <kuofvfjb77.fsf@lasipalatsi.fi> <000001c0a64c$497df250$a9807018@cr706570a> <mailman.983997159.4842.python-list@python.org> <Ntyp6.1479$y6.244056@ruti.visi.com>
Message-ID: <20010307174646.A8@harmony.cs.rit.edu>

On Wed, Mar 07, 2001 at 10:19:25PM +0000, Grant Edwards wrote:
| In article <mailman.983997159.4842.python-list at python.org>, D-Man wrote:
| 
| >Even with my newfound understanding, provided by the Martellibot, I
| >agree with this.  Calling a method on an object is fine.  Calling it
| >on a literal looks weird.
| 
| What makes you think a "literal" is not an object?

Literals are special syntax for creating objects.  So they are
objects, they just look different.  I guess a better way of expressing
it is "Calling a method on a named object is fine." except it doesn't
cover objects returned by a ctor that aren't named yet.

(just for an example)
>>> from UserString import UserString
>>> UserString( "." ).join( ['a' , 'b' , 'c' ] )

-D




From pearu at cens.ioc.ee  Sat Mar 10 20:13:31 2001
From: pearu at cens.ioc.ee (Pearu Peterson)
Date: Sun, 11 Mar 2001 03:13:31 +0200
Subject: Can anyone recomend a good intoduction to C...
In-Reply-To: <98eamb02ug2@news1.newsguy.com>
References: <mailman.983894288.19965.python-list@python.org> <ewwp6.1466$y6.241005@ruti.visi.com> <slrn9aeblb.cg1.scarblac@pino.selwerd.nl> <djagat4kk3cs9hg0c3k39cudk6goblhkg6@4ax.com> <989mbk$24nuv$1@fido.engr.sgi.com> <slrn9ah66m.k09.qrczak@qrnik.zagroda> <98auhn0270j@news2.newsguy.com> <slrn9aiaq7.rmu.qrczak@qrnik.zagroda> <98dpoe12ckn@news1.newsguy.com> <Pine.LNX.4.21.0103102137300.7964-100000@kev.ioc.ee> <98eamb02ug2@news1.newsguy.com>
Message-ID: <Pine.LNX.4.21.0103110251240.8954-100000@kev.ioc.ee>

On Sat, 10 Mar 2001, Alex Martelli wrote:

> Glad to see somebody seconding my opinion -- particularly somebody
> who was so adamantly set _against_ C++ before the Boost experience
> (remember how astonished you were that I even _considered_ C++ for
> gmpy...?  

(Not because I was so against C++, I just was more used to Python, C,
Fortran; and these rumors about C++...;).

> It would be done by now, and well on its way to a second
> and richer release, if I had chosen C++ rather than C...:-).

I am now creating an automatic interface generator to GiNaC
(www.ginac.de) library. GiNaC uses CLN
(http://clisp.cons.org/~haible/packages-cln.html) number classes.
CLN itself can be built to use GMP and it is much more complete
library than GMP. So, if you have any intention to use C++ for wrapping
MP numbers to Python, I would suggest CLN.

Regards,
	Pearu



From tim.one at home.com  Sat Mar 31 22:06:55 2001
From: tim.one at home.com (Tim Peters)
Date: Sat, 31 Mar 2001 22:06:55 -0500
Subject: PythonLabs Team Moves to ActiveState
In-Reply-To: <3AC68B5A.E2A41D0E@NoNoSpammytheworld.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCIEIOJJAA.tim.one@home.com>

[David Lees]
> Could someone explain exactly what this means?  I read over PEP 228 and
> am wondering if this is what you are referring to below with respect to
> the 'Perl numerical model'?
>
> I assume Python will still keep long integers, integers and complex.
> All that will change is coersion to float on the division of integers?
>
> Would it be possible to clarify this change?

I sincerely doubt it.

you're-welcome-ly y'rs  - tim




From dev_null at email.com  Sat Mar  3 12:19:50 2001
From: dev_null at email.com (Dev_NuLL)
Date: Sat, 03 Mar 2001 17:19:50 GMT
Subject: Intersesting Questions involving extending and embedding
References: <3a9c6f89.9613214@news.digital.net> <0D1n6.15927$Qb7.2600092@newsb.telia.net> <3a9f0af7.37387355@news.digital.net> <508n6.15966$Qb7.2607706@newsb.telia.net> <3a9dc27e.1325053@news.digital.net> <SYln6.16080$Qb7.2629142@newsb.telia.net> <3a9f00e8.2826131@news.digital.net> <A7In6.16279$Qb7.2658942@newsb.telia.net>
Message-ID: <3aa125e1.2738901@news.digital.net>

Damn, your dense
Who the hell said anything about compiling Python code in C ?? For the
last time I want to access FUNCTIONS in another Python module (which
happens to be inside the Blender executable) from within my module.
others have done it, but they won't release the source code.
BTW, you can execute Python code in C modules:

PyRun_SimpleString("print 'Hello Python World'");

>- If you're writing C or C++ code, you're *extending* Python.

Or embedding can be done in C (maybe you should read the docs)

>Read the docs again, and look as the samples shipped with the
>source distribution.  

Like I said, the examples in the docs are too Mickey Mouse.

Please, don't waste your time replying if your post are as usless as
the previous ones were. I can't believe a person can not even get a
simple answer to a simple question. Thank god for the kind sould who
pointed mt to starship.


From jlowe at mentos.urbana.css.mot.com  Wed Mar 14 17:52:05 2001
From: jlowe at mentos.urbana.css.mot.com (Jason Lowe)
Date: 14 Mar 2001 22:52:05 GMT
Subject: Python, threads, and signals (oh my!)
References: <98jqeh$6ut$1@newshost.mot.com> <98oh0j$tik$1@nntp6.u.washington.edu>
Message-ID: <98osml$sul$1@newshost.mot.com>

Donn Cave <donn at u.washington.edu> writes:

>I ran some experiments on other UNIX platforms, and Solaris is not
>alone if signals aren't handled well when there are multiple concurrent
>threads.  In general, signals were trapped but Python handler functions
>weren't executed.

I am also seeing unstable situations on Linux with threading and
signals, but I haven't looked into those problems as much as with the
Solaris case.

>My take is that it's complicated.  Does it make sense to split out
>readline's special problems as an issue that could be addressed
>separately?  In my experiments, without readline, there were still
>these issues, and even without readline, signal handling at the
>interpreter prompt differed from an interpreted disk file.

I don't think the readline module issue with threads is separate,
although there might be ways to make the readline module more robust
in the presence of threading without addressing the whole
threads/signals issue.  Specifically, the readline module is using
longjmp() to get back into the call_readline() context.  If the wrong
thread handles the SIGINT signal and performs the longjmp(), we're
hosed on many thread implementations.

>I can't think of any reason why it would hurt anything to block
>signals by default per non-main thread.  I don't think it will
>work for many platforms, but if it works for anyone's I guess
>that's better than nothing.  (And the one I would be most concerned
>about, BeOS, doesn't seem to need it, as signals seem to work more
>or less as intended in multithreaded programs there.)

Agreed.  The POSIX thread books I've seen all indicate that you want
this setup anyway.  It appears the only way to safely handle signals and 
threads is to make sure all asynchronous signals are delivered to only
one thread.  They even go so far as to recommend a separate thread whose
only purpose is to catch signals synchronously with sigwait().  The
signal-handling thread can then dispatch work to other threads using
condition variables, mutexes, and other thread
communication/synchronization primitives.

>I think the ability to block signals in general should be added,
>in the form of the POSIX 1003.1 sigprocmask function.  It would
>be easy enough to do, if anyone cared enough.  Is that how you
>masked off the signal from the new thread - do new pthreads copy
>their signal handling from the parent at creation?

I didn't use sigprocmask(), as I'm not sure that's technically portable
in a POSIX thread environment.  Instead, I used pthread_sigmask(),
although I suspect sigprocmask() == pthread_sigmask() when threads are
used on most platforms.

New pthreads do indeed inherit their signal mask from the thread that
creates them.  I masked off all signals before creating the new thread,
then restored them after the new thread was created.  Therefore, if the
main thread creates the new thread, the main thread will have its
original signal mask restored but the new thread will have all signals
blocked.  If another Python thread creates a thread, all signals for
both threads will be blocked.

The ability to control signal masks from Python would be great, although 
I suspect this has portability concerns leading to its current
omission from the standard library. If signal mask controls aren't 
added to the Python standard library, it would be nice if Python could
create new threads with all signals masked on platforms where this is 
appropriate, so signals are only handled by one thread (the main thread
or a signal-handling thread).  This would be desirable on any platform
that uses POSIX threads, in accordance with the POSIX thread
documentation.

Thoughts?

Jason Lowe
--
Jason Lowe                                    Urbana Design Center
Motorola Personal Communications Sector       1800 South Oak Street
jlowe at urbana.css.mot.com                      Champaign, IL  61820-6947
                                              (217) 384-8513


From mickeyz at qwest.net  Thu Mar  1 23:37:56 2001
From: mickeyz at qwest.net (Randy A. Ynchausti)
Date: Thu, 1 Mar 2001 21:37:56 -0700
Subject: Collection interfaces
References: <3a90fa97.0@news1.mweb.co.za> <3A933830.7F98D118@ix.netcom.com> <CgJk6.515$v77.66568@newsread2.prod.itd.earthlink.net> <220220010813598042%cmh@bDistributed.com> <ud7cakvi8.fsf@roundpoint.com> <MPG.150312ceb68f1ada98969f@news.earthlink.net> <8Gem6.473353$U46.14105062@news1.sttls1.wa.home.com> <MPG.1503548620e632919896a4@news.earthlink.net> <MCjm6.474652$U46.14163288@news1.sttls1.wa.home.com> <MPG.150394195f7f04a19896a7@news.earthlink.net> <3A9A86B1.9A8C69A2@ix.netcom.com> <MPG.1504d14563dce6499896af@news.earthlink.net> <97fjla128ir@news2.newsguy.com> <MPG.1505a76831c34b749896b3@news.earthlink.net> <97h5s40vhc@news2.newsguy.com> <MPG.1506007591cf818b9896bc@news.earthlink.net> <slrn99opop.gr.amk@mira.erols.com> <MPG.1506265ebf6d934d9896be@news.earthlink.net> <97ibmc0k47@news1.newsguy.com> <MPG.1506cf3293b094b49896c1@news.earthlink.net> <97m0c7080c@news2.newsguy.com>
Message-ID: <znFn6.4962$aH2.599554@news.uswest.net>

Alex,

> Been there, done that, more or less [snip ] -- after all, I _have_ been
> programming for well over a quarter of a century, in a
> reasonably vast variety of paradigms.  From the jump table
> of well-constructed, speedy machine-code, to the SQL Table
> of relational-DB programming, I've done more tables than
> the average Parisian cafe has on the sidewalks on a bright
> and warm spring day...

I enjoyed that Parisian cafe line -- what superb imagery!

But I don't think he will get the underlying meat.

Great line though...

Regards,

Randy




From mcalla at home.com  Wed Mar 21 21:36:32 2001
From: mcalla at home.com (Mike Callahan)
Date: Thu, 22 Mar 2001 02:36:32 GMT
Subject: Catching exceptions in Tkinter's mainloop
References: <%dws6.9947$Q47.2723350@news1.rdc1.tn.home.com> <drMs6.32482$HR6.3694604@nnrp4.clara.net> <owzt6.21061$Q47.5935326@news1.rdc1.tn.home.com> <6QNt6.37618$g63.4509118@nnrp3.clara.net>
Message-ID: <Qydu6.26214$Q47.7466405@news1.rdc1.tn.home.com>

Richard,

On my system, the exception is captured and printed in test.log but mainloop
is not exited. Is this what happens on your system?

Mike C.

"Richard Townsend" <NOstarfighterSPAM at freeuk.com> wrote in message
news:6QNt6.37618$g63.4509118 at nnrp3.clara.net...
> "Mike Callahan" <mcalla at home.com> wrote in message
> news:owzt6.21061$Q47.5935326 at news1.rdc1.tn.home.com...
> > I tried the following with no luck:
> >
>
> > import Tkinter as T
> > import sys
> >
> > class Gui:
> >     def __init__(self, master):
> >         T.Button(master, text='Crash2', command=self.crash).pack()
> >
> >     def crash(self):
> >         z = zz
> >
> > class Catcher:
> >     def __init__(self, func, subst, widget):
> >         self.func = func
> >         self.subst = subst
> >         self.widget = widget
> >
> >     def __call__(self, *args):
> >         try:
> >             if self.subst:
> >                 args = apply(self.subst, args)
> >             return apply(self.func, args)
> >         except SystemExit, msg:
> >             raise SystemExit, msg
> >         except:
> >             import traceback
> >             traceback.print_exc()
> >             self.widget.quit()
> >
> > root = T.Tk()
> > f = open('test.log', 'w')
> > sys.stderr = f
> > T.CallWrapper = Catcher
> > app = Gui(root)
> > root.mainloop()
> >
> > What am I doing wrong?
> >
>
> I've just tried your code and it wrote the following to the 'test.log'
file:
>
> Traceback (most recent call last):
>   File "catcher.py", line 21, in __call__
>     return apply(self.func, args)
>   File "catcher.py", line 9, in crash
>     z = zz
> NameError: There is no variable named 'zz'
>
>
> Is that not what you wanted ??
>
> Richard
>
>
>




From uche at ogbuji.net  Thu Mar 22 17:19:14 2001
From: uche at ogbuji.net (Uche Ogbuji)
Date: Thu, 22 Mar 2001 22:19:14 GMT
Subject: XLink/XPointer processing in Python
References: <mailman.985163671.1382.python-list@python.org>
Message-ID: <3ABA7A63.7B6FEDAF@ogbuji.net>

Lars von Wedel wrote:

> I want to build an XML processing app which shall be
> able to process "simple" links according to the XLink
> spec (using SAX).
> 
> I haven't found any information on how these are handled
> in the processing application, so I assume, I must handle
> the xlink:href attribute content on 'my' own? Are there
> any reusable pieces of software to do so?

4Suite (http://4Suite.org) contains an XLink/XPointer processor that
handles most of those specs, and certainly simple links.


-- 
Uche Ogbuji
Personal:   uche at ogbuji.net		http://uche.ogbuji.net
Work:       uche.ogbuji at fourthought.com	http://Fourthought.com


From dwig at advancedmp.net  Wed Mar  7 19:21:48 2001
From: dwig at advancedmp.net (Don Dwiggins)
Date: 07 Mar 2001 16:21:48 -0800
Subject: Can Python replace Visual Basic? Should it?
In-Reply-To: "Brad Bollenbach"'s message of "Tue, 06 Mar 2001 04:23:29 GMT"
References: <5DZo6.11126$hn5.1578410@news1.rdc1.mb.home.com>
Message-ID: <u8zmhyusj.fsf@advancedmp.net>

Brad Bollenbach writes:
>     Can Python replace Visual Basic? Should it?

> But really, very specifically, were YOU a former Visual Basic (or
> Powerbuilder, or Delphi) programmer that is now using Python to write the
> complex GUI's VB used to handle for you? I'm an on again, off again, student
> of Python, who really wants to give this a go, if there's a good chance that
> I'll get the return on my investment: a programming tool that can take over
> where VB, Powerbuilder, and Delphi used to play.

Funny you should ask.  In our tiny development team, we've been using VB
(and VBScript) exclusively, until recently.  Being a polyglot by nature and
experience, and reacting to VB's habit of running out of power while getting
very complex, I've been using Python for prototyping for a month or so.
Now, I've been asked to share what I've learned, including helping the other
guys learn the language.  Frankly, we could have done everything in our web
site in Python (except for client-side Javascript) and been better off for
it, for many reasons.  Indeed, we may evolve in that direction.

I'd say go for it -- get the Python on Windows book and convert one of your
VB, Powerbuilder, or Delphi projects to Python and see what you think (or if
you're adventurous, just use the documentation included in the
distribution).  I converted one of our ASP pages to Python from VBScript to
see if it would be faster (it wasn't), and was pleasantly surprised at how
easy it was.

I should say, though, that we haven't used any of the GUI toolkits -- maybe
someone else can recommend the best one to use when migrating from VB Forms.

-- 
Don Dwiggins                    "Solvitur Ambulando"
Advanced MP Technology
dwig at advancedmp.net




From oliver at centralnet.ch  Sat Mar 24 10:01:08 2001
From: oliver at centralnet.ch (Oliver Marugg)
Date: Sat, 24 Mar 2001 16:01:08 +0100
Subject: File in a directory
References: <3ABB6E13.403B963B@marugg.net> <AHKu6.3195$4N4.475205@newsc.telia.net>
Message-ID: <3ABCB6B3.4E60F9D0@centralnet.ch>

Fredrik,

Thank you that was the glue I needed... now it works.

Oliver

Fredrik Lundh wrote:

> Oliver Marugg wrote:
> > Has anyone an idea how to solve that with the size or timestamp
> > verification?
>
>   os.listdir(dir) => list of files in directory dir
>   os.path.getsize(file) => size of file
>   os.path.getmtime(file) => time of last modification (in seconds since epoch)
>   time.time() => current time (in seconds since epoch)
> + five to ten minutes of your time
> -----------------------------
> = solution
>
> reference info here:
> http://www.python.org/doc/current/lib/module-os.html
> http://www.python.org/doc/current/lib/module-os.path.html
> http://www.python.org/doc/current/lib/module-time.html
>
> Cheers /F



From topmind at technologist.com  Sat Mar 24 19:33:06 2001
From: topmind at technologist.com (topmind)
Date: Sun, 25 Mar 2001 00:33:06 GMT
Subject: Collection interfaces
References: <3a90fa97.0@news1.mweb.co.za>   <MPG.150394195f7f04a19896a7@news.earthlink.net> <3A9A86B1.9A8C69A2@ix.netcom.com> <MPG.1504d14563dce6499896af@news.earthlink.net> <97fjla128ir@news2.newsguy.com> <MPG.1505a76831c34b749896b3@news.earthlink.net> <97h5s40vhc@news2.newsguy.com> <MPG.1506007591cf818b9896bc@news.earthlink.net> <slrn99opop.gr.amk@mira.erols.com> <MPG.1506265ebf6d934d9896be@news.earthlink.net> <97ibmc0k47@news1.newsguy.com> <MPG.1506cf3293b094b49896c1@news.earthlink.net> <97m0c7080c@news2.newsguy.com> <MPG.15085517debf4b649896ca@news.earthlink.net> <zb%n6.31033$o%6.586026@news1.rdc1.bc.home.com> <MPG.150b383cea226d409896d0@news.earthlink.net> <kNko6.6986$eq3.707121@news.uswest.net> <MPG.150c46db9cec283e9896d4@news.earthlink.net> <3AA2A898.450B659E@mail.com> <gRCt6.12224$Im6.1308560@newsread1.prod.itd.earthlink.net> <3AB755F8.67E84126@mail.com> <%a%t6.17245$227.1601569@newsread2.prod.itd.earthlink.net> <3AB8A730.4BE13D22@mail.com>
Message-ID: <61bv6.8693$P4.676800@newsread1.prod.itd.earthlink.net>

"James A. Robertson" <jarober at mail.com> wrote in message
news:3AB8A730.4BE13D22 at mail.com...
> topmind wrote:
> >
>
> > > http://www.cincom.com/smalltalk
> > >
> >
> > I only see a front page, not a parsing example.
> >
>
> Follow the tutorial link - it's right there on that page.
>


I cannot find any "tutorial" link.

Perhaps you are afraid that if you give me the full tutorial link, you
will have to expose to the world the FACT that the site runs on
Microsoft ASP VSscript instead of Smalltalk?

Do Chevy salesmen drive Fords?

-tmind-






From cce at clarkevans.com  Tue Mar 20 20:49:30 2001
From: cce at clarkevans.com (Clark C. Evans)
Date: Tue, 20 Mar 2001 20:49:30 -0500 (EST)
Subject: Yet Another PEP:  Query Protocol Interface or __query__
In-Reply-To: <998r4j02vmb@news2.newsguy.com>
References: <Pine.LNX.4.21.0103201837570.21343-100000@clarkevans.com> <998r4j02vmb@news2.newsguy.com>
Message-ID: <Pine.LNX.4.21.0103202032080.14634-100000@clarkevans.com>

On Wed, 21 Mar 2001, Alex Martelli wrote:
> > Please have a look at the PEP below, it is intended to be a
> > light-weight solution to "typing" based on programmer intent.
> 
> At least when read at 1 AM -- sounds cool.  Further, the
> builtin query function could support _adding interfaces to
> an object post-facto, non-invasively_, much like the
> 'typeclass'-based ideas that were floating in my mind
> recently and surfaced in some of my recent posts. 

Right.

> I.e., the query() builtin would both try getting the interface
> from the _object_, by passing that interface to the obj's
> query method, AND also try doing it the other way
> 'round -- asking the interface if it knows how to wrap
> the object so it (the interface) gets implemented. 

That is a very cool idea!  And the reverse lookup could
even be added much later as a seperate PEP.  I was trying
to stay simple and to de-couple this PEP as much as possible
from the other possibilities being tossed about in the
type special interest group.  I'd hate to have a dependency
emerge when one really isn't necessary.

> That would require interface-identifiers to be richer objects
> than just string, or a registry which associates a string
> identifying an interface with such a richer object.

I'd go for a registry, as this keeps the coupling low so
that the __query__ can be delivered without requiring 
a specific interface class to be worked out and agreed upon.
A string is good enough for many purposes... 

Of course, I love the idea of having an interface class
with the check() method as proposed by Tim Hochberg 
and refined by by Paul Prescod and Tim Peters.
This same class could also have a wrap(object) as
you suggest.  And then be linked somehow to the 
__query__ identifier.  How cool.  

You've taken this far past what I had visioned.  
I was just trying to solve a simple problem; how 
do I get an object having protocol X from an
object Y.  Right now, there is no general way
to do this and I feel adding built-in interface
getters like __iter__ is not the best long term
solution.

> Since 'interface' is so deucedly overloaded a word, maybe
> 'protocol' might be be better -- so we only get confused
> with FTP, HTTP and the like:-).

Hard one. 

Best,

Clark




From twestley at buffalo.veridian.com  Thu Mar  1 14:36:58 2001
From: twestley at buffalo.veridian.com (Terry Westley)
Date: Thu, 1 Mar 2001 14:36:58 -0500
Subject: [ANNOUNCE] pyAda, first public release
References: <3A9DB9E1.8DD0E36D@bigfoot.de>
Message-ID: <wxxn6.2168$434.251256@news.uswest.net>

Gerhard H?ring <gerhard.nospam at bigfoot.de> wrote in message
news:3A9DB9E1.8DD0E36D at bigfoot.de...
> I am glad that I can finally announce a first version of pyAda. As the
> name implies, it's an Ada 95 binding to the Python interpreter. With
> pyAda, you can write Python extensions in Ada 95, and, much more
> important, you can also embed a Python interpreter in your Ada
> applications. The current release is far from finished, but the thin
> binding works already quite well for me. The thin binding is a
> one-to-one mapping of the Python -> C interface.
>
> website:  http://pyada.sourceforge.net
> release:  http://download.sourceforge.net/pyada/pyAda-2001-03-01.tar.gz
> docs:     included in release

Congratulations.  Certainly Python is getting a lot of attention and
to have an Ada binding is great.

If/when you want to create a thick binding, take a look at the latest
version of TASH and its thick binding for some ideas.  You can
either download and try it from http://www.adatcl.com or read the
thick binding tutorial that I've started on the mailing list:
http://groups.yahoo.com/group/tash.

P.S. Thanks for the reference in your Comparison page to TASH!

--
Terry Westley
Veridian Engineering
twestley at buffalo.veridian.com





From paulp at ActiveState.com  Sat Mar 10 00:34:47 2001
From: paulp at ActiveState.com (Paul Prescod)
Date: Fri, 09 Mar 2001 21:34:47 -0800
Subject: tkinter, activepython, win32
References: <OHiq6.4422$tP3.71012@news1.rdc1.bc.home.com>
Message-ID: <3AA9BCF7.C60C56C4@ActiveState.com>

Luke wrote:
> 
> python newbie!
> cannot get tkinter programs to run...
> 
> did 'normal' install of activepython 2.0 build 203
> installed Tcl/Tk 8.3 for Windows, Binary Distribution

I would suggest you use the package manager to get tk and try again.
Just start the package manager and type "install tk". Or else you could
have selected "install tk" when you installed ActivePython.

What Tk app would you like to run? Also, how are you trying to run it?

> am i missing something here?
> 
> wxpython works fine, should i stick with it?

Either of them should work fine so you shouldn't choose between them
based on that.

-- 
Python:
    Programming the way
    Guido
    indented it.



From sholden at holdenweb.com  Mon Mar  5 00:52:40 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Mon, 5 Mar 2001 00:52:40 -0500
Subject: Overriding UserList methods
References: <b566atck0ruu4tae8n1srrkntbmme26gor@4ax.com>
Message-ID: <aUFo6.14917$ML1.731870@e420r-atl2.usenetserver.com>

"Daniel Klein" <danielk at aracnet.com> wrote in message
news:b566atck0ruu4tae8n1srrkntbmme26gor at 4ax.com...
> I must be missing something, cos I can't figure out how to override a
method
> using the UserList module. Here is some code:
>
> import UserList
> class MyList(UserList.UserList):
>       __def __init__(self):
>         UserList.UserList.__init__(self)
>
While this would work, I don't see it's strictly necessary. Why not just
inherit the __init__ method from UserList?

>     def append(self, object):
>         self.append(object)
>
> Understandably, this goes into a recursive frenzy until it hits the stack
> limit. :-(
>
Well, if you want *you* append method to make use of UserList's method you
would have to write

    def append(self, object):
        UserList.append(self, object)

Since all you are doing is reflecting the underlying UserList's append()
method, however, it would seem to make more sense to simply allow append()
to be inherited as well!

> The only way I could get this to work was to give the method another name,
ie
>
>     def add(self,object):   ## This works but I would rather use 'append'
>         self.append(object)   ## as the method name.
>
> So how can I override the append() method so that I can add my specialized
code
> to it?
>
> Daniel Klein

The clue is to use the unbound method from the superclass, as I showed you
above.

Another way would be to create a UserList object as an instance variable,
and then delegate methods to the UserList which you don't recognize, but
this might be going too far.  Hope this helps.

regards
 Steve





From aleaxit at yahoo.com  Mon Mar 19 07:35:37 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Mon, 19 Mar 2001 13:35:37 +0100
Subject: stripping punctuation from tuples
References: <3AB57F0C.594CF3E9@northwestern.edu>
Message-ID: <994uem027sm@news1.newsguy.com>

"Louis Luangkesorn" <lluang at northwestern.edu> wrote in message
news:3AB57F0C.594CF3E9 at northwestern.edu...
> How do you strip the punctuation from tuples?  In particular, say I have
> a tuple a[10] (which happened to come from a not very well designed
> database)
>
> >>> a[10]
> ('Alan Tsang',)

Note that, here, there is in fact no punctuation in the tuple a[10].
The parentheses and comma[s] are inserted by 'repr' (which is being
implicitly called by the interactive interpreter to show you the
results of expressions you type interactively).


> How do I get something to say 'Alan Tsang'  with out the parens or the
> ','

a[10][0], the first and only element of the tuple, has this value.


> Or, what I'm really trying to do is make this so
> >>>b = breakname(a[10])
> >>>b[0], b[1]
>
> 'Alan','Tsang'

If you know for sure that the tuple has exactly one item (or you
know it has at least one, and only care about the first one),

def breakname(atup):
    return atup[0].split()

To work for any number of items in the tuple, e.g.:

def breaknames(atup):
    return [ x for item in atup for x in item.split() ]

Of course, if you use repr (implicitly or explicitly) to view
the results of either function, you will see punctuation again,
since repr will insert it for clarity:-).


Alex






From jkraska1 at san.rr.com  Sun Mar 18 23:34:23 2001
From: jkraska1 at san.rr.com (Courageous)
Date: Mon, 19 Mar 2001 04:34:23 GMT
Subject: Wrong reasons to enhancing the language
References: <mailman.984971990.29857.python-list@python.org>
Message-ID: <d73bbt8otnhhqdfitvfof6d96s09s5btn7@4ax.com>

>In my opinion, none of the above reasons is good enough to justify alone 
>changes to the language. There are other criteria that need to be met. Are 
>the extensions useful to many people? Are they generic and extensible? Do 
>they cause any compatibility issue?

Do they adhere to the principles which make Python uniquely Python?

Sometimes I feel it's as if the implementors or the users just need to add
features in order to have stuff to do and things to learn. I, however, prefer
to not do any learning with my computer programming language. I want
to use my language to CREATE THINGS; the language is an incidental
tool on the path to achieving some other objective.

That's one of the reasons I like Python so much. It's a demure language,
always soft-spoken and not under foot.

C//



From aleaxit at yahoo.com  Thu Mar  1 10:17:04 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 1 Mar 2001 16:17:04 +0100
Subject: Py_INCREF / DECREF not needed in simple C functions?
References: <3a9e5301.1251797@news.newsguy.com>
Message-ID: <97lp5g02uun@news2.newsguy.com>

"Owen F. Ransen" <ransen_spam_me_not at nemo.it> wrote in message
news:3a9e5301.1251797 at news.newsguy.com...
> Trundling along here and reading the docs I've come
> to the conclusion that if I pass simple values to
> my C extensions to Python I do not need to worry
> about Py_INCREF and Py_DECREF.
>
> Am I right? For example if I only pass, say, 3
> double values to a C extension function, the values
> of which I extract using PyArgs_ParseTuple, then
> I do not need to worry about reference counts?
    [snip]
>     return (Py_None) ;

Actually, you _should_ incref None if you want to
return it -- or, use
    return Py_BuildValue("");
to return None, and then you can keep not worrying:-).


Alex





From bobhicks at nospam.adelphia.net  Tue Mar 27 09:54:56 2001
From: bobhicks at nospam.adelphia.net (Robert Hicks)
Date: Tue, 27 Mar 2001 14:54:56 GMT
Subject: python, OS X, and fcntl
References: <mailman.985641394.29795.python-list@python.org>
Message-ID: <bobhicks-2703010954550001@va-frontroyal1a-202.wch.adelphia.net>

http://www.zope.org/Members/jshell/ZopeOnOSX

This goes into exactly what you are asking about...

-- 
Bob

Remove: the nospam. in my address if you want to email me


From db3l at fitlinxx.com  Wed Mar  7 14:09:39 2001
From: db3l at fitlinxx.com (David Bolen)
Date: 07 Mar 2001 14:09:39 -0500
Subject: Name of the Script
References: <20010304.185354.2114937732.8277@localhost.localdomain> <NJCo6.6720$qv3.2834377@nnrp5-w.sbc.net> <FpmjOjvwe5bm0Bg09LDTMPhh5vji@4ax.com> <Pine.LNX.4.21.0103051715150.18263-100000@kev.ioc.ee> <cbKjOqPRwNza00jOm7A1HBDkRYj7@4ax.com>
Message-ID: <upuft9z0s.fsf@ctwd0143.fitlinxx.com>

Paul Moore <paul.moore at uk.origin-it.com> writes:

> OK, so you could say "don't do that". But if I'm looking at some code
> deep in a big program, can I be *sure* that no-one has changed
> directory on me?

You can't, but then again you really shouldn't be looking at your
startup arguments (IMHO) inside code deep in a big program.  That's
the sort of thing that should be cleanly isolated to startup
processing, although some of the processing can be deferred to other
functions if necessary (e.g, sort of the way the X library did
pre-processing on its known arguments in advance of local code).

> Interestingly, although the __name__ symbol has the value __main__ in
> a script being run as a "main program", the __file__ symbol doesn't
> exist. That's a pity, as it would probably be what I wanted...

Right - if you're being imported you can reference __file__, but a top
level script really only has argv[0], I believe.

But argv[0] always references the path used to start the script which,
at the point of startup, should always be a path that you can use to
reference the same location.

I'd just suggest that if you need this sort of thing later that you
identify your location at startup, and then make that information
available in a standard way to the rest of the application independent
of sys.argv.  And if you know you might be moving directories, store
the os.path.abspath() of sys.argv[0] at startup rather than just
sys.argv[0].

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


From thomas at xs4all.net  Wed Mar  7 17:18:07 2001
From: thomas at xs4all.net (Thomas Wouters)
Date: Wed, 7 Mar 2001 23:18:07 +0100
Subject: Python 2.1b1 dumps core
In-Reply-To: <3aa6808b.962524627@news.okc1.ok.home.com>; from dwndgg@opti.cgi.net on Wed, Mar 07, 2001 at 06:46:09PM +0000
References: <3aa6808b.962524627@news.okc1.ok.home.com>
Message-ID: <20010307231807.D404@xs4all.nl>

On Wed, Mar 07, 2001 at 06:46:09PM +0000, Dan wrote:

> I have a pure Python based web server app that dumps core when run
> under Python 2.1b1.  It runs fine under Python 2.0.  I can send a core
> dump to someone if that is useful to anyone.  If not then I might be
> able to narrow it down to a certain part of the Python code and post a
> sample that breaks..

I'm pretty sure we'd want to see the traceback of the core in any case. The
corefile itself would be pretty useless unless you happen to run a very
vanilla system of a type that developers have access to -- you need to match
them up with the exact same binaries and libraries and all.

What you can do in any case is provide us with the traceback of the
application. If you are on a system that has 'gdb', or you have a 'dbx' or
something other that does the same job, you can do it with something like
this:

> gdb <python binary> <core>
<copyright information>
<loading of libraries>
<loading of symbols>
#0  0x4009a161 in sigaction () from /lib/libc.so.6
(gdb) 

(gdb) bt
#0  0x4009a161 in sigaction () from /lib/libc.so.6
#1  0x4009b5a1 in sigset () from /lib/libc.so.6
#2  0x80b9fdb in unset_signals () at bsd.c:1102
#3  0x80ba2e1 in sighandler (sig=11) at bsd.c:1269
#4  0x4009a098 in sigaction () from /lib/libc.so.6
#5  0xffffffff in ?? ()
Cannot access memory at address 0x30

(This example is from the only corefile I have at hand, which is a useless
one. Yours will hopefully give a proper traceback :) You can send the
traceback to me, if you want, or to a developer you happen to know has the
same system or experience in the area that is breaking Python here. I/we
might need more information too, btw, like the value(s) of some of the
locals in one of the frames or other, so better not delete the corefile yet
:)

Unfortunately, 2/3rds of the python development team and a good portion of
the active non-development community is at the Python conference so you
might have to be patient.

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

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



From claird at starbase.neosoft.com  Tue Mar 27 10:59:16 2001
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 27 Mar 2001 09:59:16 -0600
Subject: Running external command in sequential order
References: <Pine.LNX.4.10.10103271016010.1194-100000@cage.rug.ac.be> <985684788.135533.1517.nullmailer@ev.netlab.zetabits.com>
Message-ID: <23A194E3E7016ED0.21A1D161AA2E3CEF.8DC7E3D487CD1937@lp.airnews.net>

In article <985684788.135533.1517.nullmailer at ev.netlab.zetabits.com>,
Yukihiro Matsumoto <matz at zetabits.com> wrote:
>HI,
>
>In message "[ruby-talk:13231] Running external command in sequential order"
>    on 01/03/27, Paul Pladijs <ppladijs at cage.rug.ac.be> writes:
>
>|The system method of the Kernel module creates a subshell,
>|starts the requested command (= parameter of system method).
>|After creating  the subshell, the following statements of
>|the ruby script are executed. So, after invoking the system
>|method, two processes (the ruby interpreter and the subshell)
>|are running in parallel, at least in a multitasking
>|environnement. (I hope this is correct)
>
>system function in Ruby waits the process to terminate.  So it's just
>like system2 you desired.
>
>							matz.

Mr. Pladijs has been asking this in at least a couple of
newsgroups.  I'll attempt a bit of clarification.  I hope
I dispel, not increase, the muddiness.

Mr. Pladijs, your comments on the parallel action of
system are NOT correct for Ruby and Python (or Tcl or
Perl or ...).  The basic external-command-invocation in
each of these libraries is sequential and synchronous, as
what you labeled "system2" in your original post.

You expressed a desire to minimize process count.  This is
a commendable instinct.  However, I recommend that you first
model your application in working code.  If you still have a
concern about minimization of processes, yes, there are ways
to achieve that.  I think you'll be in a more informed posi-
tion to discuss them after you've deepened your experience
of conventional uses of library facilities.
-- 

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


From qrczak at knm.org.pl  Fri Mar 16 16:46:29 2001
From: qrczak at knm.org.pl (Marcin 'Qrczak' Kowalczyk)
Date: 16 Mar 2001 21:46:29 GMT
Subject: Who's minister of propaganda this week?
References: <B98041329020E007.7DA84CAD9D0562AB.817BBC572F480869@lp.airnews.net> <98nu1m$12e@dispatch.concentric.net> <660ACCA1E4816F7E.C359CDF4168EF3FB.3B6B05538D619770@lp.airnews.net> <3AAFA13D.76E32C3F@divalsim.it> <98oa4o01i4v@news2.newsguy.com> <mailman.984619039.7856.python-list@python.org> <98q9h60122v@news2.newsguy.com>
Message-ID: <slrn9b52dj.8it.qrczak@qrnik.zagroda>

Thu, 15 Mar 2001 12:37:18 +0100, Alex Martelli <aleaxit at yahoo.com> pisze:

> More generally: we might erroneously pass to foo some object that
> does not even _implement_ First and Second methods with acceptable
> signatures, or some that _does_ implement the methods with signatures
> that appear good BUT not with the semantics that foo needs
[...]
> No compiler wards you at compile-time against all, or even _most_,
> errors of this very common kind.

This is not true. In statically typed languages I use most of my errors
(over 80%) are detected at compile time.

Most of them are type errors (e.g. mismatching the number of
arguments) or name binding errors (e.g. forgotten module imports or
using a parameter which is not present in this function after some
reorganization).

I sometimes pass an object which doesn't implement the right interface.
This is checked statically. I never pass an object which implements
a different interface under the same name (because interfaces are
not matched by method names but by interface identities).

-- 
 __("<  Marcin Kowalczyk * qrczak at knm.org.pl http://qrczak.ids.net.pl/
 \__/
  ^^                      SYGNATURA ZAST?PCZA
QRCZAK


From bglenden at hotmail.com  Sat Mar 24 15:25:38 2001
From: bglenden at hotmail.com (Brian Glendenning)
Date: Sat, 24 Mar 2001 13:25:38 -0700
Subject: Best Python documentation ->HTML?
Message-ID: <pg0qbt0e598g1ecnfj5lj0lpa1q1pqvagq@4ax.com>

I am a Python newbie and am confused about what the best way to turn
Python documentation into HTML is (i.e., a "javadoc" equivalent). If
it could also make good printed documentation (e.g., page numbers) as
well as HTML, all the better.

Pointers and suggestions welcome! -

Cheers,
Brian Glendenning, bglenden at nrao.edu


From apec at chinanusa.com  Wed Mar 28 20:34:17 2001
From: apec at chinanusa.com (APEC)
Date: Wed, 28 Mar 2001 17:34:17 -0800
Subject: An Invitation from the Asia-Pacific Rim
Message-ID: <WEB-DEV1ElR3n4DVvm200001901@web-dev1>

 	
Aiming at Asia-Pacific Market?	 
Here's your one and only chance this year to target 
21 developed and developing member economies in the Asia-Pacific rim
<http://www.apec2001.org/apec_member.asp> .	

   	 	
 Industrial Science and Technology Cooperation
 Conference and Exhibition	
 September 21 - 25, 2001
 Suzhou, China

	
 An hour's drive from Shanghai, the biggest city in China 	
 	
Who Should Attend	
*	Technology-oriented companies who want to     demonstrate
advanced technologies or to seek     joint ventures
*	Venture capitalists and other investors who     want to identify
promising high-tech     companies and investment opportunities
*	Entrepreneurs who are seeking market     expansion to the
Asia-Pacific region
 	
 	
Best Chance to Showcase Technologies	
*	Exhibition & transaction of high technology &     hi-tech
products.
*	Bid for technological cooperation.
*	Discussions & negotiations on technological     transaction and
cooperation.
*	APEC forum on science and technology.
 	
 	
All the Opportunities!	
*	Technomart is endorsed by APEC's 21     member economies that
had a combined     Gross Domestic Product of over US$18 trillion     in
1999 and 43.85 percent of global trade.
*	In Technomart III, attendees included 1,000+     hi-tech
exhibitors and recorded 37,000 visitors.
*	In Technomart III, over 5,000 business     negotiations
occurred, involving trade in     excess of about US$ 100 million over
the     course of the five-day event.
*	Direct connection to China market.
 	
 	Service Center	
*	Technomart representation
*	Exhibition Booth & Expense
<http://www.apec2001.org/apec_booth.asp> 
*	Booth arrangements and allocation
*	Transportation
*	Hotels <http://www.apec2001.org/apec_hotel.asp> 
*	Sourcing/manufacturers visits
*	On-site public relation facilitation, media     coverage
*	Collateral translation
*	Escort interpretation
*	Leisure touring in Suzhou and its vacinities
 	
 	
Key Technological Areas	
*	Biotechnology
*	Environmental and cleaner production     technologies
*	Communication
*	Information technologies / electronics
*	Advanced materials
*	Mechatronics
*	Transportation
*	Resource management technology
*	Energy
*	Sustainable agriculture
*	Emergency preparedness and climate     prediction
*	Exploitation of natural resources
 	
  For more information about the upcoming APEC Technomart IV, please
visit:
http://www.apec2001.org	
  For inquiring a brochure, please visit:
http://www.apec2001.org/apec_register.asp	
  For Technomart IV services, please visit:
http://www.apec2001.org/apec_serve.asp	
  For exhibition booth information, please visit:
http://www.apec2001.org/apec_booth.asp	
  For on-line registration information, please click:
<http://www.apec2001.org/apec_registernow.asp> 
For further questions, please email us at apec at CHINAnUSA.com
<mailto:apec at CHINAnUSA.com>  or call Jessica Pan at (510) 497-8783. 	
  _____  

CHINAnUSA.com is proud to represent APEC Technomart IV committee here in
the US to work closely with you. It is our pleasure to assist you in
every aspect for the most beneficial trade show.

The past Technomart initiated trillions of transactions. Register today
<http://www.apec2001.org/apec_registernow.asp> ! 
We look forward to your participation and seeing you in Suzhou, China.

Authorized representative of APEC Technomart IV

Unsubscribe:
If you received this message in error or wish to unsubscribe from this
subject, just click here: Click here.
<http://www.apec2001.org/unsub/unsubscribe.asp?mail_id=pyt9801100413532>

Or you can reply to this email with 'Unsubscribe' as subject title.

All trademarks are the property of their respective owners (C)2001 All
rights reserved. 



-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20010328/0cd97621/attachment.html>

From parz at home.com  Fri Mar  2 01:08:53 2001
From: parz at home.com (Parzival Herzog)
Date: Fri, 02 Mar 2001 06:08:53 GMT
Subject: Unresolved symbols in MSVCRT.DLL when running a py2exe packaged program.
References: <RKln6.6630$hn5.853203@news1.rdc1.mb.home.com> <hVBn6.7888$UZ4.3012833@news4.rdc1.on.home.com>
Message-ID: <VNGn6.7720$hn5.948267@news1.rdc1.mb.home.com>

Thank you for your kind replies. I have not yet found what the actual problem
is, my app will indeed run quite nicely on several different machines, with
several different MSVCRT.DLL's, and even on this offending machine, provided
that I do not use py2exe. I can't figure out what is different about this machine,
being essentially a brand new machine churned out by a disinterested
corporate IT support group, and foisted on users loaded with antivirus
and corporate policy software, but otherwise untouched, and unconfigured.

Thanks to the responders, I now know that there is
a difference in DLL policy between Win2k and Win NT. Anyways I'm not that
interested in replacing MSVCRT.DLL in the system directory of my client's machines,
that's just what we tried to do in order to find the cause of the problem.
I just want my packaged Python app to work on any machine it is distributed to
without trouble. Has no one experienced this sort of problem for themselves?
Has anyone resolved it? How?

- Parzival


"Tom" <NoSpam at NoSpam.com> wrote in message news:hVBn6.7888$UZ4.3012833 at news4.rdc1.on.home.com...
> One difference (between the Win2K & WinNT machines) is the new system file
> protection 'feature' in Win2K.  It is supposed to disallow inappropriate
> updates to system DLL's (of which MSVCRT.DLL is one).  Another aspect of
> this feature is that apps can load their own versions of system DLL's, so
> multiple different versions MSVCRT.DLL can be memroy resident at the same
> time (but not on NT, just 2K).
>
> Someone suggested deleting the DLL to see if it is in use.  I'm not sure
> about DLL's, but with drivers (.SYS) you can delete it even while it is in
> use (wierd).  There are utilities around that will tell you what DLL's your
> app has loaded.  Maybe even the sysinfo utility, included in Win2K, might
> tell you this?
>





From f.g.van.dorp at freeler.nl  Mon Mar 19 09:16:14 2001
From: f.g.van.dorp at freeler.nl (Fred)
Date: Mon, 19 Mar 2001 14:16:14 GMT
Subject: [Pythonwin] Scintilla  ID(L)Eal
Message-ID: <bu4cbts8mjliub4iobb2bvkd1u3q4m4392@4ax.com>

REGEXP support and "Find in Files" (recursively)
as in IDLE would be nice !



From emile at fenx.com  Thu Mar 22 20:42:56 2001
From: emile at fenx.com (Emile van Sebille)
Date: Thu, 22 Mar 2001 17:42:56 -0800
Subject: Good Editor? (What's the big deal with syntax colouring?)
References: <mailman.985291829.27633.python-list@python.org> <99dqqo$gpi$1@news.mathworks.com>
Message-ID: <99e9sv$usmp$1@ID-11957.news.dfncis.de>

No!  Black on pale yellow.  Routines not exceed one screen's worth.

Now-if-I-could-only-read-5-point-type-on-my-21"-monitor-ly y'rs


Emile van Sebille
emile at fenx.com

--
-----

"Joshua Marshall" <jmarshal at mathworks.com> wrote in message
news:99dqqo$gpi$1 at news.mathworks.com...
> John Schmitt <jschmitt at vmlabs.com> wrote:
> ...
> > The one-and-only colour combination that works best for source code is
> > bright green on black.
> ...
>
> Yes, you're right of course.  How silly of us.




From aleaxit at yahoo.com  Tue Mar 20 06:43:39 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Tue, 20 Mar 2001 12:43:39 +0100
Subject: Functionalism, aesthetics Was:(RE: I come to praise .join, not to bury it...)
References: <mailman.985067371.2582.python-list@python.org>
Message-ID: <997fp901i44@news2.newsguy.com>

"Clark C. Evans" <cce at clarkevans.com> wrote in message
news:mailman.985067371.2582.python-list at python.org...
> Alex Martelli [mailto:aleaxit at yahoo.com] wrote:
> > it DOES mean each of these professionals need not fall back
> > on vague generalities and empty hand-waving if and when their
> > design choices get challenged.
>
> I've been confronted on too many occasions where my
> "aesthetics" differed from another party's.  Perhaps
> our functions were different.  But in any case, I was
> able to clearly articulate what my functions were and
> why the form I choose supported those functions.

...which exactly matches my (optimistic:-) expectations
from a real professional (in _most_ fields; I would not
expect _all_ arts to be amenable to such patiently
spelled-out, verbalized analysis... just _most_:-).

> The other party typically rests soley on dogma,
> giving in to "god", "universal patterns", or "beauty"
> that I obviously don't comprehend.  Oh well.

I have some problems envisaging a realistic situation
where a technical design choice for a system's architecture
might be challenged on theological grounds, but I guess
that might happen, if one was in the business of designing
churches, for example!-)


> > Somebody who's unable to do anything but talk of aesthetics
> > when discussing a choice in the architecture of a software
> > component is simply not contributing to the discussion.  It
> > looks good to you, it looks bad to him, it looks indifferent
> > to her, yawn, great debate, now let's get beyond this to the
> > REAL meat of the issue, please -- the technical implications
> > of this, that, or the other choice in a specific matter.
>
> Most often those bringing up aesthetics need to go
> and do their homework, detailing those functions that
> they would expect or like to have from the software.
> And then an analysis as to how each form satisfies those
> particular functions.

Amen to that.  Should technology be shamefully hidden,
disappear from view, become invisible (a la Don Norman),
or should it be displayed, without either pride or shame,
exactly for what it is and does (a la Renzo Piano's,
Beaubourg first and foremost)?  I'm not qualified to
judge, I guess, but I _do_ know who designed Osaka's
Kansai Air Terminal (which survived the Kobe earthquake,
quite destructive even by Japanese standards), and who,
on the other hand, used to work for a computer company
whose newest, company-saving line of products later relied
on a brilliant new look shamelessly _displaying_ internals
and technology through colourful translucent exteriors:-).


> Another approach, is a user community survey, explaining
> both sides, the options and implications, and asking
> each user to think deeply before casting a vote.  Usually
> if there is any core "aesthetic" it will pop out of
> a stastical survey.

I'm less optimistic than you about the pragmatics of user
surveys, perhaps in part because of my personal experience
with them.  More than once, I've seen a firm saved by a
small dissenting group proceeding stealthily to develop
an alternative, user-survey-condemned architecture, while
the user-survey-blessed choice foundered (often on rocks
of excessive complexity and acute featuritis) -- when the
official product had to be canceled, indefinitely postponed,
or even recalled, the simpler, pure-technology-oriented,
stealthworks-developed alternate by the technonerds was
at hand to save the day.  I might even suggest that, in
some fields, a productive strategy might be to proceed to
determine by traditional means all that the users _say_
they want, then do exactly the opposite (but that would
be pure provocation on my part, so I won't:-).


Alex





From bkline at rksystems.com  Thu Mar 29 14:46:02 2001
From: bkline at rksystems.com (Bob Kline)
Date: Thu, 29 Mar 2001 14:46:02 -0500 (EST)
Subject: forcing dictionary order
In-Reply-To: <3AC380AE.B225C07F@uniserve.com>
Message-ID: <Pine.LNX.4.10.10103291444410.31908-100000@rksystems.com>

On Thu, 29 Mar 2001, Bob van der Poel wrote:

> 
> I'd like (not a must!) to walk a dictionary in the same order in
> which it was created. Example:
> 
> 	foo={ key1: (), key2: (), key3: () }
> 
> If I do:
> 	
> 	for a in foo.keys(): ...
> 
> the order is not "key1, key2, key3", which is what I want. Note, I
> can't do something simple like sort() on the keys since the are not
> in any alphanumeric order. I'm just using this to generate a largish
> table for another program, and the actual order doesn't matter for
> it to work properly, but it would nicer if it were in the correct
> logical order. So, I'm not going to spend a lot of time on this, but
> if there is a simple way????

Sure.  Maintain a parallel array of references to the entries in the
dictionary as the dictionary is populated.

-- 
Bob Kline
mailto:bkline at rksystems.com
http://www.rksystems.com




From Mike_B at T-Online.de  Mon Mar 26 09:35:52 2001
From: Mike_B at T-Online.de (Michael Bauer)
Date: Mon, 26 Mar 2001 16:35:52 +0200
Subject: daemon app in win32
Message-ID: <99nk49$a2q$02$1@news.t-online.com>

Hi there,

i am writing a daemon application that runs both under windows and unix.  
Since i am not very experienced in win32 - programming, i really would 
appreciate some hints how to register my app in win32 that i can do some 
cleanup-functions when windows is shut down.

At the moment it is just a standard console application that simply gets 
killed when windows shuts down. The app does not have/need a GUI, it should 
just run in the background.

thank you, Mike



From aahz at panix.com  Fri Mar  2 19:08:57 2001
From: aahz at panix.com (Aahz Maruch)
Date: 2 Mar 2001 16:08:57 -0800
Subject: cataloging words in text file
References: <3AA0124E.74A39783@comm.mot.com> <v9Vn6.396$z6.33544@ruti.visi.com>
Message-ID: <97pcmp$fuk$1@panix2.panix.com>

In article <v9Vn6.396$z6.33544 at ruti.visi.com>,
Grant Edwards <grante at visi.com> wrote:
>
>#!/usr/local/bin/python2.1
>import sys
>d={}
>for w in sys.stdin.read().split():
>    if d.has_key(w):
>        d[w] += 1
>    else:
>        d[w] = 1
>print d    

Since you're using 2.1, try this

for w in sys.stdin.read().split():
    d[w] = d.setdefault(w,0) + 1
-- 
                      --- Aahz (Copyright 2001 by aahz at pobox.com)

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

Nostalgia just ain't what it used to be


From tchur at optushome.com.au  Fri Mar  9 03:14:14 2001
From: tchur at optushome.com.au (Tim Churches)
Date: Fri, 09 Mar 2001 19:14:14 +1100
Subject: Who am I: can a class instance determine its own name?
References: <mailman.984034150.9635.python-list@python.org> <slrn9af71n.d6p.scarblac@pino.selwerd.nl> <mailman.984081385.20244.python-list@python.org> <la%p6.1337$54.1249@www.newsranger.com>
Message-ID: <3AA890D6.93E3F10F@optushome.com.au>

Gregory Jorgensen wrote:
> 
> Emile van Sebille's very cool hack aside, I don't understand why you would do
> this. I'm not criticizing; you've got me curious. I've never needed this in
> years and years of programming. I don't think of a variable name as an attribute
> of the object it refers to, because (as one poster pointed out) objects may have
> no name, one name, or several names at any time, and the names referring to an
> object may change during the object's lifetime through assignment, parameter
> passing, etc.
...snip...

Fade to scene of Domestic Bliss...5 year old Wunderkind quietly
supervising the evolution of assorted Pokemon, Better Half reading some
obscure journal, Your Correspondent hacking away on the Faithful Laptop,
frogs croaking in the rain outside. Better Half, who, like Your
Correspondent, is an epidemiologist and not a professional programmer,
asked (out of genuine curiosity, I think) what was so great about
Python. Better half's only exposure to programming languages has been to
SAS (a large, venerable, sprawling, very expensive and very proprietary
business/scientific data management/statistics suite for which the main
3.5GL is an odd but curiously effective mixture of PL/1, Fortran and
BASIC mixed with lots of prepackaged high-level stats and graphics
procedures). 

So Clever Clogs (aka Your Correspondent) proceeds to demonstrate some of
the elegance of Python and the advantages of an O-O approach...

from PureGenius import *

# Define a dataset
measles_cases = Dataset("measles_cases")

# Define some variables
sex = Var("sex")
agegroup = Var("agegroup")

# Embed the variables in the dataset
measles_cases.addVar(sex)
measles_cases.addVar(agegroup)

...at which point, stifling a yawn, Better Half asks "Why do you have to
type the variable names twice?". At which point Wunderkind excitedly
announces that Charmeleon has just evolved into Charazard, providing a
convenient excuse for a change of subject and a stern lecture on why
Darwin was right and Lamarck was wrong (Pokemon exhibit Lamarckian
evolution).

But Better Half's question was far from stupid (hello sweetheart if you
are subscribed to this list) and it got me thinking...

You did ask.

Tim C



From aleaxit at yahoo.com  Sat Mar 31 02:22:01 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sat, 31 Mar 2001 09:22:01 +0200
Subject: newbie variables question
References: <mailman.986005960.10805.python-list@python.org>
Message-ID: <9a40is028ua@news1.newsguy.com>

"Michael Hall" <py.list at mulga.com.au> wrote in message
news:mailman.986005960.10805.python-list at python.org...
>
> I've been trying to do this:
>
> SYSPRODIR = /home/admin/systemprofile
> os.system('mv /tmp/install.log SYSPRODIR/install.log')
>
> No matter how I play around with single and double quotes, this won't
> work.
>
> But this does work:
>
> os.system('mv /tmp/install.log %s/install.log' %s SYSPRODIR)
>
> Is this the normal/best way to do this in Python ... seems a little
> convoluted me, like there must be a simpler way. Bash for example would be

There are several ways, but I think this one is simplest.  Alternatives:

os.system('mv /tmp/install.log '+SYSPRODIR+'/install.log')
os.system('mv /tmp/install.log %(SYSPRODIR)s/install.log' % vars())
os.system('mv /tmp/install.log FOO/install.log'.replace('FOO',SYSPRODIR))


Alex






From garrett at cs.colostate.edu  Thu Mar  1 04:51:26 2001
From: garrett at cs.colostate.edu (Deon Garrett)
Date: Thu, 01 Mar 2001 09:51:26 GMT
Subject: To the point of defeat.....IDLE
References: <3A9AE62B.5070402@yahoo.com>
Message-ID: <slrn99s6tu.8rl.garrett@falcon.cs.colostate.edu>

The best way to go is to completely wipe any traces
of python rpms from your system before installing 
Python 2.0 from source.

rpm -e python

will fail because a bunch of stuff needs python.  But
it will list all the packages that need it.  You should
be able to look through the dependencies and find all
the stuff you want to remove.

rpm -e --nodeps python python-tools tkinter.......

I assume that the python executable is installed in 
/usr/bin on RedHat 7.  If so, then run configure in 
the python 2.0 root directory with the --prefix=/usr
option and all your scripts should still work.

cd your/python2.0/root/dir
./configure --prefix=/usr (and any other options you want)
make
make install

Then build any add-ons you want (Tkinter, NumPy, PIL, etc).

--
Deon Garrett


From jwbnews at scandaroon.com  Mon Mar 19 23:48:33 2001
From: jwbnews at scandaroon.com (John W. Baxter)
Date: Mon, 19 Mar 2001 20:48:33 -0800
Subject: What IDE are you'all using for Python 2.1 for Windows?
References: <995gai$2bv$1@news.chatlink.com>
Message-ID: <jwbnews-3748ED.20483319032001@corp.supernews.com>

In article <995gai$2bv$1 at news.chatlink.com>,
 "Luis Cortes" <lcortes at aodinc.com> wrote:

>   Just wanted to know what "good" ide's are out there for python.  Please
> let me know which ones you prefer.

I'm not qualified to answer, since Windows is the environment in which I 
haven't used Python.  (I generally use emacs on Unix/Linux and BBEdit on 
Mac.)

  --John

-- 
John W. Baxter   Port Ludlow, WA USA  jwbnews at scandaroon.com


From stephen_purcell at yahoo.com  Thu Mar 22 06:22:12 2001
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Thu, 22 Mar 2001 12:22:12 +0100
Subject: Why Python/Jython?
In-Reply-To: <99cl9e$b2b$1@Merlin.et.bocholt.fh-gelsenkirchen.de>; from freier@xtramind.com on Thu, Mar 22, 2001 at 11:48:48AM +0100
References: <99cl9e$b2b$1@Merlin.et.bocholt.fh-gelsenkirchen.de>
Message-ID: <20010322122212.A5609@freedom.puma-ag.com>

Andreas Freier wrote:
> In several articles I read that Python/Jython imporoves the programmers
> performance many times. Why?

Simple: Python code is usually many times shorter than the corresponding
Java code.

Programmer performance is usually gauged by coding time and
number of bugs, both of which are vaguely proportional to the number of
lines of code in most sane languages*.

* Perl is a noteworthy exception

> I am considering to use Jython for my job as Java programmer. How should I
> use it to achieve this performance?

Just use it and see if it works for you.

-Steve

-- 
Steve Purcell, Pythangelist
Get testing at http://pyunit.sourceforge.net/
Any opinions expressed herein are my own and not necessarily those of Yahoo



From ddb at crystal.uwa.edu.au  Sun Mar 25 21:01:29 2001
From: ddb at crystal.uwa.edu.au (Douglas du Boulay)
Date: Mon, 26 Mar 2001 10:01:29 +0800
Subject: Jython and jpythonc2
References: <3AB58BDF.56EA463C@crystal.uwa.edu.au>
Message-ID: <3ABEA2F9.69142925@crystal.uwa.edu.au>

Douglas du Boulay wrote:
[snip]
> been processed, the compiler has consumed around 80 Megabytes of ram.
> At that point on a 128Meg(RAM) Compaq Alpha the process apparently
> gets stashed in the swap space. Thereafter, instead of running with 98%
> of the CPU time it runs at about 2%, with continuous disk thrashing.
> At the moment, after 3 hours real time, the process has only
> had 8 minutes of actual CPU time and is using 180 megabytes of virtual
> RAM. (There are about 80 source files involved here)

In the end it took about 5-6 hours for the java file generation to
finish 
and then it bombed out for lack of memory when it spawned a javac job to
compile the java sources.

The problem can be avoided by not using the COMPAQ True64 supplied JVM 
(both 1.3.0 and 1.2.2). On an intel linux box running the HotSpot 1.3.0
JVM it completed in 10 minutes and consumed about 50Megs of RAM.

Doubtless no one else will ever encounter this problem. 
This message was just for completeness.
Doug


From stephen_purcell at yahoo.com  Thu Mar  8 03:39:17 2001
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Thu, 8 Mar 2001 09:39:17 +0100
Subject: Who am I: can a class instance determine its own name?
In-Reply-To: <saa7c5ef.027@dohprimo.ccf.health.nsw.gov.au>; from TCHUR@doh.health.nsw.gov.au on Thu, Mar 08, 2001 at 05:48:24PM +1100
References: <saa7c5ef.027@dohprimo.ccf.health.nsw.gov.au>
Message-ID: <20010308093917.A2886@freedom.puma-ag.com>

Tim CHURCHES wrote:
> ...can an instance of a class determine the name of the variable to which it is assigned? For example:
> 
> ###########################
> class Foo:
>     def whoami(self):
>          return "You are a Foo() but I do not know your name"
> 
> FooBar = Foo()
> 
> print FooBar.whoami()
> ###########################
> 
> How does one define the method whoami() so that it returns "FooBar"?


Hmmm. There's no way to do this without telling Foo its name explicitly.
If there are two names for a Foo instance, what would you like 'whoami' to
print?

>>> foobar = Foo()
>>> another_name_for_that_foo = foobar
>>> foobar
<__main__.Foo instance at 807e0b8>
>>> another_name_for_that_foo 
<__main__.Foo instance at 807e0b8>
>>> 


See the problem?

-Steve

-- 
Steve Purcell, Pythangelist
Get testing at http://pyunit.sourceforge.net/
Any opinions expressed herein are my own and not necessarily those of Yahoo



From ben.hutchings at roundpoint.com  Tue Mar 20 17:43:40 2001
From: ben.hutchings at roundpoint.com (Ben Hutchings)
Date: 20 Mar 2001 14:43:40 -0800
Subject: ASCII encoding error: ordinal not in range(128)
References: <91o5ih$2kv$1@news.wrc.xerox.com> <20001219172933.B29727@ActiveState.com> <mailman.985123475.18421.python-list@python.org>
Message-ID: <u7l1krrhf.fsf@roundpoint.com>

"Jose Isaias Cabrera" <jicman at cinops.xerox.com> writes:

> Greetings all!
> 
> I have a python program that interfaces with Microsoft Outlook.  I
> am running into a problem when I have attachments and subject titles
> with different characters that are not your normal everyday
> character.
<snip>
> UnicodeError: ASCII encoding error: ordinal not in range(128)
> 
> I've search on python sites for this, and though there is something about
> Unicode, there is no python example code to use.
>
> Does anyone have any idea how to fix this?

Not much, because you didn't post any code.  But you should probably
be looking at the codecs.EncodedFile class, which wraps a non-ASCII
text file (or file-like object) and translates whatever encoding the
file uses (which you must discover and specify) to and from Unicode.

-- 
Any opinions expressed are my own and not necessarily those of Roundpoint.


From neil.benn at cambridgeantibody.com  Wed Mar 28 08:04:16 2001
From: neil.benn at cambridgeantibody.com (Neil Benn)
Date: Wed, 28 Mar 2001 14:04:16 +0100
Subject: Defining class files
Message-ID: <3ac33343$0$12248$ed9e5944@reading.news.pipex.net>

Hello,

            Sorry to ask such a basic question but -

I'm wokring through the tutorial and have reached the section on classes in
Python.  Working through the example on classes I have entered the class
using the interpreter:-

class MyClass:
    "A simple example class"
    i = 12345
    def f(x):
        return 'hello world'

    This worked fine when I typed:-

>>> x=MyClass()
>>> x.f()
'Hello World'

    However, I then tried to write the class as an external text file,
using:-

class YourClass:
    "A simple example class"
    i = 12345
    def f(x):
        return 'hello world'

    I then imported the class, tried to assign the class and ivoke a method
:-

>>> y = YourClass()
Traceback (innermost last):
  File "<console>", line 1,
TypeError: call of non-func

    This seemed a bit strange, so I investigated the Your class and it
seemed to be a module.

>>> YourClass
<module YourClass at 2584541>

    The text file is saved as YourClass.py - is this the problem, should
class files have different terminaters in their filename??

    Any/all help would be most appreciated!!

Cheers,

Neil Benn
Automation Informatics Scientist
Cambridge Antibody Technology
Cambs
UK




From phys137 at mailcity.com  Sun Mar  4 23:09:29 2001
From: phys137 at mailcity.com (phys137 at mailcity.com)
Date: Mon, 5 Mar 2001 15:09:29 +1100
Subject: Java guy interested in Python
References: <mailman.983675705.3042.python-list@python.org>
Message-ID: <97v3fm$diu$1@mail.pl.unisys.com>

personally for me the most useful features were the ones related to list processing.

List comprehensions and (limited) lambdas allow you to write one line instead of 10 - and debug much less. 

Otherwise Java platform has a lot of things that Python doesn't (like language-defined multithreading, platform defined 3D graphics etc.)

Still, Python syntax allows for many less lines of code in many situations. And that's important. Also, it is scripting - you don't have to bother with type declarations. 
  "Kyle Wayne Kelly" <kknbk2 at bellsouth.net> wrote in message news:mailman.983675705.3042.python-list at python.org...
  Can anyone tell me some of the things that Python offers which are not available in Java?  I like Python's syntax, but what can Python do that Java can not do?  Any comments are helpful.

  Kyle Wayne Kelly
  Computer Science Student
  University of New Orleans
  http://www.cs.uno.edu/~kkelly
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20010305/bbcd2f4c/attachment.html>

From rjroy at takingcontrol.com  Mon Mar  5 20:46:45 2001
From: rjroy at takingcontrol.com (Robert Roy)
Date: Tue, 06 Mar 2001 01:46:45 GMT
Subject: Trouble running a Grayson Example
References: <3AA3E573.6DBE6D2E@theworld.com>
Message-ID: <3aa43d78.258996734@news1.on.sympatico.ca>

On Mon, 05 Mar 2001 19:13:50 GMT, David Lees
<deblNonOspammy at theworld.com> wrote:

>I am trying to run example 8.9 (Example_8_9.py) from Graysons book,
>under Python 2.0 and get the following errors.  Can someone tell me what
>is going on?  Perhaps this has to do with changes between 1.5.2 and 2.0
>Python?
>
>Thanks in advance.
>
>david lees
>
>
>Traceback (most recent call last):  File
>"d:\Python20\Pythonwin\pywin\framework\scriptutils.py", line 301, in
>RunScript
>    exec codeObject in __main__.__dict__
>  File "E:\Grayson1\Chapter08\Example_8_9.py", line 79, in ?
>   ddnotebook = DDNotebook()
>  File "E:\Grayson1\Chapter08\AppShell.py", line 55, in __init__
>  self.__createInterface()
>  File "E:\Grayson1\Chapter08\AppShell.py", line 282, in
>__createInterface
>   self.createInterface()
>  File "E:\Grayson1\Chapter08\Example_8_9.py", line 75, in
>createInterface
>    self.createNotebook()
>  File "E:\Grayson1\Chapter08\Example_8_9.py", line 26, in
>createNotebook  Pmw.NoteBookR,(self.interior(),),)
>  File "E:\Downloads\pmw\Pmw\Pmw_0_8_4\lib\PmwLoader.py", line 174, in
>__getattr__
>AttributeError: NoteBookR

The examples in the Grayson book were written using  an older version
of Pmw probably 0_8_1. NoteBookR ceased to exist in 0_8_3 and you are
using 0_8_4. There were some fundamentat changes to the way Notebook
works.  

The following code will work.



    def createNotebook(self):
# changed Pmw.NoteBookR to Pmw.NoteBook
        self.notebook = self.createcomponent('notebook', (), None,
                                         Pmw.NoteBook,
(self.interior(),),)
        self.notebook.pack(side=TOP, expand=YES, fill=BOTH, padx=5,
pady=5)
        self.formwidth = self.root.winfo_width()
        
    def addPage(self, dictionary):
        table, top, anchor, incr, fields, \
            title, keylist = dataDict[dictionary]
# there is no label attribute anymore but a tab_text attribute
#(optional)
#        self.notebook.add(table,label=title)
        self.notebook.add(table, tab_text=title)
        self.current = 0
        ypos = top
        idx = 0
        for label, field, width, proc, valid, nonblank in fields:
            pstr = 'Label(self.notebook.page(table),'\
                   'text="%s").place(relx=%f,rely=%f, anchor=E)\n' % \
                   (label, (anchor-0.02), ypos)
            if idx == keylist[0]:
# self.notebook.page(table).interior() to self.notebook.page(table)
# did this in 3 places
                pstr = '%sself.%s=Entry(self.notebook.page(table),'\
                       'text="",insertbackground="yellow",
width=%d+1,'\
                       'highlightthickness=1)\n' % (pstr,field,width)
            else:
                pstr = '%sself.%s=Entry(self.notebook.page(table),'\
                       'text="", insertbackground="yellow",'\
                       'width=%d+1)\n' % (pstr,field,width)
            pstr = '%sself.%s.place(relx=%f, rely=%f,'\
                   'anchor=W)\n' % (pstr,field,(anchor+0.02),ypos)
            exec '%sself.%sV=StringVar()\n'\
                 'self.%s["textvariable"] = self.%sV' % \
                                          (pstr,field,field,field)
        ypos = ypos + incr
        idx = idx + 1





From quinn at riyal.ugcs.caltech.edu  Thu Mar 15 15:14:07 2001
From: quinn at riyal.ugcs.caltech.edu (Quinn Dunkan)
Date: 15 Mar 2001 20:14:07 GMT
Subject: Process details
References: <mailman.984439220.26863.python-list@python.org> <slrn9b14cv.fae.quinn@riyal.ugcs.caltech.edu> <mailman.984677900.23419.python-list@python.org>
Message-ID: <slrn9b28kf.hio.quinn@riyal.ugcs.caltech.edu>

On Thu, 15 Mar 2001 09:40:38 -0800, Timothy Grant <tjg at exceptionalminds.com>
wrote:
>...I just went and looked at my code and discovered the major
>bottleneck--I think. The problem comes due to the fact that not
>all process list all entries in the /status file. This means
>that a sequential read assigning to variables won't work, so I
>am actually looking for entry names in a loop.
>
>I just added a break statement and shaved three seconds of the
>time! I'm sure there is more room to optimize, but exiting my
>loop after I found what I was looking for sure helps.

Not that you asked or anything <wink>, but the 'stat' file has mostly the same
info and is more machine-readable.  You should be able to do something like:

class Proc:
    def __init__(self, stats):
        self.pid = int(stats[0])
        self.name = stats[1] # strip parens if swapped out
        self.state = stats[2]
        # etc.

procs = [ Proc(open('/proc/%s/stat' % s).read().split())
    for s in os.listdir('/proc') if s.isdigit() ]

for p in procs:
    print p.pid, p.name, p.state


From chris.gonnerman at usa.net  Thu Mar  1 09:08:45 2001
From: chris.gonnerman at usa.net (Chris Gonnerman)
Date: Thu, 1 Mar 2001 08:08:45 -0600
Subject: how can i copy a sublist
References: <3A9E4C6F.E2DD63C3@bg-sys.de>
Message-ID: <006901c0a259$261d0900$a100000a@local>

The key word is copy.  Python lists are copied with the [:] (all-slice)
notation, as in:

    b.append(a[0][:])

Hope this helps...

----- Original Message ----- 
From: "Daniel Orlowski" <daniel.orlowski at bg-sys.de>
Subject: how can i copy a sublist


> Hi,
> 
> i try to copy a sublist from list "a" to list "b". After that i alter a
> value in "b" and oops! value in "a" is also altered.
> so it looks like that:
> >>> a=[[1,2],[3,4]]
> >>> b=[]
> >>> b.append(a[0])
> >>> b
> [[1, 2]]
> >>> b[0][0]=3
> >>> a
> [[3, 2], [3, 4]]
> 
> how can i copy a sublist without
> "get_a_value_from_a_list_and_put_it_into_another"
> 
> Regards,
> Daniel
> 
> 
> -- 
> http://mail.python.org/mailman/listinfo/python-list
> 




From Laura_01 at MailAndNews.com  Fri Mar  9 23:31:13 2001
From: Laura_01 at MailAndNews.com (Laura Lewin)
Date: Fri, 9 Mar 2001 23:31:13 -0500
Subject: true newbie
Message-ID: <3AAEB1E2@MailAndNews.com>

Hi,
You can try: How to think like a computer scientist, Python Version by Allen 
Downey and Jeffrey Elkner (www.ibiblio.org/obp/thinkCSpy/).  
Laura
LLewin at oreilly.com

>===== Original Message From "Elaine Sweeny" <bobbo at milwpc.con> =====
>can anyone help me find a tutorial for python thats for people just starting
>out in programming. The one one that comes with it is ok but its still too
>difficult.
>        thanks
>



From stephen_purcell at yahoo.com  Thu Mar 15 09:54:30 2001
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Thu, 15 Mar 2001 15:54:30 +0100
Subject: Random Access Files
In-Reply-To: <9e4s6.42245$p66.13534448@news3.rdc1.on.home.com>; from colinmeeks@home.com on Thu, Mar 15, 2001 at 02:21:57PM +0000
References: <9e4s6.42245$p66.13534448@news3.rdc1.on.home.com>
Message-ID: <20010315155430.A1259@freedom.puma-ag.com>

Colin Meeks wrote:
> Is it possible to have random access files as opposed to serial access.

Yes. 

The built-in function 'open' is described at

  http://www.python.org/doc/lib/built-in-funcs.html

and returns a file object as desribed by

  http://www.python.org/doc/lib/bltin-file-objects.html

If you open a file for 'updating' (e.g. using mode 'r+'), it's a random access
file.

-Steve

-- 
Steve Purcell, Pythangelist
Get testing at http://pyunit.sourceforge.net/
Any opinions expressed herein are my own and not necessarily those of Yahoo



From akenn at microsoft.com  Mon Mar 26 08:58:07 2001
From: akenn at microsoft.com (Andrew Kennedy)
Date: Mon, 26 Mar 2001 14:58:07 +0100
Subject: CfP: Babel01 Workshop on Multi-Language Infrastructure and Interoperability
Message-ID: <3abf4ae1@msrnews.research.microsoft.com>

			       BABEL'01
			   CALL FOR PAPERS

		   First Workshop on Multi-Language
		 Infrastructure and Interoperability

	   http://research.microsoft.com/~nick/babel01.htm

			 Satellite to PLI'01

		  Firenze, Italy, 8th September 2001


AIMS AND SCOPE

Recent years have seen a resurgence of interest in multi-language
tools and intermediate languages, and in interoperability between
programs and components written in different programming
languages. Shared infrastructure such as code generators, analysis
tools and garbage collectors can greatly ease the task of producing a
high-quality implementation of a new programming language, whilst
being able to interoperate easily with code written in existing
languages is essential for such an implementation to be useful in
practice.  This workshop aims to bring together researchers and
developers working on multi-language integration. Contributions are
welcome on topics including, but not limited to:


* Compilation of high level languages to common executable formats
  such as Sun's Java Virtual Machine, the DRA's ANDF or Microsoft's
  .NET Common Language Runtime.

* Defining and using bindings for component middleware such as OMG's
  CORBA or Microsoft's COM.

* Language constructs to support interoperability between different
  languages, particularly from different paradigms (e.g. OO/functional).

* Multi-language backends for compilation and/or analysis
  (e.g. MLRISC, FLINT, C--, TAL, SUIF).

* Type theoretic and semantic foundations for multi-language work.

* Multi-language development environments and tools (e.g. debuggers,
  profilers).

Submissions may address any programming paradigm.  Experience papers
which describe the use of multiple languages in application
development are also welcome. Authors unsure of the appropriateness of
a potential submission should email the PC chair (nick at microsoft.com).


PROGRAMME COMMITTEE

Nick Benton (chair)          Microsoft Research
Fergus Henderson             University of Melbourne
Andrew Kennedy (organiser)   Microsoft Research
Greg Morrisett               Cornell University
Martin Odersky               Ecole Polytechnique Federale de Lausanne
John Reppy                   Bell Labs
Andrew Tolmach               Portland State University
David Wakeling               University of Exeter


INVITED SPEAKERS

Zhong Shao                   Yale University


IMPORTANT DATES

Deadline for submission      1st June 2001
Notification of acceptance   9th July 2001
Final version due            10th August 2001


SUBMISSION DETAILS

Papers should be at most 14 pages and should be submitted in
Ghostscript-compatible PostScript or PDF format and be printable on
both US letter and A4 paper.  Authors are strongly encouraged to use
ENTCS style files (see http://math.tulane.edu/~entcs/).  Papers should
be submitted electronically from the workshop website given above.
Submissions should not overlap significantly with work which has
already been published or submitted to any other conference or
journal.  An informal proceedings will be published as a technical
report and distributed at the workshop.  It is intended that a final
proceedings will be published in a volume of ENTCS.





From robin.thomas at starmedia.net  Wed Mar 21 07:23:09 2001
From: robin.thomas at starmedia.net (Robin Thomas)
Date: Wed, 21 Mar 2001 07:23:09 -0500
Subject: Yet Another PEP:  Interface Adapter Mechanism __adapt__
Message-ID: <4.3.1.2.20010321072228.00d94740@exchange.starmedia.net>

At 04:30 AM 3/21/01 -0500, Clark C. Evans wrote:

>Thank you all for your feedback and support with the first
>pass of this PEP.  Below is a second attempt.  The context and
>motivation sections are identical.  Here are the major changes:
>
>     * __query__ is renamed __adapt__  (adapter pattern)
>     * __adapt__ now takes a class instance (instead of a DNS string)
>     * __adapt__ now returns None if the lookup was unsuccessful
>     * straw-man adapt function is detailed
>     * the title of this PEP is updated respectively

This is very good. It actually can assimilate my idea for abstract type 
checking via a __type__ method.

I also see that adapt() works to assimilate isinstance(), and probably 
could assimilate issubclass(). I support that work, but I see that it gives 
some complexity to the meaning of the return value -- in adapter-like 
cases, you want to return self or an adapter of self; in test-like cases, 
you want to return a boolean. I also think that is very cool.

I like "adapt" better than "query".

I also think that this function is a candidate to be an operator in the 
future. The two quotes you mention in the PEP, to me at least, boil down to:

Quote 1: "The object may have the answer, so ask it about the identifier."
Quote 2: "The identifier may have the answer, so ask it about the object."

This makes me think of adapt() as a binary op just as +, ^, divmod, etc.

A word that works nicely as an operator name, especially a simple word that 
novices can grasp quickly, would be ideal as a name for this new feature.

Suggestions:

1) "like"

like(obj, ident, options=None)
# future operator "like" and "not like"
a like b    # equiv to like(obj,ident)
a not like b # equiv to not like(obj,ident,"test")

2) "is" with builtin isa() or even just adapt()

a is b     # adapt(a,b) as adapt written below
a is not b # not adapt(a,b,"test")
adapt(obj, ident, options=None)


Suggested impls for adapt:

# the like version
# ! uses builtins/ops as shorthand

def like(obj, ident, options=None):
     # if same object, TRUE
     if obj is ident: return 1
     obj_type = type(obj)
     # if one is the concrete type of another,
     # or if they have the same concrete type, TRUE
     if obj_type is ident: return 1
     ident_type = type(obj)
     if obj_type is ident_type or obj is ident_type: return 1

     # seems that isinstance() and issubclass() can be implemented
     # here if we return true cases and let the false cases fall
     # through to the __like__ check.

     retval = 0
     if ident_type is ClassType:
         # isinstance
         if obj_type is InstanceType and isinstance(obj, ident): return 1
         # issubclass
         elif obj_type is ClassType and issubclass(obj, ident): return 1

     # __like__, if a "builtin", is actually in the type methods
     # table, not a real named attribute. obj.__like__ is just
     # the way for instances to emulate the builtin method. Right?
     # So the code below is pseudo-code for what would happen
     # at the C level.

     # try the obj's type first
     if obj_type.slot('__like__'):
         retval = obj_type.slot('__like__')(obj, ident)
     # else try the ident
     if ident_type.slot('__like__'):
         retval = ident_type.slot('__like__')(ident, obj)

     # options hook (if still needed)
     return retval

And the implementations for some built-in types:

InstanceType:
     if hasattr(self, "__like__"): return self.__like__(ident)

All the others are wide open. If the types module gets some abstract type 
objects like Mapping or Sequence or even things like Mutable, the __like__ 
slot exists to implement a type hierarchy.

I considered using ClassType's slot function to implement isinstance() and 
issubclass(), or at least issubclass(). Maybe that's still a good idea.


--
Robin Thomas
Engineering
StarMedia Network, Inc.
robin.thomas at starmedia.net




From akuchlin at mems-exchange.org  Fri Mar 30 22:51:07 2001
From: akuchlin at mems-exchange.org (A.M. Kuchling)
Date: Fri, 30 Mar 2001 22:51:07 -0500
Subject: New ZODB package
Message-ID: <200103310351.WAA06149@mira.erols.com>

I've synced with Zope 2.3.1final, released earlier today, and made a
new tarball of the standalone ZODB distribution.  It's available from
the usual place:

http://www.amk.ca/files/zodb/

--amk




From tomega at earthlink.net  Fri Mar 16 00:49:31 2001
From: tomega at earthlink.net (Simon Kesenci)
Date: Fri, 16 Mar 2001 05:49:31 GMT
Subject: Cookies
References: <mailman.984707839.22839.python-list@python.org>
Message-ID: <LPhs6.617$Im6.98112@newsread1.prod.itd.earthlink.net>

You need the script to send a cookie header.

import Cookie
import httplib

C["compose"] = "as thou wilt"
h = httplib.HTTP("www.microsoft.com")
h.putrequest("GET", "whatever.html")
h.putheader("Accept", "text/html")
h.putheader("Cookie", C)
h.endheaders()
reply = h.getreply()

I haven't tried this.  Let me know if it works.

--S.

Ken Seehof wrote:

> Short version:
> 
> Is it possible for a stand-alone python application to read a cookie
> without the help of a browser?"
> 
> Long version:
> 
> I have a web page W and a client side python script P.  I want W to
> generate a GUID (perhaps using javascript) which is then sent to a server
> S, where it is to be used as a key in a database.  Now I run P (on the
> client, and not in a browser) and I want P to somehow find the value of
> the GUID, so that it can query the database on S.
> 
> 
> Alternate question:
> 
> Is there a technique common to both javascript and python that will obtain
> a unique identifier for a computer (e.g. IP address).  It must not be
> browser specific.
> 
> 
> Motivation:
> 
> I need to write a record of data from an html form, keyed by some kind of
> unique client id, to the server and then be able to query that record from
> a stand alone python script running on the client.  The missing link is
> getting the html form and the python script to agree on a value for the
> id; the rest is easy.
> 
> 
> - Ken
> 
> 
> 



From aahz at panix.com  Wed Mar 14 00:01:34 2001
From: aahz at panix.com (Aahz Maruch)
Date: 13 Mar 2001 21:01:34 -0800
Subject: Perl test
Message-ID: <98mtve$enl$1@panix3.panix.com>

http://www.bbspot.com/News/2001/03/perl_test.html
-- 
                      --- Aahz <*>  (Copyright 2001 by aahz at pobox.com)

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

Python:
    Programming the way
    Guido indented it


From futurb at nh.km.ua  Thu Mar  8 05:02:50 2001
From: futurb at nh.km.ua (Anton A. Zotov)
Date: Thu, 8 Mar 2001 12:02:50 +0200
Subject: virtual hosting in Zope
Message-ID: <987lbo$119t$1@infocom.km.ua>

Urgent: I'd appreciate a person who would tell me how to implement virtual
hosting under Zope/Medusa environment




From sholden at holdenweb.com  Sat Mar 10 00:39:04 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Sat, 10 Mar 2001 00:39:04 -0500
Subject: class func & var name overlap
References: <3AA9AEC2.12BA1E95@nd.edu>
Message-ID: <L9jq6.47179$1D5.2262702@e420r-atl1.usenetserver.com>

I've added a few comments which might make things plainer.

"Changsen Xu" <xu.20 at nd.edu> wrote in message
news:3AA9AEC2.12BA1E95 at nd.edu...
> hi all,
>
> I just now tried function name and variable name overlaping
> in Python's script (1) and (2):
>
> (1)
> class x:
>         x = 3                        # bind x to 3
>         def x(self): return 5    # bind x to a function that returns 5
> y=x()                                # bind y to an instance of class x
> print y.x                            # print the x (instance method) of y
>
>
> (2)
> class x:
>         x = 3                        # bind x to 3
>         def x(self): return 5    # bind x to a function that returns 5
>         x = 6                        # bind x to 6
> y=x()                                # bind y to an instance of class x
> print y.x                            # print the x (class variable) of y
>
>
> Then if I run 1), I got screen echo as
>         <method x.x of x instance at 80e17d4>

Correct. The "def" statement assigns the function body (and the function is
an instance method) to the function name. This behavior is competely
dynamic, because def is an executable statement. So you can define the same
function several times, and the last definition befoer the call is the one
which gets called.

> while for 2), I got
>         6
>

Correct. In this case you are referring to a name which was assigned during
the class definition. Like def, class is an executable statement, and names
(whether of methods or simple variables) bound during the definition are
associated with the class. In looking for attributes of an instance Python
looks in the class definition if it doesn't find it in the instance. If ti
doesn't find it there, it looks in the superclass, and so on. So you get the
last value assigned to the class variable.

> I'm here curious why Python doesn't even report an error
> for this obvious mistake ?
>
That's because it's not a mistake. It's perfectly legitimate to bind
functions to several names.

Hope his helps.

regards
 Steve





From grumble at mailcity.com  Thu Mar  1 18:28:58 2001
From: grumble at mailcity.com (jcm)
Date: 1 Mar 2001 23:28:58 GMT
Subject: Lower bounds of lists
References: <2mjt9tcfa0o55ap99p452trjj3u7h3qst6@4ax.com>
Message-ID: <97mlvq$gi2$1@news.mathworks.com>

Michael Prager <Mike.Prager at noaa.gov> wrote:
> As a newcomer to Python, I'm curious about the choice to number
> list elements starting at zero, rather than one.  My impression
> is that there is no user option to change that in a specific
> program.  Correct?

> It seems an odd choice, as humans count from one, and Python in
> other respects seems quite logical and well thought out.
> Starting lists at zero also seems to have had ripple effects,
> such as 

> range(1,100) giving (1, 2, 3, ... , 99)

> which leads to such odd conscructions as 

> for i in range(start ,stop+1):
> 	#do something

> Though I don't consider quirks in computer software lovable, I'm
> convinced that Python's virtues far outweigh any inconvenience
> caused by this.  I am curious, though, how this choice came
> about.  Can anyone refer me to source material on that, or
> explain the logic behind it?

I like the idea of list indices beginning with 0, since the index can
then be thought of as an offset.  Not sure if this was the original
motivation though.

range's behavior is also nice, because

  range(a, b)

evaluates to the same thing as

  range(a, n) + range(n, b)

does.  No need for and +1 or -1 adjustments.
Assuming a <= n <= b of course.


From jasonsmith at jps.net  Tue Mar 13 17:34:13 2001
From: jasonsmith at jps.net (Jason D. Smith)
Date: Tue, 13 Mar 2001 14:34:13 -0800
Subject: How do I get Win32 to locate DLLs elsewhere (not really a Python thing)?
References: <mailman.984506597.12469.python-list@python.org>
Message-ID: <98m7d7$8hg$1@news2.symantec.com>

"Skip Montanaro" <skip at pobox.com> wrote in message
news:mailman.984506597.12469.python-list at python.org...
|
| How do I perform that trick so that Microsoft's equivalent of ld.so can
| find them at run-time?  I assume there is some key in the registry I can

--- REGEDIT 4 Syntax ----
[HKEY_CURRENT_USER\Environment]
PATH=(Your path here)


Jason.

--

 Hi.  I'm Bob.  I'm a tomato and I'm here to help you.




From bsb at winnegan.de  Wed Mar  7 07:44:11 2001
From: bsb at winnegan.de (Siggy Brentrup)
Date: 07 Mar 2001 13:44:11 +0100
Subject: urllib.urlopen and proxies
In-Reply-To: <vBKmOkB47iH8JfPSAvl1kJShzae+@4ax.com>
References: <vse4atgcgcfafth7c5tq3po7srd3l32f3h@4ax.com>
	<008301c0a57d$1331cde0$a100000a@local>
	<mailman.983804894.17415.python-list@python.org>
	<BfakOm9OYyn2V0Hg9h1SSS=a7t7Y@4ax.com>
	<mailman.983901970.11969.python-list@python.org>
	<vBKmOkB47iH8JfPSAvl1kJShzae+@4ax.com>
Message-ID: <87k8614ulg.fsf@winnegan.de>

Paul Moore <paul.moore at uk.origin-it.com> writes:


[...]

> >>> f = urllib2.urlopen('http://www.python.org/')
> Traceback (most recent call last):
>   File "<stdin>", line 1, in ?
>   File "c:\applications\python\lib\urllib2.py", line 137, in urlopen
>     return _opener.open(url, data)
>   File "c:\applications\python\lib\urllib2.py", line 325, in open
>     '_open', req)
>   File "c:\applications\python\lib\urllib2.py", line 304, in
> _call_chain
>     result = func(*args)
>   File "c:\applications\python\lib\urllib2.py", line 747, in http_open
>     raise URLError(err)
> urllib2.URLError: <urlopen error host not found>
> 
> I'll see if I can understand what is going on when I get a chance, but
> I don't know how far I'll get...

Well, obviously I should have tested before posting :-( There's one
omission in my code as well as three typos in urllib2.py (sample and
patch attached)

Siggy
-------------- next part --------------
A non-text attachment was scrubbed...
Name: proxy_auth.py
Type: application/octet-stream
Size: 559 bytes
Desc: proxy authorization
URL: <http://mail.python.org/pipermail/python-list/attachments/20010307/c1125dc3/attachment.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: urllib2.diff
Type: text/x-patch
Size: 654 bytes
Desc: Fix typos in urllib2.py
URL: <http://mail.python.org/pipermail/python-list/attachments/20010307/c1125dc3/attachment.bin>

From robin.thomas at starmedia.net  Fri Mar 30 13:15:21 2001
From: robin.thomas at starmedia.net (Robin Thomas)
Date: Fri, 30 Mar 2001 13:15:21 -0500
Subject: best practice? null/None (correction)
In-Reply-To: <4.3.1.2.20010330113737.00e39ab0@exchange.starmedia.net>
References: <9a2b9v+lrgp@eGroups.com>
Message-ID: <4.3.1.2.20010330131233.0256b3c0@exchange.starmedia.net>

At 11:58 AM 3/30/01 -0500, Robin Thomas wrote:
>HTTP query arguments are of one type: STRING. There is no concept of a 
>"null" value; the closest thing you have is an "empty" string like "". In 
>most database engines, there is a clear difference between NULL and "" in 
>any textual or character data column.

I'm sorry to be outright WRONG here, fluxent.

Many databases consider "" to be NULL for textual columns. Oracle 7 and up 
certainly does.

So maybe you have a really good point here.


--
Robin Thomas
Engineering
StarMedia Network, Inc.
robin.thomas at starmedia.net




From tanzer at swing.co.at  Wed Mar 14 11:51:49 2001
From: tanzer at swing.co.at (Christian Tanzer)
Date: Wed, 14 Mar 2001 17:51:49 +0100
Subject: Buffered/Unbuffered on HP-UX (was: Yet another Python vs. Perl speed issue/quest 
In-Reply-To: Your message of "Wed, 14 Mar 2001 16:06:17 GMT."
             <98o4tp+eq6o@eGroups.com> 
Message-ID: <m14dEV3-000wcBC@swing.co.at>

nanotech at europa.com wrote:

> Running the correct code (you are definitely right Christian, I posted
> "bad" Python code!), buffered/unbuffered on Linux is just as fast as
> the Perl (which *is* correct Christian; I am using Perl 5.004, what is
> yours?)

My Linux installation is not quite as old as my machine but its still
ancient. My perl is version 5.003_07 -- sorry for picking on your Perl
script.

Interesting that `map' changes so much between Perl versions.

Hint: starting python with option `-S' shaves off another 20 percent
of real time (for test2.py on my machine).

I hope you'll enjoy Python.

-- 
Christian Tanzer                                         tanzer at swing.co.at
Glasauergasse 32                                       Tel: +43 1 876 62 36
A-1130 Vienna, Austria                                 Fax: +43 1 877 66 92




From bk at xk7.com  Thu Mar  1 09:04:49 2001
From: bk at xk7.com (Burkhard Kloss)
Date: Thu, 1 Mar 2001 14:04:49 -0000
Subject: how can i copy a sublist
References: <3A9E4C6F.E2DD63C3@bg-sys.de>
Message-ID: <983455490.755326@master.nyc.kbcfp.com>

Look at copy.copy
>>> from copy import copy
>>> a=[[1,2],[3,4]]
>>> b = []
>>> b.append (copy (a [0]))
>>> b
[[1, 2]]
>>> b[0][0]=3
>>> b
[[3, 2]]
>>> a
[[1, 2], [3, 4]]

Also have a look at the documentation for copy.deepcopy to see which is more
appropriate in your case.

"Daniel Orlowski" <daniel.orlowski at bg-sys.de> wrote in message
news:3A9E4C6F.E2DD63C3 at bg-sys.de...
> Hi,
>
> i try to copy a sublist from list "a" to list "b". After that i alter a
> value in "b" and oops! value in "a" is also altered.
> so it looks like that:
> >>> a=[[1,2],[3,4]]
> >>> b=[]
> >>> b.append(a[0])
> >>> b
> [[1, 2]]
> >>> b[0][0]=3
> >>> a
> [[3, 2], [3, 4]]
>
> how can i copy a sublist without
> "get_a_value_from_a_list_and_put_it_into_another"
>
> Regards,
> Daniel
>
>




From loewis at informatik.hu-berlin.de  Sun Mar  4 05:55:01 2001
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 04 Mar 2001 11:55:01 +0100
Subject: GC and extension types
References: <jKio6.8061$3F2.2558420@typhoon.ne.mediaone.net>
Message-ID: <j4wva57qii.fsf@informatik.hu-berlin.de>

"David Abrahams" <abrahams at mediaone.net> writes:

> Anyway, I'm hoping there is authoritative text somewhere that describes
> what's really going on... and that the text isn't just 'C' code!

It may not be authoritive, but the book "Python 2" describes the GC in
some detail. That is of help only if you can read German, of course.

Regards,
Martin



From olive at conectiva.com.br  Thu Mar 15 08:53:56 2001
From: olive at conectiva.com.br (Fabio Olive Leite)
Date: Thu, 15 Mar 2001 10:53:56 -0300
Subject: This math scares me
In-Reply-To: <erRr6.659$lz1.20662@ruti.visi.com>; from grante@visi.com on Wed, Mar 14, 2001 at 09:31:54PM +0000
References: <20010312170356.B3513@better.net> <mailman.984480856.18097.python-list@python.org> <984486395.407379@newsmaster-04.atnet.at> <slrn9asf1d.ja3.qrczak@qrnik.zagroda> <984518199.127140@newsmaster-04.atnet.at> <jwbnews-09D75F.13313313032001@corp.supernews.com> <98ohuc$du9$1@news.udel.edu> <7bQr6.644$lz1.17686@ruti.visi.com> <98oku1$f6a$1@news.udel.edu> <erRr6.659$lz1.20662@ruti.visi.com>
Message-ID: <20010315105356.B1757@conectiva.com.br>

Hi there,

On Wed, Mar 14, 2001 at 09:31:54PM +0000, Grant Edwards wrote:
) No.  The binary floating point representation of many decimal
) fraction values is not exact.  That's what started this whole
) thread.  You can't represent 0.1 exactly in binary FP.  You can
) in BCD.

Just to make matters worse in this discussion, let me add some more noise.

Since base 10 contains base 2 and base 5 (`factor 10`), every number coming
from any of those bases is exactly represented in base 10. Conversely, base
2 contains only itself, and thus it cannot represent correctly some numbers
that are exactly representable in other bases, like 0.1 in base 10.

The usual 1/3 example is a problem for base 10, but not so for base 12, for
example, since 12 contains 3 2 2. Thus:

1_{12} / 3_{12} == 0.4_{12}

And then base 30 would let us count exactly fractions of base 2, 3, 5, 6,
10, 12, 15, 16 and all the others you can think of that have only 2, 3 and
5 as factors of the base. But since no one (usualy) has 30 fingers, that
base is not terribly popular. :)

Please shoot me if this is just a bad dream.

F?bio
-- 
( F?bio Oliv? Leite -*- http://www.conectiva.com.br/~olive )



From grante at visi.com  Wed Mar  7 12:26:11 2001
From: grante at visi.com (Grant Edwards)
Date: Wed, 07 Mar 2001 17:26:11 GMT
Subject: Can anyone recomend a good intoduction to C...
References: <3aa366c6_2@nnrp1.news.uk.psi.net> <8gNo6.440$y6.128357@ruti.visi.com> <3aa6551e.212870101@news.laplaza.org>
Message-ID: <Taup6.1292$y6.235409@ruti.visi.com>

In article <3aa6551e.212870101 at news.laplaza.org>, Mats Wichmann wrote:

>(oops, I smell a flood of "I can top that" comments such as "you had a
>terminal? I learned on a Teletype", "you had a Teletype? I had to
>toggle code in using switches", "you had switches?  I had to use
>Forth!".  Consider that joke already played out, eh).

You had 1's?  In my day we had to write everything using nothign but 0's!

[That's out of some Dilbert strip I believe]

-- 
Grant Edwards                   grante             Yow!  On the road, ZIPPY
                                  at               is a pinhead without a
                               visi.com            purpose, but never without
                                                   a POINT.


From loewis at informatik.hu-berlin.de  Thu Mar 22 04:41:48 2001
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 22 Mar 2001 10:41:48 +0100
Subject: New PEP: The directive statement
References: <mailman.985085006.24236.python-list@python.org> <3AB82078.8060005@caltech.edu.is.my.e-mail.address>
Message-ID: <j4hf0mywbn.fsf@informatik.hu-berlin.de>

Nathan Gray <n8gray at caltech.edu.is.my.e-mail.address> writes:

> Excellent work, Martin.

Thanks!

> Would you consider using this sort of syntax instead?
> 
>    directive_statement: "directive" directive_ID ("," directive_ID)* 
>                             [";"] NEWLINE
>    directive_ID:        identifier ("." identifier)*
> 
> I understand that there are those who want to keep things as dead-simple as
> possible, and there are good reasons for *not* wanting people to put multiple
> directives on one line

It would be certainly possible to make the directive statement a
simple_stmt, so you could write

directive foo; print "Hello"; directive bar "value"

As for extending the syntax of the directive statement itself: That
can be done when need occurs. As it stands, the compiler needs to be
taught every new directive on its own, so it may turn out that the
syntax is insufficient one day - just as the import syntax turned out
to be insufficient twice (adding packages first, and "import as"
recently).

Please note that the third element of a directive is an atom, which
allows quite some expressivenes.

> My second comment is that you may want to consider adopting a form of the 
> __future__ proposal's runtime semantics for directive statements.  Perhaps 
> 
>     directive transitional.nested_scopes
> 
> could create a "_Feature object" called nested_scopes identical to the one
> proposed in "back to the __future__".
> 
>     print nested_scopes.getOptionalRelease()   # prints "(2, 1, 0, 'beta', 1)"

I have my problems with that part of PEP 236 also. While it is
certainly possible to provide the release when a feature was first
released, it is not possible to give an accurate answer to the
question when it will become mandatory. The intent to make it
mandatory in the future should go into the documentation, IMO.

Furthermore, what is the use of that information to a program?

Regards,
Martin


From aahz at panix.com  Thu Mar 15 10:40:48 2001
From: aahz at panix.com (Aahz Maruch)
Date: 15 Mar 2001 07:40:48 -0800
Subject: This math scares me
References: <20010312170356.B3513@better.net> <erRr6.659$lz1.20662@ruti.visi.com> <mailman.984664520.26463.python-list@python.org> <HR4s6.877$lz1.73512@ruti.visi.com>
Message-ID: <98qnq0$qc2$1@panix3.panix.com>

In article <HR4s6.877$lz1.73512 at ruti.visi.com>,
Grant Edwards <grante at visi.com> wrote:
>In article <mailman.984664520.26463.python-list at python.org>, Fabio Olive Leite wrote:
>>
>>The usual 1/3 example is a problem for base 10, but not so for
>>base 12, for example, since 12 contains 3 2 2. Thus:
>>
>>1_{12} / 3_{12} == 0.4_{12}
>>
>>And then base 30 would let us count exactly fractions of base
>>2, 3, 5, 6, 10, 12, 15, 16 and all the others you can think of
>>that have only 2, 3 and 5 as factors of the base. But since no
>>one (usualy) has 30 fingers, that base is not terribly popular.
>>:)
>
>One step further is base-60, which is still used for many
>things.  Time and angular measurement for example is partially
>a base-60 system.  I believe that one or more of the ancient
>Aztecs/Incans/Mayans used a base-60 number system.

Paging Ivan Van Laningham.....
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

"The overexamined life sure is boring."  --Loyal Mini Onion


From tim.one at home.com  Sat Mar 24 16:47:48 2001
From: tim.one at home.com (Tim Peters)
Date: Sat, 24 Mar 2001 16:47:48 -0500
Subject: PEP 245: Python interfaces
In-Reply-To: <Pine.LNX.4.32.0103241227160.26302-100000@localhost.localdomain>
Message-ID: <LNBBLJKPBEHFEDALKOLCKEBOJIAA.tim.one@home.com>

[Martin von Loewis, on
 http://python.sourceforge.net/peps/pep-0245.html
]
> Even though this is quite a long document, it is incomplete.

[Michel Pelletier]
> Yes, *sigh*.  I know I need more formalism, but should this be at the
> expense of the section on goals?  I'm not sure.

Shoot for "in addition to" rather than "at the expense of".  The formal
specification sections needn't contain any motivation at all (indeed, are
probably better if they don't!); contrarily, the motivation sections can
defer to the specification sections for details.

> I was raised in an XP environment, the goals are the most important
> part to me.  This is something I've been debating myself over.  I
> obviously want the PEP to be accessable and not boring, but I also
> want it to be comprehensive.

To the extent possible, a PEP should be complete and precise enough to allow
someone from Mars to implement it, and leave you happy with the result in all
respects.

Don't forget that we're implementing a programming language here:  what the
PEP specifies has to cover all possible cases, and even the tiniest
advertised details have to remain stable for years and years to come.  Like
it or not, it's very rare that we get to do any form of refactoring in a
language or library spec once released:  people rely on the details within
hours of getting it.  Heck, look at the person-decades of turmoil caused by
the mere *suggestion* that Python make a trivial change like having 1/2
return 0.5 instead of 0 <wink>.

*these*-goals-can't-be-met-until-the-details-are-precise-ly y'rs  - tim




From gregj at pobox.com  Sun Mar 11 18:57:15 2001
From: gregj at pobox.com (Gregory Jorgensen)
Date: Sun, 11 Mar 2001 23:57:15 GMT
Subject: random number between -maxint and +maxint
References: <K2Sq6.360684$Pm2.5406628@news20.bellglobal.com>
Message-ID: <vhUq6.2211$54.2265@www.newsranger.com>

Use long integers instead of ints. The whrandom module's internal methods use
int() to make sure you passed an integer; they could just as well use long().
Look at the whrandom source.

Try this:

>>> import whrandom
>>> whrandom.seed()
>>> maxint = 2L**31
>>> r = long(whrandom.random() * (maxint + maxint + 1)) - maxint

The expression (maxint + maxint + 1) is the range of random long integers you
want: [0,maxint].


In article <K2Sq6.360684$Pm2.5406628 at news20.bellglobal.com>, Dan Parisien
says...
>
>I can't figure out a good way to do it, does anyone know a better solution 
>than
>
>from whrandom import whrandom
>rand = whrandom()
>max = 2**30 + (2**30-1) #one less than 2**31
>randomnumber = rand.randrange(0, max) - rand.randrange(0, max)
>
>thank you for your time,
>Dan

Greg Jorgensen
Deschooling Society
Portland, Oregon, USA
gregj at pobox.com


From francois.granger at free.fr  Sat Mar 10 14:01:21 2001
From: francois.granger at free.fr (=?ISO-8859-1?Q?Fran=E7ois_Granger?=)
Date: Sat, 10 Mar 2001 19:01:21 GMT
Subject: For french speaking Pythoneer
Message-ID: <1eq2jkp.1vrsbtvwllg84N%francois.granger@free.fr>

Sorry for this message in french. There is a UsenetFR vote on creating
the newsgroup fr.comp.lang.python. And I feel important to get as many
french pytoneer as possible to vote for this.


Le premier appel ? voter pour la cr?ation d'un groupe de discussion sur
Python est paru.

Allez dans  et cherchez le message dont le sujet est :
[AAV 1] Creation de fr.comp.lang.python (non-modere)
Message-ID: <1epxces.18yyneu19bvnu0N%massiot at via.ecp.fr>

dans le NG
fr.usenet.forums.annonces

et votez ;-)


PS: "Votez OUI, votez NON mais votez"
PPS: "Je vous parle d'un temps / Que les moins de vingt ans / Ne peuvent
pas connaitre..."

;-)


From donn at oz.net  Fri Mar 16 01:13:09 2001
From: donn at oz.net (Donn Cave)
Date: 16 Mar 2001 06:13:09 GMT
Subject: Should I close after popen??
References: <XFMail.20010315142704.shaleh@valinux.com> <mailman.984697040.2458.python-list@python.org> <m3lmq6ppv0.fsf@master.athome>
Message-ID: <98satl$1t8$0@216.39.151.169>

Quoth Andrew Markebo <flognat at flognat.myip.org>:
[ ... re should close file explicitly ...]

| I agree, but I tried with close, but I got the following problems:
|
| Traceback (innermost last):
|   File "<stdin>", line 1, in ?
|   File "/usr/tmp/python-18803PG", line 314, in ?
|     mymain()
|   File "/usr/tmp/python-18803PG", line 304, in mymain
|     keepALive()
|   File "/usr/tmp/python-18803PG", line 259, in keepALive
|     if PingAway():
|   File "/usr/tmp/python-18803PG", line 230, in PingAway
|     fd1.close()
| IOError: [Errno 10] No child processes
|
| The code is:
|
|
|     fd1=os.popen("ping -q -c 3 %s 2> /dev/null" % name1)
|     fd2=os.popen("ping -q -c 3 %s 2> /dev/null" % name2)
|     pid1, res1=os.wait(fd1)
|     out1=fd1.read()
|     pid2, res2=os.wait(fd2)
|     out2=fd2.read()
|     fd1.close()
|     fd2.close()

For a moment there I was awfully confused, since Python won't
even try to run that code, but I see you're running an earlier
version, before 2.0.

os.wait() has nothing to do with files.  Before 2.0, it ignored
its argument(s) altogether, so you could call it with a parameter
without any ill effect, but of course without any good effect either.
Its purpose is to reap the status from a fork's exit.  I see you
have evidently figured that much out, but 1) it's just any fork,
probably the first to exit that hasn't been reaped yet, 2) you
can reap this status only once per fork, and 3) the file object
created by popen() has a special close() method that does this
(man pclose).  Except it doesn't use wait(), but rather waitpid()
specifying its own particular process.

So you snatched the process exit status away from underneath the
pclose(), hence the error.

Files normally (in C Python, anyway) close implicitly when they
go out of scope and no reference to them remains in the system.
There's nothing wrong with an explicit close, though, and in
popen() files it also releases the process slot, which until
then (or until the parent's exit)  will appear as a "zombie".

	Donn Cave, donn at oz.net


From tjreedy at udel.edu  Mon Mar  5 00:04:35 2001
From: tjreedy at udel.edu (Terry Reedy)
Date: Mon, 5 Mar 2001 00:04:35 -0500
Subject: Newbie question about nested lists and constructor functions...
References: <W1Do6.4683$ke1.121750@jekyl.ab.tac.net>
Message-ID: <97v6sl$ov8$1@news.udel.edu>

"Bartok" <JollyRoger at hotmail.com> wrote in message
news:W1Do6.4683$ke1.121750 at jekyl.ab.tac.net...
> I need some help with nested lists when using class constructors to
> initialize their contents.  I found the tutorial to be of not much help
in
> this area and have tried everything that I can think of.

Constructing a list, nested or not, within __init__() is no different from
doing so anywhere else.

> I understand the process as I am an experienced C++ programmer but I seem
to
> be having some difficulty using the correct syntax.  Can anyone please
help?

It would be much easier to do so if you post both a specific example of
what you want to do and what you have tried.





From andreas at andreas-jung.com  Sat Mar 17 16:58:33 2001
From: andreas at andreas-jung.com (Andreas Jung)
Date: Sat, 17 Mar 2001 16:58:33 -0500
Subject: image handling
References: <Pine.LNX.4.10.10103171013150.10770-100000@localhost.localdomain>
Message-ID: <004a01c0af2d$6fb89210$9865fea9@SUXLAP>

There is no native support inside Zope for the kind of
image manipulation you need. However you can use PIL to manipulate images in
many ways.
Take a look at http://www.zope.org/Member/ajung/AJgallery as an example to
integrate PIL in Zope or Python.

Andreas
----- Original Message -----
From: "Michael Hall" <admin at mulga.com.au>
To: <python-list at python.org>
Sent: Friday, March 16, 2001 7:50 PM
Subject: image handling


>
> G'day people:
>
> I'm brand new on this list and am looking for resources about image
> processing in a web environment in Python. I'm wondering if there exists a
> module that does things like the PHP image handling functions
> (creating, copying, resizing, etc).
>
> I've had a look at the Python Functions Library but there doesn't appear
> to be much in there. I've started to use Python a bit in a web environment
> and like what I've seen so far, and from what I've heard I'm sure there
> must be some pretty flash image handling capabilities hanging around
> somewhere.
>
> Thanks
> Mick
>
> ***********************************************
> Michael Hall - mick at mulga.com.au - mulga.com.au
> ***********************************************
>
>
>
> --
> http://mail.python.org/mailman/listinfo/python-list








From jkraska1 at san.rr.com  Wed Mar  7 21:24:24 2001
From: jkraska1 at san.rr.com (Courageous)
Date: Thu, 08 Mar 2001 02:24:24 GMT
Subject: writing a C function to manipulate python lists
References: <jllmqpnclz.fsf@endor.bbb.caltech.edu>
Message-ID: <njrdat89p7cpv5fs072f98fqndfae13c0i@4ax.com>

>If anyone has some code lying around that does something like this, could
>they pass it on to me?

Get the book _Essential Python Reference_. It's section on extending
and embedding is outstanding. After you've read that, get the source
code release of python. Look at the sample object implementation, and
then at the native source for Python lists right in the raw.

It's really all very straighforward once you understand the dispatch
mechanism and reference counting (the latter of which is a bit tricky).


C//



From wayne.izatt at _nospam_myself.com  Tue Mar  6 00:18:19 2001
From: wayne.izatt at _nospam_myself.com (Wayne Izatt)
Date: Tue, 06 Mar 2001 05:18:19 GMT
Subject: Python & mySQL
References: <DJQo6.27114$c7.7825529@news3.rdc1.on.home.com>
Message-ID: <vq_o6.183437$df5.5138495@news1.crdva1.bc.home.com>

It really all depends on what your ISP is prepared to do for you. As long as
your ISP installs the MySQLda module, you should be off to the races. What
you have on your machine won't really be a factor, although for testing
you'll probably want to install MySQL and MySQLda. I believe the module is
available as a dll, so you won't need to recompile Python.

There are ample resources available to get you going from a programming
perspective. I can probably dredge up some samples.

cheers


"Colin Meeks" <colinmeeks at home.com> wrote in message
news:DJQo6.27114$c7.7825529 at news3.rdc1.on.home.com...
> Does anybody have any pointers for integrating mySQL with Python.  I've
been
> using Python a while and am looking to develop a web backend using Python
> with mySQL.  My site is hosted by a Linux based ISP who has both Python
and
> mySQL but they don't currently have a link between the two.  I'm using
> Windows2000 as my development base, so I want something that is usuable on
> both, so it's easier for me to track any problems.
>
> Ideally I want something that doesn't involved recompiling Python if
that's
> possible, as i want the installation to be as straightforward for my ISP
and
> of course me :-)
>
> Many thanks in advance
>
> Colin
>
>




From akuchlin at mems-exchange.org  Thu Mar 29 11:13:18 2001
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 29 Mar 2001 11:13:18 -0500
Subject: Threading legality/morality
References: <mailman.985831297.32578.python-list@python.org>
Message-ID: <3d66gswo2p.fsf@ute.cnri.reston.va.us>

hungjunglu at yahoo.com writes:
> For whatever it's worth, I tried to build Python2.0 from source on 
> cygwin, and it has problem with some pthread thing. I had to disable 
> thread support when doing ./configure. If I have to guess, cygwin may 
> have problems with thread support.

Jason Tishler submitted a truckload of patches for Cygwin
compatibility that went into 2.1; you could try 2.1b2 and see if it
works better.

--amk



From stephen_purcell at yahoo.com  Thu Mar 22 05:13:19 2001
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Thu, 22 Mar 2001 11:13:19 +0100
Subject: automatic cookie handling with urllib?
In-Reply-To: <3AB9BFA9.56301B22@ugcs.caltech.edu>; from boswell@ugcs.caltech.edu on Thu, Mar 22, 2001 at 01:02:33AM -0800
References: <3AB9BFA9.56301B22@ugcs.caltech.edu>
Message-ID: <20010322111319.B4949@freedom.puma-ag.com>

dustybos wrote:
> Has anyone run into a similar problem or know of a good
> library/module/whatever
> to make cookies automatically handled like redirects?

You might be able to crib some code from something I hacked up recently:

  http://webunit.sourceforge.net/

Best wishes,

-Steve

-- 
Steve Purcell, Pythangelist
Get testing at http://pyunit.sourceforge.net/
Any opinions expressed herein are my own and not necessarily those of Yahoo



From robin at alldunn.com  Fri Mar 23 16:58:02 2001
From: robin at alldunn.com (Robin Dunn)
Date: Fri, 23 Mar 2001 13:58:02 -0800
Subject: [wxPython] FW: Accessing Python stuff thru OLE...
References: <6176E3D8E36FD111B58900805FA7E0F80CCF6473@mcnc-mdm1-ex01>
Message-ID: <014001c0b3e4$55fdb6a0$0100a8c0@Rogue>

>
> I used the example code at the end of this post.  I took the MyAboutBox
from
> the wxPython demo and tried opening it thru my OLE object.  Works great
the
> first time, but I get the only 1 instance of wxApp allowed per thread
> exception afterwords.  Anyway to fix this?

What version of wxPython do you have?  I went to remove this problem a
couple days ago and found that I had already ifdef'd it out of the code.  I
had forgotten that I was already experimenting with it.  Anyway, that change
was in version 2.2.5 so it should not have the restriction.  I've pasted in
a session from PythonWin below that shows this in action.

The main thing to watch out for now is that all wxPython stuff needs to
happen in the same thread as wxWindows keeps track of what it thinks is the
"main thread" and tries to ensure that everything to do with the GUI happens
there.

One little potential problem in your sample code is that you create a Dialog
but then treat it like a frame.  One of the differences between them is that
the dialog does not automatically delete itself when it is closed, so you
will have to call dialog.Destroy when you are done with it.



PythonWin 2.0 (#8, Oct 16 2000, 17:27:58) [MSC 32 bit (Intel)] on win32.
Portions Copyright 1994-2000 Mark Hammond (MarkH at ActiveState.com) - see
'Help/About PythonWin' for further copyright information.
>>> from wxPython.wx import *
>>> app = wxPySimpleApp()
>>> f = wxFrame(None, -1, "Hello", size=(150,75))
>>> b = wxButton(f, -1, "Click Me")
>>> f.Show(true)
1
>>> def hello(evt):
...  print "HELLO:", evt
...
>>> EVT_BUTTON(f, b.GetId(), hello)
>>> app.MainLoop()
HELLO: <C wxCommandEvent instance at _12f244_wxCommandEvent_p>
HELLO: <C wxCommandEvent instance at _12f244_wxCommandEvent_p>
HELLO: <C wxCommandEvent instance at _12f244_wxCommandEvent_p>
HELLO: <C wxCommandEvent instance at _12f244_wxCommandEvent_p>
HELLO: <C wxCommandEvent instance at _12f244_wxCommandEvent_p>
HELLO: <C wxCommandEvent instance at _12f244_wxCommandEvent_p>
>>>
>>> app2 = wxPySimpleApp()
>>> f = wxFrame(None, -1, "Hello", size=(150,75))
>>> b = wxButton(f, -1, "Click Me")
>>> EVT_BUTTON(f, b.GetId(), hello)
>>> f.Show(true)
1
>>> app2.MainLoop()
HELLO: <C wxCommandEvent instance at _12f244_wxCommandEvent_p>
HELLO: <C wxCommandEvent instance at _12f244_wxCommandEvent_p>
HELLO: <C wxCommandEvent instance at _12f244_wxCommandEvent_p>
>>>

--
Robin Dunn
Software Craftsman
robin at AllDunn.com       Java give you jitters?
http://wxPython.org      Relax with wxPython!







From tariq.rashid at iname.com  Tue Mar 27 18:15:25 2001
From: tariq.rashid at iname.com (Taz)
Date: Wed, 28 Mar 2001 00:15:25 +0100
Subject: wavlets?
Message-ID: <tc27anovquqm79@corp.supernews.co.uk>

i'm about to write a wavelets module for python - does one already exist...
couldn't find one?

taz




From amitp at Xenon.Stanford.EDU  Sun Mar 18 12:07:32 2001
From: amitp at Xenon.Stanford.EDU (Amit Patel)
Date: 18 Mar 2001 17:07:32 GMT
Subject: Who's minister of propaganda this week?
References: <B98041329020E007.7DA84CAD9D0562AB.817BBC572F480869@lp.airnews.net> <3AB0176D.8070903@yahoo.com> <mailman.984663621.25421.python-list@python.org> <slrn9b9gkq.ahm.qrczak@qrnik.zagroda>
Message-ID: <992q0k$eq9$1@nntp.Stanford.EDU>

 Marcin 'Qrczak' Kowalczyk <qrczak at knm.org.pl> wrote:
| 
| In many languages (Haskell, SML, OCaml, Clean, Mercury, Eiffel) they
| are almost never used. Again, not all languages are like C++ and Java
| where casting is essential.
| 

Type casting does not seem to be essential in C++ except when dealing
with low level interfaces (to the OS, usually through C libraries).
Unfortunately "cast" is used in two different ways in C and C++.
There is the type cast used to get around the type system.  For
example, (int)(some_pointer).  But ordinary conversions, such as
converting 3 to 3.0, or from Dog& to Animal&, are labelled "cast".
The former kind is the kind we want to stay away from, and is usually
reinterpret_cast in C++.  The latter is present in other languages but
isn't called a "cast" in those languages.

So yes, you could say casting (in the C sense) is essential in C++,
but in my experience, these essential casts are almost always safe
conversions and not the unsafe type casts.

In Java on the other hand, casting is essential when working with generic
container classes.  :-(

    - Amit
-- 
--
Amit J Patel, Computer Science Department, Stanford University
http://www-cs-students.stanford.edu/~amitp/


From sholden at holdenweb.com  Wed Mar  7 18:52:28 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Wed, 7 Mar 2001 18:52:28 -0500
Subject: cursor.execute fails under mxODBC
References: <mailman.983993842.26448.python-list@python.org>
Message-ID: <tUzp6.29925$ML1.1507874@e420r-atl2.usenetserver.com>

<bill_tolbert at bigfoot.com> wrote in message
news:mailman.983993842.26448.python-list at python.org...
> I just upgraded to the latest version of mxODBC. Calling execute on a
> cursor object is hosing python. This code worked perfectly under 1.5.1
> and version 1.1.1 of mxODBC.
>
> Having trouble on both Win98 and Win2000.
>
> I can stay on windows, so I'm gonna give ADO a try, but I'd still like
> to know what's going on.
>
> Thanks,
>
> Bill
>
So would we! What error messages are you seeing, form what code?

regadrs
 Steve





From gregj at pobox.com  Sat Mar 31 21:17:42 2001
From: gregj at pobox.com (Greg Jorgensen)
Date: Sun, 01 Apr 2001 02:17:42 GMT
Subject: Couple of simple questions
References: <3ac4854a.1170697@news.is.co.za>
Message-ID: <adwx6.657193$U46.20196569@news1.sttls1.wa.home.com>

"nomad" <nomad***@***freemail.absa.co.za> wrote:

> Does anybody know what size python variabel types are?
> Eg. how many bytes is and integer?  What about a long?  I have
> searched through the docs but I cant find a reference to the byte
> sizes.  The reason is that I'm battling with struct.py and some
> complex C++ structs.  Any help here would greatly appreciated

The range of values for integers, long integers, floats, and doubles are
architecture-dependent. From the Python Reference Manual section 2.4.4:

    Plain integer decimal literals must be at most 2147483647 (i.e., the
largest
    positive integer, using 32-bit arithmetic). Plain octal and hexadecimal
literals
    may be as large as 4294967295, but values larger than 2147483647 are
    converted to a negative value by subtracting 4294967296. There is no
limit
    for long integer literals apart from what can be stored in available
memory.

So depending on the CPU and OS, an integer may occupy 1, 2, 3, or 4 bytes.
On a 32-bit OS a Python integer is equivalent to a long int in C/C++.

C and C++ don't have a built-in type analogous to Python's long integers.

You are on the right track using the struct module to map C++ structs.
Besides the relatively simple mapping of variables based on size, you may
have to deal with big-endian/little-endian bit ordering, and possibly struct
padding/alignment. You should refer to your C++ compiler documentation to
find out how it aligns struct objects.

--
Greg Jorgensen / programmer, pedant, raconteur / Portland, Oregon, USA




From tim at vegeta.ath.cx  Thu Mar  8 20:45:43 2001
From: tim at vegeta.ath.cx (Tim Hammerquist)
Date: Fri, 09 Mar 2001 01:45:43 GMT
Subject: "!=" is bad form. Re: sorry....never mind
References: <DZkp6.18811$pF4.3015652@news1.telusplanet.net> <fdlp6.16982$go.2157953@news0.telusplanet.net> <3AA6E308.5B63ECC6@cosc.canterbury.ac.nz> <3AA72253.FB021815@Lugoj.Com> <slrn9aehgr.33d.tim@vegeta.ath.cx> <984047188.296438@newsmaster-04.atnet.at>
Message-ID: <slrn9agdkv.9ms.tim@vegeta.ath.cx>

Werner Schiendl <ws-news at gmx.at> wrote:
> > I hate '<>' because it's too much like VB (and so many other BASIC
> > variants).
> 
> I think there are at least the same number of programmers that hate the
> other way because it looks to much like C.
> Maybe we should invent some completely new operator, ending up with line
> noise one day... :-]

I realize I seem to be in the minority in this group when I say I like
C/C++.  NB: I like them as _languages_, but would prefer NOT to do
everyday programming in them, as they're just a pain in the ass most of
the time.

I started out programming in BASIC on the TRS-80 and remember the utter
confusion upon seeing the '<>' operator ("It's greater than _and_ less
than?")  I had no Pascal background.  I like C's '!=' because it just
makes more sense to my logical, math-adept brain.

I have no problem with anyone else using it, and if I saw it in
someone's code, I would understand it.  I just prefer not to use it.
And since I've not heard any mention of either inequality operator being
deprecated, I have no cause to change.  ;)

-- 
-Tim Hammerquist <timmy at cpan.org>

I don't have any solution, but I certainly admire the problem.
	-- Ashleigh Brilliant


From thomas.heller at ion-tof.com  Thu Mar  1 06:49:40 2001
From: thomas.heller at ion-tof.com (Thomas Heller)
Date: Thu, 1 Mar 2001 12:49:40 +0100
Subject: Unresolved symbols in MSVCRT.DLL when running a py2exe packaged program.
References: <RKln6.6630$hn5.853203@news1.rdc1.mb.home.com>
Message-ID: <97ld0l$pmjld$1@ID-59885.news.dfncis.de>

"Parzival Herzog" <parz at home.com> wrote in message news:RKln6.6630$hn5.853203 at news1.rdc1.mb.home.com...
> I have a Python application (which uses the wxPython extension) that I have
> packaged
> for distribution with py2exe.  The dfistributed app works on my own
> computer,
> and on another machine I have, which has never seen any Python software, nor
> any
> development software (i.e. no dll's not a part of a standard Win2000
> installation are present).
>
> I delivered my app to my client, who is expected to modify some
> configuration code, then
> use py2exe to package the app, and run the packaged app on a server machine.
> My client
> has a notebook computer, running Win NT (Sp 6), plus MsVc and assorted
> development tools.
> My app worked on that noebook, and py2exe again packaged the app and the
> package was
> copied to the server machine, a Win NT sp 6 desktop computer. When we
> attempted to
> start the app, it failed immediatly, with the message that the symbol
> ??_U at YAPAXI@Z could
> not be found in MSVCRT.DLL.
[...]

You should try and simply delete the MSVCRT.DLL which py2exe copies
into the dist directory for your application and try to run the app.
MSVCRT.DLL is a system file, which means that it should live in the
system (or system32) directory, and it should only be upgraded with
proper version checking.
Installer programs (like WISE or Inno Setup) can do this for you,
you should not do it 'manually' (normally you cannot replace it manually
because it is in use by the system).

I intend to change the next version (0.2.4) of py2exe in a way that
these system files are _not_ copied into the dist directory.

Thanks for trying it,

Thomas Heller




From max at alcyone.com  Mon Mar  5 23:44:44 2001
From: max at alcyone.com (Erik Max Francis)
Date: Mon, 05 Mar 2001 20:44:44 -0800
Subject: PIL and Windows file formats?
References: <3AA45311.150DB844@alcyone.com> <GdZo6.24467$1D5.1007863@e420r-atl1.usenetserver.com> <7u8zmjil03.fsf@gnwy100.wuh.wustl.edu>
Message-ID: <3AA46B3C.F7F82F49@alcyone.com>

Christopher Lee wrote:

> Is the PIL directory part of your python path?  The png image plugin
> is
> only loaded automatically if the PIL directory is on your pythonpath.
> 
> [Are you perhaps using PIL as a package (as in import PIL.Image )?]
> 
> You can load the plugin yourself.  In a message on the image-sig list,
> /F
> suggested:
> 
>     from PIL import Image, PngImagePlugin # etc.

Aha.  That didn't occur to me; I had presumed all the relevant formats
would be loaded automatically (or, rather, PIL.Image would know to load
them as needed).

I'm seeing other issues now, but they seem to be general Windows
misbehaviors rather than this stumbling block.  Thanks!

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ Nationalism is an infantile sickness.
\__/ Albert Einstein
    Official Buh rules / http://www.alcyone.com/max/projects/cards/buh/
 The official rules to the betting card game, Buh.


From dnew at san.rr.com  Sat Mar 31 20:11:58 2001
From: dnew at san.rr.com (Darren New)
Date: Sun, 01 Apr 2001 01:11:58 GMT
Subject: list vs tuple
References: <mailman.986066384.30712.python-list@python.org> <d1rx6.1053$p5.4024@news1.rivrw1.nsw.optushome.com.au> <slrn9cciaj.3ug.qrczak@qrnik.zagroda> <lWrx6.1057$p5.4152@news1.rivrw1.nsw.optushome.com.au>
Message-ID: <3AC6805F.7E26327A@san.rr.com>

deadmeat wrote:
> Are you saying there is no difference (IN THE *RESULT*) between these?  Are
> you able to read?

a = [1, 2, 3]
b = a
b = [4, 5, 6]

a is still [1, 2, 3]. What's the problem?


-- 
Darren New / Senior MTS & Free Radical / Invisible Worlds Inc.
San Diego, CA, USA (PST).  Cryptokeys on demand.
        schedule.c:7: warning: assignment makes calendar_week 
                          from programmer_week without a cast.


From syver at NOSPAMcyberwatcher.com  Wed Mar 14 17:06:57 2001
From: syver at NOSPAMcyberwatcher.com (Syver Enstad)
Date: Wed, 14 Mar 2001 23:06:57 +0100
Subject: Package
References: <mailman.984607039.2853.python-list@python.org>
Message-ID: <98oq69$1e1$1@troll.powertech.no>

Place a file named __init__.py in a directory that is in your python path or a subdirectory of an existing package that is on your python path. 

  "Anbuselvam Palusamy" <anbu at adsc.com> wrote in message news:mailman.984607039.2853.python-list at python.org...
  Can any one explain how to create a packages in python?

  Thanks
  Anbu
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20010314/74078eec/attachment.html>

From vdanovich at empireone.net  Sun Mar 18 03:39:09 2001
From: vdanovich at empireone.net (Vyacheslav Danovich)
Date: Sun, 18 Mar 2001 08:39:09 GMT
Subject: standalone
Message-ID: <Nu_s6.244274$v.26337040@typhoon.nyroc.rr.com>

Hi!
I'm pretty new to Python.
I need to run an executable Python script in Windows environment which does
not have Python library installed.
My questions are: can it be done and if "yes" what steps should I take


Thank you very much

--
############################
Mr. Vyacheslav L. Danovich
103 Bleeker Rd Apt 7
Guilderland, NY 12084
Telephone: (518) 452-4403
Facsimile: (518) 452-4403
Mobile : (518) 221-4403
############################




From benjamin.altman at noaa.gov  Wed Mar 28 12:58:10 2001
From: benjamin.altman at noaa.gov (Benjamin.Altman)
Date: Wed, 28 Mar 2001 12:58:10 -0500
Subject: list vs tuple
References: <hGlw6.782$p5.3210@news1.rivrw1.nsw.optushome.com.au>
Message-ID: <3AC22631.1F687E46@noaa.gov>

The Tutorial "Dive Into Python" (link from the python web site
http://www.python.org/doc/Intros.html or http://diveintopython.org/toc.html)
says:

1. Tuples are faster than lists. If you're defining a constant set of values
and all you're ever going to do with it is iterate through it, use a tuple
instead of a list...
2. Tuples are one of those types [that]... can be used as keys in a
dictionary, but lists can't.
3. Tuples are used in string formatting.

Also (same source):

  You can't add elements to a tuple. Tuples have no append or extend method.

  You can't remove elements from a tuple. Tuples have no remove or pop
method.
  You can't find elements in a tuple. Tuples have no index method.
  You can, however, use in to see if an element exists in the tuple.

regards,
Ben






deadmeat wrote:

> Whats the practical difference between a list[] and a tuple() ?
>
> A tuple seems to be a very basic list, so unless it's faster, what's the
> point of it?



From Bill.Scherer at VerizonWireless.com  Thu Mar 22 07:22:25 2001
From: Bill.Scherer at VerizonWireless.com (Bill Scherer)
Date: Thu, 22 Mar 2001 07:22:25 -0500
Subject: Accessing Oracle from Python remotely ?
References: <Xns906AB9C901807PaCmAnRDLM@161.48.76.20> <eE3u6.776$9%.203510@newsr1.maine.rr.com> <99bj2o$u91$2@nntp9.atl.mindspring.net>
Message-ID: <3AB9EE81.30AF90B2@VerizonWireless.com>

Andreas Jung wrote:
> 
>  There are currently two different DCOracle versions available. DCOracle
> 1.3.x
> supports is mostly build on top of Oracle 7 but I works although with Oracle
> 8i
> except features like BLOBs and CLOBs. When you need complete support
> for Oracle 8i take a look at Matt Kromers new version of DCOracle 2. It is
> still alpha so be warned :-) You can find DCO2 on www.zope.org  - search for
> DCO2.
> 
> Andreas Jung
> Digital Creations
> 

I'll note that the above is note strictly true.  I use blobs with Oracle
8.1.6 and DCOracle 1.3.(2?) all the time and it works fine.  The syntax
for this operation is probably not DB API compliant, but it does work. I
have spoken with Matt about this and he is aware of it. I can only
assume that clobs work similarly.

- Bill



From robin.thomas at starmedia.net  Fri Mar  2 11:57:01 2001
From: robin.thomas at starmedia.net (Robin Thomas)
Date: Fri, 02 Mar 2001 11:57:01 -0500
Subject: Slice operation changes
In-Reply-To: <97ncle$tef$1@216.39.151.169>
References: <15006.46879.607192.367739@w221.z064000254.bwi-md.dsl.cnc.net>
 <200103012101.QAA18516@cj20424-a.reston1.va.home.com>
 <mailman.983485324.29650.python-list@python.org>
Message-ID: <4.3.1.2.20010302111904.00cfab10@exchange.starmedia.net>

At 05:55 AM 3/2/01 +0000, Donn Cave wrote:

>The problem for me is that when we write ob[i:j], we clearly
>must have expected that ob is a sliceable sequence.  If it is
>instead a mapping, then the expression has no sensible meaning,
>so it's an error and ought to raise an exception.

Making slice objects comparable/hashable is at cross purposes with 
forbidding their use as dictionary keys. Guido endorsed the strict 
forbiddance of slice objects as dict keys; that patch is more 
straightforward than extending the slice objects' behavior.

Proposed 2.1 patch:

- no changes to slice object
- Slice objects are rejected as dict keys, just as lists and dicts
   are rejected; this rejection is only on setitem; getitem
   will just throw suitable KeyError as it does now
- fix exception values for built-in types like DictType for
   the operation dict[i:j,]; the exception value
   strips the tupleness and I dislike that inconsistency
- change things so that obj[:] passes slice(None,None,None)
   instead of slice(0, sys.maxint, None)
   to mapping getitem (sequence getslice behavior still uses
   (0, sys.maxint) as defaults)

The last item deserves a little explanation. From the language reference 5.3.3:

"The semantics for a simple slicing are as follows. The primary must 
evaluate to a sequence object. The lower and upper bound expressions, if 
present, must evaluate to plain integers; defaults are zero and the 
sys.maxint, respectively. If either bound is negative, the sequence's 
length is added to it. The slicing now selects all items with index k such 
that i <= k < j where i and j are the specified lower and upper bounds. 
This may be an empty sequence. It is not an error if i or j lie outside the 
range of valid indexes (such items don't exist so they aren't selected)."

[Thus for simple slicing, the default lower bound is 0 and the default 
higher bound is sys.maxint.]

"The semantics for an extended slicing are as follows. The primary must 
evaluate to a mapping object, and it is indexed with a key that is 
constructed from the slice list, as follows. If the slice list contains at 
least one comma, the key is a tuple containing the conversion of the slice 
items; otherwise, the conversion of the lone slice item is the key. The 
conversion of a slice item that is an expression is that expression. The 
conversion of an ellipsis slice item is the built-in Ellipsis object. The 
conversion of a proper slice is a slice object (see section 3.2) whose 
start, stop and step attributes are the values of the expressions given as 
lower bound, upper bound and stride, respectively, substituting None for 
missing expressions."

Thus for extended slicing, default values for start, stop, and step are 
always None. If obj[i:j] finds that obj is not a sliceable sequence, the 
fallback to mapping getitem is by nature an extended slicing operation, and 
thus should obey extended slicing rules. Thus:

 >>> d = {}
 >>> d[:]
Traceback (innermost last):
   File "<pyshell#1>", line 1, in ?
     d[:]
KeyError: slice(0, 2147483647, None)

is wrong. Instead it should be:

 >>> d = {}
 >>> d[:]
Traceback (innermost last):
   File "<pyshell#1>", line 1, in ?
     d[:]
KeyError: slice(None, None, None)

Think this is a trivial issue? It's not. The whole reason for extended 
slicing was to let the Numeric Python guys have richly expressive slice 
objects: a None value always has a different and important meaning than 
some imposed default like sys.maxint. Treating [:] as some magic integer 
range is the wrong behavior, and requires implementations to write 
special-case code to handle slice(0, sys.maxint, None) as slice(None, None, 
None).

So this suggested small change just enforces the rules in section 5.3.3.

I will need a vote of confidence from the Numeric Python gang. My searches 
of NumPy source turn up no breakage as a result of the change. David A? Anyone?

I'll be unavailable for all of next week. If I can find the time today, 
I'll put together the first draft of the patch, but then I'll need to hand 
it off to someone else. Volunteers? Donn?

One last note:

With the above patches to 2.1, extended slice handling is now consistent 
with the Language Reference. Any type wishing to implement extended slicing 
has complete freedom in how to interpret the extended slice object or 
tuple. Thus we're free to add extended slice support incrementally to 
built-in types and only in ways that make sense for the type.

For 2.2 or later, I would like to add extended slice handling to several 
built-in types. I'll make a PEP. Here's a quick example to start the wheels 
turning:

 >>> s = "abcdef"
 >>> s[1:2,3:,::-1,3,765,...]
"bdeffedbcadabcdef"

I've wanted this kind of behavior for a while. :)


--
Robin Thomas
Engineering
StarMedia Network, Inc.
robin.thomas at starmedia.net
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20010302/c1058391/attachment.html>

From ws-news at gmx.at  Thu Mar  8 05:18:33 2001
From: ws-news at gmx.at (Werner Schiendl)
Date: Thu, 8 Mar 2001 11:18:33 +0100
Subject: "!=" is bad form. Re: sorry....never mind
References: <DZkp6.18811$pF4.3015652@news1.telusplanet.net> <fdlp6.16982$go.2157953@news0.telusplanet.net> <3AA6E308.5B63ECC6@cosc.canterbury.ac.nz> <3AA72253.FB021815@Lugoj.Com> <3AA72768.30243C29@alcyone.com>
Message-ID: <984046994.82647@newsmaster-04.atnet.at>

Personally I have no problem with != because I have to use C and C++ for
many task at may job.
However, I think <> is more intuitive to the novice and maybe the same
widely used...

Maybe not the very best examples, but Basic and Pascal use it too

Probably it's the best to just let two of them be valid...

regards
werner

Erik Max Francis <max at alcyone.com> wrote in message
news:3AA72768.30243C29 at alcyone.com...
> James Logajan wrote:
>
> > Actually, for Python, "!=" IS totally bogus and should be considered
> > the
> > unusual variant. Unless the logical negation operation has been
> > changed from
> > "not" to "!" (shucks, version 1.5.2 still does all I need) then "!="
> > is not
> > suggestive of anything.
>
> It's a sequence of characters, very commonly used in programming
> languages to indicate inequality.  Sometimes you actually have to learn
> things; not everything can instantly make sense without reading some
> documentation.
>
> --
>  Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
>  __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
> /  \ What do women want?
> \__/ Sigmund Freud
>     Alcyone Systems' CatCam / http://www.catcam.com/
>  What do your pets do all day while you're at work?  Find out.




From greg at cosc.canterbury.ac.nz  Tue Mar 20 00:08:46 2001
From: greg at cosc.canterbury.ac.nz (Greg Ewing)
Date: Tue, 20 Mar 2001 17:08:46 +1200
Subject: relational/boolean operator overloading
References: <98tjq6$qre$1@news.flashnet.it> <98tk4b$sbv$1@panix3.panix.com> <994h71$ohq$1@news.flashnet.it>
Message-ID: <3AB6E5DE.666F03C5@cosc.canterbury.ac.nz>

Renzo Tomaselli wrote:
> 
> The missing part is about overloading 'and'/'or' operators

That's impossible at present, and is not easy to fix. The
compiler treats these specially to implement short-circuiting.

I have some ideas on how to address this, and I might write
a PEP on it one day. 

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


From dsh8290 at rit.edu  Fri Mar 30 18:38:23 2001
From: dsh8290 at rit.edu (D-Man)
Date: Fri, 30 Mar 2001 18:38:23 -0500
Subject: Creation d'un NG francophone sur Python
In-Reply-To: <1er40bo.b8vjsq6avf3gN%francois.granger@free.fr>; from francois.granger@free.fr on Fri, Mar 30, 2001 at 11:18:56PM +0000
References: <1eqzyk6.gxd4jr1tkjykzN%francois.granger@free.fr> <9a1e6f011es@enews3.newsguy.com> <bS1x6.9225$SB2.838111@ruti.visi.com> <1er40bo.b8vjsq6avf3gN%francois.granger@free.fr>
Message-ID: <20010330183823.D15462@harmony.cs.rit.edu>

On Fri, Mar 30, 2001 at 11:18:56PM +0000, Fran?ois Granger wrote:
| Grant Edwards <grante at visi.com> wrote:
| 
| > I certainly would never vote against a French Python newsgroup.
| > But, I want to let you know that French language postings are
| > quite welcome in this group -- as are threads in any other
| > language.

ditto

| Thanks for this message and other nice messages.
| And sorry to have started a political thread about linguistic, this was
| not my aim.

Of course not, you were trying to learn something.  You should feel
comfortable posting in whatever language you like (or know, or
whatver) best.  Apparently some other people here speak french and
were able to understand (and hopefully answer!) your questions.  I, on
the other hand, am another one of those undeducated americans <wink>.
I have no trouble delete messages I can't understand regardless if
they are in french or random binary junk.

Here's a joke I once heard :

What do Europeans call someone who speaks 3 languages?
    Trilingual.
What do Europeans call someone who speaks 2 languages?
    Bilingual.
What do Europeans call someone who speaks 1 language?
    An American.

<grin>

(actually I think it was "swiss" not european in the version I
heard, but that was because it was associated with a story about a
trip to Switzerland,  substituting any locale where multi-lingual-ness
is the norm works just as well)

Have a nice day...
-D




From ct at gocept.com  Thu Mar 22 13:32:19 2001
From: ct at gocept.com (Christian Theune)
Date: Thu, 22 Mar 2001 19:32:19 +0100
Subject: python2 & curses
References: <99dge2$jpa$04$1@news.t-online.com>
Message-ID: <99dgfj$ba5$03$1@news.t-online.com>

stupid meself. .. posted this question before. sorry. ...


From nanotech at europa.com  Tue Mar 13 17:17:49 2001
From: nanotech at europa.com (nanotech at europa.com)
Date: Tue, 13 Mar 2001 22:17:49 -0000
Subject: Yet another Python vs. Perl speed issue/question
Message-ID: <98m6ad+i25s@eGroups.com>

All:

Python 2.0 on an HP (nice and big) takes ~30 seconds to lowercase all 
the lines in a 3meg file. Perl takes less than a second. What am I 
doing wrong?!?

Python
------
import sys
import string

stdin=sys.stdin
lower=string.lower

print map(lower,stdin.readlines())

Perl
----

print map(lc,<STDIN>);





From MarkH at ActiveState.com  Mon Mar 19 17:03:58 2001
From: MarkH at ActiveState.com (Mark Hammond)
Date: Mon, 19 Mar 2001 22:03:58 GMT
Subject: What IDE are you'all using for Python 2.1 for Windows?
References: <995gai$2bv$1@news.chatlink.com>
Message-ID: <3AB68390.9070200@ActiveState.com>

Luis Cortes wrote:

> Hi,
> 
>   Just wanted to know what "good" ide's are out there for python.  Please
> let me know which ones you prefer.

I'm using Pythonwin and Komodo :)  I believe Komodo will work today with 
Python 2.1, and Pythonwin is waiting for the next 2.1 beta due to a bug 
in Beta1's installer.

Mark.



From Sibylle.Koczian at Bibliothek.Uni-Augsburg.de  Wed Mar 21 10:47:53 2001
From: Sibylle.Koczian at Bibliothek.Uni-Augsburg.de (Sibylle Koczian)
Date: Wed, 21 Mar 2001 16:47:53 +0100
Subject: win32net, documentation
Message-ID: <3AB8CD29.C69CC1E2@Bibliothek.Uni-Augsburg.de>

Hello, all,

I want to do the Python equivalent of 

net use k: \\server\share

so I suppose I must call

win32net.NetUseAdd("\\server", level, data).

Level = 1 should be right, I think, but I can't get the syntax right.
I'm trying to translate the api_error messages (they seem to come
directly from my german Windows version):

- "server" without the "\\": api_error 50, the network request isn't
supported

- "\\server": api_error 123, the syntax for the file name, path name or
the drive name is wrong. I've tried:

data['local'] = 'k:', 'k:\\' or 'k'
data['remote'] = 'share', '\\server\share', '\\\\server\\share',
'\\share'
data['password'] = my normal domain password
data['asg_type'] = STYPE_DISKTREE

It's a network ressource I'm using constantly, so I'm quite sure it
exists, I've got the name right and I can connect to it.

So what's the right syntax? The help file for the Win32 extensions
hasn't got anything about the USE_INFO_* dictionaries, and the MSDN
library doesn't help either.

Thank you for help
Sibylle

-- 
----
Dr. Sibylle Koczian
Universitaetsbibliothek , Abt. Naturwiss.
D-86135 Augsburg

Tel.: (0821) 598-2400, Fax : (0821) 598-2410
e-mail : Sibylle.Koczian at Bibliothek.Uni-Augsburg.DE


From dog at ERC.MsState.Edu  Fri Mar  9 08:19:44 2001
From: dog at ERC.MsState.Edu (David OGwynn)
Date: Fri, 9 Mar 2001 07:19:44 -0600 (CST)
Subject: Problems with C API
Message-ID: <200103091319.f29DJii06393@ERC.MsState.Edu>

This is hard to explain, but I'm coding a visualization (Polygonal Mapping Tree) 
module for Python. So far, everything is going fairly swimmingly. I'm not having 
any real trouble from OpenGL. I've come into a strange problem, I have a binary 
parsing function built into the Octree object. It's fairly straightforward. And, 
when I run the code by hand 

i.e.
>>> from PMvis import *
>>> a = Octree('1',1,1,1,1)
>>> a.parsebinary('bsp1.bsp','BSP')
Done!
>>>

Everything works fine. But, when I try to run the exact same code through a 
script file (or through pdb) it segfaults on me. I'm totally lost. I really need 
this functionality built into my Octree object, given the large (30M or more) 
size of the binary files...

Thanks.

_____________
David O'Gwynn
dog at erc.msstate.edu

    "Life begins where math breaks down."
	--B.D.O.




From digitig at cix.co.uk  Mon Mar 19 16:32:00 2001
From: digitig at cix.co.uk (Tim Rowe)
Date: Mon, 19 Mar 2001 21:32 +0000 (GMT Standard Time)
Subject: Why "from __future__" stinks; a counter-offer
References: <mailman.985029808.7088.python-list@python.org>
Message-ID: <memo.20010319213204.8771C@digitig.compulink.co.uk>

In article <mailman.985029808.7088.python-list at python.org>, 
jeremy at alum.mit.edu (Jeremy Hylton) wrote:

> >>>>> "A" == Aahz Maruch <aahz at panix.com> writes:
> 
>   A> In article <slrn9bcg72.9q8.scarblac at pino.selwerd.nl>, Remco
>   A> Gerlich <scarblac at pino.selwerd.nl> wrote:
>   >> John W. Baxter <jwbnews at scandaroon.com> wrote in
>   >> comp.lang.python:
>   >>>
>   >>> For the future, the time to complain about PEP contents is when
>   >>> the PEP is posted, not after it has been posted, discussed,
>   >>> implemented, and the result distributed.
>   >>
>   >> And the place to complain is the Python-DEV list, which is not
>   >> public, although everyone can post to it and everyone can read it
>   >> using some archives...
> 
>   A> Officially, complaints need only be posted to c.l.py.  It's not
>   A> clear to me at this point what the True Practice is, but sending
>   A> e-mail to the PEP champion is supposed to work best of all.  (Not
>   A> all PEP creators are on python-dev.)
> 
> Discussion on comp.lang.python is fine, although it would usually help
> to cc the PEP author.  If there are legitimate complaints posted to
> c.l.py and the PEP author ignores them, I expect someone will cry
> foul.  Tim reads c.l.py as devotedly as anyone.  Barry and I try to
> keep up, too.
> 
> Jeremy

But there's part of the problem; egoless PEP writing is as hard as egoless 
programming, and having come up with a solution the author of that 
solution is bound to have some emotional involvement in that solution. 
There have been (rightly IMHO) howls of protest about __future__, but Tim 
has said it's ok for him, routinely dismisses all protests, presumably 
because he genuinely doesn't understand what our problem is, and he tells 
us it's going in anyway. Can you see why we're not all convinced that the 
PEP process is any help to us?


From aon.912502367 at aon.at  Mon Mar 12 04:39:54 2001
From: aon.912502367 at aon.at (Gregor Lingl)
Date: Mon, 12 Mar 2001 10:39:54 +0100
Subject: Einstein's Riddle
References: <oNPq6.396644$ge4.138479566@news2.rdc2.tx.home.com> <3AAC729F.77068663@iit.demokritos.gr>
Message-ID: <3AAC996A.862D66DA@aon.at>


Petasis George schrieb:

....

> But unfortunately, I have to agree with Greg Jorgensen: the way it is
> stated is not solvable:-) You have to make assumptions for solving it...
>
> George
>

Right!. But the ONLY assumption you have to make is, that it IS solvable!
Gregor Lingl



From Jaeho.Lee at brooks.com  Mon Mar 19 15:17:05 2001
From: Jaeho.Lee at brooks.com (Lee, Jaeho)
Date: Mon, 19 Mar 2001 15:17:05 -0500
Subject: What IDE are you'all using for Python 2.1 for Windows?
Message-ID: <857F15D7E3D8D3118D290008C7CF058602C9756B@mail-naeast1.brooks.com>

emacs.

You can get automatic indentation (not so dramatic compared with other
languages, though) and highlights syntax.

Optionally you can use IDLE or Pythonwin along with emacs. But usually emacs
is enough for me.

/Jaeho



From cce at clarkevans.com  Fri Mar 23 05:12:09 2001
From: cce at clarkevans.com (Clark C. Evans)
Date: Fri, 23 Mar 2001 05:12:09 -0500 (EST)
Subject: automatic delegation (was Re: PEP 245: Python interfaces)
In-Reply-To: <99f1h901dd8@news2.newsguy.com>
Message-ID: <Pine.LNX.4.21.0103230509090.22896-100000@clarkevans.com>

On Fri, 23 Mar 2001, Alex Martelli wrote:
> >    d) What is also missing from Python is
> >       an "auto-delegation".  
> 
> Maybe you mean you miss some favourite _syntax sugar_
> of yours for automatic delegation...?

Sugar. In Water. More Sugar.  

*blush*

Nice link, thank you, as this answers my question! 

http://pythoncookbook.activestate.com/pythoncookbook/community/data/985258114

Clark




From shonn at shonn.com  Wed Mar  7 05:23:54 2001
From: shonn at shonn.com (Shonn Gilson)
Date: Wed, 7 Mar 2001 02:23:54 -0800
Subject: Python, ASP, ADO, and Image string/binary problems
References: <ta0amgebq0sn0b@corp.supernews.com> <tabjei1p4a9481@corp.supernews.com> <984ql60tdk@news1.newsguy.com>
Message-ID: <tac37rqd7o2h64@corp.supernews.com>

Perfect. It works for both Response.BinaryWrite and doing an AppendChunk
with ADO. I noticed that it was coming in as a buffer from ADO but
everything that I saw relating to buffers was that they were an undocumented
necessary evil.
Thank you, that ends a lot of frustration.

Shonn

"Alex Martelli" <aleaxit at yahoo.com> wrote in message
news:984ql60tdk at news1.newsguy.com...
> "Shonn Gilson" <shonn at shonn.com> wrote in message
> news:tabjei1p4a9481 at corp.supernews.com...
>     [snip]
> > that '\377' is octal not decimal.). The problem seems to be that when
> Python
> > passes strings through COM that they get translated to Unicode. This
> happens
>
> Yes, this IS what COM (actually, Automation) specifies -- *all* strings
> are Unicode (note that Java specifies exactly the same thing, but COM
> was there first:-).
>
> Bravo for diagnosing that this was the problem in your case, though!
>
>
> > for things like Response.BinaryWrite along with things like recordsets
> > (objRS = Server.CreateObject("ADODB.Recordset")).  The result is that
the
> > beginning of my jpg file looks like
> > '\377\000\330\000\377\000\340\000\000\000\020\000J\000F\000I\000F'
instead
> > of '\377\330\377\340\000\020JFIF'.
> >
> > Does anyone know how to handle this?
>
> If you don't want to use *STRINGS* (normally-readable sequences of
> characters), then don't.  Python offers other datatypes, which will
> help you map (for example) *sequences of BYTES* (a byte is not the
> same thing as a character -- even though decades of programming with
> single-byte-characters strings have obscured this for many:).
>
>
> Specifically, I think that what you want is the built-in *BUFFER*
> type, which Python/COM will map into (in low-level COM terms) a
> SAFEARRAY of VT_UI8 (a sequence-of-bytes, uninterpreted).
>
>
> Try just changing this line:
>
> > > Response.BinaryWrite(bstrOut.getvalue())
>
> to:
>
> > > Response.BinaryWrite(buffer(bstrOut.getvalue()))
>
> and see if that solves it...
>
>
> Alex
>
>
>




From aleaxit at yahoo.com  Mon Mar 19 01:53:13 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Mon, 19 Mar 2001 07:53:13 +0100
Subject: Who's minister of propaganda this week?
References: <B98041329020E007.7DA84CAD9D0562AB.817BBC572F480869@lp.airnews.net> <98nu1m$12e@dispatch.concentric.net> <660ACCA1E4816F7E.C359CDF4168EF3FB.3B6B05538D619770@lp.airnews.net> <3AAFA13D.76E32C3F@divalsim.it> <98oa4o01i4v@news2.newsguy.com> <mailman.984619039.7856.python-list@python.org> <98q9h60122v@news2.newsguy.com> <mailman.984937946.5318.python-list@python.org> <9937en0bge@news2.newsguy.com> <slrn9bagbm.toq.qrczak@qrnik.zagroda>
Message-ID: <994ab801h5o@news2.newsguy.com>

"Marcin 'Qrczak' Kowalczyk" <qrczak at knm.org.pl> wrote in message
news:slrn9bagbm.toq.qrczak at qrnik.zagroda...
    [snip]
> Sun, 18 Mar 2001 21:55:59 +0100, Alex Martelli <aleaxit at yahoo.com> pisze:
>
> > For example, say that you want to organize certain objects in a
> > container/contained hierarchy, with no nesting limits.
> >
> > All existing objects will be leaves in the resulting tree.  Why does
> > this imply that such existing objects need to inherit from a certain
> > interface asserting "I may or may not be a container" (perhaps with
> > a default implementation stating "I am _not_ a container after all")?
>
> It doesn't imply that.

But if the container has a list of typed references, they have to
be typed to SOMETHING (which can be either a leaf of any
kind or another nested container).  The classic Design Pattern
uses just such a 'certain interface' as I just described.


> > In a dynamically typed language, you will define an interface
> > (and perhaps a default implementation thereof) *only for container
> > objects*, i.e., the "new" organizing superstructure.  Each container
> > will hold a set (e.g. a list, or dict) of (suitably tagged)
> > references to other objects; for recursive walks on the tree, one
> > will dynamically test whether a given object is-a container (else,
> > it's a leaf).
>
> In a statically typed language I would just recognize leafs on the
> container side, without relying on any properties of objects in
> leafs themselves.

So what (single) type do you make your container's references?

> You solved a problem which did not exist.

_I_?!  Have you reread "Design Patterns" lately...?


Alex





From aahz at panix.com  Thu Mar 22 17:21:20 2001
From: aahz at panix.com (Aahz Maruch)
Date: 22 Mar 2001 14:21:20 -0800
Subject: New PEP: The directive statement
References: <mailman.985285352.31497.python-list@python.org>
Message-ID: <99dtt0$l6t$1@panix3.panix.com>

In article <mailman.985285352.31497.python-list at python.org>,
 <James_Althoff at i2.com> wrote:
>
>ok, how about:
>
>from  __readyOrNot__  import nested_scopes
>
>or perhaps,
>
>from  __likeItOrNot__  import nested_scopes

This is in some respects excessively flippant, but that very flippancy
calls attention to it.  I'm not going to push this idea, but I do think
it's worth considering.
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

"I won't accept a model of the universe in which free will, omniscient
gods, and atheism are simultaneously true."  -- M


From deblNoNOspammy at theworld.com  Sat Mar 10 02:49:16 2001
From: deblNoNOspammy at theworld.com (David Lees)
Date: Sat, 10 Mar 2001 07:49:16 GMT
Subject: true newbie
References: <3aa9a9e9$0$65277$38c90ffe@news1.twtelecom.net>
Message-ID: <3AA9DC7B.918E9922@theworld.com>

You might look at Alan Gauld's on-line tutorial at:
http://www.crosswinds.net/~agauld/

david lees


Elaine Sweeny wrote:
> 
> can anyone help me find a tutorial for python thats for people just starting
> out in programming. The one one that comes with it is ok but its still too
> difficult.
>         thanks


From bedge at troikanetworks.com  Thu Mar  1 10:35:26 2001
From: bedge at troikanetworks.com (Bruce Edge)
Date: Thu, 01 Mar 2001 07:35:26 -0800
Subject: command line command structure representation
Message-ID: <3A9E6C3E.5844AD90@troikanetworks.com>

I'm doing a command line interface in python.
I'm going to make extensive use of the completer and readline libraries
for line editing.
In order to intelligently complete commands at any point, I need to
represent the command heirarchy tree with all available options,
arguments, etc in python. 
Is there anything out there (besides rapid logic's CLI package) to aid
in this task?

I've derived my own class from rlcompleter which looks for a list of my
commands and intercepts the completion key (tab) if one is detected. 
Each command has syntax like this

add widget <name> [[not]opt1 xxx] [[not]opt2 yyy] [[not]flag1]
[[not]flag2]

So my command tree representation will have to mark field types as:
- base command name, eg: add, delete, modify...
- non-promptable user entered string/number ie: no completion possible
- string/number from predefined list, completer cycles through these
- string/number from dynamically constructed list, completer options
built on demand.
- optional negation prefix, ie: not or no etc.

I know it's not that difficult to just create a set of nested structures
for this, but I'm trying to make something that is easily
extensible/maintanable, perhaps a way of embedding the command structure
tree in the command implementation classes themselves, and building it
at runtime though introspection.
Or, some kind of tree editor?

Any thoughts or suggestions?


Thanks, Bruce.


From MarkH at ActiveState.com  Fri Mar  2 17:24:14 2001
From: MarkH at ActiveState.com (Mark Hammond)
Date: Fri, 02 Mar 2001 22:24:14 GMT
Subject: How can I get the logged in users username on Win32?
References: <3a9f50fd.58862977@news.is.co.za> <6uIn6.16285$Qb7.2659273@newsb.telia.net>
Message-ID: <3AA01CC7.2090001@ActiveState.com>

Fredrik Lundh wrote:

> "nomad" wrote:
> 
>> After searching through the Python Docs, I found that on Un*x I could
>> use getpass.getuser(), but there is no equivalent for the Win32
>> platform.
>> 
>> In other words, how can I get the username in Windows?
> 
> 
> getpass.getuser()
> 
> (it works as long as USERNAME is set, which it usually is)

And win32api.GetUserName() should work even if it isnt :)

Mark.



From loewis at informatik.hu-berlin.de  Fri Mar 23 03:31:50 2001
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 23 Mar 2001 09:31:50 +0100
Subject: New PEP: The directive statement
References: <mailman.985322448.17782.python-list@python.org>
Message-ID: <j4elvokhs9.fsf@informatik.hu-berlin.de>

"Tim Peters" <tim.one at home.com> writes:

> > Furthermore, what is the use of that information to a program?
> 
> Primarily, it's what makes it possible to write a simple tool that nukes
> obsolete future statements.

I'd question this use of the information. It has yet to be written,
but I hope it won't operate in automatic "nuke all" mode. E.g. when
case insensitivity is the default in Python 2.5, Python 2.3 will still
be around (say). So the author of some package may want to nuke the
nested_scope directives, since nobody uses Python 2.1 anymore, but
would leave the case_insensitive directives.

In short, I hope that the program would have a dry-run mode, where it
reports all directives it finds, and a nuking mode, where it is given
a list of directives that it will remove.

> In another vein, in my own code I don't hesitate to use new features, and
> couldn't care less if rare incompatible changes require me to rewrite some
> code.  So I'll likely use future_statements much more than most people.  I
> already have code that does
> 
>     from __future__ import nested_scopes
>     import sys
>     assert nested_scopes.getMandatoryRelease() > sys.version_info
> 
> This will assert-fail as soon as I run it under a release where nested_scopes
> is the rule, and at that point I'll simply delete the future_statement and
> move on.

That sounds like a legitimite use, but I have my problems
understanding it. Are you relying on nested scopes in your code? If
so, why do put the assert statement there? Or does your code give a
warning and you want to shut it up using the future import? In that
case, wouldn't it be better to use the warning framework?

> Code like this also works fine:
> 
> import __future__
> if __future__.nested_scopes.getMandatoryRelease() > sys.version_info:
>     import old_version_of_module as amodule
> else:
>     import amodule
> 
> I don't know that I'll have an acute need for that wrt nested_scopes
> specifically

Why are you checking the mandatory release here? Doesn't amodule have
a future import in it?

> In that case I may very well have an acute need to ship one version
> of my source that exploits the new semantics but falls back to an
> older & slower module if the compiler it's running under can't play
> along.

People routinely use sys.*version* for that; I'm really curious
whether they'd prefer to investigate other variables instead.

Regards,
Martin


From aleaxit at yahoo.com  Wed Mar  7 11:42:36 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Wed, 7 Mar 2001 17:42:36 +0100
Subject: Can anyone recomend a good intoduction to C...
References: <mailman.983894288.19965.python-list@python.org> <983896344.701456@newsmaster-04.atnet.at> <m3lmqibw0a.fsf@atrus.jesus.cam.ac.uk> <983911749.217622@newsmaster-04.atnet.at> <UScp6.1097$y6.196347@ruti.visi.com> <983916572.286918@newsmaster-04.atnet.at> <9852b8015jm@news1.newsguy.com> <983967284.246700@newsmaster-04.atnet.at>
Message-ID: <985oev0216v@news1.newsguy.com>

"Werner Schiendl" <ws-news at gmx.at> wrote in message
news:983967284.246700 at newsmaster-04.atnet.at...
> > > That is of course a point, if you need to let the program inspect you
> will
> > > try to keep it as simple as possible.
> > > But given the same functionality, I think a C++ program will not be
more
> > > complex.
> >
> > The program will not be, the language is.  It's a hard-to-judge
tradeoff.
> >
>
> As to my knowledge, the application itself is inspected for safety
critical
> appliances.
> And you must not change anything afterwards in the software or you need to
> re-inspect the software.
>
> What use would it be to check the language?

Complete, detailed, total knowledge of the many tripwires that the
C++ language deploys for you to stumble on is needed on the part of
the code-inspection team (ideally, all members thereof).  This is
rarer than you might think -- I'm easily in the upper centile of C++
programmers according to Brainbench's tests, yet I'm fully aware
that there _are_ areas I'm fuzzy on (and, yes, they DO come up by
mistake time and again).

Further, a language that is very complicated is likelier to have
buggy implementations.  Again, the bugs need not be in obscure areas
*that you can easily avoid*; for example, name-resolution (and you
can hardly avoid NAMES, can you?-) does NOT fully conform to the
C++ standard in several popular C++ implementations (Microsoft, Borland,
gcc -- last time I looked at each, but that wasn't *AGES* ago:-).

A language with a simpler standard makes it likelier...:
    [A] that all code-inspectors will be fully aware of
        each and every provision of the language,
    [B] that implementations will be reasonably bug-free
        and fully implement the language standard (at
        least if said standard HAS been around a while:-).

This is what makes using a complicated language such as C++ a hard
to judge tradeoff for security-critical code.  On one hand, *IF* you
could fully trust the C++ implementation (libraries included), the
amount of code to be painstakingly checked would be vastly reduced
(a factor of two would not be surprising!); on the other hand, the
checking is going to be harder (in terms of cost-per-line-checked),
since the checkers need complete knowledge of the complicated (sigh)
language, *AND* said language has a nasty habit of implicitly
invoking code on your behalf -- it's easier to check if you SEE
everything that is happening, but, in C++, you often don't (ctors
and dtors in particular are being invoked left and right every
time you breathe, it seems at times:-).


> > I do not: C++ *forces* "any number of features" on you -- you can't
> > easily subset it and ENSURE features outside of the subset are not
> > intruding.
>
> Why this?

I guess it's because it was never designed with "ability to subset
easily and effectively" among the design-goals.  Lord knows it had
enough extremely hard design goals already, and its complication
stems from stretching to meet them all.

> You can use templates. But you can leave them at the side with no ill
> effects.

But then you say goodbye to the Standard Library -- it's just about
fully implemented in terms of templates!  Without templates, no
strings, no data structures, no algorithms, no auto-pointers...
(oops, I seem to have slipped into recitation of the Heart of
Wisdom Sutra -- but, hey, it *IS* true...!-).

In practice, templates are not optional -- you may strive to avoid
defining new ones, but, if you are to get any substantial utility
out of C++, it's a lost cause to avoid using all the ones that the
standard library is built of.

(If the choice for a project was, C++ or C, in full awareness of
all the tradeoffs, I would generally choose C++; but, if it was
"C++ without templates" or C, I would have ABSOLUTELY _NO_ DOUBTS
choosing C -- if you must cut out 90% of C++'s immediate pluses,
you might as well go all out for simplicity).

> You can use multiple inheritance. But, as well, you can leave it out if
you
> don't really need or like it.

Yes, multiple inheritance CAN be fully avoided -- it doesn't happen
"behind your backs" like other things do:-).

Things you _cannot_ avoid include, for example, exceptions.  They
are a VERY powerful mechanism (and an excellent reason to choose
C++: error-processing is absolutely essential), but _simple_ they
aren't (I'm not claiming they could have been made any simpler
within the design-goals of C++, mind you).  But, there's more...:

> You can use function/method overloading. But you can as well select a
> distinct name.

You cannot set your C++ compiler to turn off the overloading
feature.  And overloading IS an easy thing to do by mistake.

Its existence synergizes with templates, and subtleties of
name-resolution, to manufacture splendidly-subtle bugs, ones
very likely to escape even the most-careful and knowledgeable
human code-inspector (you MAY be able to get third party code
checking tools that CAN be set to warn on policy violations,
including the existence of overloading -- I have not had much
luck with my past experience of such source-checkers, but it
HAS been a while in this case, so things may have changed).

Avoiding overloads of *NON*-renameable functions (where the
language does NOT allow you to select a distinct name and
still get the same functionality), constructors above all,
is also basically impossible.  How CAN you avoid, at the
_very_ least, overloading constructors for default and copy
(for canonical classes)?!

And those (overloaded) functions *DON'T GET EXPLICITLY
NAMED IN MANY CASES* -- you find yourself _magically and
implicitly_ invoking a constructor (maybe of a temporary
object...) and the related destructor... all inserted for
you, behind the scenes, care of the compiler.  Nor can
you avoid 'slicing' if you do ANY public inheritance --
the language is designed so you can't avoid it, because
if A extends B, an A object IS accepted wherever a
const B& is desired (e.g., the copy constructor of B,
which is magically invoked anytime a B object is taken
by-value as a function argument, for example).

> You can use the new streams syntax introduced with C++, and it may have
> additional benefits.
> But nobody hinders you using the traditional APIs read, write, and
friends.

Formatting becomes a huge problem if you decide to eschew
iostreams, as the language then offers no other language-supplied
way to perform formatting (read and write are Unix _system calls_
related to UN-formatted I/O, and hardly a solution to that).
In practice, you'd have to build (or acquire and certify) a
completely separate formatting-and-IO library (as you would for
C, since its 'standard I/O' is among its weakest parts -- the
BSD I/O library, where explicit buffer-size arguments are always
passed, would be a much preferable approach).


> > > family of classes does not have that kind of problem.
> >
> > No, but it does have other issues -- e.g.,
> >     std::string oneletter('A', 1);
> > does NOT do what one might expect, nor does it give any warning
> > in the implementations I have at hand (oneletter becomes a string
> > of 65 characters all equal to '\001', not one of 1 character
> > equal to 'A' -- oops, arguments the wrong way around...!-).
>
> Yes, trading characters as integral types is maybe not a very good
decision
> from today's view on things.

But, knowing the language has made this non-changeable-now decision,
the constructors of std::string _should_ have been carefully designed
to avoid smacking heads-on against it.  This is one of the few aspects
of the standard C++ libraries that I consider an outright defect.

> However, I guess that would not do that much bad and at least not
introduce
> a security problem that turns out after years the software is in use.

If any serious testing is done (and it had BETTER be, NO MATTER
how carefully one code-inspects!), then NO problem should ever
'turn out after years the software is in use'.

If testing is skimped on, then "the buffer which is dimensioned
differently from what the code author and inspectors think it
is" MAY be excremental matter suspended over a ventilator and
just waiting for the thin thread it hangs on to snap -- today,
tomorrow, or three years from now.  E.g.,

    static const int buffer_size = 100;
    std::string largebuffer('X', buffer_size);
    if(data_to_copy.size() > buffer_size) {
        raise too_much_data_for_the_buffer;
    }
    memcpy(largebuffer.data(), data_to_copy.data(),
        data_to_copy.size());
    // or std::copy with all the .begin() and .end()
    // you wish -- just to give a false sense of
    // security...:-)

The author and inspectors may even feel particularly proud
of the named buffer_size, the careful use of const rather
than bleak old #define, the painstaking checking, and the
exception raised if the check fails...!-)

If, during tests, the data_to_copy.size() happens to be all
the time <= 88 OR > 100, then here we have a buffer overrun
just waiting to happen... the first time data_to_copy has
a size between 89 and 100, included.

You think these things don't happen in reality...?  How much
code-inspection and unit-testing have you done on C++ code?!

As the "C++ guru" (among other responsibilities) at my
employer, I routinely get called to help with mysterious
problems (and I _wish_ I was more often called for design
and code inspections *before* any problems show up, but that's
another issue:-).  I assure you that they DO happen, even to
coders who SHOULD know best -- sometimes code is written
late, by very tired people rushing against deadlines, or
(worse, for some:-) early in the morning, before enough
coffe (or, depending on taste, cola) has been imbibed to
raise the individual's awareness above amoeba-level...!-)


Note I'm *NOT* saying that C is better on this specific
task: I very specifically said:

> > The trade-off is no doubt a win for std::string here -- it
> > removes many more issues than it introduces.  But it's not


But it would be a _very_ serious mistake to assert:

> I think security is not a problem of C++...

because of COURSE it is.  The language's complexity is a
serious security MINUS, and MUST be carefully and responsibly
weighed against its pluses to choose the best too.


> You can easily write C or even Python code that does introduce security
> problems.

Of course you can.  But, the simplicity of the language
makes it easier for such problems to be caught in code
inspection (that's ASSUMING code inspection is not skimped
upon for security-critical code, or you're dogmeat ANYWAY!-).

Each language carefully avoids introducing a false sense
of security -- C is clearly so low-level everything needs
to be checked, Python does no static type-checks and so
just as clearly needs unit-testing and code-inspection.

Note that C++ does too, but it MAY lull the unwary into
easily-misinterpreted assertions such as the ones you've
made just above:-).  [Maybe Eiffel, since it DOES strive SO
darn hard (and makes such a good job, but still of necessity
incomplete... it can't solve the halting problem, can it?-)
to check everything and a half, is the worst from the point
of view of false sense of security...?-) -- but I think C++
has more foot-height tripwires ready:-)].


Alex





From aleaxit at yahoo.com  Mon Mar 12 04:35:06 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Mon, 12 Mar 2001 10:35:06 +0100
Subject: Einstein's Riddle
References: <oNPq6.396644$ge4.138479566@news2.rdc2.tx.home.com> <98glrd$1si1a$1@ID-75892.news.dfncis.de> <3AABE712.499CCF4C@home.com> <98gp5b$1u0oi$1@ID-75892.news.dfncis.de>
Message-ID: <98i5880pl2@news2.newsguy.com>

"Boris Popov" <no-spam-popov at bigserver.com> wrote in message
news:98gp5b$1u0oi$1 at ID-75892.news.dfncis.de...
> "Bruce Elrick" <belrick at home.com> wrote in message
> news:3AABE712.499CCF4C at home.com...
> > I swear there is no evidence that any of them own a fish.  The German
> > owns an emu.
> >
> > Bruce
>
> I am missing something then. I thought my table satisfies all given
> statements and it has fish right under the german...
>
> -Boris
> P.S. What's emu ?

A large flightless bird, whose meat is appreciated by more and
more connaisseurs (it's red meat, reasonably tasty, quite soft,
light in cholesterol & fats).  You can get a LOT of information
about it from the American Emu Association, http://www.aea-emu.org/.
(Particularly with the many recent scares about "mad cow's disease"
aka BSE, emu meat has received a huge market boost -- it's now
quite common for supermarkets around here to offer it, restaurants
to feature it in several dishes, etc).

Any association with Europe's "Economic and Monetary Union" is,
one can hope, totally coincidental (some people WOULD argue that
_that_ EMU _is_ 'owned' "by the German", but that's a base and
unsubstantiated canard...).


Alex





From paul.moore at uk.origin-it.com  Thu Mar  8 04:34:29 2001
From: paul.moore at uk.origin-it.com (Paul Moore)
Date: Thu, 8 Mar 2001 10:34:29 +0100 
Subject: Name of the Script
References: <20010304.185354.2114937732.8277@localhost.localdomain> <NJCo6.6720$qv3.2834377@nnrp5-w.sbc.net> <FpmjOjvwe5bm0Bg09LDTMPhh5vji@4ax.com> <Pine.LNX.4.21.0103051715150.18263-100000@kev.ioc.ee> <cbKjOqPRwNza00jOm7A1HBDkRYj7@4ax.com> <upuft9z0s.fsf@ctwd0143.fitlinxx.com>
Message-ID: <uFCnOk=ag9iJ0BBH3CPw9xrMiTRB@4ax.com>

On Wed, 7 Mar 2001 20:09:39 +0100 , David Bolen <db3l at fitlinxx.com>
wrote:

>Paul Moore <paul.moore at uk.origin-it.com> writes:
>
>> OK, so you could say "don't do that". But if I'm looking at some code
>> deep in a big program, can I be *sure* that no-one has changed
>> directory on me?
>
>You can't, but then again you really shouldn't be looking at your
>startup arguments (IMHO) inside code deep in a big program.

Agreed, in principle. However, in practice, I do have occasion to need
this sort of thing. As a specific example, suppose I had a module
which provided some form of logging service

def log(msg):
    logfile.write(timestamp + ": ")
    logfile.writeline(msg)

... you know the sort of thing. Now, I want my log file to reside in
the same directory as the calling script (not the same directory as
the logging module), but with an extension of .log instead of .py.

The obvious, and most user-friendly, way of doing this is using the
script name to generate a logfile name. This means looking at
sys.argv[0] from within the module.

A less friendly way is to require the main script to call
logger.init(os.path.abspath(sys.argv[0])) before starting processing.
(Which in turn requires sys and os imports which the main script may
otherwise not use).

This is a real-life case - I had a (VB Script, sorry!) program which
did exactly this. And the initialisation step was a definit pain...

But in the absence of a guaranteed full pathname in sys.argv[0], I
guess it's the only way to go.

Maybe having sys.argv[0] always be the full script pathname would be
something which could be added to Python? The interpreter must know
where the script is, so it would be easy to do from the startup
code...

Paul.



From max at alcyone.com  Sun Mar 25 13:00:22 2001
From: max at alcyone.com (Erik Max Francis)
Date: Sun, 25 Mar 2001 10:00:22 -0800
Subject: Newbie needs help to find bug in his first Perl program
References: <3ABCC757.6C6F17CA@earthlink.net>
Message-ID: <3ABE3236.A2D78C95@alcyone.com>

Ron Stephens wrote:

> sixtwo = raw_input ("where does %s rank for %y ?" % (option6,
> criteria2))

There is no format specifier for %y.

> Can anyone see what is wrong with this? I have several lines like this
> and they all give me error messages.

In future it would be more useful if you actually stated _what_ error
message you are getting.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ Nationalism is an infantile sickness.
\__/ Albert Einstein
    7 sisters productions / http://www.7sisters.com/
 Web design for the future.


From db3l at fitlinxx.com  Tue Mar 13 19:35:59 2001
From: db3l at fitlinxx.com (David Bolen)
Date: 13 Mar 2001 19:35:59 -0500
Subject: Poll'able Queue objects.
References: <txkr904gwu9.fsf@vortex.ukshells.co.uk> <cuRq6.240807$o91.31035696@typhoon.nyroc.rr.com> <slrn9aolc9.sci.grante@isis.visi.com>
Message-ID: <ulmq9191s.fsf@ctwd0143.fitlinxx.com>

grante at visi.com (Grant Edwards) writes:

> The mind wobbles.  One of the things that Unix got right is that you can
> select() (or poll()) on pipes, sockets, serial ports (any properly written
> character device that's properly written), or, in degenerate-case sort of
> way, "regular" files.  The fact that MS could take such a giant step
> backwards demonstrates either a monumental lack of knowledge about OS's or a
> monumental amount of conceit.  Probably a combination of both.

Actually, the depressing part is that they really did a decent job
with most of their objects but ended just shy of really covering
everything.  You can wait on a mixture of most any kernel objects
under Win32, and for other actions (like file I/O) the action can be
queued asynchronously and coupled to generic event objects that are
also waitable.  The inability to do overlapping (and non-blocking) I/O
on anonymous pipes is one of those misses that causes me no end of
annoyance.

It's really the select() implementation in WinSock that got messed up,
as it only works on sockets, which makes porting much more difficult.
But if you're willing to use the Win32 API, you can incorporate a
socket into the more general WaitForMultipleObjects by linking any of
a series of actions on the socket (e.g., ready to read, write, notify
on incoming connection, etc...) to a Win32 event object, which can
then play nicely with other objects in a general wait operation.

It's not the Unix approach of select() on file handles while
incorporating most everything under a device driver model with file
handles, but it's different, not necessarily terrible.  As someone
else pointed out, Unix has grown it's own share of system objects that
are incompatible with the select() approach.

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


From bedge at troikanetworks.com  Tue Mar 20 10:53:08 2001
From: bedge at troikanetworks.com (Bruce Edge)
Date: Tue, 20 Mar 2001 07:53:08 -0800
Subject: python suitability for large critical run forever apps
Message-ID: <3AB77CE4.78378862@troikanetworks.com>

I started using Python for come user CLI stuff. It has worked out so
well that we're thinking about extending it's use in our project.
I am not experienced enough to be able to answer some of the issues that
came up. Does anyone have any thoughts on these:


> > *data* size of each Python interpretter plus however much space
> > each object takes versus C++'s minimal data resource requirements.
> 
>     Any rough numbers (shared interpreter code, shared data, unshared data
>     and object overhead?)

Multi-threading, and mutual exclusion. Are these available, reliable,
and debuggable?

> > Is Python a "toy" language or is it industrial
> > strength ? For instance, does it facilitate recovering from exceptions,
> > and leaving objects in good states (the strong gurantee) ?  When you
> > are low on memory, will the fact that more Python operations require
> > dynamically allocated memory make implementations more brittle ? Is it
> > even possible to write a line of Python where you don't have to
> > worry about out-of-memory exceptions (the no-throw gurantee) ?
> > IOW if I'm in a low memory situation, can I perform any operations
> > to even fail gracefully ?

> > Regarding large-scale apps again, is there anything written in Python
> > which "runs forever" ? That is, it doesn't just die to clean up and
> > reset itself ?


thanks, Bruce.


From kca17 at dial.pipex.com  Thu Mar 29 16:27:20 2001
From: kca17 at dial.pipex.com (Eddie and Babs)
Date: Thu, 29 Mar 2001 22:27:20 +0100
Subject: Quoted identifiers in Python?
References: <3ac3387f$0$12248$cc9e4d1f@news.dial.pipex.com>
	 <slrn9c75sj.98.scarblac@pino.selwerd.nl>
Message-ID: <3ac3aa35$0$15026$cc9e4d1f@news.dial.pipex.com>

> Well, how would this work?

> a = "Ivory the engine"
> ax = obi.a

> What is ax?

"Ivor(y) the engine" is just an attribute name like any other, so you
couldn't do this any more than you could do:-

obj.name = "Rosy Blade"

x = "name"

y = obj.x

...and expect the y variable to contain "Rosy Blade". There is nothing
special about these quoted identifiers; they are just allowed to contain
characters other than those usually reserved for identifiers.



>
> So when I do
>
> x = "turn around"
>
> How does Python know whether I mean the string or the identifier?
>

We don't necessarily have to use the existing string quote characters to
delimit the quoted identifier. We could use angle brackets <...> or one of
the as-yet unused characters ($, @, ?). Or we could just prefix the first
quote character with something else like we do with Unicode strings. So we
could have:-

    x = i"turn around"

or
    x = _"turn around"

or whatever else you prefer. However, we run the risk of the old "static
inline friend void..." problem from C++: we might end up with things like

    iur"turn around"

Anyway, lexical analysis details aside, I think the basic idea of allowing
some form of QIs is a sound one.



From lsommerer at sewardweb.com  Fri Mar 23 12:18:01 2001
From: lsommerer at sewardweb.com (Lloyd Sommerer)
Date: Fri, 23 Mar 2001 11:18:01 -0600
Subject: Installing python on a virtual server account.
Message-ID: <3ABB8549.1987BA68@sewardweb.com>

Greetings,

I am at a bit of a lose.  I've been using python for about a month to
write scripts to handle CGI type things on a website.  My hosting
company has python 1.5.1 installed and says they will not be upgrading
(what they actually said was, "We are removing all references to Python
from our website and are offically no longer supporting it. When we
tried to upgrade to the newest version, it failed.").  They are leaving
1.5.1 in place, but I would prefer to use the current version for my
next project.  I would rather not switch hosting companies.

Does anyone have experience installing 2.0 in a shared hosting
situation?  Even telling me that I can't do it and that I should give up
would  be helpful at this point.

I am on a SunOS 5.7 machine.  Both configure and make seem to run okay.
Make install dies with the message attached at the bottom of this
message.  I'm assuming that is because I don't have the needed
permission.

Thanks for any advice.

Lloyd Sommerer

=====================================================
gcc   python.o \
          ../libpython2.0.a -ldb  -lpthread -lsocket -lnsl -ldl
-lthread -lm  -
o python
Undefined                       first referenced
 symbol                             in file
Py_Main                             python.o
ld: fatal: Symbol referencing errors. No output written to python
*** Error code 1
make: Fatal error: Command failed for target `link'
Current working directory /home/gatheri/python/Modules
*** Error code 1
make: Fatal error: Command failed for target `python'



From tuttledon at hotmail.com  Wed Mar 21 17:44:19 2001
From: tuttledon at hotmail.com (Don Tuttle)
Date: Wed, 21 Mar 2001 22:44:19 GMT
Subject: learning python...
References: <3AB8ECB5.2197E940@noaa.gov> <m77u6.56$Hp.3284@typhoon.aracnet.com> <mailman.985205007.3092.python-list@python.org>
Message-ID: <79au6.143514$__6.26351759@typhoon.southeast.rr.com>

"Timothy Grant"
> I'm am searching for a review of PP2nd. Is it a significantly
> different book from PP1st?

It's almost a total re-write.  I'm only a short way in PP2E but have already
found it very heplful.  Buy a copy and see for yourself. ;-)  BookPool.com
has it for about 43% off.

> Also. I'd love to see a new version of Beazley's book.

Python Essential Reference 2nd Edition to be released in June.

Don




From jkraska1 at san.rr.com  Thu Mar  8 19:51:23 2001
From: jkraska1 at san.rr.com (Courageous)
Date: Fri, 09 Mar 2001 00:51:23 GMT
Subject: Can anyone recomend a good intoduction to C...
References: <mailman.983894288.19965.python-list@python.org> <3AA5EF59.507F1D91@olsen.ch> <983958878.976344@newsmaster-04.atnet.at> <77up6.1289$y6.235541@ruti.visi.com> <985vqs01ipr@drn.newsguy.com> <ewwp6.1466$y6.241005@ruti.visi.com> <slrn9aeblb.cg1.scarblac@pino.selwerd.nl>
Message-ID: <djagat4kk3cs9hg0c3k39cudk6goblhkg6@4ax.com>

>All this bloat added to C++ was not necessary for making an OO language.

Echo that. What's "object oriented" about templates?

A: NOTHING.

C//



From tjreedy at udel.edu  Thu Mar 15 12:52:55 2001
From: tjreedy at udel.edu (Terry Reedy)
Date: Thu, 15 Mar 2001 12:52:55 -0500
Subject: This math scares me
References: <3aad4062.1290021@News.CIS.DFN.DE> <20010312170356.B3513@better.net> <mailman.984480856.18097.python-list@python.org> <984486395.407379@newsmaster-04.atnet.at> <slrn9asf1d.ja3.qrczak@qrnik.zagroda> <984518199.127140@newsmaster-04.atnet.at> <jwbnews-09D75F.13313313032001@corp.supernews.com> <98ohuc$du9$1@news.udel.edu> <7bQr6.644$lz1.17686@ruti.visi.com> <98oku1$f6a$1@news.udel.edu> <erRr6.659$lz1.20662@ruti.visi.com>
Message-ID: <98qv55$f5d$1@news.udel.edu>

Me:
> >> >If one counts pennies instead of dollars, then 32 bit ints (+-
> >> >$20 million) will serve for many purposes.  Exact, faster, and
> >> >more widely available than BCD
Grant Edwards:
> >> In fiancial calculations you also need to represent numbers
> >> that aren't amounts of money.  Interest rates for example.
Me:
> >So?
Grant
> So 32 bit integers with a weighting of one penny per count
> won't work for that.
Me:
> >Floats (double precision) work well enough for this, I believe.
Grant:
> No.  The binary floating point representation of many decimal
> fraction values is not exact.  That's what started this whole
> thread.  You can't represent 0.1 exactly in binary FP.  You can
> in BCD.

So, .1% gets represented as the binary equivalent of, perhaps,
.0010000000000000001.
So the interest on 2,000,000,000 pennies is calculated as
2000000.0000000002 instead of
2000000 pennies.  When rounded off, they are the same.  To me, this is, as
I claimed. 'well enough' (for many purposes, at least), even though not
'exact'.  Even if you are using 64 bit ints to allow larger amounts,  the
rounded error doesn't reach a whole penny until your dollar amounts are
about 50 quadrillion (US version) dollars (maybe give or take a power of
ten).  Haha.  The US Feds with all their BCD_equipped IBM mainframes can't
keep track of spending to within a billion dollars.

Now suppose a financial instrument carries a yearly interest rate of .077
(7.7%) with montly payments.  Then the monthly multiplier is
.0006416666666666666666666666666666666666...
Whoops, the BCD representation has to be approximate also, just like the
binary one.

Even if a monthly multiplier like .00643 (corresponding to a yearly 7.716%)
is exactly represented in BCD, there may still be fractional amounts that
need to be rounded to the nearest penny by one rule or another.  BCD does
not eliminate the problems of finite precision arithmetic.

Terry J. Reedy




From neilh at scintilla.org  Mon Mar 26 21:17:45 2001
From: neilh at scintilla.org (Neil Hodgson)
Date: Tue, 27 Mar 2001 02:17:45 GMT
Subject: warrior geeks
References: <B43D149A9AB2D411971300B0D03D7E8B0FBFE0@natasha.auslabs.avaya.com> <mailman.985653814.22028.python-list@python.org>
Message-ID: <dLSv6.26823$992.166430@news-server.bigpond.net.au>

Paul Prescod:

> But the question is: Which of Perl versus Python is more appropriate for
> shooting yourself in the foot.

   The correct calibre pearl may fit into your gun but persuading a snake to
enter a thin dark unnatural tube is more difficult. Even when drunk.

   There're reasons we're not allowed to keep Pythons as pets without a
license here in Oz. Oh those guv'm'nt bastards!

   Neil




From topmind at technologist.com  Tue Mar 20 01:33:48 2001
From: topmind at technologist.com (topmind)
Date: Tue, 20 Mar 2001 06:33:48 GMT
Subject: Collection interfaces
References: <3a90fa97.0@news1.mweb.co.za>   <MPG.150312ceb68f1ada98969f@news.earthlink.net> <8Gem6.473353$U46.14105062@news1.sttls1.wa.home.com> <MPG.1503548620e632919896a4@news.earthlink.net> <MCjm6.474652$U46.14163288@news1.sttls1.wa.home.com> <MPG.150394195f7f04a19896a7@news.earthlink.net> <3A9A86B1.9A8C69A2@ix.netcom.com> <MPG.1504d14563dce6499896af@news.earthlink.net> <97fjla128ir@news2.newsguy.com> <MPG.1505a76831c34b749896b3@news.earthlink.net> <97h5s40vhc@news2.newsguy.com> <MPG.1506007591cf818b9896bc@news.earthlink.net> <slrn99opop.gr.amk@mira.erols.com> <MPG.1506265ebf6d934d9896be@news.earthlink.net> <97ibmc0k47@news1.newsguy.com> <MPG.1506cf3293b094b49896c1@news.earthlink.net> <97m0c7080c@news2.newsguy.com> <MPG.15085517debf4b649896ca@news.earthlink.net> <zb%n6.31033$o%6.586026@news1.rdc1.bc.home.com> <MPG.150b383cea226d409896d0@news.earthlink.net> <kNko6.6986$eq3.707121@news.uswest.net> <MPG.150c46db9cec283e9896d4@news.earthlink.net> <3AA2A898.450B659E@mail.com>
Message-ID: <gRCt6.12224$Im6.1308560@newsread1.prod.itd.earthlink.net>

"James A. Robertson" <jarober at mail.com> wrote in message
news:3AA2A898.450B659E at mail.com...
> Topmind wrote:
> >
>
> <snip>
>
> >
> > Not exactly "typical custom business applications", the domain
> > in which I focus most of my whining.
> >
> > The percentage of OO fans in the business domain seems lower
> > than average. Perhaps this says something about OO and
> > the business domain.
> >
>
> 90% + of the customers using VisualWorks are writing 'typical' business
> apps.
>


Then why are most "see what OO can do" examples strangely non-business?


> [snip]

-tmind-
oop.ismad.com





From cabrera at wrc.xerox.com  Thu Mar 22 16:40:04 2001
From: cabrera at wrc.xerox.com (jicman)
Date: 22 Mar 2001 21:40:04 GMT
Subject: ASCII encoding error: ordinal not in range(128)
References: <91o5ih$2kv$1@news.wrc.xerox.com> <20001219172933.B29727@ActiveState.com> <mailman.985123475.18421.python-list@python.org> <W3Rt6.2582$4N4.344041@newsc.telia.net>
Message-ID: <99drfk$eul$1@news.wrc.xerox.com>

fredrik at pythonware.com says...

>summary: someone's handing you unicode strings (probably
>the COM interface), and you're trying to use them as 8-bit
>strings (probably by passing them to a method that expects
>8-bit data) without specifying an encoding.
>
>(if you want me to be more precise, you have to show us
>more code).

def FindIfAttachments(myMsg,spar,wID,fam):
    myAtt = myMsg.Attachments
    if myAtt.Count > 0:
        LogDir = os.path.join(logspath,'Attachments',spar)
        if os.path.isdir(LogDir):
            #do nothing for now
            nothing = 1
        else:
            os.makedirs(LogDir)
        i = 1
        while i <= myAtt.Count:
            if myAtt.Item(i).Type == 1:
                code = myAtt.Item(i).DisplayName
                name = code.encode()
                if name == '':
                    name = 'Untitled Attachment.msg'
                else:
                    name = CleanNameToSave(name)
                name = spar + ' attachment - ' + name
                name = os.path.join(LogDir,name)
                error = myAtt.Item(i).SaveAsFile(name)
            i = i + 1

myMsg is a COM pointer to an Outlook 2000 message.  myAtt is the pointer to the 
attachments.  DisplayName is the name of the attachment causes the problem.


Here is the error message:
C:\>qna0.py
Traceback (most recent call last):
  File "C:\myprograms\qna0.py", line 1570, in ?
    main()
  File "C:\myprograms\qna0.py", line 1556, in main
    ProcessAllItems(wf2,workpath,workpath,NSpace)
  File "C:\myprograms\qna0.py", line 1484, in ProcessAllItems
    FindIfAttachments(em,Rec['SPAR'],wID,fam)
  File "C:\myprograms\qna0.py", line 1156, in FindIfAttachments
    name = code.encode()
UnicodeError: ASCII encoding error: ordinal not in range(128)

>
>to solve this, you have to add an explicit conversion to the
>right place.  e.g:
>
>    s = u.encode("latin-1")
>
>where "u" is a unicode string and "s" an 8-bit ISO Latin string.
>
>(if you don't do this, Python will convert the string anyway when
>you pass it to an 8-bit function.  the default conversion assumes
>ASCII, which is why you get that exception)

So, if I am understanding this right, the line

name = code.encode()

should be

name = code.encode('latin-1')

or perhaps

name = code.encode('PC-8')

Thanks very much.

I'll try this and let you know....

thank you so very much

jos?



From jafo at tummy.com  Thu Mar  1 17:12:43 2001
From: jafo at tummy.com (Sean Reifschneider)
Date: Thu, 1 Mar 2001 15:12:43 -0700
Subject: Performance when working directory is slow network drive
In-Reply-To: <983480535.813990@clam-ext>; from m.hadfield@niwa.cri.nz on Fri, Mar 02, 2001 at 10:01:28AM +1300
References: <983411179.898268@clam-ext> <mailman.983425921.14373.python-list@python.org> <983480535.813990@clam-ext>
Message-ID: <20010301151243.A22330@tummy.com>

On Fri, Mar 02, 2001 at 10:01:28AM +1300, Mark Hadfield wrote:
>1. Why having the current directory in the search path affects the import
>time for the os module much more than it does for other modules I tested
>(like string). Is this because os is a package?

It affects it because if you do an import of "os", it follows the path trying
to find it.

Sean
-- 
 How does a girl like you get to be a girl like you?
                 -- _North_by_Northwest_
Sean Reifschneider, Inimitably Superfluous <jafo at tummy.com>
tummy.com - Linux Consulting since 1995. Qmail, KRUD, Firewalls, Python



From tim.one at home.com  Sun Mar 11 02:11:22 2001
From: tim.one at home.com (Tim Peters)
Date: Sun, 11 Mar 2001 02:11:22 -0500
Subject: Static typing (was Re: Java guy interested in Python)
In-Reply-To: <20010309175317.66881.qmail@web219.mail.yahoo.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCCENFJEAA.tim.one@home.com>

[Hamish Lawson]
> ...
> Not sure that makes a type system? Isn't that what the modern
> understanding of a type is - a list of operations that are expected to
> be supported, as with Java's idea of interfaces being distinct from
> classes?
> ...

All type systems in all programming languages are hacks, because at heart
what you *really* want to say is "variable x must take on values from set S",
and interesting specifications for S may not even be computable.  So you
approximate in more-or-less crude ways, saying "i must be an int" when you
really mean "i must be an even integer" or "i must be an odd prime"; saying
"x and y must be reals" when you really mean "x and y must be reals such that
x**2 + y**2 <= 1"; etc.  A static type system is interesting to the extent
that it closely approximates what you want to say most often, and can be
efficiently checked at compile-time (there's nothing you can say in a static
type system that can't be said today at runtime in Python).

between-what-you-want-to-say-and-what-you-can-say-are-worlds-of-
    artificial-complexity-ly y'rs  - tim




From db3l at fitlinxx.com  Fri Mar 16 18:53:08 2001
From: db3l at fitlinxx.com (David Bolen)
Date: 16 Mar 2001 18:53:08 -0500
Subject: How do I get Win32 to locate DLLs elsewhere (not really a Python thing)?
References: <mailman.984506597.12469.python-list@python.org> <3AB1F697.485A776D@yahoo.fr> <98t3sn$8t9$1@neptun.beotel.net>
Message-ID: <uwv9p8e57.fsf@ctwd0143.fitlinxx.com>

"gzeljko" <gzeljko at sezampro.yu> writes:

> Your have system provaided interface:
> ControlPanel/System/Environment  (No need for reboot or 'relog')

Actually, depending on what will be executing the code you may in fact
have to reboot.  While "apply"ing changes will affect new cmd.exe
sessions or direct execution of Python scripts, if you try to run
scripts via services or other long running processes that doesn't pick
up system changes when announced you'll have to reboot for them to
notice the change.  It's less common than older Windows platforms, but
still seems to always happen when you least want to reboot :-)

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


From ransen_spam_me_not at nemo.it  Thu Mar  1 12:04:49 2001
From: ransen_spam_me_not at nemo.it (Owen F. Ransen)
Date: Thu, 01 Mar 2001 17:04:49 GMT
Subject: SetProgramName, needs the quotes of Windows paths?
Message-ID: <3a9f799b.1500426@news.newsguy.com>

Does Py_SetProgramName, when running under 32 bit
Windows, need the quotes which you often find in the
shortcuts to windows programs?

I think that in Windows this is to allow spaces to
appear in paths without the system getting confused
over what is a space before a command line parameter 
and what is a space simply in a path name.

Since what we are passing Python is a pure path and
not a command line I'd assume that the quotes are
not needed...


--
Owen F. Ransen
http://www.ransen.com/
Home of Gliftic & Repligator Image Generators


From s2mdalle at titan.vcu.edu  Thu Mar  1 19:40:22 2001
From: s2mdalle at titan.vcu.edu (David Allen)
Date: Fri, 02 Mar 2001 00:40:22 GMT
Subject: define a new func on the fly?
References: <3A9EE927.8744154E@troikanetworks.com>
Message-ID: <WZBn6.6573$jP.677969@typhoon2.ba-dsg.net>

In article <3A9EE927.8744154E at troikanetworks.com>, "Bruce Edge"
<bedge at troikanetworks.com> wrote:

> I'm trying to define a new function on the at runtime,
> 
> this fails:
> 
>>>> eval( "def global_func_name():\n\tpass" )
> Traceback (most recent call last):
>   File "<stdin>", line 1, in ? File "<string>", line 1
>     def global_func_name():
>       ^
> SyntaxError: invalid syntax
> 
> any idea how I can do this?

Try using lambda instead for anonymous functions,
and then bind it to a variable:

>>> squaring_function = eval("lambda x: x * x")
>>> print squaring_function(5)
25

In this respect, I don't like python's required 
whitespace, because while it keeps programmers honest
when they're programming, it makes it much more of a
pain to dynamically generate a string of code and
then eval() it as opposed to, *cough*, other 
scripting languages.

-- 
David Allen
http://opop.nols.com/
----------------------------------------
Art is anything you can get away with. 
- Marshall McLuhan.


From gregj at pobox.com  Fri Mar  2 02:00:44 2001
From: gregj at pobox.com (Gregory Jorgensen)
Date: Fri, 02 Mar 2001 07:00:44 GMT
Subject: define a new func on the fly?
References: <3A9EE927.8744154E@troikanetworks.com> <o6Cn6.846$45.1115@www.newsranger.com> <3a9f0bc2.10059041@News.CIS.DFN.DE>
Message-ID: <wyHn6.925$45.1159@www.newsranger.com>

It should be a \n after 'hello', not \t. Sorry for the typo.


In article <3a9f0bc2.10059041 at News.CIS.DFN.DE>, Costas Menico says...
>
>Gregory Jorgensen<gregj at pobox.com> wrote:
>
>>eval() evaluates an expression. exec() executes a statement. Try this:
>>
>>>>> exec("def f():\n\treturn 'hello'\t")
>
>When I try the above I get a syntax error... I am using PythonWin. Is
>this a bug?
>
>Costas
>
>
>>>>> f()
>>'hello'
>>
>>>>> eval("f()")
>>'hello'
>

Greg Jorgensen
Deschooling Society
Portland, Oregon, USA
gregj at pobox.com


From bowman at montana.com  Fri Mar 16 08:24:00 2001
From: bowman at montana.com (bowman)
Date: Fri, 16 Mar 2001 06:24:00 -0700
Subject: windows script host and python
References: <%tKr6.2570$pX.26174@news1.onlynews.com> <98oq3r$1e0$1@troll.powertech.no>
Message-ID: <8Aps6.104$pl.1515@newsfeed.slurp.net>

Syver Enstad <syver at NOSPAMcyberwatcher.com> wrote in message
news:98oq3r$1e0$1 at troll.powertech.no...
>
> Just curious but what do plan to acheive by using python with wsh that you
> couldn't do using normal .py files?

even more curious: can Python hook up and use WMI from a normal .py?




From paulp at ActiveState.com  Wed Mar 14 15:11:16 2001
From: paulp at ActiveState.com (Paul Prescod)
Date: Wed, 14 Mar 2001 12:11:16 -0800
Subject: Who's minister of propaganda this week?
References: <B98041329020E007.7DA84CAD9D0562AB.817BBC572F480869@lp.airnews.net>
		<98nu1m$12e@dispatch.concentric.net> <3AAF84FE.F28A2C30@ActiveState.com> <uelw09pup.fsf@advancedmp.net>
Message-ID: <3AAFD064.A085DBB5@ActiveState.com>

Don Dwiggins wrote:
> 
>...
> I'm relatively new to Python, and this is something I'm trying to get my
> head around.  I know that closures (blocks) are extremely important in
> Smalltalk (e.g., conditionals and looping are implemented entirely in terms
> of blocks).  So, is this just an indication of Smalltalk's peculiarity, 

Well, you could argue that it's a form of elegance...but yes, it is
somewhat unique to Smalltalk. Even lisps don't, in general, treat
conditionals as a form of closure.

> or
> could Python benefit from the ability to exploit more robust closures in
> something like the manner of Smalltalk?  For example, how about implementing
> general iteration as something a container class does to a closure (as in
> select:, detect:, inject:into:, etc.)?

I don't know enough to comment in detail ... but I suspect there are
Pythonic ways to do the same sorts of things. Rather than turning blocks
into closures and closures into objects, Python would ask you to use
more traditional object oriented structures like classes and __getitem__
or __getiterator__ methods. (Note that the latter is semi-fanciful)

-- 
Python:
    Programming the way
    Guido
    indented it.
       - (originated with Skip Montanaro?)



From MarkH at ActiveState.com  Mon Mar 12 19:00:58 2001
From: MarkH at ActiveState.com (Mark Hammond)
Date: Tue, 13 Mar 2001 00:00:58 GMT
Subject: Pythonwin default directory init
References: <3aab04bc.2359756@News.CIS.DFN.DE> <3AABFC9E.3070706@ActiveState.com> <3aac185d.25472528@News.CIS.DFN.DE>
Message-ID: <3AAD6280.9090708@ActiveState.com>

Costas Menico wrote:


> I am using Windows 98. Is there someplace in the .py code that I can
> explicitly force on startup to default it to the directory from "Start
> In". It is already loaded in os.getcwd(). So it should be a simple
> matter. 

Unfortunately this is in MFC code.  It would involve moving that dialog out into .py code.

> This is a great product. Its the reason I started using Python and
> building my confidence to dump VB. The code completion (intellisense?)
> makes using Python a pleasure.
> 
> Is there a planned upgrade coming up any time soon? And is there some
> wish/comments list I can participate in?

Minor upgrades are always happening, but no major changes are planned.  There is no list for comments and suggestions - otherwise I would spend my life trying to add them all.  The best way to get new features in is to send me patches :)  Otherwise, something has to bother me personally before I do it.

Mark.



From djc at object-craft.com.au  Tue Mar  6 05:44:09 2001
From: djc at object-craft.com.au (Dave Cole)
Date: 06 Mar 2001 21:44:09 +1100
Subject: Access to sybase API
References: <obOo6.1158$f07.1696858@nnrp6.proxad.net> <3AA3B0D0.90876BA7@schlund.de>
Message-ID: <m3zoezxjly.fsf@vole.object-craft.com.au>

>>>>> "Carsten" == Carsten Gaebler <cg at schlund.de> writes:

Carsten> Neotaku wrote:
>> I search a code source for python to access Sybase thru API
>> connexion the Sybase DB-Library API or Sybase CT-Library (aka the
>> Client Library) API

Carsten> Have a look at Dave Cole's Sybase module:
Carsten> http://www.object-craft.com.au/projects/sybase/

I have been slowly rewriting this module to make it as close to 100%
DB-API 2.0 compliance as possible.  My new version has split the
module into a Python part and a C extension module which is a fairly
thin wrapper on top of the Sybase CT API.  I am getting fairly close,
but the pressures of earning money keep forcing me back onto real
work.

For some people, the new version of the module might be good enough
already to replace the current 0.12 module.  I have not placed a link
to the new module on the homepage because I want to wait until the new
module can completely replace the old module.

If you are interested in giving the new module a spin, or you are
interested in how the CT library works, you can get the work in
progress from:

        http://www.object-craft.com.au/projects/sybase/Sybase-0.21.tar.gz

If you were not happy with the speed of cursors in the old module then
you should try this one.  Cursors are now about five times faster
because I am no longer using ct_cursor().

In some ways, the new module is much closer to DB-API compliance than
the old one.

- Dave

P.S.  If you want to hekp out with the implementation, I will
      gratefully accept patches.

-- 
http://www.object-craft.com.au


From sholden at holdenweb.com  Fri Mar  9 09:07:23 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Fri, 9 Mar 2001 09:07:23 -0500
Subject: Static typing (was Re: Java guy interested in Python)
References: <mailman.984131592.17401.python-list@python.org> <98ah8301ndm@news2.newsguy.com>
Message-ID: <Xv5q6.45955$ML1.1997058@e420r-atl2.usenetserver.com>

"Alex Martelli" <aleaxit at yahoo.com> wrote in message
news:98ah8301ndm at news2.newsguy.com...
> "Hamish Lawson" <hamish_lawson at yahoo.co.uk> wrote in message
> news:mailman.984131592.17401.python-list at python.org...
>     [snip]
> > Wouldn't there be a hierarchy of interface types for file-like objects,
> > with some interfaces specifying more methods than others? You'd then
>
> It probably wouldn't be a hierarchy -- there is no special sense
> in which any of .read, .readline, .readlines, .xreadlines is "more
> fundamental" than any of the others (Haskell's typeclasses are the
> best way I know to express such common situations -- methods can be
> defined in a mutually recursive way in the typeclass, the types that
> are its specific instances may break the recursion any way they like,
> and client-code gets all the desirable usage convenience).
>
> > choose the interface type that specified the collection of methods that
> > you require of your passed object.
>
> Any of the 2**4-1 non-empty subsets of the above 4 may make
> sense.  Why invent 15 names for the subsets?  The collection
> of methods that you require is exactly the set of methods
> you will actually call.  Interfaces may have their uses (I'd
> LOVE a way to get Haskell typeclasses in Python!-) but this
> doesn't seem to be a particularly useful one.
>
To answer the original question (is static typing likely to make it into
Python)...

>From http://www.python.org/sigs/types-sig/

"""Recently, the types SIG has come under attack for inactivity, and in
response Paul Prescod has offered to guide the SIG towards accomplishing a
single specific goal: to develop an optional static typing system for
Python. The December 1999 archives of the SIG show the discussion that lead
to this decision."""

However, I'm not sure how active the SIG now is, since this kind of work is
well out of my current range of interests.

regards
 Steve





From timr at probo.com  Wed Mar 14 01:33:07 2001
From: timr at probo.com (Tim Roberts)
Date: Tue, 13 Mar 2001 22:33:07 -0800
Subject: Subject with SMTPLIB?
References: <3AADE48A.5D8AA43@daimlerchrysler.com>
Message-ID: <as2uatca7m0mpo9l0t6d2d7ockujt4dt3j@4ax.com>

root <root at mailhost.stg.rt.daimlerchrysler.com> wrote:
>
>I need to write some mail for my fax-log evaluating script. I'm
>currently using the smtplib to simply send some mails. But how 
>can I get (without studying RFCs and papers on SMTP for years)
>a SUBJECT for my mails?

There have already been several replies to this, but I'm not sure they
really explained the root principle here.  This is an important and
non-intuitive point, although perhaps a bit off-topic for this newsgroup.

<enter lecture mode>

smtplib is concerned with SMTP.  That is the protocol for transferring
e-mail messages, described in RFC 821.  SMTP cares only about the
originator and the recipients of your message.  Everything else about the
message, including the body AND all of the headers, is just random text to
SMTP.  None of the e-mail headers you are accustomed to are part of the
protocol at all.  Here is a perfectly valid SMTP session in which I send a
message to myself:

  MAIL FROM: <timr at probo.com>
  RCPT TO: <timr at probo.com>
  DATA
  From: SMTP doesn't care what I put here
  To:   SMTP doesn't care what I put here
  Subject: SMTP doesn't care what I put here
  Blech: Headers don't matter

  Message body.
  .

Everything up to the "DATA" command is called the "envelope".  This is what
SMTP servers like sendmail use to direct the transmission of the message.
The headers in the message itself (which are defined in a separate RFC,
822) are irrelevant to SMTP; they are just part of the payload.

The headers in the payload are only there for the mail reader at the other
end.  If you want the mail reader to show a subject for your message, you
need to add a Subject: header.  Further, if you want the mail reader to
show a valid sender and recipient, YOU have to include correctly formatted
"From:" and "To:" headers.  The addresses you pass to the smtplib.sendmail
method are STRICTLY for the envelope: the SMTP MAIL and RCPT commands.
smtplib.sendmail does not add the "From:" and "To:" headers, because they
are not part of SMTP.

For the benefit of modern mail readers, it is important to get the "From:"
and "To:" lines correct.  If I actually sent the message I showed above, my
mail reader would discard it as spam, because my address is not present in
the "To:" or "Cc:" headers.  However, the exchange is perfectly valid SMTP.

<exit lecture mode>
--
- Tim Roberts, timr at probo.com
  Providenza & Boekelheide, Inc.


From urner at alumni.princeton.edu  Sat Mar  3 01:24:35 2001
From: urner at alumni.princeton.edu (Kirby Urner)
Date: Fri, 02 Mar 2001 22:24:35 -0800
Subject: Adding a method to an instance
References: <G2%n6.360714$JT5.11606234@news20.bellglobal.com> <6621ato2vknc667m3iuqvmm8vcj0h99qqk@4ax.com>
Message-ID: <ie31atcusg4u31fh6qc4ukb2uh7eqo9gc1@4ax.com>

Kirby Urner <urner at alumni.princeton.edu> wrote:

>In saying f.__class__, you point back to the superclass foo.
>If you want to poke newMethod into object f alone, then you 
>should just go:
>
>f = newMethod
>

Sorry,

f.newMethod = newMethod

-- you need to have a method name on the receiving end, lest
you rebind f entirely, my mistake.

The ensuing example was more correct.

Kirby



From the_triple_i at hotmail.com  Mon Mar  5 09:04:48 2001
From: the_triple_i at hotmail.com (Alton Luder III)
Date: Mon, 5 Mar 2001 08:04:48 -0600
Subject: Open Source
Message-ID: <9806i4$2e97$1@news3.infoave.net>

Is there any place on the web that offers open source python programs. It
might be helpful for me to learn if I see other people's program's source
code.
Thanks in advance,
Alton Luder III




From francois_forest at yahoo.fr  Sun Mar  4 17:04:20 2001
From: francois_forest at yahoo.fr (Francois Forest)
Date: Sun, 04 Mar 2001 22:04:20 GMT
Subject: screen + sound access
Message-ID: <3AA2BBBC.760DE6CB@yahoo.fr>

I plan to teach programming to teen agers and would need drawing
capabilities as well as paying sounds.
Would Python be a proper candidate ?
Is there any book or tutorial about Python and drawing or Python and
accessing a device ?



From paul at fxtech.com  Sun Mar 11 13:55:10 2001
From: paul at fxtech.com (Paul Miller)
Date: Sun, 11 Mar 2001 12:55:10 -0600
Subject: any interest in type-scoped static constants?
References: <3AAB9578.4C8754E4@fxtech.com> <98g85d$cih$1@panix6.panix.com>
Message-ID: <3AABCA0E.CB3E2F4A@fxtech.com>

> Hmmmm....  The problem is that in Python, you've got a recursive
> structure definition -- is that exact syntax really legal in C++?  In

It's legal - what it's doing is defining a static instance of the class
that is INSIDE the class's scope. It's just a scoping thing.

> Python itself, the way I'd do this is to create a module Point that
> contains the Point class, and following the class definition create the
> ZERO object.  Then you refer to it as Point.ZERO and create new points
> with Point.Point().

You are right that I could do this. However, I'm talking about extension
types. If I am embedding Python in an app, I may be providing several
extension types, that provide access to my internal C++ classes.
However, I do not want to make a module for each type.

I realize I could have a "constants.py" that is automatically imported,
or I could put my constants in my module dictionary, but this is not
object-oriented. I would like to SCOPE my constants by the class/type
they relate to.

What I want is a dictionary that is associated with a TYPE, that
contains constant attributes for the TYPE.


From jkraska1 at san.rr.com  Sat Mar 24 15:37:20 2001
From: jkraska1 at san.rr.com (Courageous)
Date: Sat, 24 Mar 2001 20:37:20 GMT
Subject: static methods
References: <mailman.985383828.2942.python-list@python.org> <slrn9bpqgo.15q.grante@tuxtop.visi.com> <mailman.985457492.29908.python-list@python.org>
Message-ID: <7h1qbt8h8aos1parc090s8tqu2ooc04t4d@4ax.com>

>That it's called "static" is an artifact of C, I believe. In C, "static" 
>has been overexploited such that it has many different meanings in 
>different contexts that have nothing to do with being "static". Hideous 
>example: declaring a function as "static" in C means that it's invisible to 
>code outside of the source file and to all code following it in the same 
>file. :-0

Hrm. I'd forgotten that. In languages where there are no namespaces,
it's useful to be able to make certain that symbols don't go out and
pollute the global namespace in which ALL symbols reside. This is
what that's for.

More generally, "static" in most cases in C and C++ means "global
for the scope that I'm in", err, well, sortof. Examples:

	static class data --> global for the class
	static file data --> global for the file
	static local variables --> global across invocations of the function

I'm in agreement that the keyword "static" is confusing and in some
cases inconsistent.



C//



From piet at cs.uu.nl  Sun Mar  4 09:56:21 2001
From: piet at cs.uu.nl (Piet van Oostrum)
Date: 04 Mar 2001 15:56:21 +0100
Subject: Can SAX parser return <!-- COMMENTS -->  ???
References: <MWxn6.15038$x27.906254@news1.rdc2.on.home.com> <j4d7c0jh03.fsf@informatik.hu-berlin.de>
Message-ID: <wzwva5shuy.fsf@sunshine.cs.uu.nl>

>>>>> Martin von Loewis <loewis at informatik.hu-berlin.de> (MvL) writes:

MvL> "Bruce Fleming" <bfleming at bserv.com> writes:
>> How can I expose comments in the content handler ?
>> It does not return comments by default, and I cannot
>> find where to hook into the class..?

MvL> A ContentHandler never gets comments reported. That is what the
MvL> LexicalHandler is for (which is only defined in PyXML at the moment).

MvL> It is a matter of ongoing debate whether applications should even deal
MvL> with comments; if an application needs XML comments for proper
MvL> operation, something is wrong with that application.

What about a prettyprinter? Doesn't that count as an application?
-- 
Piet van Oostrum <piet at cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: P.van.Oostrum at hccnet.nl


From johngrayson at home.com  Tue Mar  6 18:23:32 2001
From: johngrayson at home.com (John Grayson)
Date: Tue, 06 Mar 2001 18:23:32 -0500
Subject: Need help with Tkinter bind statement
Message-ID: <3AA57174.7070107@home.com>

-----Original Message-----


>  > Hello -
>  > I have been working on an RPN calculator program for a while now, and
>  > I am having problems with some bindings.  Here is a code snippet of
>  > where I'm having trouble:


Snip..


>  > The second problem is that when I
>  > use any of these keys, the keystroke shows up in the entry widget.
>  > What I want to happen is that pressing the "+" key, for example,
>  > should add the values in the xStack and yStack (which it does) and
>  > then put the result in the yStack entry widget (which it also does).
>  > The problem is that the "+" is left in the xStack.  The contents of
>  > the xStack are deleted by the Calc function, so I thought it would
>  > also delete the "+" as well.

>  > Thanks.

>  > Mike

Here is the solution to both your problems...

from Tkinter import *

root = Tk()

def gotit(which):
    print 'Got', which
    return "break"

frame = Frame(root)
entry = Entry(frame)
entry.pack(anchor=CENTER)
frame.pack()

entry.bind('<+>', lambda event, fxn = 'y+x':gotit(fxn))
entry.bind('<Key-minus>', lambda event, fxn = 'y-x':gotit(fxn))
entry.bind('<*>', lambda event, fxn = 'y*x':gotit(fxn))
entry.bind('</>', lambda event, fxn = 'y/x':gotit(fxn))

root.mainloop()


1) You've got to use <Key-minus> since "-" is a valid token
   in the bind...

2) return "break" stops the propagagation of the event to
   the widget, so it does not insert the character.



     John Grayson







From db3l at fitlinxx.com  Mon Mar 19 19:18:14 2001
From: db3l at fitlinxx.com (David Bolen)
Date: 19 Mar 2001 19:18:14 -0500
Subject: Cookies
References: <mailman.984707839.22839.python-list@python.org> <lmq6m07j.fsf@propylon.com> <mailman.984773062.25819.python-list@python.org> <uzoel8ed8.fsf@ctwd0143.fitlinxx.com> <mailman.984791084.26139.python-list@python.org>
Message-ID: <uofux8f95.fsf@ctwd0143.fitlinxx.com>

"Ken Seehof" <kens at sightreader.com> writes:

> > Or, if it isn't going to connect to the server, how about just letting
> > the Python application open up it's own web page internally to run
> > appropriate code on your web server to generate the identifier (which
> > could run as a server script), and then Python could parse it right
> > from the resulting HTML information.
> 
> Again, it's the same catch-22.  I need the server to generate the same
> id that the javascipt generated earlier, but I need each client to have
> a unique id.

I'm still somewhat confused as to the requirement that your javascript
needs the identifier - is there something else it gets used for as
part of the web page other than to generate it on behalf of the
application?

If you support generating the id the first time via a script running
from a normal browser, couldn't you do the same thing as I suggested
above by letting your application itself be that browser - you could
just generate the id server side and parse the results right within
your application, so no need to transfer the id at all.

In other words, why not just use the web page to download the app, and
that's it.  Don't depend on the web page and browser access for
anything else.  Let your application connect to your server (since you
mention it'll do it anyway) and during its first execution get the
server to generate the id and then use it from there.

If you need to have the first generation take place in the context of
an HTTP session, that can still be encapsulated by your Python
application rather than depending on a separate Javascript execution
in a different browser.

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


From Norman_Shelley-RRDN60 at email.sps.mot.com  Mon Mar  5 17:42:33 2001
From: Norman_Shelley-RRDN60 at email.sps.mot.com (Norman Shelley)
Date: Mon, 05 Mar 2001 15:42:33 -0700
Subject: Python 2.1b1 setup.py bug for HPUX11
Message-ID: <3AA41658.AAA154F1@email.sps.mot.com>

In order to build python 2.1b1 for
% uname -a
HP-UX wshelley B.11.00 A 9000/785 2014123772 two-user license

I had to modify the setup.py script as follows. BTW, I did not request
curses or termios in Modules/Setup either.


% diff setup.py setup.py.orig
385c385
<             ###exts.append( Extension('termios', ['termios.c']) )
---
>             exts.append( Extension('termios', ['termios.c']) )
418,419c418,419
<             ###exts.append( Extension('_curses', ['_cursesmodule.c'],
<             ###                       libraries = curses_libs) )
---
>             exts.append( Extension('_curses', ['_cursesmodule.c'],
>                                    libraries = curses_libs) )




From cobrien at Radix.Net  Tue Mar 20 15:45:33 2001
From: cobrien at Radix.Net (Cary O'Brien)
Date: 20 Mar 2001 15:45:33 -0500
Subject: python suitability for large critical run forever apps
References: <3AB77CE4.78378862@troikanetworks.com> <3AB7986A.206EFB37@downright.com>
Message-ID: <998fhd$ssb$1@saltmine.radix.net>

In article <3AB7986A.206EFB37 at downright.com>,
Doug Fort  <dougfort at downright.com> wrote:
>
>--------------1AC435D54686058051C2DC36
>Content-Type: text/plain; charset=us-ascii
>Content-Transfer-Encoding: 7bit
>
>Our website loadtesting application http://www.stressmy.com has backend
>
>agents that 'run forever'. (Actually, we're doing frequent upgrades,
>
>so most agents get restarted at least once a week, but we've
>
>never had anything fail from memory or resource leaks).
>
>I find multithreaded apps much easier to develop and debug in
>
>Python than in C/C++ or Java.  However, I'm becoming convinced
>
>that it's better to use select() than threads for waiting on I/O.
>

Could you elaborate on your reasons?  I'm a bit leery about threads
too, but so far they seem to work, although I hate having all those
locks all over the place.  I really miss the tcl event-driven model
with fileevent and after.  Oh well.

-- cary


From ws-news at gmx.at  Wed Mar  7 13:14:35 2001
From: ws-news at gmx.at (Werner Schiendl)
Date: Wed, 7 Mar 2001 19:14:35 +0100
Subject: Can anyone recomend a good intoduction to C...
References: <mailman.983894288.19965.python-list@python.org> <983896344.701456@newsmaster-04.atnet.at> <m3lmqibw0a.fsf@atrus.jesus.cam.ac.uk> <983911749.217622@newsmaster-04.atnet.at> <UScp6.1097$y6.196347@ruti.visi.com> <983916572.286918@newsmaster-04.atnet.at> <9852b8015jm@news1.newsguy.com> <983967284.246700@newsmaster-04.atnet.at> <o9up6.1291$y6.235469@ruti.visi.com>
Message-ID: <983989153.204220@newsmaster-04.atnet.at>

Grant Edwards <grante at visi.com> wrote in message
news:o9up6.1291$y6.235469 at ruti.visi.com...
> In article <983967284.246700 at newsmaster-04.atnet.at>, Werner Schiendl
wrote:
>
> >> > That is of course a point, if you need to let the program
> >> > inspect you will try to keep it as simple as possible. But
> >> > given the same functionality, I think a C++ program will not be
> >> > more complex.
> >>
> >> The program will not be, the language is.  It's a hard-to-judge
> >> tradeoff.
> >
> >As to my knowledge, the application itself is inspected for
> >safety critical appliances. And you must not change anything
> >afterwards in the software or you need to re-inspect the
> >software.
> >
> >What use would it be to check the language?
>
> The point is that the more complex the language, the more
> difficult it is to inspect the application.  You can look at a
> few lines of C and have a pretty good chance at guessing what
> they do.  In C++ it's much harder to look at snippet of code
> and figure out what it's going to do.
>

At least if the design is bad.

Usually an object oriented program should be easier to read.
Just because the syntax does more closely resemble the way humans think.

(But I saw enough examples that demonstrated just the opposite...)

> --
> Grant Edwards                   grante             Yow!  Somewhere in
DOWNTOWN
>                                   at               BURBANK a prostitute is
>                                visi.com            OVERCOOKING a LAMB
CHOP!!




From sholden at holdenweb.com  Mon Mar  5 11:28:57 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Mon, 5 Mar 2001 11:28:57 -0500
Subject: I come to praise .join, not to bury it...
References: <mailman.983756766.27347.python-list@python.org>
Message-ID: <IcPo6.16785$ML1.838179@e420r-atl2.usenetserver.com>

"Delaney, Timothy" <tdelaney at avaya.com> wrote in message
news:mailman.983756766.27347.python-list at python.org...
> > I have not heard ANY _technical_ arguments opposed to
> > this last time the discussion went around -- nothing but
> > vague aesthetic, "it should be the other way", "I find it
> > ugly" kind of complaints.  Unless and until further notice,
> > then, I class these together with the complaints of people
> > who dislike 0-based indexing, on similar vague bases -- as
> > 0-based indexing _works_ better than 1-based, then, for
> > me, it is a superior choice.  Aesthetics is in the eye of the
>
> I truly have no complaints with the technical merits of ''.join().
>
> However, I have *severe* complaints with the aesthetics and consistency.
> Quite simply, the way we look at things tends to be
>
> "result" = "thing to be acted on to get the result" modified by "thing
which
> does the action"
>
> This is consistently followed in the standard python libraries in almost
all
> cases ... but ''.join() is a glaring inconsistency. As a result, it seems
> unintuitive and ugly.
>
> This is why I prefer string.join(sequence, string). It is now syntactic
> sugar for the above, and documented as such. However, it is *important*
> syntactic sugar. It is known that the method is called polymorphically on
> the second argument (a standard interface) but it feels that it is the
first
> argument which is the basis for the expression. This remains consistent
with
> the rest of the standard libraries.
>
> OTOH, I suggest an alternative syntax ...
>
> (sequence)join.''
>
> There. Everyone should be happy with that ;)
>
See the FAQ: I have added question 4.96. "Why is join() a string method when
I'm really joining the elements of a (list, tuple, sequence)?" which )for
some value of "summarises") summarises the discussions on this one. Please
feel free to shoot it down in flames. Just trying to keep the bandwidth
down.

done-my-good-deed-for-the-day-ly y'rs  - steve

PS: Have a good time at the conference while I'm teaching Unix security in
New Mexico!




From dmaas at dcine.com  Mon Mar 12 16:56:32 2001
From: dmaas at dcine.com (Dan Maas)
Date: Mon, 12 Mar 2001 21:56:32 GMT
Subject: This math scares me
References: <3aad4062.1290021@News.CIS.DFN.DE>
Message-ID: <kCbr6.217438$v.22135339@typhoon.nyroc.rr.com>

> Ok, I can see maybe division having problems. But why does addition of
> the two numbers below:
>
> 5.01+5.54
>
> give me this?
>
> 10.550000000000001

Python uses a standard 4- or 8-byte limited-precision representation for
rational numbers. You will never get more than a few significant digits of
precision. (there are techniques to get more if you really need it, but they
are slow and complicated)

If this bothers you just ignore all but the first few digits...

>>> a = 5.01 + 5.54
>>> print "%.4f" % a

10.5500

Dan




From borealis3 at home.com  Sat Mar 31 09:57:19 2001
From: borealis3 at home.com (Gary Walker)
Date: Sat, 31 Mar 2001 14:57:19 GMT
Subject: Newbie Dictionary Question
Message-ID: <jfmx6.7861$iU.1496746@news1.rdc1.md.home.com>

I *think* I want to do the following:

I want to create a list (this I'm able to do)
Then I want to dynamically create a dictionary for each item in the list,
add some keys (and values), and then put the dictionary in the list.

(This part is frustrating me, because my newbie python code is referencing
the SAME dictionary for each of the list indexes, this NOT what I want. I
want a new instance of a dictionary for each pass thru the list as I loop)

I want to do this because this way I can have an ordered list of key/value
pairs... at least that was my plan...

In other words:
How do I dynamically create a dictionary whose name is not known until
runtime?

Or, what's a better way to do this (i.e. create an ordered list of key/value
pairs)

Actually, they don't have to be ordered, so I suppose (thinking aloud here)
that I could use a dictionary of dictionaries, but again - it seems to me
that I'd need to dynamically create a dictionary whose name isn't known
until runtime...

Or am I missing the obvious? This is the most likely scenario :)

TIA,

Gary




From me at nospam.net  Fri Mar 30 14:37:19 2001
From: me at nospam.net (Scottie)
Date: Fri, 30 Mar 2001 11:37:19 -0800
Subject: Getting all the *files* from a directory -- A better way??
References: <mailman.985732718.3725.python-list@python.org> <99r6v7$4hef2$1@fido.engr.sgi.com> <mailman.985740096.16937.python-list@python.org>
Message-ID: <tc9o3kjpnavs6b@corp.supernews.com>

"Ken Seehof" <kens at sightreader.com> wrote in message
news:mailman.985740096.16937.python-list at python.org...
>... e.g.:
> f = os.path.join(root, 'doc/chapter2/spam.html')
> rather than
> f = os.path.join(root, os.path.join('doc', os.path.join('chapter2',
> 'spam.html'))))
>

I just thought I'd remind you:
   f = os.path.join(root, 'doc', 'chapter2', 'spam.html')
works as well.  It is easy to forget join is a multi-arg function.

-Scott David Daniels





From DanielK at jBASE.com  Wed Mar 21 15:30:22 2001
From: DanielK at jBASE.com (Daniel Klein)
Date: Wed, 21 Mar 2001 12:30:22 -0800
Subject: Substitute for join()
References: <Mo7u6.58$Hp.2896@typhoon.aracnet.com>
Message-ID: <Qb8u6.62$Hp.3417@typhoon.aracnet.com>

Thanks everyone for your public and private answers. This is the greatest
bunch of folks I have ever had the pleasure of being a part of!

I have indeed recoded to use string.join().

Again thanks,
Dan

"Daniel Klein" <DanielK at jBASE.com> wrote in message
news:Mo7u6.58$Hp.2896 at typhoon.aracnet.com...
> I've got to make a script (written in version 2.0) to work on a machine
> running 1.5.2. The only incompatible line is one that uses the
> "".join(mylist) method. To 'get out of the woods' quickly, I substituted
it
> with
>
> s = ''
> for element in mylist: s = s + element
>
> Is this the most efficient way to code this?
>
> Thanks,
> Daniel Klein
>
>




From phd at phd.pp.ru  Tue Mar 20 11:22:20 2001
From: phd at phd.pp.ru (Oleg Broytmann)
Date: Tue, 20 Mar 2001 19:22:20 +0300 (MSK)
Subject: python suitability for large critical run forever apps
In-Reply-To: <3AB77CE4.78378862@troikanetworks.com>
Message-ID: <Pine.LNX.4.30.0103201920030.4007-100000@fep132.fep.ru>

Hi!

   I recommend you to look briefly into some projects, written in Python.
Zope (www.zope.org) is well-know web application server. It is
multithreaded server, that can run for months without any problem. Mailman
(www.list.org) is a mailing list manager. It is so good that Free Software
Foundation adopted it is their "official" list manager.

On Tue, 20 Mar 2001, Bruce Edge wrote:
> I started using Python for come user CLI stuff. It has worked out so
> well that we're thinking about extending it's use in our project.
> I am not experienced enough to be able to answer some of the issues that
> came up. Does anyone have any thoughts on these:
>
>
> > > *data* size of each Python interpretter plus however much space
> > > each object takes versus C++'s minimal data resource requirements.
> >
> >     Any rough numbers (shared interpreter code, shared data, unshared data
> >     and object overhead?)
>
> Multi-threading, and mutual exclusion. Are these available, reliable,
> and debuggable?
>
> > > Is Python a "toy" language or is it industrial
> > > strength ? For instance, does it facilitate recovering from exceptions,
> > > and leaving objects in good states (the strong gurantee) ?  When you
> > > are low on memory, will the fact that more Python operations require
> > > dynamically allocated memory make implementations more brittle ? Is it
> > > even possible to write a line of Python where you don't have to
> > > worry about out-of-memory exceptions (the no-throw gurantee) ?
> > > IOW if I'm in a low memory situation, can I perform any operations
> > > to even fail gracefully ?
>
> > > Regarding large-scale apps again, is there anything written in Python
> > > which "runs forever" ? That is, it doesn't just die to clean up and
> > > reset itself ?

Oleg.
----
     Oleg Broytmann            http://phd.pp.ru/            phd at phd.pp.ru
           Programmers don't die, they just GOSUB without RETURN.




From syver at NOSPAMcyberwatcher.com  Mon Mar 26 03:54:26 2001
From: syver at NOSPAMcyberwatcher.com (Syver Enstad)
Date: Mon, 26 Mar 2001 10:54:26 +0200
Subject: Threading legality/morality
Message-ID: <fsDv6.1247$NR.100201@news3.oke.nextra.no>

How is calling print from more than one thread looked upon?




From fellowsd at cs.man.ac.uk  Wed Mar 14 10:39:13 2001
From: fellowsd at cs.man.ac.uk (Donal K. Fellows)
Date: Wed, 14 Mar 2001 15:39:13 +0000
Subject: Einstein's Riddle
References: <oNPq6.396644$ge4.138479566@news2.rdc2.tx.home.com> <UdSq6.2184$54.2166@www.newsranger.com> <984380185.768920@news.aei.ca> <uP%q6.2375$54.2207@www.newsranger.com> <4w6r6.193645$df5.5192565@news1.crdva1.bc.home.com> <3aae4114.63604234@nntp.interaccess.com> <3AAE4BFD.9A7B0D9C@cs.man.ac.uk> <3AAF1240.C446D19@iit.demokritos.gr> <wi67l1spme5.fsf@merlin.irisa.fr>
Message-ID: <3AAF90A1.AF5B238F@cs.man.ac.uk>

Yoann Padioleau wrote:
> Petasis George <petasis at iit.demokritos.gr> writes:
>> Well, try this:-)
>> Written by a friend of mine. Its amazing how small can programs be if you
>> know STL. To bad I don't know stl :-)
>> Of course it will need some time, as it randomly checks all combinations...
> But your program dont really do the job cos it is too slow.
> You first generate the permutations and then test, the pb is that
> there is too many permutations :
> p[nation][english] can be either 0 1 2 3 4 or 5, same for the other
> variables you have 5*(5!) possibililty = 24 883 200 000
> if you are able to do 1 000 000 ValidConstraints per second (cos you
> have perhaps a very very fast computer) you will need 24 883 seconds
> =~ 6hours to compute (in the worst case i admit)
> 
> the prolog program i propose take 0.1 second to find the solution :)
> who say that prolog is slower than c++ :))

Having read both C++ and Prolog versions, do either of you wish to take on
the extension and write a program to take a natural language description of
the problem and solve it?  In this caffeine-hyped .com-ready electronically
enabled speed-of-light online world of ours, I might not have time to take
your existing code and adapt it to the specific problem I'm interested in.

Scanner support is optional.  >:^)

Donal.
- 
Donal K. Fellows    http://www.cs.man.ac.uk/~fellowsd/    fellowsd at cs.man.ac.uk
-- With a complex beast like Swing, it's not just a matter of "What button
   should I push", but rather "How do I put myself into a nice metamorphosis
   so that I am deemed acceptable by the Swing Gods."             -- Anonymous


From whisper at oz.net  Wed Mar 21 12:29:47 2001
From: whisper at oz.net (Dave LeBlanc)
Date: 21 Mar 2001 17:29:47 GMT
Subject: Little red lines in PythonWin!
References: <mailman.985107113.4466.python-list@python.org> <uu24ojk5h.fsf@ctwd0143.fitlinxx.com> <9998c2$a16$0@216.39.170.247>
Message-ID: <99aoeb$9en$0@216.39.170.247>

Ugh - take it all back! I wuz wrong!

\t\s != \s\t since \t\s is (# of tabspaces + \s) and \s\t is only (#
of tabspaces - leading \s) since it "rounds off" to the multiple and
therefore the # of spaces inserted varies depending on leading spaces.

My bad - sincere apologies. I'll study Perl in penance.

Dave LeBlanc


On 21 Mar 2001 03:49:22 GMT, whisper at oz.net (Dave LeBlanc) wrote:

>>"... if spaces get mixed in it's bound to cause problems.  So in
>>general that's nasty..."
>
>Urban legend perhaps?
>
>Seems like, so long as you actually indent dependent lines, such as
>those comprising the body of an if statement, as little as one space
>or tab character then you're legal.
>
>So long as a text processor expands tabs to the same number of spaces,
>then I don't see how any tab+space sequence that only varies by the
>number of explicit space characters can ever possibly result in the
>same or less whitespace to the left of text on a line. This makes it
>hard to understand why one wouldn't want to use a mix of tabs and
>spaces, especially "because of the nature of text editors on non-UNIX
>platforms" (quote from 2.1.7).
>
>\t\t\t\s\s\s and \s\s\s\t\t\t where \t --> \s\s\s\s (and \s represents
>a single space character) will always create the same amount of space
>in any text editor (as contrasted to a word processor) i've ever used
>on a wide variety of OSs, including places where sp != 0x20 and tab !=
>0x09 (EBCDIC et al.). Even text editors that allow the use of
>proportional fonts seem to have a fixed idea of the screen area taken
>up by a space character. (People who use word processors for text
>editing, especially source code, deserve what they get.)
>
>Frankly, using that many spaces (8) to indent seems like a waste of
>screen real estate, not to mention making the code actually harder to
>read. Perhaps 8 spaces/tab made sense when the primary output medium
>was a decwriter or a tty, which was the last time I used so many
>spaces to replace a tab, but not anymore.
>
>Just my opinion...
>
>Dave LeBlanc
>
>On 20 Mar 2001 14:49:14 -0500, David Bolen <db3l at fitlinxx.com> wrote:
>
>>In theory you really shouldn't get into trouble using tabs as long as
>>you follow Python's parser definition of what a tab is (treat as
>>moving to next multiple of 8 - Language Reference 2.1.7).  The trouble
>>is mostly when an editor is using an explicit tab character but
>>internally adjusting its dimension to something other than that rule.
>>Even that might work as long as only tabs are used for indentation,
>>but if spaces get mixed in it's bound to cause problems.  So in
>>general that's nasty, but as long as you untabify in that same
>>editor/IDE before distribution, since it knows its own rules, the
>>untabified version should be fine.
>>
>



From cce at clarkevans.com  Sun Mar 25 17:09:34 2001
From: cce at clarkevans.com (Clark C. Evans)
Date: Sun, 25 Mar 2001 17:09:34 -0500 (EST)
Subject: Yet Another PEP:  Query Protocol Interface or __query__
In-Reply-To: <99lnc6$ck4$1@tyfon.itea.ntnu.no>
References: <Pine.LNX.4.21.0103201837570.21343-100000@clarkevans.com> <mailman.985490195.2917.python-list@python.org> <99lnc6$ck4$1@tyfon.itea.ntnu.no>
Message-ID: <Pine.LNX.4.21.0103251655070.4503-100000@clarkevans.com>

On Sun, 25 Mar 2001, Magnus Lie Hetland wrote:
> "Clark C. Evans" <cce at clarkevans.com> wrote in message:
> >     This proposal puts forth a light-weight explicit mechanism
> >     for the adaptation (including verification) of an object to
> >     a context where a specific type, class, interface, or other
> >     protocol is expected.
> 
> I like the generality here -- nice PEP overall, IMO :)

Thank you.

I've asked for a location on source forge for it.  
After talking with a few people, I think it would be 
appropriate to field test it for a while; if the
component falls into a resonable amount of usage
then I'll ask for it to be built-in.  This will also
allow for any changes to be made before it is 
"permanent".  

> >     To enable the first case listed in the requirements,
> >     the adapt function first checks to see if the objects
> >     type or the objects class are identical to the protocol.
> >     If so, then the adapt function returns the object directly
> >     without further ado.
> 
> Hm. So in some cases the protocol object is merely an example
> (prototype) of an object satisfying the protocol, and in other
> cases it represents the protocol itself (with __prodapt__ etc)?

Sorry if this is unclear, a protocol is a TypeType or a TypeClass.
If an __adapt__ method wants to use another python type for
a protocol, then the framework doesn't care.

> In that case, one could perhaps use adapt(obj,SomeClass), and
> the function would check if obj was an instance of SomeClass;
> it would *not* check whether SomeClass and obj were instances
> of the *same* class (which seems semantically weird in this
> context to me...)

Right.
 
> *Or*, the intransitive version of __adapt__ could be
> called __conform__, which is indeed intransitive. Then the
> transitive version could be called __adjust__ or __adapt__ or
> whatever. I think that gets my vote:
> 
>   __conform__ - method in an object which can conform to
>                 a protocol
>   __adapt__   - method in a protocol used to adapt an
>                 object to it

Very Nice.  Much better than __prodapt__, which is almost a joke.

;) Clark




From tim.one at home.com  Wed Mar 14 17:08:07 2001
From: tim.one at home.com (Tim Peters)
Date: Wed, 14 Mar 2001 17:08:07 -0500
Subject: Who am I: can a class instance determine its own name?
In-Reply-To: <98o8fu$2s4dj$1@ID-11957.news.dfncis.de>
Message-ID: <LNBBLJKPBEHFEDALKOLCAELNJFAA.tim.one@home.com>

[Emile van Sebille]
> Has locals changed so that it may be written to reliably?  My
> documentation still warns:
>
> locals ()
> Return a dictionary representing the current local symbol table. Warning:
> The contents of this dictionary should not be modified; changes may not
> affect the values of local variables used by the interpreter.

No, this has not changed.  As the docs say, writing to locals() is not
allowed.

and-more-likely-than-not-will-be-enforced-someday-too-ly y'rs  - tim




From slinkp23 at yahoo.com  Wed Mar 14 02:08:17 2001
From: slinkp23 at yahoo.com (Paul Winkler)
Date: Wed, 14 Mar 2001 07:08:17 GMT
Subject: Einstein's Riddle
References: <mailman.984369618.12493.python-list@python.org> <3AACBB41.5854666A@yahoo.com>
Message-ID: <3AAF1885.FF28CC21@yahoo.com>

Steve wrote:
(snip)
> Only one IQ test, the Stanford-Binet form L-M, has
> been shown to reliably discern among IQ's higher than about 150,

If there are no other reliable measures of IQ > 150, how can the S-B
test be proven reliable?

-- 
.................    paul winkler    ..................
slinkP arts:   music, sound, illustration, design, etc.
           web page:  http://www.slinkp.com
      A member of ARMS:   http://www.reacharms.com


From francois.granger at free.fr  Wed Mar 28 13:37:47 2001
From: francois.granger at free.fr (=?ISO-8859-1?Q?Fran=E7ois_Granger?=)
Date: Wed, 28 Mar 2001 18:37:47 GMT
Subject: Creation d'un NG francophone sur Python
Message-ID: <1eqzyk6.gxd4jr1tkjykzN%francois.granger@free.fr>

Sorry for this french message. Translation follow.

Il est encore temps de voter pour la cr?ation d'un NG en fran?ais sur
Python. Allez dans 'fr.usenet.forums.annonces' et cherchez le message
'[AAV 3] Creation de fr.comp.lang.python (non-modere)'
votez en r?pondant ? ce message.

[en]
It is time for voting to create the french speaking newsgroup
'fr.comp.lang.python'. Go to 'fr.usenet.forums.annonces' and look for
the message '[AAV 3] Creation de fr.comp.lang.python (non-modere)'.
Votes by replying to this message.

-- 
"La connaissance est le chemin de la tol?rance, c'est valable pour 
tous, en toutes saisons."
- Raymond Page


From urner at alumni.princeton.edu  Sat Mar  3 20:08:50 2001
From: urner at alumni.princeton.edu (Kirby Urner)
Date: Sat, 03 Mar 2001 17:08:50 -0800
Subject: Hitting two targets:  OO + group theory (pedagogy)
References: <u2k0atkgkiatdp9ukj3mpd1ej0b648ldsm@4ax.com> <3aa1077a.1110803@nntp.sprynet.com> <0932atgt9v8d8aboq90evoi5ipj5c0sk8s@4ax.com> <3aa11a91.5998558@nntp.sprynet.com> <ko72atgf5hmg66suefbenp8lvn6vhblajv@4ax.com> <3aa13152.173635120@news.okstate.edu>
Message-ID: <u2s2atcprnkj5tfl5q2gvg6s9nh0ip6vme@4ax.com>

ullrich at math.okstate.edu (David C. Ullrich) wrote:

>Right. I didn't say it was inappropriate. Although in fact
>something about the bit about "If you've studied Genesis"
>or however you put it seems a little off-putting; if
>you were referring to Bugs Bunny or Zeus you'd
>just refer to them.

I understand.  Actually I was trying to sound international
and multi-ethnic -- not presuming my reader had ever heard
of Cain/Abel and suggesting a search engine might pull up
some of that Bible stuff -- just as I'd need to search re
characters in other storylines.  But I understand your 
sensitivity.

I do expect this OO thing eventually might inspire _other_ 
authors to write stuff like:

   class Man(God):

      self.lastwords = "To thee I commend my spirit..."

      def __init__(self):
          self.soul = God.lend(self)
          self.years_to_live = random.choice(range(113))

      def live_for_a_day(self):
          self.days_to_live -= 1
          if self.days_to_live == 0:
              self.die()

      def die(self):
          print "Last words: %s " % self.lastwords
          God.getback(self.soul)
            
but don't worry, I'm not planning to do _that_ website.

>Never mind - could be I've been in Okllahoma
>too long.
>

That's somewhere near Texas right? :-D

>>In my example, it's the elements of a group which
>>get instantiated as separate objects, with each 
>>element inheriting the same meaning for *, whatever
>>that might be. 
>
>Right. That's actually what I meant, a class Group
>instances of which would represent elements
>of a generic group - of course there's no such
>thing so you'd never create instances of Group.
>Then subclasses of Group represent actual
>groups; instances of the subclasses being
>elements of actual groups.
>
>Not important, just seems to me like what it
>"should" be.

After further consideration, I think you're right.  

However, a Group and a member of a Group are somewhat
distinct ideas, so an instance of a Group would not
be a member, but a kind of group, e.g. a group of 
coprimes.  Internal to each kind of group, you have 
your conception of a member -- meaning you endow your
paradigm member with knowledge of how to be fruitful
and multiply (*) in a closed universe :-D.

In Java we have what are called abstract classes, which,
as you say, are never implemented directly -- only 
their subclasses might be.  Another option is to define
an "interface", which is a list of names specifying the
methods you'd need to implement in order to legitimately 
advertise you support said interface (like a canned 
control panel -- these are the controls we'd expect 
of any foo (any so-called 'car' must have a way to 
'steer').

So a finite Group interface might include:

(C) List all members                    getall()
(A) multiply two members (associative)  mul(a,b)
(I) Supply the inverse of a member      a.inv()
(N) output the identity element         id()

Based on these ruminations, I went over my Python code 
to create a completely revamped version:

  http://www.inetarena.com/~pdx4d/ocn/python/groups_draft2.html

which is rather more advanced than before (might still 
have a bug or two though -- this is definitely beta).

The [finite] Group class now contains the code for spitting 
out a table of all possible products, but it will be 
looking to a subclass to define what this means.

The Coprimes class inherits from Group, and contains an
internal representation of a Member (another class), 
wherein __mul__, __pow__, and inv() are giving the 
relevant definitions for this particular kind of group 
(positive integers coprime to and less than a modulus).

A main advantage of coding in this way is I can get several 
Coprimes groups instantiated (as objects) and work with 
them all in memory simultaneously (example below).

A next challenge would be to define my Permutations subclass
as also inheriting from Group, with its own internal conception 
of a Member.  That'd give an even better idea of the super->sub
(in principle -> specialcase) hierarchy.

Thanks for prompting me to take a second and third look at 
this issue.

Kirby

===============================

Sample transcript:


 Python 2.1a2 (#10, Feb  2 2001, 16:01:03) [MSC 32 bit (Intel)] on win32
 Type "copyright", "credits" or "license" for more information.
 IDLE 0.6 -- press F1 for help

 >>> import primes2             # what I call the module on my end
 >>> g17 = primes2.Coprimes(17) # create group of 0<k<17 where gcd(k,17)=1
 >>> len(g17)                   # this method at the Group level
 16
 >>> g20 = primes2.Coprimes(20) # another group instance, coprimes of 20
 >>> len(g20)                   # invokes same Group-level method on self
 8
 >>> g20.getproducts()          # getproducts() is at the Group/top level
 
       1  3  7  9 11 13 17 19
    ----------------------------
  1|   1  3  7  9 11 13 17 19
  3|   3  9  1  7 13 19 11 17
  7|   7  1  9  3 17 11 19 13
  9|   9  7  3  1 19 17 13 11
 11|  11 13 17 19  1  3  7  9
 13|  13 19 11 17  3  9  1  7
 17|  17 11 19 13  7  1  9  3
 19|  19 17 13 11  9  7  3  1
 >>> g20.explist()              # exponent of each member, method in Coprimes
 [(1, 1), (3, 4), (7, 4), (9, 2), (11, 2), (13, 4), (17, 4), (19, 2)]
 >>> g17.explist()              # same method, but different self/instance
 [(1, 1), (2, 8), (3, 16), (4, 4), (5, 16), (6, 16), (7, 16), (8, 8), 
  (9, 8), (10, 16), (11, 16), (12, 16), (13, 4), (14, 16), (15, 8), (16, 2)]
 >>> a=g20[2]
 >>> a                          # indexing is from 0
 7
 >>> a*a                        # powering a member of g20 (2nd power)
 9
 >>> a*a*a                      # powering a member of g20 (3rd power)
 3
 >>> a*a*a*a                    # powering a member of g20 (4th power)
 1
 >>> a*a.inv()                  # a.inv() = a*a*a, since (a*a*a)*a = 1
 1




From moshez at zadka.site.co.il  Mon Mar 19 12:13:59 2001
From: moshez at zadka.site.co.il (Moshe Zadka)
Date: Mon, 19 Mar 2001 19:13:59 +0200
Subject: Python-related standards
In-Reply-To: <3dae6h21h8.fsf@ute.cnri.reston.va.us>
References: <3dae6h21h8.fsf@ute.cnri.reston.va.us>
Message-ID: <E14f3EF-000579-00@darjeeling>

On 19 Mar 2001, Andrew Kuchling <akuchlin at mems-exchange.org> wrote:

> It struck me today that it might be a good idea to create a series of
> Python-related standard documents that would be a counterpart to PEPs.
> We already have some documents that would fit into this series:
> 
>         * The DB-API.
>         * The DO-SIG's CORBA mapping.
>         * The SAX and DOM mappings.

FWIW, these could be "informational" PEPs.

> for Python packages (currently being done in PEP 241), or defining a
> structured text format for docstrings.  IMHO such topics aren't really
> suitable for a PEP, since they don't necessarily have anything to do
> with the core interpreter or language.

Which is exactly what "informational" (as opposed to standards track)
PEP were meant to be.

-- 
"I'll be ex-DPL soon anyway so I'm        |LUKE: Is Perl better than Python?
looking for someplace else to grab power."|YODA: No...no... no. Quicker,
   -- Wichert Akkerman (on debian-private)|      easier, more seductive.
For public key, finger moshez at debian.org  |http://www.{python,debian,gnu}.org



From emile at fenx.com  Sat Mar 31 20:58:49 2001
From: emile at fenx.com (Emile van Sebille)
Date: Sat, 31 Mar 2001 17:58:49 -0800
Subject: PythonLabs Team Moves to ActiveState
References: <mailman.986086501.5576.python-list@python.org>
Message-ID: <9a626p$3qrct$1@ID-11957.news.dfncis.de>

Guido and team:

I'm sorry things didn't work out with Digital Creations.  I had hoped that
Python core development had found a natural home, and was looking forward to
seeing the results of a combined effort.  Somehow, imagining a future with
enhanced acquisition is much more interesting than a future with curly
braces.  Can IFEND and DEFEND or even a fork be far behind? (stackless?) I
think that the "build it and they will come" philosophy has worked nicely so
far.

Some clarification may help.  Perhaps you can reconcile the *optional* curly
braces and the "from __future__ import" discrepancy.  It may just be my
misunderstanding, but isn't __future__ designed to allow bleeding-edgers to
write code for what will become python?  That is, won't all __future__
thingys become mandatory?  I believe this is how it has been presented so
far, and makes me think that *optional* curly braces won't be after a short
transition period.

It also occurs to me that ActiveState is already deep into .Net and python
for .Net.  Couldn't you simply provide a 'python for perlites' obfustication
mode?  ;-)

The good news is that I believe that ActiveState will do good things for
Python, and as you say, should provide a stable home for python development.
You join a team of highly respected python people in this move, and the
resulting products will be far from common, and that is exciting.

Good luck to you all,

--

Emile van Sebille
emile at fenx.com

---------
"Guido van Rossum" <guido at python.org> wrote in message
news:mailman.986086501.5576.python-list at python.org...
> PythonLabs Team Moves to ActiveState
>
> To all Python users and developers:
>
>    Less than half a year ago, I moved with my team to Digital Crations, in
the
>    hope of finding a new permanent home for Python development. At
>    Digital Creations we've done several good things for Python, such as
forming
>    the Python Software Foundation and the management of the Python 2.1
release.
>
>    Unfortunately, Digital Creations's original plans for PythonLabs didn't
>    work out as hoped, and we weren't able to reach mutual agreement on
>    workable alternative plans - despite trying for months.
>
>    I am proud to have found a new home for my entire team: starting
>    today, Tim Peters, Barry Warsaw, Jeremy Hylton, Fred Drake and myself
>    are working for ActiveState.

<snip>




From jacobus at geo.co.za  Tue Mar 13 07:56:15 2001
From: jacobus at geo.co.za (Jacobus van der Merwe)
Date: Tue, 13 Mar 2001 14:56:15 +0200
Subject: Writing redirected e-mail to file
Message-ID: <3AAE18EF.6083CD74@geo.co.za>

Hi,

I use sendmail to redirect e-mails to a certain e-mail address to a
script.
This script must 'look' at the headers and body of the e-mail, and then
write the e-mail body to a file.

I can't figure out exactly how to do this:

I use rfc822.Message to get the headers, but how do I get the body?

Can one use mimetools.Message?   What does 'fp' in mimetools.Message
parameters mean?  Once read with mimetools.Message, how do I write
it to a textfile.

The body will consist of only plain text.

Regards,

Jacobus




From tdelaney at avaya.com  Wed Mar  7 18:33:46 2001
From: tdelaney at avaya.com (Delaney, Timothy)
Date: Thu, 8 Mar 2001 10:33:46 +1100 
Subject: I come to praise .join, not to bury it...
Message-ID: <B43D149A9AB2D411971300B0D03D7E8B0FBF3C@natasha.auslabs.avaya.com>

> comma.join()
> period.join()
> bang.join()
> etc.

What's a period? I would look at this as

comma.join()
fullstop.join()

That's one of the problems - even in English, we sometimes have completely
different words for the same symbols, let alone spelling. I mean, *why*
would you drop all the 'u's?

Tim Delaney
Avaya Australia

All recipient(s) of this email have permission to forward or reply to this
email, quoting this email in full or in part.



From t*b*r*u*m*l*e*y at f*u*s*e.n*e*t  Sat Mar 24 20:38:36 2001
From: t*b*r*u*m*l*e*y at f*u*s*e.n*e*t (Troy Brumley)
Date: Sat, 24 Mar 2001 20:38:36 -0500
Subject: Collection interfaces
References: <3a90fa97.0@news1.mweb.co.za>   <MPG.150394195f7f04a19896a7@news.earthlink.net> <3A9A86B1.9A8C69A2@ix.netcom.com> <MPG.1504d14563dce6499896af@news.earthlink.net> <97fjla128ir@news2.newsguy.com> <97h5s40vhc@news2.newsguy.com> <MPG.1506007591cf818b9896bc@news.earthlink.net> <slrn99opop.gr.amk@mira.erols.com> <MPG.1506265ebf6d934d9896be@news.earthlink.net> <97ibmc0k47@news1.newsguy.com> <MPG.1506cf3293b094b49896c1@news.earthlink.net> <97m0c7080c@news2.newsguy.com> <MPG.15085517debf4b649896ca@news.earthlink.net> <zb%n6.31033$o%6.586026@news1.rdc1.bc.home.com> <MPG.150b383cea226d409896d0@news.earthlink.net> <kNko6.6986$eq3.707121@news.uswest.net> <MPG.150c46db9cec283e9896d4@news.earthlink.net> <3AA2A898.450B659E@mail.com> <gRCt6.12224$Im6.1308560@newsread1.prod.itd.earthlink.net> <3AB755F8.67E84126@mail.com> <%a%t6.17245$227.1601569@newsread2.prod.itd.earthlink.net> <3AB8A730.4BE13D22@mail.com> <61bv6.8693$P4.676800@newsread1.prod.itd.earthlink.net>
Message-ID: <B6E2B64C.D8E3%t*b*r*u*m*l*e*y@f*u*s*e.n*e*t>

in article 61bv6.8693$P4.676800 at newsread1.prod.itd.earthlink.net, topmind at
topmind at technologist.com wrote on 3/24/01 7:33 PM:

> 
> "James A. Robertson" <jarober at mail.com> wrote in message
> news:3AB8A730.4BE13D22 at mail.com...
>> topmind wrote:
>>> 
>> 
>>>> http://www.cincom.com/smalltalk
>>>> 
>>> 
>>> I only see a front page, not a parsing example.
>>> 
>> 
>> Follow the tutorial link - it's right there on that page.
>>
> 
> I cannot find any "tutorial" link.

This may be a victim of putting the site up on the in development "web
works" system. This part of our site is now powered by Smalltalk.

http://www.cincom.com:80/scripts/smalltalk.exe/gettingstarted/index.asp?cont
ent=trysmalltalknc

Pardon the word wrapping my news client does. There's no spaces in the
above.

That should get you to a place where you can download the lesson browser
(thanks again to Mark Windholtz, great job!) and the tutorial (thanks to Jim
F. as well).

> Perhaps you are afraid that if you give me the full tutorial link, you
> will have to expose to the world the FACT that the site runs on
> Microsoft ASP VSscript instead of Smalltalk?

Naw. Jim isn't afraid of anything. At least, nothing on Usenet!

Me either :)

The rest of the Cincom site, and the old Smalltalk portion of the site, does
indeed not run on Smalltalk. We'll get it all moved over eventually :)

Heck, my own business unit, which is very into VisualWave, understands that
we have to use JavaScript on our pages to drive the browsers. There's
nothing wrong with being multi-lingual.

> 
> Do Chevy salesmen drive Fords?

Some do. I work for Cincom and use both our Smalltalk products, I actually
prefer ObjectStudio when I'm doing Windows specific work, but I also have
Squeak and Dolphin around and have experimented with both.

> -tmind-



From grante at visi.com  Tue Mar  6 16:44:20 2001
From: grante at visi.com (Grant Edwards)
Date: Tue, 06 Mar 2001 21:44:20 GMT
Subject: Can anyone recomend a good intoduction to C...
References: <mailman.983894288.19965.python-list@python.org> <983896344.701456@newsmaster-04.atnet.at> <m3lmqibw0a.fsf@atrus.jesus.cam.ac.uk> <983911749.217622@newsmaster-04.atnet.at>
Message-ID: <UScp6.1097$y6.196347@ruti.visi.com>

In article <983911749.217622 at newsmaster-04.atnet.at>, Werner Schiendl wrote:

>> Yes, you're missing the fact that C++ sucks large rocks through very
>> small straws.  There's some merit in learning C for when low-levelness
>> is wanted, eg. when developing Python.  If you're writing an
>
>I do not see the point, why not use C++ for low-levelness.
>Years ago I had a lot of discussions why use C instead of Assembler.
>The arguments were mainly for performance.
>And still C made the race, guess why?
>It is easier, more powerful, more safe (the compiler can catch more bugs
>then the assembler)

C++ is far more complex that C without being much safer.  Since
the safety has to be "inspected" into the program, you want
to make things as easy to inspect as possible: a nice simple
language.

-- 
Grant Edwards                   grante             Yow!  Have my two-tone,
                                  at               1958 Nash METRO brought
                               visi.com            around...


From gunturu_neelima at rediffmail.com  Tue Mar  6 02:26:02 2001
From: gunturu_neelima at rediffmail.com (neelima  g)
Date: 6 Mar 2001 07:26:02 -0000
Subject: (no subject)
Message-ID: <20010306072602.9821.qmail@mailweb7.rediffmail.com>

after creating the empty database,how to enter the values into it  which is having morethan two columns.
                       confirm 881077


_____________________________________________________
Chat with your friends as soon as they come online. Get Rediff Bol at
http://bol.rediff.com







From corour01 at motorola.com  Fri Mar 23 06:10:13 2001
From: corour01 at motorola.com (O'Rourke Clodagh-corour01)
Date: Fri, 23 Mar 2001 11:10:13 -0000
Subject: Tkinter - Widget Colours
Message-ID: <13EE655665F4D311AB4D0008C789498A01EE2C9C@zei02exm02.cork.cig.mot.com>

Hi all,

I was wondering does anyone know how the bag/fag colours can be changed after the widget is packed.  

Thanks,
Clodagh



From jeremy at alum.mit.edu  Mon Mar 19 14:22:13 2001
From: jeremy at alum.mit.edu (Jeremy Hylton)
Date: Mon, 19 Mar 2001 14:22:13 -0500 (EST)
Subject: Why "from __future__" stinks; a counter-offer
In-Reply-To: <995h3l$2kp$1@panix6.panix.com>
References: <mailman.984975803.9006.python-list@python.org>
	<jwbnews-3793B2.08394319032001@corp.supernews.com>
	<slrn9bcg72.9q8.scarblac@pino.selwerd.nl>
	<995h3l$2kp$1@panix6.panix.com>
Message-ID: <15030.23653.43360.232412@w221.z064000254.bwi-md.dsl.cnc.net>

>>>>> "A" == Aahz Maruch <aahz at panix.com> writes:

  A> In article <slrn9bcg72.9q8.scarblac at pino.selwerd.nl>, Remco
  A> Gerlich <scarblac at pino.selwerd.nl> wrote:
  >> John W. Baxter <jwbnews at scandaroon.com> wrote in
  >> comp.lang.python:
  >>>
  >>> For the future, the time to complain about PEP contents is when
  >>> the PEP is posted, not after it has been posted, discussed,
  >>> implemented, and the result distributed.
  >>
  >> And the place to complain is the Python-DEV list, which is not
  >> public, although everyone can post to it and everyone can read it
  >> using some archives...

  A> Officially, complaints need only be posted to c.l.py.  It's not
  A> clear to me at this point what the True Practice is, but sending
  A> e-mail to the PEP champion is supposed to work best of all.  (Not
  A> all PEP creators are on python-dev.)

Discussion on comp.lang.python is fine, although it would usually help
to cc the PEP author.  If there are legitimate complaints posted to
c.l.py and the PEP author ignores them, I expect someone will cry
foul.  Tim reads c.l.py as devotedly as anyone.  Barry and I try to
keep up, too.

Jeremy



From tim.one at home.com  Sat Mar  3 17:47:15 2001
From: tim.one at home.com (Tim Peters)
Date: Sat, 3 Mar 2001 17:47:15 -0500
Subject: I come to praise .join, not to bury it...
In-Reply-To: <97rka002k9@news1.newsguy.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCAEMAJDAA.tim.one@home.com>

[Alex Martelli, with a wonderfully on-target explanation of why it's
 sep.join(seq) and not seq.join(sep)]

Another reason could be that Guido is a dictator, and, like all dictators,
he's interested in killing off the various opposition groups that arise from
time to time.  Now any dictator would be mad to believe that the opposition
will be cooperative enough to implement a method that lets him kill them off
by passing an instrument of execution:

    future_haters.kill(knife)
    curly_brace_lovers.kill(slow_strangulation)

This puts it outside his control!  Much better for him if opposition groups
merely agree to enumerate themselves (and they probably need to have such a
protocol anyway, if for no other reason than to count their burgeoning
numbers).  Then he can satisfy his blood lust via simply teaching his weapons
of mass destruction how to enumerate collections of people:

    poison.kill(coalition_for_less_frequent_releases)
    neutron_bomb.kill(coalition_for_more_frequent_releases)

Guido can deal with the enumeration mechanics just once then, in his
LethalWeapons base class.

Unfortunately, plain "kill" on its own doesn't make clear which is being
killed and which is the instrument of death.  Ditto for plain "join".
Perhaps this whole brouhaha could have been avoided had we named
sep.join(seq) as sep.joining(seq) instead?  Yuck.  Anyone who suggests that
will be shot <wink>.

hoping-i'm-not-spilling-too-many-of-guido's-secrets-ly y'rs  - tim




From cce at clarkevans.com  Wed Mar 21 04:23:29 2001
From: cce at clarkevans.com (Clark C. Evans)
Date: Wed, 21 Mar 2001 04:23:29 -0500 (EST)
Subject: Yet Another PEP:  Interface Adapter Mechanism __adapt__
In-Reply-To: <Pine.LNX.4.21.0103201837570.21343-100000@clarkevans.com>
References: <Pine.LNX.4.21.0103201837570.21343-100000@clarkevans.com>
Message-ID: <Pine.LNX.4.21.0103210418490.21377-100000@clarkevans.com>

Thank you all for your feedback and support with the first 
pass of this PEP.  Below is a second attempt.  The context and
motivation sections are identical.  Here are the major changes:

    * __query__ is renamed __adapt__  (adapter pattern)
    * __adapt__ now takes a class instance (instead of a DNS string)
    * __adapt__ now returns None if the lookup was unsuccessful
    * straw-man adapt function is detailed
    * the title of this PEP is updated respectively

Regards,

Clark Evans

PEP: XXX
Title: Interface Adapter Mechanism
Version: $Revision$
Author: Clark Evans
Python-Version: 2.2
Status: Draft
Type: Standards Track
Created: 21-Mar-2001
Updated: 22-Mar-2001

Summary

    This paper asserts that "interface typing" can be carved 
    into two separable concerns, that of (a) protocol, which is 
    all about behavior/expectations, and that of (b) signature 
    which is about method existence and argument type checking.

    This proposal puts forth a declarative method for interface 
    protocol discovery that could be orthogonal and complementary
    to a more signature based approach as being developed by the
    types special interest group.

    The proposal is modeled after the adapter pattern as described
    design patterns book by Gamma, Helm, Johnson and Vlissides.

Context

    Python is a very dynamic language with powerful introspection
    capabilities.  However, it has yet to formalize an interface
    or abstract type checking mechanism. A consensus for a user
    defined type system may be far off into the future.

    Currently, existence of particular methods, particularly those
    that are built-in such as __getitem__, is used as an indicator
    of support for a particular interface.  This method may work
    for interfaces blessed by GvR, such as the new enumerator interface
    being proposed and identified by a new built-in __iter__.

    However, this current method does not admit an infallible way
    to identify interfaces lacking a built-in method.

Motivation

    In the recent type special interest group discussion [1], there
    were two complementary quotes which motivated this proposal:

       "The deep(er) part is whether the object passed in thinks of
        itself as implementing the Foo interface. This means that
        its author has (presumably) spent at least a little time
        about the invariants that a Foo should obey."  GvR [2]

    and

       "There is no concept of asking an object which interface it
        implements. There is no "the" interface it implements. It's
        not even a set of interfaces, because the object doesn't 
        know them in advance. Interfaces can be defined after objects
        conforming to them are created." -- Marcin Kowalczyk [3]

    The first quote focuses on the intent of a class, including 
    not only the existence of particular methods, but more 
    importantly the call sequence, behavior, and other invariants.
    Where the second quote focuses on the type signature of the
    class.  These quotes motivate a distinction between interface
    as a "declarative, I am a such-and-such" construct, as opposed
    to a "descriptive, It looks like a such-and-such" mechanism.

    Furthermore, it is clear that both aspects of interface are
    important, if not completely orthogonal and complementary.
    For purposes of this proposal, the word "protocol" is aligned
    with Guido's deep issue, and "signature" with Marcin's vision
    of a type system.  In this way, we clearly demarcate the two
    attitudes of the word "interface" in a type system.

    Clearly, detailing an exhaustive method of interface 
    signatures is very difficult problem, especially for 
    a dynamic language like Python where the interface 
    signature for a class may change over time.  However, a 
    simple declarative mechanism to inquire which interface 
    protocols an object supports at a given time is relatively 
    straight forward and could bring immediate benefit.

Details

    A new built-in method, __adapt__ is proposed.  This method has 
    a single argument, an adapter identifier, and either returns 
    an object supporting the given protocol or it returns None.
    As of this PEP, an adapter identifier may only be a class object.
    Future PEPs may expand the scope to allow other types of objects
    and will detail how they are treated.
 
    Further. a built-in function, adapt is suggested.  This function
    takes two arguments, an object and an identifier.  Initially, this
    function calls the __adapt__ method of the object, passing in
    the identifier.  If the __adapt__ method provides an object, then
    this is returned.  Otherwise, if the __adapt__ method is not found,
    or if the __adapt__ method returns None, then the objects class
    ancestry is checked to see if the identifier is a member.  If so,
    then the object itself is returned.  Otherwise, if no other PEPs
    apply, then None is return value of the adapt function.  Following
    is a sample implementation of this built-in.

    def adapt(obj,ident, options =  None):
        global check_base
        def check_base(bas,cmp):
            if bas == cmp: return 1
            for base in bas.__bases__:
                if check_base(base,cmp): return 1
            return 0
        if hasattr(obj, '__adapt__'):
            retval = obj.__adapt__(ident)
        if retval == None:
            if check_base(obj.__class__,ident):
                retval = obj
        # options flag used to enable:
        #   - reverse lookup via ident (future PEP) goes here?
        #   - signature based lookup (future PEP) goes here?
        #   - automatic signature checking (future PEP) goes here?
        return retval

Example Usage

    >>> class KnightsWhoSayNi: pass

    >>> class EggsOnly:  # an unrelated class/interface
            def eggs(self,str): print "eggs!" + str

    >>> class HamOnly:  # used as an interface, no inhertance
            def ham(self,str): pass
            def _bugger(self): pass  # irritating a private member

    >>> class SpamOnly: # a base class, inheritance used
            def spam(self,str): print "spam!" + str

    >>> class EggsSpamAndHam (SpamOnly):
            def ham(self,str): print "ham!" + str
            def __adapt__(self,cls):
                if cls == HamOnly:
                    return self       # HamOnly implicit, no _bugger
                if cls == EggsOnly:
                    return EggsOnly() # Knows how to create the eggs!
                return None

    >>> import adapter.example
    >>> from adapter import adapt
    >>> x = adapter.example.EggsSpamAndHam()
    >>> adapt(x,adapter.example.SpamOnly).spam("Ni!")
    spam!Ni!
    >>> adapt(x,adapter.example.EggsOnly).eggs("Ni!")
    eggs!Ni!
    >>> adapt(x,adapter.example.HamOnly).ham("Ni!")
    ham!Ni!
    >>> adapt(x,adapter.example.EggsSpamAndHam).ham("Ni!")
    ham!Ni!
    >>> adapt(x,adapter.example.KnightsWhoSayNi).spam("Ni!")
    Traceback (innermost last):
      File "<interactive input>", line 1, in ?
        AttributeError: 'None' object has no attribute 'spam'

Relationships To Iterator Proposal:
 
    The iterator special interest group is proposing a new built-in
    called "__iter__", which could be replaced with __adapt__ if an
    an Interator class is introduced.  Following is an example.

    >>> from adapter import adapt
    >>> import adapter.example
    >>> x = adapter.example.IteratorTest(3)
    >>> iter = adapt(x,adapter.example.Iterator)
    >>> iter.next()
    1
    >>> iter.next()
    2
    >>> iter.next()
    Traceback (innermost last):
      File "<interactive input>", line 1, in ?
      File "c:\work\adapter\example.py", line 39, in next
        return Iterator.next(self)
      File "c:\work\adapter\example.py", line 24, in next
        raise IndexError
    IndexError:

Relationships To Microsofts Query Interface:

    Although this proposal may sounds similar to Microsofts 
    QueryInterface, it differs by a number of aspects.  First, 
    there is not a special "IUnknown" interface which can be used
    for object identity, although this could be proposed as one
    of those "special" blessed interface protocol identifiers.
    Second, with QueryInterface, once an object supports a particular
    interface it must always there after support this interface; 
    this proposal makes no such guarantee, although this may be 
    added at a later time. Third, implementations of Microsofts
    QueryInterface must support a kind of equivalence relation. 
    By reflexive they mean the querying an interface for itself 
    must always succeed.  By symmetrical they mean that if one 
    can successfully query an interface IA for a second interface 
    IB, then one must also be able to successfully query the 
    interface IB for IA.  And finally, by transitive they mean if 
    one can successfully query IA for IB and one can successfully
    query IB for IC, then one must be able to successfully query 
    IA for IC.  Ability to support this type of equivalence relation
    should be encouraged, but may not be possible.  Further research 
    on this topic (by someone familiar with Microsoft COM) would be
    helpful in further determining how compatible this proposal is.

Backwards Compatibility

    There should be no problem with backwards compatibility.  
    Indeed this proposal, save an built-in adapt() function, 
    could be tested without changes to the interpreter.

Future Compatibility

    It appears that this proposal could be implemented orthogonal
    to the protocol checking system being constructed [4] by Paul
    Prescod and company.  In particular, a new PEP could be added
    which allows for Interface objects to be used as the adapter
    identifier.  Then, the __check__ method could be used within
    the adapt() method.

    It is less clear to me how this proposal would work with the
    the more ambitious signature declaration and checking approach[5]
    taken by Michel Pelletier.  This requires further investigation.

Questions and Answers

    Q:  Why was the name changed from __query__ to __adapt__ ?  

    A:  It was clear that significant QueryInterface assumptions were
        being laid upon the proposal, when the intent was more of an 
        adapter.  Of course, if an object does not need to be adapted
        then it can be used directly and this is the basic premise.

    Q:  This is just a type-coercion proposal.

    A:  No. Certainly it could be used for type-coercion, such
        coercion would be explicit via __adapt__ or adapt function. 
        Of course, if this was used for iterator interface, then the
        for construct may do an implicit __adapt__(Iterator) but
        this would be an exception rather than the rule.

    Q:  Why did the author write this PEP?

    A:  He wanted a simple proposal that covered the "deep part" of
        interfaces without getting tied up in signature woes.  Also, it
        was clear that __iter__ proposal put forth is just an example
        of this type of interface.  Further, the author is doing XML 
        based client server work, and wants to write generic tree based
        algorithms that work on particular interfaces and would
        like these algorithms to be used by anyone willing to make
        an "adapter" having the interface required by the algorithm.

    Q:  Why not call this __queryinterface__ ?

    A:  Too close to Microsofts QueryInterface, especially given the
        semantic differences which may not be reconcilable.

    Q:  Is this in opposition to the type special interest group?

    A:  No.  It is meant as a simple, need based solution that could
        easily complement the efforts by that group.

    Q:  Why was the identifier changed from a string to a class?

    A:  This was done on Michel Pelletiers suggestion.  This mechanism
        appears to be much cleaner than the DNS string proposal, which 
        caused a few eyebrows to rise.  

    Q:  Why not allow any object to be an identifier, why just classes?

    A:  It would be hard to get forward compatibility with new behaviors
        described in other PEPs.

    Q:  It seems that a reverse lookup could be used, why not add this?

    A:  There are many other lookup and/or checking mechanisms that
        could be used here.  However, the goal of this PEP is to be 
        small and sweet ... having any more functionality would make
        it more objectionable to some people.  However, this proposal
        was designed in large part to be completely orthogonal to other
        methods, so these mechanisms can be added later if needed

Copyright

    This document has been placed in the public domain.


References and Footnotes

    [1] http://www.zope.org/Members/michel/types-sig/TreasureTrove
    [2] http://mail.python.org/pipermail/types-sig/2001-March/001105.html
    [3] http://mail.python.org/pipermail/types-sig/2001-March/001206.html
    [4] http://mail.python.org/pipermail/types-sig/2001-March/001223.html

-----------------------------------------------------
adapter/__init__.py
-----------------------------------------------------
def adapt(obj,cls, options =  None):
    global check_base
    def check_base(bas,cmp):
        if bas == cmp: return 1
        for base in bas.__bases__:
            if check_base(base,cmp): return 1
        return 0
    if hasattr(obj, '__adapt__'):
        retval = obj.__adapt__(cls)
    if retval == None:
        if check_base(obj.__class__,cls):
            retval = obj
    # options flag used to enable:
    #   - reverse lookup via cls (future PEP) goes here?
    #   - signature based lookup (future PEP) goes here?
    #   - automatic signature checking (future PEP) goes here?
    return retval

------------------------------------------------------
adapter/example.py
------------------------------------------------------

class EggsOnly:  # an unrelated class/interface
    def eggs(self,str): print "eggs!" + str

class HamOnly:  # used as an interface, no inhertance
    def ham(self,str): pass
    def _bugger(self): pass  # irritating a private member

class SpamOnly: # a base class, inheritance used
    def spam(self,str): print "spam!" + str

class Intermediate (SpamOnly): pass

class EggsSpamAndHam (Intermediate):
    def ham(self,str): print "ham!" + str
    def __adapt__(self,cls):
        if cls == HamOnly:
            return self       # implements HamOnly implicitly, no _bugger
        if cls == EggsOnly:
            return EggsOnly() # Knows how to create the eggs!
        return None

class Iterator:
    def next(self):
        raise IndexError

class IteratorTest:
    def __init__(self,max):
        self.max = max
    def __adapt__(self,cls):
        if cls == Iterator:
            class IteratorTestIterator(Iterator):
                def __init__(self,max):
                    self.max = max
                    self.count = 0
                def next(self):
                    self.count = self.count + 1
                    if self.count < self.max:
                      return self.count
                    return Iterator.next(self)
            return IteratorTestIterator(self.max)
        return None









From Eugene.Leitl at lrz.uni-muenchen.de  Wed Mar  7 08:47:49 2001
From: Eugene.Leitl at lrz.uni-muenchen.de (Eugene Leitl)
Date: Wed, 7 Mar 2001 14:47:49 +0100 (MET)
Subject: advice on infrastructure for porting an application
Message-ID: <Pine.GSO.4.03.10103071348140.14273-100000@sun1.lrz-muenchen.de>

I'm porting a legacy cheminformatics application (currently available in
two incarnations: a pure Windows and a cgi-bin (Perl and C) under
Solaris). It lets you paint a target molecule in an editor, and offers you
several synthesis paths to it (taken from a proprietary database
containing millions of reactions). User-driven iteration of above on
educts (synthetic precursors) results in a synthesis tree.

Here's a mockup of a session, in case you're interested:

	http://www.lrz.de/~ui22204/mockup.gif

The new architecture has the following features:

* crossplatform (at the very least *nix, Windows), GNU gcc as target
* should be able to run both stand-alone and over network
* should be driven via a browser

What I would like is to deliver a (single or several) binary with a
system-dependent installer script. It would be of course nice if it itself
would be written in Python, and frozen into standard C.

I've looked at CGIHTTPServer, which was trivially easy to freeze. The
performance and functionality is sufficient to be run over localhost or
network, as the query will be the bottleneck. I'm looking at
CGIHTTPServer.py right now, but don't quite understand how it handles
cgi-bin scripts which are written in Python. If I freeze the http server,
it already contains a complete Python interpreter, so presumably it would
be able to fork (on architectures which support that, that is) and execute
the script natively. But all it seems to do is a os.execve(scriptfile,
args, env). If I'm getting this right, has anybody a version which doesn't
involve the overhead of calling a large frozen binary (for obviously I
can't contaminate customer's machine with random Python installations).
Btw, I rather liked the discussion of how to package stuff you want to
ship to your customer found here http://www.mcmillan-inc.com/install1.html

The parts of the legacy app which written in C (which I would want later
to rewrite in Python) could be incorporated by extending Python, right?

I'll be probably needing a number of other applications, such as Gadfly
for the session list manager, which will probably reside in several
independent instances, each burdened with it's own Python interpreter and
whatever else freeze will see fit to include. This is not pretty, but I 
guess it will do as long the redundancy will remain tolerable. Plus,
calling a binary each time afresh gets rid of memory leaks, if any.

I'd welcome your comments to above thoughts. Yes, I have no practical
experience with developing nontrivial under Python.

TIA,

-- Eugene

______________________________________________________________
ICBMTO  : N48 10'07'' E011 33'53'' http://www.lrz.de/~ui22204 
57F9CFD3: ED90 0433 EB74 E4A9 537F CFF5 86E7 629B 57F9 CFD3




From tuttledon at hotmail.com  Sat Mar 10 12:46:54 2001
From: tuttledon at hotmail.com (Don Tuttle)
Date: Sat, 10 Mar 2001 17:46:54 GMT
Subject: tkinter, activepython, win32
References: <OHiq6.4422$tP3.71012@news1.rdc1.bc.home.com> <mailman.984202333.22527.python-list@python.org>
Message-ID: <iMtq6.63293$__6.12624413@typhoon.southeast.rr.com>

> Luke wrote:
> >
> > python newbie!
> > cannot get tkinter programs to run...
> >
> > did 'normal' install of activepython 2.0 build 203
> > installed Tcl/Tk 8.3 for Windows, Binary Distribution
>>
> "Paul Prescod"
> I would suggest you use the package manager to get tk and try again.
> Just start the package manager and type "install tk". Or else you could
> have selected "install tk" when you installed ActivePython.

Seem ActiveState's TK build is broken:

>>> from Tkinter import *
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
  File "d:\python20\lib\lib-tk\Tkinter.py", line 34, in ?
    import FixTk # Attempt to configure Tcl/Tk without requiring PATH
  File "d:\python20\lib\lib-tk\FixTk.py", line 1, in ?
    import sys, os, _tkinter
ImportError: dynamic module does not define init function (init_tkinter)

Don






From sknhom at flashmail.com  Thu Mar 22 19:40:59 2001
From: sknhom at flashmail.com (sknhom at flashmail.com)
Date: Fri, 23 Mar 2001 00:40:59 GMT
Subject: On your own?  2929
Message-ID: <vYwu6.53239$KL5.91263@NewsReader>

Please may I apologise for contacting you by such a direct method.

www.ice121.com is a completely new Dating/Contact site where you can 
mail,chat online or arrange to meet your perfect partner. You can even view members' photographs. Why not take a couple of minutes to check out www.ice121.com   
Join free for one year saving you $115/?79. This offer is for a limited period only. 
Come and find out why everyone is talking about www.ice121.com  


Kindest regards
  
Frank Kia

qgktgnxlsqwwscsjotiubhvdlctpldcuppwcdezkogeczkghonqeynftjzfxonomkrisjbobnphyoetjijytjpjhunbjpkubuetogieukkikrmvzox



From quinn at riyal.ugcs.caltech.edu  Sat Mar 31 13:13:30 2001
From: quinn at riyal.ugcs.caltech.edu (Quinn Dunkan)
Date: 31 Mar 2001 18:13:30 GMT
Subject: Newbie Dictionary Question
References: <jfmx6.7861$iU.1496746@news1.rdc1.md.home.com>
Message-ID: <slrn9cc7i9.9n3.quinn@riyal.ugcs.caltech.edu>

On Sat, 31 Mar 2001 14:57:19 GMT, Gary Walker <borealis3 at home.com> wrote:
>I *think* I want to do the following:
>
>I want to create a list (this I'm able to do)
>Then I want to dynamically create a dictionary for each item in the list,
>add some keys (and values), and then put the dictionary in the list.
>
>(This part is frustrating me, because my newbie python code is referencing
>the SAME dictionary for each of the list indexes, this NOT what I want. I
>want a new instance of a dictionary for each pass thru the list as I loop)

list = []
for elt in thingsies:
    list.append({'key': elt})

>I want to do this because this way I can have an ordered list of key/value
>pairs... at least that was my plan...

Well, then a dictionary isn't going to help you, it's not ordered.  You might
want a list of pairs:

a = [
    ('foo', 'a'),
    ('bar', 'q'),
    ('baz', 'j'),
    ('faz', 't'),
]

Or, if you mean 'sorted' when you say 'ordered', you could sort a dictionary's
keys and iterate over that:

d = { ... }
ks = d.keys()
ks.sort()
for k in ks:
    print d[k]

>In other words:
>How do I dynamically create a dictionary whose name is not known until
>runtime?

Use braces:

{'foo': bar}

Nitpick: dictionaries don't have names.  No objects have names.  Dictionaries
might be bound to one, two, three, a million, or zero names, but they
themselves don't have names.  So "{'foo': bar}" creates a dictionary without a
name.  "d = {'foo': bar}" creates a dictionary without a name *and* a name
in the current environment (global or local) that refers to that dictionary.
"def f(x): ..." creates a function without a name and a name in the current
environment that refers to that function.  "a = b = {}" creates a function
without a name, and two names in the current environment that refer to it.
"a[0] = {}" creates a dictionary without a name and asks 'a' to store a
reference to it in its 0th element.

>Actually, they don't have to be ordered, so I suppose (thinking aloud here)
>that I could use a dictionary of dictionaries, but again - it seems to me
>that I'd need to dynamically create a dictionary whose name isn't known
>until runtime...
>
>Or am I missing the obvious? This is the most likely scenario :)

You should post some example code.  I'm not sure I know what you're talking
about.


From gradha at iname.com  Fri Mar  9 18:20:06 2001
From: gradha at iname.com (gradha at iname.com)
Date: Sat, 10 Mar 2001 00:20:06 +0100
Subject: Any documentation about Python daemons?
References: <e0gq79.8og.ln@127.0.0.1>
Message-ID: <6fob89.68d.ln@127.0.0.1>

Thanks for the pointers, they will keep me busy for enough time ;-)

-- 

 Grzegorz Adam Hankiewicz   gradha at iname.com - http://gradha.infierno.org
 Other web pages:  http://glub.ehu.es/  -  http://welcome.to/gogosoftware/


From Laura_01 at MailAndNews.com  Sat Mar 10 02:10:57 2001
From: Laura_01 at MailAndNews.com (Laura Lewin)
Date: Sat, 10 Mar 2001 02:10:57 -0500
Subject: Python In A Nutshell
Message-ID: <3AB1700C@MailAndNews.com>

Hi,
We all know Alex--hi Alex!--uses lots and lots of words to say things. 
<grin>
But he certainly knows how to make a point....(wait and see!)  It will be a 
nutshell not a treatise.  No schedule yet.
Laura
LLewin at oreilly.com

>===== Original Message From "Don Tuttle" <tuttledon at hotmail.com> =====
>> [Don Tuttle]
>> >Planning stages?  We've all seen his posts (a.k.a. treatises) here.  I'd
>say
>> >he should have the final draft to you in 3 weeks tops. ;-)
>>
>>[Aahz ]
>> ITYM "first draft".  It will take another three-to-six months for Alex
>> to edit it down.  ;-)
>
>Good point Aahz. And of course another six months for O'reilly to cut it
>down to a mere 1,200 pages!  All in all, it MIGHT be out just in time for
>Christmas 2002  ;-)
>
>Don
>
>



From fredrik at pythonware.com  Fri Mar 23 02:01:30 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 23 Mar 2001 07:01:30 GMT
Subject: Scrolling tkinter Canvas
References: <B6DFE101.52A%daniel769@earthlink.net>
Message-ID: <exCu6.3135$4N4.461044@newsc.telia.net>

Daniel Chamberlin wrote:
> I am unable to scroll a canvas.  could someone look at the
> code squib below and point out what needs to be changed?

to enable scrolling, you need to set the "scrollregion" option.

    canvas.config(scrollregion="0 0 100 100")
    canvas.config(scrollregion=canvas.bbox(ALL))

http://www.pythonware.com/library/tkinter/introduction/x1643-concepts.htm
=> concepts => coordinate systems

Cheers /F




From dsh8290 at rit.edu  Thu Mar 29 13:07:07 2001
From: dsh8290 at rit.edu (D-Man)
Date: Thu, 29 Mar 2001 13:07:07 -0500
Subject: Python GUI with Swing
In-Reply-To: <001e01c0b812$43f00760$340210ac@winson>; from winson@mdsserv.mds.com.tw on Thu, Mar 29, 2001 at 01:36:50PM +0800
References: <001e01c0b812$43f00760$340210ac@winson>
Message-ID: <20010329130707.A12014@harmony.cs.rit.edu>

On Thu, Mar 29, 2001 at 01:36:50PM +0800, ?i???q wrote:
| 
|    Is there any good solution for using Java Swing in Python ?? I

Jython would be it.  It is as simple as

import javax.swing
f = javax.swing.JFrame()
f.getContentPane().add( javax.swing.JLabel( "Hello World" ) )
f.setVisible( 1 )


|    surveyed Jython, PyJava, JPE and JPI. Jython sounds great but I only
|    needs Swing as my GUI and the whole Jython is based on Java, speed
|    will be a issue. Actually, I would like to find out a module can
|    access Java's Swing package.

If speed is an issue, don't use Swing.  Swing is quite slow (from
experience) as it is implemented entirely in Java.  If you only need
to support a single platform you can use Jython+Swing+JNI+C using JNI
to connect Java to the speed critical stuff written in C.  Otherwise I
would recommend looking at other toolkits such as PyGTK and wxPython.

-D




From aahz at panix.com  Tue Mar 27 11:32:15 2001
From: aahz at panix.com (Aahz Maruch)
Date: 27 Mar 2001 08:32:15 -0800
Subject: Eat this one
References: <3AC0B349.EE0AA9B2@hack.org>
Message-ID: <99qfaf$lkg$1@panix6.panix.com>

In article <3AC0B349.EE0AA9B2 at hack.org>,
Rikard Bosnjakovic  <bos at hack.org> wrote:
>
>What way would be better and _faster_ to wipe out the five chars from
>the list?

string.translate()
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

"Boost the stock market -- fire someone"


From Simon.Budig at unix-ag.org  Thu Mar  8 14:34:40 2001
From: Simon.Budig at unix-ag.org (Simon Budig)
Date: 8 Mar 2001 20:34:40 +0100
Subject: Apache and os.fork() in CGI-Binaries
Message-ID: <3aa7ded0@si-nic.hrz.uni-siegen.de>

Hi all.

I am currently developing a CGI-Binary which starts a second
(long running) process. The problem is: Apache keeps the
connection open until the fork()'ed child exits.

Discussing this thing with friends we came to the conclusion
that some open filedescriptors keep the connection between the forked
child and the apache alive:

1) Apache gets a request
2) Apache starts CGI-Script (Python). This is the parent process
3) Parent process forks and creates the child
4) Parent process exits
5) Child process exits
6) Apache closes the connection to the browser.

But I need the closed connection after step 4. Since the Child
inherits all filedescriptors, this is probably the thing that 
tells apache to keep the connection open.

I tried various things:

sys.stdout.close()
sys.stdin.close()
sys.stderr.close()

then I added

sys.__stdout__.close()
sys.__stdin__.close()
sys.__stderr__.close()

and this did not help. Finally i tried:

os.close (sys.stdout.fileno ())
os.close (sys.stdin.fileno())
os.close (sys.stderr.fileno ())

But at that time the Child does not run properly to the end.
Of course I do something a la

logfile = open (mebong.basedir + "/logfile", "a")
sys.stdout = sys.stderr = logfile

immediately afterwards, but this does not help.

So - how can I release all filedescriptors inherited from the parent?

The relevant parts from my code are:

-------------------------------
# this is from the CGI-Binary
childpid = os.fork ()
if not childpid:
   os.execv ("/path/to/jobcontrol", ["jobcontrol", arg1, arg2])
-------------------------------

-------------------------------
# this is from jobcontrol
os.close (sys.stdin.fileno ())
os.close (sys.stdout.fileno ())
os.close (sys.stderr.fileno ())

logfile = open ("/path/to/logfile", "a")
sys.stdout = sys.stderr = logfile
mypid = os.getpid ()
print "%d started: %s" % (mypid, repr(sys.argv))
-------------------------------

Thanks for hints!

Bye,
        Simon

-- 
      Simon.Budig at unix-ag.org       http://www.home.unix-ag.org/simon/


From stadt at cs.utwente.nl  Sat Mar  3 20:54:46 2001
From: stadt at cs.utwente.nl (Richard van de Stadt)
Date: Sun, 04 Mar 2001 02:54:46 +0100
Subject: CGI & Python (error)
References: <3A9F5FF5.34E01148@usa.net> <pkNn6.4257$c7.1344814@news3.rdc1.on.home.com>
Message-ID: <3AA1A066.D4F67A16@cs.utwente.nl>

Colin Meeks wrote:
> 
> > When accessing  the  script I get the error:
> >
> > File "getdate2.py", line 12, in ?
> >     day=form["day"].value
> >   File "/usr/lib/python1.5/cgi.py", line 907, in __getitem__
> >     raise KeyError, key
> > KeyError: day
> >
> > The script works from the command line.
> >
> > My problems is:  the script I am trying to run is based on another one.
> > This other script works. I can't  figure out the what is causing the
> problem
> > as the 2 scripts are identical  in every aspect execpt for 1 variable.
> Both
> > scripts are run in a page which has all the required parameters passed to
> it.
> > Help. Please...
> If this is what I think it is (sorry it's not immediately clear), but when I
> try
> to parse anything from CGI, if nothing is entered in one of the boxes, this
> seems to be especially for radio buttons, then the variable isn't
> initialised.
> 
> The thing I always do is something like
> 
> try:
>     day=form["day"].value
> except:
>     day=1 #or what ever you want the day variable initialised as
> 
> I hope this is of some help.

You can also use this:

if form.has_key("day") and form["day"].value) <> "":
	has_day_action(form["day"].value)
else:
	doesnot_have_day_action()

Note that neither of these options work in case the webform field
is a checkbox list. In such cases you shouldn't use ".value" directly,
but initially only the field name, like this:

if aForm.has_key("checkbox_list") and aForm["checkbox_list"] <> "":
	has_checkbox_list_action ()

To get the values of the selected items of a checkbox list (after you
tested if the fields exist like state above), do this:

checkbox_items =  aForm["checkbox_list"]
if type(checkbox_items) <> type ([]):
	checkbox_items = [checkbox_items]

After this, you will have the item(s) of the checkbox list in an array.
To get the values of the items, walk through the array, and *then* use
".value", e.g.:

checkbox_item_values = []
for item in checkbox_items:
	checkbox_item_values.append(item.value)

Cheers,

Richard.


From bedge at troikanetworks.com  Wed Mar 14 13:20:19 2001
From: bedge at troikanetworks.com (Bruce Edge)
Date: Wed, 14 Mar 2001 10:20:19 -0800
Subject: Dynamic construction of a print format string
Message-ID: <3AAFB663.57FCCE7@troikanetworks.com>

I'm using:

  fmtstr = "%s%s%s" % ( "    %", namelen, "s - %s" )

to get:

  "    %19s - %s" 

Is there a better way?

TIA, Bruce.


From debl at nonospammytheworld.com  Sat Mar 31 21:32:34 2001
From: debl at nonospammytheworld.com (David Lees)
Date: Sun, 01 Apr 2001 02:32:34 GMT
Subject: PythonLabs Team Moves to ActiveState
References: <mailman.986086501.5576.python-list@python.org> <3AC68B5A.E2A41D0E@NoNoSpammytheworld.com>
Message-ID: <3AC69350.FC1899BE@nonospammytheworld.com>

OK, in rereading the original post by Guido, I am pretty sure I just
responded to an April Fools joke.

Good one.

David Lees


David Lees wrote:
> 
> Could someone explain exactly what this means?  I read over PEP 228 and
> am
> wondering if this is what you are referring to below with respect to the
> 'Perl numerical model'?
> 
> I assume Python will still keep long integers, integers and complex.
> All that will change is coersion to float on the division of integers?
> 
> Would it be possible to clarify this change?
> 
> Thanks,
> 
> David Lees
> 
>    "...Also, after some talks with Larry, I've decided to adopt the Perl
>    numerical model for Python. The Perl model is simple, predictable and
>    easy to explain. This is a small enough change that it does not need
>    a PEP, and Moshe will implement it before 2.1rc1 is out."


From jafo at tummy.com  Thu Mar 15 01:35:35 2001
From: jafo at tummy.com (Sean Reifschneider)
Date: Wed, 14 Mar 2001 23:35:35 -0700
Subject: news from python9!!!
In-Reply-To: <uJRp6.46512$__6.10501931@typhoon.southeast.rr.com>; from tuttledon@hotmail.com on Thu, Mar 08, 2001 at 08:13:14PM +0000
References: <mailman.984074414.26983.python-list@python.org> <uJRp6.46512$__6.10501931@typhoon.southeast.rr.com>
Message-ID: <20010314233535.O30951@tummy.com>

On Thu, Mar 08, 2001 at 08:13:14PM +0000, Don Tuttle wrote:
>Hey, you gotta run with what you can find, seeing how all you conventioneers
>where having way too much fun to drop into the newsgroup and give us an
>update! <wink>

That was just because they didn't have *REAL* wireless networking at the
show...  ;-)  For my release of the 2.1b1 RPMs, I built them in the bar
while hanging out, then went to the terminal room to schlep them up to
the server and send out the announcement...

Sean
-- 
 I can't understand why people are frightened of new ideas.  I'm frightened
 of the old ones.  -- John Cage
Sean Reifschneider, Inimitably Superfluous <jafo at tummy.com>
tummy.com - Linux Consulting since 1995. Qmail, KRUD, Firewalls, Python



From Eugene.Leitl at lrz.uni-muenchen.de  Wed Mar  7 06:42:02 2001
From: Eugene.Leitl at lrz.uni-muenchen.de (Eugene Leitl)
Date: Wed, 7 Mar 2001 12:42:02 +0100 (MET)
Subject: GadFly and gfplus.py working with Python 2.0?
In-Reply-To: <Pine.GSO.4.03.10103071148001.13488-100000@sun1.lrz-muenchen.de>
Message-ID: <Pine.GSO.4.03.10103071230540.13488-100000@sun1.lrz-muenchen.de>

On Wed, 7 Mar 2001, Eugene Leitl wrote:

> On Wed, 7 Mar 2001, Oleg Broytmann wrote:
> 
> >    Why not just try yourself? (Works for me, YMMV)
> 
> GadFly seems to be working, but gfplus.py doesn't seem
> to accept simple valid SQL. 

Somebody whack me with a cluestick: of course it works. Comes
from asking Oracleites about pidgin SQL: they wouldn't know it
if it bit them in the behind.

I've just made a session with a frozen gfplus.py:

[eleitl at pc27 test2]$ ./gfplus
 
gfplus 1.1 -- Interactive gadfly shell
 
Wednesday March 07, 2001 12:24 PM
 
Using:
DB Name: foobar
DB Location: /tmp/ooga
Unable to locate database "foobar" at location "/tmp/ooga".
Create? (Yy/Nn) Y
GF> desc user_tables;
 
COLUMN_NAME
===========
 
GF> create table test (nr integer, name varchar(5));
GF> insert into test values (1, 'blah');
GF> insert into test values (2, 'ookook');
GF>  select * from test;
 
NR | NAME
===========
1  | blah
2  | ookook
 
GF>   

So it does work. Whew. I've also tested the client/server mode
with ./gfplus admin 2222 admin localhost 
and it also seems to behave sanely sofar. Double whew.




From qrczak at knm.org.pl  Sat Mar 31 16:48:07 2001
From: qrczak at knm.org.pl (Marcin 'Qrczak' Kowalczyk)
Date: 31 Mar 2001 21:48:07 GMT
Subject: list vs tuple
References: <mailman.986066384.30712.python-list@python.org> <d1rx6.1053$p5.4024@news1.rivrw1.nsw.optushome.com.au> <slrn9cciaj.3ug.qrczak@qrnik.zagroda> <lWrx6.1057$p5.4152@news1.rivrw1.nsw.optushome.com.au>
Message-ID: <slrn9cck4n.49p.qrczak@qrnik.zagroda>

Sat, 31 Mar 2001 21:24:33 GMT, deadmeat <root@[127.0.0.1]> pisze:

> >>> a = 1
> >>> b = a
> >>> a = 2
> >>> b
> 1
> 
> >>> a = [1,2,3]
> >>> b = a
> >>> a[0] = 0
> >>> b
> [0, 2, 3]
> 
> Are you saying there is no difference (IN THE *RESULT*) between these?

Of course there is, because you assigned to 'a' in the first case
and to 'a[0]' in the second. How could the result be the same if
you execute different statements?

-- 
 __("<  Marcin Kowalczyk * qrczak at knm.org.pl http://qrczak.ids.net.pl/
 \__/
  ^^                      SYGNATURA ZAST?PCZA
QRCZAK


From aleaxit at yahoo.com  Sun Mar 18 15:55:59 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sun, 18 Mar 2001 21:55:59 +0100
Subject: Who's minister of propaganda this week?
References: <B98041329020E007.7DA84CAD9D0562AB.817BBC572F480869@lp.airnews.net> <98nu1m$12e@dispatch.concentric.net> <660ACCA1E4816F7E.C359CDF4168EF3FB.3B6B05538D619770@lp.airnews.net> <3AAFA13D.76E32C3F@divalsim.it> <98oa4o01i4v@news2.newsguy.com> <mailman.984619039.7856.python-list@python.org> <98q9h60122v@news2.newsguy.com> <mailman.984937946.5318.python-list@python.org>
Message-ID: <9937en0bge@news2.newsguy.com>

"Michael Chermside" <mcherm_python at yahoo.com> wrote in message
news:mailman.984937946.5318.python-list at python.org...
    [snip]
> First of all, I completely agree that telling whether an object has
> the right method names and signatures (an "interface" in the java
> sense) is only a very limited subset of defining the signatures,
> pre-requisites, post-conditions, and invariants. Eiffel allows
> something like this... let's call the idea a "contract". I
> think it would GREAT if we could specify our strict typing in terms
> of "contracts" instead of the more limited "interfaces".

I would frame it in a slightly different way: that each interface
ideally needs to be 'completed' by including the contracts on
its methods (and abstract-state invariants).

> Compilers could NOT be expected to verify at compiletime that the
> pre-requisites were being satisfied, but they could at LEAST
> verify that the object being used claimed to implement that
> contract.

"Explicit is better than implicit" supports this.  Right now, in
Python, we only have _implicit_ interfaces -- and we pay some
price for that (for example, when we mention a 'sequence
interface', it's often not 100% clear if we only need suitable
behavior of the __getitem__ method, or also a __length__,
etc, etc).

However, the language constraints needed to make even this
'assertion to implement' verifiable *at compile time* are still
too strong for what little further advantage you gain.  The
documentation advantages of contract-ish interfaces are not
dependent on their compile-time vs run-time behavior, and
decent unit-tests will ensure the run-time checks are in fact
exercised (with clearer diagnostics than currently one can get
from typical contract-violation errors).

The extra advantage in terms of programmer productivity of
getting the diagnostics earlier may be, say, 5% or so -- with
a cost on the same scale of, say, 10% or so, for the extra
constraints of static typing.  Which is where we come in:-).

In other words: the undoubted advantages of interfaces with
programming-by-contract annotations do not really militate
in favour of statically typed languages vs dynamically typed
ones.


> On the occasions when I want to write something which works on
> all objects of a certain behavior, I define an interface for
> the behavior (even in languages which don't have direct support
> for "interface") and specify that all appropriate objects
> implement the interface. In fact, I often write multi-paragraph
> comments on how the interface should behave, so that it's really
> more of a "contract" than an interface. When I'm being good,
> (and I'm good much more often lately) I even write unit tests to
> enforce it.

Excellent (and static typing does not really help you check
your comments:-).  But, in large-scale software development,
you do not fully control this; there will be existing components,
that might well work just fine with other components... except
that, being developed separately and independently (and/or
not tolerating mutual dependencies) will not be inheriting from
each other's interfaces.  I'm not saying this _can't be solved_ --
I _am_ saying that solves it takes work _and_ complication
(templates, multiple inheritance, delegation, wrapping) and a
10% productivity impact is a _conservative_ evaluation of it.

Further -- your 'design space' itself (within a component, or
between components you control) gets constrained by the
language and object-model limitations coming from the static
typing characteristics thereof.  You may well be even considering
only about half of the architectures -- and there is a chance
that the optimal architecture may be among those that static
typing doesn't even let you _think_ of.

For example, say that you want to organize certain objects
in a container/contained hierarchy, with no nesting limits.

All existing objects will be leaves in the resulting tree.  Why
does this imply that such existing objects need to inherit
from a certain interface asserting "I may or may not be a
container" (perhaps with a default implementation stating
"I am _not_ a container after all")?  This is invasive, and
forces a logically uneeded dependency on the existing objects;
they shouldn't have to know about these organizing
superstructures at all!  In a dynamically typed language,
you will define an interface (and perhaps a default
implementation thereof) *only for container objects*, i.e.,
the "new" organizing superstructure.  Each container will
hold a set (e.g. a list, or dict) of (suitably tagged) references
to other objects; for recursive walks on the tree, one will
dynamically test whether a given object is-a container (else,
it's a leaf).  It's as simple as this; no headaches.  Any
headaches that DO come with this useful pattern are all
costs of using it with stating typing...!-)

If one ever considered only "green-field" development,
where no "existing objects" may possibly exist, maybe
the costs would be lower (there would still be some, as
the might-want-to-be-contained objects would still
have to multiply inherit the appropriate interface for
the _sole_ purpose of satisfying the static checks, a
pure overhead cost in terms of complication).  But this
is not a realistic hypothesis anyway -- standard and
other useful libraries have just too much useful stuff
to make it sensible to eschew them for the sake of
design purity:-).


> ... I'm just learning this stuff. In fact, I have been working
> with dynamically typed languages only for a very short time. So
> my own take-away is that I should really try to exercise the
> ability to make more extravagant use of polymorphism when working
> in dynamically typed languages, because I may discover (if my
> experience is similar to yours) that I'm saving 10% of my time,
> by avoiding awkward contortions that I don't even realize I'm
> doing now.

Good point.  I may keep talking for ages, but nothing I say will
substitute for actually getting some hands-on experience!-)


Alex





From robin at jessikat.fsnet.co.uk  Fri Mar  9 19:50:21 2001
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Sat, 10 Mar 2001 00:50:21 +0000
Subject: gah! I hate the new string syntax
References: <mailman.983615170.23313.python-list@python.org>
 <tN5o6.2746$eH3.14913@NewsReader>
 <Koco6.7518$1D5.244602@e420r-atl1.usenetserver.com>
 <mailman.983997252.5628.python-list@python.org>
 <m2vgpkyfjo.fsf@mycroft.actrix.gen.nz>
 <mailman.984063199.18779.python-list@python.org>
 <98an7i01tgp@news2.newsguy.com>
 <e7eq6.47946$ML1.2142738@e420r-atl2.usenetserver.com>
Message-ID: <3QQBdWANpXq6EwX6@jessikat.fsnet.co.uk>

In article <e7eq6.47946$ML1.2142738 at e420r-atl2.usenetserver.com>, Steve
Holden <sholden at holdenweb.com> writes
>The topic drift on this thread could only be explained by the Gondwanaland
>theory!
>
>property-is-theft-ly y'rs  - steve
nobody has mentioned Hitler yet. Doh! :(
-- 
Robin Becker


From mcherm_python at yahoo.com  Wed Mar 14 20:14:21 2001
From: mcherm_python at yahoo.com (Michael Chermside)
Date: Wed, 14 Mar 2001 19:14:21 -0600
Subject: Who's minister of propaganda this week?
References: <B98041329020E007.7DA84CAD9D0562AB.817BBC572F480869@lp.airnews.net> <98nu1m$12e@dispatch.concentric.net> <660ACCA1E4816F7E.C359CDF4168EF3FB.3B6B05538D619770@lp.airnews.net> <3AAFA13D.76E32C3F@divalsim.it> <98oa4o01i4v@news2.newsguy.com>
Message-ID: <3AB0176D.8070903@yahoo.com>

Alex Martelli wrote:
         ... [snip]...

  >
  > So, all the compile-time checking is buying is catching (a small
  > subset of) the errors that would be caught in testing anyway, a
  > little bit earlier (thus, a little bit cheaper) -- it's never the
  > case that one has to write a test which would not be needed at
  > all if type-checking was static, since "the object has the right
  > type" is a small subcase of "the object _behaves_ per the specs".
  >
I'm really not sure I see it this way. If the method  foo(x) is known to
take a FancyDateObject
there are three kinds of errors we could make. One is that foo() is
written badly so it
doesn't do what it's supposed to. The unit tests of foo() need to guard
against this.
Another possible error is that FancyDateObject isn't written properly.
The unit tests
of FancyDateObject need to guard against this. And the third type of
error is that
that somewhere where we CALL foo(), we might pass it a DateObject
instead... or
even a String... which will cause it to perform wrong. To guard against
this in a
dynamically typed language, we have to write unit tests for *every
single place*
that we call foo(). Of course, we'd be writing unit tests for those
functions anyway,
but we won't be able to assume that foo() works properly, and will need
extra
tests to ensure this.

In a staticly typed language, we still need the unit tests for cases 1
and 3, but the
third type of error is caught by the compiler. And in my mind, ANY TIME that
I can have a machine do my work for me it's better... I can be lazy, and
the machine
never gets tired after a long day and forgets to test sometimes. Of
course, I can
put an assert at the top of foo() which asserts that x is of type
FancyDateObject,
but if I always assert the types of my arguments then I'm basically
using a staticly
typed language.

There are times when a dynamically typed language is more flexible, and
it's certainly
nice not to have to declare everything just to specify its type, but
there ARE
advantages to static typing, and this, I believe, is the biggest one.

-- Michael Chermside




_________________________________________________________
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com




From aahz at panix.com  Fri Mar 16 11:57:33 2001
From: aahz at panix.com (Aahz Maruch)
Date: 16 Mar 2001 08:57:33 -0800
Subject: list.sort(cmpfunc) question
References: <m1ae6lg5qt.fsf@pclabs188.labs.trema.com> <98t9dq$qvu$1@panix6.panix.com> <98tf9d$g3v$1@news.mathworks.com>
Message-ID: <98tglt$bk3$1@panix6.panix.com>

In article <98tf9d$g3v$1 at news.mathworks.com>,
Joshua Marshall  <jmarshal at mathworks.com> wrote:
>
>Seems the word "should" should never appear in documentation.

Go argue that one in news.groups and news.admin.misc.  ;-)




(There's a perpetual floating flamewar over Son-of-1036 (the netnews
RFC) about whether munged e-mail addresses are permitted.)
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

Three sins: BJ, B&J, B&J


From jwbnews at scandaroon.com  Tue Mar  6 12:39:14 2001
From: jwbnews at scandaroon.com (John W. Baxter)
Date: Tue, 06 Mar 2001 09:39:14 -0800
Subject: I come to praise .join, not to bury it...
References: <mailman.983890030.6303.python-list@python.org> <983894109.2077137603@news.silcom.com>
Message-ID: <jwbnews-AA1BA7.09391406032001@corp.supernews.com>

With luck and good will, we ought not have so many of this sort of 
discussion after the fact in the future.

We now have the PEP process...and the discussion of the PEP is the right 
place for alarms to be raised, both technical ("that fails when...") and 
otherwise ("I'd rather see this than that.").

Previously, much of the community* didn't see the developer discussions, 
so the changes tended to spring upon us already decided and implemented.
[* I'm overweight enough that even if I'm alone in that, it's still 
"much of the community".]

We'll see in the next couple of rounds (post 2.1) how well the above 
works out it practice.  Even in Python, the later in the product cycle a 
change is made, the more it costs.

   --John

-- 
John W. Baxter   Port Ludlow, WA USA  jwbnews at scandaroon.com


From hinsen at cnrs-orleans.fr  Mon Mar 19 04:49:37 2001
From: hinsen at cnrs-orleans.fr (Konrad Hinsen)
Date: 19 Mar 2001 10:49:37 +0100
Subject: Extending using Numeric, a quest for patterns
References: <98u4q9$kip$1@slb6.atl.mindspring.net>
Message-ID: <m3pufegkb2.fsf@chinon.cnrs-orleans.fr>

"Greg Landrum" <gReGlAnDrUm at earthlink.net> writes:

> The core of the Python for the bit I'm interested in converting is:
>           posArray = array(pos1+pos2)
>           avgVect = sum(posArray)/len(posArray)
>           devs = posArray-avgVect
>           devs *= devs
>           e = sum(sum(devs))
> where pos1 and pos2 are python lists.

Keep the first line in Python, there is nothing to gain by converting
it to C. The rest then becomes a single C function which takes
avgVect as input and returns e. But don't expect miracles, unless
your array is rather short, the above Python code should be quite
efficient already.

> What I think would make a lot of this easier is if I could figure out how to
> use the ufuncs from C code.  I pored over the source and the docs, but I was

That's not worth the effort, and you'd build in unnecessary overhead.
Just access the array data directly and do your rather simple
calculations in straightforward C code.

Konrad.
-- 
-------------------------------------------------------------------------------
Konrad Hinsen                            | E-Mail: hinsen at cnrs-orleans.fr
Centre de Biophysique Moleculaire (CNRS) | Tel.: +33-2.38.25.56.24
Rue Charles Sadron                       | Fax:  +33-2.38.63.15.17
45071 Orleans Cedex 2                    | Deutsch/Esperanto/English/
France                                   | Nederlands/Francais
-------------------------------------------------------------------------------


From ransen_spam_me_not at nemo.it  Fri Mar  2 04:32:12 2001
From: ransen_spam_me_not at nemo.it (Owen F. Ransen)
Date: Fri, 02 Mar 2001 09:32:12 GMT
Subject: Bug in Py_Finalize() ?
Message-ID: <3a9f636d.2908314@news.newsguy.com>

Trying to find a bug in my C extension I ended up writing
this loop:

    for (UINT i = 0 ; i < 10 ; i++) {
        Py_Initialize();
        Py_Finalize() ;
    }

which crashes Python when i=8 with the message:
Windows 32, VC++ program.

Fatal Python error: UNREF invalid object

My understanding is that I can use Py_Initialize() and
Py_Finalize() in pairs to free up memory and to start
with a clean interpreter each time I need it. And that
is why I'd like to use Py_Finalise....




--
Owen F. Ransen
http://www.ransen.com/
Home of Gliftic & Repligator Image Generators


From carribeiro at yahoo.com  Thu Mar 15 08:48:06 2001
From: carribeiro at yahoo.com (Carlos Ribeiro)
Date: Thu, 15 Mar 2001 10:48:06 -0300
Subject: Thoughts about extensions to the Python language
In-Reply-To: <15023.62110.155494.764345@w221.z064000254.bwi-md.dsl.cnc.n
 et>
References: <5.0.2.1.0.20010314181331.0213eac0@mail.inet.com.br>
 <5.0.2.1.0.20010314181331.0213eac0@mail.inet.com.br>
Message-ID: <5.0.2.1.0.20010315094519.0216aec0@mail.inet.com.br>

At 17:37 14/03/01 -0500, jeremy at alum.mit.edu wrote:
>You seem to have a very abstract concern about the speed of
>develpoment and the fact that we write design documents that describe
>changes to the language.  I voted for "just right" in the poll, so we
>probably don't agree on the former.  As for the latter, what
>alternative is there?  Make changes to the language without
>describing the changes in advance? <wink>  I don't expect Guido's good
>judgement is impaired by having written proposals to consider.  He's
>been known to reject PEPs.

  Ok, my concern about written proposals was mostly a joke. And no, I dont 
have any alternative. However, from my practical experience, I *do* believe 
that is fairly easy to to be impressed by junk ideas just because they are 
well written and documented. You'd better keep your wits sharp.

Some of the proposal dont make much sense to me. Someone else (I'm afraid 
that it was Guido :-) has just pointed out that improving the library is 
not as much fun as trying to improve the language. Ok, so lets try to find 
a generic approach to make language *easily* extensible. I see some areas 
where this could be done.

1) This topic has been beaten to death, but anyway... the "import from 
future" problem should be solved by a more generic and natural approach. 
There is a possibility that some code may break, but this is a 
"transitional" effect, and nothing stops people to keep using older 
versions of Python in this case. On the other side, the extension of the 
meaning of the import directive would be definitive, and would affect all 
code written from now on. We could:

- why not a generic pragma or directive keyword? a lot of people suggested 
doing that, but I dont see any reason not to do it, except perhaps the 
possibility of breaking code.
- why not extend the Python compiler itself with Python? We already have 
the __main__ symbol for the top-level script, why not a __compiler__ symbol?

(please take this as suggestions only - I could be the one making junk 
proposals here :-)

2) One area that I would like to see more discussion is about the 
implementation of a generic notation for new fundamental types. Python does 
a nice job on fundamental types - we have strings, list, sequences, 
mappings. And the syntax is clean, which makes Python so easy to learn. The 
problem is that all the reasonable delimiters were spent in this proccess, 
and any proposal to extend or create new fundamental types has to use a 
kind of kludge. For instance this already happened with the Unicode string. 
Why not think of a generic approach for that?


Carlos Ribeiro
carribeiro at yahoo.com
cribeiro at mail.inet.com.br


_________________________________________________________
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com




From jarober at mail.com  Sun Mar 25 11:56:57 2001
From: jarober at mail.com (James A. Robertson)
Date: Sun, 25 Mar 2001 16:56:57 GMT
Subject: Collection interfaces
References: <3a90fa97.0@news1.mweb.co.za>   <MPG.1504d14563dce6499896af@news.earthlink.net> <97fjla128ir@news2.newsguy.com> <97h5s40vhc@news2.newsguy.com> <MPG.1506007591cf818b9896bc@news.earthlink.net> <slrn99opop.gr.amk@mira.erols.com> <MPG.1506265ebf6d934d9896be@news.earthlink.net> <97ibmc0k47@news1.newsguy.com> <MPG.1506cf3293b094b49896c1@news.earthlink.net> <97m0c7080c@news2.newsguy.com> <MPG.15085517debf4b649896ca@news.earthlink.net> <zb%n6.31033$o%6.586026@news1.rdc1.bc.home.com> <MPG.150b383cea226d409896d0@news.earthlink.net> <kNko6.6986$eq3.707121@news.uswest.net> <MPG.150c46db9cec283e9896d4@news.earthlink.net> <3AA2A898.450B659E@mail.com> <gRCt6.12224$Im6.1308560@newsread1.prod.itd.earthlink.net> <3AB755F8.67E84126@mail.com> <%a%t6.17245$227.1601569@newsread2.prod.itd.earthlink.net> <3AB8A730.4BE13D22@mail.com> <61bv6.8693$P4.676800@newsread1.prod.itd.earthlink.net> <B6E2B64C.D8E3%t*b*r*u*m*l*e*y@f*u*s*e.n*e*t> <o1jv6.9635$P4.789083@newsread1.prod.itd.earthlink.net>
Message-ID: <3ABF74C4.3EDB3C39@mail.com>

topmind wrote:
> 
> >
> >
> http://www.cincom.com:80/scripts/smalltalk.exe/gettingstarted/index.asp?cont
> > ent=trysmalltalknc
> >
> 
> It seems to be wacked out right now.

No, it was up last night, and is up now.  Are you sure you copied the
entire link?


> 
> > Pardon the word wrapping my news client does. There's no spaces in the
> > above.
> >
> > That should get you to a place where you can download the lesson browser
> > (thanks again to Mark Windholtz, great job!) and the tutorial (thanks to
> Jim
> > F. as well).
> >
> 
> Lesson browser download?
> 
> The last thing I need is yet more foriegn software on my PC making it crash
> more and confusing other DLL's.
> 
> Please, HTML.
> 
> >...site runs on
> > Microsoft ASP VSscript instead of Smalltalk?
> >
> > Naw. Jim isn't afraid of anything. At least, nothing on Usenet!
> >
> > Me either :)
> >
> > The rest of the Cincom site, and the old Smalltalk portion of the site,
> does
> > indeed not run on Smalltalk. We'll get it all moved over eventually :)
> >
> > Heck, my own business unit, which is very into VisualWave, understands
> that
> > we have to use JavaScript on our pages to drive the browsers. There's
> > nothing wrong with being multi-lingual.
> >
> > >
> > > Do Chevy salesmen drive Fords?
> >
> > Some do. I work for Cincom and use both our Smalltalk products, I actually
> > prefer ObjectStudio when I'm doing Windows specific work, but I also have
> > Squeak and Dolphin around and have experimented with both.
> >
> > > -tmind-
> >
> >

-- 
James A. Robertson
Product Manager (Smalltalk), Cincom
jarober at mail.com
<Talk Small and Carry a Big Class Library>


From tim.one at home.com  Fri Mar 16 19:14:54 2001
From: tim.one at home.com (Tim Peters)
Date: Fri, 16 Mar 2001 19:14:54 -0500
Subject: Pep 238 Why new // operator (Non-integer Division)
In-Reply-To: <98tc2s$qir$1@news2.ottawa.cyberus.ca>
Message-ID: <LNBBLJKPBEHFEDALKOLCMEDGJGAA.tim.one@home.com>

[Gordon Williams]
> Why do we need a new // operator to do the divide operation.  If
> we want to do an integer divide with two reals why wont people just
> use int(a/b).  For

1. int() truncates.  Current i/j returns the floor.

2. There is no bound on the sizes of i and j; it's very easy to
   get Python long ints that don't fit in a Python float.




From jcollins at endeavors.com  Fri Mar  9 14:33:48 2001
From: jcollins at endeavors.com (Jeff Collins)
Date: Fri, 9 Mar 2001 12:33:48 -0700
Subject: Python & Wireless Apps
In-Reply-To: <98b4ra$107$1@bashir.peak.org>; from louis@decipherinc.com on Fri, Mar 09, 2001 at 09:46:36AM -0800
References: <98b4ra$107$1@bashir.peak.org>
Message-ID: <20010309123348.D1139@endeavors.com>

We just released Pippy, a port of Python to the Palm.  Please see
http://endeavors.com/pippy for more information.

On Fri, Mar 09, 2001 at 09:46:36AM -0800, prince.kattare.com wrote:
> Hello!   I just got back from the Python Conference and glad to see that
> there is a lot of motivation for the language.  I am just starting out with
> python and was wondering how it's being used in the wireless space, meaning
> mostly palm devices.  Has anyone ported an app to a palm pilot?  What kind
> of restrictions should I be considering when thinking about this?
> 
> Thanks for any insight you might have!
> 
> Louis
> louis at decipherinc.com

-- 
Jeffery D. Collins, Ph.D.
Sr. Software Developer
Endeavors Technology, Inc.
http://www.endeavors.com



From debl at NoNospammytheworld.com  Fri Mar 23 20:41:28 2001
From: debl at NoNospammytheworld.com (David Lees)
Date: Sat, 24 Mar 2001 01:41:28 GMT
Subject: Good GUI for Python
References: <mailman.985314033.5120.python-list@python.org> <AADu6.436656$f36.14223632@news20.bellglobal.com>
Message-ID: <3ABBFB52.1FB59ABA@NoNospammytheworld.com>

What are the licensing issues referred to below on pyQT?

david lees


Bryan Mongeau wrote:
<snip>
> 
> There is also pyQT available from thekompany.com but the legality of the
> windows licensing remains a thornbush.
> 
<snip>


From aleaxit at yahoo.com  Fri Mar 16 08:43:27 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 16 Mar 2001 14:43:27 +0100
Subject: gah! I hate the new string syntax
References: <mailman.983485923.1136.python-list@python.org> <97sudi01i26@news1.newsguy.com> <98elrk$li8$1@nntp.Stanford.EDU> <98fc070thh@news1.newsguy.com> <98iqo2$56o$1@nntp.Stanford.EDU> <98iv3s01sc2@news2.newsguy.com> <slrn9avqlm.8ee.quinn@retch.ugcs.caltech.edu> <Pine.LNX.4.21.0103161350510.2809-100000@bast.jhb.equinox.co.za>
Message-ID: <98t5a20l6s@news1.newsguy.com>

"Pieter Nagel" <pieter at equinox.co.za> wrote in message
news:Pine.LNX.4.21.0103161350510.2809-100000 at bast.jhb.equinox.co.za...
> On Mon, 12 Mar 2001 17:56:33 +0100, Alex Martelli <aleaxit at yahoo.com>
wrote:
>
> > Focusing on the polymorphism needs, I see the issue in the mirror
> > way from you: it's a _weakness_ (of Python and most other languages)
> > that I can't add methods 'post-facto, from the outside' to existing
> > objects -- it impedes the most natural, elegant, typeswitch-free way
> > to add functionality.
>
> Not true. See my other message in this post for how to extend Python
> classes with new methods at runtime.

Ah, _classes_ (and, actually, also, if need be, _instances_), *NOT*
_all_ objects -- the type/class split again.  For example, you can't
add methods to 'all strings', as they belong to a type, not a class;
nor to compiled-regular-expressions, file-objects, complex-numbers,
dictionaries, tuples...


Alex





From owen at astrono.junkwashington.emu  Tue Mar 20 13:38:27 2001
From: owen at astrono.junkwashington.emu (Russell E. Owen)
Date: Tue, 20 Mar 2001 10:38:27 -0800
Subject: Functionalism, aesthetics Was:(RE: I come to praise .join, not
References: <mailman.985052844.17993.python-list@python.org>
Message-ID: <998835$vq6$1@nntp6.u.washington.edu>

I completely agree. There is some elitism from some of the cognoscenti. 
And the pace of change is fast enough now that I think some ugly stuff 
is getting into the language despite the protests.

Having said that, I hasten to add that I am not part of the Python 
development team, but rather one of unwashed and sometimes-squashed 
masses. I hope someday to make a contribution, but for now, I'm a user. 
Hence there's only so much complaining one can do productively. I'm glad 
there are folks who care enough and are willing to put the time into 
maintaining the language.

-- Russell

In article <mailman.985052844.17993.python-list at python.org>,
 Carlos Ribeiro <carribeiro at yahoo.com> wrote:

>I'm lost for words... This summarizes a lot of concerns that sometimes I 
>find so hard to explain. Anyone who has actually *read* Christopher 
>Alexander's books on design should know that the best solutions are the 
>ones that take into account the inner sense of satisfaction of their end 
>users. As Alexander puts it itself, it is a rewarding experience born out 
>of the "completeness" of the solution. The best solutions also share 
>another common trait: they look *and feel* good to the untrained eye. 
>Beware of works that can only be appreciated by experts - they are way too 
>trained to see the obvious.
...


From cdwom at mpinet.net  Thu Mar 29 17:55:55 2001
From: cdwom at mpinet.net (Corey Woodworth)
Date: Thu, 29 Mar 2001 22:55:55 GMT
Subject: Newbie needs a little help with Tkinter
Message-ID: <%3Pw6.108357$fC.12218204@bin1.nnrp.aus1.giganews.com>

I'm brand new here, brand new to python, and brand new to programing in
general (I bet you get these alot :) Anywho. I've learned enough to tackle
my first big project: A bigger badder Notepad. I figgured it'd be an easy
project to start with, but it could progress into something very complex, so
it'd be my guide to learning python. I've peeked at the IDLE source a bit,
and thats helped me alot, but there are still a few things I don't know how
to do (probably a lot more than a few):

1. How can add Copy Cut and Paste to an edit menu? I know how to make the
menu items, I just don't know how to assign them to Cut, Copy, and Paste
actions.

2. I know that you can askopenfilename and asksaveas to bring up premade
Open and Save As Dialogue boxes. Is there a similar one for print?

3. How can I add Keyboard shortcuts to menu items? Particuarly Cut, Copy,
Paste, Save, and Open if they are prebuilt, but I'd also like to know how to
generate my own (for ex. If I added custom menuitem and wanted a shortcut to
it)

4. Is there an easy way to implement an INS mode for test widgets?

ummm.. thats all I can think of for now, I'm sure there will be tons more.
This Python stuff is great though!

Thanks in advance for any help!!

So long, and thanks for all the fish!
Corey

Check out my Webpage at: schitzomedia.y0ru.net
and vote for the top 10 games of all time!!!





From mwh21 at cam.ac.uk  Sun Mar 18 10:26:09 2001
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 18 Mar 2001 15:26:09 +0000
Subject: Console column Width
References: <hhAr6.3889$L7.27822@news1.sshe1.sk.home.com> <m3snkg7n54.fsf@atrus.jesus.cam.ac.uk> <DITr6.4002$L7.30409@news1.sshe1.sk.home.com>
Message-ID: <m38zm3hze6.fsf@atrus.jesus.cam.ac.uk>

"Nathan Heagy" <gervaserules at yahoo.com> writes:

> Yes, this does work, except on MacOSX, which appears to not have
> TIOCGWINSZ in its TERMIOS module. Is this true for FreeBSD/other
> BSDs?

Does it have TIOCGSIZE?  I don't know anything about BSDs, but I do
know that ncurses tries to use that and then falls back to TIOCGWINSZ.

Cheers,
M.

-- 
  It could be argued that since Suitespot is infinitely terrible,
  that anything else, by very definition of being anything else,
  is infinitely superior.                -- ".", alt.sysadmin.recovery


From tuttledon at hotmail.com  Thu Mar 22 20:45:09 2001
From: tuttledon at hotmail.com (Don Tuttle)
Date: Fri, 23 Mar 2001 01:45:09 GMT
Subject: Standalone executables
References: <998igr$4al$1@niagara.nonet> <iXWt6.156032$Z8.33024358@typhoon.southeast.rr.com> <99aldn$7dh$1@niagara.nonet> <99aonq$a4c$0@216.39.170.247> <Wbfu6.145012$__6.26818921@typhoon.southeast.rr.com> <906C967DEgmcmhypernetcom@199.171.54.195>
Message-ID: <FUxu6.150271$__6.27719240@typhoon.southeast.rr.com>

"Gordon McMillan"
> I use my installer with wxPython all the time (I rarely use TK).
> It doesn't need any special support (well, there was a time when
> what Robin was doing and what I was doing conflicted, but it
> seems to be OK again now).
>
> On Linux, the packager may have to take care of wxWindows on the
> target machine (if it's not statically linked); it's only on Windows
> that the intaller will trace the (system) dependencies of extension
> modules.

Glad to hear it!  Thanks for setting the record straight.

Don




From claird at starbase.neosoft.com  Thu Mar 15 10:49:08 2001
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 15 Mar 2001 09:49:08 -0600
Subject: PIREP: LAF-HYS-FMN-LGB-DRO-HYS-LAF (Python 9)
References: <98pknq$m7h$1@mozo.cc.purdue.edu>
Message-ID: <DE3F5D617FE68B1B.2DF8FECD0DEB5C1A.C14786543AB610AE@lp.airnews.net>

In article <98pknq$m7h$1 at mozo.cc.purdue.edu>,
Kyler Laird <laird at gunsmoke.ecn.purdue.edu> wrote:
>Some notes from our family trip to the Python 9 conference
>in Long Beach:
>	http://www.ecn.purdue.edu/~laird/flying/flights/2001/Long_Beach-A
>
>--kyler

Incidentally, I'd welcome notice of all such URLs; I'm committed to
maintaining an index to 'em at
<URL: http://starbase.neosoft.com/~claird/comp.lang.python/9IPC.html >
-- 

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


From chris.gonnerman at usa.net  Sat Mar 24 15:53:19 2001
From: chris.gonnerman at usa.net (Chris Gonnerman)
Date: Sat, 24 Mar 2001 14:53:19 -0600
Subject: static methods
References: <mailman.985383828.2942.python-list@python.org> <4.3.1.2.20010324124702.01b5c100@exchange.starmedia.net>
Message-ID: <00a701c0b4a4$7d2fe160$a100000a@local>

----- Original Message -----
From: "Robin Thomas" <robin.thomas at starmedia.net>
Subject: Re: static methods
> At 04:38 PM 3/24/01 +0000, you wrote:
> Hideous
> example: declaring a function as "static" in C means that it's invisible
to
> code outside of the source file and to all code following it in the same
> file. :-0

Huh?  static functions in C *are* visible in the same source file, just not
externally (unless you use something like the method table used in Python C
extensions).

How would you use a function if it is invisible in all code scopes?





From smigowsky at dspace.de  Wed Mar 28 03:18:28 2001
From: smigowsky at dspace.de (Stefan Migowsky)
Date: Wed, 28 Mar 2001 09:18:28 +0100
Subject: AW: How does one set a value via win32com
Message-ID: <84257D042AA7D411B3F700D0B7DB9B7C13A9D6@newsintern.dspace.de>

Hi,

you ran on a problem with dynamic dispatch. It can't use properties
with parameters. In a makepy generated wrapper properties with parameters
are translated in methods. So you must use a wrapper to use such parameters.
To ensure a wrapper use the gencache module, in your case

import win32com.client
import regex
wcc = win32com.client

# Ensure that a makepy generated wrapper is available
wcc.gencache.EnsureDispatch("COMAdmin.COMAdminCatalog")

adm = wcc.Dispatch( "COMAdmin.COMAdminCatalog" )
app = adm.GetCollection("Applications")
app.Populate()
 
for thingy in app:
  if regex.match("^Hello World.*",thingy.Name)!= -1:
     print "what we have here is ...",thingy.Name
     # Use generated method which is not available dynamically
     # since the COM object does not export SetValue
     thingy.SetValue("Name", thingy.Name + "t0")
     print thingy.Value("Name")
     break
  else:
     print thingy.Name
app.SaveChanges()


	Stefan


> -----Urspr?ngliche Nachricht-----
> Von: Richard Weth [mailto:richard at loudcloud.com]
> Gesendet: Dienstag, 27. M?rz 2001 21:24
> An: python-list at python.org
> Betreff: How does one set a value via win32com
> 
> 
> All
> 
> I have been banging away on an assignment that requires me to 
> alter com
> plus settings
> (something one should never do normally). Here is the pertinent code
> fragment:
> import win32com.client
> import regex
> wcc = win32com.client
> 
> adm = wcc.Dispatch( "COMAdmin.COMAdminCatalog" )
> app = adm.GetCollection("Applications")
> app.Populate()
> 
> for thingy in app:
>   if regex.match("^Hello World.*",thingy.Name)!= -1:
>     print "what we have here is ...",thingy.Name
>     thingy.Value("Name") = thingy.Name + "t0"          #
> <================ THE PROBLEM
>     break
> 
>   else:
>     print thingy.Name
> app.SaveChanges()
> 
> 
> Now the code in VBS
>     thingy.Value("Name") = thingy.Name & "t0"
> works just fine but in python I get:
> >>> thingy.Value("Name") = "joemama"
> SyntaxError: can't assign to function call
> 
> Is there a way to set that (and other) properties in Python via
> win32com. I really
> would like to learn as little VB as humanly possible and not 
> write this
> in it's own VBScript.
> Any and all suggestions welcome!
> 
> --
> Sincerely
>   Richard A. Weth
>   Sr. Software Engineer
>   LoudCloud, Inc
>   richard at loudcloud.com
> 
> 
> 
> -- 
> http://mail.python.org/mailman/listinfo/python-list
> 



From sholden at holdenweb.com  Fri Mar  9 09:23:42 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Fri, 9 Mar 2001 09:23:42 -0500
Subject: Foot in yourself shoot [was: learning C]
References: <mailman.983972829.23481.python-list@python.org> <983975751.18833@newsmaster-04.atnet.at> <3AA82422.53026D3D@cosc.canterbury.ac.nz> <20010308221519.A4367@harmony.cs.rit.edu> <mailman.984113052.19927.python-list@python.org> <984135226.858948@newsmaster-04.atnet.at> <mailman.984146352.13033.python-list@python.org>
Message-ID: <eL5q6.45993$ML1.2000092@e420r-atl2.usenetserver.com>

"Chris Gonnerman" <chris.gonnerman at usa.net> wrote in message
news:mailman.984146352.13033.python-list at python.org...
> ----- Original Message -----
> From: "Werner Schiendl" <ws-news at gmx.at>
> Subject: Re: learning C
[ ... ]
> > Why not have a look at all of it and use what's suited for the
application
> > best?
>
> Like I said, been there, done that, couldn't afford the T-shirt.  Rewiring
> my
> brain for C++ is just too hard.  I'm glad that it works for you, and many
> others,
> but there are also many like me that can't see why you are pushing so
hard.
> I
> guess you just can't see our point of view, though I think I see yours
just
> fine.
>

I have for a long time been a pragmatist when it comes to language choice --
use the most appropriate language for a particular task, assuming the
learning curve for a new language doesn't make it easier to use an existing
but somewhat less-appropriate one.

The computing community has known for a long time that each language is
better for some tasks than others, vide

        http://www.cs.bgu.ac.il/~omri/Humor/shoot.html

(quoted below). As with much good humo(u)r, there are real points of
comparison available for those who care to look for them. I think the C++
quote is especially relevant. To which I would only add:

Python: After half an hour's design work you realise that foot is simply a
special case of limb.extremity, so you use a standard library to solve your
problem!

regards
 Steve

"""
The proliferation of modern programming languages (all of which seem to have
stolen countless features from one another) sometimes makes it difficult to
remember what language you're currently using. This handy reference is
offered as a public service to help programmers who find themselves in such
a dilemma.

TASK: Shoot yourself in the foot.
C: You shoot yourself in the foot.

C++: You accidentally create a dozen instances of yourself and shoot them
all in the foot. Providing emergency medical assistance is impossible since
you can't tell which are bitwise copies and which are just pointing at
others and saying, "That's me, over there."
"""

[Many other languages omitted: follow the link if these are new to you].





From aleaxit at yahoo.com  Sat Mar  3 10:49:23 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sat, 3 Mar 2001 16:49:23 +0100
Subject: Can I capture console output from an os.system() call?
References: <97p7nt$a2s@news.or.intel.com>
Message-ID: <97rf7u02t5u@news1.newsguy.com>

"Brian Geddes" <brian.j.geddes at intel.com> wrote in message
news:97p7nt$a2s at news.or.intel.com...
> All -
>
> When I call a foreign program using an os.system("program.exe") call, is
> possible to get at the console output of "program.exe" from my python
> script?

If it's actually *console* output (performed through the Windows API
for doing console output), no.  If it is, as in the following example,
output performed to standard-handles, then, yes, but you will have
to use os.popen rather than os.system.


> If "program.exe" spits out a message "Hello World!" to the screen when run
> from the DOS command line, and I call 'os.system("program.exe")' as a line
> of my Python script, how do I get at the "Hello World!" message?

If program.exe chooses to perform its output directly 'to the screen'
(via the Console API's of Windows, or other means, such as poking
into the simulated screen memory etc), you're out of luck.  If it more
normally chooses to write to 'standard output', then run it via
xx=os.popen("program.exe","r") -- the returned xx object is a file
like object from which you can read from.


Alex





From mvanier at endor.bbb.caltech.edu  Sun Mar  4 01:03:06 2001
From: mvanier at endor.bbb.caltech.edu (Michael Vanier)
Date: 03 Mar 2001 22:03:06 -0800
Subject: jython question: accessing protected superclass methods
Message-ID: <jl66hqvzol.fsf@endor.bbb.caltech.edu>

Hi,

So, I've been playing with jython for a few hours, and I'm *really*
impressed.  This is a kick-ass system for doing portable graphics
programming.

However, the online docs are a bit sparse, and one problem I've been having
is in trying to call a protected superclass method from a subclass which
overrides the method.  The offending superclass is javax.swing.JPanel.  I
have a subclass of this (let's call it MyJPanel) which needs to override the
protected paintComponent() method of JPanel.  The naive approach:

class MyJPanel:
    # ... lots of code ...

    def paintComponent(self, g):  # g == Graphics object
        JPanel.paintComponent(self, g)
        # extra code goes here

doesn't work, because paintComponent is not accessible from the MyJPanel
namespace.  The online jython docs say this:

    In Python, if I want to call the foo method in my superclass, I use the
    form:

        SuperClass.foo(self)

    This works with the majority of methods, but protected methods cannot be
    called from subclasses in this way.  Instead you have to use the
    "self.super_foo()" call style.

Well, I couldn't get this to work.  I'm not sure if the above statement means
that you're supposed to write (in my case) "self.super_paintComponent(g)" or
self.JPanel_paintComponent(g) but neither one works.  What is the magic
invocation?

Also, I was surprised to find that MyJPanel, just by subclassing JPanel, gets
its own public version of the protected JPanel methods!  So if I don't
override the method, I can call it directly.  I assume this is intentional,
but maybe the jython developers can comment.

Thanks,

Mike

--------------------------------------------------------------
Mike Vanier	mvanier at bbb.caltech.edu
Department of Computation and Neural Systems, Caltech 216-76


From aleaxit at yahoo.com  Wed Mar 28 14:41:29 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Wed, 28 Mar 2001 21:41:29 +0200
Subject: list vs tuple
References: <hGlw6.782$p5.3210@news1.rivrw1.nsw.optushome.com.au> <3AC1EB93.7BC87292@hack.org> <wEmw6.785$p5.3426@news1.rivrw1.nsw.optushome.com.au> <3AC1F93A.BAC9E34B@hack.org>
Message-ID: <99teuu02p5f@news1.newsguy.com>

"Rikard Bosnjakovic" <bos at hack.org> wrote in message
news:3AC1F93A.BAC9E34B at hack.org...
> deadmeat wrote:
>
> > so is that the only difference?
>
> Both are sequences, but tuples are "read-only", that's the only
> difference (except for the in-place-methods which requires mutability).

Actually, tuples have NO methods -- including those that would
require no mutability at all:

D:\pybridge\pybr>python
Python 2.0 (#8, Oct 16 2000, 17:27:58) [MSC 32 bit (Intel)] on win32
Type "copyright", "credits" or "license" for more information.
>>> dir([])
['append', 'count', 'extend', 'index', 'insert', 'pop', 'remove', 'reverse',
'sort']
>>> dir(())
[]
>>>

Note the lack of 'count' and 'index', which do not mutate a list
when applied to it, but are still missing from tuples.


Alex






From Pahud.bbs at openbazaar.net  Mon Mar 19 06:28:44 2001
From: Pahud.bbs at openbazaar.net (©¬­J¼w)
Date: 19 Mar 2001 11:28:44 GMT
Subject: HTML escape in Python 1.5.2
Message-ID: <3fQTii$l4N@openbazaar.net>

Hello there,

I want to do the html escaping in python, however, HTMLgen class
does not ship with python 1.5.2. Is there any alternative?
Thanks in advance.
--
Patrick Hsieh                     * Email:  pahud @ python.nu
CyberworkSolution Inc.            * TEL:    02-8771-2261
System Engineer                   * http://www.cyberworksolution.com
--
?? Origin: ???????uBBS <openbazaar.net> 
?? From: earth.cyberworksolution.com


From Michael.Hearne at noaa.gov  Sat Mar 24 09:01:14 2001
From: Michael.Hearne at noaa.gov (Michael Hearne)
Date: Sat, 24 Mar 2001 09:01:14 -0500
Subject: problems installing tkinter 2.0 on linux
Message-ID: <3ABCA8AA.CC6E97D8@noaa.gov>

I have RH7 linux on an Intel PC.  I downloaded Python 2.0, and Tkinter. 
When I try to install
Tkinter, I get a message telling me that it has dependencies on tcl8.0
and tk8.0.

I uninstalled tcl8.3 (was this a bad thing?), and found a src.rpm for
tcl8.0.5.  I tried to install it with rpm -i and nothing happened.  So I
dug through the man page for rpm, then tried:

rpm --rebuild

It took an awfully long time and spewed a bunch of messages on the
screen, and in the end, failed to install.

Does anyone have

1) any tips?
2) a binary rpm for tcltk8.0?

Thanks,

Mike


From kens at sightreader.com  Wed Mar 28 12:56:53 2001
From: kens at sightreader.com (Ken Seehof)
Date: Wed, 28 Mar 2001 09:56:53 -0800
Subject: diff lists
References: <3AC1DD4E.4BA7D421@aon.at>
Message-ID: <001901c0b7b0$792bbad0$04090a0a@upcast.com>

It's often a good idea to use dictionaries instead of lists for
this kind of thing if performance is important.  A dictionary
is a better model for a set than a list is, although this may seem
intuitively strange at first.  People should use dictionaries more;
they are very powerful, and not just for lookup tables.

I am assuming your lists won't have duplicates (e.g. ['a','b','b'])
since you seem to be interested in set operations.

Try this:

>>> def andnot(a, b):
...  d = {}
...  for x in a:
...   d[x]=0 # value doesn't matter, we're just using the key
...  for x in b:
...   if d.has_key(x):
...    del d[x]
...  return d.keys()
... 
>>> a = ['a', 'b', 'c', 'd', 'e', 'f']
>>> b = ['e', 'c', 'f']
>>> andnot(a,b)
['d', 'b', 'a']

This will be super fast (linear time) even for really big lists.

Note that if b is small compared to a, you are significantly
better off using dictionaries in the first place rather than
translating as in the above example.

This one uses dictionaries instead of lists:
>>> def andnot2(a, b):
...  d = a.copy()
...  for x in b.keys():
...   if d.has_key(x):
...    del d[x]
...  return d
... 
>>> a = {'a':0, 'b':0, 'c':0}
>>> b = {'b':0}
>>> andnot2(a,b)
{'c': 0, 'a': 0}

----- Original Message ----- 
From: "Oliver Vecernik" <vecernik at aon.at>
Newsgroups: comp.lang.python
To: <python-list at python.org>
Sent: Wednesday, March 28, 2001 4:47 AM
Subject: diff lists


> Hi,
> 
> I've got following two lists:
> 
> ['a', 'b', 'c', 'd', 'e', 'f']
> ['e', 'c', 'f']
> 
> I'd like to have the result:
> 
> ['a', 'b', 'd']
> 
> The list need not to be ordered. ['d', 'a', 'b'] will also be ok. What
> is the most effective way to achive that result?
> 
> Oliver
> -- 
> http://mail.python.org/mailman/listinfo/python-list
> 




From hzhu at mars.localdomain  Wed Mar 21 03:41:17 2001
From: hzhu at mars.localdomain (Huaiyu Zhu)
Date: Wed, 21 Mar 2001 08:41:17 GMT
Subject: Yet Another PEP:  Query Protocol Interface or __query__
References: <Pine.LNX.4.21.0103201837570.21343-100000@clarkevans.com> <998r4j02vmb@news2.newsguy.com> <Pine.LNX.4.21.0103202032080.14634-100000@clarkevans.com>
Message-ID: <slrn9bgprm.rv.hzhu@mars.localdomain>

On Tue, 20 Mar 2001 20:49:30 -0500, Clark C. Evans <cce at clarkevans.com> wrote:
>On Wed, 21 Mar 2001, Alex Martelli wrote:
>> 
>> At least when read at 1 AM -- sounds cool.  Further, the
>> builtin query function could support _adding interfaces to
>> an object post-facto, non-invasively_, much like the
>> 'typeclass'-based ideas that were floating in my mind
>> recently and surfaced in some of my recent posts. 
>
>Right.
>
>> I.e., the query() builtin would both try getting the interface
>> from the _object_, by passing that interface to the obj's
>> query method, AND also try doing it the other way
>> 'round -- asking the interface if it knows how to wrap
>> the object so it (the interface) gets implemented. 
>
>That is a very cool idea!  And the reverse lookup could
>even be added much later as a seperate PEP.  I was trying
>to stay simple and to de-couple this PEP as much as possible
>from the other possibilities being tossed about in the
>type special interest group.  I'd hate to have a dependency
>emerge when one really isn't necessary.

I have an implementation of this idea at

http://www.geocities.com/huaiyu_zhu/python/Interface/typecheck.txt
(You can try it.  This is in fact a .py file, but geocities does not allow
that extension.  Should get a Starship account someday.)

The explantions are here
http://www.geocities.com/huaiyu_zhu/python/Interface/README.txt

Summary of points:

1. It should be possible to define interface after the fact, without
   redefining the objects.

2. It should be possible to use it for builtin types and other things, not
   just class objects.  Eg., it allows saying a particular instance of a
   class satisfies the interface without dragging all other instances of
   this class in.

3. It should allow "reversed inheritance".  That is, saying things in this
   order would allow f(sys.stdin) to work:

      sys.stdin is a file
      file is a kind of stream
      f want a stream as argument

   In other words, you can build up interfaces by going from specific to
   general, in contrast to class inheritance which goes the other way.

If you are interested, there are a few other files at
http://www.geocities.com/huaiyu_zhu/python/Interface/


-- 
Huaiyu Zhu   <hzhu at users.sourceforge.net>


From bbollenbach at homenospam.com  Wed Mar  7 21:12:46 2001
From: bbollenbach at homenospam.com (Brad Bollenbach)
Date: Thu, 08 Mar 2001 02:12:46 GMT
Subject: Can Python replace Visual Basic? Should it?
References: <5DZo6.11126$hn5.1578410@news1.rdc1.mb.home.com> <u8zmhyusj.fsf@advancedmp.net> <mailman.984016452.13746.python-list@python.org>
Message-ID: <yUBp6.14006$hn5.2035563@news1.rdc1.mb.home.com>

"Ken Seehof" <kens at sightreader.com> wrote in message
news:mailman.984016452.13746.python-list at python.org...
> That's easy.  Yes.

Are you sure? What practical experience (see my original post) with
replacing VB GUI programming with doing it in Python do you have?

> Also, I recommend wxPython (www.wxpython.org) as your GUI framework (in
> python you have several good ones to choose from).  It's cross platform
and
> a much cleaner design than the Microsoft stuff.  Be sure to run the
wxPython

Yes, I know of "all the major" GUI toolkits for Python thanks, that's not
what I was asking. :)

> demo program; it will save you lots of time learning.

Yes, I've already run the demo many, many times, in a galaxy far. far away.




From emile at fenx.com  Thu Mar 29 15:12:09 2001
From: emile at fenx.com (Emile van Sebille)
Date: Thu, 29 Mar 2001 12:12:09 -0800
Subject: Quoted identifiers in Python?
References: <3ac3387f$0$12248$cc9e4d1f@news.dial.pipex.com> <3AC35C84.FBED128@alcyone.com> <3ac38552$0$15025$cc9e4d1f@news.dial.pipex.com> <9a04sn$34lka$1@ID-11957.news.dfncis.de>
Message-ID: <9a054e$30cld$1@ID-11957.news.dfncis.de>

[Note: post appeared as edited by my spell checker ;-)]

a = "Ivor the engine"
x = obj.a

What is x?

Just-glad-it-wasn't-a-bigger-post-ly y'rs

--

Emile van Sebille
emile at fenx.com

---------
"Emile van Sebille" <emile at fenx.com> wrote in message
news:9a04sn$34lka$1 at ID-11957.news.dfncis.de...
> Well, how would this work?
>
> a = ."Ivory the engine"
> ax = obi.a
>
> What is ax?
>
> --





From gunturu_neelima at rediffmail.com  Mon Mar  5 23:51:40 2001
From: gunturu_neelima at rediffmail.com (neelima  g)
Date: 6 Mar 2001 04:51:40 -0000
Subject: (no subject)
Message-ID: <20010306045140.6588.qmail@mailweb32.rediffmail.com>

sir,

     i learnt from your website a lot about python.i was given a job to work on python.i have to work on database.please make me wise how to create a database which is having more than 2 columns using gdbm.please reply me.
                                thanking you,
                                   neelima g.

_____________________________________________________
Chat with your friends as soon as they come online. Get Rediff Bol at
http://bol.rediff.com







From svpeters at t-online.de  Sun Mar  4 19:06:06 2001
From: svpeters at t-online.de (Sven Peters)
Date: Mon, 5 Mar 2001 01:06:06 +0100
Subject: Static class methods
References: <rdzo6.154$gc4.80445@news.pacbell.net> <ta5ka1ks0buc3b@corp.supernews.com>
Message-ID: <97ul9k$n2l$02$1@news.t-online.com>

Chris Armstrong wrote:

> I have trouble believing that something like this is really needed.
> Could you give a real-world example?
Factoty methods, for getting objects created earlier and no longer needed, 
yet expensive to create e.g. database connection pooling.
Sure, you can place this elsewhere(module level or other objects).
Interresting:
You CAN have class methods:
def bert:
        print "Bert"
class.method = bert
class.method()
"Bert"

... but no language support to create them as easy as instance methods

-- 
Sven Peters


From robinjim at earthlink.net  Sat Mar 31 08:57:30 2001
From: robinjim at earthlink.net (robin and jim)
Date: Sat, 31 Mar 2001 13:57:30 GMT
Subject: unable to install MySQLdb (version 0.3.5 ) using ActivePython (build 203)  on Win 98
Message-ID: <enlx6.12199$aP5.1155318@newsread2.prod.itd.earthlink.net>

Hello,

I just downloaded the 2001-03-28 build of MySQLdb (version 0.3.5 ) from
http://home.t-online.de/home/err666/ (thanks Gerhard!) and issued the
command:

python setup.py install

using ActivePython (build 203) on a Windows 98 (2nd edition) computer.

Below is the result.  Any suggestions about what I'm doing wrong.  Thanks.

running install
running build
running build_py
not copying MySQLdb.py (output up-to-date)
not copying MySQLdb.pyc (output up-to-date)
not copying CompatMysqldb.py (output up-to-date)
not copying _mysql_const\__init__.py (output up-to-date)
not copying _mysql_const\CR.py (output up-to-date)
not copying _mysql_const\CR.pyc (output up-to-date)
not copying _mysql_const\FIELD_TYPE.py (output up-to-date)
not copying _mysql_const\FIELD_TYPE.pyc (output up-to-date)
not copying _mysql_const\ER.py (output up-to-date)
not copying _mysql_const\ER.pyc (output up-to-date)
not copying _mysql_const\FLAG.py (output up-to-date)
not copying _mysql_const\FLAG.pyc (output up-to-date)
not copying _mysql_const\REFRESH.py (output up-to-date)
not copying _mysql_const\REFRESH.pyc (output up-to-date)
not copying _mysql_const\CLIENT.py (output up-to-date)
not copying _mysql_const\CLIENT.pyc (output up-to-date)
running build_ext
skipping '_mysql' extension (up-to-date)
running install_lib
not copying build\lib.win32-2.0\MySQLdb.py (output up-to-date)
not copying build\lib.win32-2.0\CompatMysqldb.py (output up-to-date)
not copying build\lib.win32-2.0\_mysql_const\__init__.py (output up-to-date)
not copying build\lib.win32-2.0\_mysql_const\CR.py (output up-to-date)
not copying build\lib.win32-2.0\_mysql_const\FIELD_TYPE.py (output
up-to-date)
not copying build\lib.win32-2.0\_mysql_const\ER.py (output up-to-date)
not copying build\lib.win32-2.0\_mysql_const\FLAG.py (output up-to-date)
not copying build\lib.win32-2.0\_mysql_const\REFRESH.py (output up-to-date)
not copying build\lib.win32-2.0\_mysql_const\CLIENT.py (output up-to-date)
not copying build\lib.win32-2.0\_mysql.pyd (output up-to-date)
byte-compiling C:\PYTHON20\MySQLdb.py to MySQLdb.pyc

[... maybe lost some output here -- jv]

  File "c:\python20\lib\distutils\core.py", line 138, in setup
    dist.run_commands()
  File "c:\python20\lib\distutils\dist.py", line 829, in run_commands
    self.run_command(cmd)
  File "c:\python20\lib\distutils\dist.py", line 849, in run_command
    cmd_obj.run()
  File "c:\python20\lib\distutils\command\install.py", line 483, in run
    self.run_command(cmd_name)
  File "c:\python20\lib\distutils\cmd.py", line 328, in run_command
    self.distribution.run_command(command)
  File "c:\python20\lib\distutils\dist.py", line 849, in run_command
    cmd_obj.run()
  File "c:\python20\lib\distutils\command\install_lib.py", line 93, in run
    self.byte_compile(outfiles)
  File "c:\python20\lib\distutils\command\install_lib.py", line 130, in
byte_com
pile
    verbose=self.verbose, dry_run=self.dry_run)
  File "c:\python20\lib\distutils\util.py", line 423, in byte_compile
    compile(file, cfile, dfile)
  File "c:\python20\lib\py_compile.py", line 60, in compile
    codeobject = __builtin__.compile(codestring, dfile or file, 'exec')
TypeError: compile, argument 1: expected string without null bytes, string
found





From barry at digicool.com  Tue Mar 27 02:44:51 2001
From: barry at digicool.com (Barry A. Warsaw)
Date: Tue, 27 Mar 2001 02:44:51 -0500
Subject: PEP 1, PEP Purpose and Guidelines
References: <mailman.985196613.7994.python-list@python.org>
	<99ifgu$jei$1@panix3.panix.com>
	<200103241546.KAA19937@panix3.panix.com>
Message-ID: <15040.17651.954045.580823@anthem.wooz.org>

>>>>> "AM" == Aahz Maruch <aahz at panix.com> writes:

    AM> [posted & e-mailed]

[Aside: say Aahz, when you post and email, can you do it with CC's so
it's easier to follow up? -baw]

    >>  Note the addition of the Replaces: and Replaced-By: headers
    >> for formalizing the PEP replacement policy (thanks to Andrew
    >> Kuchling for the idea and patch).

    AM> After taking a closer look at Moshe's numeric proposals, I
    AM> suggest adding a Depends-On: header.  (E.g. PEP 240 depends on
    AM> PEP 239.)

What about Requires:?  Or Dependencies: where each entry on the line
would be a PEP number (which we could hyperlink fairly easily).

-Barry



From anduguid at NOThome.com  Sat Mar 10 00:11:10 2001
From: anduguid at NOThome.com (Luke)
Date: Sat, 10 Mar 2001 05:11:10 GMT
Subject: tkinter, activepython, win32
Message-ID: <OHiq6.4422$tP3.71012@news1.rdc1.bc.home.com>

python newbie!
cannot get tkinter programs to run...

did 'normal' install of activepython 2.0 build 203
installed Tcl/Tk 8.3 for Windows, Binary Distribution

am i missing something here?

wxpython works fine, should i stick with it?

luke
--
remove the NOT from the email address




From skip at mojam.com  Thu Mar  8 17:42:57 2001
From: skip at mojam.com (Skip Montanaro)
Date: Thu, 8 Mar 2001 16:42:57 -0600 (CST)
Subject: Get terminal size in perl and python
In-Reply-To: <20010308194028.A1616@bork.demon.nl>
References: <20010308194028.A1616@bork.demon.nl>
Message-ID: <15016.2801.964727.460101@beluga.mojam.com>

    Egbert> My perl correspondent uses:
    Egbert>     use Term::ReadKey;
    Egbert>     ($cols,$lines)=GetTerminalSize()
    Egbert> What is the python equivalent ?
    Egbert> If I don't come up with one, python looses ten points.

Looks like this will work.  (I've never used curses, so there may be a more
appropriate way to do it.)

    import curses
    stdscr = curses.initscr()
    rows, cols = stdscr.getmaxyx()

For more details, check out the online docs:

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

-- 
Skip Montanaro (skip at mojam.com)
Support Mojam & Musi-Cal: http://www.musi-cal.com/sponsor.shtml
(847)971-7098



From thomas at xs4all.net  Fri Mar 16 17:15:20 2001
From: thomas at xs4all.net (Thomas Wouters)
Date: Fri, 16 Mar 2001 23:15:20 +0100
Subject: redefining stdout
In-Reply-To: <3AB240E4.1EB6E144@dvinc.com>; from mark.limber@dvinc.com on Fri, Mar 16, 2001 at 09:35:48AM -0700
References: <3AB240E4.1EB6E144@dvinc.com>
Message-ID: <20010316231519.G29286@xs4all.nl>

On Fri, Mar 16, 2001 at 09:35:48AM -0700, Mark Limber wrote:

> >>> import t # short file names = less typing!

In Python 2.0 and later, you can do

>>> import mystdoutreplacementthatunfortunatelyisnotworking as t

to avoid the confusion (and name clashing!) of short filenames.

-- 
Thomas Wouters <thomas at xs4all.net>

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



From tuttledon at hotmail.com  Fri Mar  9 20:22:41 2001
From: tuttledon at hotmail.com (Don Tuttle)
Date: Sat, 10 Mar 2001 01:22:41 GMT
Subject: Python In A Nutshell
References: <nG5q6.54910$__6.11376601@typhoon.southeast.rr.com> <Ufcq6.140786$47.2078615@news.bc.tac.net>
Message-ID: <Blfq6.60716$__6.11886688@typhoon.southeast.rr.com>

Planning stages?  We've all seen his posts (a.k.a. treatises) here.  I'd say
he should have the final draft to you in 3 weeks tops. ;-)
Don

<LLewin at oreilly.com> wrote in message
news:Ufcq6.140786$47.2078615 at news.bc.tac.net...
> Hi,
> Yep--Python Nutshell is "forthcoming"...we're just in the planning stages
at this point--very early on in the process.  Any suggestions are welcome!
> Laura
> LLewin at oreilly.com
>
> > Hey Alex,  congrats on landing "Python In A Nutshell" book deal.  One
> > problem.  I went down to the book store and it's not there.  What gives?
> > <wink>
> >
> > Don
> >
> >
>
>
> ==================================
> Posted via http://nodevice.com
> Linux Programmer's Site




From jurgen.defurne at philips.com  Wed Mar 28 01:18:38 2001
From: jurgen.defurne at philips.com (jurgen.defurne at philips.com)
Date: Wed, 28 Mar 2001 08:18:38 +0200
Subject: newbie question
Message-ID: <0056900016935967000002L072*@MHS>

moshez at zadka.site.co.il@SMTP at python.org on 23/03/2001 20:26:40
Sent by:	python-list-admin at python.org
To:	claird at starbase.neosoft.com@SMTP
cc:	python-list at python.org@SMTP 
Subject:	Re: newbie question
Classification:	


On 23 Mar 2001, claird at starbase.neosoft.com (Cameron Laird) wrote:

<snipped a lot of wise things about CGI and Apache>
>> That's a lot more exciting
>> for them than hearing that, with a mere week or two of
>> study, they can install Zope on their own.

>That's pure FUD. Zope takes <depending on your internet connection> time
>to install, about 5 minutes to compile and 30 seconds to startup. And
>using it is not necessarily harder then learning CGI. I'm only mentioning
>it because "Zope is hard" seems to be a commonly uttered phrase, one that
>I believed until I actually had to get a Zope server up and running.

I have given an introductory course on Linux. Most of the people involved
had no previous experience with it. I mentioned Zope only once, and at the
end of the course one of the trainees came to me and said :
'We have been experimenting with Zope, very nice, very easy'

>Hooking up Zope to Apache (say, with reverse proxying) is very easy, though
>it does use some Apache configuration knowledge. This can all be learned
>in a day, and anyone seriously using Apache needs to know about this
>stuff.
--
"I'll be ex-DPL soon anyway so I'm        |LUKE: Is Perl better than Python?
looking for someplace else to grab power."|YODA: No...no... no. Quicker,
   -- Wichert Akkerman (on debian-private)|      easier, more seductive.
For public key, finger moshez at debian.org  |http://www.{python,debian,gnu}.org

--
http://mail.python.org/mailman/listinfo/python-list





From aahz at panix.com  Fri Mar  9 17:24:55 2001
From: aahz at panix.com (Aahz Maruch)
Date: 9 Mar 2001 14:24:55 -0800
Subject: how to make a function thread-safe ?
References: <3AA94A6A.50138001@noaa.gov>
Message-ID: <98bl7n$cn6$1@panix2.panix.com>

In article <3AA94A6A.50138001 at noaa.gov>,
j vickroy  <jim.vickroy at noaa.gov> wrote:
>
>If a function may be applied (simultaneously) by multiple threads is the
>safe (correct) procedure to wrap the function inside a class and have
>each thread create its own instance of the class?

As long as the function uses strictly local data, there's no need to do
any wrapping or locking, just as it's safe to use functions from the
string module.  If the function uses shared data, things get a lot more
complicated, and you'll need to give us more info if you want advice.
-- 
                      --- Aahz (Copyright 2001 by aahz at pobox.com)

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

"..., and some of you may regard all women as evil traps that exist only
to tease, torture, and suck out your very soul."  --DrMax


From s2mdalle at titan.vcu.edu  Fri Mar  2 12:42:58 2001
From: s2mdalle at titan.vcu.edu (David Allen)
Date: Fri, 02 Mar 2001 17:42:58 GMT
Subject: Tkinter:  image doesn't exist.
References: <1iBn6.1691$Ry6.649664@typhoon2.ba-dsg.net> <A7In6.16280$Qb7.2658868@newsb.telia.net>
Message-ID: <CYQn6.2149$Ry6.967802@typhoon2.ba-dsg.net>

In article <A7In6.16280$Qb7.2658868 at newsb.telia.net>, "Fredrik Lundh"
<fredrik at pythonware.com> wrote:

> David Allen wrote:
>> What kind of a crappy error message is "image doesn't exist?"
> 
> it means exactly what it says: when created, the PhotoImage class registers
> a photoimage object with Tk, but when you try to use it in FooModule, it's
> no longer there.
> 
> are you perhaps using multiple root windows?

That is *exactly* what I'm doing.

> (each root has its own interpreter context, and images registered
> in one context aren't visible in other contexts.  use Toplevel in- stead).

D'oh!  Thank you very much for the feedback.  I kept
RTFM'ing, and I wasn't finding anything.  How you're
supposed to know these types of things without someone
telling you is beyond me.  :)

-- 
David Allen
http://opop.nols.com/
----------------------------------------
It is impossible to sharpen a pencil with a blunt axe.  It is equally
vain to try to do it with ten blunt axes instead.
                -- Edsger W. Dijkstra


From danielk at aracnet.com  Tue Mar  6 10:04:28 2001
From: danielk at aracnet.com (Daniel Klein)
Date: Tue, 06 Mar 2001 07:04:28 -0800
Subject: Overriding UserList methods
References: <b566atck0ruu4tae8n1srrkntbmme26gor@4ax.com> <aUFo6.14917$ML1.731870@e420r-atl2.usenetserver.com>
Message-ID: <qqu9atsitacnuacu29k25rvtrieofeg6nk@4ax.com>

On Mon, 5 Mar 2001 00:52:40 -0500, "Steve Holden" <sholden at holdenweb.com>
wrote:

>"Daniel Klein" <danielk at aracnet.com> wrote in message
>news:b566atck0ruu4tae8n1srrkntbmme26gor at 4ax.com...
>> I must be missing something, cos I can't figure out how to override a
>method
>> using the UserList module. Here is some code:
>>
>> import UserList
>> class MyList(UserList.UserList):
>>       __def __init__(self):
>>         UserList.UserList.__init__(self)
>>
>While this would work, I don't see it's strictly necessary. Why not just
>inherit the __init__ method from UserList?

One of the reasons for subclassing UserList is to provide additional instance
variables. So I need my own __init__ method to initialize this, don't I?

>>     def append(self, object):
>>         self.append(object)
>>
>> Understandably, this goes into a recursive frenzy until it hits the stack
>> limit. :-(
>>
>Well, if you want *you* append method to make use of UserList's method you
>would have to write
>
>    def append(self, object):
>        UserList.append(self, object)
>
>Since all you are doing is reflecting the underlying UserList's append()
>method, however, it would seem to make more sense to simply allow append()
>to be inherited as well!

Again, I need to provide append() with some specialized code, ie I need to
ensure that only certain types of objects are in the list.

>> The only way I could get this to work was to give the method another name,
>ie
>>
>>     def add(self,object):   ## This works but I would rather use 'append'
>>         self.append(object)   ## as the method name.
>>
>> So how can I override the append() method so that I can add my specialized
>code
>> to it?
>>
>> Daniel Klein
>
>The clue is to use the unbound method from the superclass, as I showed you
>above.
>
>Another way would be to create a UserList object as an instance variable,
>and then delegate methods to the UserList which you don't recognize, but
>this might be going too far.

Good suggestion! I don't need this for here but I might use it elsewhere.

> Hope this helps.

Yes, it did. Thanks.

Daniel Klein



From emile at fenx.com  Wed Mar 21 18:17:02 2001
From: emile at fenx.com (Emile van Sebille)
Date: Wed, 21 Mar 2001 15:17:02 -0800
Subject: Simple regex that has me banging my head against a wall
References: <mailman.985213768.20178.python-list@python.org>
Message-ID: <99bcv9$iqe3$1@ID-11957.news.dfncis.de>

In 2.0,

'$1$asdfjke$lkjasdkjfsdasdfj'.replace('$',r'\$')

but then I don't tend to use re's for this kind of thing.  Is there some
advantage I'm not seeing?

Emile van Sebille
emile at fenx.com

"Timothy Grant" <tjg at exceptionalminds.com> wrote in message
news:mailman.985213768.20178.python-list at python.org...
> Hi again.
>
> I actually thought I had this regex thing down pretty well, but
> I'm getting beat up by what I thought should be a simple one.
>
> >>> x = '$1$asdfjke$lkjasdkjfsdasdfj'
> >>> re.sub(r'\$', r'\\\$', x)
> '\\\\$1\\\\$asdfjke\\\\$lkjasdkjfsdasdfj'
>
> All I want to do is substitute \$ for $.
>
> Thanks.
>
> --
> Stand Fast,
>     tjg.
>
> Timothy Grant                         tjg at exceptionalminds.com
> Red Hat Certified Engineer            www.exceptionalminds.com
> Avalon Technology Group, Inc.         <><       (503) 246-3630
> >>>>>>>>>>>>>Linux, because rebooting is *NOT* normal<<<<<<<<<
> >>>>This machine was last rebooted:  64 days  2:43 hours ago<<
>




From Jaeho.Lee at brooks.com  Fri Mar 16 11:32:18 2001
From: Jaeho.Lee at brooks.com (Lee, Jaeho)
Date: Fri, 16 Mar 2001 11:32:18 -0500
Subject: Yet another C++ binding (using Extension Classes)
Message-ID: <857F15D7E3D8D3118D290008C7CF058602C9755A@mail-naeast1.brooks.com>

> About a year ago, I had to bind a C++ template class into objects that
> could be used from within python.  I made a wrapper for Extension
> Classes that extended the basic structure into a C++ class.  Any objects
> that subclassed this class would automagically become valid python
> objects and could be extended in much the same way as Extension Classes.

I am researching on how to expose "complex" C++ class in python. I found
that SWIG does not support well my specific purpose. If you can open your
source code (even with document - how great !!), it will be very helpfu.

Thanks,
/Jaeho Lee



From stephen_purcell at yahoo.com  Fri Mar 16 13:27:39 2001
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Fri, 16 Mar 2001 19:27:39 +0100
Subject: Killing threads
In-Reply-To: <3ab23ea7.328575630@News.CIS.DFN.DE>; from costas@meezon.com on Fri, Mar 16, 2001 at 05:30:42PM +0000
References: <3ab23ea7.328575630@News.CIS.DFN.DE>
Message-ID: <20010316192738.A5550@freedom.puma-ag.com>

costas at meezon.com wrote:
> I have th program below. How does one go about kill a thread once
> started? Why is there no method for killing threads? I thought del
> would do it but obviously it doesn't.

Thread cancellation is generally considered a bad thing. If I'm not mistaken,
the argument against allowing thread cancellation is concerned with
correctly releasing locks shared between threads.  (I can trust Mr. Martelli
to set me straight on this if I'm wrong :-)

You should think about using a synchronisation mechanism
(threading.Condition, threading.Semaphore, Queue.Queue) so that the worker
thread can exit itself when told to do so by the main thread changing the
synchronising object.

-Steve

-- 
Steve Purcell, Pythangelist
Get testing at http://pyunit.sourceforge.net/
Any opinions expressed herein are my own and not necessarily those of Yahoo



From max at alcyone.com  Mon Mar  5 22:01:37 2001
From: max at alcyone.com (Erik Max Francis)
Date: Mon, 05 Mar 2001 19:01:37 -0800
Subject: PIL and Windows file formats?
Message-ID: <3AA45311.150DB844@alcyone.com>

I installed Python 2.0 and PIL for Python 2.0 on my machine and am
trying to manipulate images.  I try to open a (valid) PNG image (with a
standard .png file extension) with PIL.Image.open and get a "cannot
identify image type" error.  I then debug through PIL.Image.py only to
discover, much to my surprise, the only formats it seems to know about
(in PIL.Image.ID) are GIF, JPEG, and TIFF.

Any idea what in the world is going on here?

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ All the people in my neighborhood turn around and get mad and sing
\__/ Public Enemy
    Crank Dot Net / http://www.crank.net/
 Cranks, crackpots, kooks, & loons on the Net.


From gradha at iname.com  Sat Mar 10 12:57:57 2001
From: gradha at iname.com (gradha at iname.com)
Date: Sat, 10 Mar 2001 18:57:57 +0100
Subject: How do I check return value of popen2.* ?
Message-ID: <5vpd89.4ei.ln@127.0.0.1>

Hello.

I have tried the following test. File t2.py:

	#!/usr/bin/env python
	
	import sys, popen2
	
	sal, en, err = popen2.popen3 ("./test.py")
	val1 = en.close()
	print sal.readlines()
	val2 = sal.close()
	print err.readlines()
	val3 = err.close()
	
	print val1, val2, val3
	
File test.py:

	#!/usr/bin/env python
	
	import sys
	
	sys.exit(2)
	
Now, when I run t2, the output is:

	[]
	[]
	None None None

How am I supposed to check the return value of the opened pipe?

-- 

 Grzegorz Adam Hankiewicz   gradha at iname.com - http://gradha.infierno.org
 Other web pages:  http://glub.ehu.es/  -  http://welcome.to/gogosoftware/


From bos at hack.org  Tue Mar 27 01:15:25 2001
From: bos at hack.org (Rikard Bosnjakovic)
Date: Tue, 27 Mar 2001 08:15:25 +0200
Subject: compiling questions, a la lisp
References: <LQTv6.92647$Ch.13131089@typhoon.we.rr.com>
Message-ID: <3AC02FFD.96FD027A@hack.org>

Pete Shinners wrote:

> these folks say that the LISP language (which i know nothing about),
> can be compiled natively and provide a substantial speedup. these

I suggest you to read the following URL:

	http://www.norvig.com/python-lisp.html


-- 
Rikard Bosnjakovic - http://a214.ryd.student.liu.se/cv/ - ICQ: 1158217

Anyone sending unwanted advertising e-mail to my address will be
charged $250 for network traffic and computing time. By extracting my
address from this message or its header, you agree to these terms.


From urner at alumni.princeton.edu  Sat Mar  3 14:05:12 2001
From: urner at alumni.princeton.edu (Kirby Urner)
Date: Sat, 03 Mar 2001 11:05:12 -0800
Subject: Thinking Pythonically (was Re: gah! I hate the new string   syntax)
References: <mailman.983485923.1136.python-list@python.org> <B7In6.16281$Qb7.2658902@newsb.telia.net> <97ohh2$polt0$1@ID-77331.news.dfncis.de> <97opsb$nhu$1@nntp6.u.washington.edu> <97oupq$qijrh$1@ID-77331.news.dfncis.de> <97p3mh$bkm$1@tyfon.itea.ntnu.no> <B6C58CBD.212F%f8dy@diveintopython.org> <97q14l$jbi$0@216.39.151.169> <7k31at0hbkuqgjtnsfts9nieskove1nli4@4ax.com> <B6C60DB9.48DA%f8dy@diveintopython.org> <97rbvb$lrj$0@216.39.151.169>
Message-ID: <87f2atgq3rk4vfrc7q3mpp3m30plp2p39h@4ax.com>

Donn Cave <donn at oz.net> wrote:

>I used to use map, filter, reduce, but a couple of years ago
>I gave that up because I felt it would make my code harder
>for other people to read.  An explicit loop is better.
>
>	Donn Cave, donn at oz.net

I think it's possible to go overboard with the 'functional
programming' type operators (if we can call 'em that), but
I prefer to use them within reason vs. not use them at all
-- and I think the new list comprehension syntax makes the
code even more readable.

For example, when doing sigma (sum of terms), passing in 
the function and lower/upper term indices, you could write:

   from operator import add  # used below
   def f(x): return x*x      # example function

   def sigma(f,stop,start=1):
      result = 0
      for i in range(start,stop+1):
	 result += f(i)
      return result

or

   def sigma(f,start,stop,start=1):
      return reduce(add,map(f,range(start,stop+1)))

or

   def sigma(f,stop,start=1):
      return reduce(add,[f(x) for x in range(start,stop+1)])

This last is my personal favorite.  I don't find
it harder to read than the first.

And I _really_ like the combination of functions 
below for listing out numbers < n that are relatively 
prime to n:

  def gcd(n,m):
      """
      Euclid's Algorithm for GCD
      """
      while m:
         n,m = m,n%m
      return n

  def relprimes(n):
      """
      List coprimes < n
      """
      return [x for x in range(1,n) if gcd(n,x)==1]

So neat, so clean!  Pythonic math is kewl.

Kirby
	


From cribeiro at mail.inet.com.br  Wed Mar 28 19:04:56 2001
From: cribeiro at mail.inet.com.br (Carlos Alberto Reis Ribeiro)
Date: Wed, 28 Mar 2001 21:04:56 -0300
Subject: What *instance* called me & proxy methods
In-Reply-To: <99tpim$2rrba$1@ID-11957.news.dfncis.de>
References: <3AC1BE92.ACE82741@ua.ac.be>
 <mailman.985798623.5866.python-list@python.org>
Message-ID: <5.0.2.1.0.20010328205752.021aa010@mail.inet.com.br>

At 14:42 28/03/01 -0800, Emile van Sebille wrote:
>It seems to me that what you're trying to do is discover the container of an
>instance.  Whether the instance is wrapped in a class or held in a list,
>when the instance's methods are invoked, self is self and not the container.
>If you take a look at pep 232
>http://python.sourceforge.net/peps/pep-0232.html you may find that, if
>instance method attribute setting were allowed, this could be what you're
>looking for.

Thanks for the pointer. Note that I have two related problems, and this PEP 
may solve both of them.

1) I want to have a way to hide the im_* pointers from the code. The 
intention is to hide the implementation details of the object that is being 
'proxied'.

2) I want to set some special property to indicate who's the 'owner'. The 
problem is that instances may be referenced in several places. So my idea 
is to create a proxy, that is owned by the container, but points to the 
same instance method.


>Perhaps you can take this PEP to the next step if it's what you're after

Maybe I'll give it a try - not now, but in a few weeks.


Carlos Ribeiro





From erno-news at erno.iki.fi  Sun Mar  4 17:36:23 2001
From: erno-news at erno.iki.fi (Erno Kuusela)
Date: 05 Mar 2001 00:36:23 +0200
Subject: Concurrency and shelve
References: <3AA26D5E.2FD28B40@stroeder.com> <kuk865l6e1.fsf@lasipalatsi.fi> <3AA2C1F0.980A23F7@stroeder.com>
Message-ID: <kuelwdkvq0.fsf@lasipalatsi.fi>

In article <3AA2C1F0.980A23F7 at stroeder.com>, Michael Str?der
<michael at stroeder.com> writes:

| Hmm, it seems to me that gdbm does prevent other processes from
| opening the database at all.

yes.

| BTW: I'm trying to use shelve to store web session data. How does
| others solve this problem with the the standard lib without being to
| DB-specific? Unpossible?

don't keep it open all the time.

  -- erno


From jwbnews at scandaroon.com  Sat Mar 17 22:18:45 2001
From: jwbnews at scandaroon.com (John W. Baxter)
Date: Sat, 17 Mar 2001 19:18:45 -0800
Subject: More accurate summations [was Re: This math scares me]
References: <mailman.984726142.2994.python-list@python.org> <98umht$7e0$1@agate.berkeley.edu> <slrn9b7b6k.10s.grante@tuxtop.visi.com>
Message-ID: <jwbnews-8D08E1.19184417032001@corp.supernews.com>

In article <slrn9b7b6k.10s.grante at tuxtop.visi.com>,
 grante at visi.com (Grant Edwards) wrote:

> That was (I think) the original topic: BCD floating point vs.
> binary floating point.

The origin of the thread was the 101,278,513th co-discovery that 
obviously exact results aren't exact.  [Counting more than just Python 
users, of course...not to mention my making up the number just now.]  
Triggered, again, by the interpreter's (wholly proper, IMHO) use of 
repr() to display a result.

  --John

-- 
John W. Baxter   Port Ludlow, WA USA  jwbnews at scandaroon.com


From pinard at iro.umontreal.ca  Thu Mar 29 21:32:35 2001
From: pinard at iro.umontreal.ca (=?iso-8859-1?q?Fran=E7ois?= Pinard)
Date: 29 Mar 2001 21:32:35 -0500
Subject: Should I close after popen??
In-Reply-To: <fhOs6.510$Fb.777424@typhoon2.ba-dsg.net>
References: <m38zm6r8ih.fsf@master.athome>
	<fhOs6.510$Fb.777424@typhoon2.ba-dsg.net>
Message-ID: <oqsnjwm1fg.fsf@lin2.sram.qc.ca>

[David Allen]

> In article <m38zm6r8ih.fsf at master.athome>, "Andrew Markebo"
> <flognat at flognat.myip.org> wrote:

> > I am hacking away a small background program, using popen to call
> > commands, should I close the file in some way when done or it is taken
> > care of automagically??

> IIRC when garbage collected, all file handles are closed automagically
> by python.  But that doesn't mean that it's OK to do it, that's just for
> safety.  You should ALWAYS close your own handles, so you can keep track of
> where they're good and where they're no good.  [...]  It's good practice.

Explicit closing is probably not always good practice, and is not necessarily
safer.  I would completely agree with the above for someone who wants to
write Jython, or write Python that could be quickly turned into Jython.
But if you stick with Python, you can safely rely on the fact that a file
is automagically close as soon as you do not refer to it anymore.

[Donn Cave]

> That's the party line, since the Java implementation can't finalize on
> reference counts and for that matter there's no guarantee that the C
> implementation always will.  But at present it does.  In C Python, you
> may rely on destructor timing if you want, it will happen exactly when
> it should.  The garbage collection option that was introduced in 2.0
> does not change that.  You have to understand that it's not official,
> but it's so useful that it's hard to pretend it isn't there.

In fact, if I remember well, Tim Peters confirmed (on this list) that
Python, the C implementation, is very not likely to change on this.

Code like:

    open(logfile, 'a').write('Done!\n')

or maybe:

    for line in open(logfile).readlines():
        process(line)

is very legible, and would rather be uselessly encumbered with explicit
closes and intermediate variables.  It is not worth writing these in the
style of other languages where closes are explicitely needed.

Of course, there are also cases where explicit closes are welcome and useful.
Yet, in my experience so far, this does not happen frequently.

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



From urner at alumni.princeton.edu  Sat Mar  3 00:58:19 2001
From: urner at alumni.princeton.edu (Kirby Urner)
Date: Fri, 02 Mar 2001 21:58:19 -0800
Subject: shortcut for 'self' wish
References: <3aa050ad.35416419@News.CIS.DFN.DE> <97pnpf$99$1@news.udel.edu>
Message-ID: <ot11atghgm5cccc9oc8tm2pq453c1n8ukh@4ax.com>

"Terry Reedy" <tjreedy at udel.edu> wrote:
>
>This is entirely a social convention for publicly displayed code.  For my
>own work, I use s for self (and o for other in __add__(), etc.).  Python is
>not a straightjacket.
>

I agree with Pilgrim in 'Dive Into Python' that we should
honor the 'self' convention.  These standards make source
more readable across the board and innovating some local
idiosyncratic convention just to save keystrokes (or 
whatever) strikes me as self-indulgent.

Kirby




From dave at superluminal.com  Wed Mar 21 13:44:29 2001
From: dave at superluminal.com (Dave Seidel)
Date: Wed, 21 Mar 2001 13:44:29 -0500
Subject: help building extension with C++
Message-ID: <99asob$jj4$1@pyrite.mv.net>

I'm trying to build a small extension to Python 2.0 as a shared library.
This is on Linux (RH6.2), although the code was originally written on
Windows.

Following the directions in the "Extending and Embedding..." chapter, I
made a local copy of Misc/Makefile.pre.in, made a simple Setup file, and
ran "make -f Makefile.pre.in boot". However, the resulting Makefile has a
problem: it uses $(CCC) as the compiler command for my source files (which
have the extension .cpp), and CCC isn't defined anywhere in the makefile.

My Setup looks like this :

<snip>
*shared*

MRHashTable MRHash.cpp MRHashTable.cpp PythonExtension.cpp

</snip>

What am I missing or doing wrong?

TIA,

- Dave


From tim.one at home.com  Sat Mar  3 05:25:15 2001
From: tim.one at home.com (Tim Peters)
Date: Sat, 3 Mar 2001 05:25:15 -0500
Subject: gah! I hate the new string syntax
In-Reply-To: <97opsb$nhu$1@nntp6.u.washington.edu>
Message-ID: <LNBBLJKPBEHFEDALKOLCKEKGJDAA.tim.one@home.com>

[Donn Cave]
> ...
> Pseudo-code is the language you don't have to know, to understand,
> right?

No such language.

> Familiar symbols, unambiguous syntax.  Could you articulate
> your thoughts on this, how do list comprehensions improve on their
> alternatives, there?

Donn, listcomps are a minor variant of set-builder notation; in wide use at
latest since the 19th century; in America, it appears this is normally
introduced in the 6th or 7th grades (ages 11-12, for those of you in inferior
countries <wink>).  Search the web, and you'll find intros sprinkled with
cheery encouragements like:

    Math is fun ...
    and who can doubt it ...
    If we have ...
    no doubt about it!!!

(quoted without permission from Miss LeBrocq's Sixth Grade Math Syllabus).

Try explaining compositions of higher-order functions to an 11-year old.
That's "their alternatives" in Python practice -- at least to judge from
Kirby's map-happy code <wink>.

if-listcomps-were-a-bad-idea-i-vote-we-send-the-language-straight-
    to-hell-ly y'rs  - tim




From sholden at holdenweb.com  Fri Mar  9 18:06:46 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Fri, 9 Mar 2001 18:06:46 -0500
Subject: Pythonwin Initialization
References: <3aa95fb8.350774771@News.CIS.DFN.DE>
Message-ID: <Wpdq6.46403$1D5.2154843@e420r-atl1.usenetserver.com>

"Costas Menico" <costas at meezon.com> wrote in message
news:3aa95fb8.350774771 at News.CIS.DFN.DE...
> I would like to execute a .py file that initializes my Pythonwin or
> Python environment. Is there some recommended way of doing this
> automatcially when Python starts up?
>
>From the PythonWin readme:

"""
The following command line arguments are recognised. Note that these are
implemented in "intpyapp.py", so should be pretty easy to change!

/run scriptname args
Runs the named script. Works just like "Python.exe scriptname args", except
it is run in the GUI environment, and output goes to the interactive window.

/runargs scriptname args
As above, except the normal Pythonwin 'Run Script' dialog is shown.
"""

regards
 Steve





From francois.granger at free.fr  Sun Mar  4 17:26:12 2001
From: francois.granger at free.fr (=?ISO-8859-1?Q?Fran=E7ois_Granger?=)
Date: Sun, 04 Mar 2001 22:26:12 GMT
Subject: A counter-proposal to __future__ in PEP 236
References: <3AA08F31.55CB6ADB@destiny.com> <mailman.983610546.9172.python-list@python.org> <AP9o6.6393$1D5.198210@e420r-atl1.usenetserver.com>
Message-ID: <1epr4je.1u9drswyvbh8kN%francois.granger@free.fr>

Steve Holden <sholden at holdenweb.com> wrote:

> "Tim Peters" <tim.one at home.com> wrote in message
> news:mailman.983610546.9172.python-list at python.org...
> [ ... ]
> >
> > can-make-it-foolproof-but-not-damned-fool-proof<wink>-ly y'rs  - tim
> >
> Besides which, every time we invent a new and foolproof syste the world has
> a way of producing a complementary new and improved fool who finds another
> hole.
> 
> there's-no-end-to-improving-ease-of-use-ly y'rs  - steve

"Programming today is a race between software engineers striving to
build bigger idiot-proof programs, and the Universe trying to produce
bigger and better idiots. So far, the Universe is winning." - Rich Cook


-- 
Une faq de frj: http://faq.jardin.free.fr/, pr?s de l'encyclop?die
http://nature.jardin.free.fr/
MacPython beginners http://francois.granger.free.fr/MacPython/faq/


From tchur at optushome.com.au  Sat Mar 24 21:44:47 2001
From: tchur at optushome.com.au (Tim Churches)
Date: Sun, 25 Mar 2001 12:44:47 +1000
Subject: Referencing a container object's attributes
References: <200103250239.SAA10240@mail.novus-tele.net>
Message-ID: <3ABD5B9F.666A4C96@optushome.com.au>

Brian Quinlan wrote:
> 
> I gather that you are describing a situation similar to the following:
> 
> class a:
>         def doSomething(self):
>                 # I need to access the "b" instance that instantiated me here!
> 
> class b:
>         __init__(self):
>                 self.a = a()
>                 self.a.doSomething( )
> 

Yes, precisely.

> Is so, there there is no language feature that would allow you to do
> this without passing an instance of 'b' to 'a'. Standard practice would
> be to do something like this:
> 
> class a:
>         def __init__(self, parent):
>                 self.parent = parent
> 
>         def doSomething(self):
>                 self.parent.blah( )
> class b:
>         __init__(self):
>                 self.a = a( self )
> 

OK, many thanks. 

One question: does the circularity inherent in the above (in which a
reference to the container object is placed in an attribute of each
containee) have any potential for causing problems - particularly since
I want to pickle the container objects?

Cheers,

Tim C


> On Saturday, March 24, 2001, at 06:15 PM, Tim Churches wrote:
> 
> > Is there any way for a method of a class instance which is embedded in a
> > container class instance to look up attributes of that container class
> > instance (i.e. the one which contains it)? At the moment I am putting
> > the name of the container instance in an attribute of the containee (at
> > the time when the containee is instantiated). Is there a neater way? All
> > the Python texts cover referencing an object's superclass(es), but not
> > its container object(s).
> >
> > Tim C
> >
> > --
> > http://mail.python.org/mailman/listinfo/python-list
> >



From aleaxit at yahoo.com  Wed Mar  7 02:50:26 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Wed, 7 Mar 2001 08:50:26 +0100
Subject: How to use <>?
References: <DZkp6.18811$pF4.3015652@news1.telusplanet.net>
Message-ID: <984p9b0ri1@news1.newsguy.com>

"Victor Louie" <vlouie at telusplanet.net> wrote in message
news:DZkp6.18811$pF4.3015652 at news1.telusplanet.net...
> Hi,
>
> I was just wondering if any knew what <> does....I've got some sample code

It's an exact and obsolescent synonym for != -- it tests
whether the two operands are different.


Alex





From max at alcyone.com  Thu Mar  8 12:11:41 2001
From: max at alcyone.com (Erik Max Francis)
Date: Thu, 08 Mar 2001 09:11:41 -0800
Subject: Can Python 1.5.2 code run on Python 2.x?
References: <mailman.984069070.9978.python-list@python.org>
Message-ID: <3AA7BD4D.5AFE8EE8@alcyone.com>

Laura Creighton wrote:

> Big change ... string conversion of long int no longer tacks an L on
> the end.  Since I just ripped off the last digit wholesale all over
> this code, I have many lines to change.

Use repr instead.

max at charmaine:~% python
Python 2.0 (#5, Nov 10 2000, 21:47:15) 
[GCC egcs-2.91.66 19990314/Linux (egcs-1.1.2 release)] on linux2
Type "copyright", "credits" or "license" for more information.
>>> l = 299792458l      
>>> l
299792458L
>>> str(l)
'299792458'
>>> repr(l)
'299792458L'

This change in string conversions of longs was one of the documentation
changes, by the way.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ Do you like what you see / Do you like yourself
\__/ Neneh Cherry
    Product's Quake III Arena Tips / http://www.bosskey.net/
 Tips and tricks from the absolute beginner to the Arena Master.


From grante at visi.com  Wed Mar  7 17:19:25 2001
From: grante at visi.com (Grant Edwards)
Date: Wed, 07 Mar 2001 22:19:25 GMT
Subject: I come to praise .join, not to bury it...
References: <kuofvfjb77.fsf@lasipalatsi.fi> <000001c0a64c$497df250$a9807018@cr706570a> <mailman.983997159.4842.python-list@python.org>
Message-ID: <Ntyp6.1479$y6.244056@ruti.visi.com>

In article <mailman.983997159.4842.python-list at python.org>, D-Man wrote:

>Even with my newfound understanding, provided by the Martellibot, I
>agree with this.  Calling a method on an object is fine.  Calling it
>on a literal looks weird.

What makes you think a "literal" is not an object?

> Take for example something like this
>
>three = 1.__add__( 2 )

Looks just fine to a ex-Smalltalker: everything is an object.

-- 
Grant Edwards                   grante             Yow!  These PRESERVES
                                  at               should be FORCE-FED to
                               visi.com            PENTAGON OFFICIALS!!


From grante at visi.com  Tue Mar 13 17:08:30 2001
From: grante at visi.com (Grant Edwards)
Date: Tue, 13 Mar 2001 22:08:30 GMT
Subject: This math scares me
References: <3aad4062.1290021@News.CIS.DFN.DE> <20010312170356.B3513@better.net> <mailman.984480856.18097.python-list@python.org> <984486395.407379@newsmaster-04.atnet.at> <slrn9asf1d.ja3.qrczak@qrnik.zagroda> <984518199.127140@newsmaster-04.atnet.at> <jwbnews-09D75F.13313313032001@corp.supernews.com>
Message-ID: <yTwr6.1235$Tg.167429@ruti.visi.com>

In article <jwbnews-09D75F.13313313032001 at corp.supernews.com>, John W. Baxter wrote:
>In article <984518199.127140 at newsmaster-04.atnet.at>,
> "Werner Schiendl" <ws-news at gmx.at> wrote:
>
>> True, but even 0.1 and 0.2 need not necessarily be exact values.
>
>Of the single decimal digit numbers, 20% are exact (x.0 and
>x.5, for not huge x).  For two decimal places, it's down to 4%
>(x.00, x.25, x.50, x.75) and it spirals quickly downward from
>there.
>
>Bill Gates got this right in the early days, delivering two
>Basic interpreters: binary floating point for speed, and
>decimal floating point for money and such.  (I never looked at
>how well his interpreters did...I just used them.)

I don't know if you can credit Gates with doing anything other
than following the pack.  Back when I first started doing
software (CP/M 1.7), all language packages for "personal"
computers came with both BCD and binary FP math libraries so
you could do exact monetary calculations if you wanted.

Of course we'd never imagined we'd ever see hardware FP and BCD
libraries were only marginally slower than binary ones. Now
that FP is usually in hardware, BCD FP (or fixed point) will be
pretty slow by comparison.

-- 
Grant Edwards                   grante             Yow!  Imagine--a WORLD
                                  at               without POODLES...
                               visi.com            


From greg at cosc.canterbury.ac.nz  Wed Mar  7 20:50:12 2001
From: greg at cosc.canterbury.ac.nz (Greg Ewing)
Date: Thu, 08 Mar 2001 14:50:12 +1300
Subject: Static class methods
References: <9825od0omq@news1.newsguy.com> <002201c0a6de$e00442a0$0300000a@acer>
Message-ID: <3AA6E554.DFB04DBF@cosc.canterbury.ac.nz>

> Alex Shindich wrote:
> 
> This seems weird, since the level
> above the class-- the modules, can have functions. So can the level
> below-- the instance methods. But the classes cannot.

The class scope is already special with respect to
nested scopes anyway, since it doesn't count as a
scope when you're looking out from inside a method.
So even if there were class methods, you wouldn't
be able to call them from within a method without
qualification of some sort.

However, class scopes *do* behave like other scopes
in that the code in the body of the following function:

  def g():
    def f():
      print "foo"
    f()

works just the same way when *executed* in a class
scope:

  class C:
    def f():
      print "foo"
    f()

Try it and you'll see!

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


From gunturu_neelima at rediffmail.com  Tue Mar  6 02:07:29 2001
From: gunturu_neelima at rediffmail.com (neelima  g)
Date: 6 Mar 2001 07:07:29 -0000
Subject: (no subject)
Message-ID: <20010306070729.24325.qmail@mailweb20.rediffmail.com>

sir,

   i was given a job to create a database using gdbm module in python with more than two columns.the fields are
  userid,from,subject,body,attachment information.

i could not find the way to ctreate the database with these fields and how to enter the data into that database.i mean i did not find any write function in that module.
                               thanking you,
                                  neelima g.


_____________________________________________________
Chat with your friends as soon as they come online. Get Rediff Bol at
http://bol.rediff.com







From jsoons at juilliard.edu  Tue Mar 27 10:26:13 2001
From: jsoons at juilliard.edu (Jonathan Soons)
Date: Tue, 27 Mar 2001 10:26:13 -0500
Subject: warrior geeks
Message-ID: <B7C74E1A6FBAD411BFC60002B309DDF4E43F@mailbox.juilliard.edu>


-----Original Message-----
From: Delaney, Timothy [mailto:tdelaney at avaya.com]
Sent: Monday, March 26, 2001 7:16 PM
To: python-list at python.org
Subject: RE: warrior geeks



> 
> What are you?  Some kind of Commie?  Guns form the very foundation of
> the Pax Americana that gives you the freedom to use Python.

I hope there was a <wink> there that I missed ...

 My father had a .22 bolt-action rifle when we
lived in the country for shooting foxes - it was kept in 3 pieces in various
parts of the house. Getting the bullets required taking a whole pile of
stuff out of the pantry, taking up a board in the floor and reaching down to
the bottom of a hole which an adult could only just reach.

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

This is irresponsible. What if a fox happens along?
After seeing many grown men get their throats ripped out by foxes
we keep a Kalashnikov (in fully auto with the safety off) propped
outside the front door.



From dwig at advancedmp.net  Fri Mar 16 12:29:56 2001
From: dwig at advancedmp.net (Don Dwiggins)
Date: 16 Mar 2001 09:29:56 -0800
Subject: site.py on Windows
In-Reply-To: "Chris Gonnerman"'s message of "Thu, 15 Mar 2001 20:54:24 -0600"
References: <%tKr6.2570$pX.26174@news1.onlynews.com>
	<98oq3r$1e0$1@troll.powertech.no> <u3dcfagwh.fsf_-_@advancedmp.net>
	<013501c0adc4$6a94b620$a100000a@local>
Message-ID: <uae6l1v1n.fsf@advancedmp.net>

Chris Gonnerman writes in response to my efforts to make use of site.py:
> Frankly, I just hack site.py like so:

> # site.py, Python 2.0, about line 259
> addsitedir("c:/python/lib/site-packages")

> It may not be the right way to do it, but it works for me.  Don't
> forget to substitute your Python install directory.

Thanks for that.  Digging a little deeper, I was able to avoid modifying
site.py.  It turns out that you can put .pth files in the install directory
itself (c:/python in your case) rather than in the site-packages directory,
and they'll be recognized.  Not as clean as the documentation describes, but
it does get the job done.

Finally, can anyone in the "core" team enlighten us as to why site.py treats
Unix differently from other OS's?  As I mentioned before, I'd be willing to
work on making this more uniform (at least for Windows -- I don't have
access to a Mac) if I knew what the issues are.

-- 
Don Dwiggins                    "Solvitur Ambulando"
Advanced MP Technology
dwig at advancedmp.net




From aleaxit at yahoo.com  Wed Mar  7 10:06:26 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Wed, 7 Mar 2001 16:06:26 +0100
Subject: Python, ASP, ADO, and Image string/binary problems
References: <ta0amgebq0sn0b@corp.supernews.com> <tabjei1p4a9481@corp.supernews.com> <984ql60tdk@news1.newsguy.com> <tac37rqd7o2h64@corp.supernews.com>
Message-ID: <985ipk01otv@news1.newsguy.com>

"Shonn Gilson" <shonn at shonn.com> wrote in message
news:tac37rqd7o2h64 at corp.supernews.com...

    [about using buffer for binary-blob args to/from COM methods]

> Perfect. It works for both Response.BinaryWrite and doing an AppendChunk
> with ADO. I noticed that it was coming in as a buffer from ADO but
> everything that I saw relating to buffers was that they were an
undocumented
> necessary evil.
> Thank you, that ends a lot of frustration.

You're welcome.  I've seen derogatory comments (and very little
doc) about buffer too, not sure why, but it sure IS necessary
if you want to exchange binary blobs to/from COM/Automation:-).


Alex





From dwig at advancedmp.net  Fri Mar 23 13:07:35 2001
From: dwig at advancedmp.net (Don Dwiggins)
Date: 23 Mar 2001 10:07:35 -0800
Subject: PyUnit in Python 2.1 standard library
In-Reply-To: "Don Tuttle"'s message of "Fri, 23 Mar 2001 01:50:23 GMT"
References: <3ABA8A23.85EA76CE@downright.com>
	<zZxu6.150312$__6.27726222@typhoon.southeast.rr.com>
Message-ID: <u7l1gcqag.fsf@advancedmp.net>

Don Tuttle writes:
> "Doug Fort"
>> Just recieved word that Steve Purcell has checked PyUnit  into the
>> standard library.  Aside from being a useful tool to me PyUnit typifies
>> the kind of development that ESR advocates in 'The Cathedral and the
>> Bazaar'.   A project started to 'scratch an itch', then shared with the
>> communityto evolve and grow.  None of the stuff I  work on seems to come
>> out so clean.  Congratulations Steve!  PyUnit is an example to us all.

> So what kind of itch does PyUnit scratch?

Taking the question at face value, check out
http://members.pingnet.ch/gamma/junit.htm and http://c2.com/cgi/wiki?search=UnitTest

-- 
Don Dwiggins                    "Solvitur Ambulando"
Advanced MP Technology
dwig at advancedmp.net




From amitp at Xenon.Stanford.EDU  Sun Mar 18 12:13:21 2001
From: amitp at Xenon.Stanford.EDU (Amit Patel)
Date: 18 Mar 2001 17:13:21 GMT
Subject: Who's minister of propaganda this week?
References: <B98041329020E007.7DA84CAD9D0562AB.817BBC572F480869@lp.airnews.net> <98nu1m$12e@dispatch.concentric.net> <3AAFA861.C773D977@san.rr.com>
Message-ID: <992qbh$et7$1@nntp.Stanford.EDU>

 Darren New  <dnew at san.rr.com> wrote:
| Phlip wrote:
| > Ruby trumps Python in the simple matter of
| > closures (imagine my shock when I discovered Python didn't have them). 
| 
| Can someone summarize the important difference between closures and object
| instances? Other than convenience, I mean, it would seem that something like
| 
| class yadda:
|   def __init__(self, a, b):
|     self.a = a ; self.b = b
|   def __call__(self, x, y):
|     return (self.a + x * self.b * y)
| 
| would seem to be quite a lot like a closure with a and b bound and x and y
| free?

Don't underestimate convenience.  :-) If you had to declare each
string value in your program as a class like the above, it would make
strings less useful in a practical sense although they'd still be
equivalent in a theoretical sense.  C++ suffers from this when it
comes to container types -- building a list or array takes a lot more
syntax than just "[1, 2, 3]" so people use them less casually.
Function objects in C++ are even worse, although perhaps the "Lambda
library" (http://lambda.cs.utu.fi/) may help.



       - Amit
-- 
--
Amit J Patel, Computer Science Department, Stanford University
http://www-cs-students.stanford.edu/~amitp/


From fgeiger at datec.at  Tue Mar 13 12:35:33 2001
From: fgeiger at datec.at (Franz GEIGER)
Date: Tue, 13 Mar 2001 18:35:33 +0100
Subject: Want to catch pywintypes.api_error
References: <98kftn$jf9$1@newsreaderm1.core.theplanet.net> <3AAE208D.4010103@ActiveState.com>
Message-ID: <98llta$2ea$1@newsreaderg1.core.theplanet.net>

Thank you, Mark!

How could I have figured out this by myself? I did a text search for
api_error over my ActivePython distribution, w/o success. Is it burrie din
those pyd-files?

Regards
Franz


"Mark Hammond" <MarkH at ActiveState.com> wrote in message
news:3AAE208D.4010103 at ActiveState.com...
> Franz GEIGER wrote:
>
> > When I start win32api.FindFiles(filespec) with a non-existing filespec
it
> > throws an exception of class pywintypes.api_error. But a "except
> > pywintypes.api_error, e:" does not work. Using "except Exception, e:"
does
> > the job, because Exception is the base class. But how can I catch
api_error?
> > What do I miss?
> >
>
> win32api.error
>
> Mark.
>




From n8gray at caltech.edu.is.my.e-mail.address  Tue Mar  6 03:19:20 2001
From: n8gray at caltech.edu.is.my.e-mail.address (Nathaniel Gray)
Date: Tue, 6 Mar 2001 00:19:20 -0800
Subject: I come to praise .join, not to bury it...
References: <mailman.983756766.27347.python-list@python.org> <97vnu3076q@news2.newsguy.com> <slrn9a89dk.2rm.hzhu@rocket.knowledgetrack.com>
Message-ID: <9826te$ra5@gap.cco.caltech.edu>

Huaiyu Zhu wrote:

> It is all nice and well to say that sep.join(list) is good for
> polymorphism, except that there are several practical annoyances:
> 
> 1. The word join can be either transitive or nontransitive:
>    sep joins list.
>    list joins with sep.
>    
>    Might sep.glue(join) be clearer?  Well, compatibility with str.join?

I actually always liked ":".joins(list), which you used above.  It's much 
harder to get backwards than "join".  Alas, history has a way of trumping 
clarity, and if the name changed from "join", to "joins" during the 
transition to string methods, just as many people would be complaining 
about the switch as now complain about sep.join(list).

-n8

-- 
_.~'^`~._.~'^`~._.~'^`~._.~'^`~._.~'^`~._
             Nathaniel Gray
   California Institute of Technology
     Computation and Neural Systems
     n8gray <at> caltech <dot> edu
_.~'^`~._.~'^`~._.~'^`~._.~'^`~._.~'^`~._



From aleaxit at yahoo.com  Thu Mar 15 06:37:18 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 15 Mar 2001 12:37:18 +0100
Subject: Who's minister of propaganda this week?
References: <B98041329020E007.7DA84CAD9D0562AB.817BBC572F480869@lp.airnews.net> <98nu1m$12e@dispatch.concentric.net> <660ACCA1E4816F7E.C359CDF4168EF3FB.3B6B05538D619770@lp.airnews.net> <3AAFA13D.76E32C3F@divalsim.it> <98oa4o01i4v@news2.newsguy.com> <mailman.984619039.7856.python-list@python.org>
Message-ID: <98q9h60122v@news2.newsguy.com>

"Michael Chermside" <mcherm_python at yahoo.com> wrote in message
news:mailman.984619039.7856.python-list at python.org...
> Alex Martelli wrote:
>          ... [snip]...
>   > So, all the compile-time checking is buying is catching (a small
>   > subset of) the errors that would be caught in testing anyway, a
>   > little bit earlier (thus, a little bit cheaper) -- it's never the
>   > case that one has to write a test which would not be needed at
>   > all if type-checking was static, since "the object has the right
>   > type" is a small subcase of "the object _behaves_ per the specs".
>   >
> I'm really not sure I see it this way. If the method  foo(x) is known to
> take a FancyDateObject

Assume that is an abstract interface (no gain in terms of
functionality assurance if it's concrete) and (without loss
of generality) that it has two methods First and Second that
foo uses (it may have others that foo is ignoring, but they
don't affect the following argument).  OK so far?  Good.

> there are three kinds of errors we could make. One is that foo() is
> written badly so it
> doesn't do what it's supposed to. The unit tests of foo() need to guard
> against this.

Right.  Specifically, they'll test the combinations of calls
to First and Second methods of the x argument that foo needs
to actually perform, say in certain cases First only, in
others Second only, in others yet, First then Second (if foo
never needs to call Second before, and First after, on its
argument, then its unit-tests will not exercise that path).

> Another possible error is that FancyDateObject isn't written properly.

As I assumed this is an interface, let's say you're talking
about some specific implementation thereof -- FDO_impl1, say.

> The unit tests
> of FancyDateObject need to guard against this.

Right again -- specifically, they'll test that FDO_impl1's
implementations of First and Second support the call patterns
that the specification demands.  For example, if that is what
the specs say, the implementation will work fine if only First
is called, or if only Second is called, or if Second is called
before and _then_ First is called -- 'First before, Second after'
may not be in the specs and thus doesn't get exercized by unit
tests.

> And the third type of
> error is that
> that somewhere where we CALL foo(), we might pass it a DateObject
> instead... or
> even a String... which will cause it to perform wrong. To guard against

More generally: we might erroneously pass to foo some object that
does not even _implement_ First and Second methods with acceptable
signatures, or some that _does_ implement the methods with signatures
that appear good BUT not with the semantics that foo needs -- e.g.,
it does not let First be called earlier and Second later (because
there is a mismatch in semantics specs between the specs that foo
requires and the specs that x actually ensures -- just as in the
other cases, regarding existence or signature of the methods).

No compiler wards you at compile-time against all, or even _most_,
errors of this very common kind.  (The existence of more than one
method is not needed to have this behavior -- just a single method
suffices to exhibit this error, e.g. it could take an argument i
with a prereq of i>23 and be called with i==23 by foo; or foo might
call it 7 times when the semantics specify it must be called no
more than 6 times; etc, etc).

Another way to express it: an interface is not just, not even
_mostly_, about existence and signature of methods -- it's mostly
about prereq's, post-conditions, and invariants; and nobody can
check those at compile-time in enough cases to make a difference
to your software's reliability.

So, statically-typed languages make a LOT of the not-very-important
issues of method-existence and signature -- because those issues
are THE ones they can check statically, not because their importance
(wrt the importance of real semantics issues, the programming-by-
contract parts of the interface) warrants special attention.  It's
like the drunkard who was looking for his housekeys at one end of
the street, opposite to the end where he had dropped them, because
the end he was searching at was the one that a streetlamp lit...


> this in a
> dynamically typed language, we have to write unit tests for *every
> single place*
> that we call foo(). Of course, we'd be writing unit tests for those
> functions anyway,
> but we won't be able to assume that foo() works properly, and will need
> extra tests to ensure this.

You can never 'assume that foo(x) works properly' for a given x
in an untested case -- and syntactic-level compatibility between
the methods x offers (& their signatures), and the ones foo
requires of its arguments, doesn't buy you much, since the likely
and troublesome errors are with contract-expectations mismatches.

If the implementation of foo and its call are inside the same
component, then unit-tests should exercize the relevant paths
(or else the component is being released in a state of incomplete
testing -- thus, dubious reliability, whether with or without
static typechecks).

If the implementation of foo and its call are in different
components, then you have a system-integration problem (again,
one that remains independently of static type-checking) and
thus more 'strategic' kinds of troubles.  You *STILL* need
'extra' tests to ensure the actually-implemented semantics
of a given actual argument and the ones foo requires of its
formal argument match -- for all distinct cases that occur
in the (system-level) acceptance criteria tests.  If then, in
later system operation, you meet a failing case that was not
tested (makes no difference whether the mismatch is in
method existence, signature, or semantics), then your acceptance
tests were insufficient -- and static checking would not have
made them sufficient.


> In a staticly typed language, we still need the unit tests for cases 1
> and 3, but the
> third type of error is caught by the compiler.

Not in most cases of interest, no.

> And in my mind, ANY TIME that
> I can have a machine do my work for me it's better... I can be lazy, and
> the machine
> never gets tired after a long day and forgets to test sometimes. Of

Extra checks (even if theoretically redundant) are good insurance EXCEPT
where they breed exactly this kind of 'complacency'.  Release procedures
MUST 'never .. forget to test' either -- and it's not that hard a problem
to setup your build/release environment so that a machine ensures this.

> course, I can
> put an assert at the top of foo() which asserts that x is of type
> FancyDateObject,
> but if I always assert the types of my arguments then I'm basically
> using a staticly typed language.

If your 'assert ... is of type' actually runs a sensible albeit small
'type'-testing procedure, which exercizes the whole contract that the
interface implements, then you have gone statically typed languages
one better -- unfortunately, this is most often impractical (as such
infinitely-repeated tests are far too slow, can't be made non-invasive,
etc, etc).  And it doesn't buy you all that much either (a bit more
than just statically checking types, but not all that much more) --
you still need to have *exercised* the call-cases of interest.

Asserting a single, specific, invariable concrete type would be on
a different plane -- perhaps feasible (at non-inconsiderable cost)
for some within-the-component cases (where you can commit to never
needing ANY polymorphism EVER), definitely unfeasible across any
component-boundary (cfr. Lakos' "Large Scale" book, again -- it's
still the best treatment I know of dependency management in large
scale software development, and, in particular, of the inevitability
of purely abstract interface-classes across component boundaries
to manage those dependencies; Martin's articles, which can be found
on his objectmentor site, are more readable, although not quite as
deep and extensive as Lakos' big book of course).


> There are times when a dynamically typed language is more flexible, and
> it's certainly
> nice not to have to declare everything just to specify its type, but
> there ARE
> advantages to static typing, and this, I believe, is the biggest one.

We agree that the (redundant, but earlier) checks performed by the
statically-checking compiler are the 'biggest' (least small?-)
advantage of said compiler (well, apart from speed issues, which
may at times be paramount for certain well-identified components).

We disagree on how big that 'biggest' is.  I estimate that (speed
apart) this specific advantage may buy me about a 5% productivity
improvement -- so I agree it's an advantage, and I agree it's
larger than any other advantages of static checks (none of which,
it seems to me, may make even a 1% further improvement even when
all taken together -- again, speed of resulting code apart), but
I don't think it's worth anywhere near the _bother_ (productivity
impact, _negative_ improvement) of the contortions I have to
perform to satisfy the checks (which cost me _at least_ 10% of
my sw-lifetime-coding-productivity, even in the cases where I'm
least interested in specifically taking advantage of dynamism).

I come to these (tentative) conclusions after a lifetime spent
working _mostly_ in statically-typed languages -- because the
performance characteristics of the machines I was targeting just
didn't afford me the luxury of doing otherwise (when they DID,
I repeatedly tried out dynamically typed languages, and, over
and over again, I was tickled pink at how well they worked --
most people are surprised when they learned how much of the
'background processing' of the programs I was doing in the
mid-80's [on IBM mainframes] was done in Rexx, but then, they
are equally surprised at how fast and reliably I delivered:-).

Today, I'm still relying mostly on C++ to earn my daily bread
(3-D modeling &c being pretty compute-intensive even by today's
standards -- and 3D for mechanical engineering is that what
we're mostly doing here at my current employer), but more and
more dynamically typed code 'sneaks in' (thanks be!-)...


Alex





From bsass at freenet.edmonton.ab.ca  Sat Mar 31 15:51:51 2001
From: bsass at freenet.edmonton.ab.ca (Bruce Sass)
Date: Sat, 31 Mar 2001 13:51:51 -0700 (MST)
Subject: list vs tuple
In-Reply-To: <uGqx6.1052$p5.3997@news1.rivrw1.nsw.optushome.com.au>
References: <hGlw6.782$p5.3210@news1.rivrw1.nsw.optushome.com.au><3AC20CD3.265A2D94@alcyone.com><ru69cto61jtmrqsf809e38hb3pbm2lhi7g@4ax.com><bC1x6.4319$4N4.905946@newsc.telia.net><71q9ct4le63anefmiajj5cbh3mg04qu1ga@4ax.com><mailman.985991202.5578.python-list@python.org><7Bfx6.989$p5.3791@news1.rivrw1.nsw.optushome.com.au><Ukhx6.4421$4N4.950603@newsc.telia.net>
 <mailman.986049281.32365.python-list@python.org>
 <uGqx6.1052$p5.3997@news1.rivrw1.nsw.optushome.com.au>
Message-ID: <Pine.LNX.4.33.0103311326520.7534-100000@bms>

On Sat, 31 Mar 2001, deadmeat wrote:

> > - There is an difference between "copy" and "deep copy". What happens if
> > you make copy a structure that contains pointers in it?
>
> then you got a problem :)
>
> but thats not the point - which when the statement a = b copies the data
> across for one type it's expected to do it for others.  It doesn't, so it's
> inconsistant, contrary to the original claim.

But that is not what is happening!

	a = 1

points "a" to an object
(which just happens to have the name "1" and an integer value of 1)

	b = a

points "b" to the same object "a" points to (not to "a" itself)

	a = 2

points "a" to a different object,
but "b" is unaffected because the only relationship it has with "a"
is/was the fleeting one of having referred to the same object at some
point during the program's execution.

You seem to be thinking of "a = 1" as a statement which creates an
integer object with the value of 1 and the name "a", when it is
actually just creating a pointer labeled "a" to an object whose value
is int(1).  It may seem subtle, but it is a big distinction, and is
why your int vs list example a few posts back was flawed... in the
first case (integers)  you were changing what objects the labels were
refering to, in the second example (lists) you were modifying the
contents of an object referred to by more than one label -- apples and
oranges, eh.

HTH


- Bruce




From sk+10 at cs.brown.edu  Thu Mar  8 17:32:47 2001
From: sk+10 at cs.brown.edu (Shriram Krishnamurthi)
Date: 08 Mar 2001 17:32:47 -0500
Subject: free high school teacher summer workshops
Message-ID: <w7dd7brki28.fsf@cs.brown.edu>

[Since there is some interest in educational issues in the Python
 community, this workshop announcement should be of interest.  This
 project shares many goals with some Python educators, but has a much
 stronger pedagogic slant.]

     	   TeachScheme! -- The Pre-AP Curriculum
     	   =====================================

The National Science Foundation has awarded a large grant to the
TeachScheme! Project.  Over the past few years, the Project has
pioneered a new approach to problem solving with a minimum of syntax
and strong methodological support from software.  The innovative
program has attracted widespread attention and rapid growth.  Several
teachers now use TeachScheme! as their pre-AP (or sole) curriculum.
The Project has also developed a gentle and direct transition path to
Java, suitable for all students.

The Project provides

- free workshops
- free on-line book (also available in print from the MIT Press)
- free solution sets and guides
- free innovative software that grows with students' knowledge
- free round-the-clock consultation

Using the grant, the Project will pay flight, hotel, meals, and
stipends for up to 100 qualified participants per year.  We welcome
pairs of teachers from the same school.  Teachers who wish to earn
graduate credit may apply for a tuition waiver at the participating
institutions.

To learn more about the Project, please visit

    http://www.teach-scheme.org/

The workshops will take place in Boston, MA (Northeastern University);
Garden City, Long Island, NY (Adelphi University); Providence, RI
(Brown University and Worcester Polytechnic Institute); and Salt Lake
City, UT (University of Utah), in July and August, 2001.


From stephen_purcell at yahoo.com  Sun Mar 18 06:35:11 2001
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Sun, 18 Mar 2001 12:35:11 +0100
Subject: Help with 500 error on cgi script
In-Reply-To: <5739bt04e50ku01rn3dptgqch3l07p2cnf@4ax.com>; from sheila@spamcop.net on Sun, Mar 18, 2001 at 10:29:08AM +0000
References: <quv8bto7r8s6f1tckc7qpmuir0kvufp1nb@4ax.com> <mailman.984909623.8814.python-list@python.org> <5739bt04e50ku01rn3dptgqch3l07p2cnf@4ax.com>
Message-ID: <20010318123511.A9960@freedom.puma-ag.com>

Sheila King wrote:
> 	print "<H1>Good Result!</H1>

Closing quote is missing.

Try running the scripts from the command line first to weed out such syntax
errors.

-Steve

-- 
Steve Purcell, Pythangelist
Get testing at http://pyunit.sourceforge.net/
Any opinions expressed herein are my own and not necessarily those of Yahoo



From jcm at bigskytel.com  Wed Mar  7 19:07:48 2001
From: jcm at bigskytel.com (David Porter)
Date: Wed, 7 Mar 2001 17:07:48 -0700
Subject: Extracting cell contents from html tables
In-Reply-To: <3AA67658.B39580CB@comm.mot.com>; from stepheb@comm.mot.com on Wed, Mar 07, 2001 at 11:56:40AM -0600
References: <3AA67658.B39580CB@comm.mot.com>
Message-ID: <20010307170748.A4283@bigskytel.com>

* Stephen Boulet <stepheb at comm.mot.com>:

> Is there a module for extracting cell contents from html tables
> (preferably into an array of strings)?

This might be useful:
http://starship.python.net/crew/jjkunce/python/htmlTableParse.py


David



From dusty at strike.wu-wien.ac.at  Thu Mar  8 12:37:43 2001
From: dusty at strike.wu-wien.ac.at (Hermann Himmelbauer)
Date: Thu, 8 Mar 2001 18:37:43 +0100
Subject: Documentation/Examples about the htmllib?
Message-ID: <3aa7c372$0$22464@SSP1NO25.highway.telekom.at>

Hi,
I want to parse a htmlpage with python, so I thought using the htmllib 
would be good for that task.

In my html-page I have this tag:
<area href="/html/page.html?key" .... >

What I want to do is extract this link parameter "key".

It would be perfect if the htmllib would extract me all the <area ...> tags 
into a list so that I could simply find the rigth tag and extract the key.

I did manage to get data between a tag like <title> data </title> but could 
not get the data "in" the tag itself. Of course I could do this with a 
regular expression but I thought using this module would give me better 
results, what do you think?

Does anyone have a clue?

                Best Regards,
                Hermann

-- 
 ,_,
(O,O)     "There is more to life than increasing its speed."
(   )     -- Gandhi
-"-"--------------------------------------------------------------


From sholden at holdenweb.com  Tue Mar  6 19:03:12 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Tue, 6 Mar 2001 19:03:12 -0500
Subject: Open Source Organic Chemistry software project
References: <taaseuid59k047@corp.supernews.com>
Message-ID: <QYep6.29320$1D5.1247980@e420r-atl1.usenetserver.com>

Don't forhet to solicit the chemictry community, much of which won't read
c.l.py

not-done-organic-chemistry-in-thirty-five-years-ly y'rs  - steve

"mapMonkey" <jziniti at speakeasy.org> wrote in message
news:taaseuid59k047 at corp.supernews.com...
> Pythonistas,
>
> I am encouraging my current employer to GPL at least some of
> our code and I would like to see what kind of interest I can
> generate from the Python community ( everyone's favorite
> language ;).  The amount of interest is very likely to swing the
> vote ...
>
> The current code-base is geared to Organic Chemistry molecules
> and the IUPAC nomenclature system and it works quite well.  It
> was written entirely in Python, and entirely by me. My
> goals are to re-engineer this code for added flexibility, and to
> increase the scope of the model to include all (or most) of organic
> chemistry (reactions, submolecular thingies, etc.).
>
> Anyone? Anyone?
>
> Thanks for your time,
>
> John Ziniti




From fredrik at pythonware.com  Wed Mar 14 17:22:16 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 14 Mar 2001 22:22:16 GMT
Subject: tiny extension crashes
References: <Pine.SOL.4.30.0103141957240.10526-100000@mimosa.csv.warwick.ac.uk>
Message-ID: <saSr6.332$4d3.76176@newsb.telia.net>

John J. Lee wrote:
> static PyObject *iround(PyObject *self, PyObject *args) {
>     /* round to integer type, 0.49->0, 0.5->1 */
>     double number;
>     double *intp;
>
>     if (!PyArg_ParseTuple(args, "d", &number))
>         return NULL;
>     if (fabs(modf(number, intp)) >= 0.5) {
>         *intp >= 0 ? (*intp)++ : (*intp)--;
>     }
>     return Py_BuildValue("l", (long)*intp);

umm.  aren't you supposed to pass a pointer to an existing
double as the second argument to modf? (instead of an un-
initialized pointer...)

> static PyObject *iround(PyObject *self, PyObject *args) {
>     /* round to integer type, 0.49->0, 0.5->1 */
>     double number;
=     double intpart;
>
>     if (!PyArg_ParseTuple(args, "d", &number))
>         return NULL;
=     if (fabs(modf(number, &intpart)) >= 0.5) {
=         intvalue >= 0 ? intvalue++ : intvalue--;
>     }
=     return Py_BuildValue("l", (long) intvalue);

Cheers /F




From neilh at scintilla.org  Fri Mar 23 15:18:56 2001
From: neilh at scintilla.org (Neil Hodgson)
Date: Fri, 23 Mar 2001 20:18:56 GMT
Subject: Install Python on Apache
References: <99fph5$a2n$1@rex.ip-plus.net>
Message-ID: <QcOu6.19903$992.122447@news-server.bigpond.net.au>

Sharp:

> [Fri Mar 23 15:54:50 2001] [error] [client xxx.xxx.xxx.xxx] Premature end
of
> script headers: c:/web/cgi-bin/python/python.exe
>
> the script is looking like this
>
> #!c:/web/cgi-bin/Python/python.exe
> ...

   I had success with:
#! c:\Program Files\Python\Python
   so try adding the space, using backslashes or dropping the .exe

   Neil




From parkw at better.net  Wed Mar  7 18:28:59 2001
From: parkw at better.net (William Park)
Date: Wed, 7 Mar 2001 18:28:59 -0500
Subject: I come to praise .join, not to bury it...
In-Reply-To: <986f2h$1legl$1@fido.engr.sgi.com>; from pj@sgi.com on Wed, Mar 07, 2001 at 11:09:05PM +0000
References: <kuofvfjb77.fsf@lasipalatsi.fi> <000001c0a64c$497df250$a9807018@cr706570a> <mailman.983997159.4842.python-list@python.org> <986f2h$1legl$1@fido.engr.sgi.com>
Message-ID: <20010307182859.A30508@better.net>

On Wed, Mar 07, 2001 at 11:09:05PM +0000, Paul Jackson wrote:
> Time to get used to simply:
> 
>     ",".join(x)

The best way to remember the syntax is
    Using ',', join the following ['...', '...', ...].

---William Park, Open Geometry Consulting, Linux/Python, 8 CPUs.



From flognat at flognat.myip.org  Fri Mar  9 17:35:44 2001
From: flognat at flognat.myip.org (Andrew Markebo)
Date: Fri, 09 Mar 2001 22:35:44 GMT
Subject: I get lost with "signal"
References: <mailman.984104651.27664.python-list@python.org>
Message-ID: <m3d7bqfu6j.fsf@master.athome>

By doing 

import signal
dir(signal) 

you can see which signals are there.. 

Or.. 

Rehack the ftp and socket module to use unblocked sockets, and do a
select call when waiting for the server to connect, I have done such
hacks for http.

Shouldn't be too hard copying ftplib.py and socket.py and mess around
with setblocking (before doing connect), and select.. for the first
queries to the server.

        /Andy

/ "Andrea Ciuffoli" <ACIUFFO1 at irf.com> wrote:
| Hi,
| I'm trying to use "signal" library to control that an FTP connection is established within a specified time (e.g. 3 sec) but I got an Attribute error: SIGALRM.
| Is that module supported by Windows NT and 98?
| If not, is there another method to control that the connection is established without problem? How can I know if the server is down?
| Thanks,
| Andrea.


From tomega at earthlink.net  Tue Mar 13 18:07:51 2001
From: tomega at earthlink.net (Simon Kesenci)
Date: Tue, 13 Mar 2001 23:07:51 GMT
Subject: Missing libX11.a on AIX
References: <AEFF34C2E454D311993B00902745E4AB05498C6B@news.idx.com>
Message-ID: <bLxr6.8496$6p5.742977@newsread1.prod.itd.earthlink.net>

Open the Setup file in the Modules directory, and comment out the lines for 
the following modules:

gl
sv
tkinter

The section for the last module traverses several lines.  Comment them all 
out and recompile.

--S.

Dan Moran wrote:

> Hi Folks.
> 
> When running python on a AIX box (4.3.3), I get the following message:
> 
> exec(): 0509-036 Cannot load program ./python because of the following
> errors:
>         0509-150   Dependent module libX11.a(shr4.o) could not be loaded.
>         0509-022 Cannot load module libX11.a(shr4.o).
>         0509-026 System error: A file or directory in the path name does
>         not
> exist.
> 
> This box was not setup with the X11 libraries.
> Can Python be compiled to exclude its dependency on X11?
> 
> ::Dan Moran
> 
> 
> 
> 
> 
> 



From jjlucsy at concentric.net  Fri Mar 23 16:10:02 2001
From: jjlucsy at concentric.net (Joel Lucsy)
Date: Fri, 23 Mar 2001 16:10:02 -0500
Subject: Python Script to dial a VPN connection, automate VPN routing table updates.
References: <FCNu6.66711$A6.13912607@news4.rdc1.on.home.com>
Message-ID: <CYOu6.24703$G72.558271@e420r-atl1.usenetserver.com>

try "rasdial /?" at a command prompt

--
Joel Lucsy (jjlucsy at concentric.net)

"Warren Postma" <warren-postma at home.com> wrote in message
news:FCNu6.66711$A6.13912607 at news4.rdc1.on.home.com...
> Sorry for the nearly-off-topic nature of this post, but ....
>
> I work from home now, and there's something in Windows 2000 that's
> annoyingly manual:
>
> I have a permanent internet connection via @home cable modem, and I "dial"
> through that to create a VPN connection to the office.  The problems are
> two:
>
> (1) Windows won't accept or save my userid, password and the domain name,
so
> I have to enter it manually.
>
> (2) After connecting, I have two default gateways in my routing tables
> (ROUTE PRINT) and I need to update it so only the traffic that is destined
> to the office goes through that interface.  The IP addresses I get given
are
> different each time, making a batch file insufficient, so voila, a perfect
> use for python.
>
> So after solving #2 above in Python, I wondered, does anyone know how, in
> Python win32 extensions, one might dial an RAS connection, and if it's
> possible to do so  through the win32 API somewhere or through COM? Then I
> could solve #1 above as well.
>
> Anyone else think this might make a nice little Python applet if I build
in
> some more VPN/RAS stuff and maybe even Gui-fy it a bit in wxPython or
> tkInter.
>
> Warren Postma
>
>
>
>





From aahz at panix.com  Mon Mar 19 11:29:09 2001
From: aahz at panix.com (Aahz Maruch)
Date: 19 Mar 2001 08:29:09 -0800
Subject: Killing threads
References: <mailman.985014274.27176.python-list@python.org>
Message-ID: <995c4l$gn7$1@panix3.panix.com>

In article <mailman.985014274.27176.python-list at python.org>,
Dave Brueck <dbrueck at edgix.com> wrote:
>Aahz:
>>
>> I'm confused.  Is there some reason you're not using Queue.Queue()?
>
>Queue is overkill for simple one-to-one workflows - it's built to be
>multi-consumer and multi-producer so every get/put/check requires acquiring
>and releasing the Queue's mutex (granted, in most cases that probably won't
>make too big of a different, but...).

That's partly true; there are still single producer/consumer cases where
you need a mutex.  The only reason you can get away with it in Python is
because the GIL (global interpreter lock) does a lot of the work for
you.  The flip side is that the GIL means a mutex really adds little
overhead in the single producer/consumer case.

>Anyway, the point here is killing separate threads and my WorkQueue class
>was that was the first example that came to mind. Note that whether or not
>you use Queue.Queue makes little difference in solving this problem:

Not overall, no, but it does mean you don't have to muck with mutexes.
I'm a big believer in simple code.  One problem with the code you posted
is that it has a polling loop (with time.sleep()) in it; that's not good
for efficiency.  Use another queue!  ;-)

I'm also not a big fan of dispatching threads on functions/methods.
It's almost always clearer to just subclass from threading.Thread.
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

There's a difference between a person who gets shit zie doesn't deserve
and a person who gets more shit than zie deserves.  --Aahz


From ws-news at gmx.at  Fri Mar  2 10:29:21 2001
From: ws-news at gmx.at (Werner Schiendl)
Date: Fri, 2 Mar 2001 16:29:21 +0100
Subject: Communicating with serial port
References: <ndNn6.5161$t21.152593@news3.oke.nextra.no>
Message-ID: <983547237.143131@newsmaster-04.atnet.at>

last time i needed serial comm's on Windows (NT in my case) i opened the
interface as follows

        import win32file

        portname = 'COM2'
        paramstring = "baud=38400 parity=N data=8 stop=1 xon=on"
        self._hFile = win32file.CreateFile(portname, win32file.GENERIC_READ
| win32file.GENERIC_WRITE,
                                           0, None, win32file.OPEN_ALWAYS,
0,0)
        if paramstring:
            dcb = win32file.GetCommState(self._hFile)
            win32file.BuildCommDCB(paramstring, dcb)
            win32file.SetCommState(self._hFile, dcb)

afterwards you can use win32file.WriteFile and win32file.ReadFile to access
the port
dont forget to close the handle with win32file.CloseHandle

i'm afraid i do not know your library
but maybe you come along with that
there is also a demo that comes with the win extensions for serial comm's

hth
werner


Peter Blom <pet-blo at online.no> wrote in message
news:ndNn6.5161$t21.152593 at news3.oke.nextra.no...
> I shall write a python program for downloading data from a memory card in
a
> measuring instrument, via a serial port.
> For this purpose, I have downloaded SioModule.zip - a serial comm
extension
> for windows 95, upgraded to Python 1.5.1.
> This does not seem to work - when importing the module 'serial.py', my
> system hangs (Windows 98, Activestate Python 2.0).
> Could anybody give a hint?
>
> Peter (a newbie)
>
>




From akuchlin at mems-exchange.org  Mon Mar 19 17:27:58 2001
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 19 Mar 2001 17:27:58 -0500
Subject: Python-related standards
References: <3dae6h21h8.fsf@ute.cnri.reston.va.us> <mailman.985022189.18857.python-list@python.org>
Message-ID: <3d7l1l1jip.fsf@ute.cnri.reston.va.us>

Moshe Zadka <moshez at zadka.site.co.il> writes:
> FWIW, these could be "informational" PEPs.

OK.  Then the only thing missing from PEP 1 are 'Supersedes' and
'Superseded-by' headers, for replacing obsolete versions of documents.
I'll send Barry a note.

--amk


From mcalla at home.com  Thu Mar 22 16:09:33 2001
From: mcalla at home.com (Mike Callahan)
Date: Thu, 22 Mar 2001 21:09:33 GMT
Subject: Catching exceptions in Tkinter's mainloop
References: <%dws6.9947$Q47.2723350@news1.rdc1.tn.home.com> <drMs6.32482$HR6.3694604@nnrp4.clara.net> <owzt6.21061$Q47.5935326@news1.rdc1.tn.home.com> <6QNt6.37618$g63.4509118@nnrp3.clara.net> <Qydu6.26214$Q47.7466405@news1.rdc1.tn.home.com> <oWru6.48063$HR6.5435948@nnrp4.clara.net>
Message-ID: <hStu6.28419$Q47.7934789@news1.rdc1.tn.home.com>

OK Richard,

I will do some more digging on this end. Thanks for all your help. By the
way, do you know the difference between mainloop() and mainloop(1)? What is
the optional argument for?

Mike C.

"Richard Townsend" <richard at starfighter.freeuk.com> wrote in message
news:oWru6.48063$HR6.5435948 at nnrp4.clara.net...
> "Mike Callahan" <mcalla at home.com> wrote in message
> news:Qydu6.26214$Q47.7466405 at news1.rdc1.tn.home.com...
> > Richard,
> >
> > On my system, the exception is captured and printed in test.log but
> mainloop
> > is not exited. Is this what happens on your system?
> >
> > Mike C.
> >
>
> I've tested it on Win98 and Linux (both using Python2.0 & Tk8.3) and it
> definitely exits the mainloop in both cases. I have no idea why it doesn't
> do the same for you...
>
> Richard
>
>
>




From qrczak at knm.org.pl  Mon Mar  5 11:26:20 2001
From: qrczak at knm.org.pl (Marcin 'Qrczak' Kowalczyk)
Date: 5 Mar 2001 16:26:20 GMT
Subject: Static class methods
References: <rdzo6.154$gc4.80445@news.pacbell.net> <ta5ka1ks0buc3b@corp.supernews.com> <97ul9k$n2l$02$1@news.t-online.com>
Message-ID: <slrn9a7fhc.c9f.qrczak@qrnik.zagroda>

Mon, 5 Mar 2001 01:06:06 +0100, Sven Peters <svpeters at t-online.de> pisze:

> You CAN have class methods:
> def bert:
>         print "Bert"
> class.method = bert
> class.method()
> "Bert"

You can't. The first line is a syntax error. And the following:

class C: pass
def bert(): print "Bert"
C.method = bert
C.method()

produces an error (unbound method must be called with class instance
1st argument).

I proposed a language change which allows class methods in a thread
titled "An alternative approach to bound methods" two weeks ago.
With this proposal the above would work, as well as putting the
function definition directly in the class.

-- 
 __("<  Marcin Kowalczyk * qrczak at knm.org.pl http://qrczak.ids.net.pl/
 \__/
  ^^                      SYGNATURA ZAST?PCZA
QRCZAK


From aahz at panix.com  Thu Mar 29 16:17:44 2001
From: aahz at panix.com (Aahz Maruch)
Date: 29 Mar 2001 13:17:44 -0800
Subject: Evaluating *EVERYTHING*
References: <eBLw6.3059$j_3.1383318@typhoon2.ba-dsg.net>
Message-ID: <9a08po$icf$1@panix6.panix.com>

In article <eBLw6.3059$j_3.1383318 at typhoon2.ba-dsg.net>,
David Allen <mda at idatar.com> wrote:
>
>How do I get the interpreter to 'validate' code by
>trying to evaluate *everything* so I can catch bugs?

Try http://www.chat.net/~jeske/Projects/PyLint/ or
http://www.lfw.org/python/pylint
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

"Boost the stock market -- fire someone"


From xyzmats at laplaza.org  Mon Mar  5 09:42:16 2001
From: xyzmats at laplaza.org (Mats Wichmann)
Date: Mon, 05 Mar 2001 14:42:16 GMT
Subject: Can anyone recomend a good intoduction to C...
References: <3aa366c6_2@nnrp1.news.uk.psi.net>
Message-ID: <3aa3a3f0.36440038@news.laplaza.org>

On Mon, 5 Mar 2001 10:14:35 -0000, "Simon Brunning"
<sbrunning at trisystems.co.uk> wrote:

>... for a Python programmer?
>
>I'm going to have to bite the bullet and learn a bit of C. I'd at
>least like to learn the *right* way to do it.

My advice is similar to the one that gets given to prospective Python
programmers:  start with the source - K&R (Kernighan & Ritchie). It's
kind of similar to the Python tutorial in ways: concise but completely
to the point, from people who know why the language was written that
way. This is what everyone learned the language from in the early
days. Stay away from those 1000-page doorstops with CDroms included.


Mats Wichmann

(Anti-spam stuff: to reply remove the "xyz" from the
address xyzmats at laplaza.org. Not that it helps much...)


From pj at sgi.com  Fri Mar  2 17:02:27 2001
From: pj at sgi.com (Paul Jackson)
Date: 2 Mar 2001 22:02:27 GMT
Subject: Unresolved symbols in MSVCRT.DLL when running a py2exe packaged program.
References: <mailman.983556668.16192.python-list@python.org>
Message-ID: <97p59j$7ulbj$1@fido.engr.sgi.com>

Tim wrote:
> My understanding is that SFP isn't so easily fooled; e.g.,
>                                                           
>     http://support.microsoft.com/support/kb/articles/Q253/5/71.ASP
>                                                                 
>     http://www.microsoft.com/hwdev/sfp/wfp.htm

For those coming from the Unix/Linux/... side, who don't know
what SFP is, and for whom the above Microsoft web pages are
completely blank (some incompatibility between Netscape and
many microsoft.com pages, so far as I can guess), SFP is System
File Protection.  The following web page is one of many that
describe it a little more:

    http://www.winmag.com/windows/winme/beta3/08.htm
-- 
                          I won't rest till it's the best ...
                          Manager, Linux System Software
                          Paul Jackson <pj at sgi.com> 1.650.933.1373


From costas at meezon.com  Fri Mar  2 21:09:50 2001
From: costas at meezon.com (Costas Menico)
Date: Sat, 03 Mar 2001 02:09:50 GMT
Subject: shortcut for 'self' wish
Message-ID: <3aa050ad.35416419@News.CIS.DFN.DE>

It would be nice if there was someway to cut down on the typing of
self.something inside def methods.

 VB and other languages you are allowed to define a 'WITH' prefix so
all you type is the dot followed by the method/property.E.g.

with this
  .edit
     < more stuff here >
  .update
endwith

Since Python lets you declare your own self it maybe better to change
the 'self' to a single letter standard.  Maybe m.property or m.method
would be a good one.

so instead of

def myMethod(self, arg1, arg2):

we could have

def myMethod(m, arg1, arg2):

This would save a lot of typing and encourage using OOP more,

Any other ideas? Does anyone have experience with this?

Costas




From gregj at pobox.com  Thu Mar  8 02:17:17 2001
From: gregj at pobox.com (Gregory Jorgensen)
Date: Thu, 08 Mar 2001 07:17:17 GMT
Subject: Who am I: can a class instance determine its own name?
References: <mailman.984034150.9635.python-list@python.org>
Message-ID: <1mGp6.905$54.989@www.newsranger.com>

Class instances don't have names. Names are bound to class instances. Think
about this:

x = foo()
y = x

x and y are both bound to the same foo instance. What is the name of the
instance of foo that both x and y are bound to? The question doesn't make sense.
Not every reference has a name bound to it, an object can have different names
in different scopes, etc.


In article <mailman.984034150.9635.python-list at python.org>, Tim CHURCHES says...
>
>This is probably an elementary question and the answer is probably writ =
>large in multiple places in the Python documentation, but...
>
>...can an instance of a class determine the name of the variable to which =
>it is assigned? For example:
>
>###########################
>class Foo:
>    def whoami(self):
>         return "You are a Foo() but I do not know your name"
>
>FooBar =3D Foo()
>
>print FooBar.whoami()
>###########################
>
>How does one define the method whoami() so that it returns "FooBar"? This =
>sort of navel gazing is formally called introspection, I think (therefore =
>I am)?

Greg Jorgensen
Deschooling Society
Portland, Oregon, USA
gregj at pobox.com


From donn at oz.net  Thu Mar  1 01:06:51 2001
From: donn at oz.net (Donn Cave)
Date: 1 Mar 2001 06:06:51 GMT
Subject: d = {}; d[0:1] = 1; d[0:1] = 2; print d[:]
References: <mailman.983419142.24027.python-list@python.org>
Message-ID: <97kotr$su9$0@216.39.151.169>

Quoth Robin Thomas <robin.thomas at starmedia.net>:
| Using Python 2.0 on Win32. Am I the only person to be depressed by the 
| following behavior now that __getitem__ does the work of __getslice__?
|
| Python 2.0 (#8, Oct 16 2000, 17:27:58) [MSC 32 bit (Intel)] on win32
|  >>> d = {}
|  >>> d[0:1] = 1
|  >>> d
| {slice(0, 1, None): 1}
|
| And then, for more depression:
|
|  >>> d[0:1] = 2
|  >>> d
| {slice(0, 1, None): 1, slice(0, 1, None): 2}
|
| And then, for extra extra chagrin:
|
|  >>> print d[0:1]
| Traceback (innermost last):
|    File "<pyshell#11>", line 1, in ?
|      d[0:1]
| KeyError: slice(0, 1, None)

If it helps, you ruined my day.

| So, questions:
|
| 1) Is this behavior considered a bug by the BDFL or the community at large? 
| If so, has a fix been conceived? Am I re-opening a long-resolved issue?
|
| 2) If we're still open to proposed solutions, which of the following do you 
| like:
|
|     a) make slices hash and cmp as their 3-tuple (start,stop,step),
|        so that if I accidentally set a slice object as a key,
|        I can at least re-set it or get it or del it :)
|
|     b) have dict.__setitem__ expressly reject objects of SliceType
|        as keys, raising your favorite in (TypeError, ValueError)

I think we might be able to do better.  I hacked in a quick fix
in ceval.c that looks to me like it has the desired effect without
closing the door to intentional slice keys (however unlikely.)

    Python 2.1a2 (#1, Feb 28 2001, 21:53:26)
    [GCC egcs-2.91.66 19990314 (egcs-1.1.2 release)] on netbsd1
    Type "copyright", "credits" or "license" for more information.
    >>> d = {}
    >>> d[1:4] = 7
    Traceback (most recent call last):
      File "<stdin>", line 1, in ?
    TypeError: dict object doesn't support slice assignment
    >>> f = slice(1, 4, None)
    >>> d[f] = 7
    >>> 

The change:
*** Python/ceval.c.dist Thu Feb  1 14:48:12 2001
--- Python/ceval.c      Wed Feb 28 21:52:55 2001
***************
*** 3168,3173 ****
--- 3168,3178 ----
        /* u[v:w] = x */
  {
        int ilow = 0, ihigh = INT_MAX;
+       if (u->ob_type->tp_as_mapping) {
+               PyErr_SetString(PyExc_TypeError,
+                       "dict object doesn't support slice assignment");
+               return -1;
+       }
        if (!_PyEval_SliceIndex(v, &ilow))
                return -1;
        if (!_PyEval_SliceIndex(w, &ihigh))

-End patch.

I don't know if this is the answer, but it at least demonstrates
that [:] and slice() can be separate issues at some level.

	Donn Cave, donn at oz.net


From kens at sightreader.com  Wed Mar 21 17:01:44 2001
From: kens at sightreader.com (Ken Seehof)
Date: Wed, 21 Mar 2001 14:01:44 -0800
Subject: Dynamically reloading extension modules
Message-ID: <007f01c0b252$852bb370$04090a0a@upcast.com>

Has anyone found a technique for dynamically unloading and
reloading an extension module?  I want to be able to modify,
build, and reload an extension module from within my program.

Unfortunately python apparently doesn't use FreeLibrary when
a module object is released.  Maybe I can do some kind of
hack to call FreeLibrary explicitly.  Is there a way to do this that
is portable?

----------------------------------------------------
Copyright (c) 2001 by Ken Seehof
This document may not be distributed, copied,
duplicated, or replicated in any form without
express written permission by Ken Seehof.
Express permission is hereby granted.
kens at neuralintegrator.com
----------------------------------------------------





From dsh8290 at rit.edu  Fri Mar 30 11:12:56 2001
From: dsh8290 at rit.edu (D-Man)
Date: Fri, 30 Mar 2001 11:12:56 -0500
Subject: Python 2.0 vs 68k Mac
In-Reply-To: <9a1agj$o55$1@news.cs.tu-berlin.de>; from malte.forkel@berlin.de on Fri, Mar 30, 2001 at 08:48:34AM +0200
References: <9a1agj$o55$1@news.cs.tu-berlin.de>
Message-ID: <20010330111256.D13498@harmony.cs.rit.edu>

On Fri, Mar 30, 2001 at 08:48:34AM +0200, Malte Forkel wrote:
| Has anyone build Python 2.0 for 68k Macs? Please cc: your reply to me
| (malte.forkel at berlin.de).

What OS is on your Mac m68k?  If it is MacOS I don't know.  If it is
Debian, then yes, but it is in 'unstable' (aka sid).

HTH,
-D




From idot at vt.edu  Sat Mar 31 23:09:52 2001
From: idot at vt.edu (Carl Banks)
Date: Sat, 31 Mar 2001 23:09:52 -0500
Subject: list vs tuple
References: <hGlw6.782$p5.3210@news1.rivrw1.nsw.optushome.com.au> <3AC20CD3.265A2D94@alcyone.com> <ru69cto61jtmrqsf809e38hb3pbm2lhi7g@4ax.com> <bC1x6.4319$4N4.905946@newsc.telia.net> <71q9ct4le63anefmiajj5cbh3mg04qu1ga@4ax.com> <mailman.985991202.5578.python-list@python.org> <n30cctcvjcgdir0b9ut7etjmq388gf0v9v@4ax.com>
Message-ID: <gm96a9.kv5.ln@127.0.0.1>

Mike Prager <mprager at alum.mit.edu> wrote:
> I repeat that I'm not expressing an opinion about what is
> "better," just that Python's behavior is somewhat
> idiosyncratic and should carry a caution label for beginners.

It's not idiosyncratic, though.

Someone with experience only in C, Fortran, Perl, etc. might *think*
that this is idiosyncratic out of ignorance, but as he embarks on
further exploration of programming languages, he finds that assigment
by reference is common, too, especially in interpreted languages.

In fact, there are all different sorts of assignment paradigms out
there.  Some languages have assignment by copy (C, Fortran, Ada, Perl,
Basic).  There's probably one that has assignment by deep copy
somewhere.  There's assignment by reference (Python, all dialects of
LISP that I am aware of, Smalltalk, Eiffel?).  Some have different
semantics for built in types and user-defined types (Java).  I
remember seeing a language that had different operators for reference
and copy assignment, but I forget what is was.  (BTW, should Python
ever feel a need to add assignment by reference, I pray this is how it
is done.)  Some languages even have customizable assigment semantics
(C++, Ada 95).  I'm sure other paradigms exist, also.

Nevertheless, since many people learning Python have experienced only
in C, a boldface notice or two is in order.  (I believe, if one reads
carefully, that it is mentioned in passing in the Python Tutorial.)


-- 
CARL BANKS

"What is life but a series of inspired follies?"
    -- Henry Higgins in _Pygmalion_, by George Bernard Shaw


From carribeiro at yahoo.com  Sat Mar 24 16:46:12 2001
From: carribeiro at yahoo.com (Carlos Ribeiro)
Date: Sat, 24 Mar 2001 18:46:12 -0300
Subject: reading single bits?
In-Reply-To: <3abcb2da$0$20400@SSP1NO25.highway.telekom.at>
Message-ID: <5.0.2.1.0.20010324175338.00a2f6a0@mail.inet.com.br>

At 15:40 24/03/01 +0100, Alex Berger wrote:
>How can I read/write single bits of some binary data when I know the
>position??

Let's divide the problem in two. First you need to read the information 
from the file object. This can be done by opening the file in binary mode, 
seeking to the desired position and reading the information. But (on most 
modern filesystems) files are composed of *bytes*, not bits. So you are 
going to read one byte, not one bit. You have to account for this 
difference when seeking, dividing the offset by 8 to get to the position. 
The modulus of this division is the position of the bit inside the byte. 
For example, lets assume that you want to red the bit 35:

35 div 8 = 4  (position of the byte inside the file)
35 mod 8 = 3  (position of the bit inside the byte)

Some example code:

 >>> f = open('myfile', 'rb')
 >>> f.seek(position div 8)
 >>> c = int(f.read(1))

Now you have a single byte, preferably represented as a integer (in other 
languages, the data type name may be different - byte in Pascal, char or 
shortint in C...). In order to extract a single bit, you use binary 
arithmetic operations. Let's c be the byte you read on the first step, and 
b the bit number (from 0 to 7):

Read :  bit = (c and (1 * (2 ** b))) != 0
Write:  c = c or (bit * (2 ** b))

The basic idea is to generate a binary mask for the bit you want to 
read/write, and then use binary operations over the byte. There are other 
ways to achieve the same goal more efficiently. The most popular is using 
"<<" - the "shift left" operator. Shifting 1 to the left is equivalent to 
the 2**b operation, and makes for a slightly simpler solution:

Read :  bit = (c and (1 << b)) != 0
Write:  c = c or (bit << b)

Please note that these techniques apply to most languages and operating 
systems (I cant remember of a single example in modern environments where 
it does not works). Some details may be different. For instance, the != 
logical operator returns 1 or 0 in Python. Some other languages may have 
different behavior.

This is a very basic tutorial, and several simplifications are possible.


Carlos Ribeiro


_________________________________________________________
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com




From donn at oz.net  Sat Mar  3 00:57:41 2001
From: donn at oz.net (Donn Cave)
Date: 3 Mar 2001 05:57:41 GMT
Subject: Thinking Pythonically (was Re: gah! I hate the new string  syntax)
References: <mailman.983485923.1136.python-list@python.org> <B7In6.16281$Qb7.2658902@newsb.telia.net> <97ohh2$polt0$1@ID-77331.news.dfncis.de> <97opsb$nhu$1@nntp6.u.washington.edu> <97oupq$qijrh$1@ID-77331.news.dfncis.de> <97p3mh$bkm$1@tyfon.itea.ntnu.no> <B6C58CBD.212F%f8dy@diveintopython.org>
Message-ID: <97q14l$jbi$0@216.39.151.169>

Quoth Mark Pilgrim <f8dy at diveintopython.org>:
...
| My original point, in case it got lost in the details, was that I like list
| comprehensions because they force me to think Pythonically.  By that I mean
| thinking in terms of the things that Python is good at, like lists and
| dictionaries.  These are rich data structures with lots of built-in
| capabilities, and if I don't use them, I may as well go back to programming
| in Powerbuilder: writing WHILE loops to do progressive searches through
| strings when I really want something like the above example, using SELECT
| CASE statements for jump tables when all I really want is a dictionary of
| functions, and so forth.

Ah.  You know, not only am I a grizzled, hide-bound veteran of old
python versions beyond reckoning, I'm a UNIX programmer.

I have no idea what it's like to use Visual Basic, I have never
even heard of Powerbuilder.  Like many of my gnarly old brethren,
I fooled around with Scheme dialects, Tcl and whatever came along,
until eventually I think it was Ken Manheimer was kind enough to
personally mail me a clue about Python, but the big competition
in our world was and still is Perl.  Most of us have been there,
and when we ask ourselves what Python is about, Perl is a huge
part of the background.

If you haven't already, you must learn about Perl.  You don't
need to write programs, but along with some reading about the
language you want to see what happens when people actually write
Perl to do stuff, not for publication.  This is a language that
has the lists and the dictionaries and lots more syntactical
features that save you a bundle of typing - Python is tediously
verbose by comparison.  Many people love Perl.  You may find that
you do, too, there's no reason to expect otherwise.  But it's
important to recognize that there is a huge crowd of people who
swim upstream to use Python instead, and it's not just because
we have to be different, it's not because we want to use Zope
or our boss twisted our arm to write in Python.  It's certainly
not because Python is more powerful, in the sense you have been
talking about.  I'm beating around the bush a little, because
I don't know how to talk about it directly in a way that really
does justice to it, but I think if you get it, you'll find it
more interesting than rich feature sets.  Or maybe not, I don't
know, I guess it's not for everyone, but for myself, I want to
challenge your notion of what it is to think Pythonically.

	Donn Cave, donn at oz.net


From sholden at holdenweb.com  Mon Mar  5 21:19:09 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Mon, 5 Mar 2001 21:19:09 -0500
Subject: the demise of 'from foo import * and its implications?
References: <3kVBAfATrDo6EwEM@jessikat.fsnet.co.uk><mailman.983582584.8198.python-list@python.org><z8dAGJAyqLo6EwlX@jessikat.fsnet.co.uk> <mailman.983746566.27513.python-list@python.org> <suPo6.16824$ML1.841861@e420r-atl2.usenetserver.com> <slrn9a84io.ogf.qrczak@qrnik.zagroda>
Message-ID: <iSXo6.24273$1D5.977153@e420r-atl1.usenetserver.com>

[me]
...as a reference to the module's "i", but I'm not sure that the gain is
worth
the pain. Even if such a proposal were to be PEP-ified and accepted it could
not be mandated because that would break humongous amounts of existing code.

[the martellibot]
But why not just
    import modulewithaverylongname as sh
and then use
      sh.i
...?

You can do that today, and the detailed semantics seem
likelier to be what is desired (e.g., should attribute
i of the module ever be re-bound, it's probably better
to automatically pick up the new reference, as happens
by using sh.i, than keep using the old one, as any
'from/import' approach would do -- marginal stuff, sure).

["Qrczak"]
>
> I don't see why do to this, and it's certainly not a feature to go
> into the standard language.
>
> An opposite trick can be used if a module is designed to be
> star-imported and uses name prefixes, but we dislike star-import.
> We can remove prefixes while moving objects to a separate namespace.
>
Erm, OK. So we all agree, then. What I said was not a very good idea really
*was not* a very good idea.

but-i-thought-i'd-said-that-already-ly y'rs  - steve





From will at nospam.misconception.org.uk  Mon Mar 19 13:37:30 2001
From: will at nospam.misconception.org.uk (Will Newton)
Date: Mon, 19 Mar 2001 18:37:30 +0000
Subject: Subsetting a dictionary
References: <993dkm$ttr$1@news6.svr.pol.co.uk> <993qq3$4bajg$1@ID-11957.news.dfncis.de> <995hj6$dk6$1@newsg3.svr.pol.co.uk> <995itl$2d2$1@panix2.panix.com>
Message-ID: <995jg8$k85$1@newsg1.svr.pol.co.uk>

Aahz Maruch wrote:

>>On average this makes the code ~10x slower. I haven't done an in depth
>>analysis of how the routine is called, but my "backwards" way of doing
>>things seems much faster for my task.
> 
> While it's possible you're right, my gut instinct is that Emile's way
> will be faster unless your subset almost always includes most of the
> dictionary.

I ran the profiles, trust me, my way is faster, just not fast enough. :)



From kens at sightreader.com  Mon Mar 26 22:51:20 2001
From: kens at sightreader.com (Ken Seehof)
Date: Mon, 26 Mar 2001 19:51:20 -0800
Subject: Good GUI for Python
References: <002a01c0b33f$9595ade0$340210ac@winson> <005b01c0b616$cda12640$a325fea9@his> <001f01c0b662$676079c0$340210ac@winson>
Message-ID: <00d701c0b671$2fac13e0$04090a0a@upcast.com>

The fastest way to learn about wxPython is to just download it and "Run the DEMO".
You will have a basic idea of its capabilities in 30 minutes just by playing with the demo.
You will be writing simple applications within a few hours if you have prior GUI experience.
(especially if you start with http://wxpython.org/test1.py.html).

www.wxpython.org

Also, you should take a look at wxDesigner.  It's an excellent tool for creating
forms and dialogs.  It will give you some ideas for code generation too.

http://www.roebling.de/

Before creating any tool, be sure to see if someone else has already done it.
http://www.vex.net/parnassus/
  ----- Original Message ----- 
  From: ??? 
  To: Ken Seehof 
  Sent: Monday, March 26, 2001 6:05 PM
  Subject: Re: Good GUI for Python


  Is there any detail books or resources talked about wxPython/ wxWindows architecture? I would like to code a GUI generator based on wxPython and MySQL. So I surveyed Python's GUI platform first. :)

  --
  Best Regards,

  Winson Chang

  3F, 2, 150 Lane, Sec. 5, Hsin I Rd., Taipei, Taiwan, R.O.C.

  Tel: +886-2-27225333 ext 376
  Fax: +886-2-27222330

  Email: winson at mdsserv.mds.com.tw
    ----- Original Message ----- 
    From: Ken Seehof 
    To: python-list 
    Sent: Tuesday, March 27, 2001 1:04 AM
    Subject: Re: Good GUI for Python


    wxPython is the best.

    I use PythonWin as my IDE and wxPython as my GUI platform.  I find wxPython much easier than PythonWin mostly because MFC is hard to use (even though I've been using MFC in C++ and python (win32ui) for about 10 years).  Also, wxPython is portable, unlike win32ui.   I agree that win32ui is a masterpiece, but it is dependent on MFC, which is not a masterpiece :-(.

    I am going to switch to Komodo next (which absorbs Marks expertise but without the MFC limitations).  Komodo looks really nice, but I haven't gotten it to work on my system yet (probably because I have both Python 1.5.2 and 2.0 installed).  I'm looking forward to the switch.  I will continue to use wxPython as my GUI platform.
      Hi, All
       
      I surveyed 3 GUI framework for Python including wxPython, PythonWin's included win32ui and Tkinter. My development environment is Mark Hammond's PythonWin downloaded from ActiveStates.
       
      These're my issues:
       
      First, ActiveStates' Python tool cannot support Tkinter and I confirmed with their stuff. So Tkinter maybe good, but not compatible with PythonWin.
       
      Second, Mark's masterpiece win32ui and total MFC solution is excellent, but I'd wanna bind on Windows platform.
       
      Third, wxPython sounds great but not so easy as win32ui.
       
      Is there any suggestions ? Change my develop environment or is there another good gui for me ??

      --
      Best Regards,
       
      Winson Chang
       
      3F, 2, 150 Lane, Sec. 5, Hsin I Rd., Taipei, Taiwan, R.O.C.
       
      Tel: +886-2-27225333 ext 376
      Fax: +886-2-27222330
       
      Email: winson at mdsserv.mds.com.tw
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20010326/c4991add/attachment.html>

From MarkH at ActiveState.com  Sat Mar 10 19:54:52 2001
From: MarkH at ActiveState.com (Mark Hammond)
Date: Sun, 11 Mar 2001 00:54:52 GMT
Subject: Python, ASP, ADO, and Image string/binary problems
References: <ta0amgebq0sn0b@corp.supernews.com>
Message-ID: <3AAACC1F.5020203@ActiveState.com>

Shonn Gilson wrote:

> If I do a Response.BinaryWrite(bstrIn) I get the image in the browser but if
> I do Response.BinaryWrite(bstrOut.getvalue()) I get spaces between every
> character (I've seen this with unicode problems). It's not just an

Try using BinaryWrite(buffer(bstrOut.getvalue()))

Mark.



From g1a2j at pacbell.net  Wed Mar 28 12:10:56 2001
From: g1a2j at pacbell.net (g1a2j at pacbell.net)
Date: Wed, 28 Mar 2001 17:10:56 GMT
Subject: Pure Gold!!  2239
Message-ID: <AWow6.556$E71.14720@news.pacbell.net>

Yes, Pure Gold or Actually Cash. Would You Like To Stop Loosing Money.?? Instead Earn 80 to 100%
Interest. Min. Investment $ One Million. If you Have One Million And Can Prove It. Let's talk!! No Risk!!
Gary A.<g1a2j at pacbell.net>
pseimqugnkgijxzjsqkobjqurovybxpqozpxyltoxyenvftizevgumhgimkljyrsfqkrejn



From aahz at panix.com  Mon Mar 19 15:48:24 2001
From: aahz at panix.com (Aahz Maruch)
Date: 19 Mar 2001 12:48:24 -0800
Subject: Subsetting a dictionary
References: <993dkm$ttr$1@news6.svr.pol.co.uk> <995hj6$dk6$1@newsg3.svr.pol.co.uk> <995itl$2d2$1@panix2.panix.com> <995jg8$k85$1@newsg1.svr.pol.co.uk>
Message-ID: <995rao$e4n$1@panix3.panix.com>

In article <995jg8$k85$1 at newsg1.svr.pol.co.uk>,
Will Newton  <will at nospam.misconception.org.uk> wrote:
>Aahz Maruch wrote:
>
>>>On average this makes the code ~10x slower. I haven't done an in depth
>>>analysis of how the routine is called, but my "backwards" way of doing
>>>things seems much faster for my task.
>> 
>> While it's possible you're right, my gut instinct is that Emile's way
>> will be faster unless your subset almost always includes most of the
>> dictionary.
>
>I ran the profiles, trust me, my way is faster, just not fast enough. :)

But I don't know how you coded the build-a-dictionary approach,
therefore I don't know that you used an efficient mechanism to do your
profiling.
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

There's a difference between a person who gets shit zie doesn't deserve
and a person who gets more shit than zie deserves.  --Aahz


From cce at clarkevans.com  Tue Mar 20 18:43:48 2001
From: cce at clarkevans.com (Clark C. Evans)
Date: Tue, 20 Mar 2001 18:43:48 -0500 (EST)
Subject: Yet Another PEP:  Query Protocol Interface or __query__
Message-ID: <Pine.LNX.4.21.0103201837570.21343-100000@clarkevans.com>

Please have a look at the PEP below, it is intended to be a
light-weight solution to "typing" based on programmer intent.

Regards,

Clark Evans

PEP: XXX
Title: Query Interface Protocol
Version: $Revision$
Author: Clark Evans
Python-Version: 2.2
Status: Active
Type: Standards Track
Created: 21-Mar-2001

Summary

    This paper asserts that "interface typing" can be carved
    into two separable concerns, that of (a) protocol, which is
    all about behavior/expectations, and that of (b) signature
    which is about method existence and argument type checking.

    This proposal puts forth a declarative method for interface
    protocol discovery that could be orthogonal and complementary
    to a more signature based approach as being developed by the
    types special interest group.

Context

    Python is a very dynamic language with powerful introspection
    capabilities.  However, it has yet to formalize an interface
    or abstract type checking mechanism. A consensus for a user
    defined type system may be far off into the future.

    Currently, existence of particular methods, particularly those
    that are built-in such as __getitem__, is used as an indicator
    of support for a particular interface.  This method may work
    for interfaces blessed by GvR, such as the new enumerator interface
    being proposed and identified by a new built-in __iter__.

    However, this current method does not admit an infallible way
    to identify interfaces lacking a built-in method.

Motivation

    In the recent type special interest group discussion [1], there
    were two complementary quotes which motivated this proposal:

       "The deep(er) part is whether the object passed in thinks of
        itself as implementing the Foo interface. This means that
        its author has (presumably) spent at least a little time
        about the invariants that a Foo should obey." - GvR [2]

    and

       "There is no concept of asking an object which interface it
        implements. There is no "the" interface it implements. It's
        not even a set of interfaces, because the object doesn't
        know them in advance. Interfaces can be defined after objects
        conforming to them are created." -- Marcin Kowalczyk [3]

    The first quote focuses on the intent of a class, including
    not only the existence of particular methods, but more
    importantly the call sequence, behavior, and other invariants.
    Where the second quote focuses on the type signature of the
    class.  These quotes motivate a distinction between interface
    as a "declarative, I am a such-and-such" construct, as opposed
    to a "descriptive, It looks like a such-and-such" mechanism.

    Furthermore, it is clear that both aspects of interface are
    important, if not completely orthogonal and complementary.
    For purposes of this proposal, the word "protocol" is aligned
    with Guido's deep issue, and "signature" with Marcin's vision
    of a type system.  In this way, we clearly demarcate the two
    attitudes of the word "interface" in a type system.

    Clearly, detailing an exhaustive method of interface
    signatures is very difficult problem, especially for
    a dynamic language like Python where the interface
    signature for a class may change over time.  However, a
    simple declarative mechanism to inquire which interface
    protocols an object supports at a given time is relatively
    straight forward and could bring immediate benefit.

Details

    A new built-in method, __query__ is proposed.  This method has
    a single argument, an interface protocol identifier, and either
    returns an object supporting the given protocol, or throws an
    unknown exception.

    An interface protocol identifier is a lower-case string having
    a reverse DNS host name string:   TLD '.' DOMAIN ['.' NAME ]*
    Where TLD is a top level domain such as 'com' or 'uk.co'.  DOMAIN
    is a registered name in the given TLD.  And one or more optional
    NAME separated by a period.  NAME is a sequence of one or more
    alphabetic characters including the dash '-' character.  See the
    relevant ITEF specifications for specific details.

    Note that all of the protocols above have at least one period.
    In the future it may be prudent to introduce one or more "blessed"
    interface protocol identifiers which do not have a period. Potential
    examples include "enumerator", "sequence", "map", "string", etc.

    In addition, a built-in query method could be introduced that
    calls the __query__ method on a given object.

Example Usage

    >>> class EggsOnly:
            def eggs(self,str): print "eggs!" + str
            def __query__(self,protocol):
                if protocol == "org.python.example.eggs": return self
                raise "unknown: " + protocol

    >>> class EggsAndHam:
            def ham(self,str): print "ham!"+str
            def __query__(self,protocol):
                if protocol == "org.python.example.ham": return self
                if protocol == "org.python.example.eggs":
                    return EggsOnly()
                raise "unknown" + protocol

    >>> x = EggsOnly()
    >>> x.eggs("hello")
    eggs!hello
    >>> x.__query__("org.python.example.eggs").eggs("hello")
    eggs!hello

    >>> z = EggsAndHam()
    >>> z.ham("hi")
    ham!hi
    >>> z.__query__("org.python.example.ham").ham("hi")
    ham!hi
    >>> z.__query__("org.python.example.eggs").eggs("hi")
    eggs!hi

    >>> y = x.__query__("org.python.bad")
    Traceback (innermost last):
      File "<interactive input>", line 1, in ?
      File "<interactive input>", line 5, in __query__
        unknown: org.python.bad

Relationships:

    The iterator special interest group is proposing a new built-in
    called "__iter__", which could be replaced with __query__ and
    a blessed interface protocol identifier of "enumerator".
    Therefore calls to  obj.__iter__() could be replaced with
    obj.__query__("enumerator") with no semantic difference.

    Although this proposal may sounds similar to Microsoft's
    QueryInterface, it differs by a number of aspects.  First,
    there is not a special "IUnknown" interface which can be used
    for object identity, although this could be proposed as one
    of those "special" blessed interface protocol identifiers.
    Second, with QueryInterface, once an object supports a particular
    interface it must always there after support this interface;
    this proposal makes no such guarantee, although this may be
    added at a later time. Third, implementations of Microsoft's
    QueryInterface must support a kind of equivalence relation.
    By reflexive they mean the querying an interface for itself
    must always succeed.  By symmetrical they mean that if one
    can successfully query an interface IA for a second interface
    IB, then one must also be able to successfully query the
    interface IB for IA.  And finally, by transitive they mean if
    one can successfully query IA for IB and one can successfully
    query IB for IC, then one must be able to successfully query
    IA for IC.  Ability to support this type of equivalence relation
    should be encouraged, but may not be possible.  Further research
    on this topic (by someone familiar with Microsoft COM) would be
    helpful in further determining how compatible this proposal is.

Backwards Compatibility

    There should be no problem with backwards compatibility.
    Indeed this proposal, save an built-in query() function,
    could be tested without changes to the interpreter.

Future Compatibility

    It appears that this proposal could be implemented orthogonal
    to the protocol checking system being constructed [4] by Paul
    Prescod and company.  It may not be all that compatible with
    the more ambitious signature declaration and checking approach[5]
    taken by Michel Pelletier.  This requires further investigation.

Questions and Answers

    Q:  This is just a type-coercion proposal.

    A:  No. Certainly it could be used for type-coercion, such
        coercion would be explicit via __query__ or query function.
        Of course, if this was used for iterator interface, then the
        for construct may do an implicit __query__("enumerator") but
        this would be an exception rather than the rule.

    Q:  Why did the author write this PEP?

    A:  He wanted a simple proposal that covered the "deep part" of
        interfaces without getting tied up in signature woes.  Also, it
        was clear that __iter__ proposal put forth is just an example
        of this type of interface.  Further, the author is doing XML
        based client server work, and wants to write generic tree based
        algorithms that work on particular interfaces and would
        like these algorithms to be used by anyone willing to make
        an "adapter" having the interface required by the algorithm.

    Q:  Why not call this __queryinterface__ ?

    A:  Too close to Microsoft's QueryInterface, especially given the
        semantic differences which may not be reconcilable.

    Q:  Is this in opposition to the type special interest group?

    A:  No.  It is meant as a simple, need based solution that could
        easily complement the efforts by that group.

Copyright

    This document has been placed in the public domain.


References and Footnotes

    [1] http://www.zope.org/Members/michel/types-sig/TreasureTrove
    [2] http://mail.python.org/pipermail/types-sig/2001-March/001105.html
    [3] http://mail.python.org/pipermail/types-sig/2001-March/001206.html
    [4] http://mail.python.org/pipermail/types-sig/2001-March/001223.html






From rdsteph at earthlink.net  Tue Mar 27 21:41:18 2001
From: rdsteph at earthlink.net (Ron Stephens)
Date: Wed, 28 Mar 2001 02:41:18 GMT
Subject: Semi-final code as text...
References: <3ABCC757.6C6F17CA@earthlink.net> <3ABE73B2.CAFCA177@student.liu.se> <3ABFEC14.BD3B0709@earthlink.net>
Message-ID: <3ABFF28E.EE28A322@earthlink.net>

print "Decision Analysis Beta Test 0.1"
print "To Help You Decide Amongst Various Alternatives"
print "A General Program Requiring User Input"

number = input ("How many options do you want to decide amongst? (2 to 6):")


x = 1




option1 = raw_input ("what is option1:")

x = x + 1
option2 = raw_input ("what is option2:")
x = x + 1
while x:
 if x > number: break
 option3 = raw_input ("what is option3:")
 x = x + 1
 if x > number: break
 option4 = raw_input ("what is option4:")
 x = x + 1
 if x > number: break
 option5 = raw_input ("what is option5:")
 if x > number: break
 option6 = raw_input ("what is option6:")
 break



criteria = input ("How many criteria do you want to use for this decision? (1 to
6)")
y = 1




criteria1 = raw_input ("what is criterium1:")

y = y + 1
criteria2 = raw_input ("what is criterium2:")
y = y + 1
while y:
 if y > criteria: break
 criteria3 = raw_input ("what is criterium3:")
 y = y + 1

 if y > criteria: break
 criteria4 = raw_input ("what is criterium4:")
 y = y + 1
 if y > criteria: break
 criteria5 = raw_input ("what is criterium5:")
 y = y + 1
 if y > criteria: break
 criteria6 = raw_input ("what is criterium6:")
 break

z = 1
print "Of the criteria you have listed, when prompted please give each one a
rank of importance from 1 to 100, with 1 being the least factor of relative
importance, and 100 being the highest factor of importance. After entering a
numeral form 1 to 100, hit enter."
while z:
 rank1 = input ("From 1 to 100, what is the importance of %s " %criteria1)
 z = z + 1
 rank2 = input ("From 1 to 100, what is the importance of %s " %criteria2)
 z = z + 1
 if z > criteria: break
 rank3 = input ("From 1 to 100, what is the importance of %s " %criteria3)
 z = z + 1
 if z > criteria: break
 rank4 = input ("From scale of 1 to 100, what is the importance of %s "
%criteria4)
 z = z + 1
 if z > criteria: break
 rank5 = input ("From scale of 1 to 100, what is the importance of %s "
%criteria5)
 z = z + 1
 if z > criteria: break
 rank6 = input ("From 1 to 100, what is the importance of %s " %criteria6)
 break
q = 1
while q:

 a = 1
 while a:

  oneone = input ("From 1 to 100, how does %s rank with respect to %s ?" %
(option1, criteria1))
  a = a + 1
  onetwo = input ("From 1 to 100, how does %s rank with respect to %s ?" %
(option1, criteria2))
  a = a + 1
  if a > criteria: break
  onethree = input ("From 1 to 100, how does %s rank with respect to %s ?" %
(option1, criteria3))
  a = a + 1
  if a > criteria: break
  onefour = input ("From 1 to 100, how does %s rank with respect to %s ?" %
(option1, criteria4))
  a = a + 1
  if a > criteria: break
  onefive = input ("From 1 to 100, where does %s rank for %s ?" % (option1,
criteria5))
  a = a + 1
  if a > criteria: break


  onesix = input ("From 1 to 100, where does %s rank for %s ?" % (option1,
criteria6))
  break
 q = q + 1

 if q > number: break

 b = 1

 while b:


  twoone = input ("where does %s rank for %s ?" % (option2, criteria1))
  b = b + 1
  twotwo = input ("where does %s rank for %s ?" % (option2, criteria2))
  if b > criteria: break
  b = b + 1
  twothree = input ("where does %s rank for %s ?" % (option2, criteria3))
  b = b + 1
  if b > criteria: break
  twofour = input ("where does %s rank for %s ?" % (option2, criteria4))
  b = b + 1
  if b > criteria: break
  twofive = input ("where does %s rank for %s ?" % (option2, criteria5))
  b = b + 1
  if b > criteria: break
  twosix = input ("where does %s rank for %s ?" % (option2, criteria6))
  b = b + 1
  break

 q = q + 1

 if q > number: break

 c = 1
 while c:
  threeone = input ("where does %s rank for %s ?" % (option3, criteria1))
  c = c + 1
  threetwo = input ("where does %s rank for %s ?" % (option3, criteria2))
  c = c + 1
  if c > criteria: break
  threethree = input ("where does %s rank for %s ?" % (option3, criteria3))
  c = c + 1
  if c > criteria: break
  threefour = input ("where does %s rank for %s ?" % (option3, criteria4))
  c = c + 1
  if c > criteria: break
  threefive = input ("where does %s rank for %s ?" % (option3, criteria5))
  c = c + 1
  if c > criteria: break
  threesix = input ("where does %s rank for %s ?" % (option3, criteria6))
  break
 q = q + 1

 if q > number: break


 d = 1
 while d:
  fourone = input ("where does %s rank for %s ?" % (option4, criteria1))
  d = d + 1
  fourtwo = input ("where does %s rank for %s ?" % (option4, criteria2))
  d = d + 1
  if d > criteria: break
  fourthree = input ("where does %s rank for %s ?" % (option4, criteria3))
  d = d + 1
  if d > criteria: break
  fourfour = input ("where does %s rank for %s ?" % (option4, criteria4))
  d = d + 1
  if d > criteria: break
  fourfive = input ("where does %s rank for %s ?" % (option4, criteria5))
  d = d + 1
  if d > criteria: break
  foursix = input ("where does %s rank for %s ?" % (option4, criteria6))
  break
 q = q + 1

 if q > number: break

 e = 1
 while e:

  fiveone = input ("where does %s rank for %s ?" % (option5, criteria1))
  e = e + 1
  fivetwo = input ("where does %s rank for %s ?" % (option5, criteria2))
  e = e + 1
  if e > criteria: break
  fivethree = input ("where does %s rank for %s ?" % (option5, criteria3))
  e = e + 1
  if e > criteria: break
  fivefour = input ("where does %s rank for %s ?" % (option5, criteria4))
  e = e + 1
  if e > criteria: break
  fivefive = input ("where does %s rank for %s ?" % (option5, criteria5))
  e = e + 1
  if e > criteria: break
  fivesix = input ("where does %s rank for %s ?" % (option5, criteria6))
  break
 q = q + 1

 if q > number: break


 f = 1
 while f:
  sixone = input ("where does %s rank for %s ?" % (option6, criteria1))
  f = f + 1
  sixtwo = input ("where does %s rank for %s ?" % (option6, criteria2))
  f = f + 1
  if f > criteria: break
  sixthree = input ("where does %s rank for %s ?" % (option6, criteria3))
  f = f + 1
  if f > criteria: break
  sixfour = input ("where does %s rank for %s ?" % (option6, criteria4))
  f = f + 1
  if f > criteria: break
  sixfive = input ("where does %s rank for %s ?" % (option6, criteria5))
  f = f + 1
  if f > criteria: break
  sixsix = input ("where does %s rank for %s ?" % (option6, criteria6))
  break
 break

if criteria < 4:
    onefour = 0
    onefive = 0
    onesix = 0
    twofour = 0
    twofive = 0
    twosix = 0
    threefour = 0
    threefive = 0
    threesix = 0
    fourfour = 0
    fourfive = 0
    foursix = 0
    fivefour = 0
    fivefive = 0
    fivesix = 0
    sixfour = 0
    sixfive = 0
    sixsix = 0
    rank4 = 0
    rank5 = 0
    rank6 = 0









elif criteria < 5:
    onefive = 0
    onesix = 0
    twofive = 0
    twosix = 0
    threefive = 0
    threesix = 0
    fourfive = 0
    foursix = 0
    fivefive = 0
    fivesix = 0
    sixfive = 0
    sixsix = 0

    rank5 = 0
    rank6 = 0

elif criteria < 6:
    onesix = 0
    twosix = 0
    threesix = 0
    foursix = 0
    fivesix = 0
    sixsix = 0

    rank6 = 0

else:
    print "These are the scores for each option. The highest score is the best
option"


w = 1
while w:
 valueoption1 = (oneone)*(rank1) + (onetwo)*(rank2) + (onethree)*(rank3) +
(onefour)*(rank4) + (onefive)*(rank5) +(onesix)*(rank6)
 w = w + 1
 if w > number: break
 valueoption2 = (twoone)*(rank1) + (twotwo)*(rank2) + (twothree)*(rank3) +
(twofour)*(rank4) + (twofive)*(rank5) +(twosix)*(rank6)
 w = w + 1
 if w > number: break
 valueoption3 = (threeone)*(rank1) + (threetwo)*(rank2) + (threethree)*(rank3) +
(threefour)*(rank4) + (threefive)*(rank5) +(threesix)*(rank6)
 w = w + 1
 if w > number: break
 valueoption4 = (fourone)*(rank1) + (fourtwo)*(rank2) + (fourthree)*(rank3) +
(fourfour)*(rank4) + (fourfive)*(rank5) +(foursix)*(rank6)
 w = w + 1
 if w > number: break
 valueoption5 = (fiveone)*(rank1) + (fivetwo)*(rank2) + (fivethree)*(rank3) +
(fivefour)*(rank4) + (fivefive)*(rank5) +(fivesix)*(rank6)
 w = w + 1
 if w > number: break
 valueoption6 = (oneone)*(rank1) + (sixtwo)*(rank2) + (sixthree)*(rank3) +
(sixfour)*(rank4) + (sixfive)*(rank5) +(sixsix)*(rank6)
r = 1
while r:
 print "%s is valued at %d" % (option1, valueoption1)
 r = r + 1
 if r > number: break
 print "%s is valued at %d" % (option2, valueoption2)
 r = r + 1
 if r > number: break
 print "%s is valued at %d" % (option3, valueoption3)
 r = r + 1
 if r > number: break
 print "%s is valued at %d" % (option4, valueoption4)
 r = r + 1
 if r > number: break
 print "%s is valued at %d" % (option5, valueoption5)
 r = r + 1
 if r > number: break
 print "%s is valued at %d" % (option6, valueoption6)
 r = r + 1
 if r > number: break


print "the option with the highest score is the best choice!!!"

print "Thank you for using DecisionAnalysis. Comments and suggestions are
welcome and can be addressed to Ron Stephens at rdsteph at earthlink.net"


# and then the proposed Tkinter front end

#! /usr/bin/env python

VERSION='0.1 GUI'
NUMCRITERIA=10
NUMOPTIONS=10

from Tkinter import *
import Pmw

class decision(Frame):
    def __init__(self,master):
        # initialize parent class
        Frame.__init__(self,master)

        # create list to hold widgets - 11 rows of 10 widgets
        self.buttons=[]
        self.strings=[]
        for x in range(NUMOPTIONS+2):
        # can't just create the entire list (results in the inner lists being
identical
        # which leads to losing widgets
            self.buttons.append([None]*(NUMCRITERIA+1))
            self.strings.append([None]*NUMCRITERIA)
        self.score=[None]*NUMOPTIONS
        self.master=master

        # create the Criteria Titles - as entry fields so you can set the value.

        for x in range(NUMCRITERIA):
            self.strings[0][x]=StringVar()

self.buttons[0][x]=Entry(self,bg="white",width=8,textvariable=self.strings[0][x],justify=LEFT)

            self.buttons[0][x].grid(row=0,col=x,sticky=NSEW)
     self.grid_columnconfigure(x,weight=1)
 # create the score column title - as button to activate scoring

Button(self,text="Score",padx=1,justify=LEFT,command=self.scoring).grid(row=0,col=NUMCRITERIA,sticky=NSEW)

        self.grid_columnconfigure(NUMCRITERIA,weight=1)

 # create the weighting line - as pull-down menu
        for x in range(NUMCRITERIA):
            self.strings[1][x]=StringVar()

self.buttons[1][x]=Pmw.OptionMenu(self,labelpos=W,menubutton_textvariable=self.strings[1][x],items=('1','2','3','4','5','6','7','8','9','10'),menubutton_width=8)

            self.buttons[1][x].grid(row=1,col=x,sticky=NSEW)

 # create the option lines - as entry fields
 for y in range(NUMOPTIONS):
            for x in range(NUMCRITERIA):
                self.strings[y+2][x]=StringVar()

self.buttons[y+2][x]=Entry(self,bg="white",width=8,textvariable=self.strings[y+2][x],justify=LEFT)

                self.buttons[y+2][x].grid(row=y+2,col=x,sticky=NSEW)
     # create the score column - as labels

self.buttons[y+2][NUMCRITERIA]=Label(self,text="",padx=1,justify=LEFT)

self.buttons[y+2][NUMCRITERIA].grid(row=y+2,col=NUMCRITERIA,sticky=NSEW)

    def scoring(self):
        print 'scoring'
        for y in range(NUMOPTIONS):
            self.score[y]=0
            for x in range(NUMCRITERIA):
                s=self.buttons[y+2][x].get()
                if s:
                    w=self.buttons[1][x].get()
                    if w:
                        self.score[y]=self.score[y]+(eval(s)*eval(w))
            self.buttons[y+2][NUMCRITERIA].configure(text=str(self.score[y]))

        #self.update() # draw screen
        #self.after(REFRESH,self.redraw) # refresh after ? seconds


root=Tk()  # create the main window
# set the title
root.title("Decision Analysis Beta Test"+VERSION)
window=decision(root) # create the frame
window.pack(side=TOP,expand=YES,fill=BOTH) # place the frame in the main window
root.mainloop()  # tk main loop







































From tim.one at home.com  Sat Mar 10 14:59:21 2001
From: tim.one at home.com (Tim Peters)
Date: Sat, 10 Mar 2001 14:59:21 -0500
Subject: Urgent Question about Python licensing
In-Reply-To: <98cpfe$5oj$0@216.39.170.247>
Message-ID: <LNBBLJKPBEHFEDALKOLCOELKJEAA.tim.one@home.com>

[Dave LeBlanc]
> ...
> My questions are these:
> 1. What prevents Python and GPL code from being distributed together?
> If I write some Python code that drives a GPL'd app (but which does
> not depend on that app), what prevents me from offering both my python
> code and the GPL'd app on the same medium (disk/CD/web)?

You mean other than paranoia <0.9 wink>?  Not much, but GPL Interpretation is
an arcane art, so you had better ask the FSF about this.

> 2. Is it currently so,

No.

> or will Python end up with, some license that requires that python
> scripts be covered by the GPL?

Mabye if the sun goes nova, and that cools things off enough so that hell can
finally freeze over.

> 3. Is Python 2.0 covered by that evil CNRI 1.6 license?

2.0 is covered by CNRI's 1.6 license, as well as by BeOpen.com's minor
variation on that.  I disagree that it's evil, although the way it was
imposed on the community was exceedingly ham-handed, and it has created
problems due to its purported incompatibility with the GPL.

> I avoided 1.6 mostly because of the license, and thought that 2.0 had
> "skipped" 1.6 and used a far more equitable license then the 1.6 one.

I suggest you read the licenses in question (you clearly have not yet).

> 1.6's license permits the copyright holder to suddenly decide that
> anyone using the covered software must pay a fee (or any other thing they
> so desire - this is only an example), and by virtue of your even so much
> as having read the license are implicitly bound to their unilateral
> declaration - lovely Virginia laws).

No offense intended, Dave, but this is the most concentrated pile of
absurdist FUD I've seen in a solid year of ill-informed license rants.  Even
by the FSF's standards, the CNRI 1.6 license is a "free software" license
(see "The License of Python 1.6b1 and later versions", at:

    http://www.fsf.org/philosophy/license-list.html

).

the-2.1-license-will-require-giving-your-firstborn-to-guido-ly y'rs  - tim




From hzhu at users.sourceforge.net  Mon Mar  5 18:47:38 2001
From: hzhu at users.sourceforge.net (Huaiyu Zhu)
Date: Mon, 05 Mar 2001 23:47:38 GMT
Subject: I come to praise .join, not to bury it...
References: <mailman.983756766.27347.python-list@python.org> <97vnu3076q@news2.newsguy.com>
Message-ID: <slrn9a89dk.2rm.hzhu@rocket.knowledgetrack.com>

It is all nice and well to say that sep.join(list) is good for polymorphism,
except that there are several practical annoyances:

1. The word join can be either transitive or nontransitive:
   sep joins list.
   list joins with sep.
   
   Might sep.glue(join) be clearer?  Well, compatibility with str.join?
   
2. Can we really join all kinds of lists?

>>> a = range(10)
>>> "".join(a)
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
TypeError: sequence item 0: expected string, int found

3. What's the inverse of join?  Is it split?

>>> b = [str(x) for x in a]
>>> c = "".join(b)
>>> c
'0123456789'
>>> d = "".split(c)
>>> d
['']

Oops! The other way round.  Try again
>>> d = c.split("")
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
ValueError: empty separator

What?  Oh, in this particular case, the inverse is
>>> d = list(c)
>>> d
['0', '1', '2', '3', '4', '5', '6', '7', '8', '9']

For other cases this is OK:
>>> c = ".".join(b)
>>> c
'0.1.2.3.4.5.6.7.8.9'
>>> d = c.split(".")
>>> d
['0', '1', '2', '3', '4', '5', '6', '7', '8', '9']

4. So can we just change the order of split?  No.  Consider

file.readlines()
file.read().splitlines()
file.read().split("\n")

More abstractly, when we split something, we have 
   whole - sep ==> components.
But when we join things together, we do it like
   glue + components ==> whole,
rather then the "more natural"
   components + glue => whole,
where "more natural" is defined as "most likely to be guessed by uninitiated
minds".

Of course, all these point do not detract from the fact that from
implementer's point of view, sep.join(list) is the most natural.

So what am I trying to say here?  I'd say that there's not much point in
arguing, or even explaining which way is better.  What would help is to
make things more symmetrical/smooth if possible.

Huaiyu


From colinmeeks at home.com  Thu Mar 15 09:21:57 2001
From: colinmeeks at home.com (Colin Meeks)
Date: Thu, 15 Mar 2001 14:21:57 GMT
Subject: Random Access Files
Message-ID: <9e4s6.42245$p66.13534448@news3.rdc1.on.home.com>

Is it possible to have random access files as opposed to serial access.
Basically what I would like to be able to do is be able to write data within
a file, without having to

    1.    Read the first part
    2.    Write the first part
    3.    Write the bit I want
    4.    Read the final part
    5.    Write the final part

I would like to be able to do something like

    1.    Seek the potion to write
    2.    Write
    3.    Close file

I don't mean writing code that hides the methods of the first example.  If I
remember correctly you can do random access files in GWBasic.

Anyone have any pointer??

Thanks

Colin




From cce at clarkevans.com  Tue Mar 20 21:22:45 2001
From: cce at clarkevans.com (Clark C. Evans)
Date: Tue, 20 Mar 2001 21:22:45 -0500 (EST)
Subject: Yet Another PEP:  Query Protocol Interface or __query__ 
Message-ID: <Pine.LNX.4.21.0103202122250.16274-100000@clarkevans.com>

On 20 Mar 2001, Aahz Maruch wrote:
> Clark C. Evans <cce at clarkevans.com> wrote:
> >
> >    A new built-in method, __query__ is proposed.  This method has
> >    a single argument, an interface protocol identifier, and either
> >    returns an object supporting the given protocol, or throws an
> >    unknown exception.
> >
> >    An interface protocol identifier is a lower-case string having
> >    a reverse DNS host name string:   TLD '.' DOMAIN ['.' NAME ]*
> >    Where TLD is a top level domain such as 'com' or 'uk.co'.  DOMAIN
> >    is a registered name in the given TLD.  And one or more optional
> >    NAME separated by a period.  NAME is a sequence of one or more
> >    alphabetic characters including the dash '-' character.  See the
> >    relevant ITEF specifications for specific details.
> 
> It's not at all clear to me why this domain thing is being suggested.

A unique identifier is needed.  One could use GUIDS, but I 
think that this would be less readable and maintainable.

> Is there supposed to be some kind of lookup mechanism 
> associated with it?

Yes.  I was planning that 

  query(obj,"org.python.example.eggs")

would behind the scenes call __query__ on the 
object obj.  This indirection allows for 
additional "hooks" to be added to the system 
via another PEP that could perform a wrapping
function.  See Alex's recent post.

> Most importantly, how does one find out what interfaces 
> an object *does* support?

I don't have this need -- so I didn't include it.  I suppose
we could ammend the proposal to have this functionality.
Perhaps obj.__query__() could return a sequence (or better 
yet a new iterator object).

What do you think?

> >Example Usage
> >
> >    >>> class EggsOnly:
> >            def eggs(self,str): print "eggs!" + str
> >            def __query__(self,protocol):
> >                if protocol == "org.python.example.eggs": return self
> >                raise "unknown: " + protocol
> >
> >    >>> class EggsAndHam:
> >            def ham(self,str): print "ham!"+str
> >            def __query__(self,protocol):
> >                if protocol == "org.python.example.ham": return self
> >                if protocol == "org.python.example.eggs":
> >                    return EggsOnly()
> >                raise "unknown" + protocol
> 
> This seems to be implying that it is appropriate to return a brand-new
> object with no association with the original object.  Is that correct?

Yes.  The __query__ mechansim is vague in this respect,
no use to have any artificial limitations.

> 
> Does it make sense to have something like:
> 
> class HamAndEggs:
>     def __query__(self, protocol):
>         if protocol.find("org.python.example.ham") == 0:
>             return self
> h = HamAndEggs()
> e = h.__query__("org.python.example.ham.eggs")


Let me see, a string goes in, and an object 
comes out. Sure.  Looks good to me!

Clark





From parkw at better.net  Sun Mar 11 18:47:30 2001
From: parkw at better.net (William Park)
Date: Sun, 11 Mar 2001 18:47:30 -0500
Subject: Chinese text GIF file generation
In-Reply-To: <98h00p+bqc2@eGroups.com>; from hungjunglu@yahoo.com on Sun, Mar 11, 2001 at 10:59:37PM -0000
References: <98h00p+bqc2@eGroups.com>
Message-ID: <20010311184730.A503@better.net>

On Sun, Mar 11, 2001 at 10:59:37PM -0000, hungjunglu at yahoo.com wrote:
> William Park <parkw at b...> wrote: 
> > Because there is so many variables (ie. font, style, size, ...), I
> > don't think there is package that does what you want.  However, you
> > can use LaTeX to produce Postscript format, and then convert it into
> > GIF.
> 
> There surely are existing packages. Just visit any of the Chinese 
> news portals (e.g: www.sinanet.com) and you'll see what I mean. 

They are using blocks of image in 
    <map> <area...> <area...> ...  </map>.
So, what is your question?  Do you simply want to duplicate it using
Python?  Or, do you want to display each character as separate image?

> Unlike Korean or Japanese, or even Vietnamese, for Chinese websites, 
> it is usually a better idea to present text in GIF than in multi-byte 
> encoding schemes. The thing is, there is a big need out there for 
> real-time or quasi-real-time on-the-fly GIF generation of Chinese 
> text in GIF format. Anything that involves an intermediate manual 
> process is unacceptable. (Besides, it's easy to do it manually on any 
> PC: screen capture + format conversion and you are done. It's just 
> time-consuming.) For websites containing multiple foreign languages 
> in the USA, text-to-gif converters is a real necessity: you don't 
> want to show jibberish to your visitors.

Since you seem to be talking about character-by-character display...
Okey, generation of image for each character can be automated; it's
simple script loop.  If you use LaTeX, then see near bottom of
'HTMLtag.py' package --- do the search from <www.python.org>.

Once these images are stored in /icons/chinese/, you would have to write
stand-alone script to match your input characters to corresponding
images.  It would spit out <img src=...> for each chinese character.  

---William Park, Open Geometry Consulting, Linux/Python, 8 CPUs.



From thomas at cintra.no  Fri Mar 30 08:32:06 2001
From: thomas at cintra.no (Thomas Weholt)
Date: Fri, 30 Mar 2001 15:32:06 +0200
Subject: What version of Berkley DB are used in Zope?
Message-ID: <tV%w6.1722$R6.31572@news1.oke.nextra.no>

If used at all, that is.

Thomas




From kens at sightreader.com  Wed Mar 28 15:18:23 2001
From: kens at sightreader.com (Ken Seehof)
Date: Wed, 28 Mar 2001 12:18:23 -0800
Subject: list vs tuple
References: <hGlw6.782$p5.3210@news1.rivrw1.nsw.optushome.com.au> <99stcb$483$1@panix6.panix.com> <99t9om$2o2q3$1@ID-77331.news.dfncis.de>
Message-ID: <002301c0b7c4$3d98a1e0$04090a0a@upcast.com>

Mark Pilgrim <f8dy at yahoo.com> asked:


> It was my understanding that only tuples containing immutable objects
could
> be used as dictionary keys.  For instance, (1,2,3) can be, and
("a","b","c")
> can be, but ([1,2,3], ["a","b","c"]) can not, because the underlying
objects
> could mutate and cause problems.  (Wow, that sounds like a bad sci-fi
> movie.)  Is this true?
>
> -M
> You're smart; why haven't you learned Python yet?
> http://diveintopython.org/

Exactly.  Suppose python didn't have that rule:

>>> a = [1,2]
>>> d = {}
>>> d[a] = 'parrot'  # in the real world this raises an exception
>>> a[0] = 5

Lets assume for the moment that we don't change the fact that keys are
inserted by reference.  Now in order to make things work, python would have
to rehash d[a].  This is a problem because python doesn't keep track of who
else is referencing the list object attached to the name 'a'.  Since
dictionaries store references as keys, so the object named 'a' and the
dictionary index to 'parrot' are the same object.  Some kind of very complex
scheme would have to be devised to tell the dictionary to rehash keys when
they get modified.

A rather extraordinarily bad alternative would be copy (deep copy) the key
when assigning into a dictionary, thereby attempting to keep the keys safe
from modification (this would mean the keys() and items() methods would have
to do a deep copy too).  It gets worse.  Dictionary keys would have to be
limited to objects that have a deep copy protocol.  Of course the
performance issues would be a nightmare.  Now I have a headache.  &-(

Considering the alternatives, the constaint that dictionary keys must be
immutable is quite elegant.

- Ken Seehof
kseehof at neuralintegrator.com
www.neuralintegrator.com





From squirmy at freeuk.com  Mon Mar  5 06:48:08 2001
From: squirmy at freeuk.com (Alison Myers)
Date: Mon, 05 Mar 2001 11:48:08 GMT
Subject: Questions from a beginner
Message-ID: <1103_983792888@breakdown>

Hi. I'm a beginning programmer at a fairly late age and have been looking for a language to learn.
I have decided against starting with C or C++ because I think they will be too complicated and they seem to be only really needed for writing device drivers and operating systems, which is not what I wat to do.
Is python a good language to learn first? Will my knowledge help me with other languages later?
I will be writing for win98 and Linux and I know Python is available on both of these, but would the code be transportable, especially if I write things using tcl (this is probably a long way off yet, but I'd like to know).
I have links to online tutorials, but where would I find good source code?
Do I need to buy any books to help me and if so which do you recommend?
Is there a list of programming projects anywhere or would I get these from a textbook?
I probably need to brush up on my maths, so what specific topics should I look at?
TIA

Alison



From loewis at informatik.hu-berlin.de  Wed Mar  7 05:11:33 2001
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 07 Mar 2001 11:11:33 +0100
Subject: XML Module problems
References: <mailman.982853044.8344.python-list@python.org> <memo.20010306181753.34635C@digitig.compulink.co.uk>
Message-ID: <j4y9uhnb1m.fsf@informatik.hu-berlin.de>

digitig at cix.co.uk (Tim Rowe) writes:

> That's what I have with my (ActiveState) 2.0 installation. 
> xml.parsers.expat is very like pyexpat but with some subtle differences 

??? xml.parsers.expat should be in the file xml/parsers/expat.py,
which should read

from pyexpat import *

So I'd call it *highly* unlikely that they are different.

> that make the name change appropriate (for example, IIRC expat
> returns attributes as a dictionary rather that pyexpat's list --
> expat's version makes much more sense to me).

In recent versions of pyexpat, this is a user option.

Regards,
Martin


From oliver at marugg.net  Fri Mar 23 10:38:59 2001
From: oliver at marugg.net (Oliver Marugg)
Date: Fri, 23 Mar 2001 16:38:59 +0100
Subject: File in a directory
Message-ID: <3ABB6E13.403B963B@marugg.net>

Hi,

I am a newbie in Python (reading Learning Python now) and have no idea
how I should solve the following problem:

In a local directory there are files (all the same type), which are
spooled in that directory through a stream (NFS, FTP or other). Now I
should read the content of that directory to verify if the size and/or
timestamp of one or more files has changed in past 5 minutes. If there
was a change in size wait, otherwise do something with these files.

Has anyone an idea how to solve that with the size or timestamp
verification?

Thank you.

Oliver Marugg





From francois.granger at free.fr  Fri Mar 30 18:18:57 2001
From: francois.granger at free.fr (=?ISO-8859-1?Q?Fran=E7ois_Granger?=)
Date: Fri, 30 Mar 2001 23:18:57 GMT
Subject: max filename size
References: <3ac416a6$0$4017$6e49188b@news.goldengate.net>
Message-ID: <1er40ru.1bz9tye1wk0cx1N%francois.granger@free.fr>

Voluciris <volucris at hotmail.com> wrote:

> Is there away to find the maximum filename length of the current os? On
> Windows it's 215 characters, but I don't want to hard code that in. It would
> be much more pythonic if it was portable.

You can go to the lowest common denominatore. If you target MacOS, it is
32. If you target DOS, it is 8.3 ;-)

-- 
"De quelque mani?re que l'on divise le monde, il y a toujours la m?me
proportion d'imb?ciles et de c.." - Propos de O.L.Barenton, Confiseur
A. Detoeuf


From maxm at normik.dk  Mon Mar  5 10:12:44 2001
From: maxm at normik.dk (=?iso-8859-1?Q?Max_M=F8ller_Rasmussen?=)
Date: Mon, 5 Mar 2001 16:12:44 +0100 
Subject: iis and python
Message-ID: <7BD10B680501D411B9DF009027E06F32197C1E@exchange>

Just did a test and this script runs nicely on win2K under IIS 5.0 with
Python 2.0:

<%@ LANGUAGE="Python"%>
<%

from String import join
result = []
for i in range(100):
    result.append('Hello world')

Response.Write('\n'.join(result))

%>

It's a complete page

( I use a result list and appends in the end as this is much faster than a
bunch of Response.Write()'s )

regards Max M

-----Original Message-----
From: tech [mailto:im_cne at hotmail.com]
Sent: Monday, March 05, 2001 3:28 PM
To: python-list at python.org
Subject: iis and python


I am going to do a project building a web front end  while my friend is
going to build the back end.

The backend will be MS SQL. The front end will be a web interface, IIS 5.0
or whatever comes with win2k .

I have a choice of doing php or python.

In order to use python it would have to entail cgi or ASP. Is this correct.
Or is there an alternative. I dont know if Zope is the right thing. I hate
DTML.

Is using python in ASP a pain in the but? Am I going to have to learn a
whole technology just to implelment python well into it?

I hear troubleshooting ASP pages is not fun.

The only time I have used iis is with php and it was a total snap. Painless.


-- 
http://mail.python.org/mailman/listinfo/python-list



From charlottwilliams at mediaone.net  Fri Mar 23 11:57:49 2001
From: charlottwilliams at mediaone.net (Charlotte Williams)
Date: Fri, 23 Mar 2001 16:57:49 GMT
Subject: Help with creating Python COM
References: <Dwqu6.79766$Ch.10654381@typhoon.we.rr.com> <mailman.985283909.27351.python-list@python.org>
Message-ID: <hgLu6.80562$Ch.11203214@typhoon.we.rr.com>

Paul Robinson <paul.robinson at businesscollaborator.com> wrote in message
news:mailman.985283909.27351.python-list at python.org...

> I think the first line of the tutorial sums it up nicely:
> "Last Updated 4 June 1997 9:00 a.m."
>
> The ni module has not been used for a while now. I beleive it's use was
> required pre 1.5 to enable package importing....or something like that.
>
> I would reccommend find a more upto date tutorial on Python and COM.
> e.g. the docs that come with Pythonwin, Mark Hammonds book - "Python
> Programming on win32" - ISBN: 1565926218, Mark's starship page also has
> links to sample code tutorials etc:
> http://starship.python.net/crew/mhammond/
>
> Hope that helps.
>
> Paul.
>
> Paul Robinson
> Business Collaborator Development Manager
> Enviros Software Solutions
> WWW: http://www.businesscollaborator.com
>

Thank you for your help.  I assumed the tutorial was up to date because the
author specified that Python 2.0 was needed for the example to work.  I
browsed Mark Hammond's site and think I can use some of the info there.
However, I looked at his lecture notes from a conference he gave in 2000,
and he mentions the "import ni" as well.  I'll try to see if I can get his
win32 book.  That might clear things up.





From me at nospam.net  Sat Mar 31 19:42:10 2001
From: me at nospam.net (Scottie)
Date: Sat, 31 Mar 2001 16:42:10 -0800
Subject: PEP 245: Python interfaces
References: <mailman.985377931.22542.python-list@python.org>
Message-ID: <tccubd5c7mc488@corp.supernews.com>

The one further thing I'd like to see is a way to declare that a class
implements an interface without having to edit either the interface
or the class -- Imagine the situation where I obtain two packages,
and I want to tell the one package that it can accept objects from
the other.

I am otherwise quite pleased by the proposal.  Perhaps a useful
part of thePEP would be a "File" interface, to provide a nice
concrete definition of a "file-like object."

-Scott David Daniels





From paulp at ActiveState.com  Sun Mar 11 10:39:15 2001
From: paulp at ActiveState.com (Paul Prescod)
Date: Sun, 11 Mar 2001 07:39:15 -0800
Subject: Who am I: can a class instance determine its own name?
References: <mailman.984034150.9635.python-list@python.org> <DYV4H3fqAHA.187@typhoon.i3s.net>
Message-ID: <3AAB9C23.259A2429@ActiveState.com>

Kevin Cazabon wrote:
> 
> ...
> You could also
> make it an initialization parameter.
> 
> FooBar = Foo(name="FooBar")
> 
> Just trying to help... not elegantly though.

I disagree Kevin. This is not only the canonical way to give an instance
a name, it is also the elegant way. It gives you a lot more flexibility
because the name it has at runtime is unrelated to the variable name you
happened to have given it in your code. If you need to change one or the
other for any reason, you can do so safely.

 Paul Prescod



From fcy at enteract.com  Mon Mar 19 12:44:33 2001
From: fcy at enteract.com (Fred Yankowski)
Date: 19 Mar 2001 17:44:33 GMT
Subject: VPython/Tk app works on single processor but not dual
Message-ID: <995gi1$9ce$1@bob.news.rcn.net>

I'm working on a VPython app that gives the user a Tk window with
sliders that they can use to control the application.  This program
works fine on WinNT if I "set affinity" of the python.exe process so
that it's forced to run on a single processor, but if I let it run
normally on my dual-processor system, the application hangs in short
order, refusing to respond to mouse actions in either the Tk or
VPython display windows.

My guess is that the multiple threads involved in this program are
deadlocking [well, duh].  My program seems to hang at the point where
it sets the .visible member of a VPython object to 0, causing VPython
to take the object out of the display list.  VPython uses several
read-locks while doing the display list changes, and explicitly messes
with the reference count managed by the underlying Python 'CXX'
package, which in turn wraps the yet-further-underlying Python ref
count.  But I don't see the cause of the deadlock.

So, I'm looking for ways to work around this problem.  At a minimum,
is there some way within a Python program to "set processor affinity"
so that all threads of the process are forced to run on the same
processor?  I couldn't find anything related to this in the Python
library manual.

More deeply, what is the current state of Python support for
multithreading over multiple processors?  Is this combination --
Python 2.0, VPython, tkinter -- known to have problems with multiple
processors?

The program itself is available, for now, at
<http://ontosys.com/src/residue.py>.  I'm using Python 2.0 and VPython
2001-02-27 on NT4SP6, on dual Pentium-III 800 processors.  I have
version 8.3.2 of tk83.dll and tcl83.dll.

--
Fred Yankowski





From jmg at ecs.soton.ac.uk  Wed Mar 28 11:57:09 2001
From: jmg at ecs.soton.ac.uk (Jacek Generowicz)
Date: 28 Mar 2001 17:57:09 +0100
Subject: Boost documentation.
Message-ID: <g0wv99x256.fsf@scumbag.ecs.soton.ac.uk>

Are there any sources of documentation for the Boost Python Library
besides the source examples, getting_started<1-5>.cpp ?

How would I write a C++ function which receives a python list and
converts it into a std::vector<int> for its own use ?

Thanks,

Jacek


From wcuddy at lserv.ja10629.home  Tue Mar 27 11:46:19 2001
From: wcuddy at lserv.ja10629.home (Wayne Cuddy)
Date: Tue, 27 Mar 2001 16:46:19 GMT
Subject: large arbitrary binary data in python
Message-ID: <slrn9c1hf7.m5q.wcuddy@lserv.ja10629.home>

I am using linux for this project..

I am currently developing a python script to upgrade an embedded system and
have run into a road block.  I need to contact an arbitrary ftp or http server
and grab a rather large (several megabyte) binary file and an accompanying md5
checksum file.  I need to verify the checksum, then expand the gzipped binary
file and write it to a device with the dd command.

Now, I don't have a filesystem on the embedded system big enough to receive
the binary file.  I do have 512Megs of memory in this system, so can I do this
all in python, and can I do this all in memory.  It is possible to do the
following?

    1) simply read the data into a python binary string
    2) pipe the string to md5sum to verify integrity
    3) pipe the data to a shell command like this "gunzip -dc | dd of=device"

    How will python handle the large amount of binary data?  Is there a limit to
    the size of a binary string in python?

Much thanks in advance,
Wayne



From stephen_purcell at yahoo.com  Sat Mar 31 12:20:34 2001
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Sat, 31 Mar 2001 19:20:34 +0200
Subject: restricted execution
In-Reply-To: <Z05x6.1894$bY4.6478@www.newsranger.com>; from kendall@_drakealumni_.net on Fri, Mar 30, 2001 at 07:21:29PM +0000
References: <Z05x6.1894$bY4.6478@www.newsranger.com>
Message-ID: <20010331192033.A12181@freedom.puma-ag.com>

Kendall wrote:
> 
> dict = {}
> for i in AllowedBuiltinList :
> exec "dict['" + i + "'] = " + i
> 
> __builtins__ = dict
> del dict
> 
> Where the AllowedBuiltinList is missing things like __import__, open, and
> eval.
> If I paste this in every user defined module, have I effectively prevented all
> access to the file system and network?  I'm not letting them import sys or
> os in their module either.


Have you tried to use the built-in module 'rexec' for such restricted
execution support? You might find it easier.

-Steve

-- 
Steve Purcell, Pythangelist
Get testing at http://pyunit.sourceforge.net/
Any opinions expressed herein are my own and not necessarily those of Yahoo



From db3l at fitlinxx.com  Tue Mar 13 20:52:24 2001
From: db3l at fitlinxx.com (David Bolen)
Date: 13 Mar 2001 20:52:24 -0500
Subject: Want to catch pywintypes.api_error
References: <98kftn$jf9$1@newsreaderm1.core.theplanet.net> <3AAE208D.4010103@ActiveState.com> <98llta$2ea$1@newsreaderg1.core.theplanet.net>
Message-ID: <ud7bl15if.fsf@ctwd0143.fitlinxx.com>

"Franz GEIGER" <fgeiger at datec.at> writes:

> How could I have figured out this by myself? I did a text search for
> api_error over my ActivePython distribution, w/o success. Is it burrie din
> those pyd-files?

I'm not sure if it's explicitly documented, but one good way is to do
a dir() of the module at hand - you'll find an "error" object in both
pywintypes and the individual modules such as win32api.

It is a bit confusing that the internal class is called api_error (and
that's how the instance objects display themselves) but the actual
instance is called error.  But having an "error" object in a module
for exceptions is reasonably common, so a good thing to check for.

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


From bawolk at ucdavis.edu  Wed Mar 28 21:09:37 2001
From: bawolk at ucdavis.edu (Bruce Wolk)
Date: Wed, 28 Mar 2001 18:09:37 -0800
Subject: setvbuf() ??
References: <slrn9c50mp.me7.wcuddy@lserv.ja10629.home>
Message-ID: <3AC29961.9937EC59@ucdavis.edu>

Putting a trailing comma after the print statement eliminates the newline, but
adds a trailing space.  Would periods separated by spaces work?

I would be curious if anyone knows how to eliminate the trailing space.

Bruce

Wayne Cuddy wrote:

> I need to print a period on an xterm every few seconds to provide status to the
> user, I don't want to force a newline.  How do I turn off line buffering for
> standard out?



From aleaxit at yahoo.com  Fri Mar  2 04:17:15 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 2 Mar 2001 10:17:15 +0100
Subject: Collection interfaces
References: <3a90fa97.0@news1.mweb.co.za> <3A933830.7F98D118@ix.netcom.com> <CgJk6.515$v77.66568@newsread2.prod.itd.earthlink.net> <220220010813598042%cmh@bDistributed.com> <ud7cakvi8.fsf@roundpoint.com> <MPG.150312ceb68f1ada98969f@news.earthlink.net> <8Gem6.473353$U46.14105062@news1.sttls1.wa.home.com> <MPG.1503548620e632919896a4@news.earthlink.net> <MCjm6.474652$U46.14163288@news1.sttls1.wa.home.com> <MPG.150394195f7f04a19896a7@news.earthlink.net> <3A9A86B1.9A8C69A2@ix.netcom.com> <MPG.1504d14563dce6499896af@news.earthlink.net> <97fjla128ir@news2.newsguy.com> <MPG.1505a76831c34b749896b3@news.earthlink.net> <97h5s40vhc@news2.newsguy.com> <MPG.1506007591cf818b9896bc@news.earthlink.net> <97ifd60sen@news1.newsguy.com> <3A9EEDD8.9102C3A@cosc.canterbury.ac.nz>
Message-ID: <97noer019d7@news1.newsguy.com>

"Greg Ewing" <greg at cosc.canterbury.ac.nz> wrote in message
news:3A9EEDD8.9102C3A at cosc.canterbury.ac.nz...
> Alex Martelli wrote:
> >
> > the indisputable success
> > of Smalltalk does show that single-inheritance is *practically
> > workable* for many tasks
>
> Smalltalk is like Python in that its inheritance hierarchy
> is more about implementation inheritance than subtype
> relationships. So I don't think you can conclude much
> from Smalltalk about whether single inheritance is adequate
> for classification purposes.

Sorry for lack of clarity (but I do believe the snippet you
quote here was clearer in context): I do *NOT* think "single
inheritance" (taxonomy) is good for _classification_; I +do+
believe it's _practically workable_ ('adequate', if you will)
for many actual/practical programming, aka implementation,
tasks (although multiple inheritance, allowing mixin, is often
handier, using delegation won't destroy your productivity,
particularly if you can automate it with a little bit of
metaprogramming/introspection black magick).

I consider this similar to single dispatch (as in Python or
Smalltalk) vs multimethods (as in Dylan)... the latter is a
superior approach, but, _in practice_, one can get away with
the (simpler) former approach more often than not.  (Actually,
if I somehow had to choose between multimethods and multiple
inheritance for practical use, I think I'd take multimethods;
multiple-dispatching issues often prove less tractable, without
direct language support, than implementation multi-inheritance
issues do, in my actual programming experience).


Alex





From stirzel at mailhost.stg.rt.daimlerchrysler.com  Wed Mar 14 10:47:28 2001
From: stirzel at mailhost.stg.rt.daimlerchrysler.com (Alexander Stirzel)
Date: Wed, 14 Mar 2001 16:47:28 +0100
Subject: String comparison
Message-ID: <3AAF9290.42BDAAD3@mailhost.stg.rt.daimlerchrysler.com>

Hi!

I have a string processname with "netscape" in it and I want to compare
it with the written word "netscape".

This are some lines of my source code:

(...)
processname = splitted.pop()
(... no the string netscape is in processname ...)
if ((processname=="netscape") and ((string.atof(cputimeerg))>45) and
((string.atof(cpupercwert))>45)):
(... the first comparison is false ... why? ...)

I printed processname just before this comparison and I got "netscape"
on screen. (With no spaces there, that's not the problem) But when I 
>>>> print (processname=="netscape")
I get a "0" on screen. And when I 
>>>> print (processname)
I get "netscape" on screen.

This is strange...

Thanks for any help.

-- 
Alexander Stirzel 
E-Mail: alexander.stirzel at daimlerchrysler.com
*** welcome to hell, here's your copy of windows ***


From kzaragoza at mediaone.net  Fri Mar 30 10:48:01 2001
From: kzaragoza at mediaone.net (Kris J. Zaragoza)
Date: Fri, 30 Mar 2001 15:48:01 GMT
Subject: Jython and Win2000
References: <mailman.985940862.22029.python-list@python.org>
Message-ID: <slrn9c9alm.tap.kzaragoza@kzaragoza.ne.mediaone.net>

On Fri, 30 Mar 2001 16:23:19 +0800, =?big5?B?sWm4zq9x?=
<winson at mdsserv.mds.com.tw> wrote:

>Maybe post to wrong place. But help me...

We'd be glad to.

>
>My win2000 cannot work with Jython. When I type jython under c>, =
>sometimes there nothing happened and response me a c>. Why ?? I =
>installed Sun's Java 1.3 SDK already.
>

You say that "sometimes" nothing happens.  Does the interpreter start
up some of the time, or not at all?  I have Jythin installed on both
Windows 2000 Professional and Windows 2000 Advanced Server.  I am even
embedding it into a Java application.  It runs fine in all
circumstances.

A few things to check:

1. Is your JDK properly installed?  Can you type "java" from the C:\>
prompt and have the Java interpreter properly start up?

2. Is the Jython startup batch file in your PATH?  You may want to
type "PATH" at the command prompt and check to see if the Jython
installation directory is listed.  If it's not, you may want to add it
to your PATH so you can run Jython from any directory.

>--
>Best Regards,
>
>Winson Chang
>
>3F, 2, 150 Lane, Sec. 5, Hsin I Rd., Taipei, Taiwan, R.O.C.
>
>Tel: +886-2-27225333 ext 376
>Fax: +886-2-27222330
>
>Email: winson at mdsserv.mds.com.tw
> 

I hope this helped!

-Kris

-- 
Kris J. Zaragoza           | "Unfortunately, most people can't out-think a
kzaragoza at mediaone.net     | grapefruit."  --Jon Bodner



From erno-news at erno.iki.fi  Sat Mar 24 17:49:50 2001
From: erno-news at erno.iki.fi (Erno Kuusela)
Date: 25 Mar 2001 00:49:50 +0200
Subject: syslog beeps and logs to console.... sometimes!?
References: <20010323.170444.1704365084.578@mead.troikanetworks.com>
Message-ID: <kulmpuwzn5.fsf@lasipalatsi.fi>

In article <20010323.170444.1704365084.578 at mead.troikanetworks.com>,
"Bruce Edge" <bedge at troikanetworks.com> writes:

| I use syslog for debug info. It beeps and prints a:
|   Message from syslogd at mead at Fri Mar 23 16:56:19 2001 ... mead cli:   
|   CTREE: tokenize "cr"

it is the syslog daemon. when it can't write the log for some
reason, it writes to the console. maybe your version
of syslogd is buggy, you could check for updates for your os.

maybe your syslog is not keeping up somehow. you could try rate
limiting logging and see if it changes anything, to further
diagnose the problem. but it's almost certainly not python-related.

  -- erno


From emile at fenx.com  Tue Mar 13 11:33:06 2001
From: emile at fenx.com (Emile van Sebille)
Date: Tue, 13 Mar 2001 08:33:06 -0800
Subject: Converting a text data file from positional to tab delimited.
References: <B6D37840.2124%lee@joramo.com> <98lf1t02b1q@news1.newsguy.com>
Message-ID: <98li9l$2h6gg$1@ID-11957.news.dfncis.de>

I was in the mood for some obfustication this morning:
  (Note:  this will wrap.  there are three lines, one each for srceRecs,
fldmap and converted)

srceRecs = "aaabbbccc\nddd111222\n333444eee\n"
fldmap = [('fld1',0,3),('fld2',3,6),('fld3',6,9)]
converted = [ "".join ( [x + ['', '\t'][0 + z in [ p[2]-1 for p in
fldmap ] ] for x, z in map(None, list(srcerec) ,xrange(fldmap[-1][-1])) ])
for srcerec in srceRecs.split('\n')[:-1] ]


FWIW,

didn't-get-what-I-really-wanted-ly y'rs  ;-)

Emile van Sebille
emile at fenx.com





From bedge at troikanetworks.com  Tue Mar  6 18:05:14 2001
From: bedge at troikanetworks.com (Bruce Edge)
Date: Tue, 06 Mar 2001 15:05:14 -0800
Subject: package/namespace confusion
References: <3AA3EDD2.60393490@troikanetworks.com> <usnkqle6k.fsf@roundpoint.com>
Message-ID: <3AA56D2A.6D098DFA@troikanetworks.com>

It was the nested global thing that got me. I assumed global was global,
like C/tcl/everything else.
Maybe it should be named "parent" rather than global.

Thanks for the info, Bruce.

Ben Hutchings wrote:
> 
> Bruce Edge <bedge at troikanetworks.com> writes:
> 
> > Hi,
> > I'm sure this is a real rookie issue here, but I can't seem to define a
> > global then access it in a package.
> > I'm using __init__.py to define glob_var, then load util.py.
> >
> > I would have expected util/ufunc() to have access to glob_var.
> >
> > test/__init__.py:
> >
> > glob_var=0        <- is this where I should stick package globals?
> 
> Yes.
> 
> > __all__=["util"]
> >
> >
> > test/util.py:
> >
> > def ufunc():
> >       global glob_var
> <snip>
> 
> The global namespace for this function is the namespace of the module
> (test.util) not that of the package (test).
> 
> If you want package globals to be module globals too, then use
> 'from test import *' at the top of your modules.  Alternatively you
> can 'import test' and refer to 'test.glob_var'.
> 
> --
> Any opinions expressed are my own and not necessarily those of Roundpoint.


From scarblac at pino.selwerd.nl  Wed Mar  7 19:32:47 2001
From: scarblac at pino.selwerd.nl (Remco Gerlich)
Date: 8 Mar 2001 00:32:47 GMT
Subject: learning C
References: <mailman.983972829.23481.python-list@python.org> <983975751.18833@newsmaster-04.atnet.at>
Message-ID: <slrn9adklr.bv9.scarblac@pino.selwerd.nl>

Werner Schiendl <ws-news at gmx.at> wrote in comp.lang.python:
> > I assumed the original poster wanted to become a good C code writer, or at
> > least a good C code reader.  You tell him to learn C++ instead, as if
> learning
> > C is obselete.
> 
> If it is not obsolete, it should be at least considered deprecated for new
> projects.

A baseless statement. There are many types of project for which C is still
the best choice. For one thing, think embedded systems.

I can imagine something like Java deprecating C++, but C isn't going to go
away soon.

-- 
Remco Gerlich


From jmg at ecs.soton.ac.uk  Wed Mar 14 06:11:09 2001
From: jmg at ecs.soton.ac.uk (Jacek Generowicz)
Date: 14 Mar 2001 11:11:09 +0000
Subject: Extending Python: rewriting a single method in C
References: <g0wv9vxms0.fsf@scumbag.ecs.soton.ac.uk> <slrn9apn8f.382.jepler@potty.housenet> <g0ofv6xbj4.fsf@scumbag.ecs.soton.ac.uk> <98l1i901k4u@news1.newsguy.com> <g0lmq9xxz0.fsf@scumbag.ecs.soton.ac.uk> <98m6g20fgv@news1.newsguy.com> <g0g0ggy7ct.fsf@scumbag.ecs.soton.ac.uk> <Xns90646C7A9204Dduncanrcpcouk@127.0.0.1>
Message-ID: <g0ae6oy59u.fsf@scumbag.ecs.soton.ac.uk>

duncan at rcp.co.uk (Duncan Booth) writes:

> Jacek Generowicz <jmg at ecs.soton.ac.uk> wrote in 
> <g0g0ggy7ct.fsf at scumbag.ecs.soton.ac.uk>:
> 
> >>>> print a.__getattr__("bar")
> >Traceback (most recent call last):
> >  File "<stdin>", line 1, in ?
> >AttributeError: 'foo' instance has no attribute '__getattr__'
> >
> >
> What you want is:
> >>> print getattr(a, "bar")
> 4

Ha ha ha ha !

Doh!

I definitely tried

   a.__getattr__("bar")
   __getattr__(a,"bar")

AND

   a.getattr("bar")

but not

   getattr(a, "bar").

Thanks.

Jacek


From aleb at gmx.net  Fri Mar 23 12:34:49 2001
From: aleb at gmx.net (Alex Berger)
Date: Fri, 23 Mar 2001 18:34:49 +0100
Subject: pyncurses documentation??
Message-ID: <3abb8a08$0$16654@SSP1NO25.highway.telekom.at>

Does anybody know where I can get an introduction/documentation to 
pyncurses??

Thanks
Alex


From grisha at ispol.com  Sat Mar  3 22:39:21 2001
From: grisha at ispol.com (Gregory (Grisha) Trubetskoy)
Date: Sat, 3 Mar 2001 22:39:21 -0500
Subject: python cgi syntax highlighter
Message-ID: <Pine.BSF.4.32.0103032234260.88338-100000@localhost>

Hello -

I remember coming across a script that displayed python code in HTML with
color syntax highlighting - does anyone know where I can find one? My
searches don't turn up anything.

Thanks,

Grisha



From rikbo716 at student.liu.se  Thu Mar 22 08:14:49 2001
From: rikbo716 at student.liu.se (Rikard Bosnjakovic)
Date: Thu, 22 Mar 2001 14:14:49 +0100
Subject: Good Editor?
References: <3ab9a32b$1_1@news2.one.net>
Message-ID: <3AB9FAC9.97EA420E@student.liu.se>

fjs wrote:
 
> Can someone recomend a good editor to write python in?  Im just getting
> started learning python and dont know which ones support python.

I prefer Emacs and python-mode, it's really neat. But I know a lot of
guys who won't touch Emacs even with a long stick :)



-- 
Rikard Bosnjakovic - http://a214.ryd.student.liu.se/cv/ - ICQ: 1158217

Anyone sending unwanted advertising e-mail to my address will be
charged $250 for network traffic and computing time. By extracting my
address from this message or its header, you agree to these terms.


From fredrik at pythonware.com  Fri Mar  2 02:40:16 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 02 Mar 2001 07:40:16 GMT
Subject: Intersesting Questions involving extending and embedding
References: <3a9c6f89.9613214@news.digital.net> <0D1n6.15927$Qb7.2600092@newsb.telia.net> <3a9f0af7.37387355@news.digital.net> <508n6.15966$Qb7.2607706@newsb.telia.net> <3a9dc27e.1325053@news.digital.net> <SYln6.16080$Qb7.2629142@newsb.telia.net> <3a9f00e8.2826131@news.digital.net>
Message-ID: <A7In6.16279$Qb7.2658942@newsb.telia.net>

"Dev_NuLL" wrote:
> A complied C module is much faster than a Py or PYC and when you are
> dealing with complex graphics code you need all the speed you can get,
> but I guess you cannot help me here.

I don't think anyone can help someone who wants to compile
Python source code with a C compiler, without translating the
code in any way.

But I'll make one last attempt:

- If you have Python source code (as you said in your first post),
Ship it as a PYC file, or convert it to a C module using python2c
or freeze-style techniques.

- If you're writing C or C++ code, you're *extending* Python.
Read the docs again, and look as the samples shipped with the
source distribution.  There's some 80 extension modules in the
Modules directory, so there should be plenty of inspiration.

Cheers /F




From qrczak at knm.org.pl  Sun Mar 11 03:56:42 2001
From: qrczak at knm.org.pl (Marcin 'Qrczak' Kowalczyk)
Date: 11 Mar 2001 08:56:42 GMT
Subject: Can anyone recomend a good intoduction to C...
References: <mailman.983894288.19965.python-list@python.org> <ewwp6.1466$y6.241005@ruti.visi.com> <slrn9aeblb.cg1.scarblac@pino.selwerd.nl> <djagat4kk3cs9hg0c3k39cudk6goblhkg6@4ax.com> <989mbk$24nuv$1@fido.engr.sgi.com> <slrn9ah66m.k09.qrczak@qrnik.zagroda> <98auhn0270j@news2.newsguy.com> <slrn9aiaq7.rmu.qrczak@qrnik.zagroda> <98dpoe12ckn@news1.newsguy.com> <3AAA85C1.81FB90AA@shore.net> <98eb1f02uqq@news1.newsguy.com> <3AAAE0BB.CBCC99EC@alcyone.com>
Message-ID: <slrn9amfea.u2j.qrczak@qrnik.zagroda>

Sat, 10 Mar 2001 18:19:40 -0800, Erik Max Francis <max at alcyone.com> pisze:

> Note that modern version of egcs have all the Standard language
> _features_ all settled out, but there are a few Standard compliance
> problems in the library.

What about the 'export' keyword?

-- 
 __("<  Marcin Kowalczyk * qrczak at knm.org.pl http://qrczak.ids.net.pl/
 \__/
  ^^                      SYGNATURA ZAST?PCZA
QRCZAK


From grante at visi.com  Sat Mar 31 22:05:12 2001
From: grante at visi.com (Grant Edwards)
Date: Sun, 01 Apr 2001 03:05:12 GMT
Subject: Conditional operator in Python?
References: <3AC68CB8.22AAF7B8@alcyone.com>
Message-ID: <slrn9cddsg.rp.grante@tuxtop.visi.com>

On Sat, 31 Mar 2001 18:04:40 -0800, Erik Max Francis <max at alcyone.com> wrote:

>Why isn't there a conditional operator (a ? b : c) in Python?  

Because trying to come with a semantically identical construct
in Python provides many hours of fun and a never-ending Usenet
thread.  ;)

>I've been reading comp.lang.python for a while (since just before 2.0
>shipped) and I can't really remember much discussion about it.  Is it a
>perennial request and I'm just missing it?

It comes up every month or two.

-- 
Grant Edwards                   grante             Yow!  Here I am at the flea
                                  at               market but nobody is buying
                               visi.com            my urine sample bottles...


From gerhard.nospam at bigfoot.de  Sat Mar 17 12:47:02 2001
From: gerhard.nospam at bigfoot.de (Gerhard =?iso-8859-1?Q?H=E4ring?=)
Date: Sat, 17 Mar 2001 18:47:02 +0100
Subject: Looking for web indexer in python
References: <roy-AD5737.12020617032001@news.panix.com>
Message-ID: <3AB3A316.B28D1945@bigfoot.de>

Roy Smith wrote:
> 
> The software development group I work in has a small internal web site (a
> few hundred documents) which we don't currently have any way of searching.
> I'd like to put an automated indexer/search system in place.  If possible,
> I'd like to use python, since I think it would be a nifty way to introduce
> the group to the language.  There's no funding for this, so it has to be
> free software.
> 
> A cursory look around didn't turn up anything that would fit the bill, but
> I have to assume somebody's already done a free python web indexer.  Any
> suggestions?

Strange coincidence. A few hours ago I was also looking for another
Python indexer to put here:

http://dmoz.org/Computers/Programming/Languages/Python/Modules/

This one http://sourceforge.net/projects/pymifluz/ was already in the
directory. I had some reservations, because the Sourceforge project page
says it's in pre-alpha stage. <sarcasm>Conveniently, it doesn't include
a link to the library it is interfacing</sarcasm> But mifliz can be
found on freshmeat.net.

The only other thing I found is included in ZOPE. I don't remember the
name, and I also don't know if it can be used standalone. Anybody knows
more on this part of ZOPE?

Gerhard
-- 
Sorry for the fake email, please use the real one below to reply.
contact: g e r h a r d @ b i g f o o t . d e
web:     http://highqualdev.com


From phd at phd.fep.ru  Fri Mar 30 06:11:10 2001
From: phd at phd.fep.ru (Oleg Broytmann)
Date: Fri, 30 Mar 2001 15:11:10 +0400 (MSD)
Subject: HELP urllib POST can't get it to work
In-Reply-To: <87snjvtths.fsf@winnegan.de>
Message-ID: <Pine.LNX.4.33.0103301510170.29458-100000@phd.fep.ru>

On 30 Mar 2001, Siggy Brentrup wrote:
> >From looking at 2.1b1 lib/urllib.py: use a list of pairs instead of a
> dictionary (untested)
>
> params = urllib.urlencode( [('address','07'), ('identity','me'),

   Wow! They finally implemented this!! I asked for this years ago... but
they rejected my patch that time :)

Oleg.
----
     Oleg Broytmann            http://phd.pp.ru/            phd at phd.pp.ru
           Programmers don't die, they just GOSUB without RETURN.




From gregj at pobox.com  Thu Mar  1 19:49:24 2001
From: gregj at pobox.com (Gregory Jorgensen)
Date: Fri, 02 Mar 2001 00:49:24 GMT
Subject: define a new func on the fly?
References: <3A9EE927.8744154E@troikanetworks.com>
Message-ID: <o6Cn6.846$45.1115@www.newsranger.com>

eval() evaluates an expression. exec() executes a statement. Try this:

>>> exec("def f():\n\treturn 'hello'\t")
>>> f()
'hello'

>>> eval("f()")
'hello'


In article <3A9EE927.8744154E at troikanetworks.com>, Bruce Edge says...
>
>I'm trying to define a new function on the at runtime,
>
>this fails:
>
>>>> eval( "def global_func_name():\n\tpass" )
>Traceback (most recent call last):
>  File "<stdin>", line 1, in ?
>  File "<string>", line 1
>    def global_func_name():
>      ^
>SyntaxError: invalid syntax
>
>any idea how I can do this?
>
>
>
>Thanks, Bruce.

Greg Jorgensen
Deschooling Society
Portland, Oregon, USA
gregj at pobox.com


From qrczak at knm.org.pl  Tue Mar 27 16:44:56 2001
From: qrczak at knm.org.pl (Marcin 'Qrczak' Kowalczyk)
Date: 27 Mar 2001 21:44:56 GMT
Subject: Changing every other instance of <B> in a file
References: <mailman.985726476.22406.python-list@python.org>
Message-ID: <slrn9c22em.6tv.qrczak@qrnik.zagroda>

Tue, 27 Mar 2001 15:57:04 -0500 (EST), Bob Kline <bkline at rksystems.com> pisze:

> %s GOT_B
> %%
> <GOT_B>\<B\>    { printf("</B>"); BEGIN(0); }
> \<B\>           { printf("<B>"); BEGIN(GOT_B); }
> %%
> main()          { yylex(); }
> yywrap()        { return 1; }

Nice. I think Haskell is not bad too:

------------------------------------------------------------------------
data State = Outside | Inside
main = interact (translate Outside)

translate _        []              = []
translate Outside  ('<':'B':'>':s) = "<B>"  ++ translate Inside  s
translate Inside   ('<':'B':'>':s) = "</B>" ++ translate Outside s
translate state    (c:s)           = c      :  translate state   s
------------------------------------------------------------------------

Here is an obfuscated version:

import List; main = interact $ concat . snd . mapAccumL f 1 . tails
f x ('<':'B':'>':_) = (3-x, take x "</"); f x s = (x, take 1 s)

-- 
 __("<  Marcin Kowalczyk * qrczak at knm.org.pl http://qrczak.ids.net.pl/
 \__/
  ^^                      SYGNATURA ZAST?PCZA
QRCZAK


From tjg at exceptionalminds.com  Mon Mar 19 13:52:52 2001
From: tjg at exceptionalminds.com (Timothy Grant)
Date: Mon, 19 Mar 2001 10:52:52 -0800
Subject: name of tmpfile
In-Reply-To: <B7C74E1A6FBAD411BFC60002B309DDF4E429@mailbox.juilliard.edu>; from jsoons@juilliard.edu on Mon, Mar 19, 2001 at 12:13:50PM -0500
References: <B7C74E1A6FBAD411BFC60002B309DDF4E429@mailbox.juilliard.edu>
Message-ID: <20010319105252.A3109@trufflehunter.avalongroup.net>

When I need to actually know the name of a tempfile, I do
things in two steps:

fn = tempfile.mktemp()
t = open(fn)

Then you can use fn anywhere you need the name.

On Mon, Mar 19, 2001 at 12:13:50PM -0500, Jonathan Soons wrote:
> I am trying to find out the real name of
> 
> t = tempfile.TemporaryFile()
> 
> so that I can do
> 
> os.system('llp -d CHECKS -ofmt=checks <real_tempfile_name>')   # this is how
> I print to a printer
> 
> According to the book, os.tmpfile() never gets a real name, so I can't use
> that.
> 
> And p = tempfile.gettempprefix(t)
> 
> is supposed to put a reference to the real name in p. So
> 
> how do I pass the real name to os.system()?

-- 
Stand Fast,
    tjg.

Timothy Grant                         tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.         <><       (503) 246-3630
>>>>>>>>>>>>>Linux, because rebooting is *NOT* normal<<<<<<<<<
>>>>This machine was last rebooted:  61 days 23:06 hours ago<<



From cce at clarkevans.com  Wed Mar 21 05:01:50 2001
From: cce at clarkevans.com (Clark C. Evans)
Date: Wed, 21 Mar 2001 05:01:50 -0500 (EST)
Subject: Yet Another PEP:  Query Protocol Interface or __query__
In-Reply-To: <slrn9bgrsn.rv.hzhu@mars.localdomain>
References: <Pine.LNX.4.21.0103201837570.21343-100000@clarkevans.com> <998r4j02vmb@news2.newsguy.com> <Pine.LNX.4.21.0103202032080.14634-100000@clarkevans.com> <slrn9bgprm.rv.hzhu@mars.localdomain> <slrn9bgrsn.rv.hzhu@mars.localdomain>
Message-ID: <Pine.LNX.4.21.0103210447410.21377-100000@clarkevans.com>

On Wed, 21 Mar 2001, Huaiyu Zhu wrote:
> Python supports polymorphism in a very liberal way: if it walks like a duck
> and quacks like a duck, then it must be a duck.  The archetypical examples
> is that one can usually substitute a writable file with a user defined class
> object that happens to have a write() methods with appropriate property.

But it's even stronger than that.  In python if all you want is
duck-tracks, a class having a duck-bill isn't required only a 
class having webbed-feet.  ;)
 
> In many programs it is necessary to ensure that a certain objects has
> certain properties.  But how can we express the idea that a function expect
> a file-like object, while that can mean an endless list of things, many have
> not yet been defined?
> 
> One solution promoted by some other languages is to define some abstract
> classes for type checking purpose, and require them to be inherited by any
> class that behaves like them.  This is like what interfaces do in Java.
> This solution is not satisfactory because
> 
> - It breaks polymorphism.  One has to harness the property at the time the
>   classes are defined, instead of when they are used.  For example, if an
>   application requires a property of "ReadableFile" which is satisfied by
>   sys.stdin, it is difficult to retrofit the definition of sys.stdin to be
>   an instance of such.
> 
> - It mixes two incompatible inheritances.  For example, suppose that class A
>   has method A.m() but its subclass B overrides it with B.m().  The property
>   A satisfies is not necessarily satisfied by B.

Yes.  Very insightful.  It is easy to fall into this line of 
thinking and not see the the traps.  I was heasitent when 
I put in the "base class" stuff in the update to the PEP
being promoted.... it just seemed expected.  Perhaps it
isn't the right thing to do.  Once you have classes, 
people like to treat class inheritance as interface
inheritance... 

The rest of your post resonates deeply.  Is there some
way to get a "simple" __query__ mechanism working 
without having the type system in place?  

Clark




From cfelling at iae.nl  Mon Mar 26 20:09:52 2001
From: cfelling at iae.nl (Carel Fellinger)
Date: 27 Mar 2001 03:09:52 +0200
Subject: Is it possible to sort a dictionary?
References: <GEtv6.59583$Lm2.6378825@news0.telusplanet.net> <3ABE771B.7B9E1631@student.liu.se> <99o33k$aq4$0@216.39.170.247> <99o51301qgb@news1.newsguy.com>
Message-ID: <99op90$2tr$1@animus.fel.iae.nl>

Alex Martelli <aleaxit at yahoo.com> wrote:

> Fine!  Even faster & simpler:

Simpler yes, but faster? Are you sure?  Lately I tested something
similar and was surprised when I found the dict.items function to be
soo slow that going through the whole dict indexing each entry
one by one was faster.

So let me benchmark again ( and please be gently when you tell me
were I fucked up this time :)


The result were with python 1.5.2:

   tested t_items: 4.76
   tested t_keys:  3.9

and with python 2.0 an even more dramatic advantage for dict.keys:

   tested t_items: 4.6
   tested t_keys:  2.72

And the code used was:

import time

dict = {}
for i in range(1000):
    dict[i] = i

def test(f):
    start = time.clock()
    for i in range(1000):
        f()
    return time.clock() - start


def t_items(dict=dict):
    items = dict.items()
    items.sort()
    for key, value in items:
        v = value

def t_keys(dict=dict):
    keys = dict.keys()
    keys.sort()
    for key in keys:
        v = dict[key]

print 'tested t_items:', test(t_items)
print 'tested t_keys: ', test(t_keys)


-- 
groetjes, carel


From dfan at harmonixmusic.com  Thu Mar 22 09:46:33 2001
From: dfan at harmonixmusic.com (Dan Schmidt)
Date: 22 Mar 2001 09:46:33 -0500
Subject: Is there python code highlighting in emacs ?
References: <mailman.985223430.22832.python-list@python.org> <3AB978A9.353A0E4B@yahoo.com> <3AB97F1F.70FF97DD@velseis.com.au> <mailman.985235612.31492.python-list@python.org>
Message-ID: <wkpuf9rhdi.fsf@turangalila.harmonixmusic.com>

John McMonagle <jmcmonagle at velseis.com.au> writes:

| Add the following to your .emacs file:
| 
| (add-hook 'python-mode-hook '(lambda() (font-lock-mode 1)))

Or, if you want font-lock on for all modes that support it, just use:

  (global-font-lock-mode t)

And then you'll never have to turn any text coloring on by hand
again.

-- 
http://www.dfan.org


From aahz at panix.com  Fri Mar  9 21:51:35 2001
From: aahz at panix.com (Aahz Maruch)
Date: 9 Mar 2001 18:51:35 -0800
Subject: gah! I hate the new string syntax
References: <mailman.983615170.23313.python-list@python.org> <98asrq$645$1@panix6.panix.com> <G5eq6.47941$ML1.2142034@e420r-atl2.usenetserver.com> <jwbnews-8E2915.18132009032001@corp.supernews.com>
Message-ID: <98c4rn$flq$1@panix2.panix.com>

In article <jwbnews-8E2915.18132009032001 at corp.supernews.com>,
John W. Baxter <jwbnews at scandaroon.com> wrote:
>In article <G5eq6.47941$ML1.2142034 at e420r-atl2.usenetserver.com>,
> "Steve Holden" <sholden at holdenweb.com> wrote:
>> "Aahz Maruch" <aahz at panix.com> wrote in message
>> news:98asrq$645$1 at panix6.panix.com...
>>> In article <98amlj01sp6 at news2.newsguy.com>,
>>> Alex Martelli <aleaxit at yahoo.com> wrote:
>>>>
>>>>Wrong!  You're talking about people whose seafaring skills and
>>>>traditions are (deservedly) legendary.  It's perfectly acceptable
>>>>for either an English or a Scot to be born aboard a ship...!!!
>>>
>>> Since when are *women* allowed aboard ships?
>> 
>> Women were allowed aboard British ships in the eighteenth and nineteenth
>> centuries, I believe. The phrase "shake a leg" (meaning to stop lazing about
>> and start work) is held to have been used by ships' mates to get the seamen
>> out of their bunks. A woman would show her leg to let the mate know she
>> should not be rousted out to work.
> 
>And at least one of the US Clipper ships finished a New York to San 
>Fransicso voyage under the command of the Captain's widow.  The captain 
>died en route.  The wife was aboard as navigator (being better at the 
>task than the mate), and the crew made it clear that things would go 
>much better with her in command than with the mate in command.  Sorry, I 
>don't have a reference handy (and the right Google search for this one 
>could be a bit tricky to create).

But that's a US ship.  English Ships Are Different.
-- 
                      --- Aahz (Copyright 2001 by aahz at pobox.com)

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

"..., and some of you may regard all women as evil traps that exist only
to tease, torture, and suck out your very soul."  --DrMax


From thomas at xs4all.net  Fri Mar  2 19:08:43 2001
From: thomas at xs4all.net (Thomas Wouters)
Date: Sat, 3 Mar 2001 01:08:43 +0100
Subject: A counter-proposal to __future__ in PEP 236
In-Reply-To: <3A9FBE5C.95615241@sewardweb.com>; from lsommerer@sewardweb.com on Fri, Mar 02, 2001 at 09:38:04AM -0600
References: <j47l29lpo8.fsf@informatik.hu-berlin.de> <97ll2u02qmp@news2.newsguy.com> <3A9FBE5C.95615241@sewardweb.com>
Message-ID: <20010303010843.S9678@xs4all.nl>

On Fri, Mar 02, 2001 at 09:38:04AM -0600, Lloyd Sommerer wrote:

> I should also point out that the "from  __future__ import..." is also
> confusing to a newbie, but others have said that already.

And that's exactly as it should be: you shouldn't use the experimental
feature(s) if you're a newbie :)

-- 
Thomas Wouters <thomas at xs4all.net>

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



From thomas.heller at ion-tof.com  Fri Mar 30 10:04:25 2001
From: thomas.heller at ion-tof.com (Thomas Heller)
Date: Fri, 30 Mar 2001 17:04:25 +0200
Subject: [Distutils] PEP: Use site-packages on all platforms
References: <714DFA46B9BBD0119CD000805FC1F53B01B5ADD2@ukrux002.rundc.uk.origin-it.com>
Message-ID: <0a0c01c0b92d$078e7f60$e000a8c0@thomasnotebook>

It looks good to me.

Probably you should also post this to python-dev,
since most python core developers don't read
c.l.p nor distutils-sig.

Thomas




From jmarshal at mathworks.com  Mon Mar 26 09:34:51 2001
From: jmarshal at mathworks.com (Joshua Marshall)
Date: 26 Mar 2001 14:34:51 GMT
Subject: So what's wrong with __future__?  (Was Re: Why "from __future__" stinks ...)
References: <slrn9bdka2.3j6.hzhu@rocket.knowledgetrack.com> <LNBBLJKPBEHFEDALKOLCAEDKJIAA.tim.one@home.com> <mailman.985568441.15878.python-list@python.org>
Message-ID: <99nk2b$7sm$1@news.mathworks.com>

Timothy Grant <tjg at exceptionalminds.com> wrote:

> I'm glad that it appears that even the Python gods are
> implementing this hideous travesty under protest. It is truly
> the ugliest hack ever--far worse than lambda, which to some
> extent could have been expected.

Hey now.  Be nice to lambda.


From stephen_purcell at yahoo.com  Wed Mar 28 04:45:05 2001
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Wed, 28 Mar 2001 11:45:05 +0200
Subject: Jython and JPython Questions...How similar to Python, Tkinter?
In-Reply-To: <3ABFF14A.30D0096D@earthlink.net>; from rdsteph@earthlink.net on Wed, Mar 28, 2001 at 02:35:30AM +0000
References: <3ABFF14A.30D0096D@earthlink.net>
Message-ID: <20010328114505.E8325@freedom.puma-ag.com>

Ron Stephens wrote:
> Does Jython support Tkinter? OK, maybe that question as worded is
> nonsense, let me re-phrase it: If I code a program in Python and put a
> GUI front end on it with Tkinter, how hard would it be to do the same
> program in Jython? Or is there a better solution for Jython GUI's?

There is a better solution: the now-standard Swing java extension.

Personally, I don't see platform-dependent JNI-based packages as a good thing
for Java applications, so I wouldn't suggest jtkinter. (Clever though it
undoubtedly is.)

Now if someone ever comes up with a library that gives Swing/AWT a Python
interface, that would be another story...

-Steve

-- 
Steve Purcell, Pythangelist
Get testing at http://pyunit.sourceforge.net/
Any opinions expressed herein are my own and not necessarily those of Yahoo



From grante at visi.com  Fri Mar 30 10:57:13 2001
From: grante at visi.com (Grant Edwards)
Date: Fri, 30 Mar 2001 15:57:13 GMT
Subject: Creation d'un NG francophone sur Python
References: <1eqzyk6.gxd4jr1tkjykzN%francois.granger@free.fr> <9a1e6f011es@enews3.newsguy.com> <D9305864F80B9ADF.9B74F9BE0F743AE2.DC045B55BF4E1CD8@lp.airnews.net> <mailman.985964201.6565.python-list@python.org> <UX1x6.9226$SB2.838492@ruti.visi.com>
Message-ID: <t12x6.9227$SB2.838811@ruti.visi.com>

In article <UX1x6.9226$SB2.838492 at ruti.visi.com>, Grant Edwards wrote:

>Though as a mono-lingual American, I also feel the urge to
>
>P O S T   S L O W L Y   A N D   L O U D L Y   I F   I T   W I L L
>H E L P   P E O P L E   W H O   D O N ' T   R E A D   E N G L I S H.

I'd also like to point out that I have my compose key
configured, so I ?a? als? acc?nt things or post ze articles
using zilly accents if zat vill helpen vous.  :)

-- 
Grant Edwards                   grante             Yow!  Youth of today! Join
                                  at               me in a mass rally
                               visi.com            for traditional mental
                                                   attitudes!


From sales at tooltoad.com  Sat Mar 10 00:21:28 2001
From: sales at tooltoad.com (www.tooltoad.com)
Date: Sat, 10 Mar 2001 00:21:28 -0500
Subject: GRAND OPENING     www.tooltoad.com     GRAND OPENING
Message-ID: <0G9Y00JGMUP5W0@mta4.srv.hcvlny.cv.net>

www.tooltoad.com      www.tooltoad.com     www.tooltoad.com

HELLO ,
  
    Please visit us at the GRAND OPENING of www.tooltoad.com

Come and see our ROCK BOTTOM pennies on the dollar pricing . We sell 

electronics , housewares  , security items , tools , and much more . 



    			THANK YOU 
				The management





From phrxy at csv.warwick.ac.uk  Wed Mar 14 15:49:06 2001
From: phrxy at csv.warwick.ac.uk (John J. Lee)
Date: Wed, 14 Mar 2001 20:49:06 +0000
Subject: tiny extension crashes
Message-ID: <Pine.SOL.4.30.0103141957240.10526-100000@mimosa.csv.warwick.ac.uk>

I've been using a tiny little C extension module that I wrote for a while
now, and haven't had any trouble.  Just recently, I've been getting
crashes which go away when I use my pure Python implementations of these
functions.  I only get these crashes on windows (mingw32 compiler), not on
linux.  The other odd thing is that either of the two functions in the
extension seem to trigger the crashes on their own (or both together).
As far as I can see there is no need for any Py_INCREF or Py_DECREFfing in
these functions, but these are my first extensions, so I may be mistaken.

Can anyone help out, or tell me where to start looking?

Here is the whole C extension source file:



#include <Python.h>
#include "Numeric/arrayobject.h"

/* Non-Python functions
   ie, not callable from python. */

static int array_max(PyArrayObject *array, int absolute) {
    /* Return maximum of an integer multiarray */
    int i, max, elem = 0;
    for(i=0, max=0; i < array->dimensions[0]; i++) {
        elem = *(long *)(array->data + i*array->strides[0]);
        if (absolute)
            elem = fabs(elem);
        if (elem > max)
            max = elem;
    }
    return max;
}

/* Python functions */

static PyObject *iround(PyObject *self, PyObject *args) {
    /* round to integer type, 0.49->0, 0.5->1 */
    double number;
    double *intp;

    if (!PyArg_ParseTuple(args, "d", &number))
        return NULL;
    if (fabs(modf(number, intp)) >= 0.5) {
        *intp >= 0 ? (*intp)++ : (*intp)--;
    }
    return Py_BuildValue("l", (long)*intp);
}

static PyObject *hcf(PyObject *self, PyObject *args) {
    /* return highest common factor of a 1D numpy array
       XXX this algorithm is probably the worst possible
       XXX also probably needs to take general sequence arguments */
    PyArrayObject *array;
    long i, n, factor = 0, elem;
    long length;
    if (!PyArg_ParseTuple(args, "O!", &PyArray_Type, &array))
        return NULL;
    if (array->nd != 1 || (array->descr->type_num != PyArray_SHORT &&
                           array->descr->type_num != PyArray_LONG)) {
        PyErr_SetString(PyExc_TypeError,
                        "array must be 1D and of integer type");
        return NULL;
    }
    if ((length = array->dimensions[0]) == 0) {
        PyErr_SetString(PyExc_ValueError, "array must not be of zero length");
        return NULL;
    }
    /* start at highest element and count down to 1 */
    for (n=array_max(array, 1); n > 0; n--) {
        /* is it a factor of all the rest? */
        factor = n;
        for (i=0; i < length; i++) {
            elem = *(long *)(array->data + i*array->strides[0]);
            if (elem % n != 0) {
                factor = 0;
                break; /* not a factor, try n-1 */
            }
        }
        if (factor) break; /* found hcf */
    }
    return Py_BuildValue("l", (long)factor);
}

/* Boilerplate */

static PyMethodDef cutils_methods[] = {
    {"iround", iround, METH_VARARGS},
    {"hcf", hcf, METH_VARARGS},
    {NULL, NULL}
};

void initcutils(void) {
    (void) Py_InitModule("cutils", cutils_methods);
    import_array()
}


-- 
John



From nas at python.ca  Wed Mar 28 12:58:19 2001
From: nas at python.ca (Neil Schemenauer)
Date: Wed, 28 Mar 2001 09:58:19 -0800
Subject: Is inheritance broken?
In-Reply-To: <NDBBLAKACLGJJBDNGGCMKEINCCAA.jcc.ugm@ix.netcom.com>; from jcc.ugm@ix.netcom.com on Wed, Mar 28, 2001 at 12:30:26PM -0500
References: <NDBBLAKACLGJJBDNGGCMKEINCCAA.jcc.ugm@ix.netcom.com>
Message-ID: <20010328095819.B20281@glacier.fnational.com>

Jonathan Claggett wrote:
> It looks like the inherited angle method from rectangle is being ignored.
> Isn't this wrong? Or am I just missing something.

No, its not wrong.  You don't understand multiple inheritance in
Python.  I believe there is a section in the tutorial explaining
it.  A small clue is that it works depth-first, left to right.

   Neil



From tjg at exceptionalminds.com  Tue Mar 20 11:53:17 2001
From: tjg at exceptionalminds.com (Timothy Grant)
Date: Tue, 20 Mar 2001 08:53:17 -0800
Subject: New PEP: The directive statement
In-Reply-To: <200103201042.LAA20248@pandora.informatik.hu-berlin.de>; from loewis@informatik.hu-berlin.de on Tue, Mar 20, 2001 at 11:42:34AM +0100
References: <200103201042.LAA20248@pandora.informatik.hu-berlin.de>
Message-ID: <20010320085317.P3109@trufflehunter.avalongroup.net>

I am not one of the gods of Python, but I'm not Python
illiterate either. 

When I read PEP 236. I cringed. I understand the value, but the
solution seemed far from elegant, and in Python, most things
seem elegant (except for lambda<wink>.

When I read the below suggestion it made far more sense to me,
and seemed to be a more elegant solution to the problem.


On Tue, Mar 20, 2001 at 11:42:34AM +0100, Martin von Loewis wrote:
> Please have a look at the PEP below. It is intended to replace the
> __future__ import suggested in PEP 236.
> 
> Regards,
> Martin
> 
> PEP: XXX
> Title: The directive statement
> Version: $Revision$
> Author: Martin v. L?wis
> Python-Version: 2.1
> Status: Active
> Type: Standards Track
> Created: 20-Mar-2001

-- 
Stand Fast,
    tjg.

Timothy Grant                         tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.         <><       (503) 246-3630
>>>>>>>>>>>>>Linux, because rebooting is *NOT* normal<<<<<<<<<
>>>>This machine was last rebooted:  62 days 21:05 hours ago<<



From rossi at earthling.net  Sat Mar 17 10:19:42 2001
From: rossi at earthling.net (Thorsten Roskowetz)
Date: Sat, 17 Mar 2001 16:19:42 +0100
Subject: ZPT, a next-generation template technology
References: <98ufbu$kkh$1@uranium.btinternet.com> <slrn9b5j52.i70.mgjv@martien.heliotrope.home>
Message-ID: <98vv9e$96f$01$1@news.t-online.com>

"Martien Verbruggen" <mgjv at tradingpost.com.au> wrote:
> However, cl.perl.misc gets enough offtopic crap as it is. Don't post
> future announcements here.

You might not be aware of it but because of

http://www.zope.org/Wikis/zope-perl/

Hamish's post wasn't as off-topic as you think and certainly
not crap.

Regards,
Thorsten




From cce at clarkevans.com  Sun Mar 25 08:24:09 2001
From: cce at clarkevans.com (Clark C. Evans)
Date: Sun, 25 Mar 2001 08:24:09 -0500 (EST)
Subject: win32com, handling byref 
Message-ID: <Pine.LNX.4.21.0103250804090.26927-100000@clarkevans.com>

I am trying to call the following function from an ActiveX 
object.  It has a return value, plus three output parameters.
Unfortunately, the type library doesn't mark the last three
as "out", so I'm quite sure MakePy didn't generate a wrapper
which will help me.  

        long GetObjectAtXY(
                        [in] long x, 
                        [in] long y, 
                        [in] BSTR* pName,         /* output */
                        [in] BSTR* pLink,         /* output */
                        [in] long* plReserved);   /* output */

    def GetObjectAtXY(self, x=defaultNamedNotOptArg,
                            y=defaultNamedNotOptArg,
                        pName=defaultNamedNotOptArg,
                        pLink=defaultNamedNotOptArg,
                  plReserved=defaultNamedNotOptArg):
      """Get information about the object at a set of coordinates."""
      return self._oleobj_.InvokeTypes(0x5f, LCID, 1, (3, 0), 
             ((3, 1), (3,1), (16392, 1), (16392, 1), (16387, 1)), 
               x, y, pName, pLink, plReserved)

I was wondering if there is an incantation I can use which 
will allow me to do...

In Visual Basic, one would use:
   
    ret = GetObjectAtXY(x,y,name,link,res)

I'd like to do the following in Python...

    ret, name, link, res = GetObjectAtXY(x,y)

Is this even remotely possible?  If not, is there a description
of the low-level hoops that I could possibly jump through?

Thank you so much.

Clark




From renzo.tomaselli at tecnotp.it  Tue Mar 20 04:20:09 2001
From: renzo.tomaselli at tecnotp.it (Renzo Tomaselli)
Date: Tue, 20 Mar 2001 10:20:09 +0100
Subject: relational/boolean operator overloading -> PEP 207
References: <98tjq6$qre$1@news.flashnet.it> <98tk4b$sbv$1@panix3.panix.com> <994h71$ohq$1@news.flashnet.it> <996r1u$67f$1@panix6.panix.com>
Message-ID: <9977ap$l05$1@news.flashnet.it>

Thanks to all answering my questions.
Nevertheless, even when forgetting about my topic for a while, I still
cannot understand how PEP 207 will handle 'and/or' operators applied to
"generic" objects which the new relational overloadings could return
anytime. E.g. how an expression such as "obj1 and obj2" will be evaluated by
Python 2.1.

Renzo Tomaselli

Aahz Maruch wrote in message <996r1u$67f$1 at panix6.panix.com>...
>There's some stuff being done on the numeric side of Python that may be
>relevant (in terms of set operators and such).  I don't know enough
>about it to give you a good search term, though.





From syver at NOSPAMcyberwatcher.com  Fri Mar 30 15:20:44 2001
From: syver at NOSPAMcyberwatcher.com (Syver Enstad)
Date: Fri, 30 Mar 2001 22:20:44 +0200
Subject: Python scripts as Drop targets in Windows Explorer
References: <99qogo$o6g$1@troll.powertech.no> <99qqq3$qt7$0@216.39.170.247> <99r043$qqr$1@troll.powertech.no> <vdqs99.p16.ln@haegar.fqdn.th-h.de> <8kAw6.780$R6.22057@news1.oke.nextra.no> <9a055c$n2q$1@216.39.170.247>
Message-ID: <9a2puv$man$1@troll.powertech.no>

Most strange, the attachment opens in Emacs on my machine. I'll send it
inline then for hopefully the last time:

REGEDIT4

[HKEY_CLASSES_ROOT\Python.File\shellex\DropHandler]
@="{86C86720-42A0-1069-A2E8-08002B30309D}"

"Dave LeBlanc" <whisper at oz.net> wrote in message
news:9a055c$n2q$1 at 216.39.170.247...
> Yes, i'm using NT 4.0 currently.
>
> The second posting where you said you where putting an ascii version
> was also in "gooberish"
>
> Regards,
>
> Dave LeBlanc
>
> On Thu, 29 Mar 2001 08:10:08 +0200, "Syver Enstad"
> <syver at NOSPAMcyberwatcher.com> wrote:
>
> >
> >"Michael Linnemann" <m.linnemann at addcom.de> wrote
> >
> >> Yeah, looks cool. What is it, perl? Czech? EBCDIC?
> >As it said in the posting:
> >>>default on Win2k is to save as unicode.
> >> Come on, tune your W2k a little and make it talk plain :-)
> >I don't know what you mean, isn't Unicode one of the most plain ways to
talk
> >nowadays? And I have also posted it as plain text so it shouldn't be too
> >difficult for you.
> >
> >
> >
>




From aleaxit at yahoo.com  Sat Mar 10 14:22:30 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sat, 10 Mar 2001 20:22:30 +0100
Subject: Python In A Nutshell
References: <3AB1700C@MailAndNews.com> <98cn2d$oad$1@newsreaderg1.core.theplanet.net> <obsq6.62831$__6.12513082@typhoon.southeast.rr.com>
Message-ID: <98e8ve02sqq@news1.newsguy.com>

"Don Tuttle" <tuttledon at hotmail.com> wrote in message
news:obsq6.62831$__6.12513082 at typhoon.southeast.rr.com...
    [snip]
> One last thought.  I wonder it the reason we haven't heard from Alex is
that
> there's no Unicode mapping for hand gestures? <wink>  (i.e. Just pick'n
;-)

Actually, it's been a long tussle with WinME's Internet Connection
Sharing, which I finally gave up in favor of a mix of Sygate, AnalogX,
and sockscap, but burned up all of my time until shortly ago:-) [the
poor guys who administer Win networks are underpaid, whatever it
IS that they're paid!-)].

But, yes, the ability to communicate typical italian gestures over
the net _would_ be helpful at times:-).


Alex





From gregj at pobox.com  Thu Mar  1 19:18:05 2001
From: gregj at pobox.com (Gregory Jorgensen)
Date: Fri, 02 Mar 2001 00:18:05 GMT
Subject: catching a del
References: <j%zn6.322252$f36.11513719@news20.bellglobal.com>
Message-ID: <1FBn6.837$45.1097@www.newsranger.com>

Add a delete method to the parent class; that seems the most direct solution.

Unlike "delete" in C++, Python's "del" statement doesn't delete an object, it
decrements the object's reference count and removes the deleted name.


In article <j%zn6.322252$f36.11513719 at news20.bellglobal.com>, Dan Parisien
says...
>
>class obj:
>        def __init__(self, name, parent):
>                parent.children[name] = self
>                self.parent = parent
>                self.name = name
>
>        def onDelete(self): #?!
>                del self.parent[self.name]
>
>I want to be able to catch a del of an instance of the above class so I can 
>remove it from its parent dictionary. Is there a way? (note it cannot be in 
>a destructor because there will always be a reference inside the parent 
>dict)

Greg Jorgensen
Deschooling Society
Portland, Oregon, USA
gregj at pobox.com


From loewis at informatik.hu-berlin.de  Fri Mar  2 10:42:09 2001
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 02 Mar 2001 16:42:09 +0100
Subject: A counter-proposal to __future__ in PEP 236
References: <mailman.983479743.12289.python-list@python.org> <97meh4$c0g$1@news.mathworks.com>
Message-ID: <j4ofvkjhym.fsf@informatik.hu-berlin.de>

Joshua Marshall <jmarshal at mathworks.com> writes:

> So anybody care to post an example of how this can be ambiguous (in
> context) so I don't need to go study Python's parser...?

>From the language level, it is not ambiguous. On the parser level, you have

stmt: simple_stmt | compound_stmt

and then

simple_stmt: small_stmt (';' small_stmt)* [';'] NEWLINE

small_stmt: expr_stmt | print_stmt | del_stmt | pass_stmt | flow_stmt
            | import_stmt | global_stmt | exec_stmt | assert_stmt

expr_stmt: testlist (augassign testlist | ('=' testlist)*)

A few productions later, you find that testlist can be 

atom: NAME | <more alternatives>

So when you write 'a = b', then a and b are NAME tokens, which reduce
to testlists, which then reduces to an expr_stmt, and a stmt. Your
proposal is

small_stmt: <all alternatives so far> | directive_stmt
directive_stmt: NAME NAME atom [';']

Now, the Python parser is recursive-descent, meaning that it needs to
look at the first token to descend into an alternative. If you look at
all stmt's, you find that you can pick an alternative just by looking
at the first token. E.g.

for, while, do, if, try, class, def -> compound_stmt
NAME, print, del, pass, break, continue, return, raise, import,
      global, exec, assert -> simple_stmt

Inside simple_stmt, you alway can descend to small_stmt, where you can
pick an alternative based on the first token. If you add the
directive_stmt, both expr_stmt and directive_stmt would start with a
NAME token, and you'd have to look at the next token to find out which
case it is. Since the parser is LL(1), it cannot do that.

Regards,
Martin


From ransen_spam_me_not at nemo.it  Sun Mar  4 01:38:14 2001
From: ransen_spam_me_not at nemo.it (Owen F. Ransen)
Date: Sun, 04 Mar 2001 06:38:14 GMT
Subject: Bug in Py_Finalize() ?
References: <mailman.983652125.10274.python-list@python.org>
Message-ID: <3aa1daff.724655@news.newsguy.com>

On Sat, 3 Mar 2001 15:41:18 -0500, "Tim Peters" <tim.one at home.com>
wrote:

>> I downloaded the Windows installer for 2.1b1 but there does not
>> appear to be any debug libraries...
>Sorry, we never ship debug stuff with the Windows installer, because it's yet
>another megabyte of stuff to download that the vast majority of Windows users
>have no use for.  

>    https://sourceforge.net/project/showfiles.php?group_id=5470
>Get file Python-2.1b1-Debug.zip.
Many thanks.


>Does your program run OK under the release build we did ship?
I'll check.

>People who want debug stuff are by definition developers,
>and so are strongly encouraged to build Python from source (ain't hard!).
>Nevertheless, I just uploaded all the 2.1b1 debug stuff to SourceForge:
I fully intend to do that eventually, but I'm just starting
out with both Python and embedding and extending, so I have
to do one thing at a time!


--
Owen F. Ransen
http://www.ransen.com/
Home of Gliftic & Repligator Image Generators


From f8dy at diveintopython.org  Fri Mar  2 17:45:33 2001
From: f8dy at diveintopython.org (Mark Pilgrim)
Date: Fri, 02 Mar 2001 17:45:33 -0500
Subject: Thinking Pythonically (was Re: gah! I hate the new string
	syntax)
References: <mailman.983485923.1136.python-list@python.org> <B7In6.16281$Qb7.2658902@newsb.telia.net> <97ohh2$polt0$1@ID-77331.news.dfncis.de> <97opsb$nhu$1@nntp6.u.washington.edu> <97oupq$qijrh$1@ID-77331.news.dfncis.de> <97p3mh$bkm$1@tyfon.itea.ntnu.no>
Message-ID: <B6C58CBD.212F%f8dy@diveintopython.org>

in article 97p3mh$bkm$1 at tyfon.itea.ntnu.no, Magnus Lie Hetland at
mlh at idi.ntnu.no wrote on 3/2/01 4:31 PM:
> "Mark Pilgrim" <f8dy at yahoo.com> wrote in message
> news:97oupq$qijrh$1 at ID-77331.news.dfncis.de...
> [snip]
>>>>> vobs = [line.replace('Tag:', '').strip() for line in output if
>> line[:4]=='Tag:']
>>>>> vobs
>> ['/vobs/intranet', '/vobs/project']
> 
> I can't resist fiddling with little snippets of code... Why don't you
> use:
> 
>>>> vobs = [line[4:-1] for line in output if line[:4] == 'Tag:']

Actually, for the same reason the original poster complained about the
original code snippet: my version reads more like natural language.  (It
also, I realized after posting, assumes that the VOB name doesn't contain
'Tag:', which I know it won't, but that wasn't clear from the example.)
"For each line in the output, if the line starts with 'Tag:', cut out the
'Tag:' part and strip the whitespace, and give me a list of all of those."
That's a pretty direct mapping from code to description.

My original point, in case it got lost in the details, was that I like list
comprehensions because they force me to think Pythonically.  By that I mean
thinking in terms of the things that Python is good at, like lists and
dictionaries.  These are rich data structures with lots of built-in
capabilities, and if I don't use them, I may as well go back to programming
in Powerbuilder: writing WHILE loops to do progressive searches through
strings when I really want something like the above example, using SELECT
CASE statements for jump tables when all I really want is a dictionary of
functions, and so forth.

-M
You're smart; why haven't you learned Python yet?
http://diveintopython.org/





From DanielK at jBASE.com  Wed Mar 21 18:59:04 2001
From: DanielK at jBASE.com (Daniel Klein)
Date: Wed, 21 Mar 2001 15:59:04 -0800
Subject: What is being compared?
Message-ID: <cfbu6.67$Hp.3494@typhoon.aracnet.com>

Comparing a string to an integer...

>>> 'a' < 1
0

...what is it actually comparing? Is it 'casting' anything?

Just curious,

Dan




From jamie at cctech.co.uk  Fri Mar 16 04:05:53 2001
From: jamie at cctech.co.uk (Jamie MacIsaac)
Date: Fri, 16 Mar 2001 09:05:53 -0000
Subject: Basic information requested
Message-ID: <MHks6.144$lC5.52172@news2-win.server.ntlworld.com>

Hi,

I've just started looking at Python and would like to use it to perform some
simple admin tasks on a Windows network. However, I'm having difficulty
finding out how to accomplish elementary tasks.

Can someone please point me to sample code that shows how to map a network
share and copy some files? Is there a central resource of code samples?

Thanks in advance,

Jamie MacIsaac.





From jos at trapeze.removespam.com  Thu Mar  1 18:01:09 2001
From: jos at trapeze.removespam.com (Digital Hyakugei)
Date: Thu, 1 Mar 2001 18:01:09 -0500
Subject: Newbie: Casting Int to string....
Message-ID: <3a9ed412$1@news1>

Ok, pretty simple, no? I can not find out how to do the following...

i've got a variable:

aNumber = 1

Now i have a function

nntp.stat( aVal )

where aVal has to be a string.

I cant do

nntp.stat( aNumber )

cuz it expects a string, not an int.

How, oh how, do i cast an int to a string?
I'm sure its something that is forehead slapping obvious, but i just cant
seem to find it...

Thanks very much!
jos






From aschneid at mindspring.com  Mon Mar 19 07:57:39 2001
From: aschneid at mindspring.com (Anders Schneiderman)
Date: Mon, 19 Mar 2001 07:57:39 -0500
Subject: Dragon Macros (was: RE: Help Rewriting/Refactoring/Rethinking Parsing Algorythm)
References: <mailman.984979344.16088.python-list@python.org>
Message-ID: <3AB60243.9AD48906@mindspring.com>

Thanks, Mike.  Clearly this is the way to go.  I'll still probably write a few
simple Natlink Python macros for some special  cases (e.g., "grab lines 10 through
15" to select lines 10-15), but you've saved me a bunch of coding.

Since you've obviously figured out a lot about how to use DNS, I've got one more
question.  Using file dialog boxes, either for opening a file in a different
directory or for saving a file in a different directory, is a real pain.  Any tricks
to make life easier?

Thanks,
Anders



Mike C. Fletcher wrote:

> For those who are curious, here's a set of macros that seem to work nicely
> for the names:
>
> In VocEdit, define the following, then train them in vocabulary editor (the
> items with nothing before the slash are supposed to have nothing there,
> these are non-visible elements)...
>         \pyname, LowerCase next word, Turn Capitalisation On, Turn off Spacing
> between words
>         \pycaps, Turn Capitalisation On, Turn off Spacing between words
>         \pynorm, Restore Normal Capitalisation, Restore Normal Spacing
>         \pynormal, Restore Normal Capitalisation, Restore Normal Spacing
>         if __name__ == "__main__":\pymainline, Follow with one newline character
>
> The last entry eliminates the need for a __name__ and __main__ entry, which
> frees up the pyname entry for use as a lowerUpperUpper macro (and yes, I did
> dictate that using the macro :) ).  pynormal is more natural feeling than
> pynorm when you're first getting used to it.
>
> I'm considering adding "add an extra space following this word" for pynorm
> (without it you get some less-than-attractive effects, but with it I'd need
> another phrase for "stop the caps and spaces but don't space" for when doing
> a dotted name.  Oh, pyconstant might be introduced to give an ALL-CAPS + No
> Space macro.
>
> Making a script that auto-created all those + the def, equals, triple-quote,
> etceteras and asked the user to train each a couple of times would be cool.
> Maybe when I'm out of a job I'll look into it.
>
> Anyway, enjoy all,
> Mike
>
> -----Original Message-----
> From: Mike C. Fletcher [mailto:mcfletch at home.com]
> Sent: Sunday, March 18, 2001 11:32 PM
> To: 'Boopy Bootles'; python-list at python.org
> Subject: RE: Help Rewriting/Refactoring/Rethinking Parsing Algorythm
> ...
> Looking at your list, adding:
>         py-caps  -> \No-Space-On\Caps-On
>         py-norm  -> \No-Space-Off\Caps-Off
>         py-name  -> <word> \No-Space-On\Caps-On [ not sure if you can do this
> without a macro... ]
> ...
> -----Original Message-----
> From: Boopy Bootles [mailto:aschneid at mindspring.com]
> Sent: Sunday, March 18, 2001 8:57 PM
> To: python-list at python.org
> Subject: Help Rewriting/Refactoring/Rethinking Parsing Algorythm
>
> I'm trying to write a simple piece of code to make programming by voice
> recognition software easier to do.  I wrote a very simple function that
> would, for example, convert "number customers equals 5" to
> "numberCustomers = 5".  But once I started using it, I quickly
> discovered several more cases I had to handle.  So far, this is the
> list:
> ...



From smnordby at yahoo.com  Fri Mar 16 07:03:37 2001
From: smnordby at yahoo.com (Steve)
Date: Fri, 16 Mar 2001 04:03:37 -0800
Subject: Einstein's Riddle
References: <mailman.984369618.12493.python-list@python.org> <3AACBB41.5854666A@yahoo.com> <3AAF1885.FF28CC21@yahoo.com> <3AAF6352.726046A2@yahoo.com> <XK3s6.9935$TW.39680@tor-nn1.netcom.ca>
Message-ID: <3AB20119.C0B3B671@yahoo.com>

Warren Postma wrote:
> 
> "Steve" <smnordby at yahoo.com> wrote in message
> news:3AAF6352.726046A2 at yahoo.com...
> > Reliability is an indication of whether a measurement is repeatable:
> > Does it give consistent results under the same conditions?
> 
> Ahh yes, because consistenly Wrong, or backwards, in certain circumstances,
> is Perfectly Reliable.  The system will always fail, silently, if x = 5,  3
> < y < 6, and it's tuesday. Every time.
> 
> ;-)
> 
> Warren

Something may be reliable without being valid, but something cannot be
valid without being reliable.

-SteveN-


From aahz at panix.com  Sat Mar  3 13:21:43 2001
From: aahz at panix.com (Aahz Maruch)
Date: 3 Mar 2001 10:21:43 -0800
Subject: Bug fix releases (was Re: Nested scopes resolution -- you can breathe again!)
References: <LNBBLJKPBEHFEDALKOLCMEEPJBAA.tim.one@home.com> <mailman.982897324.9109.python-list@python.org>
Message-ID: <97rcnn$nfl$1@panix3.panix.com>

[posted to c.l.py with cc to python-dev]

[I apologize for the delay in posting this, but it's taken me some time
to get my thoughts straight.  I hope that by posting this right before
IPC9 there'll be a chance to get some good discussion in person.]

In article <mailman.982897324.9109.python-list at python.org>,
Guido van Rossum  <guido at digicool.com> wrote:
>
>We have clearly underestimated how much code the nested scopes would
>break, but more importantly we have underestimated how much value our
>community places on stability.  

I think so, yes, on that latter clause.  I think perhaps it wasn't clear
at the time, but I believe that much of the yelling over "print >>" was
less over the specific design but because it came so close to the
release of 2.0 that there wasn't *time* to sit down and talk things
over rationally.

As I see it, there's a natural tension between between adding features
and delivering bug fixes.  Particularly because of Microsoft, I think
that upgrading to a feature release to get bug fixes has become anathema
to a lot of people, and I think that seeing features added or changed
close to a release reminds people too much of the Microsoft upgrade
treadmill.

>So here's the deal: we'll make nested scopes an optional feature in
>2.1, default off, selectable on a per-module basis using a mechanism
>that's slightly hackish but is guaranteed to be safe.  (See below.)
>
>At the same time, we'll augment the compiler to detect all situations
>that will break when nested scopes are introduced in the future, and
>issue warnings for those situations.  The idea here is that warnings
>don't break code, but encourage folks to fix their code so we can
>introduce nested scopes in 2.2.  Given our current pace of releases
>that should be about 6 months warning.

As some other people have pointed out, six months is actually a rather
short cycle when it comes to delivering enterprise applications across
hundreds or thousands of machines.  Notice how many people have said
they haven't upgraded from 1.5.2 yet!  Contrast that with the quickness
of the 1.5.1 to 1.5.2 upgrade.

I believe that "from __future__" is a good idea, but it is at best a
bandage over the feature/bug fix tension.  I think that the real issue
is that in the world of core Python development, release N is always a
future release, never the current release; as soon as release N goes out
the door into production, it immediately becomes release N-1 and forever
dead to development

Rather than change that mindset directly, I propose that we move to a
forked model of development.  During the development cycle for any given
release, release (N-1).1 is also a live target -- but strictly for bug
fixes.  I suggest that shortly after the release for Na1, there should
also be a release for (N-1).1b1; shortly after the release of Nb1, there
would be (N-1).1b2.  And (N-1).1 would be released shortly after N.

This means that each feature-based release gets one-and-only-one pure
bugfix release.  I think this will do much to promote the idea of Python
as a stable platform for application development.

There are a number of ways I can see this working, including setting up
a separate project at SourceForge (e.g. pythonpatch.sourceforge.net).
But I don't think this will work at all unless the PythonLabs team is at
least willing to "bless" the bugfix release.  Uncle Timmy has been known
to make snarky comments about forever maintaining 1.5.2; I think this is
a usable compromise that will take relatively little effort to keep
going once it's set up.

I think one key advantage of this approach is that a lot more people
will be willing to try out a beta of a strict bugfix release, so the
release N bugfixes will get more testing than they otherwise would.

If there's interest in this idea, I'll write it up as a formal PEP.

It's too late for my proposed model to work during the 2.1 release
cycle, but I think it would be an awfully nice gesture to the community
to take a month off after 2.1 to create 2.0.1, before going on to 2.2.



BTW, you should probably blame Fredrik for this idea.  ;-)  If he had
skipped providing 1.5.2 and 2.0 versions of sre, I probably wouldn't
have considered this a workable idea.  I was just thinking that it was
too bad there wasn't a packaged version of 2.0 containing the new sre,
and that snowballed into this.
-- 
                      --- Aahz (Copyright 2001 by aahz at pobox.com)

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

Nostalgia just ain't what it used to be


From jacy at poczta.onet.pl  Fri Mar  9 09:02:26 2001
From: jacy at poczta.onet.pl (jacek czerwinski)
Date: Fri, 09 Mar 2001 14:02:26 +0000
Subject: BCD or fixed point numbers
Message-ID: <3AA8E272.7268DCFF@poczta.onet.pl>

Has anybody some links for python with BCD or other fixed point numbers
?


--
Profesjonalny web-hosting? Tylko Serwis RUBIKON - http://rubikon.pl 


From tim.one at home.com  Fri Mar  2 17:08:53 2001
From: tim.one at home.com (Tim Peters)
Date: Fri, 2 Mar 2001 17:08:53 -0500
Subject: A counter-counter-proposal for PEP 236: #pragma( nested_scopes)
In-Reply-To: <_sTn6.15738$UZ4.4336745@news4.rdc1.on.home.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCKEILJDAA.tim.one@home.com>

[Warren Postma]
> Hmm.  I wonder how many people would be happier with this syntax
>
> from __twilightzone__ import nested_scopes
>
> <grin>

Very few.  What they really want is

    (from-future-import (quote (nested (scopes))))

nested-identifiers-can-wait-for-2.3-ly y'rs  - tim




From kens at sightreader.com  Thu Mar 15 12:47:10 2001
From: kens at sightreader.com (Ken Seehof)
Date: Thu, 15 Mar 2001 11:47:10 -0600
Subject: I come to praise .join, not to bury it...
References: <mailman.983485923.1136.python-list@python.org> <97p7iq$kk2$1@nntp6.u.washington.edu> <97rka002k9@news1.newsguy.com> <3AAF7E47.294617E1@sage.att.com> <98qeco02hn7@news1.newsguy.com> <3AB106CD.7D259EBC@san.rr.com>
Message-ID: <002f01c0ad77$f6dbb040$04090a0a@upcast.com>

Hmm ... I wonder what I'd do if I could overload __if__ ...

Hey, I know!  You could use it to implement fuzzy logic for quantum
computers.

The new __if__ would cause multiple superimposed quantum states based on the
probability wave distribution of the quantum bit argument.

----- Original Message -----
From: "Darren New" <dnew at san.rr.com>
Newsgroups: comp.lang.python
To: <python-list at python.org>
Sent: Thursday, March 15, 2001 12:15 PM
Subject: Re: I come to praise .join, not to bury it...


> Alex Martelli wrote:
> > And then again, the _appearance_ that too much emphasis
> > was being placed on syntax-sugary issues kept me *AWAY*
> > from Python for longer than it should have
>
> Actually, what I found quite cool about Python is that the procedural
syntax
> maps to OO semantics. I.e., in Smalltalk, everything looks like a method
> call, including "if". In Eiffel, everything that isn't a method call
doesn't
> look like a method call, and vica versa. But in Python, the convenient and
> familiar "x = y[5:10]" maps to a real live method call you can override.
>
> Well, *I* thought it was neat. It lets you be OO where you want, and keeps
> it out of your face otherwise.
>
> --
> Darren New / Senior MTS & Free Radical / Invisible Worlds Inc.
> San Diego, CA, USA (PST).  Cryptokeys on demand.
> --
> http://mail.python.org/mailman/listinfo/python-list
>




From kevin_cazabon at hotmail.nospamplease!.com  Sun Mar 11 23:50:08 2001
From: kevin_cazabon at hotmail.nospamplease!.com (Kevin Cazabon)
Date: Sun, 11 Mar 2001 21:50:08 -0700
Subject: Einstein's Riddle
References: <oNPq6.396644$ge4.138479566@news2.rdc2.tx.home.com> <98glrd$1si1a$1@ID-75892.news.dfncis.de> <98h62k$knn$1@abbenay.CS.Berkeley.EDU>
Message-ID: <Qk85#ArqAHA.188@typhoon.i3s.net>

You've obviously got too much time on your hands... q:]



"Brian Harvey" <bh at abbenay.cs.berkeley.edu> wrote in message
news:98h62k$knn$1 at abbenay.CS.Berkeley.EDU...
| "Boris Popov" <no-spam-popov at bigserver.com> writes:
| >house         1 2 3 4 5
| >color            yellow blue red green white
| >nationality   norweigian dane brit german swede
| >drink            water tea milk coffee beer
| >smoke        dunhill blend pall mall prince bluemaster
| >pet              cats horse birds fish dogs
| >
| >P.S. Did I have to write software to solve it ? :)
|
| I can't resist being (apparently) first to post a computer program.
| But it doesn't really belong in any of these newsgroups, because it's
| written in Logo.
|
| First get Berkeley Logo from ftp://ftp.cs.berkeley.edu/ucblogo/
| (pick the one that matches your platform)
|
| Then get the logic-puzzle-solver from
| http://www.cs.berkeley.edu/~bh/logic-code/hybrid.lg
|
| Then here's the part specific to this puzzle:
|
| ? po [einstein neighbor neighbor1]
| to einstein
| cleanup
| category "position [1 2 3 4 5]
| category "color [red green white yellow blue]
| category "nationality [British Swedish Danish Norwegian German]
| category "beverage [tea coffee milk beer water]
| category "cigar [pallmall dunhill blend bluemaster prince]
| category "pet [dog bird cat horse fish]
| verify "British "red
| verify "Swedish "dog
| verify "Danish "tea
| justbefore "green "white :position
| verify "green "coffee
| verify "pallmall "bird
| verify "yellow "dunhill
| verify 3 "milk
| verify 1 "Norwegian
| neighbor "blend "cat :position
| neighbor "horse "dunhill :position
| verify "bluemaster "beer
| verify "German "prince
| neighbor "Norwegian "blue :position
| neighbor "blend "water :position
| print []
| solution
| end
|
| to neighbor :a :b :lineup
| falsify :a :b
| neighbor1 :lineup count :lineup
| end
|
| to neighbor1 :lineup :count
| if :count=0 [stop]
| foreach bl bf bf :lineup [implies :a first :lineup "true :b ? "false]
| neighbor1 (lput first :lineup bf :lineup) :count-1
| end
|
| ? solve [einstein]
| category position [1 2 3 4 5]
| category color [red green white yellow blue]
| category nationality [British Swedish Danish Norwegian German]
| category beverage [tea coffee milk beer water]
| category cigar [pallmall dunhill blend bluemaster prince]
| category pet [dog bird cat horse fish]
| British red -> true
| Swedish red -> false
| Danish red -> false
| Norwegian red -> false
| German red -> false
| green British -> false
| white British -> false
| yellow British -> false
| blue British -> false
| Swedish dog -> true
| British dog -> false
| red dog -> false
| Danish dog -> false
| Norwegian dog -> false
| German dog -> false
| bird Swedish -> false
| cat Swedish -> false
| horse Swedish -> false
| fish Swedish -> false
| Danish tea -> true
| British tea -> false
| red tea -> false
| Swedish tea -> false
| dog tea -> false
| Norwegian tea -> false
| German tea -> false
| coffee Danish -> false
| milk Danish -> false
| beer Danish -> false
| water Danish -> false
| green 5 -> false
| white 1 -> false
| green coffee -> true
| red coffee -> false
| British coffee -> false
| white coffee -> false
| yellow coffee -> false
| blue coffee -> false
| 5 coffee -> false
| tea green -> false
| Danish green -> false
| milk green -> false
| beer green -> false
| water green -> false
| pallmall bird -> true
| dunhill bird -> false
| blend bird -> false
| bluemaster bird -> false
| prince bird -> false
| dog pallmall -> false
| Swedish pallmall -> false
| cat pallmall -> false
| horse pallmall -> false
| fish pallmall -> false
| yellow dunhill -> true
| red dunhill -> false
| British dunhill -> false
| green dunhill -> false
| coffee dunhill -> false
| white dunhill -> false
| blue dunhill -> false
| pallmall yellow -> false
| bird yellow -> false
| blend yellow -> false
| bluemaster yellow -> false
| prince yellow -> false
| 3 milk -> true
| 1 milk -> false
| 2 milk -> false
| 4 milk -> false
| 5 milk -> false
| tea 3 -> false
| Danish 3 -> false
| coffee 3 -> false
| green 3 -> false
| white 4 -> false
| beer 3 -> false
| water 3 -> false
| 1 Norwegian -> true
| 2 Norwegian -> false
| 3 Norwegian -> false
| milk Norwegian -> false
| 4 Norwegian -> false
| 5 Norwegian -> false
| white Norwegian -> false
| British 1 -> false
| red 1 -> false
| Swedish 1 -> false
| dog 1 -> false
| Danish 1 -> false
| tea 1 -> false
| German 1 -> false
| blend cat -> false
| horse dunhill -> false
| yellow horse -> false
| bluemaster beer -> true
| pallmall beer -> false
| bird beer -> false
| dunhill beer -> false
| yellow beer -> false
| blend beer -> false
| prince beer -> false
| tea bluemaster -> false
| Danish bluemaster -> false
| coffee bluemaster -> false
| green bluemaster -> false
| milk bluemaster -> false
| 3 bluemaster -> false
| water bluemaster -> false
| German prince -> true
| British prince -> false
| red prince -> false
| Swedish prince -> false
| dog prince -> false
| Danish prince -> false
| tea prince -> false
| Norwegian prince -> false
| 1 prince -> false
| pallmall German -> false
| bird German -> false
| dunhill German -> false
| yellow German -> false
| blend German -> false
| bluemaster German -> false
| beer German -> false
| Norwegian blue -> false
| 1 blue -> false
| blue 3 -> false
| milk blue -> false
| blue 4 -> false
| blend water -> false
|
|
|   Assuming 1 yellow
|   1 yellow -> true
|   2 yellow -> false
|   dunhill 2 -> false
|   3 yellow -> false
|   milk yellow -> false
|   dunhill milk -> false
|   3 dunhill -> false
|   4 yellow -> false
|   dunhill 4 -> false
|   5 yellow -> false
|   dunhill 5 -> false
|   1 dunhill -> true
|   Norwegian dunhill -> true
|   Swedish dunhill -> false
|   dog dunhill -> false
|   yellow dog -> false
|   Swedish yellow -> false
|   Danish dunhill -> false
|   tea dunhill -> false
|   water dunhill -> true
|   pallmall water -> false
|   bird water -> false
|   prince water -> false
|   German water -> false
|   yellow water -> true
|   red water -> false
|   British water -> false
|   white water -> false
|   blue water -> false
|   Swedish water -> false
|   Norwegian water -> true
|   1 water -> true
|   2 water -> false
|   4 water -> false
|   5 water -> false
|   dog water -> false
|   coffee 1 -> false
|   green 1 -> false
|   Norwegian green -> false
|   yellow Norwegian -> true
|   beer Norwegian -> false
|   bluemaster Norwegian -> false
|   1 bluemaster -> false
|   beer 1 -> false
|   horse Norwegian -> false
|   1 horse -> false
|   water horse -> false
|   blend Norwegian -> false
|   1 blend -> false
|   pallmall Norwegian -> false
|   bird Norwegian -> false
|   1 bird -> false
|   pallmall 1 -> false
|   coffee Norwegian -> false
|   Danish yellow -> false
|   tea yellow -> false
|   white 2 -> false
|   blend 4 -> false
|   blend 3 -> false
|   milk blend -> false
|   horse 4 -> false
|   horse 3 -> false
|   milk horse -> false
|
|     Assuming 5 beer
|     5 beer -> true
|     2 beer -> false
|     bluemaster 2 -> false
|     4 beer -> false
|     bluemaster 4 -> false
|     5 bluemaster -> true
|     pallmall 5 -> false
|     bird 5 -> false
|     blend 5 -> false
|     2 blend -> true
|     white blend -> false
|     pallmall 2 -> false
|     bird 2 -> false
|     prince 2 -> false
|     German 2 -> false
|     cat 2 -> false
|     cat 5 -> false
|     bluemaster cat -> false
|     beer cat -> false
|     cat 4 -> false
|     prince 5 -> false
|     German 5 -> false
|     tea 5 -> false
|     Danish 5 -> false
|
|       Assuming 5 Swedish
|       5 Swedish -> true
|       2 Swedish -> false
|       blend Swedish -> false
|       bluemaster Swedish -> true
|       beer Swedish -> true
|       coffee Swedish -> false
|       green Swedish -> false
|       German green -> true
|       prince green -> true
|       pallmall green -> false
|       bird green -> false
|       coffee bird -> false
|       pallmall coffee -> false
|       blend green -> false
|       2 green -> false
|       4 green -> true
|       cat green -> false
|       coffee cat -> false
|       horse green -> false
|       coffee horse -> false
|       red 4 -> false
|       British 4 -> false
|       coffee 4 -> true
|       tea 4 -> false
|       Danish 4 -> false
|       2 Danish -> true
|       blend Danish -> true
|       pallmall Danish -> false
|       bird Danish -> false
|       British bird -> true
|       red bird -> true
|       white bird -> false
|       pallmall white -> false
|       blue bird -> false
|       pallmall blue -> false
|       red pallmall -> true
|       British pallmall -> true
|       4 pallmall -> false
|       3 pallmall -> true
|       milk pallmall -> true
|       tea pallmall -> false
|       blend tea -> true
|       2 tea -> true
|       cat tea -> false
|       Danish cat -> false
|       bird tea -> false
|       milk bird -> true
|       3 bird -> true
|       4 bird -> false
|       dog 3 -> false
|       Swedish 3 -> false
|       milk Swedish -> false
|       dog milk -> false
|       cat 3 -> false
|       1 cat -> true
|       yellow cat -> true
|       red cat -> false
|       British cat -> false
|       white cat -> false
|       blue cat -> false
|       water cat -> true
|       milk cat -> false
|       Norwegian cat -> true
|       German cat -> false
|       prince cat -> false
|       dunhill cat -> true
|       fish dunhill -> false
|       1 fish -> false
|       yellow fish -> false
|       water fish -> false
|       Norwegian fish -> false
|       fish 3 -> false
|       milk fish -> false
|       white 3 -> false
|       red 3 -> true
|       British 3 -> true
|       German 3 -> false
|       prince 3 -> false
|       4 prince -> true
|       horse prince -> false
|       fish prince -> true
|       milk prince -> false
|       coffee prince -> true
|       blue prince -> false
|       German blue -> false
|       white prince -> false
|       bluemaster white -> true
|       5 white -> true
|       beer white -> true
|       tea white -> false
|       Danish white -> false
|       blue Danish -> true
|       Swedish blue -> false
|       dog blue -> false
|       white Swedish -> true
|       4 Swedish -> false
|       German 4 -> true
|       dog 4 -> false
|       fish 4 -> true
|       2 fish -> false
|       blend fish -> false
|       5 fish -> false
|       beer fish -> false
|       bluemaster fish -> false
|       green fish -> true
|       red fish -> false
|       British fish -> false
|       white fish -> false
|       blue fish -> false
|       Danish fish -> false
|       German fish -> true
|       tea fish -> false
|       coffee fish -> true
|       dog coffee -> false
|       beer dog -> true
|       5 dog -> true
|       2 dog -> false
|       blend dog -> false
|       bluemaster dog -> true
|       green dog -> false
|       white dog -> true
|       horse white -> false
|       milk white -> false
|       red milk -> true
|       British milk -> true
|       German milk -> false
|       coffee German -> true
|       horse German -> false
|       white German -> false
|       2 coffee -> false
|       blend coffee -> false
|       beer British -> false
|       5 British -> false
|       bluemaster British -> false
|       red bluemaster -> false
|       5 red -> false
|       beer red -> false
|       horse British -> false
|       Danish horse -> true
|       tea horse -> true
|       beer horse -> false
|       5 horse -> false
|       2 horse -> true
|       blend horse -> true
|       bluemaster horse -> false
|       British blend -> false
|       red blend -> false
|       blue blend -> true
|       bluemaster blue -> false
|       5 blue -> false
|       2 blue -> true
|       beer blue -> false
|       tea blue -> true
|       red 2 -> false
|       British 2 -> false
|       horse blue -> true
|       red horse -> false
|
| 1 yellow Norwegian water dunhill cat
| 2 blue Danish tea blend horse
| 3 red British milk pallmall bird
| 4 green German coffee prince fish
| 5 white Swedish beer bluemaster dog




From jim at interet.com  Sat Mar 24 10:19:37 2001
From: jim at interet.com (James C. Ahlstrom)
Date: Sat, 24 Mar 2001 10:19:37 -0500
Subject: Commercial Products in Python?
References: <3abb86ee.142862695@localhost>
Message-ID: <3ABCBB09.CEEA708C@interet.com>

Victor Muslin wrote:
> 
> A friend of mine is considering developing a shrink-wrapped commercial
> product. Python seems like a possible choice, but it is not clear
> whether there are any precedents (though probably similar concerns
> apply to a Perl-based product).
> 
> I would very much appreciate an advice in this area and possibly
> answers to some of the following questions:
> 
> 1. Are there any existing shirnk-wrap commercial products written in
> Python?

We have been shipping our WinLease product with a Python front end for
5 or so years.  It is commercial, but not exactly shrink wrap.  We
have found Python to be bullet proof.

> 3. What is the best way to go about it? Is it better to have customers
> install Python and give them bytecode files or use one of the programs
> that creates an executable?

We ship a pre-made Python executable plus the necessary (not all)
Python library files as a frozen module.  Our own Python
files are in a Python jar file, but this failed to get into
Python so it is a local mod.  Using freeze for your own files
would be an alternative.
 
> 4. There is an easy shrink-wrap install for Python (at least on
> Windows NT). Are there such on Unix/Linux platforms? (Typically
> customers don't like having to compile software from source).

We use the Wise installation system, and I highly recommend it.
It is not Python specific.

> 5. If the answer to question #1 is "no", why not?
> 
> 6. Are there any licensing issues to be aware of?

No!!!  Just read the license.
 
> Anything else that comes to mind in this direction would be very
> helpful.

Jim Ahlstrom


From marcin.chady at ntlworld.com  Tue Mar 13 19:04:39 2001
From: marcin.chady at ntlworld.com (Marcin Chady)
Date: Wed, 14 Mar 2001 00:04:39 +0000
Subject: Unpickling from strings
References: <qHxr6.19658$pR3.3830844@news2-win.server.ntlworld.com> <3gyr6.159$4d3.34762@newsb.telia.net>
Message-ID: <3AAEB597.4010007@ntlworld.com>

> did you pickle classes defined in your main script (aka __main__),
> and then failed to unpickle them in another script?
> 
> don't do that.
> 
> (to solve this, move the classes to a common support module, and
> import it into the pickling script)
> 
> Cheers /F


Both pickling and unpickling is done by the same script, and the classes 
are defined in a separate module imported by that script.

Marcin



From fig at monitor.net  Fri Mar 23 16:55:34 2001
From: fig at monitor.net (Stephen R. Figgins)
Date: Fri, 23 Mar 2001 13:55:34 -0800
Subject: PyUnit in Python 2.1 standard library
References: <3ABA8A23.85EA76CE@downright.com>
Message-ID: <3ABBC656.C54B0F9A@monitor.net>

Doug Fort wrote:
> 
> Just recieved word that Steve Purcell has checked PyUnit  into the
> standard library.  

This means that it will be a standard feature of the Python
distribution?  Will it come with Python 2.1?

-Stephen


From cribeiro at mail.inet.com.br  Fri Mar 30 08:03:34 2001
From: cribeiro at mail.inet.com.br (Carlos Alberto Reis Ribeiro)
Date: Fri, 30 Mar 2001 10:03:34 -0300
Subject: NumPy loop efficiency
In-Reply-To: <m3g0fv78y4.fsf@chinon.cnrs-orleans.fr>
References: <3ac44e27$0$16692$73bec57d@news.be.uu.net>
 <mailman.985951902.14535.python-list@python.org>
Message-ID: <5.0.2.1.0.20010330093758.021d9ac0@mail.inet.com.br>

At 14:11 30/03/01 +0200, Konrad Hinsen wrote:
>Something like that (but not entirely) is possible but little known:
>
>      Numeric.add(a, b, c)
>
>does the same as
>
>      c = a + b
>
>but storing the result in the preallocated array c (which must of
>course have the right dimensions). And c can well be a or b, in which
>case the data is overwritten. By using only these functions, many
>temporary arrays can be avoided, but at the cost of producing
>unreadable code.

Another take it would be to use in-place operations such as:

z = a + b
z += c

Anyway, I'm thinking about a more general approach without the need to 
implement full-fledged support for lazy evaluation. Take this example:

z = a + b + c

It is evaluated this way:

i1 = a + b
12 = b + c
z  = i2

Two temporary arrays gets allocated. The second one is returned as the 
result of the operation. This is quite expensive both in performance and 
memory terms (specially for big arrays), and leads to the adoption of some 
less-readable code to squeeze out more performance. The question is, who 
generated the temporary array?

Logic tells me that the temporary array is generated inside __add__() - the 
method that does the operation on the array. Some optimization would be 
possible, using this technique:

1) when creating a array to store temporary results, flag the array as 
"temporary";
2) when doing some operation that takes two arrays, if one of them is 
temporary, do the operation in-place;
3) when assigning a temporary array to any other variable, reset the 
"temporay" flag.

The critical operation above is step (3). My proposal is to use a hook that 
gets called on assignment (the = operator). As far as I know, this hook 
does not exist today, but it's inclusion would allow for a number of 
optimizations similar to this one.


Carlos Ribeiro





From aahz at panix.com  Fri Mar 16 12:56:27 2001
From: aahz at panix.com (Aahz Maruch)
Date: 16 Mar 2001 09:56:27 -0800
Subject: relational/boolean operator overloading
References: <98tjq6$qre$1@news.flashnet.it>
Message-ID: <98tk4b$sbv$1@panix3.panix.com>

In article <98tjq6$qre$1 at news.flashnet.it>,
Renzo Tomaselli <renzo.tomaselli at tecnotp.it> wrote:
>
>        I'm trying to use a subset of Python for querying an object
>database, so that I would like to use a relational operator between a set
>and a const value; this would yield another set. Multiple relations would be
>connected by boolean operators with a meaning of set operators and so on.
>This approach would enable a straight variable binding.
>However there seems no way to redefine such operators, only numeric binary
>operators are mentioned in the ref. manual.
>Any trick/suggestion/comment ?

Not sure what advice to give you with current Python, but Python 2.1
introduces rich comparisons, which may give you what you want.  See
PEP 207, http://python.sourceforge.net/peps/pep-0207.html

Python 2.1 will probably be released in about a month.
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

Three sins: BJ, B&J, B&J


From mal at lemburg.com  Tue Mar 13 12:32:30 2001
From: mal at lemburg.com (M.-A. Lemburg)
Date: Tue, 13 Mar 2001 18:32:30 +0100
Subject: newsgroup postings: cursor.execute fails under mxODBC
References: <3AABF77F.7B77D51@northwestern.edu>
Message-ID: <3AAE59AE.630C148C@lemburg.com>

[CCed to the Python newsgroup, since I believe this is of general
 interest; hope you don't mind.]

Louis Luangkesorn wrote:
> 
> Hello
> 
> This is regarding the mxODBC 2.0 module that you have been providing (by
> the way, thank you).  On the comp.lang.python newsgroup there has been
> some discussion regarding the trouble some of us have been having on
> using the mxODBC 2.0 packages under Windows (95, 98, 2000).  Referring
> to the Sunday 11 Marsh message from Brett Haydon, apparently something
> happened when the packages got updated to 2.0, so his installation works
> under 1.1.1 but failed when he installed the 2.0 packages.
> 
> I am wondering if this is a known problem with a fix, or if it is
> possible to obtain the 1.1.1 packages which seem to be more stable right
> now.

Thank you for forwarding the messages. I have had numerous questions
about this problem and will look into it later this week (just came
back from the Python conference and am still fighting the jet lag).

One thing I've noted about the problem reports is that all seem
to use ActiveState Python 2.0. There may be an issue here, since
the mxODBC (alas the egenix-commercial) package was compiled against
the BeOpen Python 2.0 version which is available from python.org.

Could you perhaps check whether you experience the same problems
using BeOpen Python 2.0 ?

Thanks,
Marc.
 
> Sincerely,
> 
> Louis
> 
> PS note that my reference to changing the name of the odbc.pyc refers to
> the odbc.pyc that came with the win32 extensions.
> 
> -----<forwarded newsgroup message>-----
> Subject:                Re: cursor.execute fails under mxODBC
>          Date:                Sun, 11 Mar 2001 03:17:09 +1100
>          From:               "Brett Haydon" <bbhaydon at bigpond . com>
>  Organization:                Telstra BigPond Internet Services
> (http://www.bigpond.com)
>   Newsgroups:               comp.lang.python
>    References:                1 , 2 , 3
> 
> Hmmm looks like the new version is broken.
> 
> I am running mxODBC 1.1.1 no problems with both MySQL and Access
> for the last 6 months on W2k.
> 
> I just tried the new version also with exactly the same problems.
> 
> The old one still works fine though..
> 
> regards,
> 
> bbhaydon
> 
> Subject:                Re: cursor.execute fails under mxODBC
>          Date:                Thu, 08 Mar 2001 15:52:14 -0600
>          From:               Louis Luangkesorn <lluang at northwestern.edu>
> 
>  Organization:                Northwestern IE/MS
>   Newsgroups:                comp.lang.python
>    References:                1 , 2
> 
> I have the same problem.  Below is my output.  I will first use the odbc
> 
> that comes included with ActivePython, then the mxODBC module  (by the
> way,
> I've tried to rename the odbc.pyc file, that did not help, except break
> odbc)
> 
> PythonWin 2.0 (#8, Oct 19 2000, 11:30:05) [MSC 32 bit (Intel)] on win32.
> 
> Portions Copyright 1994-2000 Mark Hammond (MarkH at ActiveState.com) - see
> 'Help/About PythonWin' for further copyright information.
> >>>
> >>> import dbi,odbc
> >>> import pprint
> >>> channel = odbc.odbc('sqltutor')
> >>> ch1 = channel.cursor()
> >>> ch1.execute('SELECT FirstName, LastName, Address, City, State FROM
> EmployeeAddressTable')
> 0
> >>> test1 = ch1.fetchall()
> >>> pprint.pprint(test1)
> [('Joe', 'Smith', '83 First Street', 'Howard', 'Ohio'),
>  ('Mary', 'Scott ', '842 Vine Ave.', 'Losantiville', 'Ohio '),
>  ('Sam', 'Jones ', '33 Elm St.', 'Paris', 'New York '),
>  ('Sarah', 'Ackerman ', '440 U.S. 110', 'Upton', 'Michigan ')]
> >>> ch1.close()
> >>> channel.close()
> >>> import mx.ODBC
> >>> channel = mx.ODBC.Windows.connect('sqltutor')
> >>> ch1 = channel.cursor()
> >>> ch1.execute('SELECT FirstName, LastName, Address, City, State FROM
> EmployeeAddressTable')
> 
> at this point the whole thing dies.  No error messages.  I didn't see
> any
> log files anywhere either.
> 
> Louis
> 
> --
> K Louis Luangkesorn
> lluang at northwestern.edu  http://pubweb.nwu.edu/~kll560  PGP:0xF3E2D362
> Whatsoever things are true, ... honest, ... just, ... pure, ... lovely,
> ... of good report; if there be any virtue, and if there be any praise,
> think on these things.- motto - Northwestern University

-- 
Marc-Andre Lemburg
______________________________________________________________________
Company & Consulting:                           http://www.egenix.com/
Python Pages:                           http://www.lemburg.com/python/



From shaleh at valinux.com  Thu Mar 22 16:27:35 2001
From: shaleh at valinux.com (Sean 'Shaleh' Perry)
Date: Thu, 22 Mar 2001 13:27:35 -0800 (PST)
Subject: Good Editor? (What's the big deal with syntax colouring?)
In-Reply-To: <008F0A63472BD311AF9800104BCD102561CD3E@minirex.vmlabs.com>
Message-ID: <XFMail.20010322132735.shaleh@valinux.com>

On 22-Mar-2001 John Schmitt wrote:
> In the dog days of DOS, I was a brief guy.  Now I use CodeWright.  It's the
> best editor I've used (and I've used vi, although not too extensively).
> 
> Why do people want syntax colouring?  I can't stand it.  One of my minor
> beefs with CodeWright is that there isn't a one button way to tell it to
> change _all_ of my colours all at once.  Am I missing something by disabling
> syntax colouring?  It always reminds me of the artwork my 6-year-old nephew
> brings home from school to show his mommy.
> 

I code in about 5 languages, each with slightly different syntax.  For the
first few moments, I find syntax highlighting helpful because I can see that
'next' is a perl keyword, not a python keyword (I need 'continue' instead).



From sholden at holdenweb.com  Thu Mar 22 21:40:22 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 22 Mar 2001 21:40:22 -0500
Subject: Functionalism, aesthetics Was:(RE: I come to praise .join,   not
References: <mailman.985052844.17993.python-list@python.org> <mailman.985145848.16117.python-list@python.org> <99cpg101iij@news2.newsguy.com>
Message-ID: <aMyu6.8895$G72.194081@e420r-atl1.usenetserver.com>

"Alex Martelli" <aleaxit at yahoo.com> wrote in message
news:99cpg101iij at news2.newsguy.com...
[ ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ...
... ... ... ... ... ... ... ...]
>
> But refusing to accept complainants who are unable or
> unwilling to work to put their complaints in specific
> and concrete terms can be by far the best approach.
>
>
> Alex
>
Would you like to write  PEP for that? [in a nutshell ...]

thinking-i-should-be-writing-something-else-ly y'rs  - steve





From sholden at holdenweb.com  Tue Mar  6 09:07:09 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Tue, 6 Mar 2001 09:07:09 -0500
Subject: I come to praise .join, not to bury it...
References: <mailman.983485923.1136.python-list@python.org> <97p7iq$kk2$1@nntp6.u.washington.edu> <97rka002k9@news1.newsguy.com> <kuu257kb1n.fsf@lasipalatsi.fi> <982a4f0su3@news1.newsguy.com> <kuofvfjb77.fsf@lasipalatsi.fi>
Message-ID: <1e6p6.27206$1D5.1119285@e420r-atl1.usenetserver.com>

"Erno Kuusela" <erno-news at erno.iki.fi> wrote in message
news:kuofvfjb77.fsf at lasipalatsi.fi...
> In article <982a4f0su3 at news1.newsguy.com>, "Alex Martelli"
> <aleaxit at yahoo.com> writes:
>
> | "Erno Kuusela" <erno-news at erno.iki.fi> wrote in message
>
> || what's wrong with vague aesthetic?-) if lots of people share it,
> || it's not to be dismissed so lightly.
>
> | Lots of people share the 'vague aesthetic' of desiring
> | to use braces instead of whitespace to delimit blocks
> | of statements, for example.
>
> | Python dismisses that lightly, and that is definitely
> | a part of what makes it such a wonderful language.
>
> it could also be argued that python uses whitespace instead
> of braces because of vague aesthetic shared by lots of people. :)
>
Well it could be argued, but you could also look at the usability data which
(I understand) was the basis of this choice when Python was originally
designed.

not-going-back-to-braces-now-ly y'rs  - steve





From loriendNOSPAM at bigpond.com  Sun Mar 18 03:18:04 2001
From: loriendNOSPAM at bigpond.com (Lorien Dunn)
Date: Sun, 18 Mar 2001 19:18:04 +1100
Subject: Meta-class inheritance problem (Python2.0 bug?)
References: <mailman.984894982.24858.python-list@python.org>
Message-ID: <uWZs6.35929$v5.76709@newsfeeds.bigpond.com>

Tim Peters wrote:

> [posted & mailed]
> 
> [David Abrahams, on Lorien Dunn's boost + metaclass mystery]
> 
> Note that Lorien opened a bug report on this (and thank you!):
> 
> http://sourceforge.net/tracker/index.php?func=detail&
>     aid=409355&group_id=5470&atid=105470

        I thought the bug report did't work. Many apologies if the bug report came 
in more than once, as the Sourceforge submit link only gave me error 
messages. I didn't think it had worked at all. Konqueror 2.1 told me that 
the connection was broken, and Netscape told me that the document contained 
no data.

Thanks,
Lorien.


From idot at vt.edu  Fri Mar  2 11:41:14 2001
From: idot at vt.edu (Carl Banks)
Date: Fri, 2 Mar 2001 11:41:14 -0500
Subject: gah! I hate the new string syntax
References: <mailman.983485923.1136.python-list@python.org> <0vpn79.331.ln@127.0.0.1> <UY5kBKANG8n6EwqG@jessikat.demon.co.uk>
Message-ID: <afio79.rj1.ln@127.0.0.1>

Robin Becker <robin at jessikat.fsnet.co.uk> wrote:
> In article <0vpn79.331.ln at 127.0.0.1>, Carl Banks <idot at vt.edu> writes
>>Sean 'Shaleh' Perry <shaleh at valinux.com> wrote:
>>> return ";".join(["%s=%s" % (k, params[k]) for k in params.keys()])
>>>
>>> every day python seems to be moving closer to the line noise aspect of
>>> coding.
>>
>>I find it ok for methodlike functions such as split.
>>
>>For unmethodlike functions such as join, I would use the old
>>string.join syntax (for regular strings, anyways).
>>
>>"".join is just creepy.
>>
> it looks plain wrong, but would ['a','b','c'].join(';') look any better?

No, but string.join(['a','b','c'],';') does look better, IMO.


> In fact there may be some benefit in being able to say
> pathJoiner = (sys.platform=='win32' and ';' or ':').join
> .....
> path1 = pathJoiner([d1,d2,.....])
> path2 = pathJoiner([e1,e2,.....])

Sure.  I'm not saying join should not be a method, only that I prefer
the function invocation as a matter of style.  Most other string
methods are ok, but this one I find supremely weird.


-- 
CARL BANKS

"I like nothing better than a good challenge.
 This is not one of them."


From fredrik at pythonware.com  Fri Mar 16 12:50:46 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 16 Mar 2001 17:50:46 GMT
Subject: list.sort(cmpfunc) question
References: <m1ae6lg5qt.fsf@pclabs188.labs.trema.com> <98t9dq$qvu$1@panix6.panix.com> <98tf9d$g3v$1@news.mathworks.com> <3AB249FB.131080CD@san.rr.com> <98thua$i40$1@news.mathworks.com>
Message-ID: <Wnss6.186$sk3.53027@newsb.telia.net>

Joshua Marshall wrote:
> What does "should" mean, in the context of RFCs?

usually something like:

     - MUST

       This word or the adjective "REQUIRED" means that the item is an
       absolute requirement of the specification.

     - SHOULD

       This word or the adjective "RECOMMENDED" means that there might
       exist valid reasons in particular circumstances to ignore this
       item, but the full implications should be understood and the case
       carefully weighed before taking a different course.

     - MAY

       This word or the adjective "OPTIONAL" means that this item is
       truly optional.  One vendor might choose to include the item
       because a particular marketplace requires it or because it
       enhances the product, for example; another vendor may omit the
       same item.

in other words, "recommended but not required"

Cheers /F




From aleaxit at yahoo.com  Thu Mar 22 03:12:08 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 22 Mar 2001 09:12:08 +0100
Subject: New PEP: The directive statement
References: <mailman.985224089.1179.python-list@python.org>
Message-ID: <99cc4q014le@news2.newsguy.com>

"Tim Peters" <tim.one at home.com> wrote in message
news:mailman.985224089.1179.python-list at python.org...
    [snip]
> if-you-can't-laugh-at-your-tools-you're-taking-them-too-seriously-ly
>     y'rs  - tim

I did try laughing at my screwdriver last time I used it, but
I don't think it saw what was it that I was finding so funny.


Alex





From Eugene.Leitl at lrz.uni-muenchen.de  Wed Mar  7 06:07:35 2001
From: Eugene.Leitl at lrz.uni-muenchen.de (Eugene Leitl)
Date: Wed, 7 Mar 2001 12:07:35 +0100 (MET)
Subject: GadFly and gfplus.py working with Python 2.0?
In-Reply-To: <Pine.LNX.4.30.0103071253340.3221-100000@fep132.fep.ru>
Message-ID: <Pine.GSO.4.03.10103071148001.13488-100000@sun1.lrz-muenchen.de>

On Wed, 7 Mar 2001, Oleg Broytmann wrote:

>    Why not just try yourself? (Works for me, YMMV)

GadFly seems to be working, but gfplus.py doesn't seem
to accept simple valid SQL. Also, I was reading

	http://groups.yahoo.com/group/gadfly-rdbms/messages

and just got spooked, I guess. It seems the problem is not serious:

	http://groups.yahoo.com/group/gadfly-rdbms/message/413

Otoh, gfplus.py is behaving strangely (?) see below. Any idea 
what might be going wrong?

[eleitl at pc27 download]$ python gfplus.py
 
gfplus 1.1 -- Interactive gadfly shell
 
Wednesday March 07, 2001 11:56 AM
 
Using:
DB Name: foo
DB Location: /tmp/frobnik
Unable to locate database "foo" at location "/tmp/frobnik".
Create? (Yy/Nn) Y
GF> desc user_tables;
                       
COLUMN_NAME
===========
 
GF> create table test (nr number,name varchar2(5));
 
SyntaxError: unexpected token sequence.near ::
' table test (nr'*' number,name va'
*******************************
current state = 79
expects:
'EXCEPT', 'UNION', 'GROUP', 'VARCHAR', 'DESC', 'ORDER', 'WHERE', 'AS',
'*', 'IN', 'INTERSECT', 'FLOAT', '+', ')', '/', ',', '-', 'AND', 'FROM',
'*', ';', 'INTEGER', 'NOT', '>', 'OR', 'HAVING', '=', 'BETWEEN', 'ASC',
'<',
('nomatch1',)
current token = ((-8, 'user_defined_name'), 'NUMBER')
 
GF> select * from user_tables;
 
NameError: 'USER_TABLES' no such relation in DB
 
GF> desc user_tables;
 
COLUMN_NAME
===========
 
GF>  




From thomas at xs4all.net  Sun Mar  4 22:51:08 2001
From: thomas at xs4all.net (Thomas Wouters)
Date: Mon, 5 Mar 2001 04:51:08 +0100
Subject: syslog broken on FreeBSD?
In-Reply-To: <10981466442.20010303155133@buz.ch>; from gabriel_ambuehl-py@buz.ch on Sat, Mar 03, 2001 at 03:51:33PM +0100
References: <10981466442.20010303155133@buz.ch>
Message-ID: <20010305045108.X9678@xs4all.nl>

On Sat, Mar 03, 2001 at 03:51:33PM +0100, Gabriel Ambuehl wrote:

> I've been trying to write to syslog using the following
> import syslog
> syslog("TEST)

> but there won't ever appear any message in the syslog. Is this module
> broken on FreeBSD 4.2 or am I doing something very wrong?

You're doing at least two things wrong: after the import, the name 'syslog'
refers to the module object, no the function 'syslog' in the module
'syslog'. And second, you didn't close your string '"TEST"' :)

If you did both those right after all, you might have been looking at the
wrong logfile. If you don't specify a facility using syslog.openlog(),
syslog.syslog() will log using the 'LOG_INFO' priority only. That makes
which logfile it ends it very system specific, because that means the
'facility' is set to 0, and which #define is zero depends on the system. On
FreeBSD, that's 'kernel messages' or the 'kern' syslog facility.

Now you have to search your /etc/syslog.conf for where 'kern.info' messages
are sent. On my FreeBSD systems, that's nowhere. So change your syslog call
to

syslog.syslog(syslog.LOG_ERR, "TEST")

and you'll see something similar to

Mar  5 04:50:14 cvs python: TEST

in your /var/log/messages. 

-- 
Thomas Wouters <thomas at xs4all.net>

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



From stephen_purcell at yahoo.com  Tue Mar 13 05:46:24 2001
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Tue, 13 Mar 2001 11:46:24 +0100
Subject: module importing and variable syntax
In-Reply-To: <3AADF3DC.1030003@dynworks.com>; from jdavis@dynworks.com on Tue, Mar 13, 2001 at 02:18:04AM -0800
References: <3AADF3DC.1030003@dynworks.com>
Message-ID: <20010313114624.A3942@freedom.puma-ag.com>

Jeff Davis wrote:
> I was previously a perl user, but now I am taking an interest in python. 

Hurrah!

> Module importing doesn't take a string argument, but just a "bare word" 
> (in perl terms). This means I have to know the name of the file in 
> advance, and it needs to be in the right directory, or I need to use eval().

You can import a module with a given name as follows:

  the_module = __import__('the_module_name')

If you need to get named items out of the module later you can use 'getattr':

  o = getattr(the_module, 'item_name')
  o.square()

> [snip] Right now the 
> concatenation of strings can be annoying compared to just (perl) 
> '$string = "a $anotherstring be";', but for a lot of programs it is well 
> worth the tradeoff.


You can try

  the_string = "a %(anotherstring)s be" % locals()

to get a similar effect but with the additional safety of an exception being
raised if 'anotherstring' is not in the current namespace.

-Steve


-- 
Steve Purcell, Pythangelist
Get testing at http://pyunit.sourceforge.net/
Any opinions expressed herein are my own and not necessarily those of Yahoo



From tgagne at ix.netcom.com  Thu Mar  1 08:56:09 2001
From: tgagne at ix.netcom.com (Thomas Gagne)
Date: Thu, 01 Mar 2001 08:56:09 -0500
Subject: Stream interfaces
References: <3A9BC3F9.1929EB9A@ix.netcom.com> <memo.20010301101637.55617A@brangdon.madasafish.com>
Message-ID: <3A9E54F9.7AF90BCA@ix.netcom.com>

Something has gone wrong only when it's gone wrong.

Dave Harris wrote:

> tgagne at ix.netcom.com (Thomas Gagne) wrote (abridged):
> >                 ExternalStream
> >                     BufferedExternalStream
> >                         ExternalReadStream
> >                             ExternalReadWriteStream
> >                         ExternalWriteStream
> >                 InternalStream
> >                     WriteStream
> >                         ReadWriteStream
>
> Do you agree this hierarchy is a bit duff? Notice how "Write" appears in
> 3 leaf classes: ExternalReadWriteStream, ExternalWriteStream,
> ReadWriteStream. These apparently share no implementation or interface.

My first thought was, "So what?"  Without studying the internals of the
classes to see what was or wasn't inherited it matters little what a class'
superclass is.  A program is only dependent on the classes' interfaces it
uses, not the super classes' interfaces.  It doesn't make sense to me to
critisize a hierarchy unless it has resulted in a faulty or unusable class.  I
haven't read any messages suggesting those throwing rocks at OO discovered
ExternalWriteStream was functionally deficient or misleadingly named.

>
> Notice also that in one place we get a trio of classes:
>
> >                         ExternalReadStream
> >                             ExternalReadWriteStream
> >                         ExternalWriteStream
>
> and in another we get a pair:
>
> >                     WriteStream
> >                         ReadWriteStream
>
> yet they would seem to be doing the same kind of thing. How do I get an
> internal stream which is readable but not writable?

Forgive my attempt at documentation.  Had I know someone would focus on the
hierarcy as closely as this I would have been a little more thorough (my
apologies to CinCom).

    InternalStream
        ReadStream

> If ReadWriteStream
> inherits from WriteStream, why doesn't ExternalReadWriteStream inherit
> from ExternalWriteStream?

I'm unclear why it matters.  Do you want an ExternalReadWriteStream?  Use it.
When you call a C function in a vendor-provided library (like an operating
system) do you analyze the code as closely or are you only thankful the
function exists so you didn't have to write it?

>
>
> Something has gone wrong here.

Your correct.  People are confusing inheritence and usefulness.  I don't have
enough spare cycles to worry about a class' inheritence unless I've discovered
missing functionality.  Since I haven't, I haven't.

Let's examine another one with a hierarchy I'm more familiar with (from the
opensource middleware isectd):

Object
    ExternalInterface
        IsdIO (low-level interface to isdio.so)
    IsdClient (high-level interface to isdio.so, more OOriented)
        IsdWorker (abstract - framework for writing a server)
            IsdEcho (a simple server that returns whatever it receives)
                IsdWorkerReplyStream (an abstract for workers that build their
replies on a stream)
                    IsdEat (a simple worker that collects all the message
fragments and returns stats to the client about them)
                    IsdSmalltalkWorker (send it Smalltalk code, it compiles
and executes it.  Very useful.  Very dangerous)
                    IsdXMLWorkerAbstract  (builds an XML Document from the
request, prints the reply XML Document back)
                        IsdGemWorker (an abstract worker for connecting to
Gemstone/S, adds convenience methods for transactions)
                            IsdEfiWorker (my current probject -- top secret)

>
>
>   Dave Harris, Nottingham, UK | "Weave a circle round him thrice,
>       brangdon at cix.co.uk      |   And close your eyes with holy dread,
>                               |  For he on honey dew hath fed
>  http://www.bhresearch.co.uk/ |   And drunk the milk of Paradise."

--
.tom





From macdonaldrj at att.net  Sat Mar  3 09:36:17 2001
From: macdonaldrj at att.net (Richard MacDonald)
Date: Sat, 03 Mar 2001 14:36:17 GMT
Subject: Collection interfaces
References: <3a90fa97.0@news1.mweb.co.za> <3A933830.7F98D118@ix.netcom.com> <CgJk6.515$v77.66568@newsread2.prod.itd.earthlink.net> <220220010813598042%cmh@bDistributed.com> <ud7cakvi8.fsf@roundpoint.com> <MPG.150312ceb68f1ada98969f@news.earthlink.net> <8Gem6.473353$U46.14105062@news1.sttls1.wa.home.com> <MPG.1503548620e632919896a4@news.earthlink.net> <MCjm6.474652$U46.14163288@news1.sttls1.wa.home.com> <MPG.150394195f7f04a19896a7@news.earthlink.net> <3A9A86B1.9A8C69A2@ix.netcom.com> <MPG.1504d14563dce6499896af@news.earthlink.net> <97fjla128ir@news2.newsguy.com> <MPG.1505a76831c34b749896b3@news.earthlink.net> <97h5s40vhc@news2.newsguy.com> <MPG.1506007591cf818b9896bc@news.earthlink.net> <slrn99opop.gr.amk@mira.erols.com> <MPG.1506265ebf6d934d9896be@news.earthlink.net> <97ibmc0k47@news1.newsguy.com> <MPG.1506cf3293b094b49896c1@news.earthlink.net> <97m0c7080c@news2.newsguy.com> <MPG.15085517debf4b649896ca@news.earthlink.net>
Message-ID: <Bj7o6.2298$yv4.130639@bgtnsc07-news.ops.worldnet.att.net>

"Topmind" <topmind at technologist.com> wrote in message
news:MPG.15085517debf4b649896ca at news.earthlink.net...
>
> Well, my experience tells me that the "task axis"
> is the most "business friendly" axis to build
> along (being that multi-axis paradigms still need
> a lot of work and thus we are forced to emphasize
> one at the expense of others for now.)
>
We'll make an OOer of you yet.
Psst. Less structural taxonomy and more task-oriented
thought. You might even think of a little "agent" doing
that task for you. Then, you'll dress that little agent up
in some nice clothes, and presto, an object.

Or component. Perhaps you should just skip OO and
go right to components. Just don't suggest components
should be wrapped in tables.

I vote YES for comp.top




From aahz at panix.com  Fri Mar 16 23:55:03 2001
From: aahz at panix.com (Aahz Maruch)
Date: 16 Mar 2001 20:55:03 -0800
Subject: ODBC for Python 2?
References: <0IBs6.29099$ZT5.789151@news.total.net>
Message-ID: <98uqn7$2ke$1@panix6.panix.com>

In article <0IBs6.29099$ZT5.789151 at news.total.net>,
Brian Alexander <crystalc at interlog.com> wrote:
>
>Is it possible to make use of Win32 ODBC drivers from Python 2? I would like
>to run updates/synchronizations on a remote SQL database.

http://www.lemburg.com/python/mxODBC.html
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

Three sins: BJ, B&J, B&J


From digitig at cix.co.uk  Wed Mar 21 17:02:00 2001
From: digitig at cix.co.uk (Tim Rowe)
Date: Wed, 21 Mar 2001 22:02 +0000 (GMT Standard Time)
Subject: python suitability for large critical run forever apps
References: <3AB77CE4.78378862@troikanetworks.com>
Message-ID: <memo.20010321220223.57471C@digitig.compulink.co.uk>

In article <3AB77CE4.78378862 at troikanetworks.com>, 
bedge at troikanetworks.com (Bruce Edge) wrote:

> I started using Python for come user CLI stuff. It has worked out so
> well that we're thinking about extending it's use in our project.
> I am not experienced enough to be able to answer some of the issues that
> came up. Does anyone have any thoughts on these:

<snip>

It depends on what you mean by "large critical run forever apps". I 
routinely deal with systems that can kill a few hundred people if they go 
wrong, and that's what /I/ mean by "critical". But then I baulk at the 
suggestion that such a system might be "large", never mind the suggestion 
that it might be multithreaded! If you need something like IEC61508 SIL4 
or the military equivalents then forget Python (and C/C++, and Perl, and 
Ruby...). That doesn't make /any/ of those a "toy" language, just not 
right for that application. I've said before -- here, I think -- that I 
don't believe in general purpose languages; to misquote Michael Jackson, 
if a language offers to helps you with /every/ problem, it won't help you 
/much/ with /any/ problem.


From pearu at cens.ioc.ee  Mon Mar  5 04:05:05 2001
From: pearu at cens.ioc.ee (Pearu Peterson)
Date: Mon, 5 Mar 2001 11:05:05 +0200
Subject: Q: Operator precedence
Message-ID: <Pine.LNX.4.21.0103051051180.12138-100000@kev.ioc.ee>

Section 5.12 in Python reference manual summarizes operator precedence.

I would like to have the following function:

def getprec(s):
     return <the precedence number of a keyword/operator `s'>

For example,

getprec('lambda') -> 0
getprec('or')     -> 1
getprec('and')    -> 2
...
getprec('+')      -> 11
getprec('*')      -> 12
...

One solution would be tabulate the table in Sec.5.12 into a Python
dictonary and use it. 
But is there a way to get this information in runtime?

Thanks,
	Pearu



From aleaxit at yahoo.com  Fri Mar 23 03:34:31 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 23 Mar 2001 09:34:31 +0100
Subject: Digest Number 1946
References: <mailman.985313369.3107.python-list@python.org> <dxCu6.3134$4N4.460882@newsc.telia.net>
Message-ID: <99f1s601dl3@news2.newsguy.com>

"Fredrik Lundh" <fredrik at pythonware.com> wrote in message
news:dxCu6.3134$4N4.460882 at newsc.telia.net...
    [snip]
> nothing's going wrong here -- the system's reminding you that
> when you're using non-blocking mode, the fact that connect
> returns doesn't mean that it has succeeded yet.

Excellent summary!


> you're supposed to catch that exception, and check for com-
> pletion using select or poll.  see your favourite socket pro-
> gramming guide for more info.

Very good Python example code can be found as a part of
the extremely-useful timeoutsocket module at:
http://www.timo-tasi.org/python/timeoutsocket.py

(It needs some fixes for at least some versions of NT, but
I haven't heard back from the author after submitting them;
still, it's SO useful that, after using it as an example of
how to work with non-blocking sockets, one may almost forget
that and use timeoutsocket's directly all of the time...!-).


Akex





From JohnH at PHM.GOV.AU  Tue Mar 20 18:40:47 2001
From: JohnH at PHM.GOV.AU (Hirsch, John)
Date: Wed, 21 Mar 2001 10:40:47 +1100
Subject: Using ActiveX Control from Python
Message-ID: <A90DD8D7B82FD311B4D00008C7DBD6EAB5EBE3@phm-exch.phm.gov.au>

>Can you provide any pointers? Even subject lines would help.

Best to do a search at Deja for posts by Alex Martelli.
He has written quite a lot about activex.



From aahz at panix.com  Wed Mar 28 09:44:27 2001
From: aahz at panix.com (Aahz Maruch)
Date: 28 Mar 2001 06:44:27 -0800
Subject: list vs tuple
References: <hGlw6.782$p5.3210@news1.rivrw1.nsw.optushome.com.au>
Message-ID: <99stcb$483$1@panix6.panix.com>

In article <hGlw6.782$p5.3210 at news1.rivrw1.nsw.optushome.com.au>,
deadmeat <root@[127.0.0.1]> wrote:
>
>Whats the practical difference between a list[] and a tuple() ?
>
>A tuple seems to be a very basic list, so unless it's faster, what's the
>point of it?

As someone said, tuples are immutable.  This leads to useful
functionality, such as being able to easily use tuples as keys in
dictionaries (ditto strings).

While it's theoretically possible to use mutable variables as dict keys,
it creates icky problems.
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

"Boost the stock market -- fire someone"


From paul.moore at uk.origin-it.com  Mon Mar  5 10:26:18 2001
From: paul.moore at uk.origin-it.com (Paul Moore)
Date: Mon, 5 Mar 2001 16:26:18 +0100 
Subject: urllib.urlopen and proxies
References: <vse4atgcgcfafth7c5tq3po7srd3l32f3h@4ax.com> <mailman.983800688.6363.python-list@python.org>
Message-ID: <OLCjOkkqlhMaJ2JeJkisvppqbKR1@4ax.com>

On Mon, 5 Mar 2001 15:03:35 +0100 , "Chris Gonnerman"
<chris.gonnerman at usa.net> wrote:

>It *appears* (from reading the source) that username and password
>information for
>the proxy server must be passed as part of the URL:
>
>    http://username:password at my.host.com/index.html
>
>for instance.  I haven't tested this yet.  The source for urllib is a bit
>dense IMHO but the answers are in there.

Doesn't work for me... I attach an example, in case there are any
pointers in the result.

Paul

--- Example session ---

>>> from urllib import urlopen
>>> print urlopen("http://UUUU:PPPP at www.python.org/").read()
<HTML><HEAD>
<TITLE>ERROR: Proxy Access Denied, authentication failed</TITLE>
<!-- $Id: ERR_CACHE_ACCESS_DENIED,v 1.1 2000/12/20 12:11:12 devet Exp
$ -->
</HEAD>
<BODY BGCOLOR="#ffffff">
<H1>ERROR</H1>
<H2>Proxy Access Denied, authentication failed</H2>
<HR>
<P>
While trying to retrieve the URL:
<A HREF="http://www.python.org/">http://www.python.org/</A>
<P>
The following error was encountered:
<UL>
<LI>
<STRONG>
Proxy Access Denied, authentication failed.
</STRONG>
</UL>
</P>

<P>Sorry, you are currently not allowed to request:
<PRE>    http://www.python.org/</PRE>
from this proxy until you have authenticated yourself.
</P>

<P>Please use your NT userid <B>without the "domain\" part</B> and
your
NT password. <B>Example</B>: "cc01234" (not "europe\cc01234"!) and
"zekret!". </P>

<P>

See the <A
HREF="http://channels.origin-it.com/newsbin/center/fulltext.asp?msgId=NE20001207
110741C111083">
announcement</A> on Channels for further information.

<P>
You need to use Netscape version 2.0 or greater, or Microsoft Internet
Explorer 3.0, or an HTTP/1.1 compliant browser for this to work.
Please
contact the the helpdesk if you have difficulties authenticating
yourself.
</P>

<br clear="all">
<hr noshade size=1>
Generated Mon, 05 Mar 2001 15:20:35 GMT by www-proxy2.nl.origin-it.com
(<a href=
"http://squid.nlanr.net/Squid/">Squid/2.2.STABLE4</a>)
</BODY></HTML>



From menno at phreaker.net  Wed Mar 14 21:55:33 2001
From: menno at phreaker.net (Breadfan)
Date: Thu, 15 Mar 2001 03:55:33 +0100
Subject: Images in a MySQL Database ?!?!
Message-ID: <98qlb3$dp1$1@news1.xs4all.nl>

Hello, all. I'm trying to store images  in a MySQL database, using a upload
form on a zope server and it doesnt store it , it stores a string like this
:

"<ZPublisher.HTTPRequest.FileUpload instance at 86550c8>"

The forrm I use looks like this :

<form action="Insert_Banner_DTML"
method="post"enctype="multipart/form-data">
<input type="file" name="file">
<input type="submit">
</form>

And the SQL I use is this :
"
file:string
---------------
INSERT into images (Image) VALUES
(
<dtml-sqlvar file type=string>
)
"

Can someone please help me out, by telling me what I'm doing wrong ?
I've tried to look for info on this matter, but i seem to find any good
material on it  : (

Greetings,
Menno






From tim.one at home.com  Wed Mar 14 03:18:13 2001
From: tim.one at home.com (Tim Peters)
Date: Wed, 14 Mar 2001 03:18:13 -0500
Subject: Atomic Event.{wait,clear}
In-Reply-To: <c22l89.2v1.ln@127.0.0.1>
Message-ID: <LNBBLJKPBEHFEDALKOLCIEJFJFAA.tim.one@home.com>

[David Given]
> I'm using the threading module, with the Event class to handle
> synchronisation. It's all working nicely and is very easy to use.
>
> However, I can't seem to find a way of atomically waiting for an Event,
> and then clearing the flag, which is very strange. Currently I do
> Event.wait() and then Event.clear(). If another thread does a Event.set()
> in between the two, the event will be lost. For my current application,
> this isn't a problem, but what shold I do in the future?

Use a more appropriate synch mechanism.  Events are usually one-shot
broadcast mechanisms.  For example, in a multi-pass algorithm over a giant
array, send 10 threads off to do the first pass, then when each is done have
each wait for the "first pass is done; move on to the second pass" Event.
This doesn't work, though, if any thread clears the event before all have
seen it.  Most times an Event never gets cleared, and (as you've discovered!)
it's quite tricky to do so safely in any but trivial cases.

Hard to guess what you're doing, but it sounds like you're using an Event as
a "hey, there's something to do!" signal, where there will be something to do
more than once over the life of your job.  If so, a Condition is very good
for that:  the Condition protocol .notify and .wait occur under the
protection of a mutex, so after a thread T does a successful
Condition.wait(), no thread can do another Condition.notify() before T does
Condition.release() (the *reason* no other thread can do another
Condition.notify() is that they have to do a Condition.acquire() first, but
before T does its .release() all other threads will block on the .acquire()).

luckily-it's-easier-to-use-than-to-explain<wink>-ly y'rs  - tim





From abrahams at mediaone.net  Sat Mar  3 22:35:43 2001
From: abrahams at mediaone.net (David Abrahams)
Date: Sun, 04 Mar 2001 03:35:43 GMT
Subject: GC and extension types
Message-ID: <jKio6.8061$3F2.2558420@typhoon.ne.mediaone.net>

Is there documentation someplace on what I must do to make my extension
types play nicely with GC? I am worried that if I store a pointer to a
PyObject* in some arbitrary memory (and increment its reference count, of
course!), the collector might decide it was unreachable anyway and collect
it.

>From looking at the source code, I have an inkling that it might try to
account for all the references in a cycle, so that objects will only be
collected when all known references can be accounted for by the (reachable)
cycles in which it participates. I guess this would create the opposite but
more benign problem of being too conservative.

Anyway, I'm hoping there is authoritative text somewhere that describes
what's really going on... and that the text isn't just 'C' code!

Any help appreciated,
Dave




From claird at starbase.neosoft.com  Mon Mar  5 14:06:11 2001
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 5 Mar 2001 13:06:11 -0600
Subject: Python and Tcl (was: Questions from a beginner)
References: <1103_983792888@breakdown> <980gm5$gau$1@news.udel.edu>
Message-ID: <175FA9530C75FE03.40C0C547F170B40E.7949D7E2AD041AC9@lp.airnews.net>

In article <980gm5$gau$1 at news.udel.edu>, Terry Reedy <tjreedy at udel.edu> wrote:
			.
			.
			.
>> especially if I write things using tcl (this is probably a long way off
>yet, but I'd like to know).
>
>I'll let a tcler answer that (but I presume so).
Alex and I both concluded that Alison intended something else in
writing "tcl" here.

However, while we're on the subject, there are now a couple of
systems that allow programs to be written in combinations of Tcl
and Python.  TclPython <URL: http://jfontain.free.fr/tclpython.htm >
is the most polished of these.
			.
			.
			.
-- 

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


From chapman at bioreason.com  Tue Mar  6 12:23:32 2001
From: chapman at bioreason.com (Mitch Chapman)
Date: Tue, 06 Mar 2001 10:23:32 -0700
Subject: absolute namespace for callbacks?
References: <3AA50AFB.83C7459F@troikanetworks.com> <983895797.296667@newsmaster-04.atnet.at> <3AA51144.3DD3FC3F@troikanetworks.com>
Message-ID: <3AA51D14.733D06D@bioreason.com>

Bruce Edge wrote:
> 
> My callback func is defined in one module, and the callback is executed
> from another module which does not know in advance what module the
> callback func resides in. So, by default, the caller cannot find the
> callback func in it's own namespace.

It doesn't need to.  It just needs somebody to pass it a reference
to the callback function so it can save it in its own namespace.

________________________________________________________________________
# Module1.py
def myCB(*args):
    print "Where the heck am I?  Then again, who cares?"
________________________________________________________________________
# Module2.py
_callback = None

def register(newCB):
    global _callback
    _callback = newCB

def notify():
    if _callback:
        _callback()
________________________________________________________________________
# Module3.py
import Module1, Module2

Module1.register(Module2.myCB)
________________________________________________________________________


> The callback registration procedure needs to specify the namespace where
> the callback func exists.

I'd modify that to say, "whoever calls the callback registration 
procedure may need access to the namespace where the callback func 
is defined."  In other words, Module1 and Module2 don't need
to know anything about each other.  Only the composer which glues
the two together needs to know about both, and even it needs only
references to the callback and registration functions.

Does this answer the right question?

--
Mitch Chapman
chapman at bioreason.com


From ben.hutchings at roundpoint.com  Thu Mar  8 23:47:18 2001
From: ben.hutchings at roundpoint.com (Ben Hutchings)
Date: 08 Mar 2001 20:47:18 -0800
Subject: "Wrap" a python-cgi
References: <3aa7c5f9$0$21482@SSP1NO25.highway.telekom.at>
Message-ID: <un1av8s6h.fsf@roundpoint.com>

Hermann Himmelbauer <dusty at strike.wu-wien.ac.at> writes:

> Hi,
> I am currently using python for several cgi programs. One problem is that 
> if my cgi-script has a bug it exits with an exception and the user gets an 
> ugly error message done by the webserver.
> 
> I would like to put some really simple "bugfree" (if something like this 
> can exist :-) "wrapper" around my cgi-scripts, let it trap Exceptions, log 
> the error and let it output a nice error page to the user.
> 
> My first approach would perhaps be to wrap the whole script into a "try: 
> except" clause but at first I don't know if nested try/except clauses are 
> allowed

They are.

> and moreover the whole script looks somehow ugly as every line is 
> indented.
> 
> Perhaps you know a better solution?

Make the wrapper a script that imports the original script as if it's
a module, inside a try block.  This also means that Python will use a
byte-compiled version of the original script if there is one, saving a
little time.

(But it shouldn't be able to create byte-compiled versions when
running the script through CGI, if you're being sufficiently paranoid
about permissions and ownership of files and processes.)

-- 
Any opinions expressed are my own and not necessarily those of Roundpoint.


From n8gray at caltech.edu.is.my.e-mail.address  Tue Mar  6 18:22:28 2001
From: n8gray at caltech.edu.is.my.e-mail.address (Nathan Gray)
Date: Tue, 06 Mar 2001 15:22:28 -0800
Subject: Long Shifty Bits
References: <kmap6.15$8Q.28605@news.pacbell.net>
Message-ID: <3AA57134.1070401@caltech.edu.is.my.e-mail.address>

Phlip wrote:

> Not hyp:
> 
> I want this expression to make sense:
> 
>         x = 1L << 200
> 
> If someone tells me "it will roll over at the 32nd bit", or mayby even "it 
> will roll over when the CPU damn well feels like it", that's okay. I'l just 
> construct a Bit Vector of some ilk and drop my bits into it. But the above 
> notation is closest to our problem space, so it would be nice if it worked 
> as expected.

[n8gray at term92 controllers]$ python
Python 1.5.2 (#1, Feb  1 2000, 16:32:16)  [GCC egcs-2.91.66 
19990314/Linux (egcs- on linux-i386
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
 >>> x = 1L << 200
 >>> x
1606938044258990275541962092341162602522202993782792835301376L
 >>> x >> 200
1L


-n8

-- 
_.~'^`~._.~'^`~._.~'^`~._.~'^`~._.~'^`~._
               Nathaniel Gray
     California Institute of Technology
      Computation and Neural Systems
       n8gray <at> caltech <dot> edu
_.~'^`~._.~'^`~._.~'^`~._.~'^`~._.~'^`~._



From tim.one at home.com  Sat Mar 24 02:17:34 2001
From: tim.one at home.com (Tim Peters)
Date: Sat, 24 Mar 2001 02:17:34 -0500
Subject: New PEP: The directive statement
In-Reply-To: <j4elvokhs9.fsf@informatik.hu-berlin.de>
Message-ID: <LNBBLJKPBEHFEDALKOLCGEADJIAA.tim.one@home.com>

[Martin von Loewis]
>>> Furthermore, what is the use of that information to a program?

[Tim]
>> Primarily, it's what makes it possible to write a simple tool
>> that nukes obsolete future statements.

[Martin von Loewis]
> I'd question this use of the information.  It has yet to be written,

True, but I've written many harder analyzers + translators for Python, and
several have been in the distribution for years (e.g., tabnanny.py,
checkappend.py, most recently reindent.py for 2.1).  So this FUD won't fly.

In return, neither is there a tool to nuke obsolete transitional directives.
The difference is that I don't know how to write the latter, in the absence
of PEP 236's __future__.py release-info database.

> but I hope it won't operate in automatic "nuke all" mode. E.g. when
> case insensitivity is the default in Python 2.5, Python 2.3 will still
> be around (say). So the author of some package may want to nuke the
> nested_scope directives, since nobody uses Python 2.1 anymore, but
> would leave the case_insensitive directives.

Sure.

> In short, I hope that the program would have a dry-run mode,

Different issue, but e.g. reindent.py has a dry-run mode, and I expect to
derive a future-nuker via transforming reindent.py.

> where it reports all directives it finds, and a nuking mode, where
> it is given a list of directives that it will remove.

Ah, note that I'm not signing up for writing a general *directive*
manipulation tool.  That's too open-ended for me.  The tool PEP 236
hypothesizes is limited to nuking obsolete future stmts, and PEP 236 is not
an open-ended scheme.  The tool is easy because the PEP is limited.

>> In another vein, in my own code I don't hesitate to use new features,
>> and couldn't care less if rare incompatible changes require me to
>> rewrite some code.  So I'll likely use future_statements much more
>> than most people.  I already have code that does
>>
>>     from __future__ import nested_scopes
>>     import sys
>>     assert nested_scopes.getMandatoryRelease() > sys.version_info
>>
>> This will assert-fail as soon as I run it under a release where
>> nested_scopes is the rule, and at that point I'll simply delete the
>> future_statement and move on.

> That sounds like a legitimite use, but I have my problems
> understanding it.  Are you relying on nested scopes in your code?

Yes, else I wouldn't have bothered to include the

    from __future__ import nested_scopes

line.

> If so, why do put the assert statement there?

I put the assert near the line it's complaining about, because the line it's
complaining about is the line I want to get rid of someday, and I want the
assert to tell me that.  Maybe a wordier version would be clearer:

    assert nested_scopes.getMandatoryRelease() > sys.version_info, \
         "no more need for the nested_scopes stmt above -- nuke it"

> Or does your code give a warning and you want to shut it up using
> the future import?

No.

> In that case, wouldn't it be better to use the warning framework?

People can do that too, or instead, if they like.  I was just giving examples
of ways the __future__ repository *can* be used.  As a passive data
repository, it's mechanism not policy.  People can *use* the info in any way
whatsoever they can dream up code to implement.

>> Code like this also works fine:
>>
>> import __future__
>> if __future__.nested_scopes.getMandatoryRelease() > sys.version_info:
>>     import old_version_of_module as amodule
>> else:
>>     import amodule
>>
>> I don't know that I'll have an acute need for that wrt nested_scopes
>> specifically

> Why are you checking the mandatory release here? Doesn't amodule have
> a future import in it?

Since I'm checking MandatoryRelease, it's irrelevant whether amodule has a
future import in it (amodule won't be imported unless the release I'm running
under is such that __future__ import of nested_scopes has no effect).

As I said above, I'll nuke obsolute future stmts in my own code ASAP, so in
*my* code checking MandatoryRelease is appropriate.  In *your* code perhaps
you'd like to check OptionalRelease instead, and keep old future stmts around
forever.  I'm not telling you which you should do, just giving an example of
what *I'd* do.  Again, __future__ is mechanism not policy:  you can code it
any way *you* like.  OK by me, because I can still code it the way *I* like
it, and I've generally got no interest in clinging to old versions.

>> In that case I may very well have an acute need to ship one version
>> of my source that exploits the new semantics but falls back to an
>> older & slower module if the compiler it's running under can't play
>> along.

> People routinely use sys.*version* for that; I'm really curious
> whether they'd prefer to investigate other variables instead.

I think PEP 236 is extremely clear that the release info in __future__.py is
in sys.version_info format:

    ...
    where, normally, OptionalRelease <  MandatoryRelease, and both are
    5-tuples of the same form as sys.version_info:

    (PY_MAJOR_VERSION, # the 2 in 2.1.0a3; an int
     PY_MINOR_VERSION, # the 1; an int
     PY_MICRO_VERSION, # the 0; an int
     PY_RELEASE_LEVEL, # "alpha", "beta", "candidate" or "final"; string
     PY_RELEASE_SERIAL # the 3; an int
    )

How could it be clearer?  Nobody complained; Barry pointed out that he
generally uses sys.hexversion (not sys.version) instead, but allowed that
sys.version_info is easier to read; and I can't imagine that anyone in their
right mind uses sys.version for stuff like this, since that's a string and is
not defined in a way that supports correct comparison (e.g., "10.0" < "2.1",
when we get that far).

BTW, note that the std test test___future__.py verifies that the info in
__future__.py *is* of the form described above.

Note too that, as you pointed out two msgs back, even if you're wedded to
sys.hexversion, it's not possible to write a hard-coded version check that's
guaranteed to *work* for these things across releases, since the
MandatoryRelease for a given feature may change across releases.  You really
need to access some form of info that ships with the compiler in order for
this to work correctly.  Putting the info in the std library makes it
painless to get at, and PEP 236 covers that too.

have-yet-to-hear-a-complaint-from-anyone-using-it<0.6-wink>-ly y'rs  - tim




From pj at sgi.com  Thu Mar  8 23:31:48 2001
From: pj at sgi.com (Paul Jackson)
Date: 9 Mar 2001 04:31:48 GMT
Subject: Can anyone recomend a good intoduction to C...
References: <mailman.983894288.19965.python-list@python.org> <ewwp6.1466$y6.241005@ruti.visi.com> <slrn9aeblb.cg1.scarblac@pino.selwerd.nl> <djagat4kk3cs9hg0c3k39cudk6goblhkg6@4ax.com>
Message-ID: <989mbk$24nuv$1@fido.engr.sgi.com>

Courageous asks:
> Echo that. What's "object oriented" about templates?
>                                                     
> A: NOTHING.
> !!


No need to shout.  Templates support generic programming,
not object oriented programming.

Bjorn's goal, clearly, was to support a number of styles,
of which O-O is just one.

On the other hand ... I agree at some level with your
frustration.  Despite years of C++ proramming, including
some non-trivial template programming (my mind has a
natural affinity for that style), I still find C++ just
plain "too big" for my modest brain.  Even though, for
some projects, it is the language of choice.

For me, the "projects of choice" are those for which
Python is suitable, not those requiring C++.

Thank-you, Guido, for one very special language.
-- 
                          I won't rest till it's the best ...
                          Manager, Linux System Software
                          Paul Jackson <pj at sgi.com> 1.650.933.1373


From fredrik at pythonware.com  Thu Mar 22 02:55:31 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Thu, 22 Mar 2001 07:55:31 GMT
Subject: re.match
References: <mailman.985169080.17724.python-list@python.org> <BN3u6.894$sk3.291185@newsb.telia.net> <Wmgu6.4572$WC2.157632@e420r-atl1.usenetserver.com>
Message-ID: <Tdiu6.2956$4N4.403069@newsc.telia.net>

Steve wrote:
> I thought Unix was Ken's attempt to extend the reach of
> regular expressions to the user base?

could be, could be.

or maybe it was an attempt to introduce more humor into
computing ("UNIX", "g/re/p", "dd", "biff", "awk", etc).  given
the recent "__future__ isn't serious" brouhaha, I'm afraid
they might have failed...

Cheers /F




From chris.gonnerman at usa.net  Sun Mar  4 23:01:54 2001
From: chris.gonnerman at usa.net (Chris Gonnerman)
Date: Sun, 4 Mar 2001 22:01:54 -0600
Subject: Newbie question about nested lists and constructor functions...
References: <W1Do6.4683$ke1.121750@jekyl.ab.tac.net>
Message-ID: <017b01c0a529$0692cca0$a100000a@local>

Could you please post example(s) so we know exactly what is giving you
trouble?  I at least don't quite know what you are asking.

The constructor function (Python calls it "__init__") is created thus:

class Spam:
    def __init__(self):
        ...

and of course __init__ might have additional arguments.  If you are creating
a nested list with known initializers, you might say:

myvar = [ [ "list", 2, 3, 4 ], [ 'A', 'B', 'C' ] ]

so if you want to create a nested list in an __init__ function:

class Spam:
    def __init__(self):
        self.eggbasket = [ [ "list", 2, 3, 4 ], [ 'A', 'B', 'C' ] ]

to assign a nested list to a class attribute.  If you are passing in
information
which you wish to assign that way, you would embed it in the given syntax.
For
instance:

class Spam:
    def __init__(self, eggs):
        self.eggbasket = [ eggs, [ "spam", "spam", "spam" ] ]

my_spam = Spam([ "egg", "egg", "egg"])

which would nest the given list of eggs within the eggbasket member of
my_spam.
You might alternately want to make a copy of the initializer instead of
storing
a reference:

class Spam:
    def __init__(self, eggs):
        self.eggbasket = [ eggs[:], [ "spam", "spam", "spam" ] ]

[:] (slice of all) is the Pythonic "shallow" list copier.

Of course this is all potentially nonsense as, again, I don't exactly know
what
you are wanting to do.

----- Original Message -----
From: "Bartok" <JollyRoger at hotmail.com>
Subject: Newbie question about nested lists and constructor functions...
> I need some help with nested lists when using class constructors to
> initialize their contents.  I found the tutorial to be of not much help in
> this area and have tried everything that I can think of.
>
> I understand the process as I am an experienced C++ programmer but I seem
to
> be having some difficulty using the correct syntax.  Can anyone please
help?





From jmg at ecs.soton.ac.uk  Wed Mar 14 05:26:07 2001
From: jmg at ecs.soton.ac.uk (Jacek Generowicz)
Date: 14 Mar 2001 10:26:07 +0000
Subject: Extending Python: rewriting a single method in C
References: <g0wv9vxms0.fsf@scumbag.ecs.soton.ac.uk> <slrn9apn8f.382.jepler@potty.housenet> <g0ofv6xbj4.fsf@scumbag.ecs.soton.ac.uk> <98l1i901k4u@news1.newsguy.com> <g0lmq9xxz0.fsf@scumbag.ecs.soton.ac.uk> <gYur6.119$4d3.23079@newsb.telia.net>
Message-ID: <g0hf0wy7cw.fsf@scumbag.ecs.soton.ac.uk>

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

> Jacek Generowicz wrote:
> > (Would it be fair to say that the documentaion is, on the whole, less
> > that satisfactory ?)
> 
> are you sure you read the C API docs before posting?

No, I'm quite sure that I didn't . . . for some reason I was buried
in http://www.python.org/doc/current/ext

(Oops! Blush)

> http://www.python.org/doc/current/api/

This is much more like it.

Thanks,

Jacek


From aleaxit at yahoo.com  Thu Mar 29 04:10:21 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 29 Mar 2001 11:10:21 +0200
Subject: Getting all the *files* from a directory -- A better way??
References: <mailman.985816117.6859.python-list@python.org>
Message-ID: <99uu6401u9c@news2.newsguy.com>

<nanotech at europa.com> wrote in message
news:mailman.985816117.6859.python-list at python.org...
> Thanks to everyone who answered!
>
> Notice that:
>
>   dir="/a/b/c"
>
>   files1=[f for f in os.listdir(dir)
>             if os.path.isfile(f)]
>   files2=[os.path.basename(f) for f in os.listdir(dir)
>                               if os.path.isfile(f)]
>
> would be much "prettier" than:
>
>   files1=[os.path.join(dir,f) for f in os.listdir(dir)
>                               if os.path.isfile(os.path.join(dir,f))]
>   files2=[f for f in os.listdir(dir)
>             if os.path.isfile(os.path.join(dir,f))]
>
> if os.listdir returned the given path prepended -- but oh well!!

So wrap it up...:

def joined_listdir(dir):
    return [os.path.join(dir,f) for f in os.listdir(dir)]

then use your wrapper:

files1 = [f for f in joined_listdir(dir) if os.path.isfile(f)]

etc.


Alex





From greg at cosc.canterbury.ac.nz  Tue Mar 20 00:06:31 2001
From: greg at cosc.canterbury.ac.nz (Greg Ewing)
Date: Tue, 20 Mar 2001 17:06:31 +1200
Subject: Why "from __future__" stinks; a counter-offer
References: <mailman.985029808.7088.python-list@python.org> <memo.20010319213204.8771C@digitig.compulink.co.uk> <9965t7$s57$1@panix6.panix.com>
Message-ID: <3AB6E557.6D9DFF05@cosc.canterbury.ac.nz>

Aahz Maruch wrote:
> 
> One of the nice things about this solution is that if it is decided to be
> an ugly wart, it can be easily dropped for Python 2.2.

So we may one day find ourselves writing

  from __future__ import no_future_statement

:-)

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


From loewis at informatik.hu-berlin.de  Thu Mar 29 02:03:08 2001
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 29 Mar 2001 09:03:08 +0200
Subject: PEP 245: Python interfaces
References: <Pine.LNX.4.32.0103222253510.26302-100000@localhost.localdomain> <99o9g4$nh3$0@216.39.170.247> <Pine.LNX.4.21.0103261430360.25260-100000@localhost.localdomain>
Message-ID: <j4wv99m503.fsf@informatik.hu-berlin.de>

Michel Pelletier <michel at digicool.com> writes:

> Because that expression list defines inheritance, and interfaces are not
> inherited.  

That depends on your viewpoint. I'd, it defines subtyping, i.e. an
is-a relationship. An object implementing foo certainly is-a foo.

Regards,
Martin


From hsauro at cxx.calxxx.edu  Thu Mar 22 01:59:59 2001
From: hsauro at cxx.calxxx.edu (Caltech News Server)
Date: Wed, 21 Mar 2001 22:59:59 -0800
Subject: socket programming
References: <99bb0g0nce@enews3.newsguy.com>
Message-ID: <99c7l9$d63@gap.cco.caltech.edu>

I think one possiblity is to send along with your string the number of
characters in your string. Then at the receving end, make two recv calls,
one to get the number of chars coming up and a second to retrieve all the
chars. I'm not sure but you may have to continue receiving until you've got
all characters.

Herbert Sauro

"Kevin X Lin" <klin at palsoccer.com> wrote in message
news:99bb0g0nce at enews3.newsguy.com...
> Hi there, I have the following problem with socket programming:
>
> I'm trying to setup a tcp client to talk to a tcp server sitting at a
> certain port. When I issue :
>
> mysocket.setblocking(1)
> mysocket.send('something')
> mysocket.recv(1024)
>
> I was expecting I can receive all the data sent by server. Unfortunately,
> this is not necessary the case. Sometimes I can't get all of data, worse,
> the application protocal doesn't define end tag of the data, which prevent
> me from doing a loop until I receive all data.
>
> But if I do this:
>
> mysocket.setblocking(1)
> mysocket.send('something')
> sleep(0.1)
> mysocket.recv(1024)
>
> I can get all data, yet I believe this is not a good solution which might
> casusing other problems in the future.
>
> Any suggestion?
>
> Thanks.
>
> Kevin
>
>
>
>




From LangaK at discoveryhealth.co.za  Mon Mar 12 00:25:05 2001
From: LangaK at discoveryhealth.co.za (Langa Kentane)
Date: Mon, 12 Mar 2001 07:25:05 +0200
Subject: List removal
Message-ID: <A69BC9CA4B39D411BF2200105A45B45B08834794@dhexchange.discoveryhealth.co.za>

Please tell me how to UNsubscribe to this list.

Thanks
Langa



From jwbnews at scandaroon.com  Fri Mar 16 10:29:30 2001
From: jwbnews at scandaroon.com (John W. Baxter)
Date: Fri, 16 Mar 2001 07:29:30 -0800
Subject: I come to praise .join, not to bury it...
References: <mailman.983485923.1136.python-list@python.org> <97p7iq$kk2$1@nntp6.u.washington.edu> <97rka002k9@news1.newsguy.com> <3AAF7E47.294617E1@sage.att.com> <98qeco02hn7@news1.newsguy.com> <98srjf$fov$1@tyfon.itea.ntnu.no> <98t5t40lle@news1.newsguy.com>
Message-ID: <jwbnews-13D058.07293016032001@corp.supernews.com>

In article <98t5t40lle at news1.newsguy.com>,
 "Alex Martelli" <aleaxit at yahoo.com> wrote:

> If you HAD used Perl a lot (and I have) you might well
> go beyond this "very surface" impression.  Basically,
> Python goes a good way towards meeting the principle
> of least astonishment.  A common and delightful experience
> of the Python learner -- you learn about something new
> but don't know the details yet -- you guess at those
> details *and more often than not, you guess right*!

I came to Python (and dragged my company along) by way of a flooded 
Tower Books store [Bellevue, WA].  They had a [slightly] water damaged 
"Programming Python" on the shelf at something like 60% off as part of 
their flood sale.  [The bottoms of most of the pages are still stained 
and oddly curled.]

Alex expresses very well above a main part of the reason we converted to 
Python from Perl (and became an early corporate member of PSA).

  --John

-- 
John W. Baxter   Port Ludlow, WA USA  jwbnews at scandaroon.com


From emile at fenx.com  Fri Mar 23 10:17:46 2001
From: emile at fenx.com (Emile van Sebille)
Date: Fri, 23 Mar 2001 07:17:46 -0800
Subject: tempfile problem on win98
References: <g3nu6.369$hb.2141@news1.belrs1.nsw.optushome.com.au> <99d0dg$os92$1@ID-11957.news.dfncis.de> <viJu6.473$hb.2279@news1.belrs1.nsw.optushome.com.au>
Message-ID: <99fpkt$12oav$1@ID-11957.news.dfncis.de>

Steven,

You can pass mktemp a suffix parameter, ala:

 id=tempfile.mktemp(suffix='spam')

HTH,
--

Emile van Sebille
emile at fenx.com

---------
"Steven Adams" <quickdry at users.sourceforge.net> wrote in message
news:viJu6.473$hb.2279 at news1.belrs1.nsw.optushome.com.au...
>
> "Emile van Sebille" <emile at fenx.com> wrote in message
> news:99d0dg$os92$1 at ID-11957.news.dfncis.de...
> > The mktemp routine guarantees to return a name unique in the target
> > directory.  Are you sure you're using fully qualified path names
> throughout
> > when subsequently discovering duplicate files?
> >
> > Emile van Sebille
> > emile at fenx.com
>
> ugh, I feel dumb, thats exactly the problem Emile, the file was
> getting an extension added and so it wasn't actually unique, is there
> a way to have a suffix included in the calculation of the filename
> instead of the prefix?
>
> thanks for opening my eyes to the glaringly obvious :-)
>
> Steven
>
>
>




From eddthompson at ssi.nospamhere.parlorcity.com  Thu Mar 29 23:28:14 2001
From: eddthompson at ssi.nospamhere.parlorcity.com (EuphoriaDj)
Date: Thu, 29 Mar 2001 23:28:14 -0500
Subject: list vs tuple
References: <hGlw6.782$p5.3210@news1.rivrw1.nsw.optushome.com.au> <99stcb$483$1@panix6.panix.com>
Message-ID: <3ac40b87$0$10580@wodc7nh0.news.uu.net>

it sound as if tuples are simular to DATA in applebasic. where you call data
one item at a time in order but with tuples you can grab from anywhere
Am I right?




From ruediger.maehl at dzsh.de  Fri Mar  9 03:20:49 2001
From: ruediger.maehl at dzsh.de (Ruediger Maehl)
Date: Fri, 9 Mar 2001 09:20:49 +0100
Subject: Database with Python ??
References: <988vmn$t77$05$1@news.t-online.com>
Message-ID: <newscache$nf8x9g$748$1@www.dzsh.de>

Thomas Mundt schrieb in Nachricht <988vmn$t77$05$1 at news.t-online.com>...
>Has Python a own database language and do someone know a link to a Tutorial
>in database-working with python?

Are you looking for a full-featured database (several GB space,
transactions,
full sql support etc) then you will probably find www.sapdb.org useful.

If you are looking for ODBC based access to databases,
www.lemburg.com/files/python/ might be useful.

But if you only want to save some data to disk then you would perhaps
be fine with the bsddb module (Python Library Reference - Optional
Operating System Services - bsddb) or even the ConfigParser module
(Python Library Reference - Miscellaneous Services - ConfigParser).

HTH, else specify your needs more precisely.
R?diger




From cribeiro at mail.inet.com.br  Fri Mar 30 11:26:33 2001
From: cribeiro at mail.inet.com.br (Carlos Alberto Reis Ribeiro)
Date: Fri, 30 Mar 2001 13:26:33 -0300
Subject: Why not an __assign__ method?
Message-ID: <5.0.2.1.0.20010330125432.0216c4b0@mail.inet.com.br>

Just a question. I did some search around at google and found a few 
messages that were exchanged while PEP203 was being discussed, but no followup.

Relevant messages:
http://groups.google.com/groups?q=__assign__&hl=pt&lr=&safe=off&rnum=1&seld=993477252&ic=1
http://groups.google.com/groups?q=__assign__&hl=pt&lr=&safe=off&rnum=2&seld=999522195&ic=1

The question is: Why not have a __assign__ method, that gets called on 
assignment? It should be called on the *right side* of the expression (at 
the left side it does not make sense - I leave the proof as an exercise for 
the reader <wink>). The return value would be then assigned to the left 
side of the expression. This is the default behavior:

class MyClass:
   def __assign__(self):
     return self

I have thought out this while pondering about some questions on 
optimization of some numerical operations. Take this expression, where a, 
b, and c are all NumPy arrays:

z = a + b + c

Then the expression is broken in steps like this (not exact syntax, but 
enough to give you the idea):

i1 = a.__add__(b)
i2 = i1.__add__(c)
return i2

There are two instantiations of temporary arrays, where only one would be 
enough. With big arrays the extra temp copy is quite an issue. I intend to 
do some image manipulatin, and that surely will help a lot.

The mechanism that I have in mind is:

1) Set a "temp" flag for objects created in the middle of an expression.
2) If a binary operation is called upon a temporary object, use it to store 
the result (of course, if the semantics allows).
3) Reset "temp" on assign.

This has several advantages:

- it allows for the NumPy optimization;
- it may allow for other similar optimizations;
- it does not break any old code, with the exception of the
   ones that implement __assign__; they should not be using
   this name anyway, as it is reserved (double leading
   underscores);
- it is consistent with other similar methods;
- it is much easier than full fledged lazy expression
   evaluation.

The main concerns are:

- it may allow for some very strange errors and bizarre
   behavior if misused;
- the method is called on the right side of the expression,
   which is the opposite of the normal behavior. However
   I don't think it's that bad; after all we have all the
   __r*__ methods.


p.s. Maybe __assign__ is not a good name; maybe something like __onassign__ 
is better, but this does not change the heart of the proposal.



Carlos Ribeiro





From Eugene.Leitl at lrz.uni-muenchen.de  Fri Mar  9 08:46:53 2001
From: Eugene.Leitl at lrz.uni-muenchen.de (Eugene Leitl)
Date: Fri, 9 Mar 2001 14:46:53 +0100 (MET)
Subject: Tk Tree Widget
Message-ID: <Pine.GSO.4.03.10103091443460.26053-100000@sun1.lrz-muenchen.de>

I'm looking for components which I could use in my (commercial)
application. Does anyone know of a (preferably plain Python) module
which does most or parts of what the Tk Tree Widget can?

	http://archive.eso.org/~abrighto/tree/tree.html

Lacking that, is there an easy way to use above Widget (in an
platform-inbdependent way, pretty please ;)

TIA,

Eugene




From tjg at exceptionalminds.com  Mon Mar 19 15:31:23 2001
From: tjg at exceptionalminds.com (Timothy Grant)
Date: Mon, 19 Mar 2001 12:31:23 -0800
Subject: pwd behaviour when new users are added
Message-ID: <20010319123123.I3109@trufflehunter.avalongroup.net>

I just ran into an interesting feature of the pwd module

say I have users fred, ethel, ricky and lucy on my system and I
issue a 'pwd.pwgetall()' this returns a tuple with user
information. Now, while my programme is running I issue a
'useradd gilligan'

if my programme now issues another 'pwd.pwgetall()' or a
'pwd.pwgetnam()' or 'pwd.pwgetuid()' the new user name is not
available.

if I 'reload(pwd)' it still doesn't work if I 'del pwd' and
then 'import pwd' it doesn't work.

What am I doing wrong?

BTW: I have posted here a couple of times on shadow password
issues, and the outcrop of those posts is that I have written
three replacement functions for the pwget*() that include
shadow information in the tuple returned. It works quite well.

-- 
Stand Fast,
    tjg.

Timothy Grant                         tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.         <><       (503) 246-3630
>>>>>>>>>>>>>Linux, because rebooting is *NOT* normal<<<<<<<<<
>>>>This machine was last rebooted:  62 days     4 hours ago<<



From sheila at spamcop.net  Sun Mar 18 04:53:45 2001
From: sheila at spamcop.net (Sheila King)
Date: Sun, 18 Mar 2001 09:53:45 GMT
Subject: Help with 500 error on cgi script
References: <quv8bto7r8s6f1tckc7qpmuir0kvufp1nb@4ax.com>
Message-ID: <lb19bt4m1ifkbal204oi8b4t08gq05pqqc@4ax.com>

On Sun, 18 Mar 2001 09:33:57 GMT, Sheila King <sheila at spamcop.net> wrote in
comp.lang.python in article <quv8bto7r8s6f1tckc7qpmuir0kvufp1nb at 4ax.com>:

:So, here is my script. I'm sure someone here will see it right off?


Additionally, here is the error log's last entry:

[Sun Mar 18 01:35:53 2001] [error] [client x.x.xxx.xxx] Premature end of
script headers: e:/apache/cgi-bin/testform.py
[Sun Mar 18 01:35:53 2001] [error] [client x.x.xxx.xxx]   File
"e:/apache/cgi-bin/testform.py", line 30

[Sun Mar 18 01:35:53 2001] [error] [client x.x.xxx.xxx]     print k, " : ",
form[k]

[Sun Mar 18 01:35:53 2001] [error] [client x.x.xxx.xxx]         ^

[Sun Mar 18 01:35:53 2001] [error] [client x.x.xxx.xxx] IndentationError:
expected an indented block

--
Sheila King
http://www.thinkspot.net/sheila/
http://www.k12groups.org/



From catunda at pobox.com  Sun Mar 11 22:51:46 2001
From: catunda at pobox.com (Marco Catunda)
Date: Mon, 12 Mar 2001 00:51:46 -0300
Subject: popen problem
Message-ID: <98hgtt$5627@NWNEWS.PCT.EDU>

What's wrong with this code?

##########################
import threading
import os

class TestPopen( threading.Thread ):

   def run( self ):
      while 1:
         f = os.popen( "ls -l /" )
         lines = f.read()
         f.close()

if __name__ == "__main__":
   for i in range(50):
      t = TestPopen()
      t.start()
###########################

Let it run about 1 minute.
And, the result is:

Exception in thread Thread-10:
Traceback (innermost last):
  File "/var/tmp/python/python-root/usr/lib/python1.5/threading.py", line
376, in __bootstrap
    self.run()
  File "teste.py", line 10, in run
    f.close()
IOError: (0, 'Error')

Exception in thread Thread-11:
Traceback (innermost last):
  File "/var/tmp/python/python-root/usr/lib/python1.5/threading.py", line
376, in __bootstrap
    self.run()
  File "teste.py", line 10, in run
    f.close()
IOError: (0, 'Error')

....



Please, how do I correct it?
I used the same idea in another more complex code and I've had the same
problem.

-- Marco Catunda










From mwh21 at cam.ac.uk  Fri Mar  9 12:48:43 2001
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 09 Mar 2001 17:48:43 +0000
Subject: I don't get it (Extending C/C++)
References: <98b3cf$apu$02$1@news.t-online.com>
Message-ID: <m38zmeal50.fsf@atrus.jesus.cam.ac.uk>

"Moritz Voss" <thygrrr at gmx.net> writes:

> I'm somewhat stumped by all the weird function names in the Tutorial
> of the 2.0 Manual...

Do you really mean the tutorial, or the extending & embeddng manual?

> Is something like this possible:
> I call a Python function from my C/C++ program. I pass it two
> pointers, one to an array with data, and one to an empty space where
> it can put its results....

Yes, certainly.  Well, actually, not *quite* as you describe.  You
have to massage the data into a form Python can cope with, e.g. a list
of floats, using the Python C API calls and get a pointer to the
Python function, build the arguments you want to pass into the
function into a tuple, and call PyEval_CallObject.  You then get a
PyObject* back as a result, or NULL if an exception was thrown.

> I don't even remotely see a way how to do this in the Tutorial, it's
> just full of (to me) irrelevant stuff about callbacks etc...I feel
> stupid here, folks ;-)

This is one of those things that is really quite easy, once you know
it.  I agree learning how to do this stuff can be a bit intimidating.
If you have the time I recommend d/l-ing the python source and
spending an hour or two learning a sort of rough outline of how
CPython works (and how to write clean, readable C!).  After all,
CPython has to do these things that you're trying to do, e.g. calling
functions, handling errors.

I've been thinking about writing a Python-Internals-HOWTO describing
this sort of thing, but it's *hard*; everything depends on everything
else, which makes for poor pedagogy.

> Also, is it advisable to invoke the interpreter (do I even do that?
> what DO I do when I execute a python function?) several hundred
> times? (running about 500 instances of several, sometimes identical
> python scripts times would be likely for medium load of my
> application).

You call PyInitialize and PyFinalize (check spellings!) once each.
Then you can call the interprater through the PyRun_* or PyEval_* APIs
as much as you like (the interpreter itself uses the Py_Eval_* API).

> I'm sorry, I just don't /understand/ the Python concept...yet.

Hang in there.  We're here to help!

> It looks all so cool, but I don't have a clue about how to use it.
> (I plan to use it as a scripting language to customize application
> component behaviour)

Cool.

Do you have any more specific problems?  "Argh, it doesn't make any
sense!" is not the easiest problem to answer!

Cheers,
M.

-- 
  I hate leaving Windows95 boxes publically accessible, so shifting
  even to NT is a blessing in some ways.  At least I can reboot them
  remotely in a sane manner, rather than having to send them malformed
  packets.      -- http://bofhcam.org/journal/journal.html, 20/06/2000


From aahz at panix.com  Thu Mar 22 11:19:36 2001
From: aahz at panix.com (Aahz Maruch)
Date: 22 Mar 2001 08:19:36 -0800
Subject: New PEP: The directive statement
References: <mailman.985085006.24236.python-list@python.org> <slrn9bk7fk.fnm.scarblac@pino.selwerd.nl>
Message-ID: <99d8mo$lta$1@panix6.panix.com>

In article <slrn9bk7fk.fnm.scarblac at pino.selwerd.nl>,
Remco Gerlich <scarblac at pino.selwerd.nl> wrote:
>
>I could go on, but noone reads this far anyway, 

That's true.
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

"I won't accept a model of the universe in which free will, omniscient
gods, and atheism are simultaneously true."  -- M


From aleaxit at yahoo.com  Sun Mar 18 15:15:11 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sun, 18 Mar 2001 21:15:11 +0100
Subject: Who's minister of propaganda this week?
References: <B98041329020E007.7DA84CAD9D0562AB.817BBC572F480869@lp.airnews.net> <98nu1m$12e@dispatch.concentric.net> <660ACCA1E4816F7E.C359CDF4168EF3FB.3B6B05538D619770@lp.airnews.net> <3AAFA13D.76E32C3F@divalsim.it> <98oa4o01i4v@news2.newsguy.com> <mailman.984619039.7856.python-list@python.org> <98q9h60122v@news2.newsguy.com> <3AB4F520.40104@yahoo.com> <mailman.984939085.7872.python-list@python.org>
Message-ID: <99351n089f@news2.newsguy.com>

"Paul Prescod" <paulp at ActiveState.com> wrote in message
news:mailman.984939085.7872.python-list at python.org...
> Michael Chermside wrote:
> >
> >...
> >
> > But I'm quite surprised that you think you pay a 10% penalty
> > for going through contortions to adhere to static typechecking.
> > You see... in my own experience, I hardly ever want to violate
> > it. I generally design type hierarchys and use objects in places
> > where I expect an object of that type or of a direct ancestor.
>
> One issue is that you often do not have control over the design of type
> hierarchies. You need to glue together two or more large frameworks and
> you have no opportunity to set up interfaces etc. as you want.

Good point.  You work around _that_ one with templates, delegation,
multiple inheritance, etc, but the productivity impact easily exceeds
10% for this specific problem, in my experience.


> Another issue is that there are certain kinds of advanced programming
> that are are exceedingly difficult in statically typed languages. For

Right again.  Besides the specific example you give, I think a very
instructive experience is comparing the base version of "Design
Pattern" (C++ oriented) with the "Design Pattern Smalltalk Companion"
book; despite only having single inheritance in Smalltalk, you will
notice that in at least about one case in three (and we're talking
specifically of pattterns designed for a statically typed language!)
the Smalltalk version is slimmer and handier (a Python one, I think,
might easily show another small advantage yet).

Yet another example is the "Acyclic Visitor" pattern (by R. Martin)
which I already mentioned recently -- it specifically uses the dynamic
typing "escape" of C++ (dynamic_cast<>) to greatly enhance the
dependency-control characteristics of the "Visitor" pattern (removing
the cycle of dependencies that make it such a bother).  Imagine
"just doing it" rather than having to set everything up so carefully
and delicately each and every time static-typing imposes such costs
on your designs (e.g., nasty dependency graphs) -- indeed, not
having to worry about it _at all_ (which leads to easy consideration
of many more architectures, with increased likelihood of finding the
optimal one) -- and I think you'll see that my claim of a 10% overall
productivity gain for dynamic typing over static is rather modest.


Alex





From rwklee at home.com  Sun Mar 25 22:08:43 2001
From: rwklee at home.com (Rick Lee)
Date: Mon, 26 Mar 2001 03:08:43 GMT
Subject: static methods
References: <mailman.985474834.2256.python-list@python.org> <3ABD7673.9D447A32@home.com> <99l7mu01ed6@news1.newsguy.com>
Message-ID: <3ABEB2B9.C2F38680@home.com>


Alex Martelli wrote:

>
> The desire to use the class-object specifically, rather than any
> other, typically seems to have more to do with familiarity with
> other languages where classes must play certain special roles
> (C++, Java, Smalltalk, Eiffel) than with any actual programming
> needs.
>
> Alex

I beg to differ.  If you pass to the caller a class reference, you can change
it to a subclass and it is all transparent to the caller.  I don't think you
have that kind of flexibility with a module reference.



From abrahams at mediaone.net  Sat Mar 17 22:58:45 2001
From: abrahams at mediaone.net (David Abrahams)
Date: Sun, 18 Mar 2001 03:58:45 GMT
Subject: Meta-class inheritance problem (Python2.0 bug?)
References: <6Kzs6.35062$v5.73895@newsfeeds.bigpond.com>
Message-ID: <VnWs6.17199$3x.1641111@typhoon.ne.mediaone.net>

"Lorien Dunn" <loriendNOSPAM at bigpond.com> wrote in message
news:6Kzs6.35062$v5.73895 at newsfeeds.bigpond.com...
> Hullo all,
>         I've been using boost::python for a little while now- long enough
to have
> seemingly successfully wrapped my extension, and I've run into a bug in
> python 2.0. The real code that the following pseudo code represents throws
> a TypeError.
>
> class MyDerivedClass(MyPurePythonClass, MyBoostClass):
>         def __init__(self):
>                 MyPurePythonClass.__init__(self)
>                 MyBoostClass.__init__(self)
>
> The problem occurs in MyPurePythonClass- Python says "unbound method must
> be called with class instance 1st argument". This makes sense: I'm passing
> a meta-class instead of a class. David Abrahams (the main author of
> boost::python) agrees that this is a Python problem, and that it should
> affect Zope as well.

Just to be absolutely clear, I never said it was a bug in Python. I'm sure
Python doesn't purport to support this feature, and as such I would never
claim it was a bug. However, it does seem place some serious -and avoidable-
limitations on the use of Python class methods on extension class instances.

> He also said that it should be an easy modification, but he doesn't know
> what ramifications it would have throughout the rest of the python
> interpreter.

It isn't completely obvious to me why the check is there at all. Removing it
would be easy, though that may not be the best way to address the problem.

I guess there's always MyBoostClass.__dict__["__init__"](self)
 :-(

-Dave




From jeremy at alum.mit.edu  Sat Mar  3 18:32:32 2001
From: jeremy at alum.mit.edu (Jeremy Hylton)
Date: Sat, 3 Mar 2001 18:32:32 -0500 (EST)
Subject: [Python-Dev] Re: Bug fix releases
In-Reply-To: <20010305011655.V9678@xs4all.nl>
References: <LNBBLJKPBEHFEDALKOLCMEEPJBAA.tim.one@home.com>
	<mailman.982897324.9109.python-list@python.org>
	<200103031821.NAA24060@panix3.panix.com>
	<200103031910.OAA21663@cj20424-a.reston1.va.home.com>
	<20010305011655.V9678@xs4all.nl>
Message-ID: <15009.32528.29406.232901@w221.z064000254.bwi-md.dsl.cnc.net>

>>>>> "TW" == Thomas Wouters <thomas at xs4all.net> writes:

  [GvR:]
  >> So that relegates us at PythonLabs to a number of things: coding
  >> new modules (boring), or trying to improve performance of the
  >> virtual machine (equally boring, and difficult to boot), or
  >> fixing bugs (did I mention boring? :-).

  TW> How can you say this ? Okay, so *fixing* bugs isn't terribly
  TW> exciting, but hunting them down is one of the best sports
  TW> around. Same for optimizations: rewriting the code might be
  TW> boring (though if you are a fast typist, it usually doesn't take
  TW> long enough to get boring :) but thinking them up is the fun
  TW> part.

  TW> But who said PythonLabs had to do all the work ? You guys didn't
  TW> do all the work in 2.0->2.1, did you ? Okay, so most of the
  TW> major features are written by PythonLabs, and most of the
  TW> decisions are made there, but there's no real reason for
  TW> it.

Most of the work I did for Python 2.0 was fixing bugs.  It was a lot
of fairly tedious but necessary work.  I have always imagined that
this was work that most people wouldn't do unless they were paid to do
it.  (python-dev seems to have a fair number of exceptions, though.)

Working on major new features has a lot more flash, so I imagine that
volunteers would be more inclined to help.  Neil's work on GC or yours
on augmented assignment are examples.

There's nothing that says we have to do all the work.  In fact, I
imagine we'll continue to collectively spend a lot of time on
maintenance issues.  We get paid to do it, and we get to hack on Zope
and ZODB the rest of the time, which is also a lot of fun.

Jeremy



From tim.one at home.com  Tue Mar 13 05:15:24 2001
From: tim.one at home.com (Tim Peters)
Date: Tue, 13 Mar 2001 05:15:24 -0500
Subject: This math scares me
In-Reply-To: <984475826.262977@newsmaster-04.atnet.at>
Message-ID: <LNBBLJKPBEHFEDALKOLCEEEPJFAA.tim.one@home.com>

[Werner Schiendl]
> I wonder why 17 digits are used to represent the numbers...
> Think I can remember the precision of double was 15 to 16 decimal digits
> (depends on the value)
>
> Did I miss something?

It's so eval(repr(x))==x on 754-conforming platforms.  Try printing random
754 doubles under %.16g instead, then read them back in, then count how many
times you don't get back the number you printed.  You may be surprised!




From mwh21 at cam.ac.uk  Mon Mar 19 14:43:55 2001
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 19 Mar 2001 19:43:55 +0000
Subject: Python style questions
References: <98tb91$9t9$1@saltmine.radix.net>
Message-ID: <m34rwpilxg.fsf@atrus.jesus.cam.ac.uk>

cobrien at Radix.Net (Cary O'Brien) writes:

> 2. One thing I've done is to define classes used only by another class
>    inside that class.  For example, if a class A needs to keep track of a
>    bunch of "things", then the class definition for the "things" is inside
>    of the class definition of A.  This is
> 
>    a. SOP
>    b. Brilliant
>    c. Idiotic

The scoping can get a bit counter-intuitive
> 3. I really miss not having a "switch" or "case" statement.  Sniff.

-- 
3. Syntactic sugar causes cancer of the semicolon.
  -- Alan Perlis, http://www.cs.yale.edu/homes/perlis-alan/quotes.html


From loewis at informatik.hu-berlin.de  Tue Mar 20 05:42:34 2001
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: Tue, 20 Mar 2001 11:42:34 +0100 (MET)
Subject: New PEP: The directive statement
Message-ID: <200103201042.LAA20248@pandora.informatik.hu-berlin.de>

Please have a look at the PEP below. It is intended to replace the
__future__ import suggested in PEP 236.

Regards,
Martin

PEP: XXX
Title: The directive statement
Version: $Revision$
Author: Martin v. L?wis
Python-Version: 2.1
Status: Active
Type: Standards Track
Created: 20-Mar-2001


Motivation

    From time to time, Python makes an incompatible change to the
    advertised semantics of core language constructs, or changes their
    accidental (implementation-dependent) behavior in some way.  While this
    is never done capriciously, and is always done with the aim of
    improving the language over the long term, over the short term it's
    contentious and disrupting.

    The "Guidelines for Language Evolution" PEP [1] suggests ways to ease
    the pain, and this PEP introduces some machinery in support of that.

    The "Statically Nested Scopes" PEP [2] is the first application, and
    will be used as an example here.

    When a new, potentially incompatible language feature is added,
    some modules and libraries may chose to use it, while others may
    not. This specification introduces a syntax where a module author
    can denote whether a certain language feature shall is used in the
    module or not.

    In discussion of this PEP, readers commented that there are two
    kinds of "settable" language features:

    a) those that are designed to eventually become the only option,
       at which time specifying use of them is not necessary anymore.
       The features for which the syntax of the "Back to the
       __future__" PEP [3] was proposed fall into this category.  This
       PEP supports declaring such features, and supports phasing out
       the "old" meaning of constructs whose semantics has changed
       under the new feature. However, it defines no policy as to what
       features must be phased out eventually.

    b) those which are designed to stay optional forever, e.g. if they
       change some default setting in the interpreter. An example for
       such settings might be the request to always emit line-number
       instructions for a certain module; no specific flags of that
       kind are proposed in this specification.

    Since a primary goal of this PEP is to support new language
    constructs without immediately breaking old libraries, special
    care was taken not to break old libraries by introducing the new
    syntax.

Syntax

    A directive_statement is a statement of the form

        directive_statement: 'directive' NAME [atom] [';'] NEWLINE

    The name in the directive indicates the kind of the directive; it
    defines whether the optional atom can be present, and whether
    there are further syntactical or semantical restrictions to the
    atom. In addition, depending on the name of the directive, certain
    additional syntactical or semantical restrictions may be placed on
    the directive (e.g. placement of the directive in the module may be
    restricted to the top of the module).

    In the directive_statement, 'directive' is a new
    keyword. According to [1], this keyword is initially considered as
    a keyword only when used in a directive statement, see "Backwards
    Compatibility" below.

Semantics

    A directive statement instructs the Python interpreter to process
    a source file in a different way; the specific details of that
    processing depend on the directive name. The optional atom is
    typically interpreted when the source code is processed; details
    of that interpretation again depend on the directive.

Specific Directives: transitional

    If a syntactical or semantical change is added to Python which is
    incompatible, [1] mandates a transitional evolution of the
    language, where the new feature is initially available alongside
    with the old one. Such a transition is possible by means of the
    transitional directive.

    In a transitional directive, the NAME is 'transitional'. The atom
    MUST be present, and it MUST be a NAME. The possible values for
    that name are defined when the language change is defined. One
    example for such a directive is

    directive transitional nested_scopes

    The transitional directive MUST occur at before any other
    statement in a module, except for the documentation string
    (i.e. it may appear as the second statement of a module only if
    the first statement is a STRING+).

Backwards Compatibility

    Introducing 'directive' as a new keyword might cause
    incompatibilities with existing code. Following the guideline in
    [1], in the initial implementation of this specification,
    directive is a new keyword only if it was used in a valid
    directive_statement (i.e. if it appeared as the first non-string
    token in a module).

Unresolved Problems:  directive as the first identifier

    Using directive in a module as

    directive = 1

    (i.e. the name directive appears as the first thing in a module)
    will treat it as keyword, not as identifier. It would be possible
    to classify it as a NAME with an additional look-ahead token, but
    such look-ahead is not available in the Python tokenizer.


Questions and Answers

    Q:  It looks like this PEP was written to allow definition of source
        code charactersets. Is that true?

    A:  No. Even though the directive facility can be extended to allow
        source code encodings, no specific directive is proposed.

    Q:  Then why was this PEP written at all?

    A:  It acts as a counter-proposal to [3], which proposes to
        overload the import statement with a new meaning. This PEP
        allows to solve the problem in a more general way.

    Q:  But isn't mixing source encodings and language changes like
        mixing apples and oranges?

    A:  Perhaps. To address the difference, the predefined
        "transitional" directive has been defined.

Copyright

    This document has been placed in the public domain.


References and Footnotes

    [1] http://python.sourceforge.net/peps/pep-0005.html

    [2] http://python.sourceforge.net/peps/pep-0227.html

    [3] http://python.sourceforge.net/peps/pep-0236.html



Local Variables:
mode: indented-text
indent-tabs-mode: nil
End:



From ejr at cs.berkeley.edu  Tue Mar 13 17:10:08 2001
From: ejr at cs.berkeley.edu (Edward Jason Riedy)
Date: 13 Mar 2001 22:10:08 GMT
Subject: This math scares me
References: <3aad4062.1290021@News.CIS.DFN.DE> <3aad4485.2348499@News.CIS.DFN.DE> <jwbnews-7CC03F.19050312032001@corp.supernews.com> <yu993dcimjt8.fsf@europa.research.att.com>
Message-ID: <98m5s0$gog$1@agate.berkeley.edu>

And Andrew Koenig writes:
 - 
 - Briefly, IEEE requires that input and output conversion have no more
 - than 0.47 LSB error; Scheme requires that input conversion be exact
 - (that is, that the internal representation always be the closest
 - (according to the rounding rules currently in effect) internal number
 - to the infinite-precision value read, and that output yield the
 - smallest number of significant digits necessary for idempotence).

The next revision of IEEE 754 will require exact conversion, as
well.  However, note that there is no concept of a language
complying with IEEE 754, or even a library...  It's just a vague
`system.'

Jason
-- 


From tjreedy at udel.edu  Sun Mar 18 22:16:31 2001
From: tjreedy at udel.edu (Terry Reedy)
Date: Sun, 18 Mar 2001 22:16:31 -0500
Subject: Deleting dictionary items
References: <98ug6n$q9d$1@news6.svr.pol.co.uk> <98vs4p$k9e$1@slb7.atl.mindspring.net> <993570$71$1@newsg2.svr.pol.co.uk>
Message-ID: <993tnu$9fh$1@news.udel.edu>

> This makes me think, would it be a sane way of doing things to use a list
> comprehension to modify a dictionary?
...
> filter(dict, lambda key: key[attr] == value)
>
> Is this possible? Desireable? Too stupid to contemplate?

Funny coincidence.  Just 3 miniute ago, before reading this, I tossed out a
similar idea, in the thread Subsetting a Dictionary, of adding 'dict
comprehensions' with syntax directly paralleling list comprehensions.

{k,v for k,v in dict if condition(k,v)}

Lists and dicts are subcategories of collection or set.  Subseting is, by
axiom, a general set operation not limited to ordered sets (lists).  So
this is hardly a stupid idea.

Terry J. Reedy





From nas at arctrix.com  Mon Mar 19 23:17:29 2001
From: nas at arctrix.com (Neil Schemenauer)
Date: Mon, 19 Mar 2001 20:17:29 -0800
Subject: Why "from __future__" stinks; a counter-offer
In-Reply-To: <9969ta$2is$1@newsg3.svr.pol.co.uk>; from will@nospam.misconception.org.uk on Tue, Mar 20, 2001 at 12:59:56AM +0000
References: <mailman.985029808.7088.python-list@python.org> <memo.20010319213204.8771C@digitig.compulink.co.uk> <9965t7$s57$1@panix6.panix.com> <9969ta$2is$1@newsg3.svr.pol.co.uk>
Message-ID: <20010319201729.B20507@glacier.fnational.com>

On Tue, Mar 20, 2001 at 12:59:56AM +0000, Will Newton wrote:
> How often is this "from __future__" expected to be used?
> 
> Just experimental code written for bleeding edge versions of Python or a 
> significant portion of code (that will eventually become legacy code)?

Argh!  Read the PEP people.  Let me try to explain.  First
scenario, you don't need nested scopes and you don't think you
have any code that breaks with them:

    1. You upgrade to 2.1.

    2. You upgrade to 2.2.

Life is good and no "ugly" __future__ statements have entered the
picture.  Scenario 2, you want to use nested scopes in 2.1:
    
    1. You upgrade to 2.1.

    2. In modules where you want to use nested scopes you use
       the __future__ mechanism to enable them.

    3. You upgrade to 2.2.

    4. You use the handy tools provided to remove all traces of
       "from __future__ import nested_scopes"

Now you have 2.2, you are using nested scopes and there are still
no __future__ statements around.  Scenario 3, you have code that
will break with nested scopes:

    1. You upgrade to 2.1.

    2. Using the nice warnings provided you fix code that
       will break with 2.2.

    3. You upgrade to 2.2.

Again, no __future__ statements.  Please read the PEP and
understand what problem is being solved.  Dispite Tim's careful
explainations people still seem to have the idea that __future__
is a general switch for enabling language features.

  Neil



From db3l at fitlinxx.com  Thu Mar  1 20:16:26 2001
From: db3l at fitlinxx.com (David Bolen)
Date: 01 Mar 2001 20:16:26 -0500
Subject: Newbie: Casting Int to string....
References: <3a9ed412$1@news1> <3a9edb8a.95346080@news.uswest.net>
Message-ID: <ug0gxm0lx.fsf@ctwd0143.fitlinxx.com>

joconnor at cybermesa.com writes:

> Yes, it is :)
> 
> You can  do either....
> 
> 	nntp.stat (str(aNumber))
> 
> or teh shorter
> 	nntp.stat (`aNumber`)
> 
> Notice that the single quotes are backwards

Note that for an single solitary number, this will yield the same
results, but under the cover, str() is using __str__ (or the
equivalent for built-ins) whereas backquotes (``) are a short form of
repr() which uses __repr__.  So while the two the methods will both
yield a string, it may not be the same string value depending on the
objects involved.

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


From p.magwene at snet.net  Thu Mar  1 22:37:03 2001
From: p.magwene at snet.net (Paul Magwene)
Date: Thu, 1 Mar 2001 22:37:03 -0500
Subject: PythonJournal.com becomes Interactive, new issue today
References: <mailman.983500743.14313.python-list@python.org>
Message-ID: <hzEn6.1869$GD.875814@typhoon.snet.net>

"editor at pythonjournal.com" <cognite at zianet.com> wrote in message
>
> NB: Incorporating new ***interactive feature sections***, it
> shows off (and tests) the python programming language's web application,
> scripting, prototyping and graphics capabilities. Focus of all parts of

Ouch, to be perfectly honest, if the above website was my first exposure to
Python's web application capabilities I'd run for the hills.  The layout of
the site is AWFUL.

Sorry to be so negative, but that is NOT the face I think the Python
community would hope to present to the world.

--Paul




From loewis at informatik.hu-berlin.de  Sat Mar 24 05:48:53 2001
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 24 Mar 2001 11:48:53 +0100
Subject: New PEP: The directive statement
References: <mailman.985418312.27803.python-list@python.org>
Message-ID: <j43dc3o31m.fsf@informatik.hu-berlin.de>

"Tim Peters" <tim.one at home.com> writes:

> True, but I've written many harder analyzers + translators for Python, and
> several have been in the distribution for years (e.g., tabnanny.py,
> checkappend.py, most recently reindent.py for 2.1).  So this FUD won't fly.

I wasn't questioning that you *can* write such a tool - I was asking
whether it should work in the way you've been outlining.

> Ah, note that I'm not signing up for writing a general *directive*
> manipulation tool.  That's too open-ended for me.  The tool PEP 236
> hypothesizes is limited to nuking obsolete future stmts, and PEP 236
> is not an open-ended scheme.  The tool is easy because the PEP is
> limited.

My point is that even though you can nuke all obsolete future
statements, it might not be desirable to do so - since the module then
become silently incorrect in an older release, instead of outright
failing.

> I put the assert near the line it's complaining about, because the line it's
> complaining about is the line I want to get rid of someday, and I want the
> assert to tell me that.  Maybe a wordier version would be clearer:
> 
>     assert nested_scopes.getMandatoryRelease() > sys.version_info, \
>          "no more need for the nested_scopes stmt above -- nuke it"

I see. Sounds like a good application of the versioning.

> >> In that case I may very well have an acute need to ship one version
> >> of my source that exploits the new semantics but falls back to an
> >> older & slower module if the compiler it's running under can't play
> >> along.
> 
> > People routinely use sys.*version* for that; I'm really curious
> > whether they'd prefer to investigate other variables instead.
> 
> I think PEP 236 is extremely clear that the release info in __future__.py is
> in sys.version_info format

That is absolutely clear. I was asking you to predict the future: Do
you think that people will prefer to write

  if __future__.nested_scopes.getOptionalRelease() >= sys.version_info:

instead of

  if sys.version_info < (2,1):

> How could it be clearer?  Nobody complained; Barry pointed out that
> he generally uses sys.hexversion (not sys.version) instead, but
> allowed that sys.version_info is easier to read; and I can't imagine
> that anyone in their right mind uses sys.version for stuff like this

I have really no problem with the format of the _Feature members; I
very much like using tuples for versioning, since it is quite elegant.
My question is whether you think that people will use those computed
version numbers instead of hard-coded ones, when comparing to
sys.version_info.

> have-yet-to-hear-a-complaint-from-anyone-using-it<0.6-wink>-ly y'rs  - tim

*Now* you get my question :-)

Martin


From m.hadfield at niwa.cri.nz  Thu Mar  1 16:01:28 2001
From: m.hadfield at niwa.cri.nz (Mark Hadfield)
Date: Fri, 2 Mar 2001 10:01:28 +1300
Subject: Performance when working directory is slow network drive
References: <983411179.898268@clam-ext> <mailman.983425921.14373.python-list@python.org>
Message-ID: <983480535.813990@clam-ext>

"Sean Reifschneider" <jafo at tummy.com> wrote in message
news:mailman.983425921.14373.python-list at python.org...
> On Thu, Mar 01, 2001 at 02:45:25PM +1300, Mark Hadfield wrote:
> >If the working directory is on one of my hard drives the elapsed times is
> >approx 0.1 seconds; if I then cd to a slow network drive (a Samba share
to a
> >Unix machine that mounts another machine's disks via NFS!) it takes 2
> >seconds.
>
> Have you tried removing the current directory from sys.path?  That'd be
> the one that's empty.

Yes, that's it! Thank you.

Now only two things puzzle me:

1. Why having the current directory in the search path affects the import
time for the os module much more than it does for other modules I tested
(like string). Is this because os is a package?

2. Why the number of empty-string entries at the head of sys.path varies
between 0 and 2 on my system depending on how the interpreter is invoked and
what I ate for breakfast (or something).

--
Mark Hadfield
m.hadfield at niwa.cri.nz  http://katipo.niwa.cri.nz/~hadfield
National Institute for Water and Atmospheric Research




From jimd at vega.starshine.org  Mon Mar 12 10:32:05 2001
From: jimd at vega.starshine.org (Jim Dennis)
Date: 12 Mar 2001 15:32:05 GMT
Subject: Color curses and Python 2?
References: <slrn9amlnn.ktp.jimd@vega.starshine.org> <slrn9anibn.i3.amk@mira.erols.com>
Message-ID: <slrn9apqvc.ktp.jimd@vega.starshine.org>

In article <slrn9anibn.i3.amk at mira.erols.com>, A.M. Kuchling wrote:
>
>On 11 Mar 2001 10:44:09 GMT, Jim Dennis <jimd at vega.starshine.org> wrote:
>> I thought I'd heard that color ncurses support would be included
>> in 1.6.  Was it?  I still haven't seen a 1.6.x package

>Yes, it was.  Perhaps the Debian version of Python is miscompiled
>(unlikely), or perhaps your terminal definition is set to something
>that doesn't support colour.

>--amk

 I can categorically guarantee that my terminal/console supports
 common ncurses color (since I use it with vim, slrn, w3m, lynx,
 xemacs, dialog, and the Perl Curses.pm module among others).

 Heck I can even use color in shell using tput!  (Please don't
 suggest that I use external calls to tput in my Python code;
 that's not the question here).

 As I said, I haven't seen a 1.6.x package for Debian.  I gather
 that the 1.5.x versions don't have ncurses/color support (I 
 compiled my own for that).  Part of the question at hand is whether
 the 2.x betas were supposed to have color support.  

 If so, where do I report this as a bug in the betas.  If not,
 why not (or what changes were made to the programming interfaces;
 is there some other module? are there different methods and 
 attributes? etc).



From cce at clarkevans.com  Sat Mar 24 18:07:32 2001
From: cce at clarkevans.com (Clark C. Evans)
Date: Sat, 24 Mar 2001 18:07:32 -0500 (EST)
Subject: static methods
In-Reply-To: <slrn9bpcb0.bm7.qrczak@qrnik.zagroda>
Message-ID: <Pine.LNX.4.21.0103241758270.3747-100000@clarkevans.com>

On 24 Mar 2001, Marcin 'Qrczak' Kowalczyk wrote:
| Fri, 23 Mar 2001 14:27:42 -0800, Paul Prescod <paulp at ActiveState.com>:
| | Probably because they are simply not neccessary. 
| | 
| | def _static():
| | 	XXX.foo=bar
| | 	return XXX.foo
| | 
| | class Class:
| | 	staticmethod = _static
| | 	...
| 
| It doesn't work. Try to call it.
| 
| The fact that you are not the first person who thinks it works
| suggests that the current Python's solution is not ideal :-)

>>> class Class:
        class StaticFudge:
            def __call__(self): 
                print "I am a static"
        staticmethod = StaticFudge()
>>> Class.staticmethod()
I am static

    It took much looking and many wrong answers before 
    Alex Martelli pointed me to this solution.  

| | This proposal asks that None be allowed when calling a
| | method using the class syntax, e.g.,  the following
| | syntax would be allowed, MyClass.method(None,...).
| 
| I would be more radical and allow anything as the first argument.
| In other words, don't touch function objects while retrieving them
| from a class. Instead, convert them to bound methods when retrieving
| them from an instance.

This was my initial position
  
| The only problem seems that some errors will be poorly detected.

This is why I reduced it to only allowing "None".

;) Clark




From aahz at panix.com  Sat Mar 24 10:46:38 2001
From: aahz at panix.com (Aahz Maruch)
Date: 24 Mar 2001 07:46:38 -0800
Subject: PEP 1, PEP Purpose and Guidelines
References: <mailman.985196613.7994.python-list@python.org>
Message-ID: <99ifgu$jei$1@panix3.panix.com>

[posted & e-mailed]

In article <mailman.985196613.7994.python-list at python.org>,
Barry A. Warsaw <barry at digicool.com> wrote:
>
>Note the addition of the Replaces: and Replaced-By: headers for
>formalizing the PEP replacement policy (thanks to Andrew Kuchling for
>the idea and patch).

After taking a closer look at Moshe's numeric proposals, I suggest adding
a Depends-On: header.  (E.g. PEP 240 depends on PEP 239.)
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

"I won't accept a model of the universe in which free will, omniscient
gods, and atheism are simultaneously true."  -- M


From owen at astrono.junkwashington.emu  Tue Mar 20 13:23:36 2001
From: owen at astrono.junkwashington.emu (Russell E. Owen)
Date: Tue, 20 Mar 2001 10:23:36 -0800
Subject: Wrong reasons to enhancing the language
References: <mailman.984971990.29857.python-list@python.org> <9960i1$tp0$1@nntp6.u.washington.edu> <99661m$scc$1@panix6.panix.com>
Message-ID: <99877b$uss$1@nntp6.u.washington.edu>

>>I believe many of the improvements in Python 2.0 fall into the category 
>>of helpful improvements. I am more doubtful about some of the stuff 
>>coming up (including nested scopes, // and rational numbers), but we'll 
>>see how it works out.
>
>None of those are Python 2.0 improvements.  Nested scopes are an
>optional feature of Python 2.1 and the others are simply *proposed*
>features.  This is a *great* time to be debating the numerics issues.

Darn, here I thought I was being so clear, but of course I always know 
what I *mean* to say. Let me try again...

I really, really like the enhancements in Python 2.0. Hence: "I believe 
many of the improvements in Python 2.0 fall into the category of helpful 
improvements." That was a bit understated, perhaps. I think 2.0 was a 
total win. Great stuff. Much appreciated.

However, I am dubious about some of the improvements scheduled for 2.1 
and 2.2, hence: "I am more doubtful about some of the stuff coming up 
(including nested scopes, // and rational numbers), but we'll see how it 
works out.

As to the numerics, I'm not upset about the changes, just a bit nervous.
- Rational numbers will be used by default in certain situations (such 
as the results of division). I can see where rational numbers and 
infinite precision are very nice to have, but I'm not so sure I want 
them being used by default in normal calculations. Maybe that's not 
rational (sorry). Perhaps my main concern is speed.

- It's hard for me to get enthused about "//". I do see the 
justification; I agree that 1/3 = 0 is a non-obvious thing and a trap 
for non-c mavens. But I'm inclined to live with it rather than add 
another operator. "//" looks confusing to me (though at least it's a 
flavor of division, not something totally unrelated such as a comment). 
I'm not going to jump up and down about it either way. Maybe Python 3000 
can have "/" be a float operator!?

I've posted a lot about nested scopes. Summary: mediocre implementation 
with funny exceptions and too many chances to shoot self in foot. I'd 
rather simplify it or dump it.

-- Russell


From dnew at san.rr.com  Fri Mar 23 14:56:48 2001
From: dnew at san.rr.com (Darren New)
Date: Fri, 23 Mar 2001 19:56:48 GMT
Subject: Good Editor? (What's the big deal with syntax colouring?)
References: <mailman.985291829.27633.python-list@python.org> <99dqqo$gpi$1@news.mathworks.com> <99e9sv$usmp$1@ID-11957.news.dfncis.de> <99emg2$qif@gap.cco.caltech.edu>
Message-ID: <3ABBAA81.FD344D5E@san.rr.com>

Nathaniel Gray wrote:
> My coworkers cringe when they see me coding.

I had a color-blind coworker who used green on hot-pink and dark red on cyan
as xterm colors. You'd get a headache just walking past his office, let
alone trying to work on his computer. :-) 

It didn't help that the RGB guns were misaligned so everything had fringes.

-- 
Darren New / Senior MTS & Free Radical / Invisible Worlds Inc.
San Diego, CA, USA (PST).  Cryptokeys on demand.
A million monkeys in a room with a million typewriters 
              will only yield half a million pregnant monkeys.


From dfb at mrao.cam.ac.uk  Tue Mar 13 07:09:36 2001
From: dfb at mrao.cam.ac.uk (David Buscher)
Date: Tue, 13 Mar 2001 12:09:36 +0000
Subject: Pmw.Blt and python 2.0/2.1
In-Reply-To: <Pine.SOL.4.30.0103091241460.28488-100000@mraos>
References: <Pine.SOL.4.30.0103091241460.28488-100000@mraos>
Message-ID: <Pine.SOL.4.30.0103131201200.11990-100000@mraos>

Responding to my own query, I found that the problem is that when the
Python 2.1 autoconfigures which modules to build, it figures out that tk
is present and so builds tkinter, but doesn't figure out that BLT is also
there (despite the fact that libBLT.so is in the same directory as
libtk.so). It also turns out that the BeOpen Python 2.0 tkinter is built
without BLT enabled. Editing the Modules/Setup to enable tk and BLT works.

This highlights a problem with the new autoconfigure system in 2.1: how do
you find out what was actually built? Modules/Setup does not even show
that _tkinter.so should be built, and yet it is built. Where is the
autoconfigure system's idea of what is going to be built kept?

David

On Fri, 9 Mar 2001, David Buscher wrote:

> I've used Pmw.Blt with Python 1.5.2 with no problems, but I can't get it
> to work with Python 2.0 or 2.1. I can get Pmw to work fine, but not
> Pmw.Blt. Is there a known problem with this? Here is the session
> transcript:
>
> Python 2.1b1 (#1, Mar  8 2001, 20:37:37)
> [GCC egcs-2.91.66 19990314/Linux (egcs-1.1.2 release)]
> on linux2
> Type "copyright", "credits" or "license" for more
> information.
> >>> from Tkinter import *
> >>> root=Tk()
> >>> import Pmw
> >>> Pmw.Blt.haveblt(root)
> 0
> >>> root.tk.call('package', 'require', 'BLT')
> Traceback (most recent call last):
>   File "<stdin>", line 1, in ?
> TclError: couldn't load file "/usr/lib/libBLT.so.2.4":
> /usr/lib/libBLT.so.2.4: undefined symbol:
> Tk_CanvasTagsParseProc
> >
>
> The same dialogue on python1.5 has no problems
>
> Python 1.5.2 (#1, Feb  1 2000, 16:32:16)  [GCC
> egcs-2.91.66 19990314/Linux (egcs- on linux-i386
> Copyright 1991-1995 Stichting Mathematisch Centrum,
> Amsterdam
> >>> from Tkinter import *
> >>> root=Tk()
> >>> import Pmw
> >>> Pmw.Blt.haveblt(root)
> 1
> >>> root.tk.call('package', 'require', 'BLT')
> '2.4'
>
> I'm running Redhat 6.2, and the 1.5.2 python is 'out of the box', while
> the 2.1 is compiled from source - could this cause problems?
>
> My deduction from this output is that this is a problem with the way
> tkinter loads shared libraries - has this changed from 1.5 to 2.0/2.1? I'm
> using the same shared libraries for tk and blt in all cases.
>
> thanks
> David
>
> --------------------------------------+----------------------------------
>  Cavendish Laboratory                 |  Phone +44 1223 337302
>  Madingley Road                       |  Fax   +44 1223 354599
>  Cambridge CB3 0HE, UK                |  http://www.mrao.cam.ac.uk/~dfb
> --------------------------------------+----------------------------------
>
>



From carribeiro at yahoo.com  Sun Mar 25 22:23:59 2001
From: carribeiro at yahoo.com (Carlos Ribeiro)
Date: Mon, 26 Mar 2001 00:23:59 -0300
Subject: Yet Another PEP:  Query Protocol Interface or __query__
In-Reply-To: <Pine.LNX.4.21.0103251655070.4503-100000@clarkevans.com>
References: <99lnc6$ck4$1@tyfon.itea.ntnu.no>
 <Pine.LNX.4.21.0103201837570.21343-100000@clarkevans.com>
 <mailman.985490195.2917.python-list@python.org>
 <99lnc6$ck4$1@tyfon.itea.ntnu.no>
Message-ID: <5.0.2.1.0.20010325210220.00a2cc00@mail.inet.com.br>

I'm thinking about a slightly different approach for the __adapt__ PEP. 
However, I need some help to clarify my ideas - english is not my first 
language, so please be patient :-) I also did my homework, and I have a 
very nice  that works fine for me.

0) Let's start from the scratch (I'm not sure if we're trying to solve the 
same problem)

1) What is a protocol? The simplest definition that I cant think of is: a 
protocol is just a bunch of methods. An object is said to support a 
protocol if it implements all methods of the protocol specification. The 
specification defines the names and parameters of all methods.

2) A 'protocol instance' is the minimal object instance that completely 
supports the protocol.

3) Now we're getting closer. My proposal is to have a method called 
__adapt__ that returns a 'protocol instance'.

4) So __adapt__ must convert a 'object instance' into a 'protocol 
instance'. If the object implements the standard names with the default 
names, this is pretty straighforward - the object could be used as a 
protocol instance.

5) The problem occurs when the object supports the protocol using methods 
with different names. Assume that we have two different protocols asking 
for methods with the same names, but with different semantics and/or 
parameters. An object that supports both protocols will need some way to 
map its internal methods to the ones defined by each protocol.

6) We can solve (5) using a 'proxy', or 'adapter'.

EXAMPLE

Assume that we have these protocols:

a) protocol 'IntegerDivision'
    supports the __div__() method for integer division
b) protocol 'FloatDivision'
    supports the __div__() method for floating point division

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

class Protocol:
   """ empty class that can be used for all protocol adaptation """
   pass

def MakeProtocolDefault(protocol_methods, adaptee):
   """ Returns the default protocol adapter for an object
       The default uses the same name for all methods in the protocol
       and in the object instance that is being adapted
   """
   p = Protocol()
   for pm in protocol_methods:
     om = getattr(adaptee, pm)
     setattr(p, pm, om)
   return p

def MakeProtocol(protocol_methods, object_methods):
   """ Returns the a protocol adapter for an object, mapping
       each protocol_method to a object_method
   """
   p = Protocol()
   for pm, om in zip(protocol_methods, object_methods):
     setattr(p, pm, om)
   return p

class Number:
   """ supports both protocols: 'IntegerDivision' & 'FloatDivision' """
   def __init__(self, p_value):
     self.value = p_value

   def div_int(self, b):
     # integer division
     return int(self.value)/int(b)

   def div_float(self, b):
     # floating point division
     return float(self.value)/float(b)

   def __adapt__(self, protocol):
     if protocol == 'IntegerDivision':
       return MakeProtocol(['__div__'], [self.div_int])
     elif protocol == 'FloatDivision':
       return MakeProtocol(['__div__'], [self.div_float])
     else:
       return None

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

 >>> n = Number(2)
 >>> in = n.__adapt__('IntegerDivision')
 >>> fn = n.__adapt__('FloatDivision')
 >>> in/3
0
 >>> fn/3
0.66666666666666663
 >>> n.value = 5
 >>> fn/3
1.6666666666666667
 >>>

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

Some notes:

a) Please keep this in mind: I tried to keep the implementation as simple 
as possible. I'm not trying to emulate any specific language construct. My 
intention is just to provide a simple 'protocol adapter' layer for Python.

b) I'm using strings to represent protocol names. I fell that this is a 
very simple and extensible approach. For example, the string could be a URL 
pointing to the canonical description of the interface, made using XML.

c) The MakeProtocol helper function matches all methods desired in the 
protocol with the ones implemented in the object. The actual syntax may me 
different (eg passing a single sequence containing the pairs, or passing a 
dictionary), but this is the basic idea.

d) The first parameter of MakeProtocol can be thought of as the 'protocol 
definition'. However I dont have any specific suggestion on this right now 
- thats why I'm using a simple list.

e) The Protocol class is empty; this was done to avoid potential name 
clashes and to ensure that *any* protocol can be adapted using it.

f) A very simple optimization would be to prepare all 'protocol adapters' 
on __init__, returning them inside __adapt__. This can be done if you 
expect a lot of __adapt__ calls.


p.s. I think that it is possible to devise a highly optimized 
implementation. In fact, if all protocols are known at compile time, this 
should make it easier to optimize method calls, but this need to be 
explored more carefully. This does not need to be made right now - as Clark 
pointed out, may be it's better to see if people actually uses this.

pp.ss. It was not my original intention, but we could use this mechanism, 
as exposed in my example, as a way to implement the PEPs that propose 
semantic changes on the numeric types.


Carlos Ribeiro


_________________________________________________________
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com




From bedge at troikanetworks.com  Thu Mar 22 14:42:45 2001
From: bedge at troikanetworks.com (Bruce Edge)
Date: Thu, 22 Mar 2001 11:42:45 -0800
Subject: providing a python shell from an app, with app defined objects available.
Message-ID: <20010322.114245.1314218593.7262@mead.troikanetworks.com>

I apologize for the length, but here's the simples example I can make:

file xx.py contains:
def run_python():
	banner = "Use Ctrl-D to exit subshell"	
	import code
	import __main__
	locals = __main__.__dict__
	x = code.InteractiveConsole(locals)
	x.interact( banner )

run_python()

>>> from xx import *
Use Ctrl-D to exit subshell
>>> dir()
['__builtins__', '__doc__', '__name__', 'atexit', 'histfile', 
'os', 'readline', 'rlcompleter', 'savehist', 'sys']
>>> 

Note: no run_python defined

Ctrl-D hit here, back to main shell.

>>> 
>>> dir()
['__builtins__', '__doc__', '__name__', 'atexit', 'histfile', 
'os', 'readline', 'rlcompleter', -> 'run_python' <-, 'savehist', 'sys']

Now run_python is defined.

How can I get that function into the InteractiveConsole name space?

If I re-run run_python now:

>>> run_python()
Use Ctrl-D to exit python and return to CLI
>>> dir()
['__builtins__', '__doc__', '__name__', 'atexit', 'histfile', 
'os', 'readline', 'rlcompleter', 'run_python', 'savehist', 'sys']
>>> 

It _is_ defined. But, I need it to work without having to drop back to
the main interpreter shell first.

This is needed to allow a user to get access to all the objects the app defines,
but never give them direct access to the main shell.

If this is a case of RTFM I would greatly appreciate a pointer to the
section of the FM.

Thanks, Bruce.


From jarober at mail.com  Sun Mar 25 23:00:38 2001
From: jarober at mail.com (James A. Robertson)
Date: Mon, 26 Mar 2001 04:00:38 GMT
Subject: Collection interfaces
References: <3a90fa97.0@news1.mweb.co.za>   <97h5s40vhc@news2.newsguy.com> <MPG.1506007591cf818b9896bc@news.earthlink.net> <slrn99opop.gr.amk@mira.erols.com> <MPG.1506265ebf6d934d9896be@news.earthlink.net> <97ibmc0k47@news1.newsguy.com> <MPG.1506cf3293b094b49896c1@news.earthlink.net> <97m0c7080c@news2.newsguy.com> <MPG.15085517debf4b649896ca@news.earthlink.net> <zb%n6.31033$o%6.586026@news1.rdc1.bc.home.com> <MPG.150b383cea226d409896d0@news.earthlink.net> <kNko6.6986$eq3.707121@news.uswest.net> <MPG.150c46db9cec283e9896d4@news.earthlink.net> <3AA2A898.450B659E@mail.com> <gRCt6.12224$Im6.1308560@newsread1.prod.itd.earthlink.net> <3AB755F8.67E84126@mail.com> <%a%t6.17245$227.1601569@newsread2.prod.itd.earthlink.net> <3AB8A730.4BE13D22@mail.com> <61bv6.8693$P4.676800@newsread1.prod.itd.earthlink.net> <B6E2B64C.D8E3%t*b*r*u*m*l*e*y@f*u*s*e.n*e*t> <o1jv6.9635$P4.789083@newsread1.prod.itd.earthlink.net> <3ABF74ED.BC09EC8@mail.com> <7iyv6.11924$ue1.1104528@newsread2.prod.itd.earthlink.net>
Message-ID: <3AC01057.57920FA6@mail.com>

topmind wrote:
> 
> "James A. Robertson" <jarober at mail.com> wrote in message
> news:3ABF74ED.BC09EC8 at mail.com...
> > As to the LessonBrowser, it's part of the non-commercial product.  When
> > you open the system up, there it is
> >
> >
> 
> What exactly would I be "downloading"?
> 
> Is it a Java applet or what?

You would download VisualWorks non-commercial - the Smalltalk
development system

> 
> What do you mean by "open the system up"?

As in, start the non-commercial development system

> 
> Being non-commercial does not necessarily make it
> safer (less DLL's etc.)

We don't link to many dll's, and we introduce zero new ones

> 
> Is there an HTML version?
>

There are tutorials online, but they assume that you have a running
system in front of you
 
> I clicked on the "visualworks online tutorial" link, it asked for
> my email ( foo at bar.com, of course :-), and then it brought me
> back to the front page.
> 

The new site is still under construction; we are in the process of
updating it

> The second try gave me an HTTP error.
> 
> -tmind-

-- 
James A. Robertson
Product Manager (Smalltalk), Cincom
jarober at mail.com
<Talk Small and Carry a Big Class Library>


From schaffer at optonline.net  Fri Mar 23 08:20:31 2001
From: schaffer at optonline.net (Les Schaffer)
Date: Fri, 23 Mar 2001 13:20:31 GMT
Subject: VPython/Tk app works on single processor but not dual
References: <995gi1$9ce$1@bob.news.rcn.net>
Message-ID: <acjmbtsg82rm7biao0e3cmnchff8et6qqr@4ax.com>

Fred:

email to you keeps bouncing....

sorry, got super busy at work.

would you like me to send along my compiled versions of the dll's for
VPython?

otherwise, i can probably send you my VC++ project file, or what are
you using for compilation.

one last question: what video card are you using?

les schaffer

p.s. am interested too in the math side of your application, but will
focus on mundane things first.





From geoff at ou043085.otago.ac.nz  Thu Mar 29 06:03:58 2001
From: geoff at ou043085.otago.ac.nz (Geoff Low)
Date: Thu, 29 Mar 2001 11:03:58 GMT
Subject: Gaussian line profile using numeric python
Message-ID: <3ac31698@zfree.co.nz>

Hi.  

I have an extensive set of xy data that I want to convolute with a gaussian
profile.  Below is my algorithm,  which doesn't work.  Can anyone see where
I'm being dumb?

import Numeric
from MLab import sum
from Numeric import arange,zeros,transpose,shape,exp,power,pi
from TableIO import readTableAsArray,TableFromColumns,writeTable
from math import sqrt

def get_data(filename):
    d = readTableAsArray(filename,'#')
    x = d[:,0]
    y = d[:,1]
    del d
    return x,y

def main(inpfile,outfile='res.out',sfreq=350,efreq=5050,stepf=2,fwhm=2.5):
    fin,iin = get_data(inpfile)
    xpt = arange(sfreq,efreq,stepf,'d')
    ypt = calcspec(fin,iin,xpt,fwhm)
    del fin,iin
    put_data(outfile,xpt,ypt)
    
def put_data(filename,x,y):
    tmp = TableFromColumns([x,y])
    tmp['filename'] = filename
    writeTable(tmp)
    del tmp
    
def calcspec(fin,iin,x,fw):
    norm,expterm = 0.0,0.0
    inten = zeros((len(x)),'d')
    norm = float(1/(sqrt(2.0*pi)*fw))
    expterm = float(1/(2*(fw**2)))
    for i in range(len(x)):
        print "Starting step ",i+1,"of ",len(x)
        print "sum fin: ",sum(fin)," sum iin: ",sum(iin) 
        print "x[",i,"] = ",x[i],"; SUM(x[",i,"] - fin) = ",sum((x[i]-fin[:]),0)
        #inten[i] = sum((norm*iin[:]*exp(-1.0*(power((expterm*(xpts[i]-fin[:])),2)))),0)
        inten[i] = norm*sum(iin[:]*exp(-1.0*(power(((x[i]-fin[:])*expterm),2))))
        #for j in range(len(fin)):
        #inten[i] = inten[i] + float(norm*iin[j]*exp(-1.0*((expterm*(xpts[i]-fin[j]))**2)))
    return inten

P.S.  I'm no math wizard but this is my interpretation of the gaussian lineshape
function
G(v) = 1/(sqrt(2*PI)*FWHM)*exp(-(v-vo)^2/(2*FWHM^2))

Cheers 
Geoff


http://www.zfree.co.nz

######
Fortune favours the brave....
  but keeps an axe aside for the stupid!
######


From lkalesch at gmx.net  Mon Mar  5 05:35:32 2001
From: lkalesch at gmx.net (Lars)
Date: Mon, 05 Mar 2001 11:35:32 +0100
Subject: HDF4
Message-ID: <3AA36BF4.25C0BF89@gmx.net>

Hello, 

i am looking for a Python module for reading HDF4 data.
I know that there exist a module for HDF5, but unfortunately 
HDF5 is not compatible to HDF4. 

I tried the following for reading HDF4 data, but this
is faulty and dowdy:

def hdfdump(hdffile,var):
    key="Data-Set-"
    os.system("ncdump -v "+key+var+" "+hdffile+"> /tmp/hdffile")
    f=open('/tmp/hdffile','r')
    a=f.read()
    f.close()
    findexp=key+var+" ="# Data-Set-4 = 23424,23123,..
    s1=find(a,findexp)
    s2=find(a,";",s1)
    b=a[s1+len(findexp):s2]
    return b



Thank you very much for your help,

Regards,

Lars Kaleschke
Lars at Seaice.de
http://www.seaice.de


From robin.thomas at starmedia.net  Thu Mar  1 01:41:00 2001
From: robin.thomas at starmedia.net (Robin Thomas)
Date: Thu, 01 Mar 2001 01:41:00 -0500
Subject: is and == (is not other_thread)
In-Reply-To: <3A9DD10F.C8ECF803@seebelow.org>
Message-ID: <4.3.1.2.20010301010759.00d37100@exchange.starmedia.net>

At 10:33 PM 2/28/01 -0600, Grant Griffin wrote:
>Hi Gang,
>
>I've had a few of my own little questions about "is" and "==", vis-a-vis
>strings:

a is b # are a and b two references to a single object
a == b # are a and b references to "equivalent" objects or to a single object

if a is b, then a == b

>1) Are two strings that are equal in the "==" sense always guaranteed to
>be equal in the "is" sense?

No. There's a previous post explaining it well, I'll summarize:

String literals are "interned" by Python, so that it uses a single string 
object for every unique string literals it finds in the code. Strings built 
at run-time by concatenation or other means are not interned, and thus are 
always different objects.

 >>> s1 = "ab"
 >>> s2 = "ab"
 >>> s1 is s2
1
 >>> s1 == s2
1
 >>> s3 = "a" + "b"
 >>> s1 is s3
0
 >>> s1 == s3
1


>2) Is "is" faster than "==" for strings?

"is" is speedier than "==" for all object pairs save a pair of integers. 
Python does "is" and "==" with the same speed to a pair of integers.

>  Of hand, it would seem so.
>That's why I'm tempted to use it for string equality.  (OTOH, if strings
>can be "==" equal without being "is" equal, than I guess that's wildly
>false economy. <wink>)

Don't use the identity operator "is" to test object equality, just as you 
would not use the bitwise xor operator to test object equality. "is" and 
"==" are different operations meaning different things. Don't be tempted.

>3) Internally, does "==" for strings always begin with an "is"
>operation?

Comparison in CPython does an implicit "is" operation first for any objects 
except integers, which are speedily handled beforehand in a special case.


>(btw,-nobody-ever-answers-every-part-of-my-silly-multipart-questions,
>    -so-i-dare-you-folks-to-buck-the-trend-<wink>)-ly y'rs,

Use the source, Luke. All multipart questions are thus answered. This reply 
was your freebie.


--
Robin Thomas
Engineering
StarMedia Network, Inc.
robin.thomas at starmedia.net




From aleaxit at yahoo.com  Thu Mar 29 08:39:22 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 29 Mar 2001 15:39:22 +0200
Subject: Defining class files
References: <3ac33343$0$12248$ed9e5944@reading.news.pipex.net>
Message-ID: <99vdvh02dbb@news2.newsguy.com>

"Neil Benn" <neil.benn at cambridgeantibody.com> wrote in message
news:3ac33343$0$12248$ed9e5944 at reading.news.pipex.net...
    [snip]
>     However, I then tried to write the class as an external text file,
    [snip]
>     I then imported the class, tried to assign the class and ivoke a
method

Not really -- you imported the MODULE which contains the class:

> >>> y = YourClass()
> Traceback (innermost last):
>   File "<console>", line 1,
> TypeError: call of non-func
>
>     This seemed a bit strange, so I investigated the Your class and it
> seemed to be a module.
>
> >>> YourClass
> <module YourClass at 2584541>

So, YourClass is a module object -- it no doubt refers to a
class object (called, from the outside, YourClass.YourClass).

>     The text file is saved as YourClass.py - is this the problem, should
> class files have different terminaters in their filename??

No problem in having the same name.  But, after an
    import YourClass
you must use YourClass.YourClass to refer to the
_class_ object; or else (normally less advisable),
    from YourClass import YourClass
will let you use just YourClass to refer to the
class object (but you have no name for the module
object, which in some cases is helpful to use).


Alex





From itamarst at omnisky.net  Sat Mar  3 22:34:38 2001
From: itamarst at omnisky.net (itamarst)
Date: Sat, 03 Mar 2001 22:34:38 -0500
Subject: moshe zadka's birthday party now at ipc9
Message-ID: <OMMD_CRM_SQL01EvNPl00006275@OMMD_CRM_SQL01.aethertest.com>

We are in the hotel lobby (its 7:30 pm).

--
Sent wirelessly using OmniSky




From root at rainerdeyke.com  Sun Mar 11 11:56:00 2001
From: root at rainerdeyke.com (Rainer Deyke)
Date: Sun, 11 Mar 2001 16:56:00 GMT
Subject: Static typing (was Re: Java guy interested in Python)
References: <mailman.984294734.21888.python-list@python.org>
Message-ID: <A6Oq6.395631$ge4.138241941@news2.rdc2.tx.home.com>

"Tim Peters" <tim.one at home.com> wrote in message
news:mailman.984294734.21888.python-list at python.org...
>
> All type systems in all programming languages are hacks, because at heart
> what you *really* want to say is "variable x must take on values from set
S",
> and interesting specifications for S may not even be computable.  So you
> approximate in more-or-less crude ways, saying "i must be an int" when you
> really mean "i must be an even integer" or "i must be an odd prime";
saying
> "x and y must be reals" when you really mean "x and y must be reals such
that
> x**2 + y**2 <= 1"; etc.  A static type system is interesting to the extent
> that it closely approximates what you want to say most often, and can be
> efficiently checked at compile-time (there's nothing you can say in a
static
> type system that can't be said today at runtime in Python).

When I type 'int i;' in C++, what I really mean is "'i' behaves like an
integer variable.".  I'm not saying anything about the values 'i' can take,
except that the values must support integer semantics.  The limited range of
type 'int' is part of its defined semantics, as is its fast performance.  If
I want to restrict 'i' to a subset of its possible values, I use an
assertion.


--
Rainer Deyke (root at rainerdeyke.com)
Shareware computer games           -           http://rainerdeyke.com
"In ihren Reihen zu stehen heisst unter Feinden zu kaempfen" - Abigor




From iddwb at imap1.asu.edu  Mon Mar 26 20:27:03 2001
From: iddwb at imap1.asu.edu (iddwb)
Date: Tue, 27 Mar 2001 01:27:03 +0000
Subject: math.choice attribute error
Message-ID: <Pine.LNX.4.21.0103270124290.3010-100000@moroni.pp.asu.edu>

from the ref man math.choice says it takes a sequence object and returns a
random choice. perfect for my needs. the follow happens:

>>> import math
>>> list = ['apple', 'orange', 'pear']
>>> print math.choice(list)
Traceback ....
attributerror: choice

what have I missed?

David Bear
College of Public Programs/ASU



From moshez at zadka.site.co.il  Fri Mar 23 06:47:32 2001
From: moshez at zadka.site.co.il (Moshe Zadka)
Date: Fri, 23 Mar 2001 13:47:32 +0200
Subject: python, perl, lisp, tcl / got a decision to make, can you give me a  view point please ? and a joke for python users : )
In-Reply-To: <TPvu6.3114$4N4.450174@newsc.telia.net>
References: <TPvu6.3114$4N4.450174@newsc.telia.net>, <3aba3076.14395287@news.vif.com> <mailman.985284892.29036.python-list@python.org>
Message-ID: <E14gQ2W-0008KW-00@darjeeling>

On Thu, 22 Mar 2001, "Fredrik Lundh" <fredrik at pythonware.com> wrote:

> doesn't work; it imports both "warnings_are_errors" and
> "from_import_star_is_evil", and as I've pointed out before,
> catching ParadoxErrors is really hard.

As you've pointed out in the future, you mean? 2005, I think it 
was.

-- 
"I'll be ex-DPL soon anyway so I'm        |LUKE: Is Perl better than Python?
looking for someplace else to grab power."|YODA: No...no... no. Quicker,
   -- Wichert Akkerman (on debian-private)|      easier, more seductive.
For public key, finger moshez at debian.org  |http://www.{python,debian,gnu}.org



From urner at alumni.princeton.edu  Sat Mar  3 01:52:02 2001
From: urner at alumni.princeton.edu (Kirby Urner)
Date: Fri, 02 Mar 2001 22:52:02 -0800
Subject: Adding a method to an instance
References: <G2%n6.360714$JT5.11606234@news20.bellglobal.com> <6621ato2vknc667m3iuqvmm8vcj0h99qqk@4ax.com> <Sc0o6.361217$JT5.11613656@news20.bellglobal.com>
Message-ID: <ej41atcongof4vpi5i2vkrilalbdbc06m4@4ax.com>

Bryan Mongeau <bryan at eevolved.com> wrote:

>Kirby Urner wrote:
>
>Thanks Kirby,
>
>> In saying f.__class__, you point back to the superclass foo.
>> If you want to poke newMethod into object f alone, then you
>> should just go:
>> 
>> f = newMethod
>
>This I do not want. Maybe I should have been more clear. I would like to 
>add "newMethod" to the instance of class foo so that later I can call:
>
>f.newMethod()
>

Right, I typo'ed here too, making it worse.

>> 
>> Here's a transcript that addresses your question:
>> 
>>  >>> class foo:
>> def duh(self):
>>             print "Duh!"
>
>Is my question really that bad? :)
> 

No, it's a good question -- didn't mean to imply otherwise
with the "Duh!".  That's something I often use on my end, when
I'm doing 'foo' type testing (a local convention you might say).

>>             
>>  >>> o = foo()
>>  >>> o
>>  <__main__.foo instance at 00B9719C>
>>  >>> k = lambda x: x+2
>>  >>> k(2)
>>  4
>>  >>> o.othermethod = k
>
>k is a lambda function. I would need the new method to be able to access 
>internal class attributes, thus require the "self" parameter of class 
>methods.
>

OK, that's the nub of the issue, and a sense in which my answer
failed to answer your question.

>Okay, based on these statements and referring to my previous example, I 
>should be able to do this:
>
>def newMethod(self):
>  print self.bar
>
>f.newMethod = newMethod
>
>Problem is:
>
>f.newMethod()
>TypeError: not enough arguments; expected 1, got 0 
>
>Understandably though, this works in this case:
>f.newMethod(f)
>
>
>> I bet you get it now.
>
>Sorry, I must be really dense or something... :) I don't see how your 
>method can assist me.

Let me try again:

 >>> class foo:
	def duh(self, message):
	    self.msg = message
            print self.msg

            
 >>> o = foo()
 >>> o.duh("1-2-3 testing")
 1-2-3 testing
 >>> o.msg
 '1-2-3 testing'
 >>> def newMethod(self,newtext):
 	return self.msg + newtext

 >>> o.newMethod = newMethod
 >>> o.newMethod(o," how about it?")
 '1-2-3 testing how about it?'
 >>> dir(o)
 ['msg', 'newMethod']
 >>> dir(foo)
 ['__doc__', '__module__', 'duh']

When you invoke o.newMethod, you still have to pass your 
object as an instance, so that the internal 'self' has 
something to work with.  Is that so bad?  Only your instance
has this new method, and you are only consulting its own
internals by passing a reference to that object as the 
first parameter.

Kirby



From emile at fenx.com  Wed Mar 28 17:42:46 2001
From: emile at fenx.com (Emile van Sebille)
Date: Wed, 28 Mar 2001 14:42:46 -0800
Subject: What *instance* called me & proxy methods
References: <3AC1BE92.ACE82741@ua.ac.be> <mailman.985798623.5866.python-list@python.org>
Message-ID: <99tpim$2rrba$1@ID-11957.news.dfncis.de>

"Carlos Alberto Reis Ribeiro" <cribeiro at mail.inet.com.br> wrote in message
news:mailman.985798623.5866.python-list at python.org...
<snip>

> and it has some undesirable side effects because it is not completely
> transparent, and it needs some intermediate processing. The only way that
I
> found to signal to the proxied object about the existence of the proxy was
> to create a temporary "proxy" attribute (see point (3) above). However, it
> serves it purpose as a proof-of-concept:
>

It seems to me that what you're trying to do is discover the container of an
instance.  Whether the instance is wrapped in a class or held in a list,
when the instance's methods are invoked, self is self and not the container.
If you take a look at pep 232
http://python.sourceforge.net/peps/pep-0232.html you may find that, if
instance method attribute setting were allowed, this could be what you're
looking for.

Perhaps you can take this PEP to the next step if it's what you're after

Regards,

--

Emile van Sebille
emile at fenx.com

---------




From loewis at informatik.hu-berlin.de  Fri Mar 23 03:14:24 2001
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 23 Mar 2001 09:14:24 +0100
Subject: New PEP: The directive statement
References: <mailman.985301265.2425.python-list@python.org>
Message-ID: <j4hf0kkilb.fsf@informatik.hu-berlin.de>

Bruce Sass <bsass at freenet.edmonton.ab.ca> writes:

> For example, I'd rather do...
> 
> use:
>     new_math.//
>     result = new_math_based_function()
> except NewMathError:
>     result = old_math_based_function()
> 
> ...and have the compiler use the appropriate source depending on what
> the interpreter can handle, rather than do a check of the
> "sys.version"[1] or try to catch a run-time exception.

And you would supply both new_math_based_function and
old_math_based_function in the same module, outside of the use: block?
That is probably difficult to implement, as the compiler *when
processing new_math_based_function* needs to know what / means.

Regards,
Martin


From timr at probo.com  Sun Mar 18 00:58:10 2001
From: timr at probo.com (Tim Roberts)
Date: Sat, 17 Mar 2001 21:58:10 -0800
Subject: HTMLparsing abnormal html pages
References: <98pvp1$15t$1@news.netmar.com> <98ua8j$a6h$1@panix2.panix.com> <B6D81A6D.4FE6%f8dy@diveintopython.org>
Message-ID: <6ej8btssj4tf5t14gca2ge4dt6bdbajbtc@4ax.com>

Mark Pilgrim <f8dy at diveintopython.org> wrote:
>
>In fact, the BaseHTMLProcessor class I define in my book can be used to
>properly quote all attribute values, since it works by breaking down the
>entire HTML (via sgmllib) and building up equivalent HTML with proper quotes
>around attribute values.

I have been searching for an HTML pretty-printer; something where I can
feed an arbitrary page and get a more structured, indented view.  I wrote a
simple one myself, based on sgmllib; it does a fair job, but it is easily
confused by such common offenses as omitted </p> tags.  It sounds like your
BaseHTMLProcessor might be such a thing.  Is it available yet?

If not, is anybody aware of a fair HTML cleaner-upper?
--
- Tim Roberts, timr at probo.com
  Providenza & Boekelheide, Inc.


From danielk at aracnet.com  Mon Mar  5 10:12:54 2001
From: danielk at aracnet.com (Daniel Klein)
Date: Mon, 05 Mar 2001 07:12:54 -0800
Subject: Static class methods
References: <rdzo6.154$gc4.80445@news.pacbell.net>
Message-ID: <aqa7at0b6cbqjaq23mli0flmumlp8n6s9s@4ax.com>

On Sun, 4 Mar 2001 14:22:10 -0800, "Alex Shindich" <alex at shindich.com> wrote:

>Just out of my curiosity, have any of you ladies and gentlemen experienced a
>need for an equivalent of C++/Java static method?
>Because I have, but it is impossible add them to Python classes -- all class
>methods are treated as instance methods.

The Quick Python Book by Daryl Harms and Kenneth McDonald outlines a way to
accomplish this by breaking out the class into its own module and then define
the 'class methods' as top-level functions.

myclass.py

def myClassMethod():
	# do some class thing

class Foo:
	def __init__(self):
                  pass
	def bar(self):
		pass

Daniel Klein


From phd at phd.pp.ru  Wed Mar  7 10:05:46 2001
From: phd at phd.pp.ru (Oleg Broytmann)
Date: Wed, 7 Mar 2001 18:05:46 +0300 (MSK)
Subject: Python script for deMSWordification?
In-Reply-To: <985h02$1pbm$1@swen.emba.uvm.edu>
Message-ID: <Pine.LNX.4.30.0103071802380.6080-100000@fep132.fep.ru>

On 7 Mar 2001 kelvin.chu at uvm.edu wrote:
> Numerous colleagues and administration types send memos, colloqium
> announcements and other things around in Micro$oft word format.  Has anyone
> written a filter in Python that can convert these things into normal text?
>
> If not, is there a basic set of rules for decoding .doc format files
> anywhere that I can look at?

   http://www.ice.ru/~vitus/catdoc/
   http://www.wvware.com/
   http://www.winfield.demon.nl/
   http://arturo.directmail.org/filtersweb/
   http://snake.cs.tu-berlin.de:8081/~schwartz/pmh/

Oleg.
----
     Oleg Broytmann            http://phd.pp.ru/            phd at phd.pp.ru
           Programmers don't die, they just GOSUB without RETURN.




From jwbnews at scandaroon.com  Fri Mar  9 21:08:40 2001
From: jwbnews at scandaroon.com (John W. Baxter)
Date: Fri, 09 Mar 2001 18:08:40 -0800
Subject: gah! I hate the new string syntax
References: <mailman.983615170.23313.python-list@python.org> <tN5o6.2746$eH3.14913@NewsReader> <Koco6.7518$1D5.244602@e420r-atl1.usenetserver.com> <mailman.983997252.5628.python-list@python.org> <m2vgpkyfjo.fsf@mycroft.actrix.gen.nz> <mailman.984063199.18779.python-list@python.org> <98an7i01tgp@news2.newsguy.com> <mailman.984154996.23254.python-list@python.org>
Message-ID: <jwbnews-CCCF8A.18083909032001@corp.supernews.com>

In article <mailman.984154996.23254.python-list at python.org>,
 D-Man <dsh8290 at rit.edu> wrote:

> On Fri, Mar 09, 2001 at 02:52:53PM +0100, Alex Martelli wrote:
> | "D-Man" <dsh8290 at rit.edu> wrote in message
> | news:mailman.984063199.18779.python-list at python.org...
> | > that I am not really American since none of my ancestors came from
> | > America.  "Real" Americans would be the so-called "Indians", or more
> | 
> | Why?  *THEIR* ancestors came from Asia over the Bering Straits,
> | after all (and following the line a while longer, one gets back
> | to Africa).
> 
> So we think, but no one has such records.  Actually, we are all
> "people" and came from the Garden of Eden to take the line back to its
> origin.  :-)
> 
> -D

Part of the annoyance which the "Kennewick Man" remains cause stems from 
their implication that the folks who came from Asia over the Bering 
Strait land bridge may have found humans here ahead of them.

Another part stems from the fact that if Kennewick Man wasn't here 
before that group, then he was of them and his remains should be 
properly respected as such.

This has been a long struggle, with at one point a court order requiring 
the remains to be reburied without further study.

[Kennewick, Washinton, USA, that is.  I'm sure a Google search on 
Kennewick will turn up something about it.]

  --John

-- 
John W. Baxter   Port Ludlow, WA USA  jwbnews at scandaroon.com


From aahz at panix.com  Tue Mar 13 10:49:24 2001
From: aahz at panix.com (Aahz Maruch)
Date: 13 Mar 2001 07:49:24 -0800
Subject: module importing and variable syntax
References: <3AADF3DC.1030003@dynworks.com>
Message-ID: <98lfi4$74p$1@panix2.panix.com>

In article <3AADF3DC.1030003 at dynworks.com>,
Jeff Davis  <jdavis at dynworks.com> wrote:
>
>Module importing doesn't take a string argument, but just a "bare word" 
>(in perl terms). This means I have to know the name of the file in 
>advance, and it needs to be in the right directory, or I need to use eval().
>
>What I am trying to do is have an arbitrary string with an arbitrary 
>type (type with respect to my program, not python) access the function 
>associated with the string in the module associated with the type.
>
>For example: someone writes an extension to my program creating a new 
>object of name "O". Then some other part of the program (i.e. another 
>extended module) tries to square an instance of that object, like:
>o = O()
>o.square()
>
>However, my program is completely unaware of either "O" or "square" 
>until someone writes the module(s) and registers them with the program 
>(not writing to the main program's source).

One of the most important Python rules is "explicit is better than
implicit".  So if a person wants to register a generic class with your
module, provide a function for them to do it!

module A:

    object = None
    def register(newClass):
        global object
        object = newClass


module B:

    import A
    class Foo:
        def square(self):
            pass
    A.register(Foo)


Module C:

    import A
    import B
    o = A.object()
    o.square()
-- 
                      --- Aahz <*>  (Copyright 2001 by aahz at pobox.com)

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

Python:
    Programming the way
    Guido indented it


From aleaxit at yahoo.com  Tue Mar 27 14:08:10 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Tue, 27 Mar 2001 21:08:10 +0200
Subject: PEP 240 scares me
References: <Pine.LNX.4.21.0103222338530.29360-100000@kev.ioc.ee><99e2h61aao@news2.newsguy.com> <15040.53203.949420.930912@beluga.mojam.com>
Message-ID: <99qp6401mnb@news1.newsguy.com>

"Skip Montanaro" <skip at pobox.com> wrote in message
news:15040.53203.949420.930912 at beluga.mojam.com...
>
>     Alex> AFAIK -- 1.2 would be the rational 6/5
>
> Okay, dumb question time.  Suppose the compiler sees the string "1.2" in
the
> input stream.  How does it translate that into a rational number (6/5 or
> 12/10 or 24/20 or ...)?  Does it simply move the decimal point over
> sufficiently and create the corresponding rational number that contains
the
> appropriate power-of-ten denominator?

Moshe will have to confirm his intent, but I'd be astonished if it
did anything but that.

> What about rationals that can't be
> represented correctly using decimal notation (e.g. how does one get from
> 0.3333 to 1/3, or is it simply approximated by 3333/10000?)

0.3333 is _exactly_ represented by 3333/10000; the approximation
would be that of this exact rational with the different one 1/3.


Alex





From shaleh at valinux.com  Thu Mar 22 16:25:45 2001
From: shaleh at valinux.com (Sean 'Shaleh' Perry)
Date: Thu, 22 Mar 2001 13:25:45 -0800 (PST)
Subject: expanding a variable
In-Reply-To: <B7C74E1A6FBAD411BFC60002B309DDF4E435@mailbox.juilliard.edu>
Message-ID: <XFMail.20010322132545.shaleh@valinux.com>

On 22-Mar-2001 Jonathan Soons wrote:
> 
> I am trying to give my cv group (who all have logins like cv???.)
> all the same .profile. The variable in the last line of the script
> doesn't seem to be expanding, even though this is very similar to
> an example in "Core Python Progr..."
> 
> 
> d = commands.getoutput('ls /u')
> dirs = string.split(d, '\n')
> for dir in dirs:
>     if re.search('^cv', dir):
>         os.system('cp  /u/test/.profile /u/dir/')
>                                            ^^^

os.system('cp /u/test/.profile /u/%s/' % dir)

python does not expand variables in strings.



From fredp at mygale.org.nospam  Thu Mar 22 03:22:26 2001
From: fredp at mygale.org.nospam (Fred Pacquier)
Date: 22 Mar 2001 08:22:26 GMT
Subject: Good Editor?
References: <3ab9a32b$1_1@news2.one.net>
Message-ID: <Xns906C5F868E6B7PaCmAnRDLM@161.48.76.20>

"fjs" <fjs6 at usa.net> said :

>Can someone recomend a good editor to write python in?  Im just getting
>started learning python and dont know which ones support python.

You don't mention the platform. Under Unix/Linux there are a number of 
options (including Vim, Emacs, etc.). Under Windows, I found WScite 
(http://www.scintilla.org) the easiest to get started with...

-- 
YAFAP : http://www.multimania.com/fredp/


From kalle at gnupung.net  Thu Mar  8 14:37:09 2001
From: kalle at gnupung.net (Kalle Svensson)
Date: Thu, 8 Mar 2001 20:37:09 +0100
Subject: How to call C function in shared library?
In-Reply-To: <vi81ys8gke1.fsf@wally.fnal.gov>; from kriol@wally.fnal.gov on Thu, Mar 08, 2001 at 12:56:06PM -0600
References: <vi81ys8gke1.fsf@wally.fnal.gov>
Message-ID: <20010308203709.C6106@father>

Sez Oleg Krivosheev:
> i already have shared library
> with simple function which takes double
> and returns double. Is it simple way 
> to call it directly from Python?

You could take a look at SWIG (http://www.swig.org/), or make your own
wrapper.  Something like (extremely untested):

/************  BEGIN **************/
#include "Python.h"
#include "myheader.h"

static PyObject *
wrap_myfunction(PyObject *self, PyObject *args)
{
    double arg, ret;
    if (!PyArg_ParseTuple(args, "d", &arg))
        return NULL;
    ret = myfunction(arg);
    return Py_BuildValue("d", ret);
}

/* Module initialisation stuff goes here...     */
/* Check http://www.python.org/doc/current/ext/ */
/* and http://www.python.org/doc/current/api/   */
/* for more information...                      */
/************  END **************/

> Or some python encapsulation for dlopen/dlsym/dlclose ?

There is the dl module:
http://www.python.org/doc/current/lib/module-dl.html

Peace,
  Kalle
-- 
Email: kalle at gnupung.net     | You can tune a filesystem, but you
Web: http://www.gnupung.net/ | can't tune a fish. -- man tunefs(8)
PGP fingerprint: 0C56 B171 8159 327F 1824 F5DE 74D7 80D7 BF3B B1DD
[Not signed due to braindamage.  Blame Microsoft Outlook Express.]



From jorn.verwey at psi.ch  Thu Mar 29 05:23:26 2001
From: jorn.verwey at psi.ch (Jorn Verwey)
Date: Thu, 29 Mar 2001 12:23:26 +0200
Subject: Numeric question, putting one matrix into another
Message-ID: <3AC30D1C.620E197B@psi.ch>

Hello,

Using Numeric-arrays I have made a (3x3) rotation-matrix. Now I would like to put this rotation
matrix in a (4x4) transformation matrix. I believe I have to use the put function (the alternative
would be assigning each element of the matrix the proper value), but if so I don't get the indices
set properly. Am I just not bright enought  or is this the wrong approach?

Regards,

Jorn

--
Jorn Verwey
PSI - Life Sciences WMSA/B14
CH 5232 Villigen PSI
Switzerland
tel. +41 56 310 4246
fax. +41 56 310 3132     -o)
jorn.verwey at psi.ch       /\\
jorn.verwey at bigfoot.com _\_v





From danielk at aracnet.com  Wed Mar 21 09:05:14 2001
From: danielk at aracnet.com (Daniel Klein)
Date: Wed, 21 Mar 2001 06:05:14 -0800
Subject: Pick Systems D3 Database
References: <rG%t6.555$45.2416@newsfeeds.bigpond.com>
Message-ID: <pfchbtkf3j0p7prf0sp45vi21fro3bs7ad@4ax.com>

There is no way to communicate with D3 other than thru ODBC as it is a very
insulated database. Could you find out if your customer would be willing to
move the data to jBASE (which is an open database that uses the same data
structure as D3; this should be a trivial operation)? If so, then I have
written a Python client which provides classes and methods to access and
manipulate the database directly without going thru any sort of external
driver.

Daniel Klein
Portland OR USA

On Wed, 21 Mar 2001 23:17:36 +1100, "Ben" <bencr at bigpond.com> wrote:

>Hi,
>
>I have not been able to find any direct information about this so I thought
>I would ask here in the slim chance someone may have had to do something
>similar ... what would be the best way to push/pull information from a D3
>(PickSystems) database running on AIX using Python? Is ODBC my only option
>or are there others? If you have done something similar how succesful was
>your solution?
>
>Thanks in advance
>



From grey at despair.rpglink.com  Fri Mar 23 22:38:55 2001
From: grey at despair.rpglink.com (Steve Lamb)
Date: Sat, 24 Mar 2001 03:38:55 -0000
Subject: Good Editor? (What's the big deal with syntax colouring?)
References: <mailman.985291829.27633.python-list@python.org>
Message-ID: <slrn9bo5me.48h.grey@teleute.dmiyu.org>

On Thu, 22 Mar 2001 12:10:59 -0800, John Schmitt <jschmitt at vmlabs.com> wrote:
>Why do people want syntax colouring?  I can't stand it.

    Trust me, you'd understand the first time in a fit of coding you left a
quote unclosed somewhere and all of a sudden the first half of your code and
the second half of your code don't match, color-wise, at all and it is divided
on that line, right there, at that character.  

    Either that or the 2nd half of your code is all "String" color and you
know where it is as well.

    1/3rd the benefit of syntax highlighting is added context to know what is
what, the other 2/3rds is pure bug highlighting before you're even out of the
code at the syntax level.

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


From MarkH at ActiveState.com  Sat Mar 10 20:04:05 2001
From: MarkH at ActiveState.com (Mark Hammond)
Date: Sun, 11 Mar 2001 01:04:05 GMT
Subject: PyRun_SimpleFile - impossible to call it under Windows?
References: <3aa07704.2672581@news.newsguy.com> <3AA08A11.95BC7745@bigfoot.de> <3aa0bd3f.708696@news.newsguy.com>
Message-ID: <3AAACE47.8000802@ActiveState.com>

Owen F. Ransen wrote:

> I am already linking to Python20_d.lib, is that not the correct
> version?

I hope I am not too late - but make sure your extension is compiled with 
"/MD" for release builds and "/MDd" for debug builds.

Mark.



From aleaxit at yahoo.com  Fri Mar  2 04:34:52 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 2 Mar 2001 10:34:52 +0100
Subject: Lower bounds of lists
References: <2mjt9tcfa0o55ap99p452trjj3u7h3qst6@4ax.com> <mailman.983493962.23590.python-list@python.org>
Message-ID: <97nph901am9@news1.newsguy.com>

"Chris Gonnerman" <chris.gonnerman at usa.net> wrote in message
news:mailman.983493962.23590.python-list at python.org...
    [snip]
> > It seems an odd choice, as humans count from one, and Python in
> > other respects seems quite logical and well thought out.
>
> Humans count from 1.  Hackers count from 0.

Zero proved an elusive concept in the history of the human race
for many thousand of years -- that's why arithmetic lagged for
so long behind other disciplines such as geometry.  Fortunately,
somewhat over a thousand years ago, some Indian theoreticians
at last came up with the concept of zero (helped, no doubt, by
the crucial role that Nothingness plays in their philosophies --
just as much as earlier Greek thinkers were hampered by
Parmenides' insistence that "Not-Being does not exist nor
cannot it exist").

Arabs took it up from them, then Italians from Arabs, then the
rest of Europe -- with decimal notation for numbers, and one
other key innovation, double-entry bookkeeping, arguably an
Italian invention (but it would never have flown without
decimal-notation numbers to ease the sums:-).  The whole
Renaissance, and, later, the flourishing of modern science,
were enabled by the introduction of Zero, and the various
related technologies (e.g., the explosion in finance, banking,
entrepreneurship, which marks the late Italian Middle Ages and
directly ushered in the Renaissance, had double-entry
bookkeeping as one of its key enablers).

Zero works better -- it better reflects underlying arithmetic
reality.  The tens of thousands of years it took human beings
to come to terms with it confirms it's hard for our minds -- just
as, say, reading and writing, or even more, as it got another
few millennia's delay.  Fortunately, memes do spread -- these
days, we get socially exposed to Zero right from our birth,
and so, fortunately, many of us get a precious leg up -- enough
to overcome our biological limitations in the matter.

I agree with Python's design decision in this regard, to cater
more for underlying arithmetical realities than for biological
limitations of human beings.  If Python was designed around the
10th Century, then "Zero" might well be considered far too new,
radical, and unproven a concept for widespread consumption, no
doubt; but, thankfully, a thousand years' worth of history has
widely ameliorated things in this regard, so that a programming
language CAN now afford to better match undelying reality.



> > Starting lists at zero also seems to have had ripple effects,
> > such as
> >
> > range(1,100) giving (1, 2, 3, ... , 99)
>
> This gives me headaches sometimes also; JavaScript does similar things.

"Lower-limit-included, Upper-limit-excluded" is a very good
mantra.  It has all sorts of useful invariants, such as
range(X,Y) having Y-X items rather than, say, Y-X+1 -- that
'+1' would give far worse headaches and off-by-one errors.


Alex





From ab007 at hszk.bme.hu  Tue Mar  6 06:56:43 2001
From: ab007 at hszk.bme.hu (Balint Aradi)
Date: Tue, 6 Mar 2001 12:56:43 +0100
Subject: Python 2.0 on CrayT3E
Message-ID: <Pine.GSO.4.21.0103061230590.20793-100000@ural2>

Hi!

 Does anyone has experience with compiling Python 2.0 on Cray T3E
machines? 

I had problems with following modules:
  ucnhash
  mmapmodule (sys/mman.h unavailable)
  shamodule
  md5

After commenting out these modules the compilation succeeds, but starting
Python results the  error message:

Fatal Python error: Unicode configuration error: sizeof(Py_UNICODE) != 2
bytes
Abort (core dumped)


Thanks,

  Balint



From neilh at scintilla.org  Mon Mar 19 16:19:54 2001
From: neilh at scintilla.org (Neil Hodgson)
Date: Mon, 19 Mar 2001 21:19:54 GMT
Subject: [Pythonwin] Scintilla  ID(L)Eal
References: <bu4cbts8mjliub4iobb2bvkd1u3q4m4392@4ax.com>
Message-ID: <_Jut6.11470$992.69764@news-server.bigpond.net.au>

Fred:
> REGEXP support and "Find in Files" (recursively)
> as in IDLE would be nice !

   PythonWin has a "Find in Files" command - its just a bit hard to find.

   From the "File" menu select "New".
   A "New" dialog pops up with two choices "Python Script" and "Grep".
   Choose "Grep" and OK the dialog.
   Fill in the fields.
   If you want the find to be recursive select "Subdirectories".
   Click OK.
   A new window appears, some disk churning occurs and you are presented
with a list of matches.
   Double click on any of the matches to open the file.

   Another useful tool is the "Across open files" option in the Find dialog
which allows the find to move from the current file to others that are
currently open.

   Regexp support would be good but is unlikely to happen unless a user
cares enough to add the feature. There is some support in Scintilla for
regexp searching but is not turned on currently due to time constraints.
Alternatively, PythonWin could do its own regexp searching based on Python's
re module which is much larger and better than Scintilla's will be when it
is turned on.

   Neil




From phlip_cpp at my-deja.com  Tue Mar 13 12:47:42 2001
From: phlip_cpp at my-deja.com (Phlip)
Date: Tue, 13 Mar 2001 09:47:42 -0800
Subject: PyArg_ParseTuple on steroids?
References: <YObr6.839$dq1.77192@news.pacbell.net>
Message-ID: <I5tr6.11$4U5.21866@news.pacbell.net>

Grant Munsey wrote:

> I have been doing a lot of Python to C library adapters recently.
> Currently I'm generating the code for each Python cover routine with
> a Python program which uses a crude IDL to drive the generator. I
> build the IDL descriptors by hand or with another Python program in
> the case where the API is described in a "regular" way.
> 
> Currently the C source that is output by the generator looks pretty
> much like code that would be generated by hand  ... with
> PyArg_ParseTuple() at the beginning and Py_BuildValue() at the end.
> 
> I was thinking of changing this to have the generator generate a table
> of parameter descriptions for each procedure and one call to an "uber"
> argument handler ... that would check the parameters, get the C values,
> call the function, and convert the results back into a Python return
> value. Kind of a PyArg_ParseTuple(), call function, Py_BuildValue() all
> rolled into one but that could handle many more parameter types than the
> standard routines do.
> 
> This would have the benefit of smaller overall code generation and the
> various parameter handlers would be in one place instead of spread all
> over a large Python program which generates code.
> 
> Anyone done this in the past and have any advice?

You are "adding epicicyles".

Switch to C-style C++, and use the Boost Python Library. We auditioned many 
solutions and skidded to a halt at BPL when we saw how it used C++ style 
reflection to identify our function arguments at C++ compile time. No extra 
languages or IDL files or anything. We can now RefactorMercilessly our 
bindings without risking weird bugs or double-checking everything in 
multiple files.

  www.boost.org

-- 
  Phlip                          phlip_cpp at my-deja.com
============== http://phlip.webjump.com ==============
  --  MCCD - Microsoft Certified Co-Dependent  --


From max at alcyone.com  Thu Mar  8 01:32:08 2001
From: max at alcyone.com (Erik Max Francis)
Date: Wed, 07 Mar 2001 22:32:08 -0800
Subject: "!=" is bad form. Re: sorry....never mind
References: <DZkp6.18811$pF4.3015652@news1.telusplanet.net> <fdlp6.16982$go.2157953@news0.telusplanet.net> <3AA6E308.5B63ECC6@cosc.canterbury.ac.nz> <3AA72253.FB021815@Lugoj.Com>
Message-ID: <3AA72768.30243C29@alcyone.com>

James Logajan wrote:

> Actually, for Python, "!=" IS totally bogus and should be considered
> the
> unusual variant. Unless the logical negation operation has been
> changed from
> "not" to "!" (shucks, version 1.5.2 still does all I need) then "!="
> is not
> suggestive of anything.

It's a sequence of characters, very commonly used in programming
languages to indicate inequality.  Sometimes you actually have to learn
things; not everything can instantly make sense without reading some
documentation.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ What do women want?
\__/ Sigmund Freud
    Alcyone Systems' CatCam / http://www.catcam.com/
 What do your pets do all day while you're at work?  Find out.


From joconnor at cybermesa.com  Thu Mar  1 14:59:37 2001
From: joconnor at cybermesa.com (Jay O'Connor)
Date: Thu, 01 Mar 2001 12:59:37 -0700
Subject: [Q] Question from Python tutorial
References: <Lgxn6.6964$dL4.91439@vixen.cso.uiuc.edu> <3A9EA472.F15AF847@cybermesa.com> <hxxn6.6969$dL4.91744@vixen.cso.uiuc.edu>
Message-ID: <3A9EAA29.8152BDE5@cybermesa.com>


Young-Jin Lee wrote:

> I typed them manually. I dind't copy and paste.

OK, then.  It's still a code formatting issue.

If you typed it into something like the command line inertactive environment,
and not IDLE.  There should be some >>> and ... notation

If this is what you have:

>>> x.counter = 1
>>>while x.counter < 10:
...    x.counter = x.counter + 1
...print x.counter


This will raise a syntax error because the interpreter expects that the line
starting with ... is part of the while block and the print statement is not
properly indented, so it raises a syntax error

Based on the formatting in the tutorial, you should just hit enter on that
second line to tell the interpreter that your done with the block.

Your code should look like this instead

>>> x.counter = 1
>>>while x.counter < 10:
...    x.counter = x.counter + 1
...
>>>print x.counter

This will work properly. Note IDLE on Windows (and Linux  for all I
know...never used it) does not use the ... notation, but does use the >>>
notation
--
Jay O'Connor
joconnor at cybermesa.com
http://www.cybermesa.com/~joconnor

Python Language Forum -
http://pub1.ezboard.com/fobjectorienteddevelopmentpython




From loewis at informatik.hu-berlin.de  Thu Mar  1 06:09:17 2001
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 01 Mar 2001 12:09:17 +0100
Subject: writing a C function to manipulate python lists
References: <jllmqpnclz.fsf@endor.bbb.caltech.edu>
Message-ID: <j44rxdlp9e.fsf@informatik.hu-berlin.de>

Michael Vanier <mvanier at endor.bbb.caltech.edu> writes:

> If anyone has some code lying around that does something like this, could
> they pass it on to me?  I haven't seen this kind of thing in the official
> python docs.  I'd also prefer to use the C representation of python lists
> directly for speed, because using the C API would probably slow things down
> considerably.

I recommend to look at the "Embedding and Extending" tutorial, and the
Python/C API reference. The functions you will need in your code are

- PyArg_ParseTuple to extract the lists from the argument tuple
  (although PyTuple_GetItem will also do)
- PyList_New to create a result list (if necessary)
- PyList_GetItem and PyList_SetItem to manipulate lists
- potentially any other function defined in listobject.h, and
  perhaps other API, depending on you problem (most likely, Py_INCREF
  and Py_DECREF in the right places).

Hope this helps,
Martin



From ws-news at gmx.at  Mon Mar 12 10:43:27 2001
From: ws-news at gmx.at (Werner Schiendl)
Date: Mon, 12 Mar 2001 16:43:27 +0100
Subject: learning C
References: <mailman.983972829.23481.python-list@python.org> <983975751.18833@newsmaster-04.atnet.at> <slrn9adklr.bv9.scarblac@pino.selwerd.nl> <984047775.592522@newsmaster-04.atnet.at> <9896hk$ig8$1@verence.demon.co.uk> <984134612.781779@newsmaster-04.atnet.at> <3AACE22D.3A9EEB12@bt.com>
Message-ID: <984412092.336020@newsmaster-04.atnet.at>

The main use of a higher level language is to hide some of the bare bones
programming from you.
If you use some library in python (maybe one provided through an extension
dll) you do not know how things work in detail and you still use it. Why is
this more safe than the C++ way of things.

I agree that you have more control in C, this is true of course.
Even more control is provided in assembler and finally, nobody hinders you
to type in hex codes yourself ;-).

But I found numerous problems in C with uninitialized data (or
make-sure-it-is-initialized routines invoked everywhere and eating up
performance, except for the location where the problem really appeared in
the end).
I consider the guaranteed contruction of objects in C++ an obvious
advantage.

The decision to add the reference style parameters is not what I like most
about C++ because of the reasons you mentioned.
I tend to avoid non-const reference parameters whereever possible.

Passing arguments by value is almost everywhere avoidable by passing const
reference parameters.
Implicit conversion and construction of (temporary) objects can be avoided
by the implicit keyword, class factories and other means.

C++ design is maybe more difficult than C for small systems.
I think large systems are complex anyway, no matter which language you use.

By defining your own types that represent the problem space and hiding the
bits from the higher levels of the system C++ helps you structure your
system and keep it manageable much better than a set of procedures where
only documentation tells you what belongs together.

Object oriented design is better suited for large systems than procedural
programming from my point of view.
The next step would be component technology, but this leaves the programming
languages issue.

A badly designed system is bad, not matter which language.
In object oriented designs, tools may be able to get more information from
the sources.
But the only things that helps is to improve that thing (or redo it, if this
is cheaper)

regards
werner





From no-spam-popov at bigserver.com  Sun Mar 11 16:38:39 2001
From: no-spam-popov at bigserver.com (Boris Popov)
Date: Sun, 11 Mar 2001 13:38:39 -0800
Subject: Einstein's Riddle
References: <oNPq6.396644$ge4.138479566@news2.rdc2.tx.home.com> <98glrd$1si1a$1@ID-75892.news.dfncis.de> <3AABE712.499CCF4C@home.com> <98gp5b$1u0oi$1@ID-75892.news.dfncis.de>
Message-ID: <98gr9p$1v74k$1@ID-75892.news.dfncis.de>

LoL... Oh well... I see what you mean now... :) But all of them have a
certain pet. Even though we have no clue what kind of pet german owns, we
can assume it's fish cause no one else has it... :)))

-Boris

"Boris Popov" <no-spam-popov at bigserver.com> wrote in message
news:98gp5b$1u0oi$1 at ID-75892.news.dfncis.de...
> "Bruce Elrick" <belrick at home.com> wrote in message
> news:3AABE712.499CCF4C at home.com...
> > I swear there is no evidence that any of them own a fish.  The German
> > owns an emu.
> >
> > Bruce
>
> I am missing something then. I thought my table satisfies all given
> statements and it has fish right under the german...
>
> -Boris
> P.S. What's emu ?
>
> >
> > Boris Popov wrote:
> > >
> > > "Steve Graham" <js.graham at home.com> wrote in message
> > > news:oNPq6.396644$ge4.138479566 at news2.rdc2.tx.home.com...
> > > > Who would be interested in using his/her brain (and his computer) to
> solve
> > > > the following?
> > > >
> > > >
> > > > Steve Graham
> > > >
> > > > ===
> > > >
> > > > Einstein's Riddle
> > > >
> > > > Albert Einstein wrote this riddle this century [ed. 20th century].
He
> > > said
> > > > 98% of the world could not solve it.
> > > >
> > > > There are 5 houses in 5 different colors.  In each house lives a
> person
> > > with
> > > > a different nationality.  The 5 owners drink a certain type of
> beverage,
> > > > smoke a certain brand of cigar, and keep a certain pet.  No owners
> > > > have the same pet, smoke the same brand of cigar or drink the same
> > > beverage.
> > > >
> > > > The question is: "Who owns the fish?"
> > > >
> > > > Hints:
> > > >
> > > > The Brit lives in the red house.
> > > >
> > > > The Sweed keeps dogs as pets.
> > > >
> > > > The Dane drinks tea.
> > > >
> > > > The green house is on the left of the white house.
> > > >
> > > > The green house's owner drinks coffee.
> > > >
> > > > The person who smokes Pall Mall rears birds.
> > > >
> > > > The owner of the yellow house smokes Dunhill.
> > > >
> > > > The man living in the center house drinks milk.
> > > >
> > > > The Norwegian lives in the first house.
> > > >
> > > > The man who smokes Blends lives next to the one who keeps cats.
> > > >
> > > > The man who keeps the horse lives next to the man who smokes
Dunhill.
> > > >
> > > > The owner who smokes Bluemasters drinks beer.
> > > >
> > > > The German smokes Prince.
> > > >
> > > > The Norwegian lives next to the blue house.
> > > >
> > > > The man who smokes Blends has a neighbor who drinks water.
> > > >
> > >
> > > I don't beleive he said that only 2% of the world could solve it. It's
> not
> > > hard, you just need to pay attention to the details. The German owns
the
> > > fish. It was supposed to be a table that I used to layout all the
> details,
> > > but I prefer to post plain-text:
> > >
> > > house         1 2 3 4 5
> > > color            yellow blue red green white
> > > nationality   norweigian dane brit german swede
> > > drink            water tea milk coffee beer
> > > smoke        dunhill blend pall mall prince bluemaster
> > > pet              cats horse birds fish dogs
> > >
> > > -Boris
> > > P.S. Did I have to write software to solve it ? :)
>
>




From smnordby at yahoo.com  Mon Mar  5 08:33:03 2001
From: smnordby at yahoo.com (Steve)
Date: Mon, 05 Mar 2001 05:33:03 -0800
Subject: Questions from a beginner
References: <1103_983792888@breakdown>
Message-ID: <3AA3958F.2CEFAE9A@yahoo.com>

Python is a good choice.  I returned to programming after forgetting
everything I learned 20 years ago in college, dabbled in Java enough to
get confused, then got swallowed by Python and love it.  Python teaches
good programming habits which can be carried over to other languages. 
The problem I am having is that I like Python so much, I have to force
myself to use anything else!

Yes, the Python code you write will be portable beteen Win98 and Linux,
provided you don't use any operating system specific features.  When in
doubt, the documentation that comes with the interpreter specifies when
a feature is not available on all platforms.

Lutz and Ascher's _Learning Python_ is a good book to take you beyond
the tutorial.  Lutz's new edition of _Programming Python_ should be out
soon, and should cover things more in depth.  Both have good snipets of
source code.  If you are going to stick with Python for any length of
time, then _Python Essential Reference_ is indeed THE book.  If you want
to use Tk GUI's, then get John Grayson's _Python and Tkinter
Programming_.

As far as math, if you can remember first year algebra, you are in good
shape to start programming.

HTH,
-SteveN-

Alison Myers wrote:
> 
> Hi. I'm a beginning programmer at a fairly late age and have been looking for a language to learn.
> I have decided against starting with C or C++ because I think they will be too complicated and they seem to be only really needed for writing device drivers and operating systems, which is not what I wat to do.
> Is python a good language to learn first? Will my knowledge help me with other languages later?
> I will be writing for win98 and Linux and I know Python is available on both of these, but would the code be transportable, especially if I write things using tcl (this is probably a long way off yet, but I'd like to know).
> I have links to online tutorials, but where would I find good source code?
> Do I need to buy any books to help me and if so which do you recommend?
> Is there a list of programming projects anywhere or would I get these from a textbook?
> I probably need to brush up on my maths, so what specific topics should I look at?
> TIA
> 
> Alison


From aleaxit at yahoo.com  Thu Mar 29 08:00:43 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 29 Mar 2001 15:00:43 +0200
Subject: PEP 245: Python interfaces
References: <Pine.LNX.4.32.0103222253510.26302-100000@localhost.localdomain> <99o9g4$nh3$0@216.39.170.247> <Pine.LNX.4.21.0103261430360.25260-100000@localhost.localdomain> <j4wv99m503.fsf@informatik.hu-berlin.de>
Message-ID: <99vbm002b8m@news2.newsguy.com>

"Martin von Loewis" <loewis at informatik.hu-berlin.de> wrote in message
news:j4wv99m503.fsf at informatik.hu-berlin.de...
> Michel Pelletier <michel at digicool.com> writes:
>
> > Because that expression list defines inheritance, and interfaces are not
> > inherited.
>
> That depends on your viewpoint. I'd, it defines subtyping, i.e. an
> is-a relationship. An object implementing foo certainly is-a foo.

But is-a is not what inheritance currently asserts in Python: one
may perfectly well inherit for purely 'implementation' purposes, with
no need whatsoever to keep Liskov substitutability (one may inherit
to use methods A, B and C, while overriding method D with a version
that requires more from its arguments -- just for example!).

In C++, one distinguishes "is-a" inheritance from 'just-implementation'
inheritance by having the first kind public, the second kind private.

Fortunately, Python eschews such subtleties (which breed their own
complications); adopting the explicit and very clear distinction that
Java draws between 'extending' (a class) and 'implementing' (an interface)
seems much preferable to me than either totally blurring (in the Python
source) this crucial distinction, or adopting C++-inspired half-remedies
to the is-a/vs/implementation confusion.


Alex





From mcfletch at home.com  Thu Mar 22 16:28:40 2001
From: mcfletch at home.com (Mike C. Fletcher)
Date: Thu, 22 Mar 2001 16:28:40 -0500
Subject: python, perl, lisp, tcl / got a decision to make, can you give me a view point please ? and a joke for python users :)
In-Reply-To: <99dkkd018e0@news1.newsguy.com>
Message-ID: <000001c0b317$2060d310$a9807018@cr706570a>

I agree to the Python and Javascript characterisations (never used Haskell).
I've written production-level Javascript-in-VRML (which adds a whole other
level of aggravation) code for 3 years now, and if you want a workout,
that's it.  I'd also add Prolog to the list of "mind expanders", it's really
fun to figure out how to make "what you want done" into a side-effect of a
unification engine's "logic".

Lisp is fun, but once you get past the program-is-data stuff it's not really
mind-expanding either.  Maybe that's just 'cause I only ever used AutoLISP,
could be there's some consciousness-building stuff somewhere in the more
exotic variants...

Perl's... well, I only used it for about 2 weeks (historical note: I was
reading through a "learn perl in 21 days" (I know, shameful, but it was on
sale, and I was a broke student) book to be able to work on a particular
project, and on the last page of the book was a 1 paragraph description of
Python "This language has a cleaner syntax and may eventually replace
Perl.", or something like that.  That was my introduction to Python.  I
downloaded the then-current interpreter over my 14.4 modem, and I haven't
written a line of Perl since :) .  I consider the book to have been worth
the $20 I paid.)... well, it's Perl.  Some people are into that kind of
sadism, and they _claim_ it brings them closer to the ultimate, but I wonder
about those kinds of claims when we find rooms full of dead monks with
punctuations all over their bodies...

Python, however, is just too simple and straightforward to be of any use to
anyone seeking to pierce the bubble of the universe.
*pop*

-----Original Message-----
From: Alex Martelli [mailto:aleaxit at yahoo.com]
Sent: Thursday, March 22, 2001 2:45 PM
To: python-list at python.org
Subject: Re: python, perl, lisp, tcl / got a decision to make, can you
give me a view point please ? and a joke for python users :)


"Yvon Boulianne" <mysticsguy at yahoo.ca> wrote in message
...
Unfortunately, Python is far too good a language to serve this
purpose.  It's so simple you won't get much mental exercise
learning it, and so powerful it will make it too easy for you to
write all kinds of programs.  For your purpose [2],
...
or javascript (chaotically and ramblingly hard
to learn, yet wimpy enough that using it to write significant
programs is going to be serious exercise each and every time).
...




From kzaragoza at mediaone.net  Wed Mar 14 23:52:13 2001
From: kzaragoza at mediaone.net (Kris J. Zaragoza)
Date: Thu, 15 Mar 2001 04:52:13 GMT
Subject: Who's minister of propaganda this week?
References: <B98041329020E007.7DA84CAD9D0562AB.817BBC572F480869@lp.airnews.net> <98nu1m$12e@dispatch.concentric.net> <660ACCA1E4816F7E.C359CDF4168EF3FB.3B6B05538D619770@lp.airnews.net> <3AAFA13D.76E32C3F@divalsim.it>
Message-ID: <slrn9b0ijs.ccb.kzaragoza@kzaragoza.ne.mediaone.net>

On Wed, 14 Mar 2001 17:50:05 +0100, Nicola Musatti
<objectway at divalsim.it> wrote:
>Hi,
>I'm a Python curious not yet sold to the advantages of typeless
>languages; I tend to think that typelessness is suited to a set of tasks
>that has only a limited overlap with the tasks for which static typing
>is preferable. I have a question and a provocation for you.

Just a minor nit:  The languanges that are being discussed are not
actually typeless.  They are typed, but dynamically.  Every object in
Python, for example, has some type associated with it, and as a
result, the semantics that follow from that type.  You cannot thus put
a string in the middle of a mathematical expression.  Relatively few
of the "scripting" languages in popular use today are weakly typed or
typeless (Perl and TCL come to mind.)

>
>Cameron Laird wrote:
>> 
>> In article <98nu1m$12e at dispatch.concentric.net>,
>> Phlip  <phlip_cpp at my-deja.com> wrote:
>>                         .
>> >Robert C. Martin meant two things. A> If you write wall-to-wall unit tests
>> >when you code, if you use a typeless language you will add bullet-proof
>> >features faster than if you use a statically typed language, even though
>> >the former has less built-in error checking at type conversion time. The
>> >latter supports refactoring easier.
>
>But those things that the compiler used to check for you may well have
>become new tests for you to write: if the trade-off between the decrease
>in actual code size and the increase in test number is negative, even
>the better support for refactoring turns out to be an illusion. 

But this assumes that a reduction in the number of lines of code is
the goal.  It is not.  The goal is to build a robust system as quickly
as possible in the face of constantly changing requirements.
Refactoring and constantly running unit tests are two techniques that
allow for this speed and robustness.  Having the compiler check type
semantics doesn't contribute nearly as much.

Then, of course, there is the matter of how expressive and legible the
resulting code is.  See below for more on this.

>
>[...]
>>   Expressiveness ... is a far better aid to
>>   correctness ...
>
>Here I'm puzzled, but maybe it's only a "language" problem. What do you
>mean by "Expressiveness"? I ask because until now I thought
>expressiveness was an advantage of strongly typed languages (namely,
>C++) over typeless ones.
>
>Best regards,
>Nicola Musatti

Expressiveness means different things to different people.  There is
(currently) no objective standard by which it can be measured.  My
personal take on expressiveness is fairly simple:  How easily can I
express my data, models, and algorithms without getting bogged down in
language details?

Consider C++ as an example.  I actually like C++, despite all its
warts.  It's an incredibly powerful language, and allows for some very
efficient code generation in resource constrained environments.  I
like the feeling of being close to the machine when writing C++ code.
However, I don't consider it a very expressive language.  An
inordinately large amount of any C++ source file is dedicated to
either bookkeeping code for the type system, or various other
syntactical details.  This extra code results in clutter that does not
necessarily---in fact, rarely---contribute to clearly expressing the
intent of the code.

Consider Python as the contrary example.  (This is c.l.python, after
all.)  Python provides a highly interactive environment, allowing one
to type code in at a prompt and have it immediately executed without
need for a potentially lengthy compile and link cycle.  This allows
for a very different style of programming.  As you said above, there
are problem domains where this is extremely appropriate.  Exploratory
programming and algorithmic experimentation are, I believe, two such
areas.  These are, however, a critical part of any non-trivial
project.  Understanding the problem domain, properly modelling data,
and selecting the most appropriate algorithms can provide for enormous
benefits.  The best way to do these things is often to simply
experiment until things fall into place.  Python allows for a great
deal of experimentation in a very short amount of time, far shorter
than with C++.  In experimenting with more alternatives, you may be
able to find the most optimal selection of algorithms and data models
to solve the problem at hand.  This translates to a significant
competetive advantage in the business world.  The right selection of
algorithms could result in orders of magnitude increases in speed as
compared to competitors.  Likewise, the correct selection of data
models and architectures can lead to significantly greater
flexebility, allowing for much easier adaptation whenever the
requirements of the system change.  

Of course the only thing missing, then, is the robustness of the
system.  As others have already said, however, the judicious
application of unit tests (especially automated unit tests that are
regularly run) helps achieve this robustness at a level that static
type checking alone will never achieve.

I will admit that this isn't a formal proof.  Nor is it, one could
say, a strong argument.  I am, quite literally, half asleep as I type
this, so I hope you will forgive the lack of well-structured
arguments.  However, I believe that what I've stated could prove to be
good food for thought.

Salud!

Kris

-- 
Kris J. Zaragoza           | "Unfortunately, most people can't out-think a
kzaragoza at mediaone.net     | grapefruit."  --Jon Bodner


From francois.granger at free.fr  Sat Mar 17 11:20:43 2001
From: francois.granger at free.fr (=?ISO-8859-1?Q?Fran=E7ois_Granger?=)
Date: Sat, 17 Mar 2001 16:20:43 GMT
Subject: Python style questions
References: <98tb91$9t9$1@saltmine.radix.net> <98tdu1$cfk$1@panix2.panix.com> <1eqdo5u.1xxmzfi2kjk0qN%francois.granger@free.fr> <98u24p0a98@news2.newsguy.com>
Message-ID: <1eqfb5l.10gfjcu1qi9xrwN%francois.granger@free.fr>

Alex Martelli <aleaxit at yahoo.com> wrote:

> If you give us a specific, similarly-small example that
> you dream to be able to code with a switch (and are
> presumably now coding with if/elif), we can help you
> translate this into some form of dispatching...

Thanks for the proposal. And let expose my realy bad coding style :-)
(I am not a programmer, and most if not all I do in Python is at home,
by night, and for my pleasure)

In order to exercise some Neural Network and Genetic Algorithm that I am
currently trying to code, I wrote a simple TicTacToe game. The idea is
that I want to have the following possibilities:
- one player can choose a token between 'X' and 'O'
- games can be human against one of NN or GA or NN against GA or even
two different NN or GA against each other. These NN or GA are handled by
the player class wich in turn call the corresponding class.
- one of them is the starter of the game.

To call a non human player, I have a simple class 'player' with a
methode play() wich return the move as a tuple of integer between 0 & 2.
It has also to handle the feed back about win or lost to allow for NN or
GA to "learn".

I am currently on the following dispatch issue wich is far from corect:

def game():
    print 'Starting game.'
    a = tictactoe() # create a new game
    ans = raw_input('Who start (h/c) : ')
    if  (ans[0]=='q' or ans[0]=='Q'):
        return
    human = userinput   # proc to handle human interaction
    c = player('X') # create a machine player
    computer = c.play
    if (ans[0]=='h' or ans[0]=='H'):
        first, second = human, computer
    else:
        first, second = computer, human
    while 1:
        if a.turn % 2:
            move, token = second(a)
        else:
            move, token = first(a)
        a.valid(move, token)
        a.display()
        win = a.won()
        if win == 'tie':
            print 'tie'
            # call player with tie
            # exit game
        elif win == 'X':
            print 'X'
            # call player with X
            # exit game
        elif win == 'O':
            print 'O'
            # call player with O
            # exit game
        else:
            continue



-- 
Une faq de frj: http://faq.jardin.free.fr/, pr?s de l'encyclop?die
http://nature.jardin.free.fr/
MacPython beginners http://francois.granger.free.fr/MacPython/faq/


From dbrueck at edgix.com  Fri Mar  9 09:14:09 2001
From: dbrueck at edgix.com (Dave Brueck)
Date: Fri, 9 Mar 2001 07:14:09 -0700
Subject: "!=" is bad form. Re: sorry....never mind
In-Reply-To: <3AA86D0F.E36B0BDC@Lugoj.Com>
Message-ID: <BLEOJGEMBNBNEAKMIFDHEELCPLAA.dbrueck@edgix.com>

> [mailto:python-list-admin at python.org]On Behalf Of James Logajan
> Sent: Thursday, March 08, 2001 10:42 PM
> To: python-list at python.org
> Subject: Re: "!=" is bad form. Re: sorry....never mind
> Ironically, I like C and have been coding in C on and off for it
> seems just
> about forever (well, 18 some years I think). And I find programming in C a
> mostly pleasurable experience. Yet as I stated earlier, for Python "<>"
> feels "righter" to me than "!=". I suspect it is a triviality for both
> programmers to understand, and the language to support, both forms. While
> deprecating "<>" in Python would not mean the end of civilization
> as we know
> it, it would be a damn shame anyway.
>
>
> (a > b) or (a == b)  --->  (a >= b)
> (a < b) or (a == b)  --->  (a <= b)
> (a < b) or (a >  b)  --->  (a <> b)

I don't really care if Python keeps <>, but the reason I don't like it is
because of the example listed above: to me seeing "<>" implies something
about the ordering of a and b, namely (a < b) or (a > b), even when such an
ordering does not exist (e.g. - it doesn't make sense to say one instance of
a cooking recipe object is "greater than" or "less than" another one (unless
maybe you order them by yumminess), but it does make sense to say they
"aren't equal"). I know <> doesn't _really_ imply any such thing in Python,
but still..

-Dave




From syver at NOSPAMcyberwatcher.com  Tue Mar 27 17:32:18 2001
From: syver at NOSPAMcyberwatcher.com (Syver Enstad)
Date: Wed, 28 Mar 2001 00:32:18 +0200
Subject: daemon app in win32
References: <99nk49$a2q$02$1@news.t-online.com> <mailman.985619314.10299.python-list@python.org> <99nsn3$q39$06$1@news.t-online.com>
Message-ID: <99r4hq$sq3$1@troll.powertech.no>

> Hmm, seems like this is only available under NT. Is there a more general
> way to accomplish this task independently of the underlying
Windows-version?

I remember reading some time about rudimentary supprort for services under
Win9x so maybe you can use one version after all. Check out the platform
sdk, or msdn for details.




From wfocke at phoenixdsl.com  Thu Mar  8 17:17:20 2001
From: wfocke at phoenixdsl.com (Warren Focke)
Date: 8 Mar 2001 14:17:20 -0800
Subject: How to installing NumPy?
References: <3AA57C0D.BBC5E956@itu.edu.tr> <9842vc$4bb$1@mothership.telocity.com> <3AA632AA.7D15888C@itu.edu.tr>
Message-ID: <9890dg$cg1$1@mothership.telocity.com>

Doesn't look familiar.  I'd suggest rebuilding and reinstalling Python
and NumPy now that you know which packages you need and what to do with
them.

w

In article <3AA632AA.7D15888C at itu.edu.tr>,
Nuzhet Dalfes wrote:
>I'll appreciate some more help; is this a known problem?
[...]
>In file included from /usr/local/include/python2.0/Python.h:54,
>                 from Src/_numpymodule.c:1:
>/usr/local/include/python2.0/pyport.h:390:2: #error "LONG_BIT definition appears wrong for platform (bad gcc config?)."
[...]
>gcc -g -O2 -Wall -Wstrict-prototypes -fpic -IInclude -I/usr/local/include/python2.0 -c Src/_numpymodule.c -o build/temp.linux-i686-2.0/Src/_numpymodule.o
>error: command 'gcc' failed with exit status 1
-- 
Wrapping up her appearance with rubber, you can feel more comfortable.
 -- directions for a set of Chinese binoculars


From aahz at panix.com  Thu Mar 15 18:10:47 2001
From: aahz at panix.com (Aahz Maruch)
Date: 15 Mar 2001 15:10:47 -0800
Subject: PEP 6: Patch and Bug Fix Releases
References: <mailman.984662760.24168.clpa-moderators@python.org>
Message-ID: <98ri5n$3nu$1@panix2.panix.com>

[Don't argue with me, I'm just the messenger.  ;-) ]

Forwarded message:
> Message-Id: <200103152242.f2FMgTZ02863 at clove12.loc.clove.com>
> To: aahz at panix.com (Aahz Maruch)
> Subject: Re: PEP 6: Patch and Bug Fix Releases 
> In-Reply-To: Message from aahz at panix.com (Aahz Maruch) 
>    of "14 Mar 2001 22:14:54 PST." <mailman.984662760.24168.clpa-moderators at python.org> 
> Date: Thu, 15 Mar 2001 17:42:28 -0500
> From: Mike Howard <mike at clove.com>
> 
> My message was cc'ed to python-announce-list and I don't have news access.
> Please post to c.l.py - if you think it is relevant; else /dev/null.
> 
> Thanks, Mike Howard <mike at clove.com>
> 
> Msg follows:
> 
> I believe that you are being too stringent in prohibiting enhancements
> from bug-fix patches.  It is sufficient that the bugs be fixed and
> no incompatibilities be introduced.  Enhancements can always be
> ignored, whereas bugs must be worked around or squashed.
> 
> I believe that by distinguishing between feature enhancements and bug
> fixes, you may discourage a bug fix from code which also includes a
> feature enhancement.
> 
> I would suggest simply outlawing incompatible changes, but allowing
> only bug fixes and ignoring any enhancements which come along with
> the fix.  And, of course, dropping the `p' suffix.
> 
> My 2 cents,
> -- 
> Mike Howard <mike at clove.com>


From aa8vb at yahoo.com  Thu Mar 29 08:38:27 2001
From: aa8vb at yahoo.com (Randall Hopper)
Date: Thu, 29 Mar 2001 08:38:27 -0500
Subject: Whats in your $PYTHONSTARTUP$ ?
In-Reply-To: <FJ2w6.5358$GD.2921338@typhoon.snet.net>; from p.magwene@snet.net
 on Tue, Mar 27, 2001 at 10:55:41AM -0500
References: <FJ2w6.5358$GD.2921338@typhoon.snet.net>
Message-ID: <20010329083827.B4611551@vislab.epa.gov>

Paul Magwene:
 |
 |Other than import statements, what little programming gems do you folks have
 |in your startup files?  I thought it would be useful to draw on the
 |collective knowledge of the Python community.  If we get enough good ideas
 |it might be worth sticking up on a web page, or even distributing (in the
 |demos or Tools section) with a later release.

Just a little fragment to colorize my python prompts.  Nowadays, I'm always
on a VT100 color capable terminal (recent xterm, aterm, FreeBSD/Linux
console, etc.).


     import sys,os
     if os.environ.get( 'TERM' ) in [ 'xterm', 'vt100' ]:
       sys.ps1 = '\001\033[0:1;32m\002>>> \001\033[0m\002'
       sys.ps2 = '\001\033[0:31m\002... \001\033[0m\002'


Randall

-- 
Randall Hopper
aa8vb at yahoo.com



From gabriel_ambuehl at buz.ch  Tue Mar 20 14:30:12 2001
From: gabriel_ambuehl at buz.ch (Gabriel Ambuehl)
Date: Tue, 20 Mar 2001 20:30:12 +0100
Subject: Various issues regarding thread
Message-ID: <132445453868.20010320203012@buz.ch>

-----BEGIN PGP SIGNED MESSAGE-----

Hello,
I've been developping a small resource monitoring syste, in Python in
the
last time and got some questions regarding threading.thread. I figured
it would probably be the easiest way (forking of new processes seems
very unwise as RAM to useage) to have a dedicated thread for
every resource that has to be monitored so it would be possible to
just have them sleep between the work and let Python do the
scheduling. Now however, I thought it could be cool to have the thing
using pickle to dump its state before it exits so it could continue
its work right where it left last time (not yet sure if this makes
much sense though ;-). The only problem there is that pickle doesn't
appear to be working on lock objects which the thread subclasses
inherently got. So is there any possibility to serialize a class that
has been inherited from threading.thread?

Another thing I couldn't figure out is how one could stop a thread
reliably. I first thought one could have a flag that gets checked by
the run method every now and then but then found this to be to slow as
it could well be that a thread sleeps for several minutes before it
notices that the flag had been set and thus should have exited some
time ago. What am I overlooking?







Best regards,
 Gabriel

-----BEGIN PGP SIGNATURE-----
Version: PGP 6.0.2i

iQEVAwUBOrehucZa2WpymlDxAQHp/AgAzVT0NZbjq3jASS36tCetRMYGunuprDCN
EsIJYgn3/lMMgVOsYyPhio+54D3pqJiEC70fTF3Oixs3COphkzODan12AhHwY/56
cB//mLfMyHkwQ3MuIYaNMGrB2gW36Dka5Fn1EJRm4jMZM1yTS9urvKR5RWiH/Q4Y
YXOJaco2RIvsay3e7f+ZgzC4zPM1hEakDs8DtS+JkXfu1rtMGA28GR1nug7JCaLH
UTDOhcv5eL/bleZKH6GxUFsTEehqQIhxJm7VLK4ODwEILG2QiDo11Xggjv3IuZfX
0lVe9zRRbt9cJiMfbJDFSfsJCBEzSZieJdZDr9tFJ0R8B0Svh0ORmQ==
=uF7d
-----END PGP SIGNATURE-----





From emile at fenx.com  Thu Mar 29 15:08:02 2001
From: emile at fenx.com (Emile van Sebille)
Date: Thu, 29 Mar 2001 12:08:02 -0800
Subject: Quoted identifiers in Python?
References: <3ac3387f$0$12248$cc9e4d1f@news.dial.pipex.com> <3AC35C84.FBED128@alcyone.com> <3ac38552$0$15025$cc9e4d1f@news.dial.pipex.com>
Message-ID: <9a04sn$34lka$1@ID-11957.news.dfncis.de>

Well, how would this work?

a = ."Ivory the engine"
ax = obi.a

What is ax?

--

Emile van Sebille
emile at fenx.com

---------
"Eddie and Babs" <kca17 at dial.pipex.com> wrote in message
news:3ac38552$0$15025$cc9e4d1f at news.dial.pipex.com...
> Ahem... sorry, I missed the crucial word out there: "__dict__"!
> I meant that
>
>     x = obj."Ivor the engine"
>
> ...would be the same as:-
>
>     x = obj.__dict__["Ivor the engine"]
>
> The point I was really making was that since the object's attributes are
> stored in a dictionary, there is no reason why you shouldn't have any
string
> you like as the key (you can, after all, do that with any other
dictionary).
> The ability to use arbitrary strings as attribute names (as opposed to
just
> the usual alpha-num-underscore characters) DOES have its uses when you are
> allowed to re-define the attribute access "dot" operator.
>
>
> ----------
> In article <3AC35C84.FBED128 at alcyone.com>, Erik Max Francis
> <max at alcyone.com> wrote:
>
>
> > Eddie and Babs wrote:
> >
> >> NOW... would this be a good thing for Python to have in the future? It
> >> would
> >> make element access consistent with subscripting:-
> >>
> >>         x = obj."Ivor the engine"
> >>
> >>             ...would be the same as
> >>
> >>         x = obj["Ivor the engine"]
> >>
> >> ..no?
> >
> > I would hope they would not be the same, since one invokes __getattr__
> > while the other invokes __getitem__.  They're different things, they
> > shouldn't be forced to be the same thing.
> >
> > --
> >  Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
> >  __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
> > /  \ Traveler, there are no paths.  Paths are made by walking.
> > \__/ Antonio Machado
> >     REALpolitik / http://www.realpolitik.com/
> >  Get your own customized newsfeed online in realtime ... for free!




From f8dy at yahoo.com  Fri Mar  2 15:13:20 2001
From: f8dy at yahoo.com (Mark Pilgrim)
Date: Fri, 2 Mar 2001 15:13:20 -0500
Subject: Thinking Pythonically (was Re: gah! I hate the new string syntax)
References: <mailman.983485923.1136.python-list@python.org> <B7In6.16281$Qb7.2658902@newsb.telia.net> <97ohh2$polt0$1@ID-77331.news.dfncis.de> <97opsb$nhu$1@nntp6.u.washington.edu>
Message-ID: <97oupq$qijrh$1@ID-77331.news.dfncis.de>

"Donn Cave" <donn at u.washington.edu> wrote in message
news:97opsb$nhu$1 at nntp6.u.washington.edu...
> Quoth "Mark Pilgrim" <f8dy at yahoo.com>:
> | 1. I used Python 1.5.x very little; I really only ramped up when 1.6/2.0
was
> | released.  I *like* list comprehensions.  I prefer them over map/lambda
> | because of scope issues, and I prefer them over for loops because they
force
> | me to think Pythonically, and they emphasize Python's strengths (less
code,
> | richer data structures).  To my eyes, list comprehensions fulfill
Python's
> | promise of being readable pseudo-code, not readable line noise.  YMMV.
>
> Pseudo-code is the language you don't have to know, to understand,
> right?  Familiar symbols, unambiguous syntax.  Could you articulate
> your thoughts on this, how do list comprehensions improve on their
> alternatives, there?

Well, list comprehensions are unambiguous in the sense that they are a
specific syntax -- they map a list into another list, and (possibly) filter
out some elements along the way.  Using a for loop and an if statement to do
this is ambiguous in the sense that you could use that physical code
structure for anything, not just for accomplishing the logical goal of
mapping a list.  Glancing at a for loop, you couldn't just say "this bit of
code maps a list into another list" without examining the code thoroughly
and making sure that's really what it was doing, that there weren't any side
effects along the way, etc.  Glancing at a list comprehension, you instantly
know what it is, because the syntax is specific instead of general.  (Note
this does not mean you know what it *does*, just that you can reasonably
rule out lots of things that it doesn't do.)

> What does "think Pythonically" mean to you?  What are you thinking of
> when list comprehensions make you cite "less code, richer data structures"
> as a strength of Python?  Can you talk us crotchety old-timers through
this?

For example, I recently had to write a Python script that locked ClearCase
VOBs (it is not required to understand what those are to follow this
example).  You can do this on the command line: ("cleartool lock vob1 vob2
vob3..."), but the only way to get the list of VOBs is with another command
line tool ("cleartool lsvob -l"), which returns the VOB name on one line
("Tag: <vob>") plus lots of information we don't care about, like this:

Tag: /vobs/intranet
  Global path: ...
  Server host: ...
Vob on host: ...
...

Tag: /vobs/project
  Global path: ...
  Server host: ...
Vob on host: ...
...

If you think of this output as one long string, you can write a for loop
that searches for each "Tag:" and parses out the rest of the line to get a
single VOB name.  But don't do that, that's Visual Basic thinking!
(Actually in my case it's Powerbuilder thinking, but it's just as bad.)
Instead, I force myself to think of the output as a list of strings, where
each line is an element.  This is very easy using readlines():

>>> output = os.popen("cleartool lsvob -l").readlines()

Now I think: OK, I have a list of lines, but I only want some of them -- the
ones starting with "Tag:" -- and I want to throw the rest away.  I could do
this with a for loop, but that's (a) more typing, and (b) more ambiguous
(see above).  Since I have the lines in a list, I can use a list
comprehension:

>>> [line for line in output if line[:4]=='Tag:']
['Tag: /vobs/intranet\\012', 'Tag: /vobs/project\\012']

Now I think: OK, I have a list of the lines I want, but I just want the
name, not the "Tag:" prefix or all the whitespace.

>>> vobs = [line.replace('Tag:', '').strip() for line in output if
line[:4]=='Tag:']
>>> vobs
['/vobs/intranet', '/vobs/project']

Now I think: great, I have a list of VOBs, now I need them space-separated
and integrated into the rest of my command line:

>>> "cleartool lock %s" % " ".join(vobs)
'cleartool lock /vobs/intranet /vobs/project'

And that's what I call thinking Pythonically.  YMMV.

-M
You're smart; why haven't you learned Python yet?
http://diveintopython.org/





From Laura_01 at MailAndNews.com  Thu Mar 15 16:46:28 2001
From: Laura_01 at MailAndNews.com (Laura Lewin)
Date: Thu, 15 Mar 2001 16:46:28 -0500
Subject: Python Programming on Win32 (was: Magnitude of the wx* market (was: Python In A Nutshell - suggestions))
Message-ID: <3AB16A79@MailAndNews.com>

Hi,
In terms of a new edition of Python Programming on Win32, we are planning 
one, 
but it is not just around the corner.  We want to make sure to cover .NET 
and 
the technology is far from ready for prime time. 
To try to address your other concerns (your mini rant), we are planning 
application-oriented books in the Python space, but these will be for 
developers.  I hear your concerns about books that cover Python apps, but 
ORA's main focus has always been for the developer, and that's where you'll 
see most of our books.  I can't speak for all of the other 
publishers/authors 
out there.  
Please do check out The Python Cookbook 
(http://pythoncookbook.activestate.com/)though.  You mention an interest in 
system administration, databases, network programming.  These are all 
sections 
in the Cookbook.  Feel free to contribute recipes, or take any with you!!
Laura
LLewin at oreilly.com

>===== Original Message From "Don Tuttle" <tuttledon at hotmail.com> =====
>"Laura Lewin" <Laura_01 at MailAndNews.com> wrote in message
>news:3AB26DAC at MailAndNews.com...
>> Hi,
>> Sure, we'll take this suggestion into consideration (on expanding wxPython
>> coverage) when we publish the next edition of "Python Programming on
>Win32".
>
>When do you plan to do that?
>
>And please greatly expand the chapter on system administration!  Better yet,
>create an entire book along the lines of Dave Roth's excellent "Win32 Perl
>Scripting: The Administrators Handbook". In fact many sections of  "Python
>Programming on Win32"  need to be entire books, not just quick intros.
>
>This leads me to a mini rant that has been building for some time now.  I'll
>try to contain myself ;-)
>
>Over the last year I've spent several hundred dollars on NON Python books in
>order to learn Python.  Or more precisely, what to do with Python.  It's not
>that Python can't do it, it's that I don't know how to do it, or for that
>matter, that I can do "it".  (I'm finding new "its" everyday.)  And most of
>the "How To" books are written for other languages. So I have many books on
>VBScript, JScript, WSH, Perl and Batch files. And this is just for Admin
>work.  I haven't even started trying to use Python with databases.- Yet- But
>I will.  (And then there's that network programming stuff...)
>
>So, all you authors and Publishers out there...Put out more topic specific
>Python "How-To" books.  Let's liberate Python from the hands of highly
>skilled developers and give it to your average power user! Power to the
>Python! <wink>
>
>Don
>
>PS  And don't forget all those pissed off VB programmers that are ripe for
>the picking!
>
>
>
>



From martin.skott at propylon.com  Wed Mar  7 09:14:24 2001
From: martin.skott at propylon.com (Martin Skott)
Date: 07 Mar 2001 14:14:24 +0000
Subject: XML Module problems
References: <mailman.982853044.8344.python-list@python.org> <memo.20010306181753.34635C@digitig.compulink.co.uk>
Message-ID: <y9uhy8cf.fsf@propylon.com>

digitig at cix.co.uk (Tim Rowe) writes:

> That's what I have with my (ActiveState) 2.0 installation. 
> xml.parsers.expat is very like pyexpat but with some subtle differences 
> that make the name change appropriate (for example, IIRC expat returns 
> attributes as a dictionary rather that pyexpat's list -- expat's version 
> makes much more sense to me). You might also want to look at pyxie which 
> will niftily dynamically bind members calls to element names -- I'm sure 
> Sean will turn up with a URL in a moment, but a Google search finds it 
> easily enough.

Sean isn't currently reading newsgroups, but I hope my answer will do
:-) http://www.pyxie.org is the site for Pyxie. You should however be
aware that some changes in later editions of pyexpat breaks pyxie. The
PythonXML rpm on the "XML Processing with Python" companion CD works
fine however.
 
-- 
Martin Skott
Software Developer
Propylon - Enabling Universal Mobility
http://www.propylon.com
Tel: +353 096 37245
Mobile: +353 087 9680370


From svpeters at t-online.de  Sun Mar 11 18:26:29 2001
From: svpeters at t-online.de (Sven Peters)
Date: Mon, 12 Mar 2001 00:26:29 +0100
Subject: Einstein's Riddle
References: <oNPq6.396644$ge4.138479566@news2.rdc2.tx.home.com> <3AABE712.499CCF4C@home.com> <98gp5b$1u0oi$1@ID-75892.news.dfncis.de> <98gr9p$1v74k$1@ID-75892.news.dfncis.de>
Message-ID: <98h1j8$sb9$05$1@news.t-online.com>

Boris Popov wrote:

> LoL... Oh well... I see what you mean now... :) But all of them have a
> certain pet. Even though we have no clue what kind of pet german owns, we
> can assume it's fish cause no one else has it... :)))
We may also assume, there is no fish, it ist never mentioned in any hint, 
and the german owns a "use your imagination". One thing we know, he doesn?t 
own a cat, horse bird or dog.

-- 
Sven Peters


From rich_somerfield at tertio.com  Fri Mar  2 10:27:22 2001
From: rich_somerfield at tertio.com (Rich Somerfield)
Date: Fri, 2 Mar 2001 15:27:22 -0000
Subject: Newbie createing .PYC for import
References: <3a9e83fa$1@news1.dbtech.net>
Message-ID: <aXOn6.1593$925.86821@news6-win.server.ntlworld.com>

try (in shell / dos prompt) :
python compile.py <filename>

e.g.
python compile.py *

Regards
Rich


in the relavent dir with the relavent wildcard / filename
Larry Bates <lbates at syscon-computers.com> wrote in message
news:3a9e83fa$1 at news1.dbtech.net...
> I have spent the last couple of hours (unsuccessfully) attempting to
figure
> out how to generate a .PYC file for a module that I've defined.
Everything
> appears to imply that it's "automatic", but I'm not getting anywhere.  I
> have a module and I wish to generate a .PYC file so that I can import it
> into future projects.  Can anyone shed any light on how to accomplish
this?
>
> -Larry
>
>




From aahz at panix.com  Fri Mar  2 19:04:28 2001
From: aahz at panix.com (Aahz Maruch)
Date: 2 Mar 2001 16:04:28 -0800
Subject: the demise of 'from foo import * and its implications?
References: <mailman.983572744.12626.python-list@python.org>
Message-ID: <97pcec$flj$1@panix2.panix.com>

In article <mailman.983572744.12626.python-list at python.org>,
Sean 'Shaleh' Perry <shaleh at valinux.com> wrote:
>On 02-Mar-2001 Tim Peters wrote:
>>
>> What gave you the idea that "import *" is going away?
>
>/me needs to take a week off apparently.  Either that or stop believing
>anything he reads.

More precisely, you should pay attention to what you're reading.
Because of the new nested scopes, using import * in a class or function
can fail; however, it is *documented* to fail.  On the gripping hand,
some scaffolding has been added so that import * will fail in fewer
cases, and in any event nested scopes are optional in 2.1.
-- 
                      --- Aahz (Copyright 2001 by aahz at pobox.com)

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

Nostalgia just ain't what it used to be


From aahz at panix.com  Thu Mar 15 01:14:54 2001
From: aahz at panix.com (Aahz Maruch)
Date: 14 Mar 2001 22:14:54 -0800
Subject: PEP 6: Patch and Bug Fix Releases
Message-ID: <mailman.984662760.24168.clpa-moderators@python.org>

[posted to c.l.py.announce and c.l.py; followups to c.l.py; cc'd to
python-dev]

Okay, folks, here it is, the first draft of the spec for creating Python
maintenance releases.  Note that I'm not on python-dev, so it's probably
better to have the discussion on c.l.py if possible.

            PEP: 6
          Title: Patch and Bug Fix Releases
        Version: $Revision: 1.1 $
         Author: aahz at pobox.com (Aahz)
         Status: Draft
           Type: Informational
        Created: 15-Mar-2001
   Post-History:
     _________________________________________________________________
   
  Abstract
  
    Python has historically had only a single fork of development,
    with releases having the combined purpose of adding new features
    and delivering bug fixes (these kinds of releases will be referred
    to as "feature releases").  This PEP describes how to fork off
    patch releases of old versions for the primary purpose of fixing
    bugs.

    This PEP is not, repeat NOT, a guarantee of the existence of patch
    releases; it only specifies a procedure to be followed if patch
    releases are desired by enough of the Python community willing to
    do the work.


  Motivation
  
    With the move to SourceForge, Python development has accelerated.
    There is a sentiment among part of the community that there was
    too much acceleration, and many people are uncomfortable with
    upgrading to new versions to get bug fixes when so many features
    have been added, sometimes late in the development cycle.

    One solution for this issue is to maintain old feature releases,
    providing bug fixes and (minimal!) feature additions.  This will
    make Python more attractive for enterprise development, where
    Python may need to be installed on hundreds or thousands of
    machines.

    At the same time, many of the core Python developers are
    understandably reluctant to devote a significant fraction of their
    time and energy to what they perceive as grunt work.  On the
    gripping hand, people are likely to feel discomfort around
    installing releases that are not certified by PythonLabs.


  Prohibitions
  
    Patch releases are required to adhere to the following
    restrictions:

    1. There must be zero syntax changes.  All .pyc and .pyo files
       must work (no regeneration needed) with all patch releases
       forked off from a feature release.

    2. There must be no incompatible C API changes.  All extensions
       must continue to work without recompiling in all patch releases
       in the same fork as a feature release.


  Bug Fix Releases
  
    Bug fix releases are a subset of all patch releases; it is
    prohibited to add any features to the core in a bug fix release.
    A patch release that is not a bug fix release may contain minor
    feature enhancements, subject to the Prohibitions section.

    The standard for patches to extensions and modules is a bit more
    lenient, to account for the possible desirability of including a
    module from a future version that contains mostly bug fixes but
    may also have some small feature changes.  (E.g. Fredrik Lundh
    making available the 2.1 sre module for 2.0 and 1.5.2.)


  Version Numbers
  
    Starting with Python 2.0, all feature releases are required to
    have the form X.Y; patch releases will always be of the form
    X.Y.Z.  To clarify the distinction between a bug fix release and a
    patch release, all non-bug fix patch releases will have the suffix
    "p" added.  For example, "2.1" is a feature release; "2.1.1" is a
    bug fix release; and "2.1.2p" is a patch release that contains
    minor feature enhancements.


  Procedure
  
    XXX This section is still a little light (and probably
    controversial!)

    The Patch Czar is the counterpart to the BDFL for patch releases.
    However, the BDFL and designated appointees retain veto power over
    individual patches and the decision of whether to label a patch
    release as a bug fix release.

    As individual patches get contributed to the feature release fork,
    each patch contributor is requested to consider whether the patch
    is a bug fix suitable for inclusion in a patch release.  If the
    patch is considered suitable, the patch contributor will mail the
    SourceForge patch (bug fix?) number to the maintainers' mailing
    list.

    In addition, anyone from the Python community is free to suggest
    patches for inclusion.  Patches may be submitted specifically for
    patch releases; they should follow the guidelines in PEP 3[1].

    The Patch Czar decides when there are a sufficient number of
    patches to warrant a release.  The release gets packaged up,
    including a Windows installer, and made public as a beta release.
    If any new bugs are found, they must be fixed and a new beta
    release publicized.  Once a beta cycle completes with no new bugs
    found, the package is sent to PythonLabs for certification and
    publication on python.org.

    Each beta cycle must last a minimum of one month.


  Issues To Be Resolved
  
    Should the first patch release following any feature release be
    required to be a bug fix release?  (Aahz proposes "yes".)

    Is it allowed to do multiple forks (e.g. is it permitted to have
    both 2.0.2 and 2.0.2p)?  (Aahz proposes "no".)

    Does it makes sense for a bug fix release to follow a patch
    release?  (E.g., 2.0.1, 2.0.2p, 2.0.3.)

    Exactly how does a candidate patch release get submitted to
    PythonLabs for certification?  And what does "certification" mean,
    anyway?  ;-)

    Who is the Patch Czar?  Is the Patch Czar a single person?  (Aahz
    says "not me alone".  Aahz is willing to do a lot of the
    non-technical work, but Aahz is not a C programmer.)

    What is the equivalent of python-dev for people who are
    responsible for maintaining Python?  (Aahz proposes either
    python-patch or python-maint, hosted at either python.org or
    xs4all.net.)

    Does SourceForge make it possible to maintain both separate and
    combined bug lists for multiple forks?  If not, how do we mark
    bugs fixed in different forks?  (Simplest is to simply generate a
    new bug for each fork that it gets fixed in, referring back to the
    main bug number for details.)


  References
  
    [1] PEP 3, Hylton, http://python.sourceforge.net/peps/pep-0003.html


  Copyright
  
    This document has been placed in the public domain.



From william.famy at noos.fr  Sat Mar 24 11:23:31 2001
From: william.famy at noos.fr (William Famy)
Date: Sat, 24 Mar 2001 17:23:31 +0100
Subject: difference between script and prompt ! looking for the Problem
Message-ID: <3ABCCA03.90871133@noos.fr>

Hi.

I this is my problem.

I have a file fonc.py where i write a class.

class comp:
	def __init__(re,im):
		self.r=re
		self.i=im



when i use a python shell i see this


>>> import fonction
>>> c=fonction.comp(3.0,4.5)
>>> c.r,c.i
(3.0, 4.5)
>>> 

but when i use the following script 

test.py
#!/usr/bin/python
import fonction
c=fonction.comp(3.0,-4.5)
c.r, c.i


i have no result

bash-2.04$ ./test.py 
bash-2.04$

Where is the problem? where is my error?

Thanks for answer.
-- 
 			     \\\|///
                           \\  - -  //
                            (  @ @  )
+-------------------------oOOo-(_)-oOOo-------------------------+
|  Famy William  36 avenue des barattes  74000 Annecy  France   |
|  email:william.famy at noos.fr william.famy at mail-enseignant.com  |
+---------------------------------Oooo--------------------------+
                           oooO   (   )
                          (   )    ) /
                           \ (    (_/
                            \_)


From reborn at neozone.net  Sun Mar 11 11:02:23 2001
From: reborn at neozone.net (syKim)
Date: Mon, 12 Mar 2001 01:02:23 +0900
Subject: Drag & Drop in Jython
Message-ID: <98g7be$q3p$1@news2-2.kornet.net>

Hello ..

I'm recently trying  DnD programming of JTree in Jython..

but, It's too difficult to me -_- (I'm newbie yet..)

I guess it's difference of programming style between jython and java..

(especially, event driven like that xxListener, xxRecognizer etc.. )

Who have DnD source that implement in Jython ?

Or, Is there a way to handle event driven program like xxRecognizer?

Please Help me..

syKim.





From Dan.Rolander at marriott.com  Fri Mar  2 08:15:44 2001
From: Dan.Rolander at marriott.com (Rolander, Dan)
Date: Fri, 2 Mar 2001 08:15:44 -0500 
Subject: Unresolved symbols in MSVCRT.DLL when running a py2exe packag
	ed program.
Message-ID: <6176E3D8E36FD111B58900805FA7E0F80CCF63ED@mcnc-mdm1-ex01>

I agree with Robin and Gordon's assessment. It's risky at best to distribute
system dll's with your application, and impossible to upgrade system dll's
on Win2K because of system file protection.

Dan

-----Original Message-----
From: Thomas Heller [mailto:thomas.heller at ion-tof.com]
Sent: Friday, March 02, 2001 8:03 AM
To: python-list at python.org
Subject: Re: Unresolved symbols in MSVCRT.DLL when running a py2exe
packaged program.


Here is a comment about a similar problem in
another py2exe-packed application, which
I received from Jeff Kunce:

>>I just tried runing a py2exe-packaged application on an older NT machine,
and
>>it choked with an "Entry Point Not Found" error on MSVCRT.dll
>
>Well, I played with it some more and got it to work.
>
>I simply excluded all the DLLs that are normally found in NTs system32
>directory (MPR.dll, MSVCIRT.dll, MSVCRT.dll, NETAPI32.dll, NETRAP.dll,
>SAMLIB.dll) - and it works now on the older machine.
>
>It looks like there is no easy "good" solution to this problem. Here's a
>comment by Gordon McMillan on the topic:
>  http://www.wxpython.org/archive/2000-June/002469.html

Thomas


-- 
http://mail.python.org/mailman/listinfo/python-list



From moshez at zadka.site.co.il  Mon Mar 26 17:53:35 2001
From: moshez at zadka.site.co.il (Moshe Zadka)
Date: Tue, 27 Mar 2001 00:53:35 +0200
Subject: Some Basic Questions about Python
In-Reply-To: <tbv8g06eg66tec@corp.supernews.com>
References: <tbv8g06eg66tec@corp.supernews.com>
Message-ID: <E14hfrj-0005CB-00@darjeeling>

On Mon, 26 Mar 2001 15:14:05 -0500, "Bill Walker" <bwalker at earth1.net> wrote:
 
> 1) If I write an application and want to pass it on to others who may not
> have Python installed, how do I do this? I understand there is no compiler
> for
> Python and Tk yet that generates .exe files.

There is, and you don't want to use it. 

> 2) Can I distribute the Python install file along with my app without
> violating a copyright or license?

Yes, and that is a much better way.
But you don't want to trust some stranger on usenet --- read the license
yourself.

> 3) Better yet, can I include only the base files needed to run my
> application with the source code? Say Pythonw.exe and only the appropriate
> DLLs, libs, etc.

Why the hell? Any windows app comes with 10mb of useless crap anyway,
why break tradition? ;-)
More seriously, disk space is cheap, your time isn't.

-- 
"I'll be ex-DPL soon anyway so I'm        |LUKE: Is Perl better than Python?
looking for someplace else to grab power."|YODA: No...no... no. Quicker,
   -- Wichert Akkerman (on debian-private)|      easier, more seductive.
For public key, finger moshez at debian.org  |http://www.{python,debian,gnu}.org



From root at [127.0.0.1]  Mon Mar 26 02:52:33 2001
From: root at [127.0.0.1] (deadmeat)
Date: Mon, 26 Mar 2001 07:52:33 GMT
Subject: file locking
Message-ID: <5zCv6.583$p5.2667@news1.rivrw1.nsw.optushome.com.au>

How do I lock a file on opening it?  the bultin open() only allows me to set
the read/write mode, not any deny modes.





From new_name at mit.edu  Thu Mar  1 16:26:03 2001
From: new_name at mit.edu (Alex)
Date: 01 Mar 2001 16:26:03 -0500
Subject: Command-line interface to the Online Oxford English Dictionary.
Message-ID: <etdhf1dchas.fsf@pickled-herring.mit.edu>

I've written a CLI for oed.com.  You need to be on a network that's
subscribed to its service.  If there's sufficient interest, and I can
get access to a machine off MIT's network, and access to an OED account,
I'll add password-based access.

You can get it at 

http://www.mit.edu/afs/sipb.mit.edu/user/alex_c/oed/oed-1.0.tgz

Here's the README file:

Online interface to oed.com.  Run the command 'oed.py' with no arguments
to get usage information.

COPYRIGHT

As far as I'm concerned, you can do as you wish with the code in this
package, but the installation process makes two files, cookied_urllib.py
and pronunciation.py, that are based on python's urllib.py and the page 
http://dictionary.oed.com/help/OED_guide/pronunciation_2e.htm, and these
are encumbered with copyrights.

ACKNOWLEDGEMENTS

Thanks to Liana Lareau for the idea for this code, and to jhawk for
pointing me at the oed URL above.  Obviously, the code would be
worthless without python and oed.com, too; thanks to everyone who's
worked on those.

INSTALLATION

Run the command 'python2.0 make_files.py'.

On unix, put cookied_urllib.py and symbol.py somewhere that will allow
them to be imported by python programs.  Put oed somewhere that your
shell can find it, and try it out.  Let me know if you have any
problems.

I'm not sure what you'd do to get this working on a windows box, but if
you're used to using command-line interfaces, hopefully you'll know
enough to figure it out.  If someone sends me instructions, I'll include
them in a subsequent README.

Enjoy. :)
Alex.

-- 
Alex Coventry (finger alex_coventry at mit.edu)


From tim.one at home.com  Wed Mar 21 19:52:05 2001
From: tim.one at home.com (Tim Peters)
Date: Wed, 21 Mar 2001 19:52:05 -0500
Subject: ASCII encoding error: ordinal not in range(128)
In-Reply-To: <3AB9480F.4020104@ActiveState.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCOEGHJHAA.tim.one@home.com>

[Tim]
>  >>> UnicodeError: ASCII encoding error: ordinal not in range(128)
> ...
> Anything is better than what it says now!
> even-a-ham-sandwich-ly y'rs  - tim

[Mark Hammond]
> I'm a little confused.

That's because you're Australian!  So all the blood flows to your head, away
from the organ most normal guys think with.

> Are you proposing:
>
> UnicodeError: ASCII encoding error: ordinal not in ham sandwich
>
> or
>
> UnicodeError: ASCII encoding error: ham sandwich not in range(128)
>
> ??

Either of those is better than "ordinal not in range(128)", yes?  In America,
kids grow up with this violation of transitivity:

    Q: Why is a ham sandwich better than happiness in life?

    A: Because nothing is better than happiness in life, and a
       ham sandwich is better than nothing.  QED

> I prefer the former as that is something you would never find in my
> samdwiches.  However, trying to predict exactly what Tim will stick in
> his mouth is a fools folly <wink>

I've even been known to get Marmite *near* my mouth -- but never actually in
it yet.  Vegamite is right out.

    UnicodeError: ASCII unpalatable error: vegamite found, ham expected

> mmmmm-ham-ly,

leave-the-fat-for-me-ly y'rs  - tim




From grante at visi.com  Thu Mar 22 16:11:57 2001
From: grante at visi.com (Grant Edwards)
Date: Thu, 22 Mar 2001 21:11:57 GMT
Subject: Good Editor? (What's the big deal with syntax colouring?)
References: <mailman.985291829.27633.python-list@python.org>
Message-ID: <xUtu6.1763$SB2.191937@ruti.visi.com>

In article <mailman.985291829.27633.python-list at python.org>, John Schmitt wrote:

>Why do people want syntax colouring?  I can't stand it. 

At first, I didn't like it either. It took some getting used
to.  Now when a string, comment, keyword, or constant isn't in
the right color, it's an immediate hint that I've mis-typed
something.

-- 
Grant Edwards                   grante             Yow!  This ASEXUAL
                                  at               PIG really BOILS
                               visi.com            my BLOOD... He's
                                                   so... so... URGENT!!


From scarblac at pino.selwerd.nl  Wed Mar 21 15:29:12 2001
From: scarblac at pino.selwerd.nl (Remco Gerlich)
Date: 21 Mar 2001 20:29:12 GMT
Subject: learning python...
References: <3AB8F4E1.184DD9C@noaa.gov>
Message-ID: <slrn9bi3jk.ejm.scarblac@pino.selwerd.nl>

Benjamin.Altman <ben at .> wrote in comp.lang.python:
> Hello.  From the perspective of someone who already has programming
> experience, would anyone know if O'reillys "Learning Python" book would
> be better than say Prentice Hall's "Core Python Programming"?  Or would
> it be better to go straight into something like "Programming Python"?

I thought Learning Python was a very good introduction, I was a CS student
with experience in several languages. There are a few things to you need to
"get" once about Python, and the book explains them well. It doesn't go very
deep, but then, there's not that much to know, Python is simplicity.

Programming Python I found a bit chaotic, never really used it much.

Don't know Core Python Programming.

-- 
Remco Gerlich


From kens at sightreader.com  Mon Mar 26 12:55:33 2001
From: kens at sightreader.com (Ken Seehof)
Date: Mon, 26 Mar 2001 09:55:33 -0800
Subject: warrior geeks
References: <3ABF72F1.98CB78F@sage.att.com>
Message-ID: <007801c0b61d$f5b7bca0$04090a0a@upcast.com>

Me too.  I'm an Aikidoist.

----- Original Message ----- 
From: "Garry Hodgson" <garry at sage.att.com>
Newsgroups: comp.lang.python
To: <python-list at python.org>
Sent: Monday, March 26, 2001 8:48 AM
Subject: warrior geeks


> 
> a funny thing happened at IPC9.  a bunch of were sitting in the
> lounge (bar) talking, and one person mentioned that he was going
> to study kendo while in japan this year.  i asked if he had previous
> martial arts experience, and he said yes, he'd studied judo.
> it then turned out that in this group of maybe 6-7 people, we had
> this judo/kendo student, another who's done karate and now kickboxing,
> another who did karate, myself (okinawan karate, aikijutsu and others),
> and a pa qua practitioner.  at the next table sat another tae kwon
> do/aikido
> student.
> 
> it seemed like an odd coincidence, but it made me wonder: is there an 
> unusually high percentage of MA practitioners among the python
> community?
> or the open source community?  or software in general?
> 
> enquiring minds want to know...
> 
> -- 
> Garry Hodgson                   Once in a while
> Senior Hacker                   you can get shown the light
> Software Innovation Services    in the strangest of places
> AT&T Labs                       if you look at it right
> -- 
> http://mail.python.org/mailman/listinfo/python-list
> 




From shaleh at valinux.com  Wed Mar  7 12:33:46 2001
From: shaleh at valinux.com (Sean 'Shaleh' Perry)
Date: Wed, 07 Mar 2001 09:33:46 -0800 (PST)
Subject: Easy way to get the X selection (as text)?
In-Reply-To: <3AA66C83.29412EE0@gmx.net>
Message-ID: <XFMail.20010307093346.shaleh@valinux.com>

On 07-Mar-2001 Christopher Arndt wrote:
> I skimmed the docs of most X GUI modules for python on how to retrieve
> the X selection, but this seems to be a non-trivial task. Is there an
> easy way to do this, without having to use tkinter (don't want a useless
> window) or is there a ready-to-use module for this?
> 
> I want to write a little utility for openening urls I have selected in
> some window.
> 

It isn't trivial.  I wrote a command line app in C to do this, around 50 lines
or so.  Mail me at shaleh at via.net and i will mail it to you.  I do not have it
with me currently.



From aahz at panix.com  Mon Mar 26 20:36:45 2001
From: aahz at panix.com (Aahz Maruch)
Date: 26 Mar 2001 17:36:45 -0800
Subject: extending the built in dictionary type
References: <umfKujAz2iu6EwHa@jessikat.demon.co.uk> <6Rzu6.12951$_46.485491@e420r-atl2.usenetserver.com> <BXz8sKATYwu6Ewg9@jessikat.fsnet.co.uk>
Message-ID: <99oqrd$ijs$1@panix2.panix.com>

In article <BXz8sKATYwu6Ewg9 at jessikat.fsnet.co.uk>,
Robin Becker  <robin at jessikat.fsnet.co.uk> wrote:
>
>perhaps I didn't make myself clear. I will use UserDict for a python
>version, but if I want to accelerate to C I find fundamental problems in
>getting useful access to the code of say dictobject; in particular
>there's no obvious way to extend/modify in a small way; eg have a dict
>with just one more method. I either have to just copy that code
>wholesale or do rather silly things like changing my object's type
>before and after the operation. 

Why not simply have your object contain a dict as one of its members and
delegate the base operations to the Python dict?
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

"Boost the stock market -- fire someone"


From cjc26 at nospam.cornell.edu  Sat Mar 10 11:35:34 2001
From: cjc26 at nospam.cornell.edu (Cliff Crawford)
Date: Sat, 10 Mar 2001 16:35:34 GMT
Subject: finding and listing the differences between two files
References: <mailman.983966710.3462.python-list@python.org>
Message-ID: <qJsq6.209137$v.20440644@typhoon.nyroc.rr.com>

* Chris Richard Adams <chrisa at ASPATECH.COM.BR> menulis:
| 
| Anyone know the equivalent in Python or am I better off sticking to
| shell.

2.1 has a module called "difflib" which you can use.  Also, a diff
program written in Python comes with the Python source
(Tools/scripts/ndiff.py).


-- 
Cliff Crawford

He who sacrifices his conscience to ambition
         burns a picture to obtain the ashes.  - Chinese proverb


From db3l at fitlinxx.com  Thu Mar  8 22:39:42 2001
From: db3l at fitlinxx.com (David Bolen)
Date: 08 Mar 2001 22:39:42 -0500
Subject: from Tkinter import *
References: <mailman.983920449.5230.python-list@python.org> <slrn9ab0g7.s6.matt@happy-hour.mondoinfo.com> <989dot$t6t$1@panix3.panix.com>
Message-ID: <uae6vk3up.fsf@ctwd0143.fitlinxx.com>

aahz at panix.com (Aahz Maruch) writes:

> Actually, I disagree with the effbot.  I used to think it was okay to
> special-case "from Tkinter import *", but I've come to the conclusion
> that "import *" is _always_ a Bad Idea in any sort of production code.
> Sure it takes more work to write

While it may not apply to Tkinter, the one place I still see a use for
"import *" even in production code is large wrapped C libraries that
already name everything according to a flat namespace (for example,
staying in the GUI vein, wxPython).  Everything in such a system
already has a common prefix, so there's no loss of clarity or
ambiguity, and the only negative is the insertion directly into the
local namespace, but for a wrapped external library I don't think that
is much of a negative.

I do also sometimes do it for modules designed for it (such as
threading), but that's a finer line for production code.

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


From shaleh at valinux.com  Thu Mar  1 18:18:17 2001
From: shaleh at valinux.com (Sean 'Shaleh' Perry)
Date: Thu, 01 Mar 2001 15:18:17 -0800 (PST)
Subject: Newbie: Casting Int to string....
In-Reply-To: <3a9ed412$1@news1>
Message-ID: <XFMail.20010301151817.shaleh@valinux.com>

On 01-Mar-2001 Digital Hyakugei wrote:
> Ok, pretty simple, no? I can not find out how to do the following...
> 
> i've got a variable:
> 
> aNumber = 1
> 
> Now i have a function
> 
> nntp.stat( aVal )
> 
> where aVal has to be a string.
> 
> I cant do
> 
> nntp.stat( aNumber )
> 
> cuz it expects a string, not an int.
> 
> How, oh how, do i cast an int to a string?
> I'm sure its something that is forehead slapping obvious, but i just cant
> seem to find it...
> 

str(foo) or '%s' % foo.  Both work.



From tim.one at home.com  Sat Mar 31 15:35:26 2001
From: tim.one at home.com (Tim Peters)
Date: Sat, 31 Mar 2001 15:35:26 -0500
Subject: list vs tuple
In-Reply-To: <uGqx6.1052$p5.3997@news1.rivrw1.nsw.optushome.com.au>
Message-ID: <LNBBLJKPBEHFEDALKOLCGEICJJAA.tim.one@home.com>

[deadmeat]
> ...
> when the statement a = b copies the data across for one type it's
> expected to do it for others.

In Python,

    a = b

never copies data.  No exceptions.  It merely establishes a binding from the
name 'a' to whatever object 'b' is bound to at the time.  The type of the
object has no bearing on this.  Immediately after

    a = b

the Python expression

    a is b

returns true, provided only that b *was* bound to *some* object (else a
subclass of NameError is raised during the "a = b" line).

>>> b = 5
>>> a = b
>>> a is b
1
>>> b = [5]
>>> a = b
>>> a is b
1
>>>

can't-get-it-straight-until-you-give-up-on-what's-twisted-ly y'rs  - tim




From g_will at cyberus.ca  Fri Mar 16 10:39:10 2001
From: g_will at cyberus.ca (Gordon Williams)
Date: Fri, 16 Mar 2001 10:39:10 -0500
Subject: Pep 238 Why new // operator (Non-integer Division)
Message-ID: <98tc2s$qir$1@news2.ottawa.cyberus.ca>

Why do we need a new // operator to do the divide operation.  If we want to
do an integer divide with two reals why wont people just use int(a/b).  For
example:

>>> 5/2
2
>>> int(5.0/2.0)
2
>>>
The only difference with what is described below is a//b will result in a
object of type a while int(a/b) will be of type int.  Does this really need
the use of another operator??

Lets keep it simple.  (And that goes for the other Peps as well)

All these changes make my head spin.  There is more to life than learning
all the fine details in a computer language.  The fewer number of ways to do
something the better.

Regards,

Gordon Williams








// Operator

    A `//' operator which will be introduced, which will call the
    nb_intdivide or __intdiv__ slots.  This operator will be
    implemented in all the Python numeric types, and will have the
    semantics of

        a // b == floor(a/b)

    Except that the type of a//b will be the type a and b will be
    coerced into.  Specifically, if a and b are of the same type, a//b
    will be of that type too.






From claird at starbase.neosoft.com  Fri Mar 23 10:42:25 2001
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 23 Mar 2001 09:42:25 -0600
Subject: email address verification
References: <99d5qi$95l$1@solaria.cc.gatech.edu> <99fjv8$bhf$1@solaria.cc.gatech.edu> <E006A65FA23EFF2C.2D74AEB35552E0B6.065EC0B29EF187C6@lp.airnews.net> <99fo48$d9h$1@solaria.cc.gatech.edu>
Message-ID: <6491BB8AD25B69BC.0BFC7C13FCE3713E.F0B7BA5A511B6DBC@lp.airnews.net>

In article <99fo48$d9h$1 at solaria.cc.gatech.edu>,
Holland King  <insanc at cc.gatech.edu> wrote:
>Cameron Laird <claird at starbase.neosoft.com> wrote:
>: I don't understand your description.  In general,
>
>ok, we run a server that holds a lot of accounts, once an account expires
>then we run a script that parses the .forward file found in the home 
>dir. the script then takes the addresses found and makes them into aliases
>so that hopefully the user will not lose any mail. we need to make sure
>that the aliases are real addresses so that at worst mail is not bounced 
>and at best it doesn't go to a mailing list (i know i am being an optimist)
It's a good question to ask--but one which
turns out to be quixotic.  Not only can one
make no guarantees by a "static" examination
of a purported e-mail address, my own exper-
ience is that there are few heuristics which
are worth implementing.  So much can and
does happen at the recipient end of an e-mail
transaction that it's nearly fruitless to
try to automate.
>
>: comp.mail.misc might be a more productive place to
>: ask these questions than c.l.p.
>
>ok thank you i will try there to, the reason i asked here is that i tried
>comp.linux.misc and they didn't help and i am writing it in python (i 
>think my orignal post mention that i had found the smtp library and i 
>was wondering if there was something similar that would work)
I understand.  It's a reasonable presumption.
			.
			.
			.
-- 

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


From aleaxit at yahoo.com  Fri Mar 23 12:31:55 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 23 Mar 2001 18:31:55 +0100
Subject: newbe   how to read in a string?
References: <3ABB4E70.54885C9F@noos.fr> <mailman.985355670.6600.python-list@python.org>
Message-ID: <99g1ac02fug@news2.newsguy.com>

"Chris Gonnerman" <chris.gonnerman at usa.net> wrote in message
news:mailman.985355670.6600.python-list at python.org...
    [snip]
>     import string
>     seq = string.split(a, ",")
>     for i in range(1, len(seq)): # this skips the 0 index,
>         seq[i] = float(seq)      # which needs to stay a string.
>
> In Python 2.0 the code might be shorter, using string methods and list
> comprehension, but I personally prefer the "old way" myself.

The 'old way' would be
    seq = map(float, string.split(a, ','))
and it still works fine of course, though I'd rather spell it
    seq = map(float, ','.split(a))

List comprehensions are great, but when you DO want to get
map's primary job (applying an already-existing function
to parameter sequences) map is still best (with list
comprehension, you can avoid using _lambda_ just in order
to be able to use map -- I think eschewing lambda makes
your code clearer, but maybe that's just me).


Alex





From sholden at holdenweb.com  Thu Mar 22 23:23:32 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 22 Mar 2001 23:23:32 -0500
Subject: Good Editor? (What's the big deal with syntax colouring?)
References: <008F0A63472BD311AF9800104BCD102561CD3E@minirex.vmlabs.com> <mailman.985295730.14145.python-list@python.org>
Message-ID: <qgAu6.13041$_46.497624@e420r-atl2.usenetserver.com>

"Timothy Grant" <tjg at exceptionalminds.com> wrote in message
news:mailman.985295730.14145.python-list at python.org...
> On Thu, Mar 22, 2001 at 12:10:59PM -0800, John Schmitt wrote:
> > Why do people want syntax colouring?  I can't stand it.  One of my minor
> > beefs with CodeWright is that there isn't a one button way to tell it to
> > change _all_ of my colours all at once.  Am I missing something by
disabling
> > syntax colouring?  It always reminds me of the artwork my 6-year-old
nephew
> > brings home from school to show his mommy.
>
> Hey, I'm colour blind, and I still need syntax highlighting
> though I can't see all of the colour differenctials.
>
> It's pretty simple. Forget to close a ("|')--or open one for
> that matter, and you automatically know somethings wrong
> because the colours are wrong, misspell(sp?) a keyword and you
> know it because it doesn't hi-light in the correct colour.
>
> It is a huge benefit to getting syntax correct before you save
> your file.
>
> Now if only else didn't highlight but else: did<wink>
>
I seem to remember that having to write your source on coding forms, have it
punched up on tape or cards and waiting six hours until the computer center
had deposited the printout in your pigenhole made me similarly careful about
my coding style.

that's-when-i-discovered-semantics-ly y'rs  - steve





From sholden at holdenweb.com  Thu Mar  8 08:57:22 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 8 Mar 2001 08:57:22 -0500
Subject: Can anyone recomend a good intoduction to C...
References: <3aa366c6_2@nnrp1.news.uk.psi.net> <8gNo6.440$y6.128357@ruti.visi.com> <3aa6551e.212870101@news.laplaza.org> <Taup6.1292$y6.235409@ruti.visi.com>
Message-ID: <ygMp6.34893$ML1.1698582@e420r-atl2.usenetserver.com>

"Grant Edwards" <grante at visi.com> wrote in message
news:Taup6.1292$y6.235409 at ruti.visi.com...
> In article <3aa6551e.212870101 at news.laplaza.org>, Mats Wichmann wrote:
>
> >(oops, I smell a flood of "I can top that" comments such as "you had a
> >terminal? I learned on a Teletype", "you had a Teletype? I had to
> >toggle code in using switches", "you had switches?  I had to use
> >Forth!".  Consider that joke already played out, eh).
>
> You had 1's?  In my day we had to write everything using nothign but 0's!
>
We'd 'ave give our right arms for a few zeroes when I were a lad. We used to
have to get up three hours before we went to bed, lick 't information
superhighway clean wi' us bare tongues, and that were just so we could have
nothing at all. I used to borrow zeroes from next door for mi' mother's
birthday, and spend an hour cleaning 'em up before I took 'em back. And yer
tell these young programmers wi' their GUIs and their IDEs and they just
don't believe yer.

born-with-an-unfair-advantage-in-a-yorkshireman-sketch-ly y'rs  - steve

PS: Mats, this is c.l.py, you surely didn't believe you could finesse this?




From not.this at seebelow.org  Sun Mar  4 20:37:00 2001
From: not.this at seebelow.org (Grant Griffin)
Date: Sun, 04 Mar 2001 19:37:00 -0600
Subject: suggestions for more docs (was Re: Bug fix releases)
References: <LNBBLJKPBEHFEDALKOLCMEEPJBAA.tim.one@home.com> <mailman.982897324.9109.python-list@python.org>  
	            <200103031821.NAA24060@panix3.panix.com> <mailman.983646726.27322.python-list@python.org>
Message-ID: <3AA2EDBC.E933E78E@seebelow.org>

Guido van Rossum wrote:
> 
...
> But I understand that you're saying that the community has grown so
> conservative that it can't stand new features even if they *are* fully
> backwards compatible.
> 
> I wonder, does that extend to new library modules?  Is there also
> resistance against the growth there?  I don't think so -- if anything,
> people are clamoring for more stuff to become standard (while at the
> same time I feel some pressure to cut dead wood, like the old SGI
> multimedia modules).

As somebody who has sometimes fussed over certain new language features
(because they can make Python harder for beginnners to read, IMHO), I
can say that, nonetheless, I'm _absolutely_ in favor of more standard
library modules.

Although Python's beautiful simplicity makes "learning" it in an
afternoon possible, there's no avoiding the fact that learning its large
standard library (or even just the most important half-dozen or so
modules--if only the beginner were told which ones <hint>) will
inevitably take awhile.  I've been at Python for about a year now, and
having once "learned" it in an afternoon (in a day, actually--I'm a slow
learner <wink>), it's taken the rest of the year for me to get halfway
good at it.  Much of that has to do with the standard library.

Anyway, since mastering a large library is inevitable, we might as well
lay back and enjoy it.  (And presumably, the larger the library is, the
more likely we are to enjoy it. <wink>)

> So that relegates us at PythonLabs to a number of things: coding new
> modules (boring), or trying to improve performance of the virtual
> machine (equally boring, and difficult to boot), or fixing bugs (did I
> mention boring? :-).
> 
> So what can we do for fun?  (Besides redesigning Zope, which is lots
> of fun, but runs into the same issues.)

>From the preceeding, I'm gonna deduce that the primary excitement the
PythonLabs team gets is implementing new language features.  But what
about writing more documentation?--*that's* pretty exciting!  In fact,
as one who publishes software myself, I can attest to the fact that it's
even _more_ exciting than implementing new features. <wink>

Luckily, new features inevitably require new documentation, so indulging
in the former becomes an excuse to indulge in the latter.  But who
really needs an excuse?  When the PythonLabs team needs to let their
brains cool off after a hard day of doing boring stuff, they might treat
themselves to writing some of the following:

- Tutorial, Part II: This would be directed at people who have practiced
Python for, say, 3 months.  It would introduce them to the Python's more
features that aren't covered in the original tutorial.  It would be
written using examples (ala K&R), which would give a sense not just of
"what and how" but more of "when and why".

- A Tour of The Standard Library: This would help beginners learn the
most important standard modules--in effect, clearing the forest of its
useful (but less used) trees.  I really think that listing the most
frequently used modules (and the most important functions _in_ those
modules) would be a _great_ service to beginners.  Off the top of my
head, they might be (in no order): string, os, os.path, sys, math, re,
etc.  (Even just providing a _list_ of these things in order of
importance would be a great service, though, admittedly, that wouldn't
provide the PythonLabs team with more than a couple of hours of
excitement. <wink>)

Also, the following are available online, but including them in the
standard distribution would help beginners a lot: 

- Python Idioms: Why should "while 1:" remain a Masonic handshake known
exclusively to us "veterans"? <wink>

- FAQs: New Pythoneers often wonder "why is there no do/while?" and "why
do we have lists _and_ tuples?".  By providing FAQs in the standard
distribution, it would be a lot easier for them to learn answers to
these things, once somebody on c.l.py had told them to "read the fine
manual".

i'd-write-this-stuff-myself-but-writing-for-dspGuru-gives-me-as-much
   -excitement-as-my-cardiovascular-system-can-stand-<wink>-ly y'rs,

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation            http://www.iowegian.com


From donn at u.washington.edu  Thu Mar 29 11:39:54 2001
From: donn at u.washington.edu (Donn Cave)
Date: 29 Mar 2001 16:39:54 GMT
Subject: setvbuf() ??
References: <slrn9c50mp.me7.wcuddy@lserv.ja10629.home>
Message-ID: <99vogq$10dk$1@nntp6.u.washington.edu>

Quoth wcuddy at lserv.ja10629.home (Wayne Cuddy):
| I need to print a period on an xterm every few seconds to provide status to the
| user, I don't want to force a newline.  How do I turn off line buffering for
| standard out?

I don't know, but you don't need that anyway - if you don't want buffered
output then don't use an output buffer like stdout!

You can write directly to standard output like this:

   os.write(1, '.')

Unit 1 is by definition standard output;  2 is error and 0 is input.
File objects stdout, stderr and stdin use these same file descriptors
eventually, via C library stdio functions, so when you write directly
to them you just omit a lot of extra stuff you don't want.

If you really want a file object for some reason, stderr normally will
be unbuffered.

	Donn Cave, donn at u.washington.edu


From robin at jessikat.fsnet.co.uk  Fri Mar 30 18:24:13 2001
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Sat, 31 Mar 2001 00:24:13 +0100
Subject: ReportLab 1.06 Released
Message-ID: <WuOtstAdWRx6EwIT@jessikat.fsnet.co.uk>

ReportLab 1.06 has been released see http://www.reportlab.com this
version has initial support for higher level graphics including widgets
and charts.
-- 
Robin Becker


From neilh at scintilla.org  Fri Mar  2 21:15:37 2001
From: neilh at scintilla.org (Neil Hodgson)
Date: Sat, 03 Mar 2001 02:15:37 GMT
Subject: Crashed PythonWin. How can I fix it?
References: <3aa04cf5.34464140@News.CIS.DFN.DE>
Message-ID: <dtYn6.57784$o85.326383@news-server.bigpond.net.au>

Costas Menico:
> I crashed PythonWin and cant get it backup and running. I was using
> Tkinter when this happened.

   It looks quite unhappy. I'd remove PythonWin and reinstall it.

   Neil





From sholden at holdenweb.com  Tue Mar 20 23:41:52 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Tue, 20 Mar 2001 23:41:52 -0500
Subject: Should __future__ statements ever be phased out?
References: <20010320173101.21650.qmail@web217.mail.yahoo.com> <mailman.985117645.6322.python-list@python.org>
Message-ID: <kmWt6.4231$d11.242907@e420r-atl1.usenetserver.com>

"Paul Prescod" <paulp at ActiveState.com> wrote in message
news:mailman.985117645.6322.python-list at python.org...
> Hamish Lawson wrote:
[...]
>  2. Always have a version declaration or feature declaration at the top
> of Python programs.
>
> You already have the ability to implement 2 if you are paranoid. You can
> check sys.version_info in your code.
>
<Erid Idle>
Ooh, beginning to sound a bit like Perl here, dearie, know what i mean,
nudge nudge? Wink wink ... just a touch too, how shall I put this, earthy?
</Eric Idle>

Seriously: how about Python 2.1 to 2.0, and 2.0 to {1.6, 1.5.2} converters?
Is it sensible to try and engineer backwards compatibilty post facto? What
idioms translate backwards through the language's development?

Also I don't _really_ mean to belittle version control issues, which I know
are all too serious a part of engineering.

regards
 Steve







From mysticsguy at yahoo.ca  Thu Mar 22 22:47:58 2001
From: mysticsguy at yahoo.ca (Yvon Boulianne)
Date: Fri, 23 Mar 2001 03:47:58 GMT
Subject: Jython is cool?
References: <20010322103954.A18539@sync.nyct.net> <mailman.985276416.27994.python-list@python.org> <99d9d0$n2a$1@panix6.panix.com>
Message-ID: <3abac714.20193877@news.vif.com>

On 22 Mar 2001 08:31:28 -0800, aahz at panix.com (Aahz Maruch) wrote:

>In article <mailman.985276416.27994.python-list at python.org>,
>Neil Schemenauer  <nas at arctrix.com> wrote:
>>
>>I'm hoping that the Mozilla guys catch the Python bug based on
>>what Active State has been doing with Komoto and adds "<script
>>language=python>" to Mozilla.
>
>Google protection: "Komodo"
>-- 
>                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)
>
>Androgynous poly kinky vanilla queer het Pythonista   http://www.rahul.net/aahz/
>Hugs and backrubs -- I break Rule 6
>
>"I won't accept a model of the universe in which free will, omniscient
>gods, and atheism are simultaneously true."  -- M

i must answers to this last quote ;-)
in fact the free will, omniscient, gods and atheism are simultaneously
true , it all depend on the Counsciousness Level you are,  the
universe is a Multidimensionnal one.

With Love And Counsciousness 
Yvon


Yvon Boulianne ; Saphi
mysticsguy at yahoo.ca
www.monunivers.net
Humour,Paradox & Changements


From robin.thomas at starmedia.net  Sat Mar 24 13:11:43 2001
From: robin.thomas at starmedia.net (Robin Thomas)
Date: Sat, 24 Mar 2001 13:11:43 -0500
Subject: static methods
In-Reply-To: <slrn9bpqgo.15q.grante@tuxtop.visi.com>
References: <mailman.985383828.2942.python-list@python.org>
Message-ID: <4.3.1.2.20010324124702.01b5c100@exchange.starmedia.net>

At 04:38 PM 3/24/01 +0000, you wrote:
>On Fri, 23 Mar 2001 16:51:10 -0500 (EST), Clark C. Evans 
><cce at clarkevans.com> wrote:
>
> >I was wondering if there is any rationale as to why
> >Python classes do not have static methods?
>
>What's a "static method?"  I assume it's some C++ contrivance?

In the simplest sense (C++ and Java), a static method is reached through 
the class, and not through instances of the class. Often used for factory 
or conversion functions, or to provide a procedural interface when 
instantiation would be a tedious formality. If there were a "static" 
keyword in Python:

class Pickler:

     def static loads(string):
         return Pickler().loads(string)

     def static dumps(object):
         return Pickler().dumps(object)

     def loads(self, string):
         # the instance method

     def dumps(self, object):
         # the instance method

That it's called "static" is an artifact of C, I believe. In C, "static" 
has been overexploited such that it has many different meanings in 
different contexts that have nothing to do with being "static". Hideous 
example: declaring a function as "static" in C means that it's invisible to 
code outside of the source file and to all code following it in the same 
file. :-0


--
Robin Thomas
Engineering
StarMedia Network, Inc.
robin.thomas at starmedia.net




From aleaxit at yahoo.com  Thu Mar  1 10:22:45 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 1 Mar 2001 16:22:45 +0100
Subject: how can i copy a sublist
References: <3A9E4C6F.E2DD63C3@bg-sys.de>
Message-ID: <97lpg302v8v@news2.newsguy.com>

"Daniel Orlowski" <daniel.orlowski at bg-sys.de> wrote in message
news:3A9E4C6F.E2DD63C3 at bg-sys.de...
> Hi,
>
> i try to copy a sublist from list "a" to list "b". After that i alter a
> value in "b" and oops! value in "a" is also altered.
> so it looks like that:
> >>> a=[[1,2],[3,4]]
> >>> b=[]
> >>> b.append(a[0])
> >>> b
> [[1, 2]]
> >>> b[0][0]=3
> >>> a
> [[3, 2], [3, 4]]
>
> how can i copy a sublist without
> "get_a_value_from_a_list_and_put_it_into_another"

I'm not quite sure what you mean, but, anyway,
    b.append(a[0][:])
will, assuming a[0] is a sequence, append to b
a _copy_ (a shallow-copy) of that sequence, rather
than a reference to the _same_ sequences as a[0]
refers to.


In general, using xxx[:] (a "slice everything"
request) works to get a shallow-copy of a sequence
to which xxx refers to (for any xxx -- it does not,
of course, matters if it's a name, an index into
some other sequence, an index into a mapping, &c).

For more general copying (deep rather than shallow,
or copying of other object kinds) see module copy.


Alex





From phd at phd.pp.ru  Tue Mar 20 05:23:57 2001
From: phd at phd.pp.ru (Oleg Broytmann)
Date: Tue, 20 Mar 2001 13:23:57 +0300 (MSK)
Subject: Thinking Outside the Box
Message-ID: <Pine.LNX.4.30.0103201317330.26091-100000@fep132.fep.ru>

   Connections that won't, upgrades that can't, hot syncs that don't,
standards that never are, wireless transmitters radiating who knows what.

      ...

Hephaestus was the Greek god of fire and crafts, which is to say technology.
Tellingly enough, he was born lame. His greatest creation was Pandora.
Hephaestus sent her -- at his father Zeus's request -- to plague mankind. She
was beautiful. But she sought knowledge. So she opened up a box that had been
entrusted to her. Out leapt the plethora of ills that afflict humanity. All
that was left inside was hope.

      ...

   http://washingtonpost.com/wp-dyn/articles/A23937-2001Mar18.html
      (found on /.)

      ...

   "There'd be different legal categories of software. If you want to rush
it to market, and have first-mover advantage and see what happens, fine.
You could label it creative software. But much of your code will be open to
the public."

   If you want to build a business empire, however, and you want the
government to throw itself into protecting your copyrights, in Lanier's
scheme you would then have to label your product "firmware."

   "If you see this thing is taking off, you have the right to say: This is
no longer creative software. This is functional. This has utility.

   "But once you do that, you owe me a buck every time it crashes."

Oleg.
----
     Oleg Broytmann            http://phd.pp.ru/            phd at phd.pp.ru
           Programmers don't die, they just GOSUB without RETURN.




From loredo at spacenet.tn.cornell.edu  Mon Mar  5 17:17:09 2001
From: loredo at spacenet.tn.cornell.edu (Tom Loredo)
Date: Mon, 05 Mar 2001 17:17:09 -0500
Subject: PIL for the Mac?
References: <3A9C6FDF.3F9422C@alcyone.com> <ty%m6.15923$Qb7.2598968@newsb.telia.net>
Message-ID: <3AA41065.29EB50C5@spacenet.tn.cornell.edu>

Fredrik Lundh wrote:
> 
> Erik Max Francis wrote:
> > Is PIL (PythonWare Image Library) available on the Mac?  I don't see any
> > reference to the Mac on its main Web page, and the source tarball
> > doesn't appear to have any Mac specific files in it ...
> 
> from what I can tell, it's shipped with the standard Mac
> distribution:

It is indeed, and it works just fine.

-Tom Loredo


From jimd at vega.starshine.org  Sun Mar 11 05:44:09 2001
From: jimd at vega.starshine.org (Jim Dennis)
Date: 11 Mar 2001 10:44:09 GMT
Subject: Color curses and Python 2?
Message-ID: <slrn9amlnn.ktp.jimd@vega.starshine.org>

 I'm playing with the Debian Linux (x86) Python v.2 beta and
 I can't seem to get color curses working.

 A minimal working sample would be nice.  The examples in 
 the curses HOWTO at python.org are simply not working (wrt
 the color functions).  The only time I've gotten color
 curses funtions working under Python is when I compiled my
 own copy of the interpreter with a replacement/patched version
 of the curses module.  

 I thought I'd heard that color ncurses support would be included
 in 1.6.  Was it?  I still haven't seen a 1.6.x package
 available for Debian.  Is it something they put in 1.6 and 
 changed for the 2.x betas?  Why is this so difficult.  Is it
 so much to want ncurses/color support?  Do I really have to 
 go back to Perl for this?


 
  



From andreas at andreas-jung.com  Wed Mar 21 17:10:35 2001
From: andreas at andreas-jung.com (Andreas Jung)
Date: Wed, 21 Mar 2001 17:10:35 -0500
Subject: Accessing Oracle from Python remotely ?
References: <Xns906AB9C901807PaCmAnRDLM@161.48.76.20> <eE3u6.776$9%.203510@newsr1.maine.rr.com>
Message-ID: <99bj2o$u91$2@nntp9.atl.mindspring.net>

 There are currently two different DCOracle versions available. DCOracle
1.3.x
supports is mostly build on top of Oracle 7 but I works although with Oracle
8i
except features like BLOBs and CLOBs. When you need complete support
for Oracle 8i take a look at Matt Kromers new version of DCOracle 2. It is
still alpha so be warned :-) You can find DCO2 on www.zope.org  - search for
DCO2.

Andreas Jung
Digital Creations

"gkiffney" <gkiffney at my-deja.com> schrieb im Newsbeitrag
news:eE3u6.776$9%.203510 at newsr1.maine.rr.com...
> There's http://www.zope.org/Products/DCOracle which should get you
started.
> There was an article by Uche Ogbuji in Linuxworld that helped me get going
> with this, but it's apparently fallen off the Internet map into the bit
> bucket.  Using the dcoracle module (which is a bit of a pain to get going,
> look at the Zope docs carefully, and if you can find the prebuilt dll's,
use
> those), you can do things like this:
>
> [uche at malatesta uche]$ python
> Python 1.5.1 (#1, Sep 3 1998, 22:51:17) [GCC 2.7.2.3] on linux-i386
> Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
> >>> import DCOracle
> >>> #connect to DB. scott/tiger are the test database username/password
> >>> dbc=DCOracle.Connect("scott/tiger at sforza") #sforza is the local Oracle
> server
> >>> c = dbc.cursor()
> >>> c.execute("SELECT * FROM emp WHERE emp.deptno = 10")
> >>> rows = c.fetchall()
> >>> print rows
> [(7782, 'CLARK', 'MANAGER', 7839, dbiDate('1981-06-09 00:00:00'), 2450.0,
> None, 10),
> (7839, 'KING', 'PRESIDENT', None, dbiDate('1981-11-17 00:00:00'), 5000.0,
> None, 10),
> (7934, 'MILLER', 'CLERK', 7782, dbiDate('1982-01-23 00:00:00'), 1300.0,
> None, 10)]
>
>
>
> "Fred Pacquier" <fredp at mygale.org.nospam> wrote in message
> news:Xns906AB9C901807PaCmAnRDLM at 161.48.76.20...
> > Hi,
> >
> > I have an Oracle (7) database running on a Windows NT 4 server. Is it
> > possible (easy, difficult ?) to read the data in those tables from a
> python
> > script running on a remote Unix (Solaris) box ?
> >
> > Sorry if the question is silly -- I do know a bit of python but I'm
> > inexperienced in databases and their connectivity. Any pointers welcome
> > though.
> >
> > Thanks,
> > fp
> >
> > --
> > YAFAP : http://www.multimania.com/fredp/
>
>
> --
> http://mail.python.org/mailman/listinfo/python-list




From winson at mdsserv.mds.com.tw  Mon Mar 19 09:36:31 2001
From: winson at mdsserv.mds.com.tw (???)
Date: Mon, 19 Mar 2001 22:36:31 +0800
Subject: Tkinter Support
References: <001901c0b046$1f1a7200$340210ac@winson> <3AB5C475.AF89FD43@ActiveState.com>
Message-ID: <000b01c0b082$02c55de0$340210ac@winson>

My code on Python is

from Tkinter import *
widget = Label(None, text='Hello GUI world!')
widget.pack()
widget.mainloop()

My message on Python is

"Traceback (most recent call last):
  File "c:\program files\python20\pythonwin\pywin\framework\scriptutils.py",
line 301, in RunScript
    exec codeObject in __main__.__dict__
  File "C:\Program Files\Python20\Oreilly\1.Intro\GUI\GUI1.PY", line 1, in ?
    from Tkinter import *
  File "c:\program files\python20\lib\lib-tk\Tkinter.py", line 34, in ?
    import FixTk # Attempt to configure Tcl/Tk without requiring PATH
  File "c:\program files\python20\lib\lib-tk\FixTk.py", line 1, in ?
    import sys, os, _tkinter
ImportError: DLL load failed:"

So maybe it's not lost Tk, it lost Tkinter, too.

--
Best Regards,

Winson Chang

3F, 2, 150 Lane, Sec. 5, Hsin I Rd., Taipei, Taiwan, R.O.C.

Tel: +886-2-27225333 ext 376
Fax: +886-2-27222330

Email: winson at mdsserv.mds.com.tw
----- Original Message -----
From: "Paul Prescod" <paulp at ActiveState.com>
To: "?i???q" <winson at mdsserv.mds.com.tw>; <python-list at python.org>
Sent: Monday, March 19, 2001 4:33 PM
Subject: Re: Tkinter Support


> ActivePython comes with TKinter but not Tk. You can use the package
> manager utility to install it (see the ActivePython start menu). Just
> type "install tk" from the PPM command line.
> --
> Take a recipe. Leave a recipe.
> Python Cookbook!  http://www.activestate.com/pythoncookbook
>
> --
> http://mail.python.org/mailman/listinfo/python-list
>




From tim.one at home.com  Sun Mar 11 13:16:34 2001
From: tim.one at home.com (Tim Peters)
Date: Sun, 11 Mar 2001 13:16:34 -0500
Subject: Static typing (was Re: Java guy interested in Python)
In-Reply-To: <A6Oq6.395631$ge4.138241941@news2.rdc2.tx.home.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCMEOAJEAA.tim.one@home.com>

[Rainer Deyke]
> When I type 'int i;' in C++, what I really mean is "'i' behaves like an
> integer variable.".

Since that's all "int i;" *can* mean in C++, seems tautological <wink>.

> I'm not saying anything about the values 'i' can take,
> except that the values must support integer semantics.

You're also explicitly agreeing to let the implementation decide which subset
of all the integers you're allowed to use, and are insisting on signed ints,
both of which say quite a bit about the values 'i' can take.  That this is
inadequate in practice is why C99 goes on to define a few dozen related
typdefs (int_least32_t, int_fast32_t, int32_t, etc).  It's still often (not
always) just approximating the set of values you really have in mind.

> The limited range of type 'int' is part of its defined semantics,
> as is its fast performance.

I expect you'll search in vain for a promise in the std that ints are "fast"
<wink>.

> If I want to restrict 'i' to a subset of its possible values, I use an
> assertion.

While if you were programming in Pascal instead, you could express subranges
directly in *its* type system; if you were programming in SLIM, you could
express concepts such as that a variable is restricted to even or prime
integer values; etc.  You use an assertion in C++ or *any other language*
because its type system isn't rich enough to capture all you want to say, not
because type systems are inherently incapable of capturing more than how big
machine registers are <0.5 wink>.

type-systems-are-to-types-as-4-bit-color-palettes-are-to-nature-ly
    y'rs  - tim




From nde at comp.leeds.ac.uk  Mon Mar 26 07:14:59 2001
From: nde at comp.leeds.ac.uk (Nick Efford)
Date: Mon, 26 Mar 2001 12:14:59 GMT
Subject: Please rate these Python books
References: <20010322.1445209@mis.configured.host> <_Meu6.8736$8O6.1818817@typhoon.austin.rr.com> <slrn9bjuha.4qv.grey@teleute.dmiyu.org> <blzu6.8974$G72.207211@e420r-atl1.usenetserver.com> <vwWu6.2238$45.10474@newsfeeds.bigpond.com> <WC%u6.1386$sk3.452235@newsb.telia.net>
Message-ID: <GAt0p3.M51@leeds.ac.uk>

On Sat, 24 Mar 2001 11:34:14 GMT, Fredrik Lundh <fredrik at pythonware.com> wrote:

> > Looking foward to the new OReilly Python book, though I fear fitting it
> > in the briefcase may be out of the question.
> 
> it's only 300 pages or so, so that shouldn't be much
> of a problem...  oh, you meant PP2E? ;-)

....Speaking of which, when is your book out?  The O'Reilly
website says April on one page, May on another.


Nick



From scarblac at pino.selwerd.nl  Fri Mar 30 01:37:12 2001
From: scarblac at pino.selwerd.nl (Remco Gerlich)
Date: 30 Mar 2001 06:37:12 GMT
Subject: forcing dictionary order
References: <3AC380AE.B225C07F@uniserve.com>
Message-ID: <slrn9c8a9s.10d.scarblac@pino.selwerd.nl>

Bob van der Poel <bvdpoel at uniserve.com> wrote in comp.lang.python: 
> I'd like (not a must!) to walk a dictionary in the same order in which
> it was created. Example:
> 
> 	foo={ key1: (), key2: (), key3: () }
> 
> If I do:
> 	
> 	for a in foo.keys(): ...
> 
> the order is not "key1, key2, key3", which is what I want. Note, I can't
> do something simple like sort() on the keys since the are not in any
> alphanumeric order. I'm just using this to generate a largish table for
> another program, and the actual order doesn't matter for it to work
> properly, but it would nicer if it were in the correct logical order.
> So, I'm not going to spend a lot of time on this, but if there is a
> simple way????

Dictionaries by themselves don't do this, they don't have any particular
order at all, and forget the order in which the items were inserted.

So you'll have to keep track of it yourself, maybe keep a 'fookeys' list
besides the dict that keeps the keys, in the desired order.

-- 
Remco Gerlich


From moshez at zadka.site.co.il  Mon Mar 19 16:02:41 2001
From: moshez at zadka.site.co.il (Moshe Zadka)
Date: Mon, 19 Mar 2001 23:02:41 +0200
Subject: Why "from __future__" stinks; a counter-offer
In-Reply-To: <995pl4$jta$1@news.mathworks.com>
References: <995pl4$jta$1@news.mathworks.com>, <mailman.984949523.1896.python-list@python.org> <4Aft6.2171$pP2.372173@paloalto-snr1.gtei.net> <mailman.985029625.6313.python-list@python.org>
Message-ID: <E14f6nZ-0006G0-00@darjeeling>

On 19 Mar 2001, Joshua Marshall <jmarshal at mathworks.com> wrote:
 
> As responses to suggestions, there's a lot of "write up a PEP or it
> won't get implemented" going around.  I can understand this, but it's
> also important that good ideas don't go to waste just because the
> proposer _didn't_ write a PEP.

Previously, the proposer had to write an implementation before even
being considered. Writing a PEP is much easier, and Barry even corrects
deviations from the PEP format himself: nonbody in PythonLabs would
make sure your C code compiles <wink>
 
> I guess I just hope that the PEP-mechanism doesn't make it so that
> good proposals get overlooked in favor of proposals with enthusiastic
> proponents.

Good proposals have always been overlooked in favor of proposals with
enthusiastic proponents. The PEP process *levels* the playing field:
with just enough time to write a document (which takes about 30 minutes,
modulu thinking up a coherent proposal), all the proposals are mentioned
in the same place, where everyone can look at them.

-- 
"I'll be ex-DPL soon anyway so I'm        |LUKE: Is Perl better than Python?
looking for someplace else to grab power."|YODA: No...no... no. Quicker,
   -- Wichert Akkerman (on debian-private)|      easier, more seductive.
For public key, finger moshez at debian.org  |http://www.{python,debian,gnu}.org



From moshez at zadka.site.co.il  Mon Mar 19 15:55:28 2001
From: moshez at zadka.site.co.il (Moshe Zadka)
Date: Mon, 19 Mar 2001 22:55:28 +0200
Subject: [Very Long (11K)] Numeric PEPs, first public posts
In-Reply-To: <slrn9bcp3h.ns2.qrczak@qrnik.zagroda>
References: <slrn9bcp3h.ns2.qrczak@qrnik.zagroda>, <mailman.984748941.31248.python-list@python.org> <Q5Ns6.432845$ge4.152574719@news2.rdc2.tx.home.com>
Message-ID: <E14f6ga-0006Ez-00@darjeeling>

On 19 Mar 2001, qrczak at knm.org.pl (Marcin 'Qrczak' Kowalczyk) wrote:

> > A simpler way to solve this problem would be to standardize Python
> > ints as 32 bit.
> 
> I would definitely not do it (especially as 64-bit platforms will be
> more popular as the time goes). Relying on the size of a type called
> int is evil. Call a type/class int32 if you want to guarantee that
> it's exactly 32 bits.

I have to wonder how come people have not noticed my request to
*discuss it in the python-numerics sig*, which has a parallel discussion
to this one.
-- 
"I'll be ex-DPL soon anyway so I'm        |LUKE: Is Perl better than Python?
looking for someplace else to grab power."|YODA: No...no... no. Quicker,
   -- Wichert Akkerman (on debian-private)|      easier, more seductive.
For public key, finger moshez at debian.org  |http://www.{python,debian,gnu}.org



From claird at starbase.neosoft.com  Tue Mar 20 13:53:05 2001
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 20 Mar 2001 12:53:05 -0600
Subject: Accessing Oracle from Python remotely ?
References: <Xns906AB9C901807PaCmAnRDLM@161.48.76.20> <mailman.985110929.20467.python-list@python.org>
Message-ID: <D04FCD180EE2F89A.577B8E2C5584DA8B.2ECADA1537F68948@lp.airnews.net>

In article <mailman.985110929.20467.python-list at python.org>,
Bill Scherer  <Bill.Scherer at VerizonWireless.com> wrote:
>Fred - 
>I've not used it with oracle v7, but DCOracle works very well with
>Oracle 8x.
>It is available as part of the ZOracleDA package for Zope
>(http://www.zope.org).  You do not need to install zope to use it.  Just
>get the ZOracleDA package and untar it.  Follow the directions inside to
>build DCOracle, then put the DCoracle directory in your PYHTHONPATH. 
>You will need a proper installation of the Oracle client to build
>DCOracle.
			.
			.
			.
The short answer is, YES, you absolutely should be
able to expect to get at an Oracle RDBMS hosted on
a WNT box from anyplace on your network.  Python
has lots of ways to involve itself.  DCOracle is a
good one; just to get started, though, and to ex-
perience some of this for yourself, you might
experiment with command-line Oracle access from
the Solaris workstation.  Python can, of course,
control Oracle that way.

Yet another alternative is through ODBC drivers,
which some people (to my surprise) seem to favor
over DCOracle.
-- 

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


From chrisw at nipltd.com  Thu Mar  8 09:23:54 2001
From: chrisw at nipltd.com (Chris Withers)
Date: Thu, 08 Mar 2001 14:23:54 +0000
Subject: Playing with .zip or .tar.gz files
Message-ID: <3AA795FA.5C222E25@nipltd.com>

Hi,

Can anyone recommend a library to me for playing with .zip or .tar.gz files.
'playing with' includes adding files to them, removing files from, and creating
them.

thanks for any help :-)

Chris



From aleaxit at yahoo.com  Thu Mar 29 03:52:30 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 29 Mar 2001 10:52:30 +0200
Subject: setvbuf() ??
References: <slrn9c50mp.me7.wcuddy@lserv.ja10629.home>
Message-ID: <99ut4h01tb1@news2.newsguy.com>

"Wayne Cuddy" <wcuddy at lserv.ja10629.home> wrote in message
news:slrn9c50mp.me7.wcuddy at lserv.ja10629.home...
> I need to print a period on an xterm every few seconds to provide status
to the
> user, I don't want to force a newline.  How do I turn off line buffering
for
> standard out?

Running the python interpreter with a -u command line switch turns
buffering off.  That is generally the easiest way, if applicable.


Alex





From tino at wildenhain.de  Sun Mar 18 15:34:46 2001
From: tino at wildenhain.de (Tino Wildenhain)
Date: Sun, 18 Mar 2001 21:34:46 +0100
Subject: Serial Port on Unix
References: <3AB637A6.FDEAE8FF@nipltd.com>
Message-ID: <3AB51BE6.64CF103A@wildenhain.de>

Hi Chris,

Chris Withers wrote:
> 
> Hi,
> 
> I need to do some interaction with a serial port on Linux (actually getting an
> SMS from a Mobile Phone attached via a PCMCIA card, if anyone has any better
> ideas of how to setup an SMS target, please let me know ;-)
> 
> Anyway, I had a hunt on python.org without much luck. A few obscure postings and
> a module from the vaults that turned out to be a broken link. What module should
> I be using to do this and where can I get it from?

Add me! Me too! AOL! ;)

If you ever find something usefull, let me know ;) 
Found some usefull solutions for win32 but no luck with *nix so far :(
May be some magic with IOCTL and open/popen?

Regards
Tino



From embed at geocities.com  Fri Mar  2 16:39:06 2001
From: embed at geocities.com (Warren Postma)
Date: Fri, 02 Mar 2001 21:39:06 GMT
Subject: Weird Language Features
References: <nnhv8t4obq30ljfgseplp7pi4khhsf9g9n@4ax.com> <3iTj6.1813$E41.169931@news.uswest.net> <c7109to8ccq3r4gmpqq1eqf9io8co0vhum@4ax.com>
Message-ID: <_pUn6.15968$UZ4.4401842@news4.rdc1.on.home.com>

> It's a separate library that gets to see the source code before it's
> compiled. For an (admittedly extreme) example of what you can do see:
>
> <http://www.csse.monash.edu.au/~damian/papers/HTML/Perligata.html>
>
> Which discusses a filter allowing you to write Perl programs in Latin!

>From the above link, a code sample, illustrating what ought to be a widely
known
programming proverb, that if a language syntax is adequately obtuse, even
comments in Perl may be helpful:

        nexto stringum reperimentum da.     # $next = pos $string;
        nextum stringo reperimento da.      # pos $string = $next;
        inserto stringum tum unum tum duo excerpementum da.
                                    # $insert = substr($string,1,2);
        insertum stringo unum tum duo excerpemento da.
                                    # substr($string,1,2) = $insert;
        keyis hashus nominamentum da        # @keys = keys %hash;
        keya hashibus nominamento da        # keys %hash = @keys;



<GRIN>

De-gustibus-non-disputantem-ly-yrs,

Warren




From kens at sightreader.com  Fri Mar 16 13:41:31 2001
From: kens at sightreader.com (Ken Seehof)
Date: Fri, 16 Mar 2001 12:41:31 -0600
Subject: Should I close after popen??
References: <XFMail.20010315142704.shaleh@valinux.com> <mailman.984697040.2458.python-list@python.org> <m3lmq6ppv0.fsf@master.athome> <98satl$1t8$0@216.39.151.169>
Message-ID: <008401c0ae48$b8edbb80$04090a0a@upcast.com>

----- Original Message ----- 
From: "Donn Cave" <donn at oz.net>
Newsgroups: comp.lang.python
To: <python-list at python.org>
Sent: Friday, March 16, 2001 12:13 AM
Subject: Re: Should I close after popen??

> . . .
> Files normally (in C Python, anyway) close implicitly when they
> go out of scope and no reference to them remains in the system.
> There's nothing wrong with an explicit close, though, and in
> popen() files it also releases the process slot, which until
> then (or until the parent's exit)  will appear as a "zombie".
> 
> Donn Cave, donn at oz.net
> -- 
> http://mail.python.org/mailman/listinfo/python-list

More specifically, they close when garbage collected, which
is at some unknown time -after- they go out of scope and no
reference to them ramains on the system.  Usually this distinction
doesn't matter, but there are some cases where this can get you
into trouble, and the result could be a nasty intermittent bug.

In general, never rely on destructor timing in python.  It was ok
in C++, but it won't always work in python.

- Ken





From nluken at NOSPAMearthlink.net  Sat Mar 10 20:08:45 2001
From: nluken at NOSPAMearthlink.net (noah luken)
Date: Sun, 11 Mar 2001 01:08:45 GMT
Subject: newbie needs help installing NumPy, reference material.
Message-ID: <xeAq6.3544$R_6.353426@newsread2.prod.itd.earthlink.net>

Hey everyone,

Just trying to get numeric python installed on a win2000 machine with python
2.0, Idle, etc installed (but without a c compiler). I've downloaded
http://download.sourceforge.net/numpy/NumPy1711_20.zip, but am unclear about
whether it's the right version (there's an 18.0 version) and how to install
it.  The documentation at www.python.org was a little outdated.

Also, what are the good references (either online or printed) for python and
windows?  I'm a design student with next to no programming experience
interested in using python with tkinter, glut, opengl, etc. for 3d and 2d
graphic work -- any suggestions for where to get started?

Thanks in advance for any help,
noah
nluken at earthlink.net




From jafo at tummy.com  Sat Mar 17 04:31:19 2001
From: jafo at tummy.com (Sean Reifschneider)
Date: Sat, 17 Mar 2001 02:31:19 -0700
Subject: how to indent blocks in vi using << and >>
In-Reply-To: <98uo2s616np@news2.newsguy.com>; from nspurrier@tummy.com on Sat, Mar 17, 2001 at 04:10:04AM +0000
References: <98uo2s616np@news2.newsguy.com>
Message-ID: <20010317023119.H29139@tummy.com>

On Sat, Mar 17, 2001 at 04:10:04AM +0000, nspurrier at tummy.com wrote:
>The >> indents a line right one tab and << indents a line left 
>one tab. You can indent a whole bunch of lines by giving a number

More precisely, ">" indents one *shift-width*.  If, for example you
do ":set sw=2 ts=8", that set's the swift width to 2 and the tabs to
8.  So, if you do a ">>", it will move it over two spaces while
leaving tabs set at 8.  I prefer ":set ts=3 sw=3" myself...

You say this is necessary for Python coding -- I actually found it just
as useful for coding in C and other languages.  People always seem
suprised by the fact that Python uses white-space for blocks.  I thought
it was weird before I used Python, but my coding style is so consistent
that I rarely have a problem (except when editing somone else's code).

>the number before the >> or <<. For example, 10>> will indent
>ten lines from the current cursor position.

Even better, you can move to the end of a block of code you want
to indent, do "ma", then move to the beginning and do "<'a" to
dedent that block, or ">'a" to indent it.  You can also do ">}"
to indent from the cursor to the next blank line, and ">{" to
indent to the PREVIOUS blank line.  Lotsa good stuff in there.
Don't forget "." to repeat the previous indent/dedent...

Sean
-- 
 Rocky: "I must be getting near-sighted!  You look all fuzzy..."
 Bullwinkle: "Let's face it, Rock...  I *AM* all fuzzy."
Sean Reifschneider, Inimitably Superfluous <jafo at tummy.com>
tummy.com - Linux Consulting since 1995. Qmail, KRUD, Firewalls, Python



From moshez at zadka.site.co.il  Tue Mar 20 15:19:12 2001
From: moshez at zadka.site.co.il (Moshe Zadka)
Date: Tue, 20 Mar 2001 22:19:12 +0200
Subject: What IDE are you'all using for Python 2.1 for Windows?
In-Reply-To: <memo.20010320184015.54039E@digitig.compulink.co.uk>
References: <memo.20010320184015.54039E@digitig.compulink.co.uk>, <mailman.985033225.14770.python-list@python.org>
Message-ID: <E14fSb1-0000pv-00@darjeeling>

On Tue, 20 Mar 2001, digitig at cix.co.uk (Tim Rowe) wrote:

> triple-quoted strings (one of the things I consider unpythonic that seems 
> to have been in Python from the start, FWIW!)

It hasn't been in Python from the start. It was added when people found
it very very painful to emit large strings to files.

Compare:

message = '''\
From: %(origin)s
To: %(dest)s
Subject: triple quotes in Python

Why are there triple quoted strings
in Python?
''' % vars()
os.popen("/usr/lib/sendmail -oi -t", 'w').write(message)

To

message = (
"From: %(origin)s\n"
"To: %(dest)s\n"
"Subject: triple quotes in Python\n"
"\n"
"Python doesn't need no\n"
"stinking triple quotes\n" ) % vars()
os.popen("/usr/lib/sendmail -oi -t", 'w').write(message)

Which version would you feel safer letting a non-programmer modify?
-- 
"I'll be ex-DPL soon anyway so I'm        |LUKE: Is Perl better than Python?
looking for someplace else to grab power."|YODA: No...no... no. Quicker,
   -- Wichert Akkerman (on debian-private)|      easier, more seductive.
For public key, finger moshez at debian.org  |http://www.{python,debian,gnu}.org



From fredrik at pythonware.com  Fri Mar 16 11:26:58 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 16 Mar 2001 16:26:58 GMT
Subject: Off-python-topic:  Cron jobs on linux...help
References: <mailman.984684081.4438.python-list@python.org>
Message-ID: <m9rs6.167$sk3.49822@newsb.telia.net>

Chris Richard Adams wrote:
> I'd like to ask two things about the cron tool that comes with Linux -
> have RedHat 6.2.

not written in Python, is it?

> 1.) I have a script that runs fine when I am logged in a the user who
> owns it.  I have env variables that are needed and these are loaded in
> the .bash_profile as that user logs in.  As that user, I set a cron job
> putting the pathname of the script.  It runs, but fails because it
> appears the  environment vars are not set.  I can work around it by
> putting all the vars in the script of course, but what might be the
> problem?
>
> 2.) I want the script to run each 20 minutes.  With only the 5 time
> parameters to set, I can't do this?

$ man 5 crontab

Cheers /F




From fredrik at pythonware.com  Sat Mar 24 06:34:14 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Sat, 24 Mar 2001 11:34:14 GMT
Subject: Please rate these Python books
References: <20010322.1445209@mis.configured.host> <_Meu6.8736$8O6.1818817@typhoon.austin.rr.com> <slrn9bjuha.4qv.grey@teleute.dmiyu.org> <blzu6.8974$G72.207211@e420r-atl1.usenetserver.com> <vwWu6.2238$45.10474@newsfeeds.bigpond.com>
Message-ID: <WC%u6.1386$sk3.452235@newsb.telia.net>

Brett Haydon wrote:
> > Prime collection is currently "Learning Python" (despite its age), "Win32
> > Programming in Python" and "Python Essential Reference". The rest have
> > their place, but it's in the bookshelf, not the briefcase.
>
> Looking foward to the new OReilly Python book, though I fear fitting it
> in the briefcase may be out of the question.

it's only 300 pages or so, so that shouldn't be much
of a problem...  oh, you meant PP2E? ;-)

Cheers /F




From bwalker at earth1.net  Tue Mar 27 19:22:11 2001
From: bwalker at earth1.net (Bill Walker)
Date: Tue, 27 Mar 2001 19:22:11 -0500
Subject: Some Basic Questions about Python
References: <tbv8g06eg66tec@corp.supernews.com> <E14hfrj-0005CB-00@darjeeling> <mailman.985653995.22464.python-list@python.org>
Message-ID: <tc2bcm86autc6d@corp.supernews.com>

OK, I downloaded the installer and tried it. Works like a charm. Some
questions:

(For reference, the source code of my test program is at the bottom)

1. I compiled using Standalone and it created a directory with all the
needed stuff. Was kinda big, over a megabyte. It zipped up to 877K with
Winzip, so
it will all fit on a disk. During compilation, it immediately spit out an
error message that it couldn't find my dll files, however, it went ahead and
compiled anyway. Is this why the directory was so big? I have
\Windows\System in my path.

2. I compiled both with a test.py and then a test.pyw filename. Both
compilations gave me a test.exe that generated that annoying dos console in
the background. How can I get rid of it?

3. Also, I tried using the -tk option as outlined in the instructions on the
web page docs, but it flagged it as an error. However, it seemed to compile
fine without the option, so I guess I don't really need it?

All-in-all, it's a great solution for me.

Thanks a million!


>
> Sure you do.  It's
http://starship.python.net/crew/gmcm/installer_dnld.html
>
> It makes executables for Windows and Linux.  Typical results on Windows
> are around 500K depending on what modules you import.  I haven't used
> it on Linux.
>
> - Ken
>
>
>

#test

import Tkinter
from math import sin,pi

tk=Tkinter
root = tk.Tk()
root.title("Space Zounds Lab")

wHeight=300
wWidth=500
wBorder=20

myCanvas=tk.Canvas(root, height=wHeight, width=wWidth, bg='white')
myCanvas.create_line(wBorder,wHeight/2,wWidth-wBorder,wHeight/2,fill='green'
)
myCanvas.create_line(wWidth/2,wBorder,wWidth/2,wHeight-wBorder,fill='green')

for a in range(0,629):
   x1 = a/pi/200.*(wWidth-2*wBorder)+wBorder
   y1 = wHeight/2 - sin(a/100.)*(wHeight-wBorder)/2
   if (a>0): myCanvas.create_line(x1, y1, x2, y2, fill='red')
   x2 = x1
   y2 = y1

myCanvas.pack()

root.mainloop()





From dinu at reportlab.com  Fri Mar 30 02:54:30 2001
From: dinu at reportlab.com (Dinu Gherman)
Date: Fri, 30 Mar 2001 07:54:30 GMT
Subject: latitude/longitude/mapping Python script
References: <3AC40E78.D57EB9BE@world.std.com>
Message-ID: <3ac43b22.398060200@news.vr.in-berlin.de>

On Thu, 29 Mar 2001 23:41:28 -0500, Will Ware <wware at world.std.com>
wrote:

>This is a multi-part message in MIME format.
>--------------1FF6E68DDA96A9621DE8BB6C
>Content-Type: text/plain; charset=us-ascii
>Content-Transfer-Encoding: 7bit
>
>import string, urllib, re, os
>
>JUST_THE_US = 0
>
>def xerox_parc_url(marklist):
>    avg_lat, avg_lon = max_lat, max_lon = marklist[0]
>    marks = "%f,%f" % marklist[0]
>    for lat, lon in marklist[1:]:
>        marks = marks + ";%f,%f" % (lat, lon)
>        avg_lat = avg_lat + lat
>        avg_lon = avg_lon + lon
>        if lat > max_lat: max_lat = lat
>        if lon > max_lon: max_lon = lon
>    avg_lat = avg_lat / len(marklist)
>    avg_lon = avg_lon / len(marklist)
>    if len(marklist) == 1:
>        max_lat, max_lon = avg_lat + 1, avg_lon + 1
>    diff = max(max_lat - avg_lat, max_lon - avg_lon)
>    D = {'height': 4 * diff,
>         'width': 4 * diff,
>         'lat': avg_lat, 'lon': avg_lon,
>         'marks': marks}
>    if JUST_THE_US:
>        url = ("http://pubweb.parc.xerox.com/map/db=usa/ht=%(height)f" +
>               "/wd=%(width)f/color=1/mark=%(marks)s/lat=%(lat)f/" +
>               "lon=%(lon)f/") % D
>    else:
>        url = ("http://pubweb.parc.xerox.com/map/color=1/ht=%(height)f" +
>               "/wd=%(width)f/color=1/mark=%(marks)s/lat=%(lat)f/" +
>               "lon=%(lon)f/") % D
>    return url
>
>def findcity(city, state):
>    Please_click = re.compile("Please click")
>    url = ("""http://www.astro.ch/cgi-bin/atlw3/aq.cgi?expr=%s&lang=e"""
>           % (string.replace(city, " ", "+") + "%2C+" + state))
>    lst = [ ]
>    inf = urllib.URLopener().open(url)
>    for x in inf.readlines():
>        quitsoon = 0
>        if Please_click.search(x) != None:
>            lst = [ ]
>            quitsoon = 1
>        L = [ ]
>        for y in string.split(x[:-1], '<'):
>            L = L + string.split(y, '>')
>        L = filter(lambda x: len(x) > 0, L)
>        lst.append(L)
>        if quitsoon and len(lst) == 20:
>            break
>    inf.close()
>    a_href = re.compile("a href")
>    okay = 0
>    for x in lst:
>        for y in x:
>            if a_href.search(y) != None:
>                okay = 1
>                break
>        if okay:
>            break
>    assert okay
>    lat, lon = x[6], x[10]
>    def getdegrees(x, dividers):
>        if string.count(x, dividers[0]):
>            x = map(eval, string.split(x, dividers[0]))
>            return x[0] + (x[1] / 60.)
>        elif string.count(x, dividers[1]):
>            x = map(eval, string.split(x, dividers[1]))
>            return -(x[0] + (x[1] / 60.))
>        else:
>            raise "Bogus result", x
>    return getdegrees(lat, "ns"), getdegrees(lon, "ew")
>
>def showcities(citylist):
>    marklist = [ ]
>    for city, state in citylist:
>        lat, lon = findcity(city, state)
>        print ("%s, %s:" % (city, state)), lat, lon
>        marklist.append((lat, lon))
>    url = xerox_parc_url(marklist)
>    os.system('netscape "%s"' % url)
>
>citylist = (("Munich", "Germany"),)  # changed from original posting
>
>showcities(citylist)

% python map.py
Traceback (most recent call last):
  File "map.py", line 85, in ?
    showcities(citylist)
  File "map.py", line 77, in showcities
    lat, lon = findcity(city, state)
  File "map.py", line 72, in findcity
    return getdegrees(lat, "ns"), getdegrees(lon, "ew")
  File "map.py", line 65, in getdegrees
    x = map(eval, string.split(x, dividers[0]))
  File "<string>", line 1
    08
    ^
SyntaxError: unexpected EOF while parsing

Dinu

-- 
Dinu C. Gherman
dinu at reportlab dot com
http://www.reportlab.com
................................................................
"The only possible values [for quality] are 'excellent' and 'in-
sanely excellent', depending on whether lives are at stake or 
not. Otherwise you don't enjoy your work, you don't work well, 
and the project goes down the drain." 
                    (Kent Beck, "Extreme Programming Explained")


From flognat at flognat.myip.org  Thu Mar 15 18:55:05 2001
From: flognat at flognat.myip.org (Andrew Markebo)
Date: 16 Mar 2001 00:55:05 +0100
Subject: Should I close after popen??
References: <XFMail.20010315142704.shaleh@valinux.com> <mailman.984697040.2458.python-list@python.org> <m3lmq6ppv0.fsf@master.athome>
Message-ID: <m366hapoyu.fsf@master.athome>

I decided to see what happened :-) So I did some experimenting.

The files seems to be closed, I checked /proc/[pid]/fd and when I
do the popen call, and chew on the fd's I see two extra files there,
when I was finished with the fd's they are gone.

Thumbs up.

        /Andy

p.s. Still worried though.. is it 'clean', I always close the doors
        after me.. You know.. raised by strict parents and so
        on.. keeping down the heat bills.


From hernan.foffani at iname.com  Fri Mar 30 10:42:38 2001
From: hernan.foffani at iname.com (Hernan M. Foffani)
Date: 30 Mar 2001 15:42:38 GMT
Subject: triple-quote :-)
Message-ID: <20010330104238.114$ul@newsreader.com>

Regarding the "Technical Advisory Board. Joining the team is: Larry Wall,
Brendan Eich, Guido van Rossum, Rasmus Lerdorf, and Jon Udell" at
http://www.activestate.com/Corporate/Communications/Releases/Press985283476
.html

Did you see in the press news that Guido (and only he) was triple-quoted on
his comments? How did he manage to do that? How he pronounce so he is
triple-quoted?

Many :-)'s

-Hernan

PS: I guess that Lori has a very nice sense-of-humor.

-- 
-------------------- http://NewsReader.Com/ --------------------
                       Usenet for the Web


From Andreas.Leitgeb at siemens.at  Mon Mar 12 08:10:30 2001
From: Andreas.Leitgeb at siemens.at (Andreas Leitgeb)
Date: Mon, 12 Mar 2001 13:10:30 +0000 (UTC)
Subject: Einstein's Riddle
References: <oNPq6.396644$ge4.138479566@news2.rdc2.tx.home.com>
Message-ID: <slrn9apj4g.c54.Andreas.Leitgeb@pc7499.gud.siemens.at>

On Sun, 11 Mar 2001 18:49:56 GMT, Steve Graham <js.graham at home.com> wrote:
>Who would be interested in using his/her brain (and his computer) to solve
>the following?
>
>Einstein's Riddle
> [riddle's description snipped]

in Germany, there's a magazine called "PM", which has one riddle of
this structure in (almost) each issue. ("Logik Trainer")

based on this kind of riddles, I've created a  tcl/tk-prog, that
helps one solving it. 

though, there are still two open problems:
  -) it is still possible to enter contradicting information.
      (this means, that not yet all logical consequences for
       each entered information are obtained - actually i have 
       a clear idea of what's missing, but i haven't coded it)
  -) because Tk cannot yet write vertical texts,  
      it is not possible to put label for the columns, which
      makes it tough to remember  what each column represents.
      Actually, I haven't even realized the horizontal row-labels, yet :-)

If someone is interested in this kind of riddles, then mail me.
(I for myself have declined from putting any further work into it, lately)



From fredrik at pythonware.com  Mon Mar 26 14:28:12 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Mon, 26 Mar 2001 19:28:12 GMT
Subject: Please rate these Python books
References: <20010322.1445209@mis.configured.host> <_Meu6.8736$8O6.1818817@typhoon.austin.rr.com> <slrn9bjuha.4qv.grey@teleute.dmiyu.org> <blzu6.8974$G72.207211@e420r-atl1.usenetserver.com> <vwWu6.2238$45.10474@newsfeeds.bigpond.com> <WC%u6.1386$sk3.452235@newsb.telia.net> <GAt0p3.M51@leeds.ac.uk>
Message-ID: <gLMv6.3691$4N4.699120@newsc.telia.net>

Nick Eff-ord wrote:
> > it's only 300 pages or so, so that shouldn't be much
> > of a problem...  oh, you meant PP2E? ;-)
>
> ....Speaking of which, when is your book out?  The O'Reilly
> website says April on one page, May on another.

either late april, or early may.  for the moment, your guess
is as good as mine ;-)

Cheers /F




From aahz at panix.com  Tue Mar 27 09:59:50 2001
From: aahz at panix.com (Aahz Maruch)
Date: 27 Mar 2001 06:59:50 -0800
Subject: PEP 1, PEP Purpose and Guidelines
References: <mailman.985196613.7994.python-list@python.org> <99ifgu$jei$1@panix3.panix.com> <200103241546.KAA19937@panix3.panix.com> <mailman.985679316.32485.python-list@python.org>
Message-ID: <99q9t6$5un$1@panix6.panix.com>

In article <mailman.985679316.32485.python-list at python.org>,
Barry A. Warsaw <barry at digicool.com> wrote:
>>>>>> "AM" == Aahz Maruch <aahz at panix.com> writes:
>
>    AM> [posted & e-mailed]
>
>[Aside: say Aahz, when you post and email, can you do it with CC's so
>it's easier to follow up? -baw]

Not easily; I'm reading from the newsgroup, not the mailing list.

>    >>  Note the addition of the Replaces: and Replaced-By: headers
>    >> for formalizing the PEP replacement policy (thanks to Andrew
>    >> Kuchling for the idea and patch).
>
>    AM> After taking a closer look at Moshe's numeric proposals, I
>    AM> suggest adding a Depends-On: header.  (E.g. PEP 240 depends on
>    AM> PEP 239.)
>
>What about Requires:?  Or Dependencies: where each entry on the line
>would be a PEP number (which we could hyperlink fairly easily).

Those would certainly be fine; I'm not wedded to any particular naming
scheme.  And, yes, I was assuming that the field could contain multiple
entries.
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

"Boost the stock market -- fire someone"


From andreas at andreas-jung.com  Sat Mar  3 22:25:41 2001
From: andreas at andreas-jung.com (andreas at andreas-jung.com)
Date: Sat, 3 Mar 2001 22:25:41 -0500
Subject: Java guy interested in Python
References: <003801c0a459$a29c97f0$63da1542@wisdom>
Message-ID: <00ac01c0a45a$f14cad80$0b0a0a0a@SUXLAP>

see http://www.python.org/doc/Comparisons.html

----- Original Message -----
From: "Kyle Wayne Kelly" <kknbk2 at bellsouth.net>
To: <python-list at python.org>
Sent: Saturday, March 03, 2001 10:17 PM
Subject: Java guy interested in Python


Can anyone tell me some of the things that Python offers which are not
available in Java?  I like Python's syntax, but what can Python do that Java
can not do?  Any comments are helpful.





From bbollenbach at homenospam.com  Fri Mar  9 17:57:34 2001
From: bbollenbach at homenospam.com (Brad Bollenbach)
Date: Fri, 09 Mar 2001 22:57:34 GMT
Subject: Can Python replace Visual Basic? Should it?
References: <5DZo6.11126$hn5.1578410@news1.rdc1.mb.home.com> <mailman.984010931.26257.python-list@python.org> <iSBp6.14005$hn5.2035325@news1.rdc1.mb.home.com> <mailman.984094514.29514.python-list@python.org> <984133550.452523@newsmaster-04.atnet.at>
Message-ID: <yddq6.193$rB2.38693@news1.rdc1.mb.home.com>

[snip]

> If you are programming for purity in code, if OO is theoretical stuff for

Purity as in "clear, clean, concise code" should be the norm, not the
exception. If your production code is ugly, start refactoring (I do on a
daily basis). If you let it /stay/ ugly, don't pretend that it's not costing
you more money than putting the effort into doing things (almost) right the
first time.

> you consisting of inheritance, polymorphism and encapsulation, or if a

Theoretical stuff? The app that pays my bills, HEAVILY incorporates the
"theory" of all three of those fundamental OO principles. And, not as
importantly in "the industry" admittedly, OO is more fun to program in for
me (if it's not to you, good job, no flames on that please).

> really wonderful intuitiv IDE with far the best autocomplete and library

The autocomplete in VB IS truly sweet. Very, very, very sweet no doubt about
it. Unfortunately, this says NOTHING about the usefulness of a language.

> browsing features I have seen so far is just pain for you because you
prefer
> coding the GUI in plain text -> do not consider Visual Basic

I like coding in text editors or "light IDE's" such as PythonWin. For
specific reasons that I've already cited in another post, IDE's just seem to
get in my way (admittedly, this is much more true of my Powerbuilder
experiences than my VB ones, but the Powerbuilder app I work on is about
5-6x more complicated).

> Otherwise its a neat tool for the purposes mentioned.
>
> I did not do much COM stuff in Python right now, but I think you cannot
> compare it to Visual Basic at this time.
[snip]

You've hardly touched COM with Python, but "I think you cannot compare it to
Visual Basic at this time"? Let the reader decide whether they value such a
well-informed opinion. :)




From aahz at panix.com  Wed Mar 21 15:49:32 2001
From: aahz at panix.com (Aahz Maruch)
Date: 21 Mar 2001 12:49:32 -0800
Subject: learning python...
References: <3AB8F4E1.184DD9C@noaa.gov> <Z97u6.16733$Im6.1761313@newsread1.prod.itd.earthlink.net> <3ab90559_6@goliath.newsfeeds.com>
Message-ID: <99b44s$m5j$1@panix6.panix.com>

In article <3ab90559_6 at goliath.newsfeeds.com>,
Bubba Brains <chocolate at toblerone.yum> wrote:
>
>Forget Learning Python?  I think it's the best reference/tutorial around.  I
>use python once every couple months, and sometimes forget the simple python
>syntax.  The Learning Python book is great for getting up to speed fast, and
>to look up common stuff.  The other OReilly book (Programming Python) is
>absolutely annoying to use.  It's not even a good reference book (good luck
>finding the info you want quickly).  Don't know about the Prentice book.

If you look at http://www.oreilly.com/catalog/python2/ you'll see that
Programming Python just got a second edition; I've heard that there were
a lot of improvements.
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

"I won't accept a model of the universe in which free will, omniscient
gods, and atheism are simultaneously true."  -- M


From scarblac at pino.selwerd.nl  Thu Mar 29 05:04:49 2001
From: scarblac at pino.selwerd.nl (Remco Gerlich)
Date: 29 Mar 2001 10:04:49 GMT
Subject: diff lists
References: <3AC1DD4E.4BA7D421@aon.at>
Message-ID: <slrn9c61vn.rh1.scarblac@pino.selwerd.nl>

Oliver Vecernik <vecernik at aon.at> wrote in comp.lang.python:
> I've got following two lists:
> 
> ['a', 'b', 'c', 'd', 'e', 'f']
> ['e', 'c', 'f']
> 
> I'd like to have the result:
> 
> ['a', 'b', 'd']
> 
> The list need not to be ordered. ['d', 'a', 'b'] will also be ok. What
> is the most effective way to achive that result?

If your lists are going to be small, speed won't be an issue and you can't
beat

c = [element for element in a if element not in b]

for readability, in my opinion. This will usually be fast enough. If the
second list gets huge though, the problem is that 'element not in b' scans
the whole second list if necessary for every element of a. If speed is an
issue, then maybe something with a dictionary like

dict = {}
for element in a:
   dict[element] = element
for element in b:
   if dict.has_key(element): 
      del dict[element]
c = dict.keys() # Is dict.values() faster?

Test the two with lists typical for your program (see the profile.run
function).

-- 
Remco Gerlich


From qrczak at knm.org.pl  Fri Mar  9 14:02:08 2001
From: qrczak at knm.org.pl (Marcin 'Qrczak' Kowalczyk)
Date: 9 Mar 2001 19:02:08 GMT
Subject: Can anyone recomend a good intoduction to C...
References: <mailman.983894288.19965.python-list@python.org> <ewwp6.1466$y6.241005@ruti.visi.com> <slrn9aeblb.cg1.scarblac@pino.selwerd.nl> <djagat4kk3cs9hg0c3k39cudk6goblhkg6@4ax.com> <989mbk$24nuv$1@fido.engr.sgi.com> <slrn9ah66m.k09.qrczak@qrnik.zagroda> <m3hf139kje.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <slrn9aia5g.rmu.qrczak@qrnik.zagroda>

09 Mar 2001 12:47:01 +0000, Michael Hudson <mwh21 at cam.ac.uk> pisze:

> > working-on-Haskell<->Python-binding-ly y'rs
> 
> Really?  I've had thoughts about doing that, but drifted away from
> haskell for a while.  Do you have anything to show for it yet?

I'm putting the current (incomplete) stuff at
<http://qrczak.ids.net.pl/hspython.tar.gz>.

It requires Glasgow Haskell Compiler >= 4.08.2 (older ones don't
support enough of foreign function interface or are too buggy).
nhc98 will be supported as soon as it supports 'foreign export dynamic'
declarations.

It needs <http://www.sourceforge.net/projects/qforeign/>. It won't
need it under ghc >= 5.0 (not released yet).

Some day I will want to use Python as a scripting language embedded
in Haskell programs.

-- 
 __("<  Marcin Kowalczyk * qrczak at knm.org.pl http://qrczak.ids.net.pl/
 \__/
  ^^                      SYGNATURA ZAST?PCZA
QRCZAK


From sheila at spamcop.net  Sun Mar 18 00:45:16 2001
From: sheila at spamcop.net (Sheila King)
Date: Sun, 18 Mar 2001 05:45:16 GMT
Subject: Pep 238 Why new // operator (Non-integer Division)
References: <98tc2s$qir$1@news2.ottawa.cyberus.ca> <mailman.984822021.13503.python-list@python.org> <vag7bt09r54tv05nohtssd0ge891v3fa02@4ax.com> <3AB42890.AD9E9FF0@wwc.com>
Message-ID: <mqi8bt8q6oadhucrot1fcdp38782di0dao@4ax.com>

On Sun, 18 Mar 2001 03:16:03 GMT, Steve Williams <stevewilliams at wwc.com> wrote
in comp.lang.python in article <3AB42890.AD9E9FF0 at wwc.com>:

:           For simplicity's sake I do all my computations in the base-pi number
:system. In
:           this system pi=10.


Um, in base pi, wouldn't pi = 1   ???

--
Sheila King
http://www.thinkspot.net/sheila/
http://www.k12groups.org/



From just at letterror.com  Tue Mar 13 10:15:46 2001
From: just at letterror.com (Just van Rossum)
Date: Tue, 13 Mar 2001 16:15:46 +0100
Subject: main in mac?
References: <20010313095339.19264.00000243@ng-dh1.aol.com>
Message-ID: <3AAE3978.3FCDC1BB@letterror.com>

Park997 wrote:
> 
> Hi,
> 
> I am using Python 2.0 on a PPC mac. I have a module
> that is set up as follows:
> 
> from Numeric import *
> 
> class Epower:
> 
>         3 class functions defined
> 
> if __name__ == '__main__':
> 
>         test class and print output
> 
> This code does nothing if the run all button is pressed
> or if it is imported into interactive python.
> 
> when the above if statement is replaced by:
> 
> if 1:
> 
> the program runs properly with output. I thought that
> if __name__ == '__main__':
> is standard syntax. Does this need to be changed for mac
> Python?

No: if you look at the option popup menu -- which is very well hidden
at the top right corner of your edit window, it looks like an arrow --
there's an option called "run as __main__". This will set the __name__
variable in the script to "__main__".

Hope this helped,

Just


From fredrik at pythonware.com  Tue Mar 20 13:02:36 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Tue, 20 Mar 2001 18:02:36 GMT
Subject: Regular expressions and non-standard character set
References: <ka8y9u01htt.fsf@cheerful.cs.hut.fi>
Message-ID: <0XMt6.2503$4N4.321668@newsc.telia.net>

Petri Mikael Kuittinen wrote:
> I want to match word boundaries using the special sequences \b and \B
> of regular expressions. They work OK when using the "standard"
> alphanumeric set [a-zA-Z0-9_]. But I would like them to work with
> character set which also contains various "national characters"
> e.g. ?, ?, ?, ?, ?, ?, ? etc. and their uppercase equivalents.
>
> Locale doesn't seem to be the proper way to do it

are you sure?

>>> import locale
>>> locale.setlocale(locale.LC_ALL, "")
'Swedish_Sweden.1252'
>>> import re
>>> re.findall(r"\b...\b", "spam, egg, bacon, and ???")
['egg', 'and']
>>> re.findall(r"(?L)\b...\b", "spam, egg, bacon, and ???")
['egg', 'and', '???']

Cheers /F




From phd at phd.fep.ru  Tue Mar 27 11:12:36 2001
From: phd at phd.fep.ru (Oleg Broytmann)
Date: Tue, 27 Mar 2001 20:12:36 +0400 (MSD)
Subject: Whats in your $PYTHONSTARTUP$ ?
In-Reply-To: <FJ2w6.5358$GD.2921338@typhoon.snet.net>
Message-ID: <Pine.LNX.4.33.0103272011500.969-100000@phd.fep.ru>

On Tue, 27 Mar 2001, Paul Magwene wrote:
> Other than import statements, what little programming gems do you folks have
> in your startup files?

   From http://groups.yahoo.com/group/python-list/message/128961:

import rlcompleter, readline
#readline.parse_and_bind('tab: complete')
readline.parse_and_bind("tab: menu-complete")

import os
histfile="%s/.python-history" % os.environ["HOME"]
readline.read_history_file(histfile)

def savehist():
   global histfile
   readline.write_history_file(histfile)

import atexit
atexit.register(savehist)

Oleg.
----
     Oleg Broytmann            http://phd.pp.ru/            phd at phd.pp.ru
           Programmers don't die, they just GOSUB without RETURN.




From fredrik at pythonware.com  Tue Mar 20 14:27:56 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Tue, 20 Mar 2001 19:27:56 GMT
Subject: What IDE are you'all using for Python 2.1 for Windows?
References: <mailman.985033225.14770.python-list@python.org> <memo.20010320184015.54039E@digitig.compulink.co.uk>
Message-ID: <0bOt6.2522$4N4.328117@newsc.telia.net>

Tim Rowe wrote:
> (one of the things I consider unpythonic that seems to have
> been in Python from the start, FWIW!)

$ more Misc/HISTORY

...
==================================
==> Release 1.0.2 (4 May 1994) <==
==================================
...
* A new kind of string literals, surrounded by triple quotes (""" or
'''), can be continued on the next line without a backslash.
...


Cheers /F




From nas at python.ca  Fri Mar 30 17:15:05 2001
From: nas at python.ca (Neil Schemenauer)
Date: Fri, 30 Mar 2001 14:15:05 -0800
Subject: Speedup hint for users of viewcvs package
In-Reply-To: <190551356909.20010330211141@buz.ch>; from gabriel_ambuehl@buz.ch on Fri, Mar 30, 2001 at 09:11:41PM +0200
References: <m14j45Y-000CnEC@artcom0.artcom-gmbh.de> <190551356909.20010330211141@buz.ch>
Message-ID: <20010330141505.A2027@glacier.fnational.com>

Gabriel Ambuehl wrote:
> one might think about going straight with mod_python anyway...

No, you want mod_fastcgi.  Witness:

  $ ab -n 20 http://spud/cgi-bin/viewcvs.cgi/
  ...
  Requests per second:    2.33
  Transfer rate:          20.02 kb/s received

  $ ab -n 20 http://spud/cgi-bin/viewcvs.fcgi/
  ...
  Requests per second:    58.31
  Transfer rate:          501.57 kb/s received


It took about 15 minutes and 15 lines of code to convert
viewcvs.cgi to viewcvs.fcgi.  I have no idea why more people
don't know about FastCGI.

  Neil



From sholden at holdenweb.com  Tue Mar 27 23:51:51 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Tue, 27 Mar 2001 23:51:51 -0500
Subject: Where is the DBI module?
References: <MPG.152ab2bb88e6482d9896a4@news.pacbell.net>
Message-ID: <k9ew6.1083$m04.196355@e420r-atl1.usenetserver.com>

"Owen Ap Owen" <no at thank.you> wrote in message
news:MPG.152ab2bb88e6482d9896a4 at news.pacbell.net...
> I'm confused about something that probably has an obvious answer.
>
> Where is the DBI module?
>
> I can "import DBI" but when I try to find the actual module in order to
> examine it I can't. Am I missing something obvious?

On Win32, I would guess you are loading the "dbi.dll" module from
<installdir>/win32.

regards
 Steve





From nksauter at lbl.gov  Tue Mar  6 12:23:57 2001
From: nksauter at lbl.gov (Nicholas K. Sauter)
Date: Tue, 06 Mar 2001 09:23:57 -0800
Subject: Pickling arbitrary character strings on Windows
Message-ID: <3AA51D2D.3251921E@lbl.gov>

It would be good to be able to pickle strings containing arbitrary ASCII
characters, for example to encode data at the C level.  But on Windows
it is impossible to do a binary unpickle of a string that has an
embedded ASCII 26 (octal 032).  This example code tells the whole story:

from pickle import dump,load

def myfunc(i):
    pro = "I_am_a_" +chr(i)+ "string"
    f = open('try.bin','w'); dump(pro,f,1); f.close()
    g = open('try.bin','r'); str = g.readline()# i=10 is a linefeed
(that's OK)
    print "read string",i,len(str),"characters"

if __name__=="__main__":
    for i in xrange(256):
      myfunc(i)

Some obvious workarounds are to use formatted output (dump(pro,f,0)) or
to use Unix instead of Windows.  But is there any way to implement a
practical binary string pickle on Windows?

Nick Sauter, LBNL
-------------- next part --------------
A non-text attachment was scrubbed...
Name: nksauter.vcf
Type: text/x-vcard
Size: 301 bytes
Desc: Card for Nicholas K. Sauter
URL: <http://mail.python.org/pipermail/python-list/attachments/20010306/6fdc1b7c/attachment.vcf>

From aleaxit at yahoo.com  Fri Mar  9 05:58:02 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 9 Mar 2001 11:58:02 +0100
Subject: class fields vs __getattr__
References: <9898al$j3j$1@news.mathworks.com>
Message-ID: <98acvm01gfa@news2.newsguy.com>

"Joshua Marshall" <jmarshal at mathworks.com> wrote in message
news:9898al$j3j$1 at news.mathworks.com...
> If you try to get a field of an object, it seems __getattr__ is called
> only if no such field exists on the class.  For exmaple:

Yes, this is the defined semantics of __getattr_: it's designed
and documented that way.

> I would have expected ob.x to return 2.  Can anyone explain the
> rationale behind this behavior?

Performance AND simplicity in coding your __getattr__ -- when
called it KNOWS the attribute being requested is not already
defined (in the instance, its class, its bases).


Alex





From chris.gonnerman at usa.net  Mon Mar  5 08:56:15 2001
From: chris.gonnerman at usa.net (Chris Gonnerman)
Date: Mon, 5 Mar 2001 07:56:15 -0600
Subject: ftp a list to a remote server
References: <tCHo6.8449$v5.32212@newsfeeds.bigpond.com>
Message-ID: <006d01c0a57c$112f1580$a100000a@local>

It depends on what exactly you mean by saving a list.  If your list contains
only strings (i.e. as from readlines()) and you want to write it as a text
file,
you might do this:  (this code is 1.5.2 + compatible)

import ftplib, StringIO, string

virtfile = StringIO.StringIO(string.join(mylist, ''))
conn = ftplib.FTP("my.host.com", "username", "password")
conn.storbinary("STOR " + myfilename, virtfile, 1024)
conn.close()

Note that newline termination is expected.  If you don't have newlines at
the
ends of your strings:

virtfile = StringIO.StringIO(string.join(mylist + [ "" ], '\n'))

On the other hand, you might want to pickle your data to the remote file.
Create
virtfile like this:

import ftplib, StringIO, string, pickle

virtfile = StringIO.StringIO('')
pickle.dump(mylist, virtfile)
virtfile.seek(0)

then proceed as above.  Of course for performance you could also use
cStringIO.

----- Original Message -----
From: "Austin Wilson" <austinw at bhpvillage.com>
Newsgroups: comp.lang.python
To: <python-list at python.org>
Sent: Monday, March 05, 2001 2:07 AM
Subject: ftp a list to a remote server


> Hi
>
> I would like to save a list as a text file on a remote server using ftp.
>
> Is this possible? If so, could you please provide some sample code.
>
> Or do I have to save the file locally first then ftp that file to the
remote
> server?
>
> Thanks
> Austin
>
>
> --
> http://mail.python.org/mailman/listinfo/python-list
>




From cribeiro at mail.inet.com.br  Wed Mar 28 00:22:54 2001
From: cribeiro at mail.inet.com.br (Carlos Alberto Reis Ribeiro)
Date: Wed, 28 Mar 2001 02:22:54 -0300
Subject: types.AnyType - describing interfaces with sequences of types
Message-ID: <5.0.2.1.0.20010328011422.02196260@mail.inet.com.br>

My problem - short question:

I would like to have a type description called AnyType, in such a way that 
it always return true when compared for equality or membership against any 
valid type. Do something like this exist? Does it make sense?

   Example:
   >>> IntType is AnyType
   1
   >>> FloatType == AnyType
   1
   >>> "" = AnyType
   0

Long question:

I was tinkering with some of my proxy experiments, trying a lot of 
different ideas to describe interfaces using only plain Python. One of them 
is the following code snippet:

from types import *

class Interface:
   """ base class for interface specification """
   pass

class InterfaceSequence(Interface):
   """ interface to implement sequence-like objects """
   __len__      = ((), (IntType,))
   __getitem__  = ((IntType,), (NoneType,))
   __setitem__  = ((IntType,NoneType), ())
   __delitem__  = ((IntType,), ())

The idea here is to define some member attributes inside the 
InterfaceSequence class. A helper function then scans 
InterfaceSequence.__dict__ to build the proxy object. Take this only as an 
example, please - the issue here is not the interface, but the types the 
are available in the the 'types' module.

My problem now is: how to describe the method signature? I'm using the 
following data structure in my example:

<method-name>    = (<parameter list>, <return-type>)
<parameter-list> = sequence of type
<return-type>    = type

For example, take the __getitem__ method:

   __getitem__  = ((IntType,), (NoneType,))

This means that __getitem__ have one parameter of IntType (for a sequence, 
remember!), and returns a value of... NoneType (?!?!). This would be much 
better:

   __getitem__  = ((IntType,), (AnyType,))

So that's the problem. I would like to use a value of AnyType for cases 
like this, however, I dont have AnyType as a valid type description. What 
is worse, there is no UserType module to extend type definitions.


p.s. I have some approaches being explored for the interface definition. 
This is one of the alternatives. As for the method signature, there are 
some other alternatives that include more information, such as:

a)   __getitem__  = ((('key', IntType), ), (AnyType,))
b)   __getitem__  = ({'key':IntType}, (AnyType,))
c)   __getitem__  = ({'key':IntType}, AnyType)

The last one seems to be the cleanest. It limits the return value 
description to asingle type, but this seems to be enough for my intentions. 
However, the problem with 'AnyType' is still there.


Carlos Ribeiro





From phd at phd.pp.ru  Fri Mar  2 08:36:48 2001
From: phd at phd.pp.ru (Oleg Broytmann)
Date: Fri, 2 Mar 2001 16:36:48 +0300 (MSK)
Subject: Intersesting Questions involving extending and embedding
In-Reply-To: <3a9f00e8.2826131@news.digital.net>
Message-ID: <Pine.LNX.4.30.0103021635370.7708-100000@fep132.fep.ru>

On Fri, 2 Mar 2001, Dev_NuLL wrote:
> >1) last seen here: http://lima.mudlib.org/~rassilon/p2c
>
> bunk URL

   http://www.mudlib.org/~rassilon/p2c, but th stuff is old - Aug 1999.

Oleg.
----
     Oleg Broytmann            http://phd.pp.ru/            phd at phd.pp.ru
           Programmers don't die, they just GOSUB without RETURN.




From mwh21 at cam.ac.uk  Thu Mar 22 07:30:05 2001
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 22 Mar 2001 12:30:05 +0000
Subject: What is being compared?
References: <mailman.985222828.15373.python-list@python.org>
Message-ID: <m366h2f0ky.fsf@atrus.jesus.cam.ac.uk>

"Tim Peters" <tim.one at home.com> writes:

> [Daniel Klein]
> > Comparing a string to an integer...
> >
> > >>> 'a' < 1
> > 0
> >
> > ...what is it actually comparing? Is it 'casting' anything?
> 
> No cast, but what is being compared isn't defined, just that cross-type
> comparisons (except for cross-type all-numeric comparisons) are "consistent".
> As a matter of implementation accident, CPython actually compares the string
> names of the *types* in such cases, so the accident du jour is that
> cmp(some_string, some_int) == cmp("string", "int"), which is greater than 0
> (i.e., any string is "bigger than" any int -- today).

Well, apart from the fact that numeric types get a name of "" in such
considerations, so that you don't get absurdities like

(1 < [] < 1L) == 1

just because "int" < "list" < "long".

> not-the-whole-truth-but-there's-not-much-future-seeking-truth-
>     in-accidents-anyway-ly y'rs  - tim

Quite.  Relying on calling .sort() on a list to get all the numbers to
the front is probably ill-advised (even before complex numbers started
throwing exceptions on comparisons!).

Cheers,
M.

-- 
  If your telephone company installs a system in the woods with no
  one around to see them, do they still get it wrong?
                                 -- Robert Moir, alt.sysadmin.recovery


From stephen_purcell at yahoo.com  Mon Mar 26 10:08:35 2001
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Mon, 26 Mar 2001 17:08:35 +0200
Subject: daemon app in win32
In-Reply-To: <99nk49$a2q$02$1@news.t-online.com>; from Mike_B@T-Online.de on Mon, Mar 26, 2001 at 04:35:52PM +0200
References: <99nk49$a2q$02$1@news.t-online.com>
Message-ID: <20010326170835.B16310@freedom.puma-ag.com>

Michael Bauer wrote:
> i am writing a daemon application that runs both under windows and unix.  
> Since i am not very experienced in win32 - programming, i really would 
> appreciate some hints how to register my app in win32 that i can do some 
> cleanup-functions when windows is shut down.
> 
> At the moment it is just a standard console application that simply gets 
> killed when windows shuts down. The app does not have/need a GUI, it should 
> just run in the background.


If you're using NT, take a look at the win32service API. I think it provides
hooks like SvcStop that Windows will call for you (if you ask it nicely).

-Steve

-- 
Steve Purcell, Pythangelist
Get testing at http://pyunit.sourceforge.net/
Any opinions expressed herein are my own and not necessarily those of Yahoo



From william.famy at noos.fr  Sun Mar 25 14:55:37 2001
From: william.famy at noos.fr (William Famy)
Date: Sun, 25 Mar 2001 21:55:37 +0200
Subject: difference between script and prompt ! looking for the Problem thx
References: <3ABCCA03.90871133@noos.fr> <99iht4$1c7hg$1@ID-11957.news.dfncis.de> <3ABCD57F.DC395F88@noos.fr> <99ikon$1dbnf$1@ID-11957.news.dfncis.de>
Message-ID: <3ABE4D39.76AD138@noos.fr>

Thanks for the reply.

It is just a minor problem.

I have to learn a lot of think about this language ;-)

William

Emile van Sebille wrote:
> 
> Well, you're printing the individual attributes.  (3.0, 4.5) is tuple
> representation.  You can print the tuple by using:
> 
> print (c.r, c.i)
> 

-- 
 			     \\\|///
                           \\  - -  //
                            (  @ @  )
+-------------------------oOOo-(_)-oOOo-------------------------+
|  Famy William  36 avenue des barattes  74000 Annecy  France   |
|  email:william.famy at noos.fr william.famy at mail-enseignant.com  |
+---------------------------------Oooo--------------------------+
                           oooO   (   )
                          (   )    ) /
                           \ (    (_/
                            \_)


From matt at mondoinfo.com  Mon Mar 12 20:12:26 2001
From: matt at mondoinfo.com (Matthew Dixon Cowles)
Date: Tue, 13 Mar 2001 01:12:26 GMT
Subject: This math scares me
References: <3aad4062.1290021@News.CIS.DFN.DE> <Hwbr6.1139$Tg.127571@ruti.visi.com> <3aad4485.2348499@News.CIS.DFN.DE> <DZbr6.17917$Qb7.3117518@newsb.telia.net> <zrcr6.1144$Tg.129254@ruti.visi.com>
Message-ID: <slrn9aqsvq.le.matt@happy-hour.mondoinfo.com>

On Mon, 12 Mar 2001 22:53:19 GMT, Grant Edwards <grante at visi.com>
wrote:

>People should have to pass a test before they're allowed to use
>floating point numbers.  1/2 ;)

But this is Python so you're not winking at all, Grant <wink>.

Matt


From scarblac at pino.selwerd.nl  Tue Mar  6 09:50:43 2001
From: scarblac at pino.selwerd.nl (Remco Gerlich)
Date: 6 Mar 2001 14:50:43 GMT
Subject: feature wish
References: <EZPn6.325779$f36.11597531@news20.bellglobal.com>
Message-ID: <slrn9a9u6k.9ag.scarblac@pino.selwerd.nl>

Dan Parisien <dan at eevolved.com> wrote in comp.lang.python:
> it seems all the special purpose overloadable methods are to provide type 
> emulation through classes
> e.g.
> __nonzero__
> __contains__
> __setitem__
> etc.
> 
> it would be cool if there was another method provided that allowed a 
> programmer to overload the value returned by an instance as well.

No; the instance *is* a value, it returns nothing. If you want something to
be a tuple, make it a tuple ;).

> class example:
>         def __init__(self, name, data):
>                 self.name = name
>                 self.data = data
> 
>         def __value__(self):
>                 return self.data
> 
> eg = example("my name", (1,2,3))
> 
> >>>print eg
> (1,2,3)
> 
> You could get to the instance behind the value by using a function like
> instance(eg)

You can make eg *look* like a tuple by providing __repr__ and __str__
functions that return (for instance) repr(self.data).

-- 
Remco Gerlich


From neotaku at free.fr  Mon Mar  5 13:41:29 2001
From: neotaku at free.fr (Neotaku)
Date: Mon, 05 Mar 2001 18:41:29 GMT
Subject: Access to sybase API
References: <obOo6.1158$f07.1696858@nnrp6.proxad.net> <3AA3B0D0.90876BA7@schlund.de>
Message-ID: <t5Ro6.1199$yg6.1667112@nnrp6.proxad.net>

    Thanks a lot

"Carsten Gaebler" <cg at schlund.de> a ?crit dans le message news:
3AA3B0D0.90876BA7 at schlund.de...
> Neotaku wrote:
> > I search a code source for python to access Sybase thru API connexion
the
> > Sybase DB-Library API or Sybase CT-Library (aka the Client Library) API
>
> Have a look at Dave Cole's Sybase module:
> http://www.object-craft.com.au/projects/sybase/
>
> cg.
>



From Philippe_Rodriguez at bevoux.grenoble.hp.com  Mon Mar 26 06:25:55 2001
From: Philippe_Rodriguez at bevoux.grenoble.hp.com (Philippe Rodriguez)
Date: Mon, 26 Mar 2001 12:25:55 +0100
Subject: How to set PYTHONHOME ???
Message-ID: <3ABF2743.ED5904F2@bevoux.grenoble.hp.com>

Hello,

I'm running Python 1.6 on hpux 10.20. I got the standard depot binary
release.

Each time I'm trying to run Python, I have the following message :
---
Could not find platform dependent libraries <exec_prefix>
Consider setting $PYTHONHOME to <prefix>[:<exec_prefix>]
---
I have set this to every conceivable option I can think of, but it still
complains.
The program seems to be OK, but my tool which use python runs a cron
every minutes
 ans so I'm receiving a message every minute ... boring !

Any ideas?

Thanks very, very much for your help !


Phil.



From kens at sightreader.com  Fri Mar 16 13:08:26 2001
From: kens at sightreader.com (Ken Seehof)
Date: Fri, 16 Mar 2001 12:08:26 -0600
Subject: Cookies
References: <mailman.984707839.22839.python-list@python.org> <lmq6m07j.fsf@propylon.com>
Message-ID: <006801c0ae44$195cb890$04090a0a@upcast.com>

Thanks for your reply.

Creating a unique id from a random number is fine.  The problem is that
there seems to be no way to communicate this number to the standalone python
app.  Thanks for the heads up on the IP address non-uniqueness problem.

Here's a restatement of the problem.  It's really about crossing the sandbox
barrier.

Does anyone know of another mailing list more specific to http questions?

1. Client downloads an execuatable (frozen python application).
2. Client downloads a web page.
3. Web page javascript generates a statistically unique random identifier.
4. Web page sends identifier to the server, which uses it as database key.
5. User runs python application
6. *** Python application somehow gets the identifier from somewhere ***
7. Now the python application and the server both have the same identifier.

The only question I am asking is how to do step 6.

Alternative:

1. Client downloads an execuatable (frozen python application).
2. Client downloads a web page.
3. *** Web page javascript makes a unique identifier for the client. ***
4. Web page sends identifier to the server, which uses it as database key.
5. User runs python application
6. *** Python application somehow generates the same identifier ***
7. Now the python application and the server both have the same identifier.

Again the only question I am asking is how to do steps 3 and 6 to get the
same number.

----- Original Message -----
From: "Martin Skott" <martin.skott at propylon.com>
Newsgroups: comp.lang.python
To: <python-list at python.org>
Sent: Friday, March 16, 2001 5:18 AM
Subject: Re: Cookies


> "Ken Seehof" <kens at sightreader.com> writes:
> > Alternate question:
> >
> > Is there a technique common to both javascript and python that will
> > obtain a unique identifier for a computer (e.g. IP address).  It must
> > not be browser specific.
>
> The other guys seems to have answered your first question so I will
> try to answer this.
> You should first of all be aware that there is no common unique
> identifier for a computer available from the net today[1]. IP
> addresses as you suggest aren't useable because of of multiple users
> being placed behind the same firewall/proxy server whos IP all their
> requests originate from. If your application is for an intranet only
> without firewalls or proxy servers then this might be different but
> generally it fits.
> A far better solution is to generate something unique for the user to
> carry around. This is easily done using something like the choice
> function of the python random interface. I know nothing about
> Javascript so how you would come around this thing using it I don't
> know but the fact that using IPs is bad is universal.
>
> [1]Why would somebody use cookies if there was?
> --
> Martin Skott
> Software Developer
> Propylon - Enabling Universal Mobility
> http://www.propylon.com
> Tel: +353 096 37245
> Mobile: +353 087 9680370
> --
> http://mail.python.org/mailman/listinfo/python-list
>





From Henri.Valeins at rmsb.u-bordeaux2.fr  Wed Mar 21 11:00:49 2001
From: Henri.Valeins at rmsb.u-bordeaux2.fr (Henri Valeins)
Date: Wed, 21 Mar 2001 17:00:49 +0100
Subject: help for newbie
Message-ID: <3AB8D031.2E54827E@rmsb.u-bordeaux2.fr>

Hi,

I try to use python program named PLUCKER for retrieve web page on
internet. When I try this program I have this message :

Retrieved failed: 404 -- [Errno url error] unknown url type: 'http'

My proxy var is correct 
http_proxy=192.168.10.254:8080

What is wrong ??

Regards
-- 
Henri Valeins
R?sonance Magn?tique des Syst?mes Biologiques
UMR 5536 CNRS/Universit? Victor Segalen Bordeaux 2 
146 rue L?o-Saignat (case 93) 
33076 BORDEAUX cedex
tel 33 (0)5 57 57 17 85
fax 33 (0)5 57 57 45 56


From vonWedel at lfpt.rwth-aachen.de  Wed Mar 28 07:19:49 2001
From: vonWedel at lfpt.rwth-aachen.de (Lars von Wedel)
Date: Wed, 28 Mar 2001 14:19:49 +0200
Subject: Uploading a file via POST ?
References: <99qdlf$a9r$1@news.rrz.Uni-Koeln.DE>
Message-ID: <3AC1D6E5.AFCC8AD7@lfpt.rwth-aachen.de>

Hi,

You might want to try cgiupload:

http://starship.python.net/crew/jbauer/cgiupload/index.html

I'm using it for a similar purpose, see if it fits your needs...

Lars

Oliver Hofmann wrote:
> 
> 'lo everyone!
> 
> I am trying to write a small client in Python which submits a file
> to a webpage via the POST method. Part of the .html reads:
> 
> ----
> <FORM METHOD=POST  ENCTYPE=multipart/form-data
>    ACTION="/cgi-bin/tst.pl_batch"
>           onsubmit="return FSubmitValidation(this)">
> 
> <INPUT TYPE=text SIZE=40 NAME=Email_Address>
> <INPUT TYPE=file NAME=UpLoad_File SIZE=50>
> ----
> 
> I've taken a look at urllib.urlopen, however it seems to be limited
> more or less to GET requests. An optional data-argument must be
> in 'standard application/x-www-form-urlencoded format'; since the
> form requires multipart form-data this is probably not going to
> work.
> 
> Is there an easy way to upload a file and pass parameters to the
> server script?
> 
> Thanks a lot,
> 
>         Oliver
> 
> --
> http://mail.python.org/mailman/listinfo/python-list
-------------- next part --------------
A non-text attachment was scrubbed...
Name: vonWedel.vcf
Type: text/x-vcard
Size: 342 bytes
Desc: Card for Lars von Wedel
URL: <http://mail.python.org/pipermail/python-list/attachments/20010328/6edca8ba/attachment.vcf>

From jschmitt at vmlabs.com  Thu Mar  8 16:42:09 2001
From: jschmitt at vmlabs.com (John Schmitt)
Date: Thu, 8 Mar 2001 13:42:09 -0800 
Subject: Python Helps Disney Write a New Script 
Message-ID: <008F0A63472BD311AF9800104BCD102561CCE6@minirex.vmlabs.com>

Use of a Different Scripting Language Leads to a Groundbreaking Application

http://python.oreilly.com/news/disney_0201.html




From jkraska1 at san.rr.com  Fri Mar 30 01:53:19 2001
From: jkraska1 at san.rr.com (Courageous)
Date: Fri, 30 Mar 2001 06:53:19 GMT
Subject: list vs tuple
References: <mailman.985789357.19622.python-list@python.org> <3ac40acd$0$10576@wodc7nh0.news.uu.net>
Message-ID: <mkb8ct4fjc2gve5o0arrhnrlmivajsdg5c@4ax.com>

>> Off the top of my head - A tuple is immutable, and therefore hashable, and
>> therefore can be used as a key to a dictionary. I'm sure that there are
>> other things...

>what do you mean by hashable?

Python 2.0 (#8, Oct 16 2000, 17:27:58) [MSC 32 bit (Intel)] on win32
Type "copyright", "credits" or "license" for more information.
IDLE 0.6 -- press F1 for help
>>> h={}
>>> mylist=[1,2,3]
>>> mytuple=(1,2,3)
>>> h[mylist]=3
Traceback (innermost last):
  File "<pyshell#3>", line 1, in ?
    h[mylist]=3
TypeError: unhashable type
>>> h[mytuple]=3
>>> h[mytuple]
3
>>> 



From donn at u.washington.edu  Fri Mar  9 14:59:48 2001
From: donn at u.washington.edu (Donn Cave)
Date: 9 Mar 2001 19:59:48 GMT
Subject: I come to praise .join, not to bury it...
References: <mailman.984074812.27842.python-list@python.org> <98aspo024jg@news2.newsguy.com> <98b75u$15les$1@ID-77331.news.dfncis.de>
Message-ID: <98bcnk$kdi$1@nntp6.u.washington.edu>

Quoth "Mark Pilgrim" <f8dy at yahoo.com>:
| "Alex Martelli" <aleaxit at yahoo.com> wrote in message
| news:98aspo024jg at news2.newsguy.com...
|> But the .join/.writelines parallel is VERY close, so why
|> would the SAME architectural choice (to have the sequence
|> object as an argument, accessing it through its ability
|> to enumerate its items, only) cause problems in one case
|> and not the other?!
|>
|> I think it's a misperception due to thinking of strings
|> as somehow 'passive', 'data', while a file-object is
|> thought of as 'active', 'a truly object-y object', or
|> something.
|
| Exactly!  I have taught Python to half a dozen people in the last few
| months, and every one of them had trouble with this concept, because they
| all came from a C background where they only thought of "string" as "packed
| array of char", or some other language (like Powerbuilder) where "string" is
| just a native datatype and isn't an object in any sense of the word.  Nobody
| minded file objects, even if they were only used to using integer file
| handles in their previous language.  I think that's because they could build
| a mental model of a file object as a kind of secretary who has a copy of a
| document on their desk and can read you various parts of it, or all of it,
| or tell you its title, or whatever.  But everybody had trouble with string
| objects, because the mental model is different.

At first it sounds like you're saying that's because of non-OOP
languages, but then those integer file handles come off like a
contradiction.  It isn't necessarily - the integer file descriptor
is actually a much more effective I/O abstraction in some ways,
because it comes from "the computer" and its devices.  On UNIX,
anyway, and for me that's one of the truly great things about UNIX.
In some common system programming situations, on UNIX, it's easier
even in Python to get things to work if you slough off the file
objects and go directly to the integer file descriptor.

My point, anyway, is that everywhere you look, you will find files
and I/O surrounded by software abstractions, if not OOP.  This is
somewhat different from the situation with strings, so not to be
surprised if some religious instruction is required.

	Donn Cave, donn at u.washington.edu


From aleaxit at yahoo.com  Mon Mar 19 11:03:26 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Mon, 19 Mar 2001 17:03:26 +0100
Subject: newbie - infinite loop
References: <985012513.18971.0.nnrp-12.c1c3e154@news.demon.co.uk>
Message-ID: <995akb02nsf@news1.newsguy.com>

"Paul Brian" <pbrian at demon.net> wrote in message
news:985012513.18971.0.nnrp-12.c1c3e154 at news.demon.co.uk...

You have diagnosed problems modifying the very list you are looping on.
Right: such modifications would indeed cause problems.  But, anyway,
your code is not an implementation of the algorithm you intend:

> I devcided to create a (neaarly) empty list, compare each name in the file
> with that list. If the name was not in, append to list, otherwise leave it
> alone.

That would be...:

    uniqueList = []

    line = 'q'

    for item in uniqueList:
        if line == item: break
    else:
        uniqueList.append(line)

Note we append in the *else* clause of the loop, that is,
if and only if the loop has NOT been interrupted by a
'break'.  Appending what IS already in the list (i.e.,
having the .append call instead of the break statement
as the guarded code of the if statement) is the other
way around from what you thought of (and your thoughts
were indeed correct).


If we did need to modify a list as well as iterate on
it, we'd do our iteration on a _copy_ of the list.  For
example, say that your target was to get a list made
up of the first 3 lines of a fine *and any further line
identical to one of the first 3, in order*.  Then, one
approach (perhaps not optimal, but simple indeed) -- if
we didn't know about the 'in' operator, etc -- might be:

def peculiarLines(afile):
    result = []
    for i in range(3):
        result.append(afile.readline())
    for line in afile.readlines():
        for existing in result[:]:
            if line==existing:
                result.append(line)
                break
    return result

OK, _way_ suboptimal, but still, should work.  The
key to making it work is the specific statement:
        for existing in result[:]:
whereby we iterate on a *copy* (a full slice) of
the result-list -- so, our result.append (as it
modifies the real result-list, not the copy) does
not break anything.

Here, of course, we can do much better, but the
idea is worth considering since at _other_ times
we may indeed need to loop on a list AND alter
it in the body of said loop as well.

For this specific case, the 'in' operator is not
really very fast either -- it still has to scan
linearly the whole result list, after all.  A
small auxiliary dictionary would make the whole
procedure O(N) rather than O(N squared):

def peculiarLines(afile):
    result = []
    seen = {}
    for i in range(3):
        line = afile.readline()
        result.append(line)
        seen[line] = 1
    for line in afile.readlines():
        if seen.has_key(line):
            result.append(line)
    return result

The auxiliary-dictionary approach is equally
helpful for your original problem:

def uniqueLines(afile):
    result = []
    seen = {}
    for line in afile.readlines():
        if not seen.has_key(line):
            result.append(line)
            seen[line] = 1
    return result

or, if you really wanted lines *which only
appear once* (in the same order in which
they appear in the file):

def reallyUniqueLines(afile):
    result = []
    seen = {}
    for line in afile.readlines():
        if seen.has_key(line):
            try: result.remove(line)
            except ValueError: pass
        else:
            result.append(line)
            seen[line] = 1
    return result

This is, again, sub-optimal, since repeated
lines cause calls to possibly-slow method
remove -- ideal might be to keep a list of
all lines seen _at least once_ and use the
auxiliary dictionary to count how many times
each line was seen, then filter appropriately
at the end.  For example, generalizing:

def seenExactlyNtimes(afile, N=2):
    result = []
    seen = {}
    for line in afile.readlines():
        count = seen.get(line)
        if count:
            seen[line] = count+1
        else:
            result.append(line)
            seen[line] = 1
    return [line for line in result if seen[line]==N]

This returns the lines seen exactly N times
_in the order in which they were first seen_.

An interesting variation is to return them
in the order of their M-th appearance, where
M is another parameter (1<=M<=N, of course)...
but I'll leave that as an exercise for the
reader!


Alex





From ben.held at staarinc.com  Tue Mar 13 10:23:38 2001
From: ben.held at staarinc.com (Ben Held)
Date: Tue, 13 Mar 2001 09:23:38 -0600
Subject: NumPY: Reading/Writing Arrays To/From Files
Message-ID: <3aae3bd6$0$42882$272ea4a1@news.execpc.com>

I would like to be able to read/write NumPY array objects to/from ascii
files.  I could not find support for this in the NumPY code.  Any ideas?

Ben




From support at internetdiscovery.com  Wed Mar 28 05:08:28 2001
From: support at internetdiscovery.com (Mike Clarkson)
Date: Wed, 28 Mar 2001 10:08:28 GMT
Subject: Jython and JPython Questions...How similar to Python, Tkinter?
References: <3ABFF14A.30D0096D@earthlink.net> <3AC1BDDB.11D8E367@accessforall.nl>
Message-ID: <3ac1b549.3763696@24.0.228.33>

On Wed, 28 Mar 2001 11:33:09 +0100, Ype Kingma
<ykingma at accessforall.nl> wrote:

>Ron,
>
>you wrote wrote:
>> 
>> Does Jython support Tkinter? OK, maybe that question as worded is
>> nonsense, let me re-phrase it: If I code a program in Python and put a
>> GUI front end on it with Tkinter, how hard would it be to do the same
>> program in Jython? Or is there a better solution for Jython GUI's?
>>
>
>From: http://www.jython.org/docs/differences.html 
>Under extension modules:
>
> However, Finn Bock has a JNI implementation called jTkinter which supports the
> full _tkinter API. Very cool stuff! 
>
>and this is the link: http://jTkinter.sourceforge.net/

Are you sure this works with jython? From the web page:

  The jTkinter module is a complete java/JNI implementation of the   
  builtin python modules _tkinter for use with the JPython-1.1
 environment. 

I think it needs bringing  up to date for jython, which would be
really nice;  I'd be happy to try to add Tix to it if it worked with
jython 2.x. Finn, can you confirm this?

I know it is not a "pure" Java solution, but there
are a lot of people like myself who need to use
some of the massive Java libraries, and are looking
to Jython to help provide a practical solution.

Moreover, it would help to extend the completeness and
interoperability of the CPython/Jython pair to include GUI work, 
which is an added plus.

Mike



From nas at arctrix.com  Wed Mar 21 20:06:36 2001
From: nas at arctrix.com (Neil Schemenauer)
Date: Wed, 21 Mar 2001 17:06:36 -0800
Subject: New PEP: The directive statement
In-Reply-To: <CN3u6.895$sk3.290814@newsb.telia.net>; from fredrik@pythonware.com on Wed, Mar 21, 2001 at 03:29:38PM +0000
References: <mailman.985117312.5806.python-list@python.org><j4u24oi4e8.fsf@informatik.hu-berlin.de><yHPt6.2553$4N4.334967@newsc.telia.net><998hv2$5ub$1@panix3.panix.com><oOQt6.2576$4N4.342378@newsc.telia.net> <uhf0njqq5.fsf@cs.uu.nl> <CN3u6.895$sk3.290814@newsb.telia.net>
Message-ID: <20010321170636.B25383@glacier.fnational.com>

[f-bot on __future__]
> All last-minute attempts to replace that with something that
> appears to be more future proof (but really isn't) falls under
> the good old X windows adage:
> 
>     "the only thing worse than generalizing from one example is
>     generalizing from no example at all"

I should write that one the wall here at work.

  Neil



From SumnerH at fool.com  Fri Mar  9 10:25:27 2001
From: SumnerH at fool.com (G. Sumner Hayes)
Date: 09 Mar 2001 10:25:27 -0500
Subject: Setting an attribute on a Python object from C
Message-ID: <200103091526.f29FQbc10149@ushqsvns-001.foolhq.com>

Hi,

I'm working on a Python extension in C and I've reached a point I can't
get past.  How do I create an attribute on an object from C?
PyObject_SetAttrString (or _SetAttr) doesn't seem to work for this.
Here's what I'm doing (implementing a Python DB 2.0 driver for Microsoft
SQL server):

self=PyObject_NEW(mssql_cursor, &mssql_cursor_type);
v=PyInt_FromLong(i);
rv=PyObject_SetAttrString((PyObject*) self, "rowcount", Py_None);

But rv comes back as -1.  Do I need to declare room in the mssql_cursor
struct and somehow declare the rowcount attribute to point at that?

Sumner




From loewis at informatik.hu-berlin.de  Tue Mar 20 15:04:11 2001
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 20 Mar 2001 21:04:11 +0100
Subject: New PEP: The directive statement
References: <mailman.985104986.26425.python-list@python.org>
Message-ID: <j4vgp4gqbo.fsf@informatik.hu-berlin.de>

Michael Chermside <mcherm_python at yahoo.com> writes:

> After all the heated discussion going on here, I'm glad SOMEONE
> finally responded to Tim's request to write it up as a PEP.

It might be that Tim, when asking for somebody to write a PEP,
actually meant *me* to write it. It is only that this is the first day
since I implemented the patch that I found any time for it...

> But thanks for writing it up...  we can't have "from __future__
> import" being the only target, now can we? <wink>

At the moment (i.e. for 2.1), I would not use it for anything but
"directive transitional nested_scopes". After that, people proposing
new uses probably have to write a PEP as well :-)

> Seriously, it looks well-written, TRIES (I think) to incorporate
> the vast majority of what people have said on the subject so 
> far, and as Tim pointed out, it was high time it got written.

Thanks. I hope it finds consideration before the next beta. I have not
listed that in the draft: a patch implementing the proposed feature is
in

http://sourceforge.net/tracker/index.php?func=detail&aid=404997&group_id=5470&atid=305470

Regards,
Martin


From carribeiro at yahoo.com  Wed Mar 21 07:17:47 2001
From: carribeiro at yahoo.com (Carlos Ribeiro)
Date: Wed, 21 Mar 2001 09:17:47 -0300
Subject: Yet Another PEP:  Query Protocol Interface or __query__
Message-ID: <5.0.2.1.0.20010321091728.021b5be0@pop.mail.yahoo.com>

Disclaimer: It's 1 AM and I'm running on brainstorm mode. Please ignore any 
dumb idea and trash accordingly <wink>

At 18:43 20/03/01 -0500, you wrote:
>     A new built-in method, __query__ is proposed.  This method has
>     a single argument, an interface protocol identifier, and either
>     returns an object supporting the given protocol, or throws an
>     unknown exception.

Do you want __query__ to act as a class factory? In your example you return 
"self" in one of the cases, and a new object instance in the other. Anyway 
I dont see the need to do it this way in Python. There is some confusion 
going into returning an object - which is an instance of a class, right? - 
and returning an interface. It is not the same in COM, but maybe it could 
be the same in Python. (It's 1 AM, so maybe I get it better tomorrow).

Of course, there's always the question of supporting multiple interfaces 
with name clashes and so on. I dont see any way to do it without resorting 
to some magic. For instance, __query__ could return an object supporting 
only the methods specified on the given interface, but pointing otherwise 
to the same attributes. The methods would then be mapped to the ones 
implemented in the object instance.

Very Simple Example (please ignore any dumb syntax mistakes :-):

class MyClass:
   def __myrepr__(self):
     pass
   def __query__(self, protocol):
     # supports enumeration methods with the default names
     if protocol == "org.python.enumeration":
       return self
     # it has a __repr__ compatible method with other name
     elif protocol == "org.python.printable":
       # we will hack the object representation
       printable_interface = self
       printable_interface.__repr__ = self.__myrepr__
       return printable_interface

Ok, this is a very simple example, and I know it's wrong. It has side 
effects - mainly, 'self' and 'printable_interface' are pointing to the same 
object instance :-) It would work for some particular cases (such as this 
one). However, the generalization would be much more complex than this. We 
would need to have an "almost exact" copy of the original object, only that 
the methods table would be different. I really dont know if this would be 
possible.

>     An interface protocol identifier is a lower-case string having
>     a reverse DNS host name string:   TLD '.' DOMAIN ['.' NAME ]*
>     Where TLD is a top level domain such as 'com' or 'uk.co'.  DOMAIN
>     is a registered name in the given TLD.  And one or more optional
>     NAME separated by a period.  NAME is a sequence of one or more
>     alphabetic characters including the dash '-' character.  See the
>     relevant ITEF specifications for specific details.

Why not use a XML DTD-like syntax? That would allow the use of XML to 
express the interface definition in some way. The interface definition 
could even include some information about expected parameter types, without 
having to change Python type(less) model.



Carlos Ribeiro


_________________________________________________________
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com




From dsh8290 at rit.edu  Tue Mar 13 12:42:33 2001
From: dsh8290 at rit.edu (D-Man)
Date: Tue, 13 Mar 2001 12:42:33 -0500
Subject: Private class and instance methods?
In-Reply-To: <B6D000FE.4D14%f8dy@diveintopython.org>; from f8dy@diveintopython.org on Sat, Mar 10, 2001 at 04:04:14PM -0500
References: <3AAA73BF.9CDCF3B2@alcyone.com> <B6D000FE.4D14%f8dy@diveintopython.org>
Message-ID: <20010313124233.D10298@harmony.cs.rit.edu>

On Sat, Mar 10, 2001 at 04:04:14PM -0500, Mark Pilgrim wrote:
| in article 3AAA73BF.9CDCF3B2 at alcyone.com, Erik Max Francis at
| max at alcyone.com wrote on 3/10/01 1:34 PM:
| > How often do people use methods beginning with __ in order to get some
| > amount of privacy with the automatic name mangling that occurs, to get
| > some kind of weak protection, with, say, accessor methods:
| > 
| > [which would of course allow C.getX and C.setX to do something different
| > in the future]?
| 
| There is a better way to do this in Python, using the __getattr__ and
| __setattr__ special methods.
| 

There is another issue to consider with this.  The name mangling
happens at compile time and can mess things up if you use inheritance.
Say, for example, you create a class with the __getattr__ and
__setattr__ methods to do what you want in a general fashion.  Then
you make a subclass of it to avoid duplicating the __setattr__ and
__getattr__ methods.  It won't work anymore due to the name mangling.  


Check the archives for more details on this.  There was a thread a
while back where Alex Martilli explained how that base class could be
made (for __getattr__ only, he left __setattr__ as an exercise for the
reader), but he used a single underscore rather than the doube
underscore (that causes name mangling).

Actually, on second thought, if you only access the variable through
the methods, it should work, but probably not as expected.  

>>> class Example :
...     def __setattr__( self , name , value ) :
...             self.__dict__[ name ] = value
...     def __getattr__( self , name ) :
...             return self.__dict__[name]
...
>>> o = Example()
>>> print o.__dict__
{'__x': 2}
>>>


-D




From sdm7g at Virginia.EDU  Thu Mar 29 14:04:45 2001
From: sdm7g at Virginia.EDU (Steven D. Majewski)
Date: Thu, 29 Mar 2001 14:04:45 -0500 (EST)
Subject: forcing dictionary order
In-Reply-To: <3AC380AE.B225C07F@uniserve.com>
Message-ID: <Pine.NXT.4.21.0103291357450.234-100000@localhost.virginia.edu>


On Thu, 29 Mar 2001, Bob van der Poel wrote:

> 
> I'd like (not a must!) to walk a dictionary in the same order in which
> it was created. Example:
> 
> 	foo={ key1: (), key2: (), key3: () }
> 
> If I do:
> 	
> 	for a in foo.keys(): ...
> 
> the order is not "key1, key2, key3", which is what I want. Note, I can't
> do something simple like sort() on the keys since the are not in any
> alphanumeric order. I'm just using this to generate a largish table for
> another program, and the actual order doesn't matter for it to work
> properly, but it would nicer if it were in the correct logical order.
> So, I'm not going to spend a lot of time on this, but if there is a
> simple way????

Once they are in the dictionary, they keys are hashed, so any info
on the order of insertion is lost. 

You can create a Class based on UserDict, and override the __setitem__
method (and to be consistent, __delitem__ and update) to keep an ordered
list of keys in parallel. 

You might also redefine popitem to return pairs in the proper order. 
( Do you want FIFO or LIFO ? ) 

-- Steve Majewski





From donn at u.washington.edu  Mon Mar 26 20:11:22 2001
From: donn at u.washington.edu (Donn Cave)
Date: 27 Mar 2001 01:11:22 GMT
Subject: fixing map(None, [1,2,3]) ?
References: <mailman.985648235.13881.python-list@python.org>
Message-ID: <99opbq$u7k$1@nntp6.u.washington.edu>

Quoth "Sean 'Shaleh' Perry" <shaleh at valinux.com>:
|
| On 26-Mar-2001 Mike Coleman wrote:
| > Shouldn't "map(None, [1,2,3])" return "[(1,), (2,), (3,)]"?
| > 
| > Is there any hope that this could be fixed?
| > 
|
| map runs the function on each element of the list it is passed and returns an
| element of a new list.  Why should it magically return a list of one-tuples?

It would be more consistent.  Right now this is a special case -
any other number N of input lists, where N != 1, returns a list of
N-tuples.

But it's also a common case, and "fixing" it would surely break lots
of code.  Can't even see how the fixed version would be more useful.

	Donn Cave, donn at u.washington.edu


From aleaxit at yahoo.com  Tue Mar 13 04:16:20 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Tue, 13 Mar 2001 10:16:20 +0100
Subject: Extending Python: rewriting a single method in C
References: <g0wv9vxms0.fsf@scumbag.ecs.soton.ac.uk> <slrn9apn8f.382.jepler@potty.housenet> <2tfr6.26915$zV3.1982567@news1.frmt1.sfba.home.com>
Message-ID: <98kohu0ol0@news2.newsguy.com>

"Alexander Gavrilov" <agavrilov at home.com> wrote in message
news:2tfr6.26915$zV3.1982567 at news1.frmt1.sfba.home.com...
    [snip]
> > However, you cannot make a builtin function be a method of a
> class/instance.
> > Instead, you could write:
> >
> > from cmodule import fast_a, fast_b
> >
> > class fastKlass(Klass):
> > def a(self, int1, int2):
> > fast_a(self, int1, int2)
> > def b(self, int1, int2):
> > fast_b(self, int1, int2)
> >
>
> Actually, you can write the following:
>
> class fastKlass(Klass):
>     a = fast_a
>     b = fast_b

You can, but it will not work as intended if fast_a or fast_b
come from an extension module (coded in C or C++).  So-called
"builtin" functions (those coded in C) are not automagically
transformed into unbound methods when assigned as class attributes.

You _can_ request the transformation explicitly if you know
what you're doing...:

>>> class Y:
...     pass
...
>>> Y.a=new.instancemethod(cmodule.fast_a,None,Y)

Note that it can't be done within the body of Y since it needs
Y itself (the class-object) as an argument, and the class object
is not defined until AFTER the class-body is done executing.


Alex





From daniel.dittmar at sap.com  Fri Mar 30 03:28:40 2001
From: daniel.dittmar at sap.com (Daniel Dittmar)
Date: Fri, 30 Mar 2001 10:28:40 +0200
Subject: Printing mixtures in integers and strings
References: <3ac38cf6$0$12246$ed9e5944@reading.news.pipex.net>
Message-ID: <9a1g3o$ma2$1@news1.wdf.sap-ag.de>

>             I'm trying to use something like this:-
>
> def move(row): #Moves the dispenser to the row on the plate
>     print "Moved to Row - " + row

Convert the int to a string:
print "Moved to Row - " + str (row)
or
print "Moved to Row - " + `row`

Daniel

--
Daniel Dittmar
daniel.dittmar at sap.com
SAP DB, SAP Labs Berlin
http://www.sapdb.org/





From djc at object-craft.com.au  Mon Mar 12 06:25:35 2001
From: djc at object-craft.com.au (Dave Cole)
Date: 12 Mar 2001 22:25:35 +1100
Subject: Problems using python.sty to document a module
Message-ID: <m31ys3qle8.fsf@vole.object-craft.com.au>

I have been trying to document my new Sybase module using the
documentation tools which come with Python but have hit a problem.

When LaTeX hits the following line it falls over.

\declaremodule{extension}{_sybase}

The messages produced look like:
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Chapter 2.
! Missing \endcsname inserted.
<to be read again> 
                   \protect 
l.8 \declaremodule{extension}{_sybase}
                                      
? 
! Emergency stop.
<to be read again> 
                   \protect 
l.8 \declaremodule{extension}{_sybase}
                                      
!  ==> Fatal error occurred, the output PDF file not finished!
Transcript written on sybase.log.
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

If I remove the _ on the module name then it seems to work.  I have
tried escaping the _ with a \, but it still carks it.

Does anyone know how I can document a module which is named with a
leading underscore?  I would prefer not to have to rename the module.

- Dave

-- 
http://www.object-craft.com.au


From frank at quantiva.com  Thu Mar  1 17:12:45 2001
From: frank at quantiva.com (Frank Stolze)
Date: Thu, 1 Mar 2001 17:12:45 -0500 (EST)
Subject: [OT - JOB AD][REPOST] Python / XML / Distributed Systems Developer
Message-ID: <Pine.LNX.4.30.0103011711120.9554-100000@localhost.localdomain>

This time without the HTML nonsense...


Sorry for the off-topic post. We are a well-funded, stealth mode startup in the
network service management field. We are building a novel, distributed system
and service that involves Internet protocol implementations (HTTP, SMTP, POP3,
DNS, ping, etc.), statistical analysis, some AI, database storage and
reporting, as well as issues such as routing, load balancing, fail-over,
firewalls, etc.

Almost all of the implementation is being done in Python. We also use XML,
XML-RPC, HTTP tunneling and a few new things. We are looking for two
enthusiastic Python & networking gurus to join a small team of hands-on
people to help us implement a great vision!

The "official" job description is below.


Regards,
Frank

===============
Company Profile:

We're onto something really big in network service management. We have
VC backing and a clear view of the future. Our stealth situation hides our
long history in this area and lets us focus on the goal. Join the team of
innovative, visionary, enthusiastic, and passionate people that will shape
our future. Share the opportunity to work with industry leading experts on
the design and creation of a next generation system.

Quantiva is looking for people who can quickly grasp new concepts,
develop new, original solutions to existing problems, and in general can
"hit the ground running."

Quantiva is located in Princeton, New Jersey.

Please send resumes to techjobs at quantiva.com.


Job Description:

The Network Software Engineer will contribute to the design and development
Quantiva's network service management software. This position will involve
active participation in the design, architecture, and implementation of the
product.


Requirements:
? Strong software development skills in a Unix environment.
? Strong network and distributed systems programming experience. Good knowledge
  of network protocols such as TCP/IP, HTTP, DNS, SMTP, POP3 and concepts such
  as firewalls, routing required.
? 4+ years of hands-on experience with scripting languages such as Perl, Python
  or Tcl. Python proficiency is required.
? Applications programming experience in object-oriented languages such as C++
  or Java.
? DBMS experience including database schemas, SQL, and database programming.
? Solid Unix experience (Solaris preferred) in a commercial environment.
? XML experience is a plus.
? BS/MS/PhD in CS, EE, CE or related field.





From nspurrier  Fri Mar 16 23:10:04 2001
From: nspurrier (at)
Date: 17 Mar 2001 04:10:04 GMT
Subject: how to indent blocks in vi using << and >>
Message-ID: <98uo2s616np@news2.newsguy.com>


I was dumbfounded when I learned this... It's so obvious.
Am I the only one? I'm sure this will seem stupid to vi veterans,
but I've used vi for years, yet I never knew about these commands:
	>>
	and
	<<
It turns out these are critical for editing Python with vi.
The << and >> commands are used to for block indenting.
The >> indents a line right one tab and << indents a line left 
one tab. You can indent a whole bunch of lines by giving a number
the number before the >> or <<. For example, 10>> will indent
ten lines from the current cursor position.

Carry on,
Noah Spurrier

==================================
Posted via http://nodevice.com
Linux Programmer's Site


From danielk at aracnet.com  Sat Mar 10 10:56:50 2001
From: danielk at aracnet.com (Daniel Klein)
Date: Sat, 10 Mar 2001 07:56:50 -0800
Subject: Python In A Nutshell
References: <3AB1700C@MailAndNews.com>
Message-ID: <fjjkatsd3kcp7dsf1n6po8qm50vrukp07c@4ax.com>

On Sat, 10 Mar 2001 02:10:57 -0500, Laura Lewin <Laura_01 at MailAndNews.com>
wrote:

>Hi,
>We all know Alex--hi Alex!--uses lots and lots of words to say things. 
><grin>
>But he certainly knows how to make a point....(wait and see!)  It will be a 
>nutshell not a treatise.  No schedule yet.

It will have to be a pretty bit 'nut' to hold Alex's stuff. :-)

Dan


From johngrayson at home.com  Tue Mar 20 12:58:11 2001
From: johngrayson at home.com (johngrayson at home.com)
Date: Tue, 20 Mar 2001 17:58:11 -0000
Subject: Fwd:  Re: Tkinter : resizing Text widget with main window
Message-ID: <9985nj+4n9g@eGroups.com>

--- In python-list at y..., Erik de Castro Lopo <nospam at m...> wrote:
Rick Pasotto wrote:
>

This does work for grid ().

This DOES work for grid ().

from Tkinter import *

class application:

    def __init__(self, master=None):
        self.master = master

        for r in range(3):
	    master.grid_columnconfigure(r, weight=1)
	    master.grid_rowconfigure(r, weight=1)
            for c in range(3):
                lbl = Text(master, width=30,height=12, bd=3,
                 relief=SUNKEN)
                lbl.grid(row = r, column = c, sticky = NSEW)
                lbl.insert(END, 1000 + r + c)

root = Tk()
root.title('Grid Weight')
g = application(root)
root.mainloop()


The default row and column weight is zero, so the widgets
do not expand to use extra space.

     John Grayson




From max at alcyone.com  Thu Mar  1 11:48:43 2001
From: max at alcyone.com (Erik Max Francis)
Date: Thu, 01 Mar 2001 08:48:43 -0800
Subject: A counter-proposal to __future__ in PEP 236
References: <j47l29lpo8.fsf@informatik.hu-berlin.de>
Message-ID: <3A9E7D6B.D3C61E7D@alcyone.com>

Martin von Loewis wrote:

> I'd appreciate any comments as to whether this directive statement is
> better or worse for the purpose of indicating that nested scopes are
> used in a module.

I think it's much more explicit and appropriate to use a directive
statement than overload the meaning from the from...import... statement. 

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ Who shall stand guard to the guards themselves?
\__/ Juvenal
    Official Omega page / http://www.alcyone.com/max/projects/omega/
 The official distribution page for the popular Roguelike, Omega.


From phrxy at csv.warwick.ac.uk  Sat Mar 31 14:55:57 2001
From: phrxy at csv.warwick.ac.uk (John J. Lee)
Date: Sat, 31 Mar 2001 20:55:57 +0100
Subject: Know of Substantial Apps Written in Python?
In-Reply-To: <mailman.986048021.30049.python-list@python.org>
References: <3ac4e4a8_1@news4.newsfeeds.com> <KE6x6.415$SR6.235983@typhoon1.ba-dsg.net> <3ac4ff12$1_2@news4.newsfeeds.com> <mailman.986048021.30049.python-list@python.org>
Message-ID: <Pine.SOL.4.30.0103312050260.15965-100000@mimosa.csv.warwick.ac.uk>

On Sat, 31 Mar 2001, Carlos Alberto Reis Ribeiro wrote:

> At 22:27 30/03/01 -0600, Grant Griffin wrote:
[...]
> Also in Python:
>
> - Infoseek/Ultraseek server (core library in C, bind to Python)
> - The RedHat installer
[...]

This reminds me: why aren't the Redhat (or any other linux / unix vendor)
boot scripts in Python, too?  Whenever I've made small changes to them,
I've wished they were!  I have seen it stated that 'obviously' boot
scripts can't be in Python, but nobody has explained to me why this is.
How much bigger is a minimal Python interpreter than bash?

I'm doubtful that there are any advantages to shell, even for short
scripts.


John




From fredrik at pythonware.com  Tue Mar 13 18:42:55 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Tue, 13 Mar 2001 23:42:55 GMT
Subject: Unpickling from strings
References: <qHxr6.19658$pR3.3830844@news2-win.server.ntlworld.com>
Message-ID: <3gyr6.159$4d3.34762@newsb.telia.net>

Marcin Chady wrote:
> I have pickled an object into a string and saved it in an hidden HTML form
> field, so that I can recover the state when the form is resubmitted.
> However, when the form comes back and I try to unpickle my object I get an
> error saying that the module in which the object's class was originally
> defined is unknown, even though the CGI script which does the unpickling has
> already imported the module!
>
> Has anyone got an idea why it is?

did you pickle classes defined in your main script (aka __main__),
and then failed to unpickle them in another script?

don't do that.

(to solve this, move the classes to a common support module, and
import it into the pickling script)

Cheers /F




From gordonj at paradise.net.nz  Sat Mar 31 02:05:43 2001
From: gordonj at paradise.net.nz (johnston112)
Date: Sat, 31 Mar 2001 19:05:43 +1200
Subject: very basic, how do you find square roots?
Message-ID: <9a408a$1mpp$1@raewyn.paradise.net.nz>

Yes, that is my question :)
I am trying to make a program that dose the quadratic equation.

johnston112




From mickey at Vanille.de  Sun Mar 18 11:40:49 2001
From: mickey at Vanille.de (Michael 'Mickey' Lauer)
Date: 18 Mar 2001 18:40:49 +0200
Subject: What has become of PEP224 ?
Message-ID: <3ab4f320@nntp.server.uni-frankfurt.de>

Hi. If I remember correctly PEP224 (the famous "attribute docstrings")
has only been postponed because Python 2.0 was in feature freeze
in August 2000. Will it be in 2.1 ? If not, what's the reason ? What
is needed for it to be included in 2.1 ?

Yours,

:M:

-- 
Michael 'Mickey' Lauer . . . . . . . . . mickey at Vanille.de
How could anyone know me - when I don't even know myself ?




From pj at sgi.com  Sun Mar 18 03:32:08 2001
From: pj at sgi.com (Paul Jackson)
Date: 18 Mar 2001 08:32:08 GMT
Subject: How do I avoid the Tkinter 30 msec polling loop.
References: <98toci$ktjh$1@fido.engr.sgi.com> <QdOs6.506$Fb.774749@typhoon2.ba-dsg.net> <3ab3d735.8139907@24.0.228.33>
Message-ID: <991rq8$14q93$1@fido.engr.sgi.com>

Thanks for the useful Tkinter URL, David.  It led
me to some online docs I had missed.

One other detail of the after() documentation that
might be lacking.  I am quite unsure of this,
because I only saw it a couple times a few days
ago, when I was playing around with after() while
in the land of the blissful newbie, but it seemed
to me that Tkinter dumped core if I tried to
delay more than 4 seconds.

What is the intended limit, if any, of the 'ms'
parameter?
-- 
                          I won't rest till it's the best ...
                          Manager, Linux System Software
                          Paul Jackson <pj at sgi.com> 1.650.933.1373


From dsavitsk at e-coli.net  Sun Mar  4 16:13:47 2001
From: dsavitsk at e-coli.net (dsavitsk)
Date: Sun, 4 Mar 2001 15:13:47 -0600
Subject: Python COM server to VB interface
Message-ID: <keyo6.156158$Ch.31627133@newsrump.sjc.telocity.net>

I am attempting to learn to use python as a com server to a VB interface.
When i run my little example i get the following error.

Run-time error'-2147467259 (800004005)':
Unexpected Python Error: AttributeError: CreateInstance

Here is the Python end of things...

    class pyserv_test:
        _public_methods_ = [ 'AddString' ]
        _reg_progid_ = 'PyServ.Test'
        _reg_clsid_ = '{EA08AFC9-E0AA-4C9B-8E42-1170A21D59AC}'

        def AddString(self, val):
            return 'the string was: ' + str(val)

    def Register():
        import win32com.server.register
        return win32com.server.register.UseCommandLine(pyserv_test)

    if __name__=='__main__':
        print "Registering COM server..."
        Register()


and here is the relevant VB code...

    Private Sub Command1_Click()
        Dim x
        Dim y
        Set x = CreateObject("PyServ.Test")
        y = x.AddString(Text2.Text)
        Text1.Text = y
    End Sub

The error occurs on the "CreateObject" line. Does anyone know what this
might be?

Thanks,
doug




From tim.hochberg at ieee.org  Thu Mar 29 10:45:12 2001
From: tim.hochberg at ieee.org (Tim Hochberg)
Date: Thu, 29 Mar 2001 15:45:12 GMT
Subject: Gaussian line profile using numeric python
References: <3ac31698@zfree.co.nz> <m3puf0re81.fsf@chinon.cnrs-orleans.fr>
Message-ID: <cMIw6.81087$Xt3.11074709@news1.rdc1.az.home.com>

Like Konrad, I'm too lazy to completely figure this out, but I do have some
comments on it below.

Konrad Hinsen is ">"
Geoff Low is ">>":
>
> > I have an extensive set of xy data that I want to convolute with a
gaussian
> > profile.  Below is my algorithm,  which doesn't work.  Can anyone see
where
> > I'm being dumb?
>
> To be honest, I am too lazy to look through this. However,

Ditto. However some thoughts / questions. Is your data evenly spaced in x?
If it is you might want to pass in deltaX instead of x to make that clear.
If it's not, you're going to need a considerably more sophisticated
algorithm. If it is, go with Numeric.convolve as Konrad suggests.

> 2) If you work with large data sets, you ought to use FFTs to
>    compute the convolution, that's O(N*log(N)) in the size of the
>    data set, instead of O(N**2) for the straightforward method.
>    Any good book on FFTs should explain how this works in detail.

One thing to watch for is that convolution using FFT makes an implicit
assumption that your data is periodic. If it's not, and you want to try this
route, look up zero padding  in that FFT book that I'm _sure_ you'll consult
before you try this <0.5 wink>.

-tim




From mlh at idi.ntnu.no  Wed Mar 14 12:46:34 2001
From: mlh at idi.ntnu.no (Magnus Lie Hetland)
Date: Wed, 14 Mar 2001 18:46:34 +0100
Subject: Instant Hacking and Instant Python
Message-ID: <98oanu$4da$1@tyfon.itea.ntnu.no>

Hi!

I tried to post this a while back, but I think I got
it messed up somehow...

Anyway: "Instant Hacking" and "Instant Python" have
moved from their previous addresses to resp.

  http://www.hetland.org/python/instant-hacking.php

and

  http://www.hetland.org/python/instant-python.php

It's not critical (since I have added forwarding
mechanisms to the old ones), but it would be nice if
any links were updated, especially in the translated
versions of the articles.

Thank you :-)

--

  Magnus Lie Hetland          http://www.hetland.org

 "Reality is what refuses to disappear when you stop
  believing in it"                 -- Philip K. Dick





From mcherm at destiny.com  Tue Mar  6 18:34:46 2001
From: mcherm at destiny.com (Michael Chermside)
Date: Tue, 06 Mar 2001 17:34:46 -0600
Subject: Missing the functional mind set
Message-ID: <3AA57414.B3CCC383@destiny.com>

WARNING:

  SPOILERS INCLUDED!

  (well, since the original poster asked for hints not answers,
   I figured it'd be polite to do this)

















Okay, as Erik Francis (and others, I think) point out, sort() is a
modify-in-place method.

So what we need is a functional version (ie, a version that returns
a new value and doesn't modify its argument) for use in functional
programming. Here's one way to write it:

>>> def mysort(list):
        newlist = list[:]
        newlist.sort()
        return newlist

And now I can solve the original problem in a "functional" fashion:

>>> d = {3: 'd', 2: 'b', 1: 'a', 0: 'z', 938302: 'w'}
>>> templist = [ d[key] for key in mysort( d.keys() ) ]
>>> templist
['z', 'a', 'b', 'd', 'w']

Now, I don't recommend this as being the most efficent way to code
the thing, but the Daniel was originally trying to teach himself
the functional technique. And you have to admit, that it's really
quite readable. This is one of the nice things about functional
techniques.

Good luck Daniel... it's ALWAYS useful to try learning new programming
techniques, they increase the number of tools in the toolkit.

-- Michael Chermside



From ujen at nino.ru  Mon Mar  5 02:17:04 2001
From: ujen at nino.ru (ujen)
Date: Mon, 5 Mar 2001 07:17:04 +0000 (UTC)
Subject: Couple questions about Tkinter
Message-ID: <01c0a544$36a33a40$3201a8c0@ws9201>

Hello!

How to generate event from script?
How to make trace variable?

regards
 Eugene Tsymbalov, Russia




From kalle at gnupung.net  Wed Mar  7 19:51:48 2001
From: kalle at gnupung.net (Kalle Svensson)
Date: Thu, 8 Mar 2001 01:51:48 +0100
Subject: Open Source
In-Reply-To: <9806i4$2e97$1@news3.infoave.net>; from the_triple_i@hotmail.com on Mon, Mar 05, 2001 at 08:04:48AM -0600
References: <9806i4$2e97$1@news3.infoave.net>
Message-ID: <20010308015148.B1140@apone.network.loc>

A non-text attachment was scrubbed...
Name: msg.pgp
Type: application/pgp
Size: 913 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/20010308/656793b3/attachment.bin>

From no-spam-popov at bigserver.com  Sun Mar 11 16:43:02 2001
From: no-spam-popov at bigserver.com (Boris Popov)
Date: Sun, 11 Mar 2001 13:43:02 -0800
Subject: Einstein's Riddle
References: <oNPq6.396644$ge4.138479566@news2.rdc2.tx.home.com> <UdSq6.2184$54.2166@www.newsranger.com>
Message-ID: <98grhv$1vtrq$1@ID-75892.news.dfncis.de>

"Gregory Jorgensen" <gregj at pobox.com> wrote in message
news:UdSq6.2184$54.2166 at www.newsranger.com...
> The 2% that can solve it must be psychic or deranged. The riddle as posed
is not
> solvable.
>
> We are told of five houses anf five people. The hints mention five colors
(red,
> green, white, yellow, blue), five drinks (tea, coffee, milk, beer, water),
five
> smokes (Pall Mall, Dunhill, Blends, Bluemaster, Prince), and five
nationalities
> (Brit, Swede, Dane, Norwegian, German). But only four types of pet are
> mentioned: dogs, birds, cats, horse. We have no evidence that any of the
five
> owns fish.
>
> More telling: how many people can't quickly determine that this puzzle is
> unsolvable?

Well.. That's the trick of the puzzle...  and I admit I did not notice it
right away :))) My mind tricked me into thinking that there's a fish even
though statements do not mention it.
I would just say that it's not *unsolvable*... It turns out to be solveable
completely except the type of the pet german owns. It's a hole in the
solution table and we can't do anything about it...  Let's just say that
puzzle is solvable but we cannot answer the question without making any
assumptions.

-Boris

>
>
> In article <oNPq6.396644$ge4.138479566 at news2.rdc2.tx.home.com>, Steve
Graham
> says...
> >
> >Who would be interested in using his/her brain (and his computer) to
solve
> >the following?
> >
> >
> >Steve Graham
> >
> >===
> >
> >Einstein's Riddle
> >
> >Albert Einstein wrote this riddle this century [ed. 20th century].  He
said
> >98% of the world could not solve it.
> >
> >There are 5 houses in 5 different colors.  In each house lives a person
with
> >a different nationality.  The 5 owners drink a certain type of beverage,
> >smoke a certain brand of cigar, and keep a certain pet.  No owners
> >have the same pet, smoke the same brand of cigar or drink the same
beverage.
> >
> >The question is: "Who owns the fish?"
> >
> >Hints:
> >
> >The Brit lives in the red house.
> >
> >The Sweed keeps dogs as pets.
> >
> >The Dane drinks tea.
> >
> >The green house is on the left of the white house.
> >
> >The green house's owner drinks coffee.
> >
> >The person who smokes Pall Mall rears birds.
> >
> >The owner of the yellow house smokes Dunhill.
> >
> >The man living in the center house drinks milk.
> >
> >The Norwegian lives in the first house.
> >
> >The man who smokes Blends lives next to the one who keeps cats.
> >
> >The man who keeps the horse lives next to the man who smokes Dunhill.
> >
> >The owner who smokes Bluemasters drinks beer.
> >
> >The German smokes Prince.
> >
> >The Norwegian lives next to the blue house.
> >
> >The man who smokes Blends has a neighbor who drinks water.
>
> Greg Jorgensen
> Deschooling Society
> Portland, Oregon, USA
> gregj at pobox.com




From nospam at ktgroup.co.uk  Thu Mar  8 05:18:10 2001
From: nospam at ktgroup.co.uk (Pat Knight)
Date: 08 Mar 2001 10:18:10 +0000
Subject: grabbing return codes from os.system() call
References: <%wFp6.9481$dL4.126404@vixen.cso.uiuc.edu>
Message-ID: <u66hky36l.fsf@ktgroup.co.uk>

Damian Menscher <menscher+python at uiuc.edu> writes:

> How do I get the return code from an os.system call?  I would have
> expected I could do something like
> 
> ---returncode---
> #/bin/csh
> echo do stuff
> exit 3
> 
> and then in my python program I could do
> 
> print os.system('./returncode')
> 
> But it prints out 768.  Not particularly useful, even after I recognize
> the trick of dividing by 256 (byte-swapping going on?  No, because a

There's no byte swapping. UNIX shifts exit codes up 8 bits to include some
other flags about how a child process exited - normally, seg violation etc.

Look at the process management section of the os module documentation in
Python. This contains functions to decode exit codes.

Also, check the UNIX man pages for system, and the exec, exit and waitpid
system calls. You'll find details of the encoding, and probably find that you
can't have the large exit status values you want.

If you're wanting to return strings or large numbers, start investigating
os.popen.

Cheers,
        Pat


From loewis at informatik.hu-berlin.de  Fri Mar  2 09:39:53 2001
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 02 Mar 2001 15:39:53 +0100
Subject: A counter-proposal to __future__ in PEP 236
References: <mailman.983479564.11788.python-list@python.org>
Message-ID: <j4snkwjkue.fsf@informatik.hu-berlin.de>

"Tim Peters" <tim.one at home.com> writes:

> I don't.  "__future__" screams "Deep Magic" because of the double
> double-underscores; it *is* a perfectly vanilla import statement at runtime;
> the word "future" helpfully suggests temporality; it works with all existing
> tools today without change (so it's not confusing even the stupidest of them
> <0.1 wink>)

No, it doesn't. Python 2.1 does not introduce __future__ alone, it introduces

from __future__ import nested_scopes

I can enumerate a number of tools which will be confused (i.e. they'll
operate in a way which is incorrect given what they were designed to
do) - tools looking at or producing byte code being the most prominent
cases.

> > directive nested_scopes
> 
> Which loses all of that.

"all of that" now being the meta information of whether nested scopes
are on-by-default? Yes, that meta information is not available when
using directives.

> I really dislike that
> 
>     directive nested_scopes
> 
> *looks* exactly the same as
> 
>     directive favor_space

So it is the keyword that bothers you? I can propose a different syntax, e.g.

incompatible_language_change nested_scopes

(which would add another keyword incompatible_language_change)

or

directive incompatible_language_change nested_scopes

(which wouldn't - the latter syntax is already supported in my parser
patches). incompatible_language_change is meant as a placeholder;
not speaking English as a first language means that I obviously fail
to propose keywords that sound right to people.

> > In this implementation, directive is only considered as a keyword if
> > it appears at the beginning of the module.
> 
> In that case it's too feeble for a pragma facility; e.g., perhaps I want to
> optimize one method for space but another for time.

With that patch, it is not possible to specify optimization for
time. However, the patch can be certainly extended in that way, e.g. to say

directive optimize_for_time foo

Whether or not that directive still has to appear at the beginning of
the source file would be an issue of the PEP proposing such a pragma;
as would be the debate whether that should be specified through a
directive, or, say, through a function attribute.

> Or perhaps I want to specify that one class in a module uses
> traditional dicts for its namespace, but another freezes its
> attributes into integer-indexable vectors at compile-time.

I don't see your point; I have to admit I can't even follow this
example.

> Etc.  A usefully general pragma facility can't be restricted to the
> top of the module, because not all useful pragmatic choices apply to
> an entire module.

Introduction of the directive keyword needs a transition
mechanism. When other uses are found, that restriction might need to
be dropped. According to PEP 5, a transition period is necessary;
starting today would be better than starting with the next release
(since in Python 2.1, the directive won't be used for anything but
requesting nested scopes).

Of course, other parsing techniques could allow to drop this
restriction even today; e.g. allowing 'name name' at indent 0, iff the
first name is 'directive'.

> Make it general enough to be useful, and it will get hairier than that.  Bite
> the bullet:  if you want a new statement, make directive a reserved word.

It *is* a reserved word. However, PEP 5 requires a transition period,
so the reservation is not enforced in Python 2.1.

> Then we can introduce it via
> 
>     from __future__ import directive_is_reserved
> 
> in the release before it breaks code <0.7 wink>.

I knew somebody would propose that; I didn't guess it would be you...

> Note that it's also not enough to allow just one or two optional tokens
> following "directive".

Is it not?

> One of the primary reasons no pragma facility ever got into Python
> is that in previous attempts nobody could agree on syntax that *was*
> sufficiently general (as a hint of the horrors you're signing up
> for, note that "optional static typing" fits-- because of the
> "optional" part --into a pragma framework; and once people realize
> that, they're going to hijack your proposal into an
> overly-generalized mess too).

I see that differently. First, I think it is quite general, as it
allows an arbitrary atom following the directive's name (e.g. a
dictionary). Furthermore, it is not necessary to agree it is general
enough for all unforeseen needs - it is sufficient if it is necessary
for today's needs, and perhaps for the foreseen ones.

> IMO it would be *much* better for everything *except* incompatible language
> changes, for which latter purpose I dislike it.  The future_statement is
> specifically aimed at the latter, and I believe will support it very well.

If you look at it carefully, you find that it does not "support
incompatible language changes". In the proposed use for nested scopes,
it actually does not - there is no incompatible change. The
incompatible change will occur when nested scopes are activated by
default, at which time the future import is not needed. Instead, it is
a boolean switch for an optional language feature.

So you seem to say that language evolution will occur in terms of
boolean switches; something which I cannot predict accurately.

> sell-it-on-its-real-merits-ly y'rs  - tim

I'm certain that supporting binary switches is one of its merits...

Regards,
Martin





From fredrik at pythonware.com  Fri Mar 30 10:28:07 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 30 Mar 2001 15:28:07 GMT
Subject: list vs tuple
References: <hGlw6.782$p5.3210@news1.rivrw1.nsw.optushome.com.au> <3AC20CD3.265A2D94@alcyone.com> <ru69cto61jtmrqsf809e38hb3pbm2lhi7g@4ax.com>
Message-ID: <bC1x6.4319$4N4.905946@newsc.telia.net>

Michael Prager wrote:
> To expand on that, be aware that Python has a rather fanciful
> use of the assignment operator for "mutable" objects

there is no assignment operator.  (it's a statement)

and it doesn't care whether the object is mutable or not.

more info here:

    http://effbot.org/guides/python-objects.htm

Cheers /F




From Patrick.Bussi at space.alcatel.fr  Thu Mar 29 07:54:34 2001
From: Patrick.Bussi at space.alcatel.fr (Patrick.Bussi at space.alcatel.fr)
Date: Thu, 29 Mar 2001 14:54:34 +0200
Subject: =?iso-8859-1?Q?R=E9f._:_Re:_Newbie_question_:_string_processing_w?=
	=?iso-8859-1?Q?ith_RE_-_Thanks?=
Message-ID: <C1256A1E.00472A39.00@vzmta01.netfr.alcatel.fr>

Thanks to all who answered concurrently. I was stuck with this issue. Now that
I've got the answer it seems obvious (as usual).

I'm amazed by the efficiency of this python-list.

---
Patrick Bussi
patrick.bussi at space.alcatel.fr

Any opinions expressed are my own and not necessarily those of my Company.





"Terry Reedy" <tjreedy at udel.edu> on 28/03/2001 07:51:36

Pour :    python-list at python.org
cc :   (ccc : Patrick Bussi/ALCATEL-SPACE)
Objet :   Re: Newbie question : string processing with RE



I believe the indicated changes will work.

>  f3 = urllib.urlopen ("www.such-a-nice-webpage.com/index.html",
encoded_data)

f3 = urllib.urlopen ("www.such-a-nice-webpage.com/index.html",
encoded_data).read()

>  p=re.compile("window.open")
>  q=re.compile("document.location =" )
>  print p,'\n',q
>  t1=p.search(f3.read())
>  t2=q.search(f3.read())

t1=p.search(f3)
t2=q.search(f3)




--
http://mail.python.org/mailman/listinfo/python-list






Cordialement / Best Regards
                                                                     Patrick
BUSSI

ALCATEL SPACE INDUSTRIES
Ground and Operation Division - Technical and Products
Network Departement Manager
Tel : +33 1 46 52 62 31  /  Fax : +33 1 46 52 30 80






From cce at clarkevans.com  Sun Mar 25 23:08:01 2001
From: cce at clarkevans.com (Clark C. Evans)
Date: Sun, 25 Mar 2001 23:08:01 -0500 (EST)
Subject: Yet Another PEP:  Query Protocol Interface or __query__
In-Reply-To: <5.0.2.1.0.20010325210220.00a2cc00@mail.inet.com.br>
Message-ID: <Pine.LNX.4.21.0103252253550.11867-100000@clarkevans.com>

On Mon, 26 Mar 2001, Carlos Ribeiro wrote:
> 1) What is a protocol? The simplest definition that I cant think of is: a 
> protocol is just a bunch of methods. An object is said to support a 
> protocol if it implements all methods of the protocol specification. The 
> specification defines the names and parameters of all methods.

This is what an interface is, and there is a proposal 
for this.  A protocol should be more broadly defined, no?

IMHO, the goal was to have this dialogue...

Hello protocol manager, do you know if this here
object supports the protocol Y?

  The protocol manager than says, "hang on a sec"
  and runs about doing stuff like this...

     Hello object X, do you support the protocol Y? 
       If yes, great, thank you!
       If no, can you make me a wrapper that does?
          If yes, great, thank you!
          If no, hmm, I'll go ask the protocol then.

     Hello protocol Y, is X compliant with your requirements?
       If yes, great, thank you!
       If no, can you make a wrapper for X that does?
          If yes, great, thank you!
          If no, well then... hmm.
  
     Perhaps the manager tries other methods (not implemented
     at this time, such as a registry)

   In any case, the manager goes back to the caller, and 
   says either:
     
     Hello Caller, here is an object Z that is compliant
     with the protocol, either it is the same object or a wrapper.

     Hello Caller, I've been unable to find a compliant
     version of this object for you... very sorry.
          
> 2) A 'protocol instance' is the minimal object instance that completely 
> supports the protocol.
> 
> 3) Now we're getting closer. My proposal is to have a method called 
> __adapt__ that returns a 'protocol instance'.
> 
> 4) So __adapt__ must convert a 'object instance' into a 'protocol 
> instance'. If the object implements the standard names with the default 
> names, this is pretty straighforward - the object could be used as a 
> protocol instance.
> 
> 5) The problem occurs when the object supports the protocol using methods 
> with different names. Assume that we have two different protocols asking 
> for methods with the same names, but with different semantics and/or 
> parameters. An object that supports both protocols will need some way to 
> map its internal methods to the ones defined by each protocol.
> 
> 6) We can solve (5) using a 'proxy', or 'adapter'.

Exactly.  The specific proposal, however, is a two-way (and perhaps
more than two way) lookup mechanism.  This is alot of legwork
to make for "smart" protocols. It shouldn't be automatic (like coersion)
but should be simple to use and extensible.
 
> EXAMPLE
> 
> Assume that we have these protocols:
> 
> a) protocol 'IntegerDivision'
>     supports the __div__() method for integer division
> b) protocol 'FloatDivision'
>     supports the __div__() method for floating point division

Hmm... This moves into Interface territory, I'd rather not go there.
I am *specifically* not defining what a protcol is!  That's why I
chose the nebulous word.  It's the bi-directional lookup mechanism
that is being proposed, no?

> 
> p.s. I think that it is possible to devise a highly optimized 
> implementation. In fact, if all protocols are known at compile time, this 
> should make it easier to optimize method calls, but this need to be 
> explored more carefully. This does not need to be made right now - as Clark 
> pointed out, may be it's better to see if people actually uses this.



> pp.ss. It was not my original intention, but we could use this mechanism, 
> as exposed in my example, as a way to implement the PEPs that propose 
> semantic changes on the numeric types.

Hmm.  This is more along the lines of what I'm thinking, 
not just method names.  Afterall, my "next()" may be 
completely different than yours.   What does yours do?
My "next()" replaces updates "self" so that it can be 
used to fetch information about the next object in a 
sequence, think Recordset.  Note, this is different from 
the upcoming iterator proposal, which has completely 
different semantics.

;) Clark         




From ws-news at gmx.at  Fri Mar  9 05:09:54 2001
From: ws-news at gmx.at (Werner Schiendl)
Date: Fri, 9 Mar 2001 11:09:54 +0100
Subject: Can anyone recomend a good intoduction to C...
References: <mailman.983894288.19965.python-list@python.org> <3AA5EF59.507F1D91@olsen.ch> <983958878.976344@newsmaster-04.atnet.at> <989cvj$sba$1@panix3.panix.com>
Message-ID: <984132875.222868@newsmaster-04.atnet.at>

What's TMTOWTDI, please?




From tsatter at purecode.com  Wed Mar  7 12:41:05 2001
From: tsatter at purecode.com (Tom Satter)
Date: Wed, 7 Mar 2001 10:41:05 -0700
Subject: Can anyone recomend a good intoduction to C...
References: <3aa366c6_2@nnrp1.news.uk.psi.net> <8gNo6.440$y6.128357@ruti.visi.com> <3aa6551e.212870101@news.laplaza.org>
Message-ID: <bpr589.8jq.ln@tethys.purecode.com>

"Mats Wichmann" <xyzmats at laplaza.org> wrote in message
news:3aa6551e.212870101 at news.laplaza.org...
> On Mon, 05 Mar 2001 14:19:48 GMT, grante at visi.com (Grant Edwards)
> wrote:
>
> Sheesh, an ADM3a.  What luxury. On that you can run vi.  The ADM3, on
> the other hand, only allowed "open" mode - a little better than a line
> editor but not much.  (a totally pointless historical tidbit: it was
> the 3a that allowed the vi mode of ex to creep out to a wider
> population at UC Berkeley and thence to become inflicted on the rest
> of the world. Closely followed by the much-less reliable HP 2621.)
>
> (oops, I smell a flood of "I can top that" comments such as "you had a
> terminal? I learned on a Teletype", "you had a Teletype? I had to
> toggle code in using switches", "you had switches?  I had to use
> Forth!".  Consider that joke already played out, eh).

...our dad and our mother would dance about on our graves,
singing 'alleluia.
And you try and tell the young people of today that.
They won't believe you.

--
tom -- just plain old tom
tsatter at purecode.com




From h-karls at telia.com  Tue Mar  6 12:09:58 2001
From: h-karls at telia.com (Håkan Karlsson)
Date: Tue, 06 Mar 2001 17:09:58 GMT
Subject: which IDE?
References: <3A95954A.72621882@cs.strath.ac.uk> <mailman.982883343.12036.python-list@python.org>
Message-ID: <GR8p6.22103$AH6.3098852@newsc.telia.net>

Lee Reilly wrote:

>I'm relatively new to Python and have been developing a product using
>Zope and experimenting using IDLE. I was wodnering if anyone could tell
>me the name of a good IDE - something similar to Symantec Visual Cafe
>perhaps, where I could develop my code and it can spice it up
>(text/variable highlighting, etc.).

You could always have a look at PythonWorks Pro at
http://www.pythonware.com/products/works

Then there is Blackadder from theKompany
and Komodo from ActiveState but I believe they are
still in beta.

/Hakan





From grante at visi.com  Mon Mar  5 12:55:54 2001
From: grante at visi.com (Grant Edwards)
Date: Mon, 05 Mar 2001 17:55:54 GMT
Subject: Coroutines, and memory leaks
References: <8oh089.7it.ln@127.0.0.1>
Message-ID: <KqQo6.463$y6.137345@ruti.visi.com>

In article <8oh089.7it.ln at 127.0.0.1>, David Given wrote:

>Two questions here. Both this relate to a project I'm currently
>working on, a Gnome MUA written in Python (see
>http://sqmail.sourceforge.net).
>
>1. Does Python do coroutines?
>
>Reason: at one part of the program I want to run a
>command-line-like routine, that processes and emits log
>messages at intervals. I want to display the result in a
>window.

If you're running an external program, the simplest thing to do
is to run the command with popen2() and attache a handler to
the file-handle.  The GTK main event loop will then call the
handler anytime there's data available. If you want an example
you can take a look at

   ftp://ftp.visi.com/users/grante/python/lcom.py

The function serialRxHandler() is attached to a file-descriptor
so that anything received on that descriptor is displayed in a
window.  In my case that file descriptor is a serial port, but
it works the same for a pipe or socket.

If you're not running an external program, you could create a
pipe, and then fork -- running your "background" stuff in the
child process, sending messages to the parent via a pipe.

>The most elegant way of connecting this routine to GTK is to
>put them in different threads; but since I don't need true
>concurrency, coroutines would be the simplest solution.

Just remember that only one thread can do GTK stuff.  Any other
"worker" threads have to go through the GUI thread to get at GTK.

The nice thing about the pipe-to-a-different-process method is
that GTK includes a mechanism for attaching handlers to file
descriptors.  There might be something similar for inter-thread
fifos, but I don't know about it.

-- 
Grant Edwards                   grante             Yow!  This PIZZA symbolizes
                                  at               my COMPLETE EMOTIONAL
                               visi.com            RECOVERY!!


From nelson at crynwr.com  Tue Mar 27 13:10:38 2001
From: nelson at crynwr.com (Russell Nelson)
Date: Tue, 27 Mar 2001 13:10:38 -0500
Subject: Linux GPS program
References: <3AC01637.4EF3A6F2@crynwr.com> <99q3ob$ehr$1@newsserver.rrzn.uni-hannover.de>
Message-ID: <3AC0D79E.ADED2832@crynwr.com>

Bernhard Reiter wrote:
> will pygps be Free Software? I could not find anything about a
> license in the cvs.

Yes, it will.  Right now, I want people to fetch it from the CVS so
they have the latest-n-greatest.

> If it was I will be happy to add it to the www.freegis.org
> link collection.

It's free software now -- I just don't want rogue versions that would be
hard
for me to integrate back into the mainline branch.  Feel free to suggest
changes.

-- 
-russ nelson will be speaking at http://www.osdn.com/conferences/brie/
Crynwr sells support for free software  | PGPok | Watch out!  He's got
an
521 Pleasant Valley Rd. | +1 315 268 1925 voice | opinion, and he's not
Potsdam, NY 13676-3213  | +1 315 268 9201 FAX   | afraid to share it!


From DavidlNo at spammraqia.com  Thu Mar 29 12:49:27 2001
From: DavidlNo at spammraqia.com (David Lees)
Date: Thu, 29 Mar 2001 12:49:27 -0500
Subject: Looking for Python equivalent to a Perl'ism
Message-ID: <3AC375A7.A4BA87F1@spammraqia.com>

I am trying to figure out how to do the Python equivalent of the
following perl code, which finds all the groups tagged by 'xx' and
replaces them with text looked up in a dictionary:
----------
my %dict;
my $string="this is xxfooxx this is junk xxmumblexx and more stuff";
%dict=(foo=>"yechfoo",mumble=>"yukkomumble");

$string =~ s/xx(\S+)xx/$dict{$1}/g;

print $string;
---------------------

My Python equivalent without the backreference dictionary lookup is:

dict ={'xxx':'foo',
       'yyy':'barf',
       }
p=re.compile(r'zz(\w+)zz')
s='this is and so is zzxxxzz abc that an zzyyyzz abc zzxxxzz or else'
print re.sub(p,'junk',s)

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

But I do not understand how to include backreferences into the regex
substitution as is conveniently done in perl.  Can someone tell me if
there is an easy Python solution to this?

Thanks in advance.

David Lees


From simon at callan.demon.co.uk  Fri Mar  2 15:07:54 2001
From: simon at callan.demon.co.uk (Simon Callan)
Date: Fri, 02 Mar 2001 20:07:54 +0000
Subject: Collection interfaces
References: <3a90fa97.0@news1.mweb.co.za> <3A933830.7F98D118@ix.netcom.com> <CgJk6.515$v77.66568@newsread2.prod.itd.earthlink.net> 
 <220220010813598042%cmh@bDistributed.com> <ud7cakvi8.fsf@roundpoint.com> <MPG.150312ceb68f1ada98969f@news.earthlink.net> 
 <8Gem6.473353$U46.14105062@news1.sttls1.wa.home.com> <MPG.1503548620e632919896a4@news.earthlink.net> 
 <MCjm6.474652$U46.14163288@news1.sttls1.wa.home.com> <MPG.150394195f7f04a19896a7@news.earthlink.net> <3A9A86B1.9A8C69A2@ix.netcom.com> 
 <MPG.1504d14563dce6499896af@news.earthlink.net> <97fjla128ir@news2.newsguy.com> <MPG.1505a76831c34b749896b3@news.earthlink.net> 
 <97h5s40vhc@news2.newsguy.com> <MPG.1506007591cf818b9896bc@news.earthlink.net> <slrn99opop.gr.amk@mira.erols.com> 
 <MPG.1506265ebf6d934d9896be@news.earthlink.net> <97ibmc0k47@news1.newsguy.com> <MPG.1506cf3293b094b49896c1@news.earthlink.net> 
 <97m0c7080c@news2.newsguy.com>
Message-ID: <6e7a18554a.simon@satin.callan.demon.co.uk>

In message <97m0c7080c at news2.newsguy.com>
          "Alex Martelli" <aleaxit at yahoo.com> wrote:

> there STILL is no silver bullet.

Years back, I became aware to the treu reason why there is no silver
bullet for programming, and why there can *never* be one. We're trying
to kill the wrong monster.

As any student of horror will tell you, silver bullets are the
solution for dealing with were-wolves, which are relatively simple
creatures, who only want to rip your throat out and eat you. Once you
manage to kill one of these, they have the decency to stay dead, and
not bother you any more.

We are not dealing with this monster.

When you are programming, you are wrestling with a vampire, a subtle,
intelligent enemy, that slowly drains the life out of you over a long
period of time. What makes it worse is that the vampire can return
from the dead, usually because some idiot (manager?) wants it back,
thinking that they can benefit from it.

When fighting a vampire, your only weapons are sunlight (knowledge?),
a sharp stake (language / IDE?), and possibly a cat[1].

This still leaves the problem of finding the right wood to make the
stakes.

Simon

[1] Vampires have risen from the deas, the grave, and the crypt, but
have never managed it from the cat - Witches Abroad, Terry Pratchett

-- 
http://www.callan.demon.co.uk/simon/


From bsass at freenet.edmonton.ab.ca  Thu Mar  1 16:37:23 2001
From: bsass at freenet.edmonton.ab.ca (Bruce Sass)
Date: Thu, 1 Mar 2001 14:37:23 -0700 (MST)
Subject: CPAN functionality for python - requirements
In-Reply-To: <ME5n6.77748$m5.4881872@news1.rdc1.ga.home.com>
Message-ID: <Pine.LNX.4.33.0103011418020.13134-100000@bms>

> > I would not use it if it means bypassing the native package management
> > system, and would be willing to wait for "official" packages rather than
> > use third-party packages.
>
> What "official" packages do you mean here?

Stuff packaged by a developer who works for the distro.
For me that is someone who has gone through Debian's New Maintainer
process (their real identity has been verified, they have demonstrated
they know how to build Debain packages, etc.).


- Bruce




From bsass at freenet.edmonton.ab.ca  Sat Mar 31 17:32:43 2001
From: bsass at freenet.edmonton.ab.ca (Bruce Sass)
Date: Sat, 31 Mar 2001 15:32:43 -0700 (MST)
Subject: list vs tuple
In-Reply-To: <lWrx6.1057$p5.4152@news1.rivrw1.nsw.optushome.com.au>
References: <mailman.986066384.30712.python-list@python.org>
 <d1rx6.1053$p5.4024@news1.rivrw1.nsw.optushome.com.au> <slrn9cciaj.3ug.qrczak@qrnik.zagroda>
 <lWrx6.1057$p5.4152@news1.rivrw1.nsw.optushome.com.au>
Message-ID: <Pine.LNX.4.33.0103311444580.7534-100000@bms>

On Sat, 31 Mar 2001, deadmeat wrote:

> > THERE IS NO DIFFERENCE.
>
> What happens to the objects is not the point, what happens to what I put
> into a and b is the point.
>

line by line...

> >>> a = 1

"a" points to "1"

> >>> b = a

"b" points to the same thing "a" points to, not to "a" itself

> >>> a = 2

"a" now points to "2"...

> >>> b
> 1

...and "b" still points to "1"

> >>> a = [1,2,3]

"a" points to the object with the representation "[1,2,3]"

> >>> b = a

"b" points to the same object "a" points to

> >>> a[0] = 0

this modifies the first item in the sequence (or mapping) "a" points
to...

> >>> b
> [0, 2, 3]

...and since "b" points to the same object, you can see the change via
the reference from "b"

> Are you saying there is no difference (IN THE *RESULT*) between these?

Of course there is... but why should playing with references to
multiple objects have the same result as mutating a single object with
multiple references?

> Are
> you able to read?

I'm sure other are wondering the same of you (since the mindset you
need to get into has been explained, a few times)...

> What goes on in the background is NOT THE POINT. I only care about the
> values I put into them, and what happens after b = a.

...there is the way it is, and the way you think it is.  What goes on
in the background is the point, because that is how the language
works.  You may only care about what the values are, but you need to
do so within the context of the language you are using.  If the
language defines "this = that" as operating on namespaces only, then
it is best to start thinking like that (instead of thinking that it
does something with data).


- Bruce




From jwbnews at scandaroon.com  Tue Mar 13 16:31:34 2001
From: jwbnews at scandaroon.com (John W. Baxter)
Date: Tue, 13 Mar 2001 13:31:34 -0800
Subject: This math scares me
References: <3aad4062.1290021@News.CIS.DFN.DE> <20010312170356.B3513@better.net> <mailman.984480856.18097.python-list@python.org> <984486395.407379@newsmaster-04.atnet.at> <slrn9asf1d.ja3.qrczak@qrnik.zagroda> <984518199.127140@newsmaster-04.atnet.at>
Message-ID: <jwbnews-09D75F.13313313032001@corp.supernews.com>

In article <984518199.127140 at newsmaster-04.atnet.at>,
 "Werner Schiendl" <ws-news at gmx.at> wrote:

> True, but even 0.1 and 0.2 need not necessarily be exact values.

Of the single decimal digit numbers, 20% are exact (x.0 and x.5, for not 
huge x).  For two decimal places, it's down to 4% (x.00, x.25, x.50, 
x.75) and it spirals quickly downward from there.

Bill Gates got this right in the early days, delivering two Basic 
interpreters:  binary floating point for speed, and decimal floating 
point for money and such.  (I never looked at how well his interpreters 
did...I just used them.)

  --John

-- 
John W. Baxter   Port Ludlow, WA USA  jwbnews at scandaroon.com


From parz at home.com  Thu Mar  1 01:12:01 2001
From: parz at home.com (Parzival Herzog)
Date: Thu, 01 Mar 2001 06:12:01 GMT
Subject: Unresolved symbols in MSVCRT.DLL when running a py2exe packaged program.
Message-ID: <RKln6.6630$hn5.853203@news1.rdc1.mb.home.com>

I have a Python application (which uses the wxPython extension) that I have
packaged
for distribution with py2exe.  The dfistributed app works on my own
computer,
and on another machine I have, which has never seen any Python software, nor
any
development software (i.e. no dll's not a part of a standard Win2000
installation are present).

I delivered my app to my client, who is expected to modify some
configuration code, then
use py2exe to package the app, and run the packaged app on a server machine.
My client
has a notebook computer, running Win NT (Sp 6), plus MsVc and assorted
development tools.
My app worked on that noebook, and py2exe again packaged the app and the
package was
copied to the server machine, a Win NT sp 6 desktop computer. When we
attempted to
start the app, it failed immediatly, with the message that the symbol
??_U at YAPAXI@Z could
not be found in MSVCRT.DLL.

That symbol is apperrently defined in the MSVCRT.DLL file in the py2exe
package, but the
MSVCRT.DLL in the package is not the same as MSVCRT.DLL in the windows
system32
directory, and that version of the dll appears to have the symbol
??2 at YAPAXCI@Z defined
(what horrible identifiers!). So we assumed that instead of finding the
py2exe supplied
dll, the packaged python app is finding the other dll, which is much older
(dated 1997).
Of course, we don't have any way of determining the dll that is actually
being searched).
So, we attempted to replace the old MSVCRT, by installing a recent copy of
MSVC++ 6.0.
This indeed did replace the dll, replacing it with a version dated July 7,
2000. However,
the problemwith the python app persisted exactly as before.

What I can tell, with the computers I have available, is that the the Python
app works
on three different machines, 2 running Windows 2000, with the latest
MSVCRT.DLL
and one running Win NT (sp 6), with an MSVCRT.DLL dated 1999. What I can't
tell
is whether or not the py2exe packeged MSVCRT.DLL is being used by the python
app,
or the system32 dll that happens to be on the machine where the app is to
run.

Can anyone help me towards resolving this problem? Truly, I program in
Python, because I
really am far too stupid to cope with ??2 at YAPAXCI@Z missing symbol problems,
and I
fancy that I shouldn't have to either. But I appear to be misguided, and I
need some help.

- Parzival






From charlie at gabrielinteractive.com  Fri Mar 16 10:48:19 2001
From: charlie at gabrielinteractive.com (Charlie Barrows)
Date: Fri, 16 Mar 2001 10:48:19 -0500
Subject: PyImport_ImportModule
Message-ID: <98tcks$7bl$1@allthetime.news.cais.net>

I'm just starting to tinker with embedding Python in a C app and I'm trying
to get a handle on where modules end up when they're imported.

If I run this code:

PyObject *module = PyImport_ImportModule("TestScript");

... which namespace is TestScript imported into?

And can someone give me some hints as to some possible parameters for
PyImport_ImportModuleEx.  Specifically, what should the globals and locals
be?


Charlie Barrows
Gabriel Interactive




From mickeyz at qwest.net  Tue Mar 20 04:01:01 2001
From: mickeyz at qwest.net (Randy A. Ynchausti)
Date: Tue, 20 Mar 2001 02:01:01 -0700
Subject: Collection interfaces
References: <3a90fa97.0@news1.mweb.co.za>   <8Gem6.473353$U46.14105062@news1.sttls1.wa.home.com> <MPG.1503548620e632919896a4@news.earthlink.net> <MCjm6.474652$U46.14163288@news1.sttls1.wa.home.com> <MPG.150394195f7f04a19896a7@news.earthlink.net> <3A9A86B1.9A8C69A2@ix.netcom.com> <MPG.1504d14563dce6499896af@news.earthlink.net> <97fjla128ir@news2.newsguy.com> <MPG.1505a76831c34b749896b3@news.earthlink.net> <97h5s40vhc@news2.newsguy.com> <MPG.1506007591cf818b9896bc@news.earthlink.net> <slrn99opop.gr.amk@mira.erols.com> <MPG.1506265ebf6d934d9896be@news.earthlink.net> <97ibmc0k47@news1.newsguy.com> <MPG.1506cf3293b094b49896c1@news.earthlink.net> <97m0c7080c@news2.newsguy.com> <MPG.15085517debf4b649896ca@news.earthlink.net> <zb%n6.31033$o%6.586026@news1.rdc1.bc.home.com> <MPG.150b383cea226d409896d0@news.earthlink.net> <kNko6.6986$eq3.707121@news.uswest.net> <MPG.150c46db9cec283e9896d4@news.earthlink.net> <L9Go6.2720$hZ4.875436@news.uswest.net> <DWCt6.12229$Im6.1309770@newsread1.prod.itd.earthlink.net>
Message-ID: <jWEt6.7867$3_.998377@news.uswest.net>

topmind,

> Okay.
>
> For the sake of argument, let's say that OO shines if the app is very
mathy.
>
> But, what about the rest?

What the rest any different from the very mathy?  The classes and objects
don't care what their behaviors are.

It use to be that people said you should never implement a compiler using OO
because for all of the "obvious" reasons.  Then someone did it and found out
that their solution was essentially as fast, easier to maintain, and
produced a smaller code base.  As far as I am concerned the same experience
was had by the "data-visualization" industry.

Now -- where is that procedural code -- I have the OO code in hand?

Regards,

Randy




From yarlevub at aol.com  Sat Mar  3 12:42:49 2001
From: yarlevub at aol.com (YarLevub)
Date: 03 Mar 2001 17:42:49 GMT
Subject: no zlib in Python 2.0
References: <3AA09400.E5174688@pacbell.net>
Message-ID: <20010303124249.10273.00000330@ng-fk1.aol.com>

>I've recently installed Python 2.0 on a Mandrake 2.2.17 system. When I
>invoke python I get the following version information:

[snip]

You need to enable the modules you want  (they are off by default).  The
instructions for doing this are in the "Configuring additional built-in
modules" section of the README file.


From kens at sightreader.com  Sun Mar 18 11:38:30 2001
From: kens at sightreader.com (Ken Seehof)
Date: Sun, 18 Mar 2001 08:38:30 -0800
Subject: standalone
References: <Nu_s6.244274$v.26337040@typhoon.nyroc.rr.com>
Message-ID: <007401c0afc9$defd6690$a325fea9@his>

Check this out:
http://www.mcmillan-inc.com/builder.html

----- Original Message -----
From: "Vyacheslav Danovich" <vdanovich at empireone.net>
Newsgroups: comp.lang.python
To: <python-list at python.org>
Sent: Sunday, March 18, 2001 12:39 AM
Subject: standalone


> Hi!
> I'm pretty new to Python.
> I need to run an executable Python script in Windows environment which
does
> not have Python library installed.
> My questions are: can it be done and if "yes" what steps should I take
>
>
> Thank you very much
>
> --
> ############################
> Mr. Vyacheslav L. Danovich
> 103 Bleeker Rd Apt 7
> Guilderland, NY 12084
> Telephone: (518) 452-4403
> Facsimile: (518) 452-4403
> Mobile : (518) 221-4403
> ############################
>
>
> --
> http://mail.python.org/mailman/listinfo/python-list
>




From urner at alumni.princeton.edu  Sat Mar  3 01:06:46 2001
From: urner at alumni.princeton.edu (Kirby Urner)
Date: Fri, 02 Mar 2001 22:06:46 -0800
Subject: Adding a method to an instance
References: <G2%n6.360714$JT5.11606234@news20.bellglobal.com>
Message-ID: <6621ato2vknc667m3iuqvmm8vcj0h99qqk@4ax.com>

Bryan Mongeau <bryan at eevolved.com> wrote:

>Quick one for you guys:
>
>I would like to add a method to an instantiated class. 
>
>class foo:
>  def __init__(self):
>    self.bar = 123
>
>f = foo()
>
>Now to add a method to the instance f of class foo:
>
>def newMethod(self):
>  print self.bar
>
>f.__class__.newMethod = newMethod

In saying f.__class__, you point back to the superclass foo.
If you want to poke newMethod into object f alone, then you 
should just go:

f = newMethod

Here's a transcript that addresses your question:

 >>> class foo:
	def duh(self):
            print "Duh!"

            
 >>> o = foo()
 >>> o
 <__main__.foo instance at 00B9719C>
 >>> k = lambda x: x+2
 >>> k(2)
 4
 >>> o.othermethod = k
 >>> o.othermethod(3)
 5
 >>> dir(o)
 ['othermethod']
 >>> dir(foo)
 ['__doc__', '__module__', 'duh']
 >>> o.__class__
 <class __main__.foo at 00BB3ECC>
 >>> foo
 <class __main__.foo at 00BB3ECC>
 >>> o
 <__main__.foo instance at 00B9719C>

Note re the last three commands:  o.__class__ is a pointer to 
the class foo in memory, so you would be adding to the superclass
to put a method in there.  You want to insert your method into
the instance/object, so don't involve its __class__.  

I bet you get it now.

Kirby



From mcherm_python at yahoo.com  Sun Mar 18 12:49:20 2001
From: mcherm_python at yahoo.com (Michael Chermside)
Date: Sun, 18 Mar 2001 11:49:20 -0600
Subject: Who's minister of propaganda this week?
References: <B98041329020E007.7DA84CAD9D0562AB.817BBC572F480869@lp.airnews.net> <98nu1m$12e@dispatch.concentric.net> <660ACCA1E4816F7E.C359CDF4168EF3FB.3B6B05538D619770@lp.airnews.net> <3AAFA13D.76E32C3F@divalsim.it> <98oa4o01i4v@news2.newsguy.com> <mailman.984619039.7856.python-list@python.org> <98q9h60122v@news2.newsguy.com>
Message-ID: <3AB4F520.40104@yahoo.com>

First of all, I want to thank Alex for engaging me in this discussion...
if nothing else, I'm sure to learn something before we're through.


Alex Martelli wrote:

> "Michael Chermside" <mcherm_python at yahoo.com> 
> <mailto:mcherm_python at yahoo.com> wrote in message
> news:mailman.984619039.7856.python-list at python.org...
> 
>> Alex Martelli wrote:
>>          ... [snip]...
>>   > So, all the compile-time checking is buying is catching (a small
>>   > subset of) the errors that would be caught in testing anyway, a
>>   > little bit earlier (thus, a little bit cheaper) -- it's never the
>>   > case that one has to write a test which would not be needed at
>>   > all if type-checking was static, since "the object has the right
>>   > type" is a small subcase of "the object _behaves_ per the specs".
>>   >
>> I'm really not sure I see it this way. If the method  foo(x) is known to
>> take a FancyDateObject
> 
> 
> 
> Assume that is an abstract interface (no gain in terms of
> functionality assurance if it's concrete) and (without loss
> of generality) that it has two methods First and Second that
> foo uses (it may have others that foo is ignoring, but they
> don't affect the following argument).  OK so far?  Good.


Yep... ok so far

> 
>> there are three kinds of errors we could make. One is that foo() is
>> written badly so it
>> doesn't do what it's supposed to. The unit tests of foo() need to guard
>> against this.
> 
> 
> 
> Right.  Specifically, they'll test the combinations of calls
> to First and Second methods of the x argument that foo needs
> to actually perform, say in certain cases First only, in
> others Second only, in others yet, First then Second (if foo
> never needs to call Second before, and First after, on its
> argument, then its unit-tests will not exercise that path).
> 
>> Another possible error is that FancyDateObject isn't written properly.
> 
> 
> 
> As I assumed this is an interface, let's say you're talking
> about some specific implementation thereof -- FDO_impl1, say.
> 
>> The unit tests
>> of FancyDateObject need to guard against this.
> 
> 
> 
> Right again -- specifically, they'll test that FDO_impl1's
> implementations of First and Second support the call patterns
> that the specification demands.  For example, if that is what
> the specs say, the implementation will work fine if only First
> is called, or if only Second is called, or if Second is called
> before and _then_ First is called -- 'First before, Second after'
> may not be in the specs and thus doesn't get exercized by unit
> tests.


Okay... we're still together here.

> 
>> And the third type of
>> error is that
>> that somewhere where we CALL foo(), we might pass it a DateObject
>> instead... or
>> even a String... which will cause it to perform wrong. To guard against
> 
> 
> 
> More generally: we might erroneously pass to foo some object that
> does not even _implement_ First and Second methods with acceptable
> signatures, or some that _does_ implement the methods with signatures
> that appear good BUT not with the semantics that foo needs -- e.g.,
> it does not let First be called earlier and Second later (because
> there is a mismatch in semantics specs between the specs that foo
> requires and the specs that x actually ensures -- just as in the
> other cases, regarding existence or signature of the methods).
> 
> No compiler wards you at compile-time against all, or even _most_,
> errors of this very common kind.  (The existence of more than one
> method is not needed to have this behavior -- just a single method
> suffices to exhibit this error, e.g. it could take an argument i
> with a prereq of i>23 and be called with i==23 by foo; or foo might
> call it 7 times when the semantics specify it must be called no
> more than 6 times; etc, etc).
> 
> Another way to express it: an interface is not just, not even
> _mostly_, about existence and signature of methods -- it's mostly
> about prereq's, post-conditions, and invariants; and nobody can
> check those at compile-time in enough cases to make a difference
> to your software's reliability.


Okay, this is an interesting point.

First of all, I completely agree that telling whether an object has
the right method names and signatures (an "interface" in the java
sense) is only a very limited subset of defining the signatures,
pre-requisites, post-conditions, and invariants. Eiffel allows
something like this... let's call the idea a "contract". I
think it would GREAT if we could specify our strict typing in terms
of "contracts" instead of the more limited "interfaces".
Compilers could NOT be expected to verify at compiletime that the
pre-requisites were being satisfied, but they could at LEAST
verify that the object being used claimed to implement that
contract.

> So, statically-typed languages make a LOT of the not-very-important
> issues of method-existence and signature -- because those issues
> are THE ones they can check statically, not because their importance
> (wrt the importance of real semantics issues, the programming-by-
> contract parts of the interface) warrants special attention.  It's
> like the drunkard who was looking for his housekeys at one end of
> the street, opposite to the end where he had dropped them, because
> the end he was searching at was the one that a streetlamp lit...
> 
> 
>> this in a
>> dynamically typed language, we have to write unit tests for *every
>> single place*
>> that we call foo(). Of course, we'd be writing unit tests for those
>> functions anyway,
>> but we won't be able to assume that foo() works properly, and will need
>> extra tests to ensure this.
> 
> 
> 
> You can never 'assume that foo(x) works properly' for a given x
> in an untested case -- and syntactic-level compatibility between
> the methods x offers (& their signatures), and the ones foo
> requires of its arguments, doesn't buy you much, since the likely
> and troublesome errors are with contract-expectations mismatches.
> 
> If the implementation of foo and its call are inside the same
> component, then unit-tests should exercize the relevant paths
> (or else the component is being released in a state of incomplete
> testing -- thus, dubious reliability, whether with or without
> static typechecks).
> 
> If the implementation of foo and its call are in different
> components, then you have a system-integration problem (again,
> one that remains independently of static type-checking) and
> thus more 'strategic' kinds of troubles.  You **STILL** need
> 'extra' tests to ensure the actually-implemented semantics
> of a given actual argument and the ones foo requires of its
> formal argument match -- for all distinct cases that occur
> in the (system-level) acceptance criteria tests.  If then, in
> later system operation, you meet a failing case that was not
> tested (makes no difference whether the mismatch is in
> method existence, signature, or semantics), then your acceptance
> tests were insufficient -- and static checking would not have
> made them sufficient.
> 
> 
>> In a staticly typed language, we still need the unit tests for cases 1
>> and 3, but the
>> third type of error is caught by the compiler.
> 
> 
> 
> Not in most cases of interest, no.
> 
>> And in my mind, ANY TIME that
>> I can have a machine do my work for me it's better... I can be lazy, and
>> the machine
>> never gets tired after a long day and forgets to test sometimes. Of
> 
> 
> 
> Extra checks (even if theoretically redundant) are good insurance EXCEPT
> where they breed exactly this kind of 'complacency'.  Release procedures
> MUST 'never .. forget to test' either -- and it's not that hard a problem
> to setup your build/release environment so that a machine ensures this.


You misunderstand. I'm not saying that type checking allows me to
be lazy... it certainly doesn't. I'm just claiming that it's better
to have a computer doing the check instead of a human whenever possible.
I believe that statement is quite general... no matter what the test,
it's better to have a computer do it than to count on the programmer.
It catches mistakes sooner and more reliably than humans and (for that
matter) expects less in the way of salary and benefits! As far as I
can tell, you agree with me on this.

>> course, I can
>> put an assert at the top of foo() which asserts that x is of type
>> FancyDateObject,
>> but if I always assert the types of my arguments then I'm basically
>> using a staticly typed language.
> 
> 
> 
> If your 'assert ... is of type' actually runs a sensible albeit small
> 'type'-testing procedure, which exercizes the whole contract that the
> interface implements, then you have gone statically typed languages
> one better -- unfortunately, this is most often impractical (as such
> infinitely-repeated tests are far too slow, can't be made non-invasive,
> etc, etc).  And it doesn't buy you all that much either (a bit more
> than just statically checking types, but not all that much more) --
> you still need to have **exercised** the call-cases of interest.
> 
> Asserting a single, specific, invariable concrete type would be on
> a different plane -- perhaps feasible (at non-inconsiderable cost)
> for some within-the-component cases (where you can commit to never
> needing ANY polymorphism EVER), definitely unfeasible across any
> component-boundary (cfr. Lakos' "Large Scale" book, again -- it's
> still the best treatment I know of dependency management in large
> scale software development, and, in particular, of the inevitability
> of purely abstract interface-classes across component boundaries
> to manage those dependencies; Martin's articles, which can be found
> on his objectmentor site, are more readable, although not quite as
> deep and extensive as Lakos' big book of course).
> 
> 
>> There are times when a dynamically typed language is more flexible, and
>> it's certainly
>> nice not to have to declare everything just to specify its type, but
>> there ARE
>> advantages to static typing, and this, I believe, is the biggest one.
> 
> 
> 
> We agree that the (redundant, but earlier) checks performed by the
> statically-checking compiler are the 'biggest' (least small?-)
> advantage of said compiler (well, apart from speed issues, which
> may at times be paramount for certain well-identified components).
> 
> We disagree on how big that 'biggest' is.  I estimate that (speed
> apart) this specific advantage may buy me about a 5% productivity
> improvement -- so I agree it's an advantage, and I agree it's
> larger than any other advantages of static checks (none of which,
> it seems to me, may make even a 1% further improvement even when
> all taken together -- again, speed of resulting code apart), but
> I don't think it's worth anywhere near the _bother_ (productivity
> impact, _negative_ improvement) of the contortions I have to
> perform to satisfy the checks (which cost me _at least_ 10% of
> my sw-lifetime-coding-productivity, even in the cases where I'm
> least interested in specifically taking advantage of dynamism).


You know.. I understood and agreed with every single point you
made up until this one. I'm agree that static typechecking probably
only buys 5% productivity (maybe less) because the kind of mistake
it catches is the kind which are easy to diagnose and correct
anyhow.

But I'm quite surprised that you think you pay a 10% penalty
for going through contortions to adhere to static typechecking.
You see... in my own experience, I hardly ever want to violate
it. I generally design type hierarchys and use objects in places
where I expect an object of that type or of a direct ancestor.
On the occasions when I want to write something which works on
all objects of a certain behavior, I define an interface for
the behavior (even in languages which don't have direct support
for "interface") and specify that all appropriate objects
implement the interface. In fact, I often write multi-paragraph
comments on how the interface should behave, so that it's really
more of a "contract" than an interface. When I'm being good,
(and I'm good much more often lately) I even write unit tests to
enforce it.

It is actually VERY rare that I do something like (and here I
dig up my most-likely-to-happen example and phrase in terms
of python) write a function expecting a "file-like-object" --
which, in Python, is reasonably-well-understood but not
formally documented behavior (well, set of behaviors). But as
you point out...

> I come to these (tentative) conclusions after a lifetime spent
> working _mostly_ in statically-typed languages


... I'm just learning this stuff. In fact, I have been working
with dynamically typed languages only for a very short time. So
my own take-away is that I should really try to exercise the
ability to make more extravagant use of polymorphism when working
in dynamically typed languages, because I may discover (if my
experience is similar to yours) that I'm saving 10% of my time,
by avoiding awkward contortions that I don't even realize I'm
doing now.

>                                                 -- because the
> performance characteristics of the machines I was targeting just
> didn't afford me the luxury of doing otherwise (when they DID,
> I repeatedly tried out dynamically typed languages, and, over
> and over again, I was tickled pink at how well they worked --
> most people are surprised when they learned how much of the
> 'background processing' of the programs I was doing in the
> mid-80's [on IBM mainframes] was done in Rexx, but then, they
> are equally surprised at how fast and reliably I delivered:-).
> 
> Today, I'm still relying mostly on C++ to earn my daily bread
> (3-D modeling &c being pretty compute-intensive even by today's
> standards -- and 3D for mechanical engineering is that what
> we're mostly doing here at my current employer), but more and
> more dynamically typed code 'sneaks in' (thanks be!-)...
> 
> 
> Alex
> 
Thanks for the detailed input... I still feel like static
types are a good thing (5% gain perhaps) but if there's
something out there that dynamic typing can give me which
is even BETTER (10%?), then I'd REALLY like to find it!

-- Michael Chermside


_________________________________________________________
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com




From park997 at aol.comnotospam  Tue Mar 13 12:32:30 2001
From: park997 at aol.comnotospam (Park997)
Date: 13 Mar 2001 17:32:30 GMT
Subject: main in mac?
References: <mailman.984497176.29067.python-list@python.org>
Message-ID: <20010313123230.19201.00000153@ng-dh1.aol.com>

>I don't know what the run all button is (not on mac), but the whole reason
>for this idiom is so it wont 'do anything' when imported. If a file named
>x.py is imported, then it's __name__ is x. If it is being run as a program,
>then it's __name__ is __main__. 
>
>
>David

I knew that was the expected behavior, and the
disadvantage of replacing if __name__ is __main__. 
with if 1: is that it does run when imported. Turning the run as main option on
or off produces the expected behavior on the mac.

Incidently, I am very impressed with the mac Python
port. It must have been pretty tricky moving everything to a no command prompt
system.

Wendell


From cribeiro at mail.inet.com.br  Tue Mar 27 12:37:59 2001
From: cribeiro at mail.inet.com.br (Carlos Alberto Reis Ribeiro)
Date: Tue, 27 Mar 2001 14:37:59 -0300
Subject: who is my caller?
In-Reply-To: <Pine.LNX.4.21.0103271010460.24308-100000@clarkevans.com>
References: <99q8nv$2aibn$1@ID-11957.news.dfncis.de>
Message-ID: <5.0.2.1.0.20010327142810.02186640@mail.inet.com.br>

At 10:15 27/03/01 -0500, Clark C. Evans wrote:
>This brings up another question, if I
>have a function can I get the module
>handle; or if I have a bound-function can
>I fetch the object that it is part of?

I suggested a similar problem on my recent comment about proxies/adapters. 
You could use a proxy to handle security. Plese check my recent posts on 
this for details:

1) External code logs in and ask the object for a interface.

2) Your object returns a proxy object containing *only* the authorized methods.

3) When the external code calls some method through the proxy, your object 
checks to see if the call comes from the proxy or from some other object. 
The test is done on the address of the caller object.

4) One advantage of this is that the signature of the proxy can't be forged 
in the stack, because it's the address of the proxy itself. This keeps 
being true even if the call is hacked through im_self (in this case, the 
caller would be im_self).


In order for this to work, we need to answer these questions:

Q1) Do we have this information on the stack? More specifically, in the 
case of a proxied call (using my MakeAdapter implementation), do we know 
that the call was done through the proxy?

Q2) Can we be sure that Python objects keep the same memory address at all 
times? I ssume so - it would be very hard to have them moving in memory at 
will :-)

I think we can make it work. May be a fix to the Python stack 
implementation would be needed for the proxied call. However, it should not 
be very hard to do it (I haven't checked the sources yet - I'm tied to my 
windows machine for now, without gcc, and I'll only turn my linux box in 
the next two weeks or so).


Carlos Ribeiro





From sholden at holdenweb.com  Tue Mar  6 19:20:42 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Tue, 6 Mar 2001 19:20:42 -0500
Subject: Can anyone recomend a good intoduction to C...
References: <mailman.983894288.19965.python-list@python.org> <983896344.701456@newsmaster-04.atnet.at> <m3lmqibw0a.fsf@atrus.jesus.cam.ac.uk> <983911749.217622@newsmaster-04.atnet.at> <roy-32CDC5.15461906032001@news.panix.com>
Message-ID: <Xcfp6.23971$ML1.1228363@e420r-atl2.usenetserver.com>

"Roy Smith" <roy at panix.com> wrote in message
news:roy-32CDC5.15461906032001 at news.panix.com...
> "Werner Schiendl" <ws-news at gmx.at> wrote:
> > Years ago I had a lot of discussions why use C instead of Assembler.
> > The arguments were mainly for performance.
>
> I wouldn't be surprised if a good optimizing C compiler can actually
> produce faster code than hand-coded assembler in many cases.

Sun were making this claim about their optimizing C compiler when I worked
for them in the 1980s. Of course, they were trying to justify unbundling it
at the time, but I believe the claims were substantially true. I don't
imagine the optimization has got worse, and assembler is certainly more
difficult for RISC machines with pipelines than on older machines.

And, of course, for any application in a high-level language which runs too
slowly, Moore's law predicts that help is just around the corner...

regards
 Steve





From ws-news at gmx.at  Tue Mar  6 17:04:54 2001
From: ws-news at gmx.at (Werner Schiendl)
Date: Tue, 6 Mar 2001 23:04:54 +0100
Subject: Can anyone recomend a good intoduction to C...
References: <mailman.983894288.19965.python-list@python.org> <983896344.701456@newsmaster-04.atnet.at> <m3lmqibw0a.fsf@atrus.jesus.cam.ac.uk> <983911749.217622@newsmaster-04.atnet.at> <UScp6.1097$y6.196347@ruti.visi.com>
Message-ID: <983916572.286918@newsmaster-04.atnet.at>

> C++ is far more complex that C without being much safer.  Since
> the safety has to be "inspected" into the program, you want
> to make things as easy to inspect as possible: a nice simple
> language.

That is of course a point, if you need to let the program inspect you will
try to keep it as simple as possible.
But given the same functionality, I think a C++ program will not be more
complex.

...as long as it is well designed of course, nobody forces you to use any
number of features at all possible

don't you think so?

Given a 'normal' program I think C++ programs will usually be somewhat safer
as long as libraries are used whereever possible.
(Well tested, quality libraries, of course).

Think about string operations: I think about 50 % of all security issues I
read about are somehow related with a (string) buffer overflow. The string
family of classes does not have that kind of problem.

regards
werner




From rikbo716 at student.liu.se  Thu Mar 22 10:57:43 2001
From: rikbo716 at student.liu.se (Rikard Bosnjakovic)
Date: Thu, 22 Mar 2001 16:57:43 +0100
Subject: Good Editor?
References: <3ab9a32b$1_1@news2.one.net> <3AB9FAC9.97EA420E@student.liu.se> <slrn9bjv7m.4qv.grey@teleute.dmiyu.org> <99d780$l2g$1@laurel.tc.umn.edu>
Message-ID: <3ABA20F7.AB71D3B4@student.liu.se>

Brian Langenberger wrote:

> (E)ventually (M)alloc()s (A)ll (C)omputer (S)torage

Escape Meta Alt Control Shift
Eighteen Megs And Constantly Swapping


> I've spent countless hours editing Python code in Emacs, really.
> Between that and the interactive interpreter, I simply don't
> use anything else.

The best part is C-c ! and C-c C-c in the python-mode. IMNSHO ofcourse.


-- 
Rikard Bosnjakovic - http://a214.ryd.student.liu.se/cv/ - ICQ: 1158217

Anyone sending unwanted advertising e-mail to my address will be
charged $250 for network traffic and computing time. By extracting my
address from this message or its header, you agree to these terms.


From chris.gonnerman at usa.net  Fri Mar 23 08:59:07 2001
From: chris.gonnerman at usa.net (Chris Gonnerman)
Date: Fri, 23 Mar 2001 07:59:07 -0600
Subject: newbe   how to read in a string?
References: <3ABB4E70.54885C9F@noos.fr>
Message-ID: <007f01c0b3a1$713243c0$a100000a@local>

----- Original Message ----- 
From: "William Famy" <william.famy at noos.fr>
Subject: newbe how to read in a string?


> Hi.
> 
> I am looking how to read in a string.
> 
> Ex
> a='2000/12/28,24.3,87.1,65.4'
> 
> I whant to extract the 87.1. In C i would code by a
> scanf("%s,%f,%f,%f\n"),...

Actually what you are asking is how to *parse* a string you have already
read.  The easy way for this particular string is:

    import string
    d,n1,n2,n3 = string.split(a, ",")
    n1 = float(n1)
    n2 = float(n2)
    n3 = float(n3)

If the list of numbers is of variable length, this might be cool:

    import string
    seq = string.split(a, ",")
    for i in range(1, len(seq)): # this skips the 0 index,
        seq[i] = float(seq)      # which needs to stay a string.

In Python 2.0 the code might be shorter, using string methods and list
comprehension, but I personally prefer the "old way" myself.





From m_konermann at gmx.de  Wed Mar 28 18:32:02 2001
From: m_konermann at gmx.de (Marcus Konermann)
Date: Thu, 29 Mar 2001 01:32:02 +0200
Subject: i need help on builtin function math
Message-ID: <3AC27472.AC25F54B@gmx.de>

hello !

i want to use the sqrt() function of the builtin funtion math in that
kind:

import math
...
a=0
a=sqrt(whatever)

and than a NameError: sqrt occured and i don?t know why ? Is it
depending on the OS (i use Windows98 and the pywin
editor). Other functions, like pow(), from the math module are working.
Can anyone help me with my problem ?

thank?s a lot
greetings
Marcus




From ixokai-nf at myNOSPAM.org  Sat Mar 24 11:24:43 2001
From: ixokai-nf at myNOSPAM.org (Ixokai)
Date: Sat, 24 Mar 2001 08:24:43 -0800
Subject: How 2 get rid of console
References: <tborq16qctlm42@corp.supernews.com>
Message-ID: <3abccd2e_6@goliath.newsfeeds.com>

Rename the .py file to .pyw, so that it will be launched with 'pythonw'
instead of 'python'. The 'pythonw' program doesn't have a console window.

--Ix

[ To respond, replace NOSPAM with 'seraph' ]

Bill Walker wrote in message ...
>I'm using Tkinter to build a GUI application. How do I get rid of the
>distracting dos console under the Tk window?
>
>




-----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
-----==  Over 80,000 Newsgroups - 16 Different Servers! =-----


From n8gray at caltech.edu.is.my.e-mail.address  Mon Mar 26 05:15:37 2001
From: n8gray at caltech.edu.is.my.e-mail.address (Nathaniel Gray)
Date: Mon, 26 Mar 2001 02:15:37 -0800
Subject: RFC: PEP243: Module Repository Upload Mechanism
References: <mailman.985475916.3683.python-list@python.org>
Message-ID: <99n57o$dkq@gap.cco.caltech.edu>

<posted & mailed>

Sean Reifschneider wrote:
>     The following are valid os names:
> 
>         aix beos debian dos freebsd hpux mac macos mandrake netbsd
>         openbsd qnx redhat solaris suse windows yellowdog
> 
> [snip]
>
>     Version is the official version string specified by the vendor for
>     the particular release.  For example, "2000" and "nt" (Windows),
>     "9.04" (HP-UX), "7.0" (RedHat, Mandrake).
> 
>     The following are valid architectures:
> 
>         alpha hppa ix86 powerpc sparc ultrasparc

Can any or all of these be set to None, or some equivalent?  Python code is 
often cross-platform, after all.

-n8

-- 
_.~'^`~._.~'^`~._.~'^`~._.~'^`~._.~'^`~._
             Nathaniel Gray
   California Institute of Technology
     Computation and Neural Systems
     n8gray <at> caltech <dot> edu
_.~'^`~._.~'^`~._.~'^`~._.~'^`~._.~'^`~._



From tjg at exceptionalminds.com  Mon Mar  5 23:44:14 2001
From: tjg at exceptionalminds.com (Timothy Grant)
Date: Mon, 5 Mar 2001 20:44:14 -0800
Subject: Processing XML
Message-ID: <20010305204414.B4641@trufflehunter.avalongroup.net>

Hi all,

For all of you at P9, I hope you are having a grand time, I
wish I could be there!

Now on to the questions:

I am far from an XML expert. I can create a well formed XML
document and that's about it.

I am working on a significantly large project where I have to
store large quantities of configuration information. Right now
I store them in a file format that can be read by LParser. It
works beautifully well, and returns a dictionary of
dictionaries of values.

However, as the project has expanded, I find myself requiring
more and more information to be stored, not only does that make
my config files more difficult to read by hand, it also makes
them more difficult to write. Hence, my interest in XML.

I have Sean McGrath's book on XML Processing with Python, and
have played a bit with Pyxie. At the moment I can accomplish
more with Pyxie than I can with PyXML on its own. However, for
some reason I'm a bit uncomfortable with it for some reason.
Part of my discomfort comes from the fact that part of my
interest in XML is that it is so broadly useful. I am curious
as to whether I will come to a point where I find Pyxie to be
limiting the usefulness of my use of XML.

If Pyxie is a good choice, how do I add nodes to a document
stored as an xTree?

Also, as stated above, I currently move all of the
configuration data into a dictionary of dictionaries. This
works very well, However, I may be able to get equally good
access to the information I need by walking the XML tree
instead of putting things in a dictionary. Is my assumption
correct, or silly, or perhaps both. Does anyone have any better
suggestion?

Thanks much for your input.

-- 
Stand Fast,
    tjg.

Timothy Grant                         tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.         <><       (503) 246-3630
>>>>>>>>>>>>>Linux, because rebooting is *NOT* normal<<<<<<<<<
>>>>This machine was last rebooted:  48 days  8:41 hours ago<<



From stephen_purcell at yahoo.com  Thu Mar 22 12:53:48 2001
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Thu, 22 Mar 2001 18:53:48 +0100
Subject: Transparent GIFs in PIL
In-Reply-To: <20010322180110.A18476@freedom.puma-ag.com>; from stephen_purcell@yahoo.com on Thu, Mar 22, 2001 at 06:01:10PM +0100
References: <newscache$m8xlag$gk9$1@proxy.bitmailer.com> <20010322180110.A18476@freedom.puma-ag.com>
Message-ID: <20010322185348.A19190@freedom.puma-ag.com>

Steve Purcell wrote:
> Marcos S?nchez Provencio wrote:
> > Is it possible to set a transparent color in a GIF file (originally not
> > transparent) in PIL?
> 
> With PIL 1.0, not directly. For later versions (if they exist), I don't know.

I just checked; the latest version of PIL (1.1.1) would also need that patch
to support GIF transparency, as far as I can tell.

-Steve

-- 
Steve Purcell, Pythangelist
Get testing at http://pyunit.sourceforge.net/
Any opinions expressed herein are my own and not necessarily those of Yahoo



From rikbo716 at student.liu.se  Sun Mar 25 17:54:19 2001
From: rikbo716 at student.liu.se (Rikard Bosnjakovic)
Date: Mon, 26 Mar 2001 00:54:19 +0200
Subject: Is it possible to sort a dictionary?
References: <GEtv6.59583$Lm2.6378825@news0.telusplanet.net>
Message-ID: <3ABE771B.7B9E1631@student.liu.se>

Victor Louie wrote:

> I was just wondering if anyone knows how to sort a dictionary by
> value...since the dictionary is <key: value> and where my value is a list.

You can't sort the actual dictionary since they're mapping. You can,
however, put the values in a list and sort the list. Something like
this:

    for key in thedict.keys():
        sortedlist.append( (key, thedict[key]) )
    sortedlist.sort()


Test-run:

>>> foo = {"foo":[1,2,3], "bar":[2,3,4], "gazonk":[3,4,5]}
>>> foo
{'foo': [1, 2, 3], 'gazonk': [3, 4, 5], 'bar': [2, 3, 4]}
>>> sorted = []
>>> for key in foo.keys():
...   sorted.append( (foo[key], key) )
... 
>>> sorted.sort()
>>> sorted
[([1, 2, 3], 'foo'), ([2, 3, 4], 'bar'), ([3, 4, 5], 'gazonk')]



-- 
Rikard Bosnjakovic - http://a214.ryd.student.liu.se/cv/ - ICQ: 1158217

Anyone sending unwanted advertising e-mail to my address will be
charged $250 for network traffic and computing time. By extracting my
address from this message or its header, you agree to these terms.


From bh at abbenay.cs.berkeley.edu  Tue Mar 13 18:47:45 2001
From: bh at abbenay.cs.berkeley.edu (Brian Harvey)
Date: 13 Mar 2001 15:47:45 -0800
Subject: Einstein's Riddle
References: <oNPq6.396644$ge4.138479566@news2.rdc2.tx.home.com> <98glrd$1si1a$1@ID-75892.news.dfncis.de> <98h62k$knn$1@abbenay.CS.Berkeley.EDU> <98hmsv$216kg$1@ID-75892.news.dfncis.de>
Message-ID: <98mbj1$282$1@abbenay.CS.Berkeley.EDU>

"Boris Popov" <no-spam-popov at bigserver.com> writes:
> The thing is that there's no fifth pet element (fish) explicitly
>defined in the puzzle.

Of course the fish is explicit in the puzzle -- it's in the question
"who owns the fish?"

This is a well-known convention among logic-puzzle fans; it's perfectly
legitimate to treat this as if there were a clue saying "someone owns
a fish."


From dwmurphy at home.com  Mon Mar 12 11:08:32 2001
From: dwmurphy at home.com (David Murphy)
Date: Mon, 12 Mar 2001 16:08:32 GMT
Subject: Einstein's Riddle
References: <oNPq6.396644$ge4.138479566@news2.rdc2.tx.home.com> <UdSq6.2184$54.2166@www.newsranger.com> <984380185.768920@news.aei.ca> <uP%q6.2375$54.2207@www.newsranger.com>
Message-ID: <4w6r6.193645$df5.5192565@news1.crdva1.bc.home.com>

This reminds me of a test my class was given WAY back in grade 6.  It was
two sheets of paper long, and at the very top, it was clearly stated that
the student should 'Read the entire test carefully before beginning'.

The point was, at the bottom of the second page, the student was told
something to the effect of: 'After reading through the test, place your
pencil on you desk and wait for further instructions'.

Well, 90%+ of the class immediately went to work on this test.  There were a
couple of questions where students had to get up and do things, and it was
rather amusing to watch them run around.  About 30 minutes later, they
started to get to the end, and they realized that they had been running
around and scribbling down answers for nothing.  It was a rather profound
lesson, at least for me.

*IF* Einstein was just talking about people's ability to deal with just the
facts, and not to make assumptions, I think 2% +/- 5% is probably a good
estimate.   Take a look at the first draft of the next project specification
you get *grin*

dave

"Gregory Jorgensen" <gregj at pobox.com> wrote in message
news:uP%q6.2375$54.2207 at www.newsranger.com...
> Assuming the unknown pet is the fish, solving the puzzle doesn't take any
great
> skill with logic or math; certainly more than 2% of the population can
figure it
> out. I think the riddle demonstrates the fallacy of unwarranted
assumptions.
> It's a trick question; you work out the solution only to be told "I never
said
> anyone had fish." The more literal-minded the reader, the more likely they
will
> be taken in.
>
>
> In article <984380185.768920 at news.aei.ca>, sulfugor says...
> >But the fish is mentionned in the question before the clues are given. We
> >are able to find who owns each of the 4 pets mentionned in the hint list.
is
> >it so unreasonable to assume then that the last pet is a fish, since we
were
> >asked about it ? The question IMplied that one of the 5 people owned a
fish.
> >
> >sulfugor
> >
> >ps = by the way, how long is this supposed to take ( solving the riddle)
?
>
> Greg Jorgensen
> Deschooling Society
> Portland, Oregon, USA
> gregj at pobox.com




From jmarshal at mathworks.com  Mon Mar  5 17:47:54 2001
From: jmarshal at mathworks.com (Joshua Marshall)
Date: 5 Mar 2001 22:47:54 GMT
Subject: Static class methods
References: <rdzo6.154$gc4.80445@news.pacbell.net> <ta5ka1ks0buc3b@corp.supernews.com> <97ul9k$n2l$02$1@news.t-online.com> <slrn9a7fhc.c9f.qrczak@qrnik.zagroda>
Message-ID: <98152q$p15$1@news.mathworks.com>

Marcin 'Qrczak' Kowalczyk <qrczak at knm.org.pl> wrote:
> Mon, 5 Mar 2001 01:06:06 +0100, Sven Peters <svpeters at t-online.de> pisze:

>> You CAN have class methods:
>> def bert:
>>         print "Bert"
>> class.method = bert
>> class.method()
>> "Bert"

> You can't. The first line is a syntax error. And the following:

> class C: pass
> def bert(): print "Bert"
> C.method = bert
> C.method()

> produces an error (unbound method must be called with class instance
> 1st argument).

But you can be super-ugly about it:

  >>> class C: pass
  ... 
  >>> def bert(): print "bert"
  ... 
  >>> C.method = bert
  >>> C.__dict__['method']()
  bert

or

  >>> C.method.im_func()
  bert

Although I can't say I recommend either of these approaches.


From lluang at northwestern.edu  Mon Mar 12 10:12:49 2001
From: lluang at northwestern.edu (Louis Luangkesorn)
Date: Mon, 12 Mar 2001 09:12:49 -0600
Subject: looking for old python books
References: <eafoat4r8rs21v1h50239iqf9sktaetnjv@4ax.com>
Message-ID: <3AACE771.3BCDCF70@northwestern.edu>

Actually, there are several books online, and it really depends on what
level of programming skill you are at. Stephen Figgins has already
documented a number of documents on the internet which are freely
available for downloading.  It is on the O'Reilly web site and the blurb
says

Python Books Online
"Want to dive into Python or learn the basics of programming with
Python? Free and open online books make learning (or teaching) Python
easy and inexpensive."
http://www.oreillynet.com/pub/a/python/2001/02/07/pythonnews.html

One introductory tutorial that is not here, nor on the python.org site
is
http://www.honors.montana.edu/~jjc/easytut/easytut/
or http://www.honors.montana.edu/~jjc/easytut/easytut.pdf

Of course, there are also the pages on the python.org web site,
http://www.python.org/doc/Intros.html, but these seem to be geared for
people who already know how to program, and they just need to get that
first step.

Last, on the Open Directory Project page, a number of tutorials are
listed:
http://search.netscape.com/Computers/Programming/Languages/Python/Documentation_and_Tutorials

Louis

dr0 wrote:

> Hi all,
>  I was wondering if any of you have any old books on python you can
> sell to me? I'm a student and the prices for online stores's a bit
> high.
> Email me if you can help.
>
> Any help is deeply appreciated, thanks.
> Kenny

--
K Louis Luangkesorn
lluang at northwestern.edu  http://pubweb.nwu.edu/~kll560  PGP:0xF3E2D362
Whatsoever things are true, ... honest, ... just, ... pure, ... lovely,
... of good report; if there be any virtue, and if there be any praise,
think on these things.- motto - Northwestern University




From fredrik at pythonware.com  Fri Mar  2 02:58:01 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 02 Mar 2001 07:58:01 GMT
Subject: Py_INCREF / DECREF not needed in simple C functions?
References: <3a9e5301.1251797@news.newsguy.com> <97lp5g02uun@news2.newsguy.com> <3a9e8250.1005349@news.newsguy.com>
Message-ID: <doIn6.16284$Qb7.2659251@newsb.telia.net>

Owen F. Ransen wrote:
> Thanks for the tip, so I should do this:
>
> static PyObject* Py_ReallyWildFunc (PyObject *self, PyObject *args)
> {
>     double x,y,z ;
>     if (!PyArg_ParseTuple(args, "ddd",&x,&y,&z)) {
>         return (NULL) ;
>     }
>
>     ...do some really wild stuff with x,y,z...
>
>     Py_INCREF (Py_None) ;
>     return (Py_None) ;
> }

correct.

(at least if you need to call your function more than
sys.getrefcount(None) times ;-)

your function "borrows" argument references, but are expected
to return new references.  see:

    http://www.python.org/doc/current/ext/ownershipRules.html

Cheers /F

<!-- (the eff-bot guide to) the standard python library:
http://www.pythonware.com/people/fredrik/librarybook.htm
-->




From tjreedy at udel.edu  Fri Mar  2 22:15:50 2001
From: tjreedy at udel.edu (Terry Reedy)
Date: Fri, 2 Mar 2001 22:15:50 -0500
Subject: shortcut for 'self' wish
References: <3aa050ad.35416419@News.CIS.DFN.DE>
Message-ID: <97pnpf$99$1@news.udel.edu>

"Costas Menico" <costas at meezon.com> wrote in message
news:3aa050ad.35416419 at News.CIS.DFN.DE...
> It would be nice if there was someway to cut down on the typing of
> self.something inside def methods.

This is entirely a social convention for publicly displayed code.  For my
own work, I use s for self (and o for other in __add__(), etc.).  Python is
not a straightjacket.




From gilh at sympatico.ca  Thu Mar 22 19:07:20 2001
From: gilh at sympatico.ca (Gil Hauer)
Date: Thu, 22 Mar 2001 19:07:20 -0500
Subject: LONG_BIT problem
Message-ID: <3ABA93B8.8088F22D@sympatico.ca>

Hi,

I've having a problem building Python 2.0 on RedHat 7.0. 

gcc version is:
	Reading specs from /local/lib/gcc-lib/i686-pc-linux-gnu/2.95.3/specs
	gcc version 2.95.3 20010315 (release)

glibc versions are:
	glibc-2.2-12
	glibc-devel-2.1.92-14
	glibc-common-2.2-12

The error message I'm getting is:

gcc -g -O2 -Wall -Wstrict-prototypes -I./../Include -I..
-DHAVE_CONFIG_H   -c -o node.o node.c
cpp0: In file included from ../Include/Python.h:54,
                 from node.c:3:
../Include/pyport.h:390:74: #error "LONG_BIT definition appears wrong
for platform (bad gcc config?)."

Can anyone help with this?

Thanks in advance,
	Gil



From olczyk at interaccess.com  Mon Mar  5 08:10:22 2001
From: olczyk at interaccess.com (Thaddeus L. Olczyk)
Date: Mon, 05 Mar 2001 13:10:22 GMT
Subject: Python 3000.  Here we go again.
References: <Pine.GSO.4.21.0101142011080.14128-100000@y.glue.umd.edu> <mailman.979546405.2262.python-list@python.org>
Message-ID: <3aa5902f.398605718@nntp.interaccess.com>

On Mon, 15 Jan 2001 18:18:02 +0200 (IST), Moshe Zadka
<moshez at zadka.site.co.il> wrote:

>
>Python 2.1 is likely to have lexical scoping.

What is meant by lexical scoping?


From apardon at trout.vub.ac.be  Mon Mar 26 07:18:39 2001
From: apardon at trout.vub.ac.be (apardon at trout.vub.ac.be)
Date: 26 Mar 2001 12:18:39 GMT
Subject: Is this an inconsistency?
Message-ID: <slrn9bud88.2le.apardon@trout.vub.ac.be>

I'm a newbie what Python is concerned so my question may be
based on misconceptions. Also I'm still using 1.5.2

To get some experience with the language and for further
use in some toy-project I am currently working on a
Vector and Matrix class in Python.

A Matrix has just one attribute which is a list of
lists. The sublists each standing for a  row. Now
I had read in the tutorial that an easy way to
change to lists into a list of pairs was:

  map(None , list1 , list2)

And since you can turn a tuple into a list with
the function list, my way of transposing a matrix
was implemented as:

  map(list , apply(map , [None] + self.Value))


The trouble is that doesn't work if I have
a single row matrix. The trouble seems to
be that although

  map(None , list1 , list2 , list3 , ... )

turns a number of lists into a list of tuples.


  map(None , list)

Doesn't turn into a list of singletons but
just returns the list.

This seems a bit odd to me because if you
use apply with a single parameter function
the second argument has to be either a singleton
or a list with one element.


My question now: Is this behaviour changed in
the following versions? If not, wouldn't it be
more consistent if it was?

-- 
Antoon Pardon


From grante at visi.com  Sat Mar 24 18:40:00 2001
From: grante at visi.com (Grant Edwards)
Date: Sat, 24 Mar 2001 23:40:00 GMT
Subject: static methods
References: <mailman.985383828.2942.python-list@python.org> <slrn9bpqgo.15q.grante@tuxtop.visi.com> <mailman.985457492.29908.python-list@python.org> <7h1qbt8h8aos1parc090s8tqu2ooc04t4d@4ax.com>
Message-ID: <slrn9bqc2r.7nq.grante@isis.visi.com>

In article <7h1qbt8h8aos1parc090s8tqu2ooc04t4d at 4ax.com>, Courageous wrote:

>>That it's called "static" is an artifact of C, I believe. In C, "static" 
>>has been overexploited such that it has many different meanings in 
>>different contexts that have nothing to do with being "static". Hideous 
>>example: declaring a function as "static" in C means that it's invisible to 
>>code outside of the source file and to all code following it in the same 
>>file. :-0
>
>Hrm. I'd forgotten that. In languages where there are no namespaces,
>it's useful to be able to make certain that symbols don't go out and
>pollute the global namespace in which ALL symbols reside. This is
>what that's for.

I'm confused -- are you saying that C doesn't have namespaces?

>More generally, "static" in most cases in C and C++ means "global
>for the scope that I'm in", err, well, sortof.

:)

In C, the keyword "static" is used for two separate, orthogonal purposes:

 1) to define the lexical scope of a variable.

 2) to define the storage lifetime of a variable. 

There is no excuse for using the same keyword to mean two unrelated things.

-- 
Grant Edwards                   grante             Yow!  Used staples are good
                                  at               with SOY SAUCE!
                               visi.com            


From spahievi at vega.bg  Fri Mar  9 15:54:27 2001
From: spahievi at vega.bg (Niki Spahiev)
Date: Fri, 9 Mar 2001 22:54:27 +0200
Subject: segfault with PY file but ok with PYC
Message-ID: <173478215.20010309225427@vega.bg>

Hello,

I have very strange problem with python2 (ActiveState and PY20) under Windows (NT4)

If i run "python v1a.pyc" it works but
"python v1a.py" gives me segfault.

I recompiled PYC file to be current.
Sometimes problem goes away if i rearrange method def's, but if i return
them back problem reappears!
Unfortunately i can't reproduce it with small set of code.

script
from v1a import main; main()
crashes first time (and produces v1a.pyc)
second time runs ok (probably from v1a.pyc).

Sometimes PY20 distro does not segfault but NOT always.

Any suggestions?

-- 
Best regards,
 Niki Spahiev





From kalle at gnupung.net  Wed Mar 21 12:02:07 2001
From: kalle at gnupung.net (Kalle Svensson)
Date: Wed, 21 Mar 2001 18:02:07 +0100
Subject: os.system output
In-Reply-To: <B7C74E1A6FBAD411BFC60002B309DDF4E42F@mailbox.juilliard.edu>; from jsoons@juilliard.edu on Wed, Mar 21, 2001 at 11:32:39AM -0500
References: <B7C74E1A6FBAD411BFC60002B309DDF4E42F@mailbox.juilliard.edu>
Message-ID: <20010321180207.A24260@father>

Sez Jonathan Soons:
> I am trying to save the output of
> 
> >>>dirs = os.system('ls')
> 
> into a variable but I do not need any output to
> the screen (altho it doesn't hurt). How do I
> go about this?

Check out os.popen().

>>> dirs = os.popen("ls").read()

Peace,
  Kalle
-- 
Email: kalle at gnupung.net     | You can tune a filesystem, but you
Web: http://www.gnupung.net/ | can't tune a fish. -- man tunefs(8)
PGP fingerprint: 0C56 B171 8159 327F 1824 F5DE 74D7 80D7 BF3B B1DD
 [ Not signed due to lossage.  Blame Microsoft Outlook Express. ]



From garry at sage.att.com  Wed Mar 14 09:20:55 2001
From: garry at sage.att.com (Garry Hodgson)
Date: Wed, 14 Mar 2001 14:20:55 GMT
Subject: I come to praise .join, not to bury it...
References: <mailman.983485923.1136.python-list@python.org> <97p7iq$kk2$1@nntp6.u.washington.edu> <97rka002k9@news1.newsguy.com>
Message-ID: <3AAF7E47.294617E1@sage.att.com>

Alex Martelli wrote:

> I have not heard ANY _technical_ arguments opposed to
> this last time the discussion went around -- nothing but
> vague aesthetic, "it should be the other way", "I find it
> ugly" kind of complaints.  Unless and until further notice,
> then, I class these together with the complaints of people
> who dislike 0-based indexing, on similar vague bases -- as
> 0-based indexing _works_ better than 1-based, then, for
> me, it is a superior choice.  Aesthetics is in the eye of the
> beholder, and the beholder _can_ be retrained and grow
> a 'technically better and more appropriate' view of the
> world -- one that will let him or her 'see' as 'beautiful'
> those approaches which offer maximal effectiveness and
> simplicity.

don't be too quick to write off Aesthetics.  it's a significant
part of what attracted me to python (i'd seen enough curly braces
and semicolons to last me a lifetime).  avoiding scaring off newbies
until they've had time to appreciate other virtues is not a Bad Thing.

-- 
Garry Hodgson                   Once in a while
Senior Hacker                   you can get shown the light
Software Innovation Services    in the strangest of places
AT&T Labs                       if you look at it right


From dnew at san.rr.com  Sun Mar 25 21:53:53 2001
From: dnew at san.rr.com (Darren New)
Date: Mon, 26 Mar 2001 02:53:53 GMT
Subject: Comparing efficiency in sending mail
References: <3A7FA5BA.82FB7593@esec.com.au>
	 <mailman.981444758.4937.python-list@python.org> <mailman.985543113.29120.python-list@python.org>
Message-ID: <3ABEAF43.E44666AD@san.rr.com>

Robin Thomas wrote:
> If you use smtplib, your program is the MTA. Your program must do the
> actual message transfer by itself: checking/munging the message to be sent,
> blocking on connect to destination MTA, handling errors, retrying if
> desired, bouncing message to a mailbox, etc.

Well, all that's usually not a problem if you SMTP to localhost rather than
doing all the MX lookups yourself and such. You're still talking to
sendmail. You're just doing it in a standards-compliant way. Then the local
sendmail handles passing it on from there.

I.e., you're using sendmail as an MUA, not an MTA, just like (say) MH uses
it.

Of course, if the sendmail at localhost refuses the message, then it's probably
a local address and you probably want to know that right away rather than
getting bounce mail back somewhere in a mailbox.

-- 
Darren New / Senior MTS & Free Radical / Invisible Worlds Inc.
San Diego, CA, USA (PST).  Cryptokeys on demand.
A million monkeys in a room with a million typewriters 
              will only yield half a million pregnant monkeys.


From donn at u.washington.edu  Fri Mar  2 13:47:39 2001
From: donn at u.washington.edu (Donn Cave)
Date: 2 Mar 2001 18:47:39 GMT
Subject: gah! I hate the new string syntax
References: <mailman.983485923.1136.python-list@python.org> <B7In6.16281$Qb7.2658902@newsb.telia.net> <97ohh2$polt0$1@ID-77331.news.dfncis.de>
Message-ID: <97opsb$nhu$1@nntp6.u.washington.edu>

Quoth "Mark Pilgrim" <f8dy at yahoo.com>:
| "Fredrik Lundh" <fredrik at pythonware.com> wrote in message
| news:B7In6.16281$Qb7.2658902 at newsb.telia.net...
|> Sean 'Shaleh' Perry wrote:
|> > return ";".join(["%s=%s" % (k, params[k]) for k in params.keys()])
|> >
|> > every day python seems to be moving closer to the line noise aspect of
|> > coding.
|>
|> you're using list comprehensions and the wrong dictionary method,
|> and then you're complaining about string methods? ;-)
|>
|> ("items()" returns a key/value tuple that you can pass directly to
|> the formatting operator.  and nobody will complain if you stick to
|> string.join for stuff like this...)
|
| I recognize this code as an example from my book "Dive Into Python",
| specifically http://diveintopython.org/odbchelper_join.html, and would like
| to (partially) defend it in that context.
|
| 1. I used Python 1.5.x very little; I really only ramped up when 1.6/2.0 was
| released.  I *like* list comprehensions.  I prefer them over map/lambda
| because of scope issues, and I prefer them over for loops because they force
| me to think Pythonically, and they emphasize Python's strengths (less code,
| richer data structures).  To my eyes, list comprehensions fulfill Python's
| promise of being readable pseudo-code, not readable line noise.  YMMV.

Pseudo-code is the language you don't have to know, to understand,
right?  Familiar symbols, unambiguous syntax.  Could you articulate
your thoughts on this, how do list comprehensions improve on their
alternatives, there?

What does "think Pythonically" mean to you?  What are you thinking of
when list comprehensions make you cite "less code, richer data structures"
as a strength of Python?  Can you talk us crotchety old-timers through this?

| 3. I *like* string methods, even .join().  I am aware that this puts me in a
| very small minority on comp.lang.python.  But I would maintain that
| ";".join(...) only seems odd to you because you see it as the result of a
| strained migration from string.join(..., ";"), instead of seeing it with
| fresh eyes.  However, I may add a footnote explaining the old style
| string.join, if only because new programmers will eventually come across old
| code that uses it.
|
| If current trends continue, *far more people* will learn Python 2.0 than
| ever learned Python 1.x.  They will all take list comprehensions and string
| methods for granted, and stare at you blankly when you reminisce about the
| good old days of Python 1.5.2.

I bet when people see ";".join(...) with fresh eyes, they still
find it strained.  This is not about the old school vs. the new,
it's a serious question of how people understand computer programming
languages.  Is it like natural language?  Can we think about it in
sort of mathematical terms, like symmetry etc.?  Does everyone handle
it basically the same way?

For sure, more people will learn 2.x.  But they're here because of
the success of previous versions, not because 2.x was finally good
enough for them to get on board.  It doesn't tell us anything about
the value of features introduced in 2.x.

	Donn Cave, donn at u.washington.edu


From pet-blo at online.no  Fri Mar  2 08:16:58 2001
From: pet-blo at online.no (Peter Blom)
Date: Fri, 2 Mar 2001 14:16:58 +0100
Subject: Communicating with serial port
Message-ID: <ndNn6.5161$t21.152593@news3.oke.nextra.no>

I shall write a python program for downloading data from a memory card in a
measuring instrument, via a serial port.
For this purpose, I have downloaded SioModule.zip - a serial comm extension
for windows 95, upgraded to Python 1.5.1.
This does not seem to work - when importing the module 'serial.py', my
system hangs (Windows 98, Activestate Python 2.0).
Could anybody give a hint?

Peter (a newbie)




From ken at lt.com  Mon Mar  5 16:49:21 2001
From: ken at lt.com (Kenneth Loafman)
Date: Mon, 05 Mar 2001 21:49:21 +0000
Subject: Revenge - Careful there
References: <97fbbm$v1o$1@snowmane.julianhaight.com> <97goaj$jj4$1@snowmane.julianhaight.com> <3A9BFE1D.B94F6AA5@lt.com> <97j7kn$p21$1@snowmane.julianhaight.com> <3AA407BA.AA9C49A7@lt.com>
Message-ID: <3AA409E1.69D0DF22@lt.com>

Please ignore the above set of messages.  Netscape and I were arguing
and it looks like Netscape won this round.

Sorry for the inconvenience.

...Ken



From sholden at holdenweb.com  Thu Mar 15 18:35:00 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 15 Mar 2001 18:35:00 -0500
Subject: This math scares me
References: <984486395.407379@newsmaster-04.atnet.at> <slrn9asf1d.ja3.qrczak@qrnik.zagroda> <984518199.127140@newsmaster-04.atnet.at> <jwbnews-09D75F.13313313032001@corp.supernews.com> <98ohuc$du9$1@news.udel.edu> <7bQr6.644$lz1.17686@ruti.visi.com> <98oku1$f6a$1@news.udel.edu> <erRr6.659$lz1.20662@ruti.visi.com> <mailman.984664520.26463.python-list@python.org> <HR4s6.877$lz1.73512@ruti.visi.com> <mailman.984689603.19242.python-list@python.org>
Message-ID: <locs6.79734$1D5.4121865@e420r-atl1.usenetserver.com>

"Fabio Olive Leite" <olive at conectiva.com.br> wrote in message
news:mailman.984689603.19242.python-list at python.org...
> On Thu, Mar 15, 2001 at 03:04:07PM +0000, Grant Edwards wrote:
> )
> ) One step further is base-60, which is still used for many
> ) things.  Time and angular measurement for example is partially
> ) a base-60 system.  I believe that one or more of the ancient
> ) Aztecs/Incans/Mayans used a base-60 number system.
>
> Indeed, they used it. And some other people too, as is seen even today in
> the most common languages of the word. French, for example, has no "real
> words" for seventy, eighty and ninety, and uses "sixty ten", "four twenty"
> and "four twenty ten".
>
> Base 12 was used a lot for commerce, as it has more integer fractions than
> 10. :)
>
What can you expect from a race who insist on calling seventeen, eighteen
and nineteen "ten-seven", "ten-eight" and "ten-nine" respectively? And let's
not get into punds, shillings and pence.

old-enough-to-remember-them-ly y'rs  - steve





From PaulS at scenicsoft.com  Mon Mar 19 16:19:29 2001
From: PaulS at scenicsoft.com (Paul Schwartz)
Date: Mon, 19 Mar 2001 13:19:29 -0800
Subject: What IDE are you'all using for Python 2.1 for Windows?
References: <995gai$2bv$1@news.chatlink.com>
Message-ID: <3849BAFD79A7D411A1CF00508B62B5BC2000CB@exchange-us.scenicsoft.com>

It costs some money (~$300 ) but it's a great editor www.slickedit.com

Paul

> -----Original Message-----
> From: newservice at 1cust214.tnt2.el-paso.tx.da.uu.net
> [mailto:newservice at 1cust214.tnt2.el-paso.tx.da.uu.net]On 
> Behalf Of Luis
> Cortes
> Posted At: Monday, March 19, 2001 9:28 AM
> Posted To: comp.lang.python
> Conversation: What IDE are you'all using for Python 2.1 for Windows?
> Subject: What IDE are you'all using for Python 2.1 for Windows?
> 
> 
> Hi,
> 
>   Just wanted to know what "good" ide's are out there for 
> python.  Please
> let me know which ones you prefer.
> 
> Thanks,
> Luis.
> 
> 
> 



From stepheb at comm.mot.com  Thu Mar  8 11:58:22 2001
From: stepheb at comm.mot.com (Stephen Boulet)
Date: Thu, 08 Mar 2001 10:58:22 -0600
Subject: Extracting cell contents from html tables
References: <3AA67658.B39580CB@comm.mot.com> <mailman.984010102.21666.python-list@python.org>
Message-ID: <3AA7BA2D.E88B8C14@comm.mot.com>

Thanks! Exactly what I needed.

-- Stephen

David Porter wrote:

> * Stephen Boulet <stepheb at comm.mot.com>:
>
> > Is there a module for extracting cell contents from html tables
> > (preferably into an array of strings)?
>
> This might be useful:
> http://starship.python.net/crew/jjkunce/python/htmlTableParse.py
>
> David



From claird at starbase.neosoft.com  Mon Mar 26 11:02:59 2001
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 26 Mar 2001 10:02:59 -0600
Subject: time to invest ???
References: <mailman.985510476.15482.python-list@python.org>
Message-ID: <5F40A50758C8DB5C.651870BC82F8A551.BC92991507480A16@lp.airnews.net>

In article <mailman.985510476.15482.python-list at python.org>,
 <py.list at localhost.localdomain> wrote:
			.
			.
			.
>For example, I'm looking at Python now to do non-web jobs that PHP isn't
>so great for, and which shell scripts can't quite handle. Perl is the
			.
			.
			.
>Mick

Mick, please e-mail me directly.  I'd like to chat privately a bit.
-- 

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


From mcfletch at home.com  Tue Mar  6 20:31:51 2001
From: mcfletch at home.com (Mike C. Fletcher)
Date: Tue, 6 Mar 2001 20:31:51 -0500
Subject: I come to praise .join, not to bury it...
In-Reply-To: <trfp6.29423$1D5.1257189@e420r-atl1.usenetserver.com>
Message-ID: <000401c0a6a6$637dbaa0$a9807018@cr706570a>

Aesthetes are a different kind of professional, I'm a design theorist, those
are art wh#$3s ;) .  Sorry, I dislike the connotations of "aesthete" (too
turn of the last century eclecticism for me), so I use aesthetic (which, I
feel, has a much nicer resonance with "ascetic", which is closer to my sense
of beauty).

I have no problem with the semantics of the string having methods.  I asked
for lists and tuples to have a consistent and complete set of methods 2 or 3
years ago if I'm recalling correctly.  Healing the class/type split is
something I would like to live to see (if the PSU doesn't get me first).

I'll have to give that Python thing a try.  Can you recommend any good books
for someone who's been stuck in a backwater language for the last 6 years?
Isn't that Python thing slow though?  I heard they don't even have static
type declarations and they don't support polymorphic function signature
dispatches.

"," . join( ) doesn't help much, as there's no graphical connection drawn
between the two items, so you lose the correspondence of the . character to
"member of" (i.e. the glyph no longer represents the concept, but must be
interpreted at a semantic level to make the construct clear).

As I mentioned, I don't expect this to change, just wanted to point out that
there is a legitimate complaint from an aesthetic standpoint.

Enjoy,
Mike


-----Original Message-----
From: Steve Holden [mailto:sholden at holdenweb.com]
Sent: Tuesday, March 06, 2001 7:36 PM
To: python-list at python.org
Subject: Re: I come to praise .join, not to bury it...
...
A professional would call themself an aesthete :-)
...
Just the same, a beginner can read the code as "some string's method is
applied to whatever argument is passed" just by knowing the syntax.
...
Well, if you like white space Python should be the perfect language for you!
...
What about the currently syntactically-acceptable

>>> ", " . join(["big", "fat", "joint"])
'big, fat, joint'
>>>

Remember that technically the dot is an operator. Personally I find the
extra whitespace intrusive, but that is a matter of style and taste. Any
better for you?
...
Not a wart, but a consistent application of object-oriented principles.
Sorry. I too wish the world looked more beautiful (to you).
...




From ehagemann at home.com  Sun Mar 18 21:04:48 2001
From: ehagemann at home.com (Eric Hagemann)
Date: Mon, 19 Mar 2001 02:04:48 GMT
Subject: Subsetting a dictionary
References: <993dkm$ttr$1@news6.svr.pol.co.uk>
Message-ID: <4Pdt6.189183$B6.41882888@news1.rdc1.md.home.com>

Rather than copy the entire dictionary, why not create a new empty one and
only add the items that match you contraint.  Seems like making the copy of
all items and then throwing part of them away is extra work.  Am I missing
something ?

Cheers

"Will Newton" <will at nospam.misconception.org.uk> wrote in message
news:993dkm$ttr$1 at news6.svr.pol.co.uk...
>
> I need to get a subset of a dictionary:
>
> D = Dictionary of objects (in this case, training examples for a decision
> tree)
>
> D2 = {D | d in D, d under some constraint}
>
> I need to be able to do this, but as it is temporary I do not want to
> destructively remove examples, so:
>
> D.subset(c)
> will return a new dictionary object that is equivalent to D under
> constraint c.
>
> I have code that does this, but it's very slow, consuming typically 75% of
> program runtime - it is called many times, which seems to be unavoidable
> without modifying the algorithm (ID3) too far.
>
> I currently have:
>
> class TrainingData:
>          data = {}
>         def subset(self, attr, value):
>                 # Copy semantics
>
>                 td = TrainingData()
>
>                 td.data = self.data.copy()
>
>                 for item in td.data.keys():
>                          if item[attr] != value:
>                                    del td.data[item]
>
>                 return td
>
> Can anyone see how I can speed this up?
>
> I could conceivably do this with a list and use filter, I'll see how that
> works.




From qrczak at knm.org.pl  Mon Mar 19 02:31:56 2001
From: qrczak at knm.org.pl (Marcin 'Qrczak' Kowalczyk)
Date: 19 Mar 2001 07:31:56 GMT
Subject: Killing threads
References: <3ab23ea7.328575630@News.CIS.DFN.DE> <mailman.984767303.16093.python-list@python.org>
Message-ID: <slrn9bbdfb.vh8.qrczak@qrnik.zagroda>

Fri, 16 Mar 2001 19:27:39 +0100, Steve Purcell <stephen_purcell at yahoo.com> pisze:

> Thread cancellation is generally considered a bad thing. If I'm
> not mistaken, the argument against allowing thread cancellation is
> concerned with correctly releasing locks shared between threads.

In Glasgow Haskell a thread can block and unblock asynchronous
exceptions, which includes killing it. So you get the ability to kill
arbitrary computation while important sections can be protected.

The interface of block and unblock is that these functions take a
computation as an argument, so they restore the blocking state after a
computation finished. In Python they would have to be plain imperative
statements instead, to be used thus (ignoring the restoration):
    try:
        thread.block()
        ...
    finally:
        thread.unblock()

Actually killing a thread is in fact throwing an exception to it.
except: and finally: clauses are executed up to the point when the
thread was created. In this case maybe just try:finally: could be
used to release locks without having explicit block and unblock.

In Glasgow Haskell you automatically get a lock around the equivalent
of finally: clause (it's a function which uses block and unblock
internally).

Disclaimer: I don't have much experience in programming with threads.

-- 
 __("<  Marcin Kowalczyk * qrczak at knm.org.pl http://qrczak.ids.net.pl/
 \__/
  ^^                      SYGNATURA ZAST?PCZA
QRCZAK


From neelk at alum.mit.edu  Tue Mar 13 19:31:43 2001
From: neelk at alum.mit.edu (Neelakantan Krishnaswami)
Date: 14 Mar 2001 00:31:43 GMT
Subject: This math scares me
References: <3aad4062.1290021@News.CIS.DFN.DE> <Hwbr6.1139$Tg.127571@ruti.visi.com> <3aad4485.2348499@News.CIS.DFN.DE> <jwbnews-7CC03F.19050312032001@corp.supernews.com> <yu993dcimjt8.fsf@europa.research.att.com> <slrn9asbvc.dfq.neelk@alum.mit.edu> <98lgkc02dso@news1.newsguy.com>
Message-ID: <slrn9atiu8.efe.neelk@alum.mit.edu>

On Tue, 13 Mar 2001 17:07:38 +0100, Alex Martelli <aleaxit at yahoo.com> wrote:
>"Neelakantan Krishnaswami" <neelk at alum.mit.edu> wrote in message
>news:slrn9asbvc.dfq.neelk at alum.mit.edu...
>    [snip]
>> Point Numbers Accurately_ and Byrger and Dybvig's _Printing
>> Floating Point Numbers Quickly and Accurately_ and submit it as a
>> patch I don't think anyone will object. :)
>
> It's Burger, not Byrger (just in case someone should be searching or
> quoting this -- the analogy with Dybvig is hard to resist!-),

That's cleverer than the real reason: it's just that the 'y' and 'u'
keys are next to each other on my keyboard. :)

> and I think it's in the proceedings of the 1996 SIGPLAN conference.

Both papers are available from the ACM Digital Library, too. (Subscribing
to it was one of the best investments I ever made.)


Neel


From tjg at exceptionalminds.com  Sat Mar 24 17:04:25 2001
From: tjg at exceptionalminds.com (Timothy Grant)
Date: Sat, 24 Mar 2001 14:04:25 -0800
Subject: Daemonizing a python programme.
In-Reply-To: <3ABBE09E.3287DEC5@qvlinc.com>; from ckrill@qvlinc.com on Fri, Mar 23, 2001 at 03:47:42PM -0800
References: <mailman.985376338.19672.python-list@python.org> <3ABBE09E.3287DEC5@qvlinc.com>
Message-ID: <20010324140425.A19121@trufflehunter.avalongroup.net>

Thanks a bunch Coy!

I'll take a close look at it today.

On Fri, Mar 23, 2001 at 03:47:42PM -0800, Coy Krill wrote:
> Timothy Grant wrote:
> 
> > Hi all,
> >
> > I have been investigating daemonizing a python programme, and
> > have discovered to bits of code that assist in this endeavour:
> > daemonize.py and Daemon.py.

-- 
Stand Fast,
    tjg.

Timothy Grant                         tjg at exceptionalminds.com
Chief Technology Officer              www.exceptionalminds.com
HyperLINq Technologies, Inc.          <><       (503) 246-3630
>>>>>>>>>>>>>Linux, because rebooting is *NOT* normal<<<<<<<<<
>>>>This machine was last rebooted:  67 days  2:19 hours ago<<



From jkraska1 at san.rr.com  Sat Mar 24 19:52:06 2001
From: jkraska1 at san.rr.com (Courageous)
Date: Sun, 25 Mar 2001 00:52:06 GMT
Subject: static methods
References: <mailman.985383828.2942.python-list@python.org> <slrn9bpqgo.15q.grante@tuxtop.visi.com> <mailman.985457492.29908.python-list@python.org> <7h1qbt8h8aos1parc090s8tqu2ooc04t4d@4ax.com> <slrn9bqc2r.7nq.grante@isis.visi.com>
Message-ID: <4lgqbto46c4c6i307lnq0e614uujd2fsss@4ax.com>

>>Hrm. I'd forgotten that. In languages where there are no namespaces,
>>it's useful to be able to make certain that symbols don't go out and
>>pollute the global namespace in which ALL symbols reside. This is
>>what that's for.
>
>I'm confused -- are you saying that C doesn't have namespaces?

Erm, well, unless C99 defined them, no.


C//



From salnikov at inp.nsk.su  Wed Mar 21 04:03:20 2001
From: salnikov at inp.nsk.su (Andy Salnikov)
Date: Wed, 21 Mar 2001 15:03:20 +0600
Subject: Base 2 to long integers and back
References: <mailman.985145066.15092.python-list@python.org>
Message-ID: <999qoh$4ocsp$1@sky.inp.nsk.su>

For string->int or long the standard library does everything (in Python 2+):

Python 2.0 (#8, Oct 16 2000, 17:27:58) [MSC 32 bit (Intel)] on win32
Type "copyright", "credits" or "license" for more information.
>>> s = '01010101010111100010'
>>> int(s,2)
349666
>>> long(s,2)
349666L
>>>

Andy.



"Tim CHURCHES" <TCHUR at doh.health.nsw.gov.au> wrote in message
news:mailman.985145066.15092.python-list at python.org...
> Can anyone suggest a fast method of converting base2 strings
> (e.g. '0000101110') to Python long integers, and back?
> I couldn't locate any functions in the standard library to do this,
> but no doubt I am overlooking something obvious.
>
> Cheers,
>
> Tim Churches
> Sydney, Australia






From michel at digicool.com  Fri Mar 23 02:02:35 2001
From: michel at digicool.com (Michel Pelletier)
Date: Thu, 22 Mar 2001 23:02:35 -0800
Subject: PEP 245: Python interfaces
Message-ID: <Pine.LNX.4.32.0103222253510.26302-100000@localhost.localdomain>

Hi!

I'm the author of PEP 245, a framework and syntax for interface objects in
Python.

http://python.sourceforge.net/peps/pep-0245.html

It has just recently been made a PEP, and I'd like to solicite some more
input from the community (you).

For those of you who don't know the formal defintion of "interface", don't
worry (there isn't one!).  There are, however, a lot of agreed and
disagreed upon definitions out there in the world, and PEP 245 is one
particular take on the situation, based on pioneering work done by the
Type-sig and Jim Fulton over the last couple of years.

There is a Wiki (referenced in the PEP) for formal discussion and, of
course, this newsgroup, which I will hanging out in for the next few days
answering any questions about the proposal.

Thanks!

-Michel



From fig at monitor.net  Fri Mar 23 09:48:18 2001
From: fig at monitor.net (Stephen R. Figgins)
Date: Fri, 23 Mar 2001 06:48:18 -0800
Subject: Puzzling input problem
References: <3AB9449C.A1D14738@monitor.net> <d5#fcH1sAHA.187@typhoon.i3s.net>
Message-ID: <3ABB6232.15DAFE3C@monitor.net>

Kevin Cazabon wrote:
> 
> How about using this instead:
> 
> answer = raw_input("prompt string: ")

Yeah, that is how I would normally do it.  In this case, there were some
variables in the prompt string so I decided to it.

My question was not, "how do I use raw_input()?" It was, "why does the
space appear where it does?"

Alex supplied the answer.  When I followed the first print statement
with a comma, it set a softspace property on standard out. raw_input()
did not use or change this property when it modified the screen.  The
second print did, however, so it printed an extra space before the
second print.

-Stephen


From aahz at panix.com  Sun Mar 25 12:36:28 2001
From: aahz at panix.com (Aahz Maruch)
Date: 25 Mar 2001 09:36:28 -0800
Subject: Coding choices
References: <2basbt4oe4tkf6hasrlmtmnr5jvbtoq2mm@4ax.com>
Message-ID: <99laas$o3d$1@panix6.panix.com>

In article <2basbt4oe4tkf6hasrlmtmnr5jvbtoq2mm at 4ax.com>,
Daniel Klein  <danielk at aracnet.com> wrote:
>
>I've seen Python coded in both of these manners. Is there some
>advantage of one over the other? Or is it just preference?
>
>if condition:
>    return something
>raise SomeException
>
>if condition:
>    return something
>else:
>    raise SomeException

The latter is a tiny bit clearer, the former is a tiny bit more
efficient.  Mostly a wash, with personal preference playing the key
factor.
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

"I won't accept a model of the universe in which free will, omniscient
gods, and atheism are simultaneously true."  -- M


From cce at clarkevans.com  Wed Mar 21 22:04:42 2001
From: cce at clarkevans.com (Clark C. Evans)
Date: Wed, 21 Mar 2001 22:04:42 -0500
Subject: Yet Another PEP:  Interface Adapter Mechanism __adapt__
In-Reply-To: <_R4u6.466172$ge4.162590641@news2.rdc2.tx.home.com>
References: <Pine.LNX.4.21.0103201837570.21343-100000@clarkevans.com> <Pine.LNX.4.21.0103210418490.21377-100000@clarkevans.com> <_R4u6.466172$ge4.162590641@news2.rdc2.tx.home.com>
Message-ID: <Pine.LNX.4.21.0103212158580.11584-100000@clarkevans.com>

On Wed, 21 Mar 2001, Rainer Deyke wrote:
> This 'check_base' stuff is a mess. 

No joke.  Sorry for being so brain dead.  The following
code also has one or two errors, but should much better
illustrate the concept...

...

Here is the "merged" version where check_only is an
optional argument to adapt.  Imagine the operator,

  a isa b -> adapt(a,b,check_only=true)

--------------------------------------------------------
adapter/__init__.py
--------------------------------------------------------

import types
def adapt(obj,ident,check_only=0):

    # check to see if the current object is ok as is

    if obj is ident: return obj
    if type(obj) is ident: return obj
    if type(obj) is type(ident): return obj
    if type(ident) is types.ClassType:
        if type(obj) is types.InstanceType:
            if isinstance(obj,ident): return obj
        if type(obj) is types.ClassType:
            if issubclass(obj,ident): return obj

       # if we get a new interface type, code should
       # be added in this spot to check if obj is an 
       # instance of ident, where ident is an types.Interface
       # *THIS IS ANOTHER PEP*

    # the obj may have the answer, so ask it about the ident
    

    adapt = getattr(obj, '__adapt__',None)
    if adapt:
        retval = adapt(ident,check_only)
        if retval: return retval

    # the ident may have the answer, so ask it about the obj

       # some code here for a class, type, or interface,
       # or whatever to check to see if it can adapt the
       # class.  Unfortunately, calling __adapt__ won't
       # work exactly here...  we could add another
       # argument, "forward"... *THIS IS ANOTHER PEP*
 
    # a registry may exist, ask it about adapters.

       # there may be a registry-based  solution 
       # adapters that can be added to the system without
       # modifying either A or B.  *THIS IS ANOTHER PEP*

--------------------------------------------------------------
adapter/example.py
--------------------------------------------------------------

class EggsOnly:  # an unrelated class/interface
    def eggs(self,str): print "eggs!" + str

class HamOnly:  # used as an interface, no inhertance
    def ham(self,str): pass
    def _bugger(self): pass  # irritating a private member

class SpamOnly: # a base class, inheritance used
    def spam(self,str): print "spam!" + str

class EggsSpamAndHam (SpamOnly):
    def ham(self,str): print "ham!" + str
    def __adapt__(self,ident,check_only):
        if ident is HamOnly:
            return self       # implements HamOnly implicitly, no _bugger
        if ident is EggsOnly and not check_only:
            return EggsOnly() # Knows how to create the eggs!

------------------------------------------------------------------
testadapter.py
------------------------------------------------------------------
import adapter.example
from adapter import adapt
def testadapter():
    x = adapter.example.EggsSpamAndHam()
    adapt(x,adapter.example.SpamOnly).spam("Ni!")
    adapt(x,adapter.example.EggsOnly).eggs("Ni!")
    adapt(x,adapter.example.HamOnly).ham("Ni!")
    adapt(x,adapter.example.EggsSpamAndHam).ham("Ni!")
    #
    # imagine operator syntax for check_only:
    #
    #   if x isa adapter.example.SpamOnly: print "SpamOnly"
    #
    if (x,adapter.example.SpamOnly,1): print "SpamOnly"
    if (x,adapter.example.EggsOnly,1): print "EggsOnly"
    if (x,adapter.example.HamOnly,1): print "HamOnly"
    if (x,adapter.example.EggsSpamAndHam,1): print "EggsAndSpam"
    if (x,adapter.example.KnightsWhoSayNi,1): print "NightsWhoSayNi"
    adapt(x,adapter.example.KnightsWhoSayNi,1).spam("Ni!")


--------------------------------------------------------------------
example run
--------------------------------------------------------------------

>>> import testadapter
>>> testadapter.testadapter()
spam!Ni!
eggs!Ni!
ham!Ni!
ham!Ni!
SpamOnly
EggsOnly
HamOnly
EggsAndSpam
Traceback (innermost last):
  File "<interactive input>", line 1, in ?
  File "c:\work\testadapter.py", line 13, in testadapter
    if (x,adapter.example.KnightsWhoSayNi,1): print "NightsWhoSayNi"
AttributeError: KnightsWhoSayNi





From mlh at idi.ntnu.no  Fri Mar 23 13:28:10 2001
From: mlh at idi.ntnu.no (Magnus Lie Hetland)
Date: Fri, 23 Mar 2001 19:28:10 +0100
Subject: Python T-shirts
References: <mailman.985019605.8781.python-list@python.org>
Message-ID: <99g4ik$lb6$1@tyfon.itea.ntnu.no>

"Guido van Rossum" <guido at digicool.com> wrote in message
news:mailman.985019605.8781.python-list at python.org...
> At the conference we handed out T-shirts with the slogan on the back
> "Python: programming the way Guido indented it".

Just thought I'd mention...

  http://www.cafepress.com/

(I'm in no way affiliated with them.)

If you send them the grahics, at least the ones of us
who weren't at the conference can see what the T-shirts
look like before buying them (from you, for that matter).

(And we can get mugs, and stuff :)

Oh, well. Just a thought.

--

  Magnus Lie Hetland          http://www.hetland.org

 "Reality is what refuses to disappear when you stop
  believing in it"                 -- Philip K. Dick





From ben.hutchings at roundpoint.com  Mon Mar  5 14:10:40 2001
From: ben.hutchings at roundpoint.com (Ben Hutchings)
Date: 05 Mar 2001 11:10:40 -0800
Subject: jython question: accessing protected superclass methods
References: <jl66hqvzol.fsf@endor.bbb.caltech.edu>
Message-ID: <ud7bw821b.fsf@roundpoint.com>

Michael Vanier <mvanier at endor.bbb.caltech.edu> writes:

> Hi,
> 
> So, I've been playing with jython for a few hours, and I'm *really*
> impressed.  This is a kick-ass system for doing portable graphics
> programming.
> 
> However, the online docs are a bit sparse, and one problem I've been having
> is in trying to call a protected superclass method from a subclass which
> overrides the method.  The offending superclass is javax.swing.JPanel.  I
> have a subclass of this (let's call it MyJPanel) which needs to override the
> protected paintComponent() method of JPanel.  The naive approach:
> 
> class MyJPanel:
>     # ... lots of code ...
> 
>     def paintComponent(self, g):  # g == Graphics object
>         JPanel.paintComponent(self, g)
>         # extra code goes here
> 
> doesn't work, because paintComponent is not accessible from the MyJPanel
> namespace.  The online jython docs say this:
> 
>     In Python, if I want to call the foo method in my superclass, I use the
>     form:
> 
>         SuperClass.foo(self)
> 
>     This works with the majority of methods, but protected methods cannot be
>     called from subclasses in this way.  Instead you have to use the
>     "self.super_foo()" call style.

> 
> Well, I couldn't get this to work.  I'm not sure if the above
> statement means that you're supposed to write (in my case)
> "self.super_paintComponent(g)" or self.JPanel_paintComponent(g) but
> neither one works.  What is the magic invocation?
> 
> Also, I was surprised to find that MyJPanel, just by subclassing
> JPanel, gets its own public version of the protected JPanel methods!

That's what 'protected' means.

> So if I don't override the method, I can call it directly.  I assume
> this is intentional,
> but maybe the jython developers can comment.
> 
> Thanks,
> 
> Mike
> 
> --------------------------------------------------------------
> Mike Vanier	mvanier at bbb.caltech.edu
> Department of Computation and Neural Systems, Caltech 216-76

-- 
Any opinions expressed are my own and not necessarily those of Roundpoint.


From fredrik at pythonware.com  Tue Mar 20 12:41:42 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Tue, 20 Mar 2001 17:41:42 GMT
Subject: Functionalism, aesthetics Was:(RE: I come to praise .join,  not
References: <mailman.985052844.17993.python-list@python.org> <997v08$s4o$1@panix3.panix.com>
Message-ID: <qDMt6.2497$4N4.319619@newsc.telia.net>

Aahz Maruch wrote:
> Carlos Ribeiro  <carribeiro at yahoo.com> wrote:
> >
> >Its unfortunate that in many cases the non-expert user keeps quiet,
> >assuming that whatever opinion the "master" haves, its going to be better
> >than theirs - after all they are the so-called experts.
>
> There's some truth to that -- but rarely on Usenet newsgroups.

I doubt most experts bother to post to Usenet newsgroups
(or read them, for that matter).

and yes, this is true for Python too...

Cheers /F

    "Unfortunately, those people who have nothing better to do
    than post on the Internet all day long are rarely the ones who
    have the most insights." -- Jakob Nielsen





From greg at cosc.canterbury.ac.nz  Tue Mar 27 00:14:42 2001
From: greg at cosc.canterbury.ac.nz (Greg Ewing)
Date: Tue, 27 Mar 2001 17:14:42 +1200
Subject: Good Editor? (What's the big deal with syntax colouring?)
References: <mailman.985291829.27633.python-list@python.org> <99dqqo$gpi$1@news.mathworks.com> <99e9sv$usmp$1@ID-11957.news.dfncis.de> <99emg2$qif@gap.cco.caltech.edu> <3ABBAA81.FD344D5E@san.rr.com>
Message-ID: <3AC021C2.8BE6BF5D@cosc.canterbury.ac.nz>

Darren New wrote:
> 
> It didn't help that the RGB guns were misaligned so everything had fringes.

When I had to use a monitor which had a convergence problem
a few years ago, I took to using yellow on green. Since
yellow = green + red, you don't get any fringes.

Any combination in which the foreground differs from the
background in only one primary colour would work, but
I liked yellow/green the best.

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


From moshez at zadka.site.co.il  Wed Mar 21 07:24:53 2001
From: moshez at zadka.site.co.il (Moshe Zadka)
Date: Wed, 21 Mar 2001 14:24:53 +0200
Subject: Checking the type
In-Reply-To: <Pine.LNX.4.30.0103211504200.15322-100000@fep132.fep.ru>
References: <Pine.LNX.4.30.0103211504200.15322-100000@fep132.fep.ru>
Message-ID: <E14fhfZ-0002fY-00@darjeeling>

On Wed, 21 Mar 2001 15:05:10 +0300 (MSK), Oleg Broytmann <phd at phd.pp.ru> wrote:

>    Yes, builtin fuction type(x):
> 
> if type(x) <> type([]):
>    print "not list!"

Or even better, isinstance

if not isinstance(x, types.ListType):
	print "not list!"

-- 
"I'll be ex-DPL soon anyway so I'm        |LUKE: Is Perl better than Python?
looking for someplace else to grab power."|YODA: No...no... no. Quicker,
   -- Wichert Akkerman (on debian-private)|      easier, more seductive.
For public key, finger moshez at debian.org  |http://www.{python,debian,gnu}.org



From aleaxit at yahoo.com  Wed Mar 21 09:12:23 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Wed, 21 Mar 2001 15:12:23 +0100
Subject: Checking the type
References: <99a493$bhl$1@sun.rhrk.uni-kl.de>
Message-ID: <99acs601nsv@news2.newsguy.com>

"Markus Reitz" <Markus_Reitz at yahoo.com> wrote in message
news:99a493$bhl$1 at sun.rhrk.uni-kl.de...
> Hi,
>
> is there a possibilty to check if a variable refers to an object of type
x?
> Does a typeof-command exist in Python:
>
>   if a typeof list:
>      #Commands only useable with list
>
> I read the Python Tutorial, but have not found a hint to this topic.

Yes,

    if type(a)==type([]):

will give you what you're asking for -- but, it's normally
*NOT* what you really want.  Why stop future client-code
from passing in an instance object that just _mimics_ a
list (rather than _being_ one), for example?

It's more Pythonic to just go ahead and treat a AS IF it
were a list -- and maybe, depending on context, catch and
handle the exceptions you'll get if a is not amenable to
such treatment.

Sometimes you can do the equivalent-checks once and for
all at the start, to avoid incorrectly doing SOME changes
on an object that doesn't support all you need, for
example...:

# simplest
def munge(alist):
    alist.append(23)
    alist.extend(range(5))
    alist.append(42)
    alist[4] = alist[3]
    alist.extend(range(2))

this is simple BUT if alist happens to be an object that
has an append method and LACKS an extend method, or the
ability to be indexed, then alist will be 'partially
modified' and THEN the AttributeError or TypeError will
be raised -- often an unpleasant situation!

Not to worry -- it's not too hard to remedy this:

# safer
def munge(alist):
    # first extract all bound-methods we'll need
    append = alist.append
    extend = alist.extend

    # then check operations such as indexing
    try: a[0]=a[0]
    except IndexError: pass    # empty alist's OK

    # and finally operate -- no exceptions expected
    append(23)
    extend(range(5))
    append(42)
    alist[4] = alist[3]
    extend(range(2))


This doesn't interfere with polymorphism (any
object supporting all needed operations with the
right semantics will do), and yet is safer than
the previous version _without_ being substantially
slower or more complicated (well, a little, I guess).
It's important to avoid overdoing the checks, but
the assert statement can come in handy for checks
we don't want to happen in production/optimized
code (e.g, one may add "assert callable(append)"
etc to the above example).

(Such versions of the look-before-you-leap idiom
is popular in C++ templates, to get compiler
errors that are clearer and earlier, but then,
I've often said that C++ templates and Python
functions show surprising parallels:-).


Alex





From dougfort at downright.com  Wed Mar 21 09:27:34 2001
From: dougfort at downright.com (Doug Fort)
Date: Wed, 21 Mar 2001 09:27:34 -0500
Subject: Html entities
References: <Sp2u6.3034$mh4.198592@news3.oke.nextra.no>
Message-ID: <3AB8BA56.9E41A333@downright.com>

Syver Enstad wrote:

> Is there an easy way to convert ISO Latin-1 characters that are above 127
> ascii to their HTML, XML entity form?

See 'htmlentitydefs' in the Python library

--
Doug Fort (dougfort at downright.com)
Senior Meat Manager
Downright Software LLC
http://www.dougfort.net





From thomas at xs4all.net  Thu Mar  1 10:50:09 2001
From: thomas at xs4all.net (Thomas Wouters)
Date: Thu, 1 Mar 2001 16:50:09 +0100
Subject: [HAL2001-Announce] Decision Day take 2 (fwd)
Message-ID: <20010301165009.S22437@xs4all.nl>

I know this is off-topic, but if enough Pythonistas are going, we can make
it a Python event ;) I was part of the organization of HIP97, and it turned
out an incredibly fun event. HAL2001 will not be the same, but I'm certain
it will surpass HIP97 in more than a few areas :)

(Besides, it is partly ontopic: the lists are run by Mailman ;-)

----- Forwarded message from Gerrit Hiddink <grit at dnd.utwente.nl> -----

Date: Thu, 1 Mar 2001 15:28:55 +0100 (CET)
From: Gerrit Hiddink <grit at dnd.utwente.nl>
Sender: announce-admin at hal2001.org
To: announce at hal2001.org
Subject: [HAL2001-Announce] Decision Day take 2

Party is on...

It is with great pride and hollow insane laughter that we announce to
you that HAL 2001 will happen. At the time of writing this, we're
still working on closing some sizeable remaining budgetary gaps. But
most of our budget is now covered by sponsorships, so we feel it's
time to announce the good news.

Some of us have tried substitutes, and boy, did they taste swell. But
nothing, absolutely nothing, can stop this once-every-four-year itch.
After 1989, 1993 and 1997, it is time for a megalomaniac,
life-changing, hacker extravaganza event in The Netherlands again.

So, this is the time to re-schedule your marriage. Cancel that cruise
to the Caribean. Do whatever else is necessary to make sure that you
are on the Campus of the University of Twente, (in the east of The
Netherlands) between August 10th and 12th of this year, 2001.

Ticket sales will start relatively soon, and early information on many
aspects of the event will be on http://www.hal2001.org shortly. Expect
announcements from different corners of the HAL organization to
increase as the entire machine picks up speed over the next months.
The website offers the opportunity to subscribe to our announcement
mailing list.


HAL 2001

The HAL2001 convention is the fourth in a series that has been running
every four years since 1989. Quite a few of the participants at "The
Galactic Hacker Party" (1989), "Hacking at the End of the Universe" (1993)
and "Hacking in Progress" (1997) have been instrumental in bringing about
the changes that are upon us today.

HAL2001 expects to receive thousands of guests from all over the world and
from many different disciplines to debate issues ranging from advanced
technical issues regarding some obscure aspect of the Internet to
easy-to-understand lectures on some of the dangers of the information
society, as well as many, many other topics. But more than debate, the
guests at HAL2001 take ample time to get on-line, relax, build and discuss
cool stuff, and engage in good old analog interfacing.

HAL2001 is for those that can truly celebrate the Internet and embrace new
technologies, without forgetting their responsability to tell others that
all these wonderful new technologies come with new risks to the individual
and to society as a whole.
Sponsors:

The HAL 2001 organization would like to thank the sponsors for making
this event possible:

     XS4ALL Internet BV        www.xs4all.nl

     XS4ALL foundation		

     University of Twente      www.utwente.nl

     NLNet foundation          www.nlnet.nl

     Usenix                    www.usenix.org

     ITSX                      www.itsx.com

     GorillaPark               www.gorillapark.com

Board of HAL foundation:
Cor Bosman
Jaco Lockhorst
Rop Gonggrijp


HAL-staff: 
Gerrit Hiddink
Eric Slachmuylders


Contact Information:
Gerrit Hiddink
grit at wwcn.org
(+31) (0)53-4892425

_______________________________________________
Announce mailing list
Announce at hal2001.org
http://www.hal2001.org/mailman/listinfo/announce


----- End forwarded message -----

-- 
Thomas Wouters <thomas at xs4all.net>

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



From barry at digicool.com  Mon Mar 12 18:52:57 2001
From: barry at digicool.com (Barry A. Warsaw)
Date: Mon, 12 Mar 2001 18:52:57 -0500
Subject: [Python-Dev] Revive the types sig?
References: <jeremy@alum.mit.edu>
	<15020.9404.557943.164934@w221.z064000254.bwi-md.dsl.cnc.net>
	<200103120711.AAA09711@localhost.localdomain>
	<15021.22659.616556.298360@anthem.wooz.org>
	<200103122341.SAA23054@cj20424-a.reston1.va.home.com>
Message-ID: <15021.24921.998693.156809@anthem.wooz.org>

>>>>> "GvR" == Guido van Rossum <guido at digicool.com> writes:

    GvR> Barry, you were supposed to throw a brick at me with this
    GvR> content at the meeting, on Eric's behalf.  Why didn't you?  I
    GvR> was waiting for someone to explain why this was a big idea,
    GvR> but everybody kept their face shut!  :-(

I actually thought I had, but maybe it was a brick made of bouncy spam
instead of concrete. :/

    GvR> I'm not sure this is viable.  I believe Jeremy's compiler
    GvR> package actually doesn't have its own parser -- it uses the
    GvR> parser module (which invokes Python's standard parse) and
    GvR> then transmogrifies the parse tree into something more
    GvR> usable, but it doesn't change the syntax!  Quixote can get
    GvR> away with this because their only change is giving a
    GvR> different meaning to stand-alone string literals.  But for
    GvR> type annotations this doesn't give enough freedom, I expect.

I thought PTL definitely included a "template" declaration keyword, a
la, def, so they must have some solution here.  MEMs guys?

-Barry



From tbryan at python.net  Sat Mar 10 23:43:02 2001
From: tbryan at python.net (Thomas A. Bryan)
Date: Sat, 10 Mar 2001 23:43:02 -0500
Subject: running the Setup
References: <mailman.983222828.12381.python-list@python.org>
Message-ID: <3AAB0256.A38932B1@python.net>

Wayne wrote:

> I'm trying to compile the interface PyGreSQL.  I have no problem with
> the install instructions until
> I get to the "make". When I run "make" I get an error that indicates
> that readline has no rules.

It helps if you actually copy/paste the screen output into your post.
Troubleshooting make/install problems without the actual error message 
is a nightmare.

> Would I mess thing up if I comment that out? 

Possibly.  It depends whether PyGreSQL really needs it.

> The other Question I have is, shouldn't this error have come up when 
> the person putting together this distribution built the Python package?

Depends what's actually causing the error.

> Can someone recommend a good book for learning "Make"?

O'Reilly has a pretty good little book.  You can learn a lot from the 
first two chapters of that book.  Then read the man/info pages for the 
version of make that you're actually using.

> P.S. I know I can get an RPM for this, but I what to learn this stuff.

It also helps if you've written makefiles for your own C programs. 
It helps you to understand what other people's makefiles are trying 
to accomplish.

have-fun-making-make-make-stuff-ly yours,
---Tom


From stephen_purcell at yahoo.com  Fri Mar 30 02:00:36 2001
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Fri, 30 Mar 2001 09:00:36 +0200
Subject: Evaluating *EVERYTHING*
In-Reply-To: <eBLw6.3059$j_3.1383318@typhoon2.ba-dsg.net>; from mda@idatar.com on Thu, Mar 29, 2001 at 06:58:18PM +0000
References: <eBLw6.3059$j_3.1383318@typhoon2.ba-dsg.net>
Message-ID: <20010330090036.B4187@freedom.puma-ag.com>

David Allen wrote:
> Is there a way to get python to look through a whole
> file and say "Whoa there bucky - I've never heard
> of FoobarExceptoin"


You can never really track down all such possible problems, because the names
used for objects in your python code could (in theory more often than in
practice) be defined only at execution time. As an extreme example, look
at the following:

    >>> raise FoobarException        # not defined yet
    Traceback (innermost last):
      File "<stdin>", line 1, in ?
    NameError: FoobarException
    >>> import new
    >>> setattr(__import__('__main__'), 'FoobarException',                          ...         new.classobj('FoobarException',(Exception,),{}))
    >>> raise FoobarException        # now it's defined
    Traceback (innermost last):
      File "<stdin>", line 1, in ?
    __main__.FoobarException


For that reason, I see things like this: in Python, there's no difference
between a typo and a coding logic mistake, because you won't find either until
you run that code in a test or 'for real'.

-Steve


-- 
Steve Purcell, Pythangelist
Get testing at http://pyunit.sourceforge.net/
Any opinions expressed herein are my own and not necessarily those of Yahoo



From loredo at spacenet.tn.cornell.edu  Thu Mar 15 15:29:37 2001
From: loredo at spacenet.tn.cornell.edu (Tom Loredo)
Date: Thu, 15 Mar 2001 15:29:37 -0500
Subject: reading binary data
References: <3AAFE69D.FF01FEC8@yahoo.fr>
Message-ID: <3AB12631.6CC8550D@spacenet.tn.cornell.edu>

Hi-

Snack might help you; an old announcement is below.

-Tom Loredo

From: Kare Sjolander <kare at speech.kth.se>
Newsgroups: comp.lang.python
Subject: ANNOUNCE: The Snack Sound Toolkit v2.0.3
Date: Mon, 13 Nov 2000 13:58:02 +0100

This release contains:

* Many bug-fixes

* Updated documentation

* Updated demos that reflect new Snack features

* Expanded test suite

* Improved Windows installers

The Snack Sound Toolkit is designed to be used with a scripting
language.
Snack adds commands to play, record, and process sound and supports
in-memory sound objects, file based audio, and streaming audio.
It handles fileformats such as WAV, MP3, AU, AIFF, and NIST/Sphere.
Snack is extensible, new commands, filters, and sound file formats can
be added using the Snack C-library. An easy to build example extension
is contained in the source distribution.
Snack also does sound visualization, e.g. waveforms and spectrograms.
The visualization objects update in real-time and can output postscript.

Snack works with Tcl8.0 - Tcl8.4 and Python 1.5.2-2.0
Platforms: Linux, Solaris, HP-UX, IRIX, NetBSD, Macintosh, and
Windows95/98/NT/2K.

Source and binaries can be downloaded from
http://www.speech.kth.se/snack/

Regards,
Kare Sjolander
kare at speech.kth.se


From chrisa at ASPATECH.COM.BR  Mon Mar 12 15:27:39 2001
From: chrisa at ASPATECH.COM.BR (Chris Richard Adams)
Date: Mon, 12 Mar 2001 17:27:39 -0300
Subject: continue/else functionality
Message-ID: <BD674CDC0A68C642B60038986FBB56DF0138B7@mailsrv.ASPATECH.COM.BR>

I have a script that reads lines out of a text file. The first line in
the file serves as a "flag-line" stating whether or not the file should
be read: 1 yes, 0 no. The rest of the file contains data lines to be
read. How can I get my script to read on the first line and upon getting
a 1 go back to the for, get the next line and read the rest from there?
For example, I tried:


	for line in f.readlines():

	    datalist = string.split(line)
          if datalist[0][0] == 1:
             continue
          elif datalist[0][0] == 0:
             break
	    
	    ## Do a bunch of string manipulation here...

As the documentation states, from 4.4 of the tutorial "The continue
statement, also borrowed from C, continues with the next iteration of
the loop. "

Here it doesn't seem to be doing that...its continuing with the current
iteration of for.

Upon reading this first line, I need to go back and get the next line to
go on...

suggestions???

Thanks!
Chris



From aahz at panix.com  Mon Mar 26 12:08:17 2001
From: aahz at panix.com (Aahz Maruch)
Date: 26 Mar 2001 09:08:17 -0800
Subject: comp.lang.python.numeric?  (was Re: Numeric PEPs, first public posts)
References: <99669h$so4$1@panix6.panix.com> <mailman.985035504.26190.python-list@python.org> <mailman.985080535.7447.python-list@python.org> <slrn9bfsa6.rv.hzhu@mars.localdomain>
Message-ID: <99nt21$fda$1@panix3.panix.com>

In article <slrn9bfsa6.rv.hzhu at mars.localdomain>,
Huaiyu Zhu <hzhu at users.sourceforge.net> wrote:
>
>Anybody having the energy and know-how to do this?

I have the energy and know-how, but I haven't been convinced that this
is a Good Idea.  What's the total combined weekly traffic of all numeric
mailing lists, averaged over three months?
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

"I won't accept a model of the universe in which free will, omniscient
gods, and atheism are simultaneously true."  -- M


From tuttledon at hotmail.com  Wed Mar 21 00:25:34 2001
From: tuttledon at hotmail.com (Don Tuttle)
Date: Wed, 21 Mar 2001 05:25:34 GMT
Subject: Standalone executables
References: <998igr$4al$1@niagara.nonet>
Message-ID: <iXWt6.156032$Z8.33024358@typhoon.southeast.rr.com>

"Mark Carroll"
> I'm investigating the possibility of using wxPython for development.
> However, I need to be able to ship single standalone executables to
> random users of 32-bit MS Windows and Linux. (It would be really nice
> if I could compile both on Linux with mingw32 or suchlike. (-:)
>
> What are the current options, and how about the near future? I'm not
> having an easy time figuring out where things stand with freeze, etc.

On the Windows side, check out
http://starship.python.net/crew/theller/py2exe/ to compile and
http://www.jrsoftware.org/isinfo.htm to install your program.  Can't help
you on the Linux side.

Don




From bos at hack.org  Wed Mar 28 09:46:18 2001
From: bos at hack.org (Rikard Bosnjakovic)
Date: Wed, 28 Mar 2001 16:46:18 +0200
Subject: list vs tuple
References: <hGlw6.782$p5.3210@news1.rivrw1.nsw.optushome.com.au> <3AC1EB93.7BC87292@hack.org> <wEmw6.785$p5.3426@news1.rivrw1.nsw.optushome.com.au>
Message-ID: <3AC1F93A.BAC9E34B@hack.org>

deadmeat wrote:

> so is that the only difference?

Both are sequences, but tuples are "read-only", that's the only
difference (except for the in-place-methods which requires mutability).

-- 
Rikard Bosnjakovic - http://a214.ryd.student.liu.se/cv/ - ICQ: 1158217

Anyone sending unwanted advertising e-mail to my address will be
charged $250 for network traffic and computing time. By extracting my
address from this message or its header, you agree to these terms.


From tim.one at home.com  Thu Mar  1 18:53:41 2001
From: tim.one at home.com (Tim Peters)
Date: Thu, 1 Mar 2001 18:53:41 -0500
Subject: A counter-proposal to __future__ in PEP 236
In-Reply-To: <slrn99tmg1.15u.wtanksle@dolphin.openprojects.net>
Message-ID: <LNBBLJKPBEHFEDALKOLCGEDEJDAA.tim.one@home.com>

[Tim]
> from __future__ import nested_scopes
> assert nested_scopes[1] > sys.version_info, "don't need this anymore"

[William Tanksley]
> Is it really important that information such as this be stored as a tuple
> rather than a dict (with symbolic names rather than numbers)?

No.  Just another attempt to avoid hypergeneralizing PEP 236 into uselessness
<0.7 wink>.

> assert nested_scopes.version_info > sys.version_info, \
>                             "don't need this anymore"

See?  You started with "dict" but have already generalized to class instance.

> I've always been taught that magic numbers are unpleasant.

That's only because they're obscure and make things hard to change <wink>.  I
think I will make it a class instance -- thanks!




From lance.hill at olsinc.net  Thu Mar 22 20:05:15 2001
From: lance.hill at olsinc.net (lance.hill at olsinc.net)
Date: Fri, 23 Mar 2001 01:05:15 GMT
Subject: CGI for generating xml pages
References: <3aba9bc5.431603914@nnrp.wiznet.ca>
Message-ID: <3abaa18c.433083957@nnrp.wiznet.ca>

On Fri, 23 Mar 2001 00:50:55 GMT, lance.hill at olsinc.net wrote:

>
>I have a python script which creates xml. If I redirect the output to
>a file, when I launch those pages from my browser all is well.
>Unfortunately, I can't seem to get it so I can write directly to my
>browser from the script. Instead, I get an internal server error.
>
>Can someone kindly give me a clue? Do I need to do some processing
>prior to printing to standard output?
>
>Thanks in advance,
>L. Hill

BTW,

Suggestions on a tutorial or book covering xml/xslt with python would
also be appreciated.
THX.



From boud at rempt.xs4all.nl  Wed Mar 14 01:37:03 2001
From: boud at rempt.xs4all.nl (Boudewijn Rempt)
Date: 14 Mar 2001 06:37:03 GMT
Subject: PyQT-2.3 compile fails...
References: <3AAEBCCA.4E53671@pfortin.com>
Message-ID: <98n3if$kol$1@news1.xs4all.nl>

Pierre Fortin <pfortin at pfortin.com> wrote:
> Hi,

> Anyone have a clue as to why others can compile PyQT-2.3; but I get errors...? 
> I've also tried the CVS files...

> I'd hate to think this is a symlink related problem:
> /usr/local/src --> /hd/b6/src

> Thanks,
> Pierre

> # make
> make  all-recursive
> make[1]: Entering directory `/hd/b6/src/Python/PyQt-2.3'
> Making all in qt
> make[2]: Entering directory `/hd/b6/src/Python/PyQt-2.3/qt'
> /bin/sh ../libtool --mode=compile c++ -DHAVE_CONFIG_H -I. -I. -I.. 
> -I/usr/local/include/python2.0 -I/usr/local/include/sip -I/usr/local/qt/include 
> -I/usr/X11R6/include  -fno-exceptions -O2 -c sipqtQString.cpp
> c++ -DHAVE_CONFIG_H -I. -I. -I.. -I/usr/local/include/python2.0
> -I/usr/local/include/sip -I/usr/local/qt/include -I/usr/X11R6/include
> -fno-exceptions -O2 -Wp,-MD,.deps/sipqtQString.pp -c  -fPIC -DPIC
> sipqtQString.cpp -o sipqtQString.lo
> sip/qstring.sip: In function `int sipConvertTo_QString(PyObject *, QString **,
> int, int *)':
> sip/qstring.sip:456: no matching function for call to `QString::setUnicodeCodes
> (short unsigned int *&, int &)'
> make[2]: *** [sipqtQString.lo] Error 1
> make[2]: Leaving directory `/hd/b6/src/Python/PyQt-2.3/qt'
> make[1]: *** [all-recursive] Error 1
> make[1]: Leaving directory `/hd/b6/src/Python/PyQt-2.3'
> make: *** [all-recursive-am] Error 2

Has sip been compiled with Qt support? Do you have a Qt > 2.x?


-- 

Boudewijn Rempt  | http://www.valdyas.org


From piet at cs.uu.nl  Mon Mar  5 11:10:23 2001
From: piet at cs.uu.nl (piet at cs.uu.nl)
Date: 05 Mar 2001 17:10:23 +0100
Subject: Can SAX parser return <!-- COMMENTS -->  ???
In-Reply-To: Martin von Loewis's message of "05 Mar 2001 11:24:14 +0100"
References: <MWxn6.15038$x27.906254@news1.rdc2.on.home.com>
	<j4d7c0jh03.fsf@informatik.hu-berlin.de>
	<wzwva5shuy.fsf@sunshine.cs.uu.nl>
	<j48zmko6nl.fsf@informatik.hu-berlin.de>
Message-ID: <uitlo2o40.fsf@cs.uu.nl>

>>>>> Martin von Loewis <loewis at informatik.hu-berlin.de> (MvL) writes:

MvL> Piet van Oostrum <piet at cs.uu.nl> writes:
MvL> It is a matter of ongoing debate whether applications should even deal
MvL> with comments; if an application needs XML comments for proper
MvL> operation, something is wrong with that application.
>> 
>> What about a prettyprinter? Doesn't that count as an application?

MvL> Sure. But why does it have to preserve comments?

Prettyprinters are mostly used to make things more readable for human
beings. Comments are meant to be read by humans. So if a prettyprinter
deletes the comments it looses a big part of its value.
-- 
Piet van Oostrum <piet at cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: P.van.Oostrum at hccnet.nl


From cwc_NO at sound_NO.net  Mon Mar 26 00:15:59 2001
From: cwc_NO at sound_NO.net (Cary Coulter)
Date: Mon, 26 Mar 2001 05:15:59 GMT
Subject: ftpmirror.py problems
References: <vfhqbtgss3kotgac8gomnn0i9muh6j9ir7@4ax.com> <mailman.985565373.12533.python-list@python.org>
Message-ID: <scjtbt0ilj6eduuf9669bgp8tqldk3na1i@4ax.com>

I appreciate your comments (and quick response).

The network is connected via a cable modem.  The workstation in
question (running ftpmirror.py) is directly connected to the firewall
which is connected to the cable modem.  The MTU on all ethernet cards
in the network is 1500.

Let me add that if I connect to the desired ftp server via an ftp
client, I can download as many files as I want (using mget *) without
any problem.  This method surely has the same impact on IP Masq and
the ftp helper module.  Why is it this works but ftpmirror.py doesn't.

Since this is perhaps getting off topic, I am willing to continue this
discussion via email.  Be sure to remove the "_NO" from my address.

Thanks
Cary

On Sun, 25 Mar 2001 18:13:51 -0600, "Chris Gonnerman"
<chris.gonnerman at usa.net> wrote:

>You are right, in that it is the OS and not Python which is causing
>the problem.  You are probably having a known problem involving NAT 
>(Network Address Translation, i.e. IPMasq) on the local host and a 
>badly configured firewall on the remote end.
>
>If you can do so, configure the Linux system to use MTU/MRU 1500 when
>connecting via PPP.  If this does not work, try 296.
>
>The explanation is long and painful, and involves Path MTU Discovery.
>
>----- Original Message ----- 
>From: "Cary Coulter" <cwc_NO at sound_NO.net>
>Newsgroups: comp.lang.python
>To: <python-list at python.org>
>Sent: Sunday, March 25, 2001 5:46 PM
>Subject: ftpmirror.py problems
>
>
>> I am trying to use ftpmirror.py to duplicate various subdirectories.
>> When I test the replication between two hosts on my local network,
>> everything works like expected.
>> 
>> However, I am having problems replicating hosts residing on the
>> internet to my local system.  My local system is on a private network
>> (192.168...) behind a Linux firewall using IP Masq.  In this
>> environment, I am experiencing "hangs" in the ftpmirror.py
>> application.  I have to kill it with ^C.
>> 
>> I can normally download 10-40 files before this happens.  It seems to
>> be more prone if the files are smaller (can download in 3-5 seconds on
>> a cable modem) rather than the 10+ MB files.
>> 
>> Is anyone else experiencing this behavior?  Since it seem to work well
>> beween two local hosts, I am suspicious of the firewall and IP Masq.
>> I would be grateful for any pointers anyone can give to help remedy
>> this situation.
>> 
>> I am relatively new to Python. I really like what I see but I'm not
>> ready to debug this application.  I assume it really works and that
>> the problem is something else.
>> 
>> Thanks in advance for your input
>> Cary
>> 
>> -- 
>> http://mail.python.org/mailman/listinfo/python-list
>> 
>



From jasonsmith at jps.net  Tue Mar 13 19:20:06 2001
From: jasonsmith at jps.net (Jason D. Smith)
Date: Tue, 13 Mar 2001 16:20:06 -0800
Subject: How do I get Win32 to locate DLLs elsewhere (not really a Python thing)?
References: <mailman.984506597.12469.python-list@python.org> <kHxr6.293$XlD.39518660@news.randori.com>
Message-ID: <98mdjr$ao5$1@news2.symantec.com>

"Robin Dunn" <robin at alldunn.com> wrote in message
news:kHxr6.293$XlD.39518660 at news.randori.com...
|
| I don't remember the search order off the top of my head, but DLLs are
| normally searched for in the following locations:

>From the MSDN docs for LoadLibrary (Win's dlopen() analog)

--------------8<--------------8<--------------
  1.. The directory from which the application loaded.
  2.. The current directory.
  3.. Windows 95/98: The Windows system directory. Use the
GetSystemDirectory function to get the path of this directory.
  Windows NT/ 2000: The 32-bit Windows system directory. Use the
GetSystemDirectory function to get the path of this directory. The name of
this directory is SYSTEM32.

  4.. Windows NT/ 2000: The 16-bit Windows system directory. There is no
function that obtains the path of this directory, but it is searched. The
name of this directory is SYSTEM.
  5.. The Windows directory. Use the GetWindowsDirectory function to get the
path of this directory.
  6.. The directories that are listed in the PATH environment variable.
Jason.

--

 Hi.  I'm Bob.  I'm a tomato and I'm here to help you.




From scott at zenplex.com  Fri Mar 30 11:00:51 2001
From: scott at zenplex.com (scott at zenplex.com)
Date: Fri, 30 Mar 2001 11:00:51 -0500
Subject: Is there a bug tracker program written in Python?
Message-ID: <20010330.110051.1782436840.1258@scoot.zenplex.com>

I'm looking for a Bug Track system like bugzilla but written in Python.
Does anybody know of such?  Just learned Python and would would like to
use a system based on Python if at all possible.
Thanks
Scott


From timr at probo.com  Tue Mar 20 01:20:06 2001
From: timr at probo.com (Tim Roberts)
Date: Mon, 19 Mar 2001 22:20:06 -0800
Subject: HTML escape in Python 1.5.2
References: <3fQTii$l4N@openbazaar.net>
Message-ID: <0htdbtg56d7glfjg6lqndugb8apdv289v3@4ax.com>

Pahud.bbs at openbazaar.net (???J?w) wrote:
>
>I want to do the html escaping in python, however, HTMLgen class
>does not ship with python 1.5.2. Is there any alternative?

Why look for an alternative?  If you're running it on a machine where you
can't install into python site-packages, just load it into the same
directory where you have your CGI scripts.

I'm a real fan of HTMLgen.
--
- Tim Roberts, timr at probo.com
  Providenza & Boekelheide, Inc.


From bedge at troikanetworks.com  Mon Mar  5 14:49:38 2001
From: bedge at troikanetworks.com (Bruce Edge)
Date: Mon, 05 Mar 2001 11:49:38 -0800
Subject: package/namespace confusion
Message-ID: <3AA3EDD2.60393490@troikanetworks.com>

Hi,
I'm sure this is a real rookie issue here, but I can't seem to define a
global then access it in a package.
I'm using __init__.py to define glob_var, then load util.py.

I would have expected util/ufunc() to have access to glob_var.


test/__init__.py:

glob_var=0        <- is this where I should stick package globals?
__all__=["util"]


test/util.py:

def ufunc():
	global glob_var
	print "glob_var =", glob_var
	glob_var=2
	print "glob_var =", glob_var



>>> from test import *
>>> glob_var 
0
>>> util.ufunc()
glob_var =
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
  File "test/util.py", line 3, in ufunc
    print "glob_var =", glob_var
NameError: There is no variable named 'glob_var'
>>> dir()
['__builtins__', '__doc__', '__name__', 'atexit', 'glob_var',
'histfile', 'os', 'readline', 'rlcompleter', 'savehist', 'sys', 'util']


From bkline at rksystems.com  Thu Mar 29 11:27:16 2001
From: bkline at rksystems.com (Bob Kline)
Date: Thu, 29 Mar 2001 11:27:16 -0500 (EST)
Subject: Printing  "%"
In-Reply-To: <99vm4k$22l$1@ctb-nnrp2.saix.net>
Message-ID: <Pine.LNX.4.10.10103291118470.31480-100000@rksystems.com>

On Thu, 29 Mar 2001, Mix wrote:

> How do you print a percenrage (%)sign in Python?
> I tried :
>   return "select * from fax where id LIKE '%s\%'" %ident

As the manual says, it works just like the C printf conversion does.
Try:
    return "select * from fax where id LIKE '%s%%'" % ident

-- 
Bob Kline
mailto:bkline at rksystems.com
http://www.rksystems.com




From dougfort at downright.com  Tue Mar 13 20:20:39 2001
From: dougfort at downright.com (Doug Fort)
Date: Tue, 13 Mar 2001 20:20:39 -0500
Subject: Unpickling from strings
References: <qHxr6.19658$pR3.3830844@news2-win.server.ntlworld.com>
Message-ID: <IHzr6.21597$Sl.962179@iad-read.news.verio.net>

Perhaps it is getting garbled in transmission.  We quote ours:

    urllib.quote_plus(cPickle.dumps(self._statsdict))

--
Doug Fort dougfort at downright.com
Senior Meat Manager
Downright Software LLC
http://www.dougfort.net

"Marcin Chady" <marcin.chady at ntlworld.com> wrote in message
news:qHxr6.19658$pR3.3830844 at news2-win.server.ntlworld.com...
> I have pickled an object into a string and saved it in an hidden HTML form
> field, so that I can recover the state when the form is resubmitted.
> However, when the form comes back and I try to unpickle my object I get an
> error saying that the module in which the object's class was originally
> defined is unknown, even though the CGI script which does the unpickling
has
> already imported the module!
>
> Has anyone got an idea why it is?
>
> Marcin
>
>




From paulp at ActiveState.com  Tue Mar 13 18:27:13 2001
From: paulp at ActiveState.com (Paul Prescod)
Date: Tue, 13 Mar 2001 15:27:13 -0800
Subject: Yet another Python vs. Perl speed issue/question
References: <98m6ad+i25s@eGroups.com>
Message-ID: <3AAEACD1.F5CE28D1@ActiveState.com>

nanotech at europa.com wrote:
> 
> All:
> 
> Python 2.0 on an HP (nice and big) takes ~30 seconds to lowercase all
> the lines in a 3meg file. Perl takes less than a second. What am I
> doing wrong?!?
> 
> Python
> ------
> import sys
> import string
> 
> stdin=sys.stdin
> lower=string.lower
> 
> print map(lower,stdin.readlines())

My lowly laptop runs this in around a second on 3MB. Something must be
misconfigured on your server. What's Python's startup time like? Just
type "python"? Does it come up quickly?

-- 
Python:
    Programming the way
    Guido
    indented it.
       - (originated with Skip Montanaro?)



From fdrake at acm.org  Fri Mar 23 15:38:55 2001
From: fdrake at acm.org (Fred L. Drake, Jr.)
Date: Fri, 23 Mar 2001 15:38:55 -0500 (EST)
Subject: Anyone using weakrefs?
Message-ID: <15035.46175.599654.851399@localhost.localdomain>

  Is anyone out there playing with the weak references support yet?
I'd *really* appreciate receiving a short snippet of non-contrived
code that makes use of weak references to use in the documentation.
  Thanks!


  -Fred

-- 
Fred L. Drake, Jr.  <fdrake at acm.org>
PythonLabs at Digital Creations




From jkraska1 at san.rr.com  Sat Mar 17 22:16:04 2001
From: jkraska1 at san.rr.com (Courageous)
Date: Sun, 18 Mar 2001 03:16:04 GMT
Subject: Who's minister of propaganda this week?
References: <B98041329020E007.7DA84CAD9D0562AB.817BBC572F480869@lp.airnews.net> <98nu1m$12e@dispatch.concentric.net> <660ACCA1E4816F7E.C359CDF4168EF3FB.3B6B05538D619770@lp.airnews.net> <3AAFA13D.76E32C3F@divalsim.it> <98oa4o01i4v@news2.newsguy.com> <3AB0176D.8070903@yahoo.com> <3AB0C70A.C6A55DF5@ActiveState.com> <3AB40BE3.3080001@yahoo.com> <mailman.984880948.1738.python-list@python.org>
Message-ID: <98a8bto9ukuv1h0if5bh4gnfeucnakje8t@4ax.com>

>helped too. The situation is the same with dynamic typing. Moore's law
>and education will bring programmers over.

Yes. For example, there are "performance thresholds" that once passed
mean that performance doesn't matter as much in a particular subject area.
For example, if a scripting language can display dialog boxes
instantaneously without any "visible" pause, then the scripting language
has passed that performance threshold. The differences between 1/100th
of a second to draw and 1/1000th of a second to draw the dialog don't
matter in any way that most people care about.

Since Python is a superior programming environment for a great many
things, time itself is Python's friend. In time, due to Moore's law, Python
will pass more and more peformance thresholds, just in the same way
that other languages eventually surpassed assembly language.



C//



From danielk at aracnet.com  Sat Mar  3 13:08:48 2001
From: danielk at aracnet.com (Daniel Klein)
Date: Sat, 03 Mar 2001 10:08:48 -0800
Subject: Missing the functional mind set
Message-ID: <mlb2at8ptrr7sh5dgq8qlk2201kquh6tfl@4ax.com>

The purpose of this code is to return a list of dictionary values sorted by
key, resulting in 'templist'.

>>> d = {3: 'd', 2: 'b', 1: 'a', 0: 'z'}
>>> templist = []
>>> for n in range(len(d)):
	templist.append(d[n])

>>> templist
['z', 'a', 'b', 'd']

Of course when I have this in its best form, it will become a class method.

My question is, is there a 'functional' (or list comprehension) solution to
this? I don't seem to have a sufficient grasp of functional programming as yet.
I don't really want the 'complete' answer as I'd like to try to work it out for
meself. I guess I'm just looking for a 'push' in the right direction; maybe
just a hint to get that lightbulb to pop on over my head.

Note: this is _not_ a school assignment; this is for a real project I am doing
for my company.

Thanks,
Daniel Klein


From mysticsguy at yahoo.ca  Sat Mar 24 19:26:20 2001
From: mysticsguy at yahoo.ca (Yvon Boulianne)
Date: Sun, 25 Mar 2001 00:26:20 GMT
Subject: time to invest ???
Message-ID: <3abd3a17.7103358@news.vif.com>

Hello, i start to learn Python some day ago, with lot of knowledge as
a system admin but little as programmer and i wonder what is the
average time/hours someone have to invest to be able to use a language
like that easily.  i talk as well about devlopping the way of thinking
of a programmer and the knowledge of the language. please don't tell
me theres no average, at least take the idea of a normal guy, not a
genius and not stupid  ;-)

Thanks verry much and sorry for my english, my first language is
french.

Yvon Boulianne ; Saphi
mysticsguy at yahoo.ca
www.monunivers.net
Humour,Paradox & Changements


From syver at NOSPAMcyberwatcher.com  Thu Mar 29 01:10:08 2001
From: syver at NOSPAMcyberwatcher.com (Syver Enstad)
Date: Thu, 29 Mar 2001 08:10:08 +0200
Subject: Python scripts as Drop targets in Windows Explorer
References: <99qogo$o6g$1@troll.powertech.no> <99qqq3$qt7$0@216.39.170.247> <99r043$qqr$1@troll.powertech.no> <vdqs99.p16.ln@haegar.fqdn.th-h.de>
Message-ID: <8kAw6.780$R6.22057@news1.oke.nextra.no>

"Michael Linnemann" <m.linnemann at addcom.de> wrote

> Yeah, looks cool. What is it, perl? Czech? EBCDIC?
As it said in the posting:
>>default on Win2k is to save as unicode.
> Come on, tune your W2k a little and make it talk plain :-)
I don't know what you mean, isn't Unicode one of the most plain ways to talk
nowadays? And I have also posted it as plain text so it shouldn't be too
difficult for you.





From topmind at technologist.com  Sun Mar  4 15:37:14 2001
From: topmind at technologist.com (Topmind)
Date: Sun, 04 Mar 2001 20:37:14 GMT
Subject: Collection interfaces
References: <3a90fa97.0@news1.mweb.co.za> <3A933830.7F98D118@ix.netcom.com> <CgJk6.515$v77.66568@newsread2.prod.itd.earthlink.net> <220220010813598042%cmh@bDistributed.com> <ud7cakvi8.fsf@roundpoint.com> <MPG.150312ceb68f1ada98969f@news.earthlink.net> <8Gem6.473353$U46.14105062@news1.sttls1.wa.home.com> <MPG.1503548620e632919896a4@news.earthlink.net> <MCjm6.474652$U46.14163288@news1.sttls1.wa.home.com> <MPG.150394195f7f04a19896a7@news.earthlink.net> <3A9A86B1.9A8C69A2@ix.netcom.com> <MPG.1504d14563dce6499896af@news.earthlink.net> <97fjla128ir@news2.newsguy.com> <MPG.1505a76831c34b749896b3@news.earthlink.net> <97h5s40vhc@news2.newsguy.com> <MPG.1506007591cf818b9896bc@news.earthlink.net> <slrn99opop.gr.amk@mira.erols.com> <MPG.1506265ebf6d934d9896be@news.earthlink.net> <97ibmc0k47@news1.newsguy.com> <MPG.1506cf3293b094b49896c1@news.earthlink.net> <97m0c7080c@news2.newsguy.com> <6e7a18554a.simon@satin.callan.demon.co.uk>
Message-ID: <MPG.150c49ce34317639896d6@news.earthlink.net>

> In message <97m0c7080c at news2.newsguy.com>
>           "Alex Martelli" <aleaxit at yahoo.com> wrote:
> 
> > there STILL is no silver bullet.
> 
> Years back, I became aware to the treu reason why there is no silver
> bullet for programming, and why there can *never* be one. We're trying
> to kill the wrong monster.
> ....

One "trick" methodology never yet formalized is what I call
"Cult Oriented Programming". Find and hire only those
developers who have a certain affinity for a given language
or design methodology. IOW, the trick may be like-minded-ness
among developers rather than the magic-paradigm-for-all.

Developers who think differently often drive each other
bats. For example, I optimize my code to prevent the
kind of mistakes that *I* make. The things that throw
others may be completely different. I am real good at
some things and totally ditzy about others, and everybody
seems to be this way, but with a different set of tradeoffs. 

Einstein often forgot to put is pants on in the morning (it
is sometimes claimed).

-tmind-


From francois_forest at yahoo.fr  Wed Mar 14 16:46:41 2001
From: francois_forest at yahoo.fr (Francois Forest)
Date: Wed, 14 Mar 2001 21:46:41 GMT
Subject: reading binary data
Message-ID: <3AAFE69D.FF01FEC8@yahoo.fr>

I would like to read and write xxx.wav files,
this assume to alternatively read characters,  integers 16 bits and
integers 32 bits.
I found that I can read a binary file and get the result in a string,
Is there a way to convert a string slice into an int16, int32, or
floatting point ?
(what is usually done in C through a cast).

More generally, how can I browse through standard libraries ?



From MarkH at ActiveState.com  Thu Mar 22 07:59:34 2001
From: MarkH at ActiveState.com (Mark Hammond)
Date: Thu, 22 Mar 2001 12:59:34 GMT
Subject: How to increase PythonWin v2.0: COM browser limits?
References: <994psh$so5$1@news1.xs4all.nl> <uy9u06vji.fsf@roundpoint.com> <3AB9444E.5040403@ActiveState.com> <99ccsk015b7@news2.newsguy.com>
Message-ID: <3AB9F880.4090609@ActiveState.com>

Alex Martelli wrote:


>> Ahh - excellent.  Thats for tracking that down - it has always worked for
> 
> me.
> 
> Me too -- the categories part.  The one that breaks down is
> the registered-libraries part (of combrowse.py).

Ahh - excellent :)

Thanks!

Mark.



From ljohnson at resgen.com  Wed Mar  7 09:27:16 2001
From: ljohnson at resgen.com (Lyle Johnson)
Date: Wed, 7 Mar 2001 08:27:16 -0600
Subject: finding and listing the differences between two files
References: <mailman.983966710.3462.python-list@python.org>
Message-ID: <tach5qh8vhgka2@corp.supernews.com>

Chris,

Unless there's more to this story, I think your best bet is to use what's
already working (diff). You can of course suck in the results by opening a
pipe in Python, e.g.

    import os

    results = os.popen("diff file1 file2")

See the standard documentation on the "os" module and pipes for more
details. Now, keep in mind that this won't necessarily work if you move your
Python script to an operating system that doesn't have "diff" available at
the command line (say, Windows).

Lyle




From cfelling at iae.nl  Thu Mar 29 18:02:21 2001
From: cfelling at iae.nl (Carel Fellinger)
Date: 30 Mar 2001 01:02:21 +0200
Subject: Accessing variables in __main__ from modules....
References: <mailman.985891179.24578.python-list@python.org>
Message-ID: <9a0ett$eqb$1@animus.fel.iae.nl>

nanotech at europa.com wrote:

> DEBUG would be searched for at the module scope. But I want it to 
> look for DEBUG back in __main__, like:

The most straightforwarth way to achieve this would be to import Debug
from main, or better to have a seperate options module and import from
there. like:

module options:

   DEBUG = 0

main module:

   import options
   ...
   options.DEBUG = 1

other module:

   from options import DEBUG

   if DEBUG:
     print "-I-: Reading '%s'....has %s lines...."%len(filename,lines)




-- 
groetjes, carel


From kens at sightreader.com  Wed Mar  7 20:52:28 2001
From: kens at sightreader.com (Ken Seehof)
Date: Wed, 7 Mar 2001 17:52:28 -0800
Subject: Can Python replace Visual Basic? Should it?
References: <5DZo6.11126$hn5.1578410@news1.rdc1.mb.home.com> <u8zmhyusj.fsf@advancedmp.net>
Message-ID: <002701c0a772$8c41b280$fbb1f7d1@mobilestar.com>

That's easy.  Yes.

Also, I recommend wxPython (www.wxpython.org) as your GUI framework (in
python you have several good ones to choose from).  It's cross platform and
a much cleaner design than the Microsoft stuff.  Be sure to run the wxPython
demo program; it will save you lots of time learning.

----- Original Message -----
From: "Don Dwiggins" <dwig at advancedmp.net>
To: "Brad Bollenbach" <bbollenbach at homenospam.com>
Cc: <python-list at python.org>
Sent: Wednesday, March 07, 2001 4:21 PM
Subject: Re: Can Python replace Visual Basic? Should it?


> Brad Bollenbach writes:
> >     Can Python replace Visual Basic? Should it?
>
> > But really, very specifically, were YOU a former Visual Basic (or
> > Powerbuilder, or Delphi) programmer that is now using Python to write
the
> > complex GUI's VB used to handle for you? I'm an on again, off again,
student
> > of Python, who really wants to give this a go, if there's a good chance
that
> > I'll get the return on my investment: a programming tool that can take
over
> > where VB, Powerbuilder, and Delphi used to play.





From scarblac at pino.selwerd.nl  Thu Mar  8 02:05:03 2001
From: scarblac at pino.selwerd.nl (Remco Gerlich)
Date: 8 Mar 2001 07:05:03 GMT
Subject: Can anyone recomend a good intoduction to C...
References: <mailman.983894288.19965.python-list@python.org> <3AA5EF59.507F1D91@olsen.ch> <983958878.976344@newsmaster-04.atnet.at> <77up6.1289$y6.235541@ruti.visi.com> <985vqs01ipr@drn.newsguy.com> <ewwp6.1466$y6.241005@ruti.visi.com>
Message-ID: <slrn9aeblb.cg1.scarblac@pino.selwerd.nl>

Grant Edwards <grante at visi.com> wrote in comp.lang.python: 
> Right!  As in most software project failures, the problem lay
> in the requirements.  If you ask for something with both the
> unsafe features required for backwards compatibility with C
> _plus_ all the bells and whistles they thought they needed for
> OO programming, there's no way you can end up with something
> that isn't an unholy mess.  I think asking for a language that
> did both low-level stuff like C and OO stuff was a mistake.

I think Objective C is a nice language, a real good "C with OO". It's just
ANSI C with support for Smalltalk-style OO.

All this bloat added to C++ was not necessary for making an OO language.

-- 
Remco Gerlich


From MarkH at ActiveState.com  Wed Mar 21 19:10:45 2001
From: MarkH at ActiveState.com (Mark Hammond)
Date: Thu, 22 Mar 2001 00:10:45 GMT
Subject: How to increase PythonWin v2.0: COM browser limits?
References: <994psh$so5$1@news1.xs4all.nl> <uy9u06vji.fsf@roundpoint.com>
Message-ID: <3AB9444E.5040403@ActiveState.com>

Ben Hutchings wrote:

> In fact, by adding a print statement to the loop that's iterating over
> the categories to add them to the tree, I discovered that it yields
> more or less the same results as OLEView (a similar program available
> from Microsoft).  The exception is only raised after these results
> have been returned.

Ahh - excellent.  Thats for tracking that down - it has always worked for me.

Does the following patch solve the problem?

Index: combrowse.py
===================================================================
RCS file: /home/cvsroot/PyWin32/com/win32com/client/combrowse.py,v
retrieving revision 1.4
diff -r1.4 combrowse.py
101,102c101,106
< 		for catid, lcid, desc in enum:
< 			ret.append(HLICategory((catid, lcid, desc)))
---
> 		try:
> 			for catid, lcid, desc in enum:
> 				ret.append(HLICategory((catid, lcid, desc)))
> 		except pythoncom.com_error:
> 			# Registered categories occasionally seem to give spurious errors.
> 			pass # Use what we already have.



Mark.



From idot at vt.edu  Sat Mar 17 14:43:33 2001
From: idot at vt.edu (Carl Banks)
Date: Sat, 17 Mar 2001 14:43:33 -0500
Subject: Deleting dictionary items
References: <98ug6n$q9d$1@news6.svr.pol.co.uk>
Message-ID: <5pe099.cn.ln@127.0.0.1>

Will Newton <will at nospam.misconception.org.uk>,
exactly what drugs were you under the influence of when you wrote:
> 
> What's the best way to delete dictionary items?
> 
> I have a dictionary with tuples as keys, and I would like to delete all 
> keys with a certain element. First I wrote:
> 
>                 for item in td.data.keys():
>                         if item[attr] != value:
>                                 del td.data[item]
> 
> 
> But will this work?

Probably not.  This might work, though:

                for item in td.data.keys():
                        if item[attr] == value:
                                del td.data[item]


-- 
CARL BANKS

"I like nothing better than a good challenge.
 This is not one of them."


From tuttledon at hotmail.com  Wed Mar 21 22:46:17 2001
From: tuttledon at hotmail.com (Don Tuttle)
Date: Thu, 22 Mar 2001 03:46:17 GMT
Subject: Please rate these Python books
References: <20010322.1445209@mis.configured.host>
Message-ID: <dAeu6.144721$__6.26735546@typhoon.southeast.rr.com>

"TM"

I'm wondering if the last books on Python are good. Thanks for your
points of view.
I'm especially thinking about :

1/Programming Python 2nd edition @ O'reilly
2/ Core Python Programming / Chun / Prentice Hall
3/ Python Essential Reference / David Beazley / New Riders Publishing

All 3 are excellent books.  You won't go wrong with any of them.  (Althought
you might want to wait til June for Beazley's 2nd edition.) For more info
see the tread "Learning Python..."

Don







From tjg at exceptionalminds.com  Wed Mar 14 19:03:02 2001
From: tjg at exceptionalminds.com (Timothy Grant)
Date: Wed, 14 Mar 2001 16:03:02 -0800
Subject: shadow password library
Message-ID: <20010314160302.F3109@trufflehunter.avalongroup.net>

Hi,

Is there a shadow password library available anywhere?

-- 
Stand Fast,
    tjg.

Timothy Grant                         tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.         <><       (503) 246-3630
>>>>>>>>>>>>>Linux, because rebooting is *NOT* normal<<<<<<<<<
>>>>This machine was last rebooted:  57 days  4:17 hours ago<<



From tanzer at swing.co.at  Fri Mar  2 02:54:17 2001
From: tanzer at swing.co.at (Christian Tanzer)
Date: Fri, 02 Mar 2001 08:54:17 +0100
Subject: str behaviour 
In-Reply-To: Your message of "Thu, 01 Mar 2001 16:32:35 EST."
             <LNBBLJKPBEHFEDALKOLCGECFJDAA.tim.one@home.com> 
Message-ID: <m14YkOH-000wcBC@swing.co.at>

"Tim Peters" <tim.one at home.com> wrote:

> > I expected str to be called recursively for the constituent types.
> > But, alas, that did not happen. Am I misunderstanding something or is
> > it the intended behaviour? If the latter is the case, could someone
> > elaborate on why this behaviour was chosen?
> 
> You're on the edge of a very deep pit.  The builtin container types (lists,
> tuples, dicts) "pass repr down" regardless of whether str() or repr() is
> applied at the top level.  I've been whining about that for years myself
> <wink>.

> The primary reason is maddeningly simple:  it's for the benefit of contained
> strings. 
(snip)
> I call it "maddening" because, to date, nobody has thought of an alternative
> that doesn't suck just as bad in one respect or another.  If DejaNews were
> still useful, searching for ssctsoos ("str special casing the snot out of
> strings") would yield more blather on the topic than you could tolerate.

Just off the top of my head.

Would it work to have a third function, say `display' (*), besides `str'
and `repr'? The container types would "pass display down" regardless
of whether str() or repr() is applied at the top level.

For classes, the designer could choose whatever meaning applied best
for `__display__' -- whether it be an alias of `str' or of `repr' or
something entirely different.

For most builtin types, `display' would be equivalent to `str', for
strings it would be equivalent to `repr'.

(*) `display' isn't really a good name for this function but nothing
    better occurred to me. This might be a bogosity warning.

-- 
Christian Tanzer                                         tanzer at swing.co.at
Glasauergasse 32                                       Tel: +43 1 876 62 36
A-1130 Vienna, Austria                                 Fax: +43 1 877 66 92




From whisper at oz.net  Sat Mar 10 03:43:26 2001
From: whisper at oz.net (Dave LeBlanc)
Date: 10 Mar 2001 08:43:26 GMT
Subject: Urgent Question about Python licensing
Message-ID: <98cpfe$5oj$0@216.39.170.247>

Hi;

While reading up on the recent Python conference on O'Reilly's site
(http://python.oreilly.com/news/pythonday1_0301.html - actually on the
'day 2' page), I cam across the following:
----------------------------------------------------
"Some hurdles that still lie ahead for Python: Python's license is
still stuck between the Scylla of risk-averse CNRI and the Charybdis
of unbending Richard Stallman. The license that CNRI agreed to for
Python 1.6 was approved by the Open Source Initiative (OSI), but the
Free Software Foundation still objects to one small clause. 

Because later versions of Python are built on 1.6, they too are
affected by this license problem. This difficulty means that Python
can't be distributed with software licensed under the GPL. Discussions
continue between CNRI and FSF, and progress is being made; but as
Guido asked during the keynote: When will it end?"
----------------------------------------------------

My questions are these:
1. What prevents Python and GPL code from being distributed together?
If I write some Python code that drives a GPL'd app (but  which does
not depend on that app), what prevents me from offering both my python
code and the GPL'd app on the same medium (disk/CD/web)?

2. Is it currently so, or will Python end up with, some license that
requires that python scripts be covered by the GPL?

3. Is Python 2.0 covered by that evil CNRI 1.6 license? I avoided 1.6
mostly because of the license, and thought that 2.0 had "skipped" 1.6
and used a far more equitable license then the 1.6 one. (1.6's license
permits the copyright holder to suddenly decide that anyone using the
covered software must pay a fee (or any other thing they so desire -
this is only an example), and by virtue of your even so much as having
read the license are implicitly bound to their unilateral declaration
- lovely Virginia laws).

Regards,

Dave LeBlanc


From ws-news at gmx.at  Fri Mar  9 08:45:04 2001
From: ws-news at gmx.at (Werner Schiendl)
Date: Fri, 9 Mar 2001 14:45:04 +0100
Subject: COM support (was Re: Can Python replace Visual Basic? Should it?)
References: <5DZo6.11126$hn5.1578410@news1.rdc1.mb.home.com> <mailman.984010931.26257.python-list@python.org> <iSBp6.14005$hn5.2035325@news1.rdc1.mb.home.com> <mailman.984094514.29514.python-list@python.org> <984133550.452523@newsmaster-04.atnet.at> <98ai2j01o8j@news2.newsguy.com>
Message-ID: <984145785.263773@newsmaster-04.atnet.at>

There are definitely areas in COM - a lot of such, to be precise - where
Visual Basic is _not_ suited (at all).
Thing is, there are a lot of things that do not need to care on how the
marshaling is done in detail (because the default is good enough for a lot
of applications, IF you have well designed interfaces - if not, writing your
own marshaling code is the wrong answer to the problem) or how the security
is handled in detail (becuase, e. g. it is not intended to use even by a
local client - only inproc like active x controls or utility classes to
provide functions Visual Basic is missing by default).

As I mentioned, I do not have much experience with Python and COM and I
_thought_, based on that very small insight, Visual Basic would be the more
convenient environment. I read a lot of rather headache stuff on generating
interface classes and things like that and this is one point Visual Basic
handles nicely. ActiveX control containment is also one thing I would have
no idea on how to start.

BTW: I intend to get a copy of O'Reillys book "Python Programming on Win32"
to deepen my insight on that topic. If you know
other links or books worth reading (apart from the docs that come with
python, of course) I would appreciate to hear about them.

regards
werner

P.S.: Of course one should (or must, to bring it to the point) understand
the technology in use. This applies heavily to COM, and I know quite a lot
about that stuff. This is one of the reasons I think that well designed
interfaces are more important for remoting than the actual programming
language. And you can use IDL defined Interfaces with VB, otherwise it would
be a headache some time...

Alex Martelli <aleaxit at yahoo.com> wrote in message
news:98ai2j01o8j at news2.newsguy.com...
> "Werner Schiendl" <ws-news at gmx.at> wrote in message
> news:984133550.452523 at newsmaster-04.atnet.at...
>     [snip]
> > I did not do much COM stuff in Python right now, but I think you cannot
> > compare it to Visual Basic at this time.
> > Of course, Mark Hammond does a really impressing job on that issue, but
> > Microsoft is still somewhat in advance with VB I think.
>
> Speaking as somebody whose main bill-paying-techolological-expertise
> right now is COM -- I disagree.  VB may well be better for those who
> _lack_ significant COM expertise and still need to handle a lot of COM
> stuff -- if nothing else, because it does a good job of handling the
> visual parts of Active/X, and comes with a zillion wizards which (not
> always successfully) strive to hide the complexities from you.  But
> if one DOES know COM (or needs to do complex-enough stuff that the
> knowledge of COM itself is not optional anyway), Python is better --
> because it does _NOT_ strive to hide things behind your back, you are
> in far better control -- essentially the same potential for "control
> of what's going on" that you get with VC++ and ATL, but without the
> horrid complexities thereof.
>
> Like with ATL, though, you do have to study the sources you're supplied;
> the docs are not bad, but they never go deep enough into the system's
> bowels -- only study of sources can take you there.
> [[However, this also applies to theoretically "simple" stuff, as soon as
> you need to answer questions that are deep enough in themselves, e.g.,
> "if I malloc a zillion byte then free them, will the memory be given
> back to the system or will my process keep it to itself for possible
> future needs" -- this can be crucial in practice, but only studying
> the sources of your C system's runtime libraries will help you answer.]]
>
>
> Alex
>
>
>




From cfelling at iae.nl  Sat Mar 10 14:52:50 2001
From: cfelling at iae.nl (Carel Fellinger)
Date: 10 Mar 2001 20:52:50 +0100
Subject: How to efficently build a nested dictionary
References: <3AAA2042.E69A53FC@gmx.de> <98d9f1$1hopn$1@ID-11957.news.dfncis.de> <3AAA7188.266516E9@schlund.de>
Message-ID: <98e0mi$3fd$1@animus.fel.iae.nl>

Carsten Gaebler <cg at schlund.de> wrote:
> I'd like to be able to print out the annual, monthly or daily traffic for
> each directory. With nested dictionaries I could easily accomplish this by
> using nested for loops. How would I do that with a compound key?

   import operator
   vs = [traffic[D,y,m,d] for (D,y,m,d) in traffic.keys()
         if D==dir and y==year ]
   total = reduce(operator.add, vs, 0)

This has the bonus feature that it's dead easy to get strange totals,
like the total of all traffic on mondays:

   import DateTime
   vs = [traffic[D,y,m,d] for (D,y,m,d) in traffic.keys()
         if D==dir and y==year and DateTime(y,m,d).day_of_week==0 ]
   total_on_mondays = reduce(operator.add, vs, 0)
-- 
groetjes, carel


From francois.granger at free.fr  Sat Mar 17 07:47:34 2001
From: francois.granger at free.fr (=?ISO-8859-1?Q?Fran=E7ois_Granger?=)
Date: Sat, 17 Mar 2001 12:47:34 GMT
Subject: Python style questions
References: <98tb91$9t9$1@saltmine.radix.net> <98tdu1$cfk$1@panix2.panix.com> <1eqdo5u.1xxmzfi2kjk0qN%francois.granger@free.fr> <98u24p0a98@news2.newsguy.com>
Message-ID: <1eqevzh.3951ay94w9e7N%francois.granger@free.fr>

Alex Martelli <aleaxit at yahoo.com> wrote:

> There are several ways you can use a dictionary to dispatch pieces
> of code;
[...]
> def toy2(fileob, operation, *data):
>     switch = {'leggi': fileob.read, 'scrivi': fileob.write,
>         'chiudi': fileob.close }
>     return switch[operation](*data)

Fine, great, i'll test it.
Thanks

-- 
[fr, en, es, ia]
Information sur la hi?rarchie usenet europa.* :
http://europa.usenet.eu.org


From wware at world.std.com  Mon Mar  5 19:20:22 2001
From: wware at world.std.com (Will Ware)
Date: Tue, 6 Mar 2001 00:20:22 GMT
Subject: MySQL-python build griefs
References: <G9qvt6.L0I@world.std.com> <xqxvgpn4xtl.fsf@cola.ca.boeing.com>
Message-ID: <G9r29y.Azn@world.std.com>

Harry George (hgg9140 at cola.ca.boeing.com) wrote:
> #===edit Setup===
> if sys.platform == "linux-i386": # Red Hat
>     include_dirs = ['/usr/local/mysql/include/mysql']
>     library_dirs = ['/usr/local/mysql/lib/mysql']
> #===end edits===

Indeed, I had the wrong settings for those directories
in the initial setup.py. But I ran into other troubles
as well. In Distutils, in sysconfig.py, I am getting
this around line 207:

    if done.has_key(n):
        after = value[m.end():]
        print "value[:m.start()]", repr(value[:m.start()])
        print "done[n]", type(done[n]), repr(done[n])
        print "after", repr(after)
        value = value[:m.start()] + done[n] + after

which is giving me:

value[:m.start()] ''
done[n] <type 'string'> 'Parser Objects Python Modules'
after ' $(SUBDIRSTOO) Ext-dummy'
value[:m.start()] 'README ChangeLog '
done[n] <type 'string'> 'configure configure.in acconfig.h config.h.in Makefile.in'
after ''
value[:m.start()] '${INSTALL} -m '
done[n] <type 'int'> 644
after ''
Traceback (innermost last):
  File "/usr/lib/python1.5/site-packages/distutils/sysconfig.py", line 210, in parse_makefile
    value = value[:m.start()] + done[n] + after
TypeError: illegal argument type for built-in operation

What's happening apparently is that a permission is being taken as an
integer rather than a string. A little scarey since the value is
supposed to be octal.

I'll poke around and see if I can get more clues. Spooky. Thanks for
the directories tip.

-- 
-----------------------------------+---------------------
 22nd century: Esperanto, geodesic | Will Ware
 domes, hovercrafts, metric system | wware at world.std.com


From kens at sightreader.com  Thu Mar 22 14:03:58 2001
From: kens at sightreader.com (Ken Seehof)
Date: Thu, 22 Mar 2001 11:03:58 -0800
Subject: Checking the type
References: <99a493$bhl$1@sun.rhrk.uni-kl.de> <99daj70aeu@enews4.newsguy.com>
Message-ID: <002d01c0b302$da37cb50$04090a0a@upcast.com>

Here's a handy trick that lets you avoid importing the types module,
and saves you from learning all the type names (ListType, et. al.)
The compare is a tiny bit slower, but looks nicer (I think).

>>> a = [2,3,4]
>>> b = "3,4,5"
>>> def foo(x):
...  if type(x) == type([]):   # just use type() with a literal
...     print 'yes'
...  else:
...     print 'no'
...
>>> foo(a)
yes
>>> foo(b)
no


> Markus Reitz <Markus_Reitz at yahoo.com> wrote in message
> news:99a493$bhl$1 at sun.rhrk.uni-kl.de...
> > Hi,
> >
> > is there a possibilty to check if a variable refers to an object of type
> x?
> > Does a typeof-command exist in Python:
> >
> >   if a typeof list:
> >      #Commands only useable with list
> >
> > I read the Python Tutorial, but have not found a hint to this topic.
> >
> > Thanks
> >
> > Markus
> >
> >
>
> Sample code:
>
>      def Show (self, object):
>           """ <object> can be number, string, list, tuple, or
dictionary"""
>           from types import *
>           kind = type (object)
>           if kind == StringType:
>               self.Writeln (object)
>           elif (kind == ListType) or (kind == TupleType):
>                for item in object:
>                     self.Writeln (item)
>           elif kind == DictType:
>                sKeys = object.keys()
>                sKeys.sort()
>                for key in sKeys:
>                     self.Writeln ('%-16s   %s' % (key, `object[key]`))
>           else:
>               self.Writeln (`object`)
>
>           self.Activate()
>
> Mike





From ransen_spam_me_not at nemo.it  Sat Mar 10 11:37:43 2001
From: ransen_spam_me_not at nemo.it (Owen F. Ransen)
Date: Sat, 10 Mar 2001 16:37:43 GMT
Subject: I don't get it (Extending C/C++)
References: <98b3cf$apu$02$1@news.t-online.com>
Message-ID: <3aa9c775.1422022@news.newsguy.com>

On Fri, 9 Mar 2001 18:30:37 +0100, "Moritz Voss" <thygrrr at gmx.net>
wrote:

>Also, is it advisable to invoke the interpreter (do I even do that? what DO
>I do when I execute a python function?) several hundred times? (running
>about 500 instances of several, sometimes identical python scripts times
>would be likely for medium load of my application).

If you are using 2.0 I've found that the Py_Initialize() Py_Finalize()
crashes on the 8th call of this pai, so you would be better to do
one Py_Initialize() at the start of your program and one Py_Finalize()
at the end.

If you are using 2.1b1 the Py_Initialize()/Py_Finalize() bug seems to
have been fixed.

I've been doing this for about 2 weeks, so I know how you feel, 
don't take my word as the word of long time experience!




--
Owen F. Ransen
http://www.ransen.com/
Home of Gliftic & Repligator Image Generators


From sholden at holdenweb.com  Fri Mar  9 08:57:01 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Fri, 9 Mar 2001 08:57:01 -0500
Subject: BCD or fixed point numbers
References: <3AA8E272.7268DCFF@poczta.onet.pl>
Message-ID: <dm5q6.45942$ML1.1994402@e420r-atl2.usenetserver.com>

"jacek czerwinski" <jacy at poczta.onet.pl> wrote in message
news:3AA8E272.7268DCFF at poczta.onet.pl...
> Has anybody some links for python with BCD or other fixed point numbers
> ?

I'm fairly sure the timbot has done some such work.

A Google search reveals

    http://mail.python.org/pipermail/python-dev/2000-October/009728.html

which is in the public domain. Thanks yet again, Tim.

regards
 Steve





From sholden at holdenweb.com  Mon Mar  5 00:58:49 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Mon, 5 Mar 2001 00:58:49 -0500
Subject: Python COM server to VB interface
References: <keyo6.156158$Ch.31627133@newsrump.sjc.telocity.net>
Message-ID: <WZFo6.14920$ML1.733222@e420r-atl2.usenetserver.com>

"dsavitsk" <dsavitsk at e-coli.net> wrote in message
news:keyo6.156158$Ch.31627133 at newsrump.sjc.telocity.net...
> I am attempting to learn to use python as a com server to a VB interface.
> When i run my little example i get the following error.
>
> Run-time error'-2147467259 (800004005)':
> Unexpected Python Error: AttributeError: CreateInstance
>
> Here is the Python end of things...
>
>     class pyserv_test:
>         _public_methods_ = [ 'AddString' ]
>         _reg_progid_ = 'PyServ.Test'
>         _reg_clsid_ = '{EA08AFC9-E0AA-4C9B-8E42-1170A21D59AC}'
>
>         def AddString(self, val):
>             return 'the string was: ' + str(val)
>
>     def Register():
>         import win32com.server.register
>         return win32com.server.register.UseCommandLine(pyserv_test)
>
>     if __name__=='__main__':
>         print "Registering COM server..."
>         Register()
>
>
> and here is the relevant VB code...
>
>     Private Sub Command1_Click()
>         Dim x
>         Dim y
>         Set x = CreateObject("PyServ.Test")
>         y = x.AddString(Text2.Text)
>         Text1.Text = y
>     End Sub
>
> The error occurs on the "CreateObject" line. Does anyone know what this
> might be?
>
It may well be due to the absence on an __init__() method in your object?

regards
 Steve





From amk at mira.erols.com  Sun Mar  4 20:44:35 2001
From: amk at mira.erols.com (A.M. Kuchling)
Date: 5 Mar 2001 01:44:35 GMT
Subject: [Python-Dev] Re: Bug fix releases
References: <LNBBLJKPBEHFEDALKOLCMEEPJBAA.tim.one@home.com> <mailman.982897324.9109.python-list@python.org> <200103031821.NAA24060@panix3.panix.com> <200103031910.OAA21663@cj20424-a.reston1.va.home.com> <mailman.983751489.10136.python-list@python.org>
Message-ID: <slrn9a5sni.gi.amk@mira.erols.com>

On Mon, 5 Mar 2001 01:16:55 +0100, Thomas Wouters <thomas at xs4all.net> wrote:
>How can you say this ? Okay, so *fixing* bugs isn't terribly exciting, but
>hunting them down is one of the best sports around. Same for optimizations:
>rewriting the code might be boring (though if you are a fast typist, it
>usually doesn't take long enough to get boring :) but thinking them up is
>the fun part. 

Plus, time not spent on new Python features is free to be spent on new
libraries or applications.  Is it more fun to hack on the core, or on
a library such as ZODB or whatever?  I'm finding the latter, and as a
bonus it's much easier since you have a clean slate to start from.

--amk



From tage at pasta.cs.uit.no  Wed Mar 14 12:56:21 2001
From: tage at pasta.cs.uit.no (Tage Stabell-Kulo)
Date: 14 Mar 2001 17:56:21 GMT
Subject: syslog broken on FreeBSD?
References: <mailman.983631064.7213.python-list@python.org>
Message-ID: <98obc5$16v$1@news.uit.no>

Gabriel Ambuehl <gabriel_ambuehl-py at buz.ch> writes:

>I've been trying to write to syslog using the following
>import syslog
>syslog("TEST)


Using syslog from Python is a pain in the back, mostly because it used
to be an optional module.  I wrote a drop-in replacement, in Python.
It is used as follows:

   try:
      import syslog
   except ImportError:
      import slog
      syslog = slog

or

   try:
       from syslog import *
   except ImportError:
       from slog import *

You can now use syslog.openlog or openlog, respectively.  That is, on
systems where syslog isn't compiled or supported by Python, your
programs will run without modification.  Furthermore, you can easily
alter the module to send your messages to another machine, rather than
to /dev/log.

Believed to be fully compatible with syslog(3).  It is based on a
module, not at all compatible with syslog(3), written in 1997 by Sam
Rushing.

It was my first Python program, so please be kind if you find ugly
code.

You can fetch your very own copy at
    ftp://ftp.pasta.cs.uit.no/pub/users/tage/Bredband/slog.py

> Gabriel

 [TaSK]

-- 
////       Tage Stabell-Kuloe        |e-mail: Tage at ACM.org (at=@)////
///Department of Computer Science/IMR|Phone : +47-776-44032         ///
//9037  University of Tromsoe, Norway|Fax   : +47-776-44580         //
/      "'oe' is '\o' in TeX"         |URL:http://www.cs.uit.no/~tage/


From michel at digicool.com  Fri Mar 23 15:06:59 2001
From: michel at digicool.com (Michel Pelletier)
Date: Fri, 23 Mar 2001 12:06:59 -0800 (PST)
Subject: PEP 245: Python interfaces
In-Reply-To: <Pine.LNX.4.21.0103230302200.19638-100000@clarkevans.com>
Message-ID: <Pine.LNX.4.32.0103231156000.26302-100000@localhost.localdomain>

On Fri, 23 Mar 2001, Clark C. Evans wrote:

> > http://python.sourceforge.net/peps/pep-0245.html
>
> This is very good work  Michel.

Thanks.

> Here are my comments
> which I've been mulling over for the last few days.
>
>   a)  First, I'm not sure a new type is needed, perhaps
>       classes work just fine (see the PyXML library
>       for current interface declaration practices...)
>
>       In general, your PEP didn't convince me why
>       the class construct couldn't be extended
>       in a reasonable way to satisify the goals
>       you enumerated without introducing a new type.

The reason behind this is that classes are shared implementation, they are
not an interface.  Any interface of a class is an implied, assumed thing.
Interface assertions make that more concrete.  Classes could in some way,
be extened to have an inline "interface" but that is mixing your interface
with your implementation.  How can you reuse that?  How can you have
mulitple implementations?

>   Assuming that the "class" type can be augmented:
>
>   b)  If the class syntax was extended, perhaps it
>       may be useful to add a new statement like
>       pass, but could be used to mark "pure"
>       methods (like = 0 in C++), perhaps:
>
>          class X:
>             def func(self,z): pure
>
>       Then, if any method in a class was marked
>       "pure", the class could not be instantiated.

Same argument as above.

>   c)  If the class syntax was extended, perhaps
>       there might be a way to signify "inherit"
>       from "implements", perhaps:
>
>          class X (Y,#Z):
>
>       The above class X "inherits" from the
>       class Y, and "implements" the public
>       attributes found in class Z, but
>       without sharing the implementation.

Once again, this is mixing interface with implementation.  Z should not
be implementation.

>    d) What is also missing from Python is
>       an "auto-delegation".  An analysis
>       of how this fits in or doesn't fit
>       in would be helpful.

What is that?  I've neve heard of auto-delegation.

>    Assuming that a new "interface" type were
>    to be added.
>
>    e) I personally feel that the syntax should
>       be much different from the syntax of a
>       class to reduce confusion.  Note that
>       maps and lists have different syntaxes.
>       The different syntaxes give better
>       context for programmers...

Ok.

>    f) If a new type is added, why not go further?

Because I don't want to risk sliding into the absurd. ;)

>       Although I don't like the idea of argument
>       types in classes... they could very well
>       make sence for interfaces.  This could help
>       to differentiate the syntax, perhaps:
>
>         interface Y:
>            decl func( number as types.IntType,
>                       string as types.StringType,
>                       any    as any,
>                       cls    as IAnotherInterface ):
>                "This function does such and such"

These goals are reasonable.  I don't like the syntax myself.

>    g)  I like the pre/post condition stuff, however,
>        this shoudl be put in a seperate PEP as it
>        is seperable and could be added later, no?

Yes, that's probably a good idea, I'll look at it.

>    h)  I don't like the string for pre/post conditions.

They're just python expressions like that used with the 'assert' keyword.
What's not to like about them?

>        They should have special syntax, perhaps as
>        these could be implemented as "local methods?"
>
>          interface Y:
>              decl func(...)
>                  def pre(self):
>                     if ... then raise PreConditionFailure
>                  def post(self): pass
>                  def test(self): pass

What if you interface wants to have pre post and test methods that have
nothing to do with type enforcement?

> Wheu!  I hope this helps!

Thanks!

-Michel




From ken at lt.com  Mon Mar  5 16:17:52 2001
From: ken at lt.com (Kenneth Loafman)
Date: Mon, 05 Mar 2001 21:17:52 +0000
Subject: Revenge - Careful there
References: <97fbbm$v1o$1@snowmane.julianhaight.com> <97goaj$jj4$1@snowmane.julianhaight.com> <3A9BFE1D.B94F6AA5@lt.com> <97j7kn$p21$1@snowmane.julianhaight.com>
Message-ID: <3AA40280.D8864084@lt.com>

It would not matter whether they had ANI or Caller ID, the 800-spammer
number would only get 1 call from any one individual as a protest call. 
The trick would be that that over a few thousand individual protest
calls, eventually they would shut down or be costed out of profit.  No
one individual would ever be guilty (under current law) of harassment
since they only called spammer-X once and made a polite request to cease
and desist.  Kind of an automated "We protest your activity." calling
plan.

...Ken

SPAMFIGHTER wrote:
> 
> Careful, again.
>  ANI is not like Caller-ID where the data is transmitted as an
> ascii 1200 bps modem tone between the first and second rings, in
> most cases.  With ANI, the identification is known BEFORE the
> phone at the other end rings.  A short call won't stop the trace
> if it rings the far end as it is grabbed (I forget how many
> milliseconds) prior to the application of the ring burst.
> Actually, even a lot of the more recent Caller ID implementations
> now connect to the called end and send the data while that phone
> is still on-hook, so the data gets there before that phone rings,
> too.  It depends on the Central Office and the version of
> software they are using and the CID box the subscriber has.  So,
> ground starts have always been covered, and now the residential
> loop starts are becoming covered, too, as the updated software
> becomes more common in digital (not analog) central offices.
> 
> Tom
> 
> "Kenneth Loafman" <ken at lt.com> wrote in message
> news:3A9BFE1D.B94F6AA5 at lt.com...
> | Matt Dragonfly Drury wrote:
> | >
> | > Many toll-free numbers use ANI for tracking inbound numbers,
> not Caller ID,
> | > which isn't affected by the star-code to disable the
> transmission of one's
> | > number. The argument being that since the company is
> effectively paying for
> | > a collect call, they've the right to know who is calling.
> | >
> | > For that matter, I recall a case a while back where someone
> phone-spammed a
> | > televangelist he didn't like, and was held liable for the
> costs of the
> | > hundreds of short calls his modem demon-dialed for him. Not
> the most
> | > discreet method, obviously.
> | >
> | > That being said, were such a tactic to be distributed - a
> telephone-like
> | > "DDoS attack" - I'm sure it'd be harder for accused spammers
> to pin it down
> | > to one person.
> | >
> | > I wonder what kind of tracing and identification is in place
> for
> | > Internet-telephony calls? I'm sure it depends on the
> provider.
> | >
> | > I've mused in the past on keeping a list of toll-free numbers
> handy whose
> | > companies I find distasteful, and giving them a piece of my
> mind on their
> | > dime when I happen to be near a phone some distance from my
> residence.
> |
> | All this discussion leads me to another interesting idea.  If I
> | *mistakenly* call a spammer and hang up, then no charge can be
> made.
> | What we need to do is apply that 1000's of times, all across
> the world.
> | A lot of us have multiple modems that we no longer use that
> could be put
> | to good use.  We could set up a Seti-like system where a work
> block is a
> | list of spammer's 800-numbers.  You would run a dialing program
> to dial
> | that number *exactly one* time, then step to the next number.
> Lots of
> | individual calls.  No single point of harassment, and the work
> block
> | would only be executed once per client system so that no single
> number
> | ever called more than once.
> |
> | Of course, we'd need to run the server somewhere outside the
> US, but
> | that's not hard to set up.  This could be fun.
> |
> | Think about it.  The ultimate revenge, especially if we could
> | incorportate telemarketers phone numbers into the mix as well.
> Those
> | would be limited to calls between 6 and 8pm to catch them at
> supper.
> | Totally distributed, just like they do to us.
> |
> | OK, some days I'm totally evil, but I'm getting tired of spam
> and
> | telemarketing, actually marketing in general.  Do unto others
> as they've
> | done unto you.
> |
> | ...Ken



From martin.skott at propylon.com  Mon Mar 19 16:56:09 2001
From: martin.skott at propylon.com (Martin Skott)
Date: 19 Mar 2001 21:56:09 +0000
Subject: Organizing Python
References: <995ogm$1u6$0@216.39.170.247>
Message-ID: <zoeh4e4m.fsf@propylon.com>

whisper at oz.net (Dave LeBlanc) writes:

> What seems to be a reasonably trivial question to me, but i've not
> been able to find an answer in any of the Python doc i've looked at.
> 
> Is there a convention for installing 3rd party add-on modules into a
> Python installation so that:
> 	
> 	It's easy to update Python without reinstalling all the extras
> 
> 	One doesn't have to have a mile-long PYTHONPATH
> 
> I've seen some mention of packages and local/packages directories wrt
> to Linux installations, but I'm looking for a Windows solution.

This might have concerned overall package management eg. not only
Python, but the entire system. My preferred setup is to have a central
systemwide direectory with common modules and then a directory in my
homedirectory with my more personal modules, but then again that is on
Unix.
 
> I think part of why I have to ask this question is not quite
> understanding how PYTHONPATH and import works. When you say "import
> foo", then foo is taken to be a file and if you say "from foo import
> bar", then foo is taken to be a directory? Would this mean that one
> would have to have to have both foo and foo's parent directory in
> PYTHONPATH?

When you say import foo you add the namespace of the module foo to
your current namespace (all with the name foo.* ofcourse). When you
say from foo import bar then you import the name bar (eg. a class)
from the module foo. A module is usually represented by a file, but it
is also possible to have it be a directory which I once read something
about on python.org, but I can't find it now. What your PYTHONPATH
contains is a list of places to search for these modules. If your
PYTHONPATH names the directories /usr/lib/python and /home/me/modules
then these directories will always be searched when you in a script
use the import command.

-- 
Martin Skott
Software Developer
Propylon - Enabling Universal Mobility
http://www.propylon.com
Tel: +353 096 37245
Mobile: +353 087 9680370


From gregj at pobox.com  Mon Mar 12 14:36:35 2001
From: gregj at pobox.com (Gregory Jorgensen)
Date: Mon, 12 Mar 2001 19:36:35 GMT
Subject: Einstein's Riddle
References: <oNPq6.396644$ge4.138479566@news2.rdc2.tx.home.com> <UdSq6.2184$54.2166@www.newsranger.com>
Message-ID: <7z9r6.2576$54.2837@www.newsranger.com>

I just looked at a bunch of web sites I turned up with Google searching for
"einstein riddle fish." All of them conclude that the German owns the fish. I
didn't find any confirmation that Einstein actually conceived the riddle or said
that only 2% of the population could solve it. Nor could I determine if the
original author wrote the riddle in English, German, or some other language.

Clearly my statement that the puzzle can't be solved puts me in a minority. If
you assume that SOMEONE keeps fish the puzzle can be solved with a little bit of
logic. Perhaps the intent of the riddle is to show how few people can devise a
useful and logical methodology.

Greg Jorgensen
Deschooling Society
Portland, Oregon, USA
gregj at pobox.com


From cce at clarkevans.com  Tue Mar 27 08:55:03 2001
From: cce at clarkevans.com (Clark C. Evans)
Date: Tue, 27 Mar 2001 08:55:03 -0500 (EST)
Subject: who is my caller?
Message-ID: <Pine.LNX.4.21.0103270853400.24308-100000@clarkevans.com>

In many mud languages, you can query the
call-stack to find out who called you.
Is there a stack of function objects
that can be obtained somehow?  Is this
non-portable black magic?  

I'm asking beacuse this technique is
often useful for implementing security
(ok, not very good security)

Thanks!

Clark




From tfa613 at yahoo.de  Sun Mar  4 07:39:22 2001
From: tfa613 at yahoo.de (Torsten Gallmeister)
Date: Sun, 4 Mar 2001 13:39:22 +0100
Subject: what refers to an object
Message-ID: <3aa2378f$1@news.uni-rostock.de>

Hallo,

I have a problem with debugging of my app.
I'm programming a ZOPE-Product. When Zope starts it is unpickling some
Objects "A". I never wanted to store the Objects in the DB. May be it was a
prior mistake, that some Objects "B" stored in the DB have a reference to
some Objects "A".
Is there any way to find out what referes to the "A"-Objects to delete them?
Tank's for any help,
                                Torsten.




From thomas at xs4all.net  Sat Mar 10 05:06:49 2001
From: thomas at xs4all.net (Thomas Wouters)
Date: Sat, 10 Mar 2001 11:06:49 +0100
Subject: Urgent Question about Python licensing
In-Reply-To: <Pine.BSF.4.21.0103100307160.26511-100000@open-systems.net>; from chris@voodooland.net on Sat, Mar 10, 2001 at 03:10:34AM -0600
References: <98cpfe$5oj$0@216.39.170.247> <Pine.BSF.4.21.0103100307160.26511-100000@open-systems.net>
Message-ID: <20010310110649.M404@xs4all.nl>

On Sat, Mar 10, 2001 at 03:10:34AM -0600, Chris Watson wrote:

> > 2. Is it currently so, or will Python end up with, some license that
> > requires that python scripts be covered by the GPL?

> That would be when Python starts shipping under something like a GPL+. And
> the day that happens I for one will drop Python like a bad habbit. I would
> prefer a BSD or BSD style license. But im sure im out-voted on that.

Yes -- but only because the CNRI has the most say over the Python licence,
currently. They want (and need) a licence that they can feel safe with, that
they can consider binding. The current Python licence is actually very close
to the BSD style licence, except for the 'choice of law' licence that is
causing the GPL incompatibility.

> GNU = Gnu's Non-portable & Unstable
> irc.openprojects.net #FreeBSD -Join the revolution!

That's kinda unfair. It's also not true. The GNU project delivered a lot of
good software. Have you noticed that FreeBSD does not come with its own
compiler, but with the GNU compiler instead ? How can you claim it's
non-portable if gcc is the most portable compiler out there ? Notice that
GNU Mailman, the mailing list software that runs this and many other
mailinglists, is written in Python, and so is almost as portable as is
Python ? And it's GNU!

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

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



From cribeiro at mail.inet.com.br  Fri Mar 30 09:59:50 2001
From: cribeiro at mail.inet.com.br (Carlos Alberto Reis Ribeiro)
Date: Fri, 30 Mar 2001 11:59:50 -0300
Subject: PEP 245: Python interfaces
In-Reply-To: <9a1v0r030up@news1.newsguy.com>
References: <Pine.LNX.4.32.0103222253510.26302-100000@localhost.localdomain>
 <99o9g4$nh3$0@216.39.170.247>
 <Pine.LNX.4.21.0103261430360.25260-100000@localhost.localdomain>
 <j4wv99m503.fsf@informatik.hu-berlin.de>
 <mailman.985899166.8262.python-list@python.org>
Message-ID: <5.0.2.1.0.20010330114231.021e9400@mail.inet.com.br>

At 14:42 30/03/01 +0200, Alex Martelli wrote:
>Multiple inheritance of implementation (mixins) is a highly
>convenient and practical programming technique; dunno what
>you mean by "legitimate" cases in this context?

I agree. Note that the use of mixins is very similar to composition. In 
fact, most "legitimate" (read: clear and unambiguous) uses of multiple 
inheritance are related to the use of mixins. The difference is that, in 
"pure" composition, you must make a explicit choice about what and how 
contained objects will be exposed. Mixins make it automatic. But even then 
you must be careful about name clashes... (that's one of the reasons I 
prefer to use the "explicit" approach).

>Implementation by one object of multiple interfaces (which
>is also expressed as multiple-inheritance in C++) is more
>than just convenient -- it's a must to enable most decent
>modeling techniques ("not relying on absolute taxonomy" is
>what I mean by "decent" here:-).

That's exactly one of the strong points of PEP245. You can implement 
several interfaces *without* actually using multiple inheritance. The 
difference between inheriting behavior (inheritance) and implementing 
behavior (interfaces) is clear.

I tend to favor other kinds of factoring when building my class libraries, 
in particular those who lead to composition and decoration. It's more 
flexible and easier to adapt for improvements - no need to change a long 
chain of hierarchy to fix or change the behavior, only subsets of the class 
library need to be changed.

(OTOH, have you ever noted how confusing most textbook examples on multiple 
inheritance are? They're badly designed. That is a big part of the problem. 
It makes it hard for the novice to understand what it is really about; I 
took a long time to realize "what it is" and "what it is not")


Carlos Ribeiro





From bdupire at seatech.fau.edu  Sat Mar 24 18:40:59 2001
From: bdupire at seatech.fau.edu (Benoit Dupire)
Date: Sat, 24 Mar 2001 18:40:59 -0500
Subject: dynamically reload a module...Solution
References: <3ABB8520.D42D5E66@seatech.fau.edu>
Message-ID: <3ABD308B.F6BCA1FF@seatech.fau.edu>

I  post here the answer to my question, as it may interest other people.
It was found by Daniel Yoo and Michael P Reilly. Thanks a lot.
Here is a summary.....

Benoit Dupire wrote:

> My goal is to dynamically import  a module called 'func' from prog1.py and
> to get the
> references to the objects created by func
>
> # prog1.py
>
> class foo:
>     def __init__(self, module):
>         self.aa=__import__(module)          # dynamically import func
>         chRef= getattr(self.aa, 'setheading')
>         chRef.run()
>
> my_foo = foo('func')
>
> # ----------------------------------
> # func.py
> class SetHeading:
>     def run(self):
>         print 'test'
>
> setheading=SetHeading()
>

> It works fine..
> Now I change func.py, replacing ' setheading=SetHeading()' with
> 'toto=SetHeading()'

>
> Normally, this should create an error when I run prog1.py, because the
> setheading object does not exist anymore...
> However:
> >>> reload(prog1)
> test
> <module 'prog1' from 'C:\Python20\prog1.pyc'>
>
> func is not reimported!
>
> How can I force prog1.py to dynamically reload  module='func' ?
> I did not see any __reload__ function in the Library Reference ....

I already tried reload() but it didn't seem to work.
prog1.py becomes:

# prog1.py
class foo:
    def __init__(self, module):
        self.aa=__import__(module)
        reload(self.aa)
        chRef= getattr(self.aa, 'stHeading')
        chRef.run()

my_foo = foo('func')


And the results....

>>> import prog1
test

That works fine...!

Now change func.py, replacing
'setheading=SetHeading()'
with 'toto=SetHeading()'

>>> reload(prog1)
test
<module 'prog1' from 'C:\Python20\prog1.pyc'>

Why does this still work ? setheading does not exist anymore....
I had no clue... until i run this from the command line, where i got the
correct
results, and not from IDLE.

Could someone explain me why IDLE is messing everything up ?


Daniel Yoo replied:

You're not going to like this:  According to the reference documentation:

"When a module is reloaded, its dictionary (containing the module's global
variables) is retained.  Redefinition of names will override the old
definitions, so this is generally not a problem.  If the new version of a
module does not define a name that was defined by the old version, the old
definition remains."

This comes from:

    http://python.org/doc/current/lib/built-in-funcs.html

So that's why: setheading persists in the module definition if it hasn't
been overridden, so erasing it won't take affect until we restart Python.
This explains the weirdness that was happening before.  Try writing your
code to change the behavior of setheading, instead of removing it
entirely.

I wonder if there's a way of "clearing" out a module's definition, by
calling "del" on its dictionary of methods.  At the moment, I don't have
easy access to a Python interpreter (the horror!  the horror!), so I can't
confirm this yet.

Hope this helps!




From idot at vt.edu  Fri Mar  2 04:42:56 2001
From: idot at vt.edu (Carl Banks)
Date: Fri, 2 Mar 2001 04:42:56 -0500
Subject: gah! I hate the new string syntax
References: <mailman.983485923.1136.python-list@python.org>
Message-ID: <0vpn79.331.ln@127.0.0.1>

Sean 'Shaleh' Perry <shaleh at valinux.com> wrote:
> return ";".join(["%s=%s" % (k, params[k]) for k in params.keys()])
>
> every day python seems to be moving closer to the line noise aspect of
> coding.

I find it ok for methodlike functions such as split.

For unmethodlike functions such as join, I would use the old
string.join syntax (for regular strings, anyways).

"".join is just creepy.


-- 
CARL BANKS

"I like nothing better than a good challenge.
 This is not one of them."


From hgg9140 at cola.ca.boeing.com  Thu Mar 22 09:27:48 2001
From: hgg9140 at cola.ca.boeing.com (Harry George)
Date: Thu, 22 Mar 2001 14:27:48 GMT
Subject: learning python...
References: <3AB8F4E1.184DD9C@noaa.gov>
Message-ID: <xqxhf0l515n.fsf@cola.ca.boeing.com>

No matter what, get a copy of Beazley's Python Essential Reference.

If you already program in several languages (e.g., are familiar with
scripting, objects, method overrides, and lisp-ish map/apply/etc.),
you may be able to learn python from that book (plus looking at code
examples for good idioms).  Basically, you would be looking up python
syntax for things you already understand in principle.

If Essential Python leaves you guessing, the Quick Python Book might
fill in the gaps.

"Benjamin.Altman" <ben at .> writes:

> Hello.  From the perspective of someone who already has programming
> experience, would anyone know if O'reillys "Learning Python" book would
> be better than say Prentice Hall's "Core Python Programming"?  Or would
> it be better to go straight into something like "Programming Python"?
> 
> Thanks,
> Ben
> 

-- 
Harry George                E-mail:  harry.g.george at boeing.com
The Boeing Company          Renton:  (425) 237-6915
P. O. Box 3707  02-CA       Everett: (425) 266-3868
Seattle, WA 98124-2207      Page:    (425) 631-8803  


From jepler at inetnebr.com  Fri Mar 23 18:31:04 2001
From: jepler at inetnebr.com (Jeff Epler)
Date: Fri, 23 Mar 2001 23:31:04 GMT
Subject: help for newbie
References: <3AB8D031.2E54827E@rmsb.u-bordeaux2.fr>
Message-ID: <slrn9bnn5n.1qf.jepler@potty.housenet>

On Wed, 21 Mar 2001 17:00:49 +0100, Henri Valeins
 <Henri.Valeins at rmsb.u-bordeaux2.fr> wrote:
>Hi,
>
>I try to use python program named PLUCKER for retrieve web page on
>internet. When I try this program I have this message :
>
>Retrieved failed: 404 -- [Errno url error] unknown url type: 'http'
>
>My proxy var is correct 
>http_proxy=192.168.10.254:8080

I have always used
	http_proxy=http://machine:port/
This seems to work w/python, lynx, wget, and others.

Windows has a different way to set the proxy.

Jeff


From bezilla at hotmail.com  Thu Mar 15 23:28:18 2001
From: bezilla at hotmail.com (John E. Barham)
Date: Fri, 16 Mar 2001 04:28:18 GMT
Subject: Python & XSLT?
Message-ID: <CDgs6.46717$tP3.760880@news1.rdc1.bc.home.com>

We have a Web-based application that we're using Python for to do the CGI
code.  We generate the HTML interface by using hand hacked HTML templates
and printf-style substitution in Python.  It works, but isn't much fun and
the interface enhancement cycle is too long.

So I want to use XSLT.  Short of wrapping a C++ XSLT engine like Apache's
Xalan, is there any way to do XSLT from within Python?  As far as I can
tell, the existing Python XML tools don't go much above DOM/SAX style
processing...




From moshez at zadka.site.co.il  Fri Mar 30 11:19:01 2001
From: moshez at zadka.site.co.il (Moshe Zadka)
Date: Fri, 30 Mar 2001 18:19:01 +0200
Subject: Is there a bug tracker program written in Python?
In-Reply-To: <20010330.110051.1782436840.1258@scoot.zenplex.com>
References: <20010330.110051.1782436840.1258@scoot.zenplex.com>
Message-ID: <E14j1c5-0007NR-00@darjeeling>

On Fri, 30 Mar 2001 11:00:51 -0500, scott at zenplex.com wrote:
> I'm looking for a Bug Track system like bugzilla but written in Python.
> Does anybody know of such?  Just learned Python and would would like to
> use a system based on Python if at all possible.
 
Roundup.
Start searching here: www.lfw.org
-- 
"I'll be ex-DPL soon anyway so I'm        |LUKE: Is Perl better than Python?
looking for someplace else to grab power."|YODA: No...no... no. Quicker,
   -- Wichert Akkerman (on debian-private)|      easier, more seductive.
For public key, finger moshez at debian.org  |http://www.{python,debian,gnu}.org



From jcm at bigskytel.com  Tue Mar 13 08:50:03 2001
From: jcm at bigskytel.com (David Porter)
Date: Tue, 13 Mar 2001 06:50:03 -0700
Subject: Writing redirected e-mail to file
In-Reply-To: <3AAE18EF.6083CD74@geo.co.za>; from jacobus@geo.co.za on Tue, Mar 13, 2001 at 02:56:15PM +0200
References: <3AAE18EF.6083CD74@geo.co.za>
Message-ID: <20010313065003.A12638@bigskytel.com>

* Jacobus van der Merwe <jacobus at geo.co.za>:

> I use sendmail to redirect e-mails to a certain e-mail address to a
> script.
> This script must 'look' at the headers and body of the e-mail, and then
> write the e-mail body to a file.
> 
> I can't figure out exactly how to do this:
> 
> I use rfc822.Message to get the headers, but how do I get the body?

Use fp.read() in rfc822.Message. 

> Can one use mimetools.Message?   What does 'fp' in mimetools.Message
> parameters mean?  Once read with mimetools.Message, how do I write
> it to a textfile.

f = open('some_file', 'w')
f.write(it)


David



From bbhaydon at bigpond.com  Sat Mar 10 10:39:03 2001
From: bbhaydon at bigpond.com (Brett Haydon)
Date: Sun, 11 Mar 2001 02:39:03 +1100
Subject: architectural issues cot'd; which http server to embed?
References: <mailman.984150374.638.python-list@python.org>
Message-ID: <5Hrq6.12053$v5.51041@newsfeeds.bigpond.com>

Not quite sure what you've got happening there..

but look at www.modpython.org for an apache/python solution.

bbhaydon

"Eugene Leitl" <Eugene.Leitl at lrz.uni-muenchen.de> wrote in message
news:mailman.984150374.638.python-list at python.org...
>
> I haven't received much feedback on the last architectural query yet,
> as most of you people are probably on the conference. Still, here's
> goes.
>
> I've been fooling around with diverse http servers written in plain (i.e.
> containing no components written in other languages) Python (specifically,
> CPython. Perhaps threaded Python, if this doesn't limit my options in
> regard to portability).
>
> I have (in my head, at least) a multi-component (XML-talking server-client
> (sockets)) architecture driven via a web front (browser plugin or Java
> client) end by user queries.  This will work transparently, whether run
> locally, or on another machine on the network. I don't need the
> multicomponent part so badly right now, but this is needed for a future
> framework, of which this will be an integrated component.
>
> It has to be able to run on every target which has GNU tools and TCP/IP
> running on it. In practice, this will mean diverse flavors of *nix, and
> Redmondware. The ideal delivery form would be a bunch of binaries bundled
> in an archive, and a (frozen) GUI-based (with a shell fallback option for
> the hard cases) installer written in Python. I can't install my own Python
> on the client machine, and there's not enough time to do it properly (I
> have certainly found enough interesting options), so I'm going to tolerate
> some amount of bloat in the distribution, at least for time being.
>
> Right now I'm researching which http server to embed (using freeze) into
> the front-end. The performance requirements are quite negligable, but it
> should be better non-blocking (Medusa alone won't do, will Medusa/bobo
> combo? Zope is way too fat, CGIHTTPServer.py would do nicely, but it
> doesn't seem to be able to support server-side Python embedding into HTML
> when being frozen), as queries can take a while, and it might see more
> than one user, or several queries from the same user. Otoh, I'd rather not
> call multi-MByte frozen binaries via cgi-bin. It would be nice to have a
> solution which allows one to embed Python code into HTML, which can get
> interpreted by the Python instance which is present in the webserver, or
> Python packaged in a shared library.
>
> Another alternative I haven't researched yet is extending Python with a
> minimal, well-debugged off-shelf webserver written in C. This would be
> probably more palatable with passing Python scripts from cgi-bin and/or
> Python embedding into server-side HTML. Discussions and pointers to code
> are very welcome.
>
> Resistancy to memory leaks is another design facet. If I can solve the
> problem of long-running Python instances slowly draining memory
> (considering my limited amount of clue a very real factor) by paying with
> startup overhead by launching them at every query, that appears tolerable.
> Even if it involves calling multi-MByte monster binaries, I'd have to bite
> the bullet (I'd rather not, of course). How bad is the memory draining
> issue, given debugging tool and Python2.0 cycle-elimination during gc?
> Should I contemplate using a better gc?
>
> Does this make sense, or I am missing some gotcha, or obvious alternative?
> Sorry if I sound confused, I probably am. This project is rather ambitious
> for my current skills, but I'd rather not tackle it in C. I'd certainly be
> unable to handle this in C on the time schedule (2..4 months), so there's
> not much choice, anyway.
>
> TIA,
> -- Eugene
>
>




From db3l at fitlinxx.com  Fri Mar  9 19:43:04 2001
From: db3l at fitlinxx.com (David Bolen)
Date: 09 Mar 2001 19:43:04 -0500
Subject: Name of the Script
References: <20010304.185354.2114937732.8277@localhost.localdomain> <NJCo6.6720$qv3.2834377@nnrp5-w.sbc.net> <FpmjOjvwe5bm0Bg09LDTMPhh5vji@4ax.com> <Pine.LNX.4.21.0103051715150.18263-100000@kev.ioc.ee> <cbKjOqPRwNza00jOm7A1HBDkRYj7@4ax.com> <upuft9z0s.fsf@ctwd0143.fitlinxx.com> <uFCnOk=ag9iJ0BBH3CPw9xrMiTRB@4ax.com> <ug0gnk548.fsf@ctwd0143.fitlinxx.com> <p6uoOgwTJk=zxrdMpt4ziF4CK8Rq@4ax.com>
Message-ID: <ubsra8ndz.fsf@ctwd0143.fitlinxx.com>

Paul Moore <paul.moore at uk.origin-it.com> writes:

> How _do_ applications locate their data files without hard-coding directory
> names, or making assumptions of the current directory? (Actually, that's a
> better statement of my issue...)

I can't speak for everyone, but for myself, data files (and log files,
and other output files) in my scripts are virtually always a
configurable item.  This would include their location (or base
directory for an assumed hierarchy) and often name, except in some
cases like log files where an automatic naming scheme or suffixes are
used.

For maximum flexibility, the value is generally set in a configuration
file (which may just be a Python module that is imported), or via
command line options.

If you want to supply default values for a user, it can just be
deployed as a default configuration file that gets installed with your
application.  If you go the Python module route for a configuration
file, then it can contain executable code as well, so you can make
reference to sys.argv or __file__ if you'd really like to.  And since
you'd just import the configuration file, your only concern for its
location is that it be on the Python sys.path (which can be augmented
in various ways when you install your scripts, including .pth files).

Since the configuration file would generally be loaded very early by
any top level script in the system, the possibility of a switch in the
current directory before you can process sys.argv can be prevented.

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


From digitig at cix.co.uk  Mon Mar 19 16:32:00 2001
From: digitig at cix.co.uk (Tim Rowe)
Date: Mon, 19 Mar 2001 21:32 +0000 (GMT Standard Time)
Subject: Why "from __future__" stinks; a counter-offer
References: <mailman.984957623.24638.python-list@python.org>
Message-ID: <memo.20010319213202.8771A@digitig.compulink.co.uk>

In article <mailman.984957623.24638.python-list at python.org>, 
mcherm_python at yahoo.com (Michael Chermside) wrote:

> Hear hear!! I agree!

I don't really want to get into a "me too" thread, but I agree too, and 
ill thought-out stuff like this threatens to kill the language. For 
Python's sake folks, take the time to work out a way of doing it that 
isn't an ugly wart on a beautiful language :-(


From phrxy at csv.warwick.ac.uk  Thu Mar 29 13:41:15 2001
From: phrxy at csv.warwick.ac.uk (John J. Lee)
Date: Thu, 29 Mar 2001 19:41:15 +0100
Subject: Linux GPS program
In-Reply-To: <slrn9c1p9t.d3g.jra@dorothy.msas.net>
References: <3AC01637.4EF3A6F2@crynwr.com> <slrn9c17k4.cfc.jra@dorothy.msas.net> <3AC0D915.83C2C0C3@crynwr.com> <slrn9c1p9t.d3g.jra@dorothy.msas.net>
Message-ID: <Pine.SOL.4.30.0103291940140.13267-100000@primrose.csv.warwick.ac.uk>

On Tue, 27 Mar 2001, Jay R. Ashworth wrote:

> This one time, in band camp,
>   Russell Nelson <nelson at crynwr.com> wrote:
[...]
> > No.  I used pygtk, which is a set of Python bindings to the gtk
[...]

Okay, he does know what it (free software) means, I was wrong.


John



From pf at artcom-gmbh.de  Fri Mar 30 13:57:35 2001
From: pf at artcom-gmbh.de (Peter Funk)
Date: Fri, 30 Mar 2001 20:57:35 +0200 (MEST)
Subject: Speedup hint for users of viewcvs package
Message-ID: <m14j45Y-000CnEC@artcom0.artcom-gmbh.de>

Greg Stein has written a very useful Python application called viewcvs
which is availble at http://www.lyra.org/viewcvs/

Greg: if you read this: Thank you.  The following note applies to 
viewcvs version 0.6

I've installed it here in our company, because compared to cvsweb viewcvs
offers syntax colorized views of the source codes. 

But people complained, that viewcvs was a lot slower compared to cvsweb.cgi.
Fortunately this was really easy to fix:
 (1) go into the viewcvs installation directory and there into the
     sub directory cgi.  you need write permissions there. Example:
        # cd /usr/local/viewcvs-dev/cgi
 (2) Rename the file viewcvs.cgi to viewcvs.py:
        # mv viewcvs.cgi viewcvs.py
 (3) fire up the interactive Python interpreter, type 'import viewcvs'
     and leave it with Ctrl-D.
 (4) fix the permissions of the byte code file 'viewcvs.pyc' generated by
     step (3) to make it world readable.
 (5) Place the following snippet into a file viewcvs.cgi there:
	 #!/usr/bin/python
	 import sys, os
	 sys.path.insert(0, os.path.dirname(sys.argv[0]))
	 import viewcvs
     you may have to fix the path to your Python interpreter.
 (6) Try it from your web browser on your and you should be able to 
     observe speed pretty similar to cvsweb.

Explanation: viewcvs.cgi is a ~ 68 kB source file.  After applying the
procedure outlined above, this is compiled into a byte code file.
So there is no longer recompilation needed on each cgi request.

Regards, Peter
-- 
Peter Funk, Oldenburger Str.86, D-27777 Ganderkesee, Germany, Fax:+49 4222950260
office: +49 421 20419-0 (ArtCom GmbH, Grazer Str.8, D-28359 Bremen, Germany)




From paul at fxtech.com  Mon Mar 12 10:31:41 2001
From: paul at fxtech.com (Paul Miller)
Date: Mon, 12 Mar 2001 09:31:41 -0600
Subject: Einstein's Riddle
References: <mailman.984369618.12493.python-list@python.org>
Message-ID: <3AACEBDD.3362AEAE@fxtech.com>

> You could cover both cheeks by answering "if anyone owns a fish, it must be
> the German".  Then again, you may prefer to complain that "on the left of"
> doesn't necessarily imply adjacent to; or that the problem statement never
> says that a person who "lives in" a house is also its "owner"; or etc etc.

Well said. The problem is much more interesting given the assuption that
one of the owners actually has the fish. Call me dumb if you wish for
making this mental leap of logic and diving in, rather than sitting
there with arms crossed claiming that nobody said one of these people
actually owned a fish.


From cce at clarkevans.com  Fri Mar 23 15:55:15 2001
From: cce at clarkevans.com (Clark C. Evans)
Date: Fri, 23 Mar 2001 15:55:15 -0500 (EST)
Subject: PEP 245: Python interfaces
In-Reply-To: <Pine.LNX.4.32.0103231156000.26302-100000@localhost.localdomain>
Message-ID: <Pine.LNX.4.21.0103231540000.3799-100000@clarkevans.com>

On Fri, 23 Mar 2001, Michel Pelletier wrote:
> >
> >          class X:
> >             def func(self,z): pure
> >
> >       Then, if any method in a class was marked
> >       "pure", the class could not be instantiated.
>
> The reason behind this is that classes are shared implementation,
> they are not an interface.  Any interface of a class is an implied, 
> assumed thing.   Interface assertions make that more concrete. 

Yes, but a class full of "pure" methods is an interface, no?
Further, you can be more granular with this approach.  For instance,
you may have an interface with four functions.  The implementation
of three functions may be completely optional or trivial.  With
the class approach, you could label the one critical method as
"pure" and then provide a reasonable, default implementation 
for the other methods.

> Classes could in some way, be extened to have an inline "interface"
> but that is mixing your interface with your implementation.

I'm not convinced that they are entirely seperate ideas.

> How can you reuse that? 

Just like you use classes today... look at the PyXML stuff,
specifically the Sax interfaces.  Providing a default, or
trivial implemenation for optional methods is a common 
technique in C++ and very valueable.  

In Java you are forced to have two constructs, an interface,
and then a "base handler".  Having the interfae and the
base implementation in two places just makes more places
where errors can pop-up.

> How can you have mulitple implementations?

Pretty easily, see the PyXML stuff, specifically
the SAX 'interfaces.

> >    f) If a new type is added, why not go further?
> 
> Because I don't want to risk sliding into the absurd. ;)
>

Yes, but your new interface type isn't providing much
that can't already be done with classes and regular 
inheritance.
 
> >       Although I don't like the idea of argument
> >       types in classes... they could very well
> >       make sence for interfaces.  This could help
> >       to differentiate the syntax, perhaps:
> >
> >         interface Y:
> >            decl func( number as types.IntType,
> >                       string as types.StringType,
> >                       any    as any,
> >                       cls    as IAnotherInterface ):
> >                "This function does such and such"
> 
> These goals are reasonable.  I don't like the syntax myself.

Something like this would make your proposal as offering
something significant which classes don't already offer.

> >    h)  I don't like the string for pre/post conditions.
> 
> They're just python expressions like that used with the 'assert' keyword.
> What's not to like about them?

For starters my syntax highlighter doesn't work with them.
Also, nested blocks and such can't be used.
If you have a new syntax... why do they have to be strings?
 
> >        They should have special syntax, perhaps as
> >        these could be implemented as "local methods?"
> >
> >          interface Y:
> >              decl func(...)
> >                  def pre(self):
> >                     if ... then raise PreConditionFailure
> >                  def post(self): pass
> >                  def test(self): pass
> 
> What if you interface wants to have pre post and test methods that have
> nothing to do with type enforcement?

I don't think the above prevents this, I was just looking
at a more "pythonic" syntax.  Checks are *code*, IMHO, 
they don't belong in strings.  The syntax above is just
an attempt to make it more code like.  

Best,

;) Clark







From costas at springmail.com  Mon Mar 12 16:34:58 2001
From: costas at springmail.com (costas at springmail.com)
Date: Mon, 12 Mar 2001 21:34:58 GMT
Subject: This math scares me
Message-ID: <3aad4062.1290021@News.CIS.DFN.DE>

Ok, I can see maybe division having problems. But why does addition of
the two numbers below:

5.01+5.54

give me this?

10.550000000000001

I hope its some setting that can take care of this. I am using Python
under Windows 98.

TIA

Costas


From tim.one at home.com  Sat Mar  3 15:41:18 2001
From: tim.one at home.com (Tim Peters)
Date: Sat, 3 Mar 2001 15:41:18 -0500
Subject: Bug in Py_Finalize() ?
In-Reply-To: <3aa10f5b.459867@news.newsguy.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCAELMJDAA.tim.one@home.com>

[Tim]
> You didn't give a complete program so it's impossible for me to
> say whether your problem is fixed.  I did put your loop in an
> extension module and boosted the loop limit to 1000.  No problem
> under Windows Python 2.1b1, under either release or debug builds.

[Owen F. Ransen]
> That was the complete program, I had narrowed it down that
> far to find out what was crashing.

It was an isolated loop (hence not a "complete program").  Whatever problem
you were having may or may not have had to do with what you didn't post.  For
example, if you returned to the Python interpreter after doing Py_Finalize(),
then of course you're going to get a crash.  But I can't guess that in the
absence of a complete program.

> I downloaded the Windows installer for 2.1b1 but there does not
> appear to be any debug libraries...

Sorry, we never ship debug stuff with the Windows installer, because it's yet
another megabyte of stuff to download that the vast majority of Windows users
have no use for.  People who want debug stuff are by definition developers,
and so are strongly encouraged to build Python from source (ain't hard!).

Nevertheless, I just uploaded all the 2.1b1 debug stuff to SourceForge:

    https://sourceforge.net/project/showfiles.php?group_id=5470

Get file Python-2.1b1-Debug.zip.

Does your program run OK under the release build we did ship?




From m.wilson at bibliocraft.com  Mon Mar  5 19:16:42 2001
From: m.wilson at bibliocraft.com (Mark Wilson)
Date: Tue, 6 Mar 2001 00:16:42 -0000
Subject: MySQL-python build griefs
Message-ID: <02df01c0a5d3$079ebd20$4a824ac3@mwdell1>

Will,

this may well have _nothing_ to do with your problem however, I experienced
similar grief with distutils on a similar box. In my case it turned out I'd
uploaded my Python/Include directory to the server using binary ftp transfer
(from a Windows machine). Duhhhhhh. All worked fine once I realised my mistake,
but I was getting similar kinds of errors.

Will Ware:
>I'm trying to build MySQL-python on a Red Hat 6.2 box. In order
>not to confuse the Red Hat installation, which depends heavily
>on Python, I've left Python at 1.5.2. I have Distutils 1.0.1
>installed. The version of MySQL-python is 0.3.0. To get things
>to work at all, I needed to build my own copy of Python-1.5.2/*
>to populate the /usr/include/python1.5 tree, running "configure"
>to produce a suitable "config.h". Now I'm getting an error on
>"python setup.py build" that looks like this:
> [snip]
>TypeError: illegal argument type for built-in operation
>
>Am I using the wrong color duct tape here? Or just not enough
>bubblegum?

Maybe Ascii coloured duct tape is what you need?

Bestoluck

Mark





From ct at gocept.com  Thu Mar 22 13:31:30 2001
From: ct at gocept.com (Christian Theune)
Date: Thu, 22 Mar 2001 19:31:30 +0100
Subject: python2 & curses
Message-ID: <99dge2$jpa$04$1@news.t-online.com>

Is it possible that the python-curses binding doesn't work for python
2.0?

i'm getting following message:

Traceback (most recent call last):
  File "./tabview.py", line 31, in ?
    import curses, re, string
  File "/usr/local/lib/python2.0/curses/__init__.py", line 15, in ?
ImportError: No module named _curses

please help me, thanks


From guido at digicool.com  Sat Mar  3 14:10:35 2001
From: guido at digicool.com (Guido van Rossum)
Date: Sat, 03 Mar 2001 14:10:35 -0500
Subject: Bug fix releases
In-Reply-To: Your message of "Sat, 03 Mar 2001 13:21:44 EST."
             <200103031821.NAA24060@panix3.panix.com> 
References: <LNBBLJKPBEHFEDALKOLCMEEPJBAA.tim.one@home.com> <mailman.982897324.9109.python-list@python.org>  
            <200103031821.NAA24060@panix3.panix.com> 
Message-ID: <200103031910.OAA21663@cj20424-a.reston1.va.home.com>

Aahz writes:
> [posted to c.l.py with cc to python-dev]
> 
> [I apologize for the delay in posting this, but it's taken me some time
> to get my thoughts straight.  I hope that by posting this right before
> IPC9 there'll be a chance to get some good discussion in person.]

Excellent.  Even in time for me to mention this in my keynote! :-)

> In article <mailman.982897324.9109.python-list at python.org>,
> Guido van Rossum  <guido at digicool.com> wrote:
> >
> >We have clearly underestimated how much code the nested scopes would
> >break, but more importantly we have underestimated how much value our
> >community places on stability.  
> 
> I think so, yes, on that latter clause.  I think perhaps it wasn't clear
> at the time, but I believe that much of the yelling over "print >>" was
> less over the specific design but because it came so close to the
> release of 2.0 that there wasn't *time* to sit down and talk things
> over rationally.

In my eyes the issues are somewhat different: "print >>" couldn't
possibly break existing code; nested scopes clearly do, and that's why
we decided to use the __future__ statement.

But I understand that you're saying that the community has grown so
conservative that it can't stand new features even if they *are* fully
backwards compatible.

I wonder, does that extend to new library modules?  Is there also
resistance against the growth there?  I don't think so -- if anything,
people are clamoring for more stuff to become standard (while at the
same time I feel some pressure to cut dead wood, like the old SGI
multimedia modules).

So that relegates us at PythonLabs to a number of things: coding new
modules (boring), or trying to improve performance of the virtual
machine (equally boring, and difficult to boot), or fixing bugs (did I
mention boring? :-).

So what can we do for fun?  (Besides redesigning Zope, which is lots
of fun, but runs into the same issues.)

> As I see it, there's a natural tension between between adding features
> and delivering bug fixes.  Particularly because of Microsoft, I think
> that upgrading to a feature release to get bug fixes has become anathema
> to a lot of people, and I think that seeing features added or changed
> close to a release reminds people too much of the Microsoft upgrade
> treadmill.

Actually, I though that the Microsoft way these days was to smuggle
entire new subsystems into bugfix releases.  What else are "Service
Packs" for? :-)

> >So here's the deal: we'll make nested scopes an optional feature in
> >2.1, default off, selectable on a per-module basis using a mechanism
> >that's slightly hackish but is guaranteed to be safe.  (See below.)
> >
> >At the same time, we'll augment the compiler to detect all situations
> >that will break when nested scopes are introduced in the future, and
> >issue warnings for those situations.  The idea here is that warnings
> >don't break code, but encourage folks to fix their code so we can
> >introduce nested scopes in 2.2.  Given our current pace of releases
> >that should be about 6 months warning.
> 
> As some other people have pointed out, six months is actually a rather
> short cycle when it comes to delivering enterprise applications across
> hundreds or thousands of machines.  Notice how many people have said
> they haven't upgraded from 1.5.2 yet!  Contrast that with the quickness
> of the 1.5.1 to 1.5.2 upgrade.

Clearly, we're taking this into account.  If we believed you all
upgraded the day we announced a new release, we'd be even more
conservative with adding new features (at least features introducing
incompatibilities).

> I believe that "from __future__" is a good idea, but it is at best a
> bandage over the feature/bug fix tension.  I think that the real issue
> is that in the world of core Python development, release N is always a
> future release, never the current release; as soon as release N goes out
> the door into production, it immediately becomes release N-1 and forever
> dead to development
> 
> Rather than change that mindset directly, I propose that we move to a
> forked model of development.  During the development cycle for any given
> release, release (N-1).1 is also a live target -- but strictly for bug
> fixes.  I suggest that shortly after the release for Na1, there should
> also be a release for (N-1).1b1; shortly after the release of Nb1, there
> would be (N-1).1b2.  And (N-1).1 would be released shortly after N.

Your math at first confused the hell out of me, but I see what you
mean.  You want us to spend time on 2.0.1 which should be a bugfix
release for 2.0, while at the same time working on 2.1 which is a new
feature release.

Guess what -- I am secretly (together with the PSU) planning a 2.0.1
release.  I'm waiting however for obtaining the ownership rights to
the 2.0 release, so we can fix the GPL incompatibility issue in the
license at the same time.  (See the 1.6.1 release.)  I promise that
2.0.1, unlike 1.6.1, will contain more than a token set of real
bugfixes.  Hey, we already have a branch in the CVS tree for 2.0.1
development!  (Tagged "release20-maint".)

We could use some checkins on that branch though.

> This means that each feature-based release gets one-and-only-one pure
> bugfix release.  I think this will do much to promote the idea of Python
> as a stable platform for application development.

Anything we can do to please those republicans! :-)

> There are a number of ways I can see this working, including setting up
> a separate project at SourceForge (e.g. pythonpatch.sourceforge.net).
> But I don't think this will work at all unless the PythonLabs team is at
> least willing to "bless" the bugfix release.  Uncle Timmy has been known
> to make snarky comments about forever maintaining 1.5.2; I think this is
> a usable compromise that will take relatively little effort to keep
> going once it's set up.

With the CVS branch it's *trivial* to keep it going.  We should have
learned from the Tcl folks, they've had 8.NpM releases for a while.

> I think one key advantage of this approach is that a lot more people
> will be willing to try out a beta of a strict bugfix release, so the
> release N bugfixes will get more testing than they otherwise would.

Wait a minute!  Now you're making it too complicated.  Betas of bugfix
releases?  That seems to defeat the purpose.  What kind of
beta-testing does a pure bugfix release need?  Presumably each
individual bugfix applied has already been tested before it is checked
in!  Or are you thinking of adding small new features to a "bugfix"
release?  That ought to be a no-no according to your own philosophy!

> If there's interest in this idea, I'll write it up as a formal PEP.

Please do.

> It's too late for my proposed model to work during the 2.1 release
> cycle, but I think it would be an awfully nice gesture to the community
> to take a month off after 2.1 to create 2.0.1, before going on to 2.2.

It's not too late, as I mentioned.  We'll also do this for 2.1.

> BTW, you should probably blame Fredrik for this idea.  ;-)  If he had
> skipped providing 1.5.2 and 2.0 versions of sre, I probably wouldn't
> have considered this a workable idea.  I was just thinking that it was
> too bad there wasn't a packaged version of 2.0 containing the new sre,
> and that snowballed into this.

So the new (2.1) sre code should be merged back into 2.0.1, right?
Fredrik, go ahead!  We'll start planning for the 2.0.1 release right
after we're back from the conference.

BTW, See you at the conference!

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



From tim.one at home.com  Mon Mar 19 20:27:34 2001
From: tim.one at home.com (Tim Peters)
Date: Mon, 19 Mar 2001 20:27:34 -0500
Subject: Why "from __future__" stinks; a counter-offer
In-Reply-To: <jwbnews-3793B2.08394319032001@corp.supernews.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCKEMCJGAA.tim.one@home.com>

[John W. Baxter]
> For the future, the time to complain about PEP contents is when the PEP
> is posted, not after it has been posted, discussed, implemented, and the
> result distributed.
>
> For the Python 2.1 batch, late rants seem excusable, since the process
> is new.

The future_statement PEP also came very late in the release process:  like
the "Import on Case-Insensitive Filesystems" PEP, it was written to address a
problem that didn't become apparent before an alpha release was already out
the door, and was darned-near a retroactive PEP.  I certainly hope that, in
the future, all PEPs get more lead time than these did.

[Tim Rowe]
> But there's part of the problem; egoless PEP writing is as hard as
> egoless programming, and having come up with a solution the author of
> that solution is bound to have some emotional involvement in that
> solution.

Sure, although in this particular case my emotional involvement doesn't
extend much deeper than the nesting of Python 2.0's namespaces <wink>.

> There have been (rightly IMHO) howls of protest about __future__,

Howls are rarely productive.

> but Tim has said it's ok for him, routinely dismisses all protests,
> presumably because he genuinely doesn't understand what our problem is,

Fine, so I'm an idiot:  write a PEP that explains your version of the
self-evident truths for the benefit of those I haven't paid off <wink>.

> and he tells us it's going in anyway.

That part is just spelling out a consequence of how the *process* works:
without an alternative, what do you expect?  That it *won't* go in?  Get
real.  2.1 *needs* a mechanism to deal with the incompatible nested-scopes
change (or at least Guido was convinced by those who said it did).  So we do
this; wait for an alternative PEP nobody shows any sign of intending to write
(I sure asked Martin to write up his "directive" alternative often enough,
and that was the only alternative I saw that was close to being thought out);
throw away the nested-scopes change; or delay 2.1 indefinitely.  Deciding
among those wasn't my call, but I believe that going with the
future_statement was the best of those options.  More importantly, Guido
believed that.

> Can you see why we're not all convinced that the PEP process is any
> help to us?

It isn't a goal of the PEP process that everybody get what they want, or even
that everybody get something they'll accept under protest.  ANSI and ISO
committees work that way (via (often grudging) consensus), but Python does
not.  The best way to oppose a PEP is to write a competing PEP.  The entire
process would work much better that way:  Guido has no qualms about picking
winners and losers, and if you think your views are being ignored, belittled,
or misunderstood, you're encouraged to write them up yourself.  Writing a PEP
is as easy as we could make it -- although, granted, nothing will ever be as
easy as posting complaints to Usenet <0.7 wink>.

win-lose-or-draw-they're-all-possible-outcomes-ly y'rs  - tim




From chrisa at ASPATECH.COM.BR  Thu Mar 15 14:24:46 2001
From: chrisa at ASPATECH.COM.BR (Chris Richard Adams)
Date: Thu, 15 Mar 2001 16:24:46 -0300
Subject: Off-python-topic:  Cron jobs on linux...help
Message-ID: <BD674CDC0A68C642B60038986FBB56DF0138C4@mailsrv.ASPATECH.COM.BR>

I'd like to ask two things about the cron tool that comes with Linux -
have RedHat 6.2.
 
1.) I have a script that runs fine when I am logged in a the user who
owns it.  I have env variables that are needed and these are loaded in
the .bash_profile as that user logs in.  As that user, I set a cron job
putting the pathname of the script.  It runs, but fails because it
appears the  environment vars are not set.  I can work around it by
putting all the vars in the script of course, but what might be the
problem?
 
2.) I want the script to run each 20 minutes.  With only the 5 time
parameters to set, I can't do this?
 
Thanks!



From jdukarm at hydracen.com  Tue Mar  6 15:30:44 2001
From: jdukarm at hydracen.com (J.J. Dukarm)
Date: Tue, 06 Mar 2001 12:30:44 -0800
Subject: Can Python replace Visual Basic? Should it?
Message-ID: <3.0.1.32.20010306123044.00685d14@pop.islandnet.com>

Python as a replacement for Visual Basic?  This is an important question
for me, because I am in the process of "betting the farm" on it. The
software product that pays all of my grocery bills was written in VB, and I
am rewriting it in Python.

I have been programming professionally since 1965, and I have done a lot of
different things in a lot of languages on everything from microprocessors
to massively parallel computers. Incidentally, I was one of the first
commercial BASIC programmers on the planet, in 1973.

Six years ago I developed an industrial software product - diagnostic
software - using Visual Basic 3 for the user interface and database
machinery, with several DLL's written in C for the mathematical stuff.
After a short period of testing and debugging, the DLL's reached a stable
state and have not needed revising or recompilation in at least five years.
The VB portion of the app went together quickly, and initially I was very
happy because I could have a nice-looking Windows user interface with all
kinds of graphical frills without learning any dirty secrets about "device
contexts" and other mysteries of the Windows API.

Unfortunately, VB 3 itself turned out to be so buggy that I ended up
learning all the Win API stuff anyway while devising work-arounds for
show-stopping bugs that cropped up in VB (not my app), delaying the release
of my product for 6 months and almost bankrupting me.  Since then, the VB
component of the product has been difficult to maintain and improve. Visual
Basic is not a language - it is a sequence of mutually incompatible
languages, and I discovered the hard way that it was more practical to keep
the VB 3 code working in new versions of Windows than to learn each new VB,
rewrite and re-debug my code, and then debug the new VB. 

In 1998, I put the renewal form for my Universal subscription to the
Microsoft Developers Network in the shredder because it was getting too
difficult and expensive to continue using MS development tools. By then, it
had also dawned on me that it was dangerous for my product's development
cycle and delivery environment to be completely under the control of
Microsoft.

I was gearing up to switch to Java, which had its own stability and
portability problems in 1998-99, when I discovered Python.  Although some
improvements have been introduced in Python since I started with 1.5.2, no
changes in Python have ever broken any of my code. I have never had to work
around a bug in Python. There are perfectly good tools (PythonWin,
PythonWorks, wxPython, wxDesigner, Webware, Reportlab are my favorites),
excellent books, and excellent support by technically-oriented people who
do not have a marketing department re-inventing the terminology for them. 

The three biggest problems for me were: (a) finding a suitable GUI (I chose
wxPython); (b) finding a robust way to communicate with commercial database
products (I am experimenting with various ways of doing ODBC, but am still
nervous); and (c) packaging the app for distribution (just started figuring
it out).

Some things in VB (especially Windows-oriented things) are extremely easy,
where you might or might not have to do a lot of digging and do-it-yourself
in Python.  On the other hand, you never know whether the things you do in
VB will work as expected - or keep working after you ship them, or after
Billy G. changes some system DLL's. The great advantages that Python has
over VB, as I see it, are stability, reliability, and portability. For
commercial software developers, those are incredibly important.

I could go on ranting, but you will start to see the practical answer to
this question in a few months. If I am still participating in this
discussion group and haven't gone bust after the all-Python version of my
product comes out, the answer will definitely be "yes".

Jim Dukarm
DELTA-X RESEARCH
Victoria BC Canada




From robin900 at yahoo.com  Fri Mar 30 14:41:43 2001
From: robin900 at yahoo.com (Robin Thomas)
Date: Fri, 30 Mar 2001 14:41:43 -0500
Subject: Why not an __assign__ method?
In-Reply-To: <5.0.2.1.0.20010330125432.0216c4b0@mail.inet.com.br>
Message-ID: <4.3.1.2.20010330115856.0256bb00@exchange.starmedia.net>

At 01:26 PM 3/30/01 -0300, Carlos Alberto Reis Ribeiro wrote:

>The question is: Why not have a __assign__ method, that gets called on 
>assignment? It should be called on the *right side* of the expression (at 
>the left side it does not make sense - I leave the proof as an exercise 
>for the reader <wink>). The return value would be then assigned to the 
>left side of the expression.

You should really get the source to CPython and start reading it.

>z = a + b + c

The "no changes to Python" implementation of your example is:

z = a + b
z += c

Think about operations, not methods.


>Then the expression is broken in steps like this (not exact syntax, but 
>enough to give you the idea):
>
>i1 = a.__add__(b)
>i2 = i1.__add__(c)
>return i2

Read the source code. File Python/ceval.c, grep for "ADD". Thrill at the 
majestic expanse of the switch statement on Python opcode values.

The "exact syntax" (sic) of the addition operation is different than above, 
and different in a way that undermines the idea of "method calls" that you 
are illustrating in order to support a new __assign__ method.

>There are two instantiations of temporary arrays, where only one would be 
>enough. With big arrays the extra temp copy is quite an issue. I intend to 
>do some image manipulatin, and that surely will help a lot.
>
>The mechanism that I have in mind is:

...something that has nothing to do with your proposed __assign__ method.

>1) Set a "temp" flag for objects created in the middle of an expression...

...objects that support the in-place version of the regular numeric 
operator represented by the opcode occurring a few bytes forward in the 
bytecode stream? Think about what "middle of an expression" means.

>2) If a binary operation is called upon a temporary object, use it to 
>store the result (of course, if the semantics allows).

See above.

>3) Reset "temp" on assign.

Think about what your "temp flag" would do under the current behavior of 
threads in Python.

Think about how you want to turn this:

z = a + b + c

into the syntax-illegal but perhaps (and perhaps not) bytecode-legal:

z = ((a + b) += c)

I say "perhaps not" because I'm not going to look at the source to 
determine whether this is feasible. You do it. You will discover why a + b 
+= c is not allowed, and whether that can and should be changed.

Think about why you consider these intermediate objects as "temporary", 
whether Python currently treats them as "temporary", and if so, how they 
are discarded.

Use the dis module for the standard library. Look at the bytecode for:

a + b + c # expression

versus

z = a + b + c # statement

versus

z = a + b
z += c  # two statements

versus

z = a + b + c % (d * e) / f ^ (g or 42)

I hope this will keep you busy for a while! :)


--
robin900 at yahoo.com


_________________________________________________________
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com




From jwbnews at scandaroon.com  Mon Mar 19 11:39:44 2001
From: jwbnews at scandaroon.com (John W. Baxter)
Date: Mon, 19 Mar 2001 08:39:44 -0800
Subject: Why "from __future__" stinks; a counter-offer
References: <mailman.984975803.9006.python-list@python.org>
Message-ID: <jwbnews-3793B2.08394319032001@corp.supernews.com>

For the future, the time to complain about PEP contents is when the PEP 
is posted, not after it has been posted, discussed, implemented, and the 
result distributed.

For the Python 2.1 batch, late rants seem excusable, since the process 
is new.

That said,
from __future__ import nested_scopes
works fine for me.

Which is just as well, because it means I can contentedly move on.

  -----
Those who "do" PEPs will have noted a loud silence from me.  That means 
three things:

1.  I'm lazy (but you knew that already)

2.  I trust you and Guido to do the right thing, mostly.  And I can deal 
with the exceptions to that "mostly".

3.  I still have to come to grips with SourceForge.

   --John (getting back into weight training in preparation for 
Programming Python 2nd Edition)

-- 
John W. Baxter   Port Ludlow, WA USA  jwbnews at scandaroon.com


From smarsh at hotmail.com  Fri Mar 16 18:33:01 2001
From: smarsh at hotmail.com (scott)
Date: Fri, 16 Mar 2001 23:33:01 GMT
Subject: Newbie confusion about 'return'
Message-ID: <3AB2A2BE.A467DE00@hotmail.com>

I'm mystified. The same code in IDLE (or at the interactive prompt)
gives a different result compared to entering it into a file and running
it. I'm using Python 2.0 from python .org on NT4. I checked on Solaris 8
with python 2.0 (interactive vs running the file) and got the same
result. The code is from "How to think like a computer scientist Chapter
7"
-----------------------------------
IDLE:

>>> def find(str, ch):
              index = 0
              while index < len(str):
                      if str[index] == ch:
                             return index
                      index = index + 1
              return -1

>>> find ("abcde", "d")
3 (in blue indicating stdout)

The above is what I expected.
------------------------------------

Running python on the file (find.py):

def find(str, ch):
      index = 0
      while index < len(str):
             if str[index] == ch:
                    return index
             index = index + 1
      return -1

find ("abcde", "d")

I ran it as: $ python find.py
*Nothing* was returned (just gave me the command prompt again), instead
of the '3' that I expected.
---------------------------------------

Why the difference? I thought one of the points of the interactive mode
was to test code to see if it works before putting it in a file...
If I add a couple of print statements in the file, I get the expected
result (3) from running 'python find.py':

def find(str, ch):
      index = 0
      while index < len(str):
             if str[index] == ch:
                   print index
                   return
             index = index + 1
      print -1
      return

find ("abcde", "d")

Thanks in advance.




From chrisw at nipltd.com  Tue Mar 20 04:07:02 2001
From: chrisw at nipltd.com (Chris Withers)
Date: Tue, 20 Mar 2001 09:07:02 +0000
Subject: Setting file attributes on file creation.
Message-ID: <3AB71DB6.9ABC7E8F@nipltd.com>

Hi,

I've got some code which unpacks a ZIP file created using the zipfile module:

      f = open(dstpath,'wb')
      f.write(zipfile.read(zipname))
      f.close()

Is there any way I can set the date and time of the file as I unpack it? 
Something like shutil.copy2, if you see what I mean.

Thanks for any help :-)

Chris



From topmind at technologist.com  Sat Mar 24 02:38:47 2001
From: topmind at technologist.com (topmind)
Date: Sat, 24 Mar 2001 07:38:47 GMT
Subject: Collection interfaces
References: <3a90fa97.0@news1.mweb.co.za>   <MPG.1504d14563dce6499896af@news.earthlink.net> <MPG.1505a76831c34b749896b3@news.earthlink.net> <MPG.1506007591cf818b9896bc@news.earthlink.net> <MPG.1506265ebf6d934d9896be@news.earthlink.net> <97ibmc0k47@news1.newsguy.com> <MPG.1506cf3293b094b49896c1@news.earthlink.net> <97m0c7080c@news2.newsguy.com> s.earthlink.net> <zb%n6.31033$o%6.586026@news1.rdc1.bc.home.com> <MPG.150b383cea226d409896d0@news.earthlink.net> <kNko6.6986$eq3.707121@news.uswest.net> <MPG.150c46db9cec283e9896d4@news.earthlink.net> <3AA2A898.450B659E@mail.com> <gRCt6.12224$Im6.1308560@newsread1.prod.itd.earthlink.net> <3AB755F8.67E84126@mail.com> <%a%t6.17245$227.1601569@newsread2.prod.itd.earthlink.net> <3AB8A730.4BE13D22@mail.com> <B6DE17D2.D6B0%t*b*r*u*m*l*e*y@f*u*s*e.n*e*t> <anhu6.18497$Im6.1938203@newsread1.prod.itd.earthlink.net> <B6DF5D0D.D7AE%t*b*r*u*m*l*e*y@f*u*s*e.n*e*t> <Q7Bu6.2721$P4.207644@newsread1.prod.itd.earthlink.net> <B6E0B412.D887%t*b*r*u*m*l*e*y@f*u*s*e.n*e*t>
Message-ID: <baYu6.5924$ue1.559361@newsread2.prod.itd.earthlink.net>

>
> [snip snip]
> > What makes people confortable with code seems to vary
> > greatly per individual.
>
> True.
>
> > .......
> >> I mention C, Pascal, and Assembly because they are my usual tool set,
or
> >> were until I discovered Smalltalk.
> >
> > I used similar stuff at first, but was glad to leave them in the dust
when I
> > discovered the concepts of Table Oriented Programming.
> >
> > I never wanted to see another fricken array again (accept as
> > hashes/dictionaries to pass single records or parameter
> > lists or attributes. That is their best use, and not data collection
> > holding.)
>
> While I'm not familiar with table oriented programming, a long time ago I
> did read a book on decision tables in software engineering.
>

I actually saw such a book in a used book store. I should have purchased
it for the "Table Oriented Programming Museum" that will be built in
2080 :-)

Accept, it was mostly about boolean tables. I extend them to
include lots of other things.

> I wonder if your advocacy of TOP is because it's the paradigm that works
> best with how your mind works?

Probably. I think it deserves a run in the limelight like any other
paradigm. I get "fan mail" from other toppers; so I know I am
not alone.

Hopefully the next big fad will be Mindfit Oriented Programming where
an attempt is made to classify, or at least analyze how an individual
thinks and programs.

Some say that if you try language/paradigm X long enough, that you
will like it. I am a bit skeptical of this. If there is not an early
hint of affection, then more of the same will probably not cut it.
(Of course, there are exceptions. But, I think many of the
"long term OO conversions" are mostly due to the fact that
software engineering is no longer taught for procedural
programming. Thus, if you crave SE thinking, then OO is
the only place to really get such info these days. Sad.)


> I know people who can do wonderful things in
> RPG that I could never do in the language, and indeed I found that working
> in it gave me headaches. These same people, bright and capable, couldn't
> work in my prefered environment of assembly language. Some assembly
language
> programmers I worked with could not transition to 4GLs like Cincom's
MANTIS.
>
> We all have different thinking styles, and I think language/environment
> preference relates directly to this. I'm a non-linear non-verbal intuitive
> thinker and I like Smalltalk. I know one verbal linear type who I couldn't
> imagine working in Smalltalk, but she's a very sharp gal.
>
> >
>  ....
> Scripting tools are wonderful, but why write one with an ugly looking C
type
> of code? Yuck! :)  [re perl]
>

I agree. C has been over-extended in its influence. There are already plenty
of C/Unix-shell type languages floating around. Time to try something new.
Even Python has hangovers such as token case sensativity. What the h*ll were
they thinking? Scripting and case sens. do NOT go hand in hand IMSO.

Thanks for the feedback,  -tmind-





From paulp at ActiveState.com  Mon Mar 19 03:33:57 2001
From: paulp at ActiveState.com (Paul Prescod)
Date: Mon, 19 Mar 2001 00:33:57 -0800
Subject: Tkinter Support
References: <001901c0b046$1f1a7200$340210ac@winson>
Message-ID: <3AB5C475.AF89FD43@ActiveState.com>

ActivePython comes with TKinter but not Tk. You can use the package
manager utility to install it (see the ActivePython start menu). Just
type "install tk" from the PPM command line.
-- 
Take a recipe. Leave a recipe.  
Python Cookbook!  http://www.activestate.com/pythoncookbook



From neelk at alum.mit.edu  Wed Mar 14 20:57:49 2001
From: neelk at alum.mit.edu (Neelakantan Krishnaswami)
Date: 15 Mar 2001 01:57:49 GMT
Subject: Who's minister of propaganda this week?
References: <mailman.984595938.11127.python-list@python.org> <98otuq0nnf@news1.newsguy.com>
Message-ID: <slrn9b0cc2.gsa.neelk@alum.mit.edu>

On Thu, 15 Mar 2001 00:09:30 +0100, Alex Martelli <aleaxit at yahoo.com> wrote:
>"Steven D. Majewski" <sdm7g at virginia.edu> wrote in message
>news:mailman.984595938.11127.python-list at python.org...
>    [snip]
>> ( OK: not having multiple dispatch in Python means that some manual
>> dispatching is sometimes required, but usually this is burried deep
>> in some classes __coerce__ method. )
>
> "Usually" in arithmetic, perhaps.  For fancier stuff, lack of
> multimethods sometimes cramps (not that it cramps any less in C++,
> mind you), though some patterns (such as 'dynamic visitor') can help
> a bit in certain (reasonably frequent) special cases.

What's a dynamic visitor? Is it a dictionary keyed on class? That's
what I usually do when emulating multimethods in Python.


Neel


From chrisa at ASPATECH.COM.BR  Mon Mar 12 16:13:12 2001
From: chrisa at ASPATECH.COM.BR (Chris Richard Adams)
Date: Mon, 12 Mar 2001 18:13:12 -0300
Subject: continue/else functionality
Message-ID: <BD674CDC0A68C642B60038986FBB56DF0138B9@mailsrv.ASPATECH.COM.BR>

I agree... in fact it seems to get worse the older I get.

For the record...the continue does work as stated - it goes for the next
iteration - my "other" problem was type comparison

 if datalist[0][0] == 1: always fails..

Needed to use  if datalist[0][0] == '1': since datalist[0][0] gives a
string.

Learn something everyday.


> -----Original Message-----
> From: Sean 'Shaleh' Perry [mailto:shaleh at valinux.com]
> Sent: Monday, March 12, 2001 5:49 PM
> To: Chris Richard Adams
> Subject: RE: continue/else functionality
> 
> 
> 
> On 12-Mar-2001 Chris Richard Adams wrote:
> > so I'm an idiot...thanks.
> > 
> 
> nah, just needed to be pointed into a better direction.
> 
> I find that programmer's (myself included) start to think 
> about a problem one
> way and just can not get out of that mindset.
> 



From costas at springmail.com  Tue Mar 13 11:47:26 2001
From: costas at springmail.com (costas at springmail.com)
Date: Tue, 13 Mar 2001 16:47:26 GMT
Subject: Sortin dictionaries in python
Message-ID: <3aae4757.68617130@News.CIS.DFN.DE>

Is there a simple way to sort a dictionary? It seems the only way to
do this is by indirectly creating  a sorted list.

And while on the topic of sorting lists. I tried sorting a list and
wanted to to pass a compare function to the list.sort() method. The
documentaion gives no example.

import string

x=['a','c','B']

def ignoreCase(left, right):
     if string.upper(left) < string.upper(right):
         return -1
     else:
         return 1

x.sort(ignoreCase(left, right))


But I get a message that left and right dont exist. How does one
define the compare function?

TIA

Costas


From nickson at centropolisfx.com  Fri Mar 23 00:28:46 2001
From: nickson at centropolisfx.com (Nickson Fong)
Date: Thu, 22 Mar 2001 21:28:46 -0800
Subject: Multiple Processes on Unix with Python
References: <3ABA8919.835D8BED@centropolisfx.com> <99eic9$i06$0@216.39.151.169>
Message-ID: <3ABADF0E.A1F9E93A@centropolisfx.com>

this is helpful - what i really want to do is>
for example: i have a 2 processor irix machine and i want to launch 2
processes. 

proc #1 = render frames 1-10
proc #2 = render frames 11-20


os.fork os.execv os.wait are things that i have been reading.....

question: why do you os.spawnv?

Donn Cave wrote:
> 
> Quoth Nickson Fong <nickson at centropolisfx.com>:
> | Can anyone show me a quick example on how to fork multiple processes on
> | a multiple cpu Irix machine. I know I need os.fork() and os.execvp() but
> | can't figure out how they work together.
> 
> The reason it's so hard, is that there are so many different things
> you might want to do.  One thing you might be missing is os.waitpid(),
> but it's hard to say.
> 
> Just for something completely different, take a look at this:
> 
>     import os
> 
>     p = []
>     p.append(os.spawnv(os.P_NOWAIT, '/bin/date', ('date',)))
>     p.append(os.spawnv(os.P_NOWAIT, '/bin/date', ('date',)))
>     p.append(os.spawnv(os.P_NOWAIT, '/bin/date', ('date',)))
> 
>     while p:
>         for i in range(len(p)):
>             pi, st = os.waitpid(p[i], os.WNOHANG)
>             if pi == p[i]:
>                 del p[i]
>                 print 'pid', pi, 'status', st
>                 break
> 
>         Donn Cave, donn at oz.net

-- 
thanks


        (:~~~ (:~~~ (:~~~    NO MORE EGGS   ~~~:) ~~~:) ~~~:)
			        JUST 
                               BADFROG

	
	      
                     Tel : (310) 204-7300 Ext 120
                         Pager : 626-582-5459
                      Office Fax : 310-204 7301

                       nickson at centropolisfx.com
	                  ubadfrog at yahoo.com


From scarblac at pino.selwerd.nl  Mon Mar 12 08:06:37 2001
From: scarblac at pino.selwerd.nl (Remco Gerlich)
Date: 12 Mar 2001 13:06:37 GMT
Subject: I have no idea
References: <g13r6.201$qt5.10920@news3.mia>
Message-ID: <slrn9apiav.9r4.scarblac@pino.selwerd.nl>

Brendhan Horne <brendhanhorne at bellsouth.net> wrote in comp.lang.python:
> In basic you use to be able to do this:
> 10 print "Hello World"
> 20 goto 10
> Run
> 
> If you did that it would run up the screen in an endless loop until you hit
> the ESC key. If you added   ; it would cover the whole screen. How do you
> write the equivalent in python?

while 1:
   print "Hello world"

There is no "goto" in Python, people noticed a long time ago that that
statement can lead to very confusing code, and that it is not necessary.

Instead of 'while 1:', you can also say 'while "true":' or something else
that will stay true forever.

The equivalent of adding a ; is adding a ,.

while 1:
   print "Hello world",

-- 
Remco Gerlich


From chris.gonnerman at usa.net  Sun Mar 25 19:13:51 2001
From: chris.gonnerman at usa.net (Chris Gonnerman)
Date: Sun, 25 Mar 2001 18:13:51 -0600
Subject: ftpmirror.py problems
References: <vfhqbtgss3kotgac8gomnn0i9muh6j9ir7@4ax.com>
Message-ID: <00ab01c0b589$a7d86e60$a100000a@local>

You are right, in that it is the OS and not Python which is causing
the problem.  You are probably having a known problem involving NAT 
(Network Address Translation, i.e. IPMasq) on the local host and a 
badly configured firewall on the remote end.

If you can do so, configure the Linux system to use MTU/MRU 1500 when
connecting via PPP.  If this does not work, try 296.

The explanation is long and painful, and involves Path MTU Discovery.

----- Original Message ----- 
From: "Cary Coulter" <cwc_NO at sound_NO.net>
Newsgroups: comp.lang.python
To: <python-list at python.org>
Sent: Sunday, March 25, 2001 5:46 PM
Subject: ftpmirror.py problems


> I am trying to use ftpmirror.py to duplicate various subdirectories.
> When I test the replication between two hosts on my local network,
> everything works like expected.
> 
> However, I am having problems replicating hosts residing on the
> internet to my local system.  My local system is on a private network
> (192.168...) behind a Linux firewall using IP Masq.  In this
> environment, I am experiencing "hangs" in the ftpmirror.py
> application.  I have to kill it with ^C.
> 
> I can normally download 10-40 files before this happens.  It seems to
> be more prone if the files are smaller (can download in 3-5 seconds on
> a cable modem) rather than the 10+ MB files.
> 
> Is anyone else experiencing this behavior?  Since it seem to work well
> beween two local hosts, I am suspicious of the firewall and IP Masq.
> I would be grateful for any pointers anyone can give to help remedy
> this situation.
> 
> I am relatively new to Python. I really like what I see but I'm not
> ready to debug this application.  I assume it really works and that
> the problem is something else.
> 
> Thanks in advance for your input
> Cary
> 
> -- 
> http://mail.python.org/mailman/listinfo/python-list
> 




From c.evans at clear.net.nz  Sat Mar 17 01:57:38 2001
From: c.evans at clear.net.nz (Carey Evans)
Date: 17 Mar 2001 19:57:38 +1300
Subject: md5 passwords
References: <mailman.984779122.3483.python-list@python.org>
Message-ID: <87elvwvq59.fsf@psyche.dnsalias.org>

Timothy Grant <tjg at exceptionalminds.com> writes:

> Can anyone explain to me how to use the md5 module to generate
> passwords for the shadow password file?
> 
> I would like to be able to be able to os.system('useradd -p
> md5/crypt username')

Have a look at Michal Wallace's `md5crypt.py' module, which should do
this for you:

    http://www.sabren.net/code/python/crypt/

-- 
	 Carey Evans  http://home.clear.net.nz/pages/c.evans/

	    "Quiet, you'll miss the humorous conclusion."


From whisper at oz.net  Sun Mar 11 13:45:33 2001
From: whisper at oz.net (Dave LeBlanc)
Date: 11 Mar 2001 18:45:33 GMT
Subject: Urgent Question about Python licensing
References: <mailman.984290534.9099.python-list@python.org>
Message-ID: <98gh4d$kft$0@216.39.170.247>

On Sun, 11 Mar 2001 01:01:08 -0500, "Tim Peters" <tim.one at home.com>
wrote:

>[Dave LeBlanc]
>> When UCITA first came out, there was a lot of press about it's evils.
>
>There still is!  There was also a lot of discussion about it on
>comp.lang.python, specifically in relation to the 1.6 CNRI license at the
>time.  You need to read the license and the actual text of UCITA to get
>anywhere with this -- or ask a lawyer.  Whether UCITA is good or bad, the 1.6
>license grants you very strong rights, and UCITA can't magically negate
>those.
>
>From what I read, UCITA can't magically negate any rights unless it's
agreed on by the parties.

>> ...
>> However, after reviewing information at http://www.ucitaonline.com/
>> and at http://www.2bguide.com, I came to the conclusion that i'd been
>> mislead. Imagine! being mislead by the press! <g>.  (I should have
>> known better - i'm a firm believer in the "believe none of what you
>> hear and only half of what you read" school of thought.)
>
>Both of those are UCITA advocacy sites, and set up by the same lawyer (Carol
>Kunze, who "attended the UCITA drafting committee meetings from 1996 -
>1999").  So on *those* sites, believe the negative halves (if you can find
>any <0.5 wink>).
>
Some of the referenced news items are negative, but I don't think the
_really_ negative ones are linked <g>

>> ...
>> As it turns out, after the fact license changes are NOT binding,
>> "shrink wrap" or "click wrap" licenses ARE binding and have been
>> upheld by the courts, can't make a bogus venue choice, and that a
>> licensor CAN unilaterally disable software but only after a 2 week
>> warning in many cases.
>
>Note that what UCITA says is irrelevant unless read *in conjunction* with
>what the 1.6 license says.  For example, section 308(2)(B) of UCITA holds
>that the license is *perpetual* if "the license expressly grants the right to
>incorporate or use the licensed information or informational rights with
>information or informational rights from other sources in a combined work for
>public distribution or public performance.".  The CNRI 1.6 license does so
>expressly grant both rights.  I am not a lawyer, but I can read -- so can
>you.  UCITA simply doesn't kick in except where the 1.6 license leaves holes,
>and CNRI didn't intend to leave any holes.  IMO (and IANAL), the GPL does
>leave holes, and that's the fundamental source of the conflict (when CNRI
>plugs a hole that the GPL doesn't, that's "a restriction" not in the GPL,
>hence automatically leaves the CNRI license "incompatible" with the latter,
>which forbids additional restrictions of *any* kind).
>
>> ...
>> Moving along, I find it ironic that the Open Software Foundation calls
>> the GPL an open source license since it violates the rules given for
>> such. GPL software has source available, but it's use is constrained
>> and contaminating (unlike the GLL library license).
>
>I see that they've added a footnote to point 9 at
>
>    http://www.opensource.org/docs/definition.html
>
>It reads:
>
>    Yes, the GPL is conformant with this requirement.  GPLed
>    libraries "contaminate" only software to which they will actively
>    be linked at runtime, not software with which they are merely
>    distributed.
>
My notion of open software is more BSD license-ish then GPL - which
places restraints upon reuse in an unacceptable (to me and a lot of
others AFAIK) way. I recall from the GPL that "mere agregation" of GPL
and non-GPL programs on the same media doesn't contaminate the non-GPL
software.

Even Sun's "Sun Internet Standards License" is more user-friendly then
GPL.

>I'm unsure whether they added that before or after your post, though <wink>.
>
>> P.S Guido is welcome to my first born if I ever have one - but he has
>> to do his own cooking.
>
>guido-does-not-eat-children-neither-cooked-nor-raw-ly y'rs  - tim
>
>
Hey, you offerred your first born to Guido first! <g>

Dave LeBlanc


From robert at kasunic.de  Wed Mar 21 10:50:14 2001
From: robert at kasunic.de (Robert Kasunic)
Date: Wed, 21 Mar 2001 16:50:14 +0100
Subject: python and mysqldb
Message-ID: <mjia99.b14.ln@braveheart.dyndns.org>

Hello,

I would like to process a file and store different things from that file
into a MySQL DB. The code stated below works as expected, but if I try to
substitute one of those values with a variable I get the following Error:

	_mysql.OperationalError: (1054, "Unknown column 'temp' in 'field list'")
							 ^^^^
	 				       temp is the name of the variable 


db = MySQLdb.connect(db='test',user='test',passwd='***secret***')
cursor = db.cursor()
statement = "INSERT INTO test_table VALUES ('test1', 'test2', 'test3')"
cursor.execute(statement)

I am quite sure that I have to change the syntax somehow, but couldn't
find anything useful on the web. Can somebody enlighten me?

Cheers
Robert


From aahz at panix.com  Tue Mar 20 17:57:44 2001
From: aahz at panix.com (Aahz Maruch)
Date: 20 Mar 2001 14:57:44 -0800
Subject: New PEP: The directive statement
References: <mailman.985117312.5806.python-list@python.org> <yHPt6.2553$4N4.334967@newsc.telia.net> <998hv2$5ub$1@panix3.panix.com> <oOQt6.2576$4N4.342378@newsc.telia.net>
Message-ID: <998n98$fig$1@panix2.panix.com>

In article <oOQt6.2576$4N4.342378 at newsc.telia.net>,
Fredrik Lundh <fredrik at effbot.org> wrote:
>Aahz Maruch wrote:
>> Fredrik:
>>>
>>>except that "import" already is a declaration statement, of course.
>>
>> Depends how you define "already".  There is no current released version
>> of Python (alphas and betas do not count as "released versions") in
>> which "import" is a declaration.
>
>import declares things, in all existing Python versions (hint: it's
>an assignment statement)

Mmmmm....  I see your point, but I think you're missing Martin's,
particularly in terms of the analogy with "global".  I also think it's
probably better at this point that I let Martin speak for himself rather
than me trying to interpret him.  ;-)
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

"I won't accept a model of the universe in which free will, omniscient
gods, and atheism are simultaneously true."  -- M


From fredrik at pythonware.com  Fri Mar 23 11:12:22 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 23 Mar 2001 16:12:22 GMT
Subject: Slightly Off-Topic: Python featured in Severance: Blade of Darkness
References: <NMHu6.144$NR.1628@news3.oke.nextra.no>
Message-ID: <GBKu6.3193$4N4.474429@newsc.telia.net>

Thomas Weholt wrote:
> A friend of mine just informed me that while starting up his copy of the
> game Blade of Darkness he saw a reference to Python 1.5.2 during the loading
> process. Anybody know anything about that? What has Python been used for?

everything, it seems.  I found a zipped script archive on a fan
site (for mod hackers who's messed up their installation ;-).

it contains about 370,000 lines of Python source code...  game
logic, sound f/x, animations, translations, you name it.

Cheers /F




From tjreedy at udel.edu  Wed Mar 21 17:13:14 2001
From: tjreedy at udel.edu (Terry Reedy)
Date: Wed, 21 Mar 2001 17:13:14 -0500
Subject: PEP 242 Released
References: <mailman.985133006.5845.python-list@python.org>
Message-ID: <99b91p$fjn$1@news.udel.edu>

"Paul F. Dubois" <paul at pfdubois.com> wrote in message
news:mailman.985133006.5845.python-list at python.org...
> PEP: 242
> Title: Numeric Kinds
...
> Rationale
>
>     Currently it is impossible in every language except Fortran 90 to
>     write a program in a portable way that uses floating point and
>     gets roughly the same answer regardless of platform -- or refuses
>     to compile if that is not possible

I have the impression that there are issues other than precision required
to meet this goal (conversion, truncation, rounding, and standard math
function implementation).


> Kind Objects
>
>     Three new standard functions are defined in a module named
>     "kinds".  They return callable objects called kind objects.

While you here present the historical Fortran-precedent reason for the
name, it is, standing alone, uninformative as to its purpose.  I suggest
something like 'numbers', 'number_kinds', or 'numkinds', which would
tell/remind people than the module has something to do with numerical
computation.

This module, whatever the name, obviously has to be a builtin written in C
(or Java) rather than Python..  As such, it should still be as optional as
any other importable module.   And as such, you and whoever are free to
write a 'kinds' module for whatever platforms you like without anyone's
approval.  You mainly need agreement among yourselves on the function and
interface.  Taking this part of your proposal at face value, the PEP
approval process is not necessary.


> The Meaning of Literal Values
>
> ... Literal decimal values are of the greatest available binary
floating-point kind.

I strongly object to this.

First, this semantic change hijacks current syntax for the, to me, fairly
trivial purpose of avoiding writing a function call that 'kinds' should
provide (max_real(1.333) or whatever).  I do not see that it adds to the
functionality of the proposed computation machinery, and it would benefit
very few users, at potential cost to others.

Second, this change would make floating point computations more rather than
less variable.  Two versions of the interpreter compiled with the same
compiler for the same platform might no longer do floating point the same
way.   This contradicts part of the declared purpose of this PEP.

Third, despite the claim that this proposal "is backward compatible with
existing code", this part of it would break any existing code that depends
on floats being, for instance, 8 bytes (for memory allocation, for
instance).  It would also break code which depends on floats being the same
as a C double on the same machine (for interface with C routines, for
instance).

Fourth, the core interpreter would have to be aware, during compile time,
of the existence and contents of a 'kinds' module that *might* be imported
during runtime and which could well exist in a separate shared library.  To
make this semantic change to the core language, I believe that the
x_kinds() functions (and the machinery behind them) would have to be
obligatory built-in functions [like len()] rather than being segregated in
an optional import module.  This would be an entirely different proposal to
which I would, again, object.


> Concerning Infinite Floating Precision
>
>     This PEP does not propose the creation of an infinite precision
>     floating point type, just leaves room for it.  Just as int_kind(0)
>     returns the long kind object, if in the future an infinitely
>     precise decimal kind is available,

There already is such a module (real.py?).  If currently only writen in
Python, it could be converted to C for more speed.


> Coercion
>
>     In an expression, coercion between different kinds is to the
>     greater kind

If the kinds were written as extension classes rather than as extension
types, this would be taken care of with __coerce__() functions.  Again, no
changes to the core and no permissions needed.


> Open Issues
>
>     The assertion that a decimal literal means a binary floating-point
>     value of the largest available kind is in conflict with other
>     proposals about Python's numeric model.  This PEP asserts that
>     these other proposals are wrong and that part of them should not
>     be implemented.

Maybe both are wrong.


Summary of Response:

This PEP combines two somewhat disjoint proposals.

First is a proposed interface for a compiled extension class module that
would extend and parallel existing modules (such as real.py and the
Numerical Python work).  This can be implemented without changes to the
core and without anyone's approval.  This approach also leaves control of
the implementation in the hands of the proposer and collaborators.  So if
there is enough interest, go ahead and do it.

Second is a proposed semantic change in the core language which I believe
would require new builtin functions even though that was not directly
proposed.  Its purpose is to add syntactic sugar for the benefit of the
first proposal.  This change would complexify the language and probably
confuse beginners even more than they are now.  It would expand and slow
down the interpreter.  It could also break existing code.  Since it would
benefit only a small fraction of users (<1%?) and since the desired
functionality can be attained through current extension mechanisms, I see
no justification for imposing the costs on everyone else.

Terry J. Reedy





From mlh at idi.ntnu.no  Wed Mar 14 15:22:30 2001
From: mlh at idi.ntnu.no (Magnus Lie Hetland)
Date: Wed, 14 Mar 2001 21:22:30 +0100
Subject: Expect -- or popen2 -- or what?
Message-ID: <98ojsa$95n$1@tyfon.itea.ntnu.no>

I'm trying to communicate with an interactive program
which seems to be a bit reluctant to flush its
output... So when I use popen2, things just seem
to hang. It seems that expy and PIPE are both quite
out of date ... I've seen ExpectPy and it seems
interesting, but perhaps a bit more than I really need.

Is there a standard way of handling this? I can handle
the parsing of the output easily myself -- I just want
to get my hands on it :)

--

  Magnus Lie Hetland          http://www.hetland.org

 "Reality is what refuses to disappear when you stop
  believing in it"                 -- Philip K. Dick





From akuchlin at mems-exchange.org  Mon Mar 26 18:07:57 2001
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: Mon, 26 Mar 2001 18:07:57 -0500
Subject: PEP 1, PEP Purpose and Guidelines
References: <mailman.985196613.7994.python-list@python.org> <99ifgu$jei$1@panix3.panix.com>
Message-ID: <E14hg5d-0005k0-00@ute.cnri.reston.va.us>

aahz at panix.com (Aahz Maruch) writes:
> After taking a closer look at Moshe's numeric proposals, I suggest adding
> a Depends-On: header.  (E.g. PEP 240 depends on PEP 239.)

What would Depends-On mean?  Does it mean "To understand PEP 240, go
read PEP 239 first"?  Or "To implement PEP 240, PEP 239 would first
have to be implemented or accepted"?  Or something else?  

--amk



From jeremy at alum.mit.edu  Sun Mar 11 20:22:04 2001
From: jeremy at alum.mit.edu (Jeremy Hylton)
Date: Sun, 11 Mar 2001 20:22:04 -0500 (EST)
Subject: [Python-Dev] Revive the types sig?
In-Reply-To: <3AAC08DB.9D4E96B4@ActiveState.com>
References: <3AAC08DB.9D4E96B4@ActiveState.com>
Message-ID: <15020.9404.557943.164934@w221.z064000254.bwi-md.dsl.cnc.net>

>>>>> "PP" == Paul Prescod <paulp at ActiveState.com> writes:

  PP> Let's discuss my proposal in the types-sig. Executive summary:

  PP> * incremental development policy
  PP> * syntax for parameter type declarations
  PP> * syntax for return type declarations
  PP> * optional runtime type checking
  PP> * goals are better runtime error reporting and method
  PP>    documentation

If your goal is really the last one, then I don't see why we need the
first four <0.9 wink>.  Let's take this to the doc-sig.

I have never felt that Python's runtime error reporting is all that
bad.  Can you provide some more motivation for this concern?  Do you
have any examples of obscure errors that will be made clearer via type
declarations?

The best example I can think of for bad runtime error reporting is a
function that expects a sequence (perhaps of strings) and is passed a
string.  Since a string is a sequence, the argument is treated as a
sequence of length-1 strings.  I'm not sure how type declarations
help, because:

    (1) You would usually want to say that the function accepts a
        sequence -- and that doesn't get you any farther.

    (2) You would often want to say that the type of the elements of
        the sequence doesn't matter -- like len -- or that the type of
        the elements matters but the function is polymorphic -- like
        min.  In either case, you seem to be ruling out types for
        these very common sorts of functions.

If documentation is really the problem you want to solve, I imagine
we'd make much more progress if we could agree on a javadoc-style
format for documentation.  The ability to add return-type declarations
to functions and methods doesn't seem like much of a win.

Jeremy



From greg at cosc.canterbury.ac.nz  Thu Mar  1 21:18:46 2001
From: greg at cosc.canterbury.ac.nz (Greg Ewing)
Date: Fri, 02 Mar 2001 15:18:46 +1300
Subject: is and == (is not other_thread)
References: <mailman.983428868.21795.python-list@python.org>
Message-ID: <3A9F0306.140052CB@cosc.canterbury.ac.nz>

Robin Thomas wrote:
> 
> String literals are "interned" by Python, so that it uses a single string
> object for every unique string literals it finds in the code.

Only for strings which happen to look like identifiers, though:

>>> a = "asdf"
>>> b = "asdf"
>>> a is b
1
>>> c = "!@#$"
>>> d = "!@#$"
>>> c is d
0

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


From fredrik at pythonware.com  Fri Mar 16 01:42:24 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 16 Mar 2001 06:42:24 GMT
Subject: Announcement: Python Programming: Taming the Snake.
References: <mailman.984699860.10366.python-list@python.org>
Message-ID: <kBis6.104$sk3.37174@newsb.telia.net>

> Open Docs is set to release "Python Programming: Taming the Snake" at the
> end of June. It will cover 2.0 including features found in 2.1.

a book written by anonymous authors?  that's interesting.

Cheers /F




From wayne.izatt at _nospam_myself.com  Mon Mar 19 03:01:10 2001
From: wayne.izatt at _nospam_myself.com (Wayne Izatt)
Date: Mon, 19 Mar 2001 08:01:10 GMT
Subject: stripping punctuation from tuples
References: <3AB57F0C.594CF3E9@northwestern.edu>
Message-ID: <a1jt6.205998$df5.5255534@news1.crdva1.bc.home.com>

Is this what you're looking for?

>>> a = []
>>> a.append(('Alan Tsang',))
>>> a
[('Alan Tsang',)]
>>> a[0]
('Alan Tsang',)
>>> b = a[0]
>>> b
('Alan Tsang',)
>>> (x, y) = string.split(str(b[0]))
>>> x, y
('Alan', 'Tsang')
>>> x
'Alan'
>>> y
'Tsang'
>>>

cheerio


"Louis Luangkesorn" <lluang at northwestern.edu> wrote in message
news:3AB57F0C.594CF3E9 at northwestern.edu...
> How do you strip the punctuation from tuples?  In particular, say I have
> a tuple a[10] (which happened to come from a not very well designed
> database)
>
> >>> a[10]
> ('Alan Tsang',)
>
> How do I get something to say 'Alan Tsang'  with out the parens or the
> ','
>
> Or, what I'm really trying to do is make this so
> >>>b = breakname(a[10])
> >>>b[0], b[1]
>
> 'Alan','Tsang'
>
> I've tried to do:
> >>> d = split(str(a[0]))
> >>> d
> ["('Aaron", "Fan',)"]
> >>> d[0]
> "('Aaron"
> >>> d[1]
> "Fan',)"
>
> But that is not quite what I want.
> Many thanks.
> Louis
> --
> K Louis Luangkesorn
> lluang at northwestern.edu  http://pubweb.nwu.edu/~kll560  PGP:0xF3E2D362
>
> Whatsoever things are true, ... honest, ... just, ... pure, ... lovely,
> ... of good report; if there be any virtue, and if there be any praise,
> think on these things.
>      - motto - Northwestern University (Phil 4:8)
>




From jfontain at free.fr  Sun Mar 18 15:00:09 2001
From: jfontain at free.fr (Jean-Luc Fontaine)
Date: Sun, 18 Mar 2001 20:00:09 GMT
Subject: ANNOUNCE: tclpython-2.0.1
Message-ID: <3AB513AF.9A940B2A@free.fr>

### CHANGES ###

--- version 2.0.1 ---

errors from Python were not always caught and properly reported to the
  Tcl interpreter
tclpython2 binary rpm was wrongly compiled against another
  installation on my computer, which caused string I/O module loading
  failure and a core dump

### README ###

tclpython version 2.0.1: a Python package for Tcl

This package allows the execution of Python code from a Tcl
interpreter, as in:


    package require tclpython
    # or: package require tclpython2
    set interpreter [python::interp new]
    $interpreter eval {print("Hello World")}
    python::interp delete $interpreter


You can actually create several Python interpreters this way, if the
tclpython package was linked against a Python library compiled with
threads support, otherwise only 1 Python interpreter can exist at a
time.

Use the tclpython2 if you are using Python 2.0 or above.

This library is free software; you can redistribute it and/or modify
it under the terms of the GNU Library General Public License as
published by the Free Software Foundation; either version 2 of the
License, or (at your option) any later version.

This library is distributed in the hope that it will be useful, but
WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.

###

you may find it now at my homepage:

http://jfontain.free.fr/tclpython-2.0.1.tar.gz
http://jfontain.free.fr/tclpython-2.0.1-1.i386.rpm
http://jfontain.free.fr/tclpython-2.0.1-1.spec
http://jfontain.free.fr/tclpython2-2.0.1-1.i386.rpm
http://jfontain.free.fr/tclpython2-2.0.1-1.spec
http://jfontain.free.fr/tclpython.htm

Enjoy and please let me know what you think.


--
Jean-Luc Fontaine  mailto:jfontain at free.fr  http://jfontain.free.fr/





From fig at monitor.net  Fri Mar 23 19:14:22 2001
From: fig at monitor.net (Stephen R. Figgins)
Date: Fri, 23 Mar 2001 16:14:22 -0800
Subject: PyUnit in Python 2.1 standard library
References: <3ABA8A23.85EA76CE@downright.com> <3ABBC656.C54B0F9A@monitor.net>
Message-ID: <3ABBE6DE.4330AE2A@monitor.net>

> This means that it will be a standard feature of the Python
> distribution?  Will it come with Python 2.1?

Oops!  Nevermind.  The 2.1b2 announcement answered my question.


From bos at hack.org  Wed Mar 28 08:48:03 2001
From: bos at hack.org (Rikard Bosnjakovic)
Date: Wed, 28 Mar 2001 15:48:03 +0200
Subject: list vs tuple
References: <hGlw6.782$p5.3210@news1.rivrw1.nsw.optushome.com.au>
Message-ID: <3AC1EB93.7BC87292@hack.org>

deadmeat wrote:

> A tuple seems to be a very basic list, so unless it's faster, what's the
> point of it?

Tuples are immutable, lists are not. With tuples you get some sort of
integrity, since you can't change a tuple "by mistake" by some other
reference in your program. With lists, there is no such safety.


-- 
Rikard Bosnjakovic - http://a214.ryd.student.liu.se/cv/ - ICQ: 1158217

Anyone sending unwanted advertising e-mail to my address will be
charged $250 for network traffic and computing time. By extracting my
address from this message or its header, you agree to these terms.


From andreas at andreas-jung.com  Wed Mar 21 17:10:31 2001
From: andreas at andreas-jung.com (Andreas Jung)
Date: Wed, 21 Mar 2001 17:10:31 -0500
Subject: Accessing Oracle from Python remotely ?
References: <Xns906AB9C901807PaCmAnRDLM@161.48.76.20> <eE3u6.776$9%.203510@newsr1.maine.rr.com>
Message-ID: <99bj2o$u91$1@nntp9.atl.mindspring.net>

 There are currently two different DCOracle versions available. DCOracle
1.3.x
supports is mostly build on top of Oracle 7 but I works although with Oracle
8i
except features like BLOBs and CLOBs. When you need complete support
for Oracle 8i take a look at Matt Kromers new version of DCOracle 2. It is
still alpha so be warned :-) You can find DCO2 on www.zope.org  - search for
DCO2.

Andreas Jung
Digital Creations

"gkiffney" <gkiffney at my-deja.com> schrieb im Newsbeitrag
news:eE3u6.776$9%.203510 at newsr1.maine.rr.com...
> There's http://www.zope.org/Products/DCOracle which should get you
started.
> There was an article by Uche Ogbuji in Linuxworld that helped me get going
> with this, but it's apparently fallen off the Internet map into the bit
> bucket.  Using the dcoracle module (which is a bit of a pain to get going,
> look at the Zope docs carefully, and if you can find the prebuilt dll's,
use
> those), you can do things like this:
>
> [uche at malatesta uche]$ python
> Python 1.5.1 (#1, Sep 3 1998, 22:51:17) [GCC 2.7.2.3] on linux-i386
> Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
> >>> import DCOracle
> >>> #connect to DB. scott/tiger are the test database username/password
> >>> dbc=DCOracle.Connect("scott/tiger at sforza") #sforza is the local Oracle
> server
> >>> c = dbc.cursor()
> >>> c.execute("SELECT * FROM emp WHERE emp.deptno = 10")
> >>> rows = c.fetchall()
> >>> print rows
> [(7782, 'CLARK', 'MANAGER', 7839, dbiDate('1981-06-09 00:00:00'), 2450.0,
> None, 10),
> (7839, 'KING', 'PRESIDENT', None, dbiDate('1981-11-17 00:00:00'), 5000.0,
> None, 10),
> (7934, 'MILLER', 'CLERK', 7782, dbiDate('1982-01-23 00:00:00'), 1300.0,
> None, 10)]
>
>
>
> "Fred Pacquier" <fredp at mygale.org.nospam> wrote in message
> news:Xns906AB9C901807PaCmAnRDLM at 161.48.76.20...
> > Hi,
> >
> > I have an Oracle (7) database running on a Windows NT 4 server. Is it
> > possible (easy, difficult ?) to read the data in those tables from a
> python
> > script running on a remote Unix (Solaris) box ?
> >
> > Sorry if the question is silly -- I do know a bit of python but I'm
> > inexperienced in databases and their connectivity. Any pointers welcome
> > though.
> >
> > Thanks,
> > fp
> >
> > --
> > YAFAP : http://www.multimania.com/fredp/
>
>
> --
> http://mail.python.org/mailman/listinfo/python-list




From carifio.nospam at nospam.usys.com  Wed Mar 14 08:36:59 2001
From: carifio.nospam at nospam.usys.com (Mike Carifio)
Date: Wed, 14 Mar 2001 13:36:59 GMT
Subject: windows script host and python
Message-ID: <%tKr6.2570$pX.26174@news1.onlynews.com>

I'm looking for some resources on using python with wsh generally.

Specifically, I'm trying to use .wsf files with python. For example,
given a "one liner" in (say) hello.py:

    WScript.Echo("hello")

then wscript //E:python hello.py works. But hello.wsf below doesn't:

<?xml version="1.0" ?>
<?job error="true" debug="true" ?>

<package id="package1">
   <job id="job1"> <!-- run by default -->
     <comment>
       blah blah
     </comment>
     <script language="python">
     <![CDATA[
         WScript.Echo("hello")
     ]]>
     </script>
   </job>
</package>

Is this known to work by a different path? Known not to work at all? Pl.
advise. Tx.




From db3l at fitlinxx.com  Fri Mar  9 20:00:17 2001
From: db3l at fitlinxx.com (David Bolen)
Date: 09 Mar 2001 20:00:17 -0500
Subject: from Tkinter import *
References: <mailman.983920449.5230.python-list@python.org> <slrn9ab0g7.s6.matt@happy-hour.mondoinfo.com> <989dot$t6t$1@panix3.panix.com> <uae6vk3up.fsf@ctwd0143.fitlinxx.com> <slrn9ah6au.k09.qrczak@qrnik.zagroda>
Message-ID: <u8zme8mla.fsf@ctwd0143.fitlinxx.com>

qrczak at knm.org.pl (Marcin 'Qrczak' Kowalczyk) writes:

> 08 Mar 2001 22:39:42 -0500, David Bolen <db3l at fitlinxx.com> pisze:
> 
> > the one place I still see a use for "import *" even in production
> > code is large wrapped C libraries that already name everything
> > according to a flat namespace (for example, staying in the GUI
> > vein, wxPython).
> 
> I would prefer to remove these prefixes while porting to a language
> with a module system.

Generally the issue arises when the library isn't being "ported" but
just wrapped, and in most cases largely in an automatic fashion.  I
certainly wouldn't object to modularizing, but it's bound to make it
harder to keep the wrappings up to date - unless tools such as SWIG
are augmented to take care of the prefix processing for you.  In the
meantime, I'd still happy to have flat wrapping than nothing.

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


From brangdon at cix.co.uk  Thu Mar  1 15:03:00 2001
From: brangdon at cix.co.uk (Dave Harris)
Date: Thu, 1 Mar 2001 20:03 +0000 (GMT Standard Time)
Subject: Stream interfaces
References: <3A9E54F9.7AF90BCA@ix.netcom.com>
Message-ID: <memo.20010301200321.16971A@brangdon.madasafish.com>

tgagne at ix.netcom.com (Thomas Gagne) wrote (abridged):
> Your correct.  People are confusing inheritence and usefulness.  I 
> don't have enough spare cycles to worry about a class'
> inheritence unless I've discovered missing functionality.  Since
> I haven't, I haven't.

Well, I have. In the streams hierarchy, too. Albeit not that vendor and 
in a beta product so I probably shouldn't say too much about it, but the 
hierarchy was very similar. The problem was analogous to methods added to 
WriteStream not magically appearing in ExternalWriteStream.

The vendor's fix was to cut&paste methods from one class to the other. 
But this is just sticking plaster. The class hierarchy is deeply wrong, 
violates "once and only once" all over the place and needs refactoring. 
It's not about missing functionality so much as how we can add new kinds 
of stream without more error-prone cut&paste.

Whether this can be fixed without breaking existing clients, or fixed at 
all without multiple inheritance, I don't know because I've not tried it. 
I hope it can be and the problem is not with Smalltalk generally but just 
this particular bit of library design. But for whatever reason, this bit 
of library design is not ideal.

  Dave Harris, Nottingham, UK | "Weave a circle round him thrice,
      brangdon at cix.co.uk      |   And close your eyes with holy dread,
                              |  For he on honey dew hath fed
 http://www.bhresearch.co.uk/ |   And drunk the milk of Paradise."


From jwbnews at scandaroon.com  Mon Mar 19 23:49:54 2001
From: jwbnews at scandaroon.com (John W. Baxter)
Date: Mon, 19 Mar 2001 20:49:54 -0800
Subject: Pep 238 Why new // operator (Non-integer Division)
References: <mailman.984811161.25962.python-list@python.org> <995tbo0oce@drn.newsguy.com>
Message-ID: <jwbnews-05F3D1.20495419032001@corp.supernews.com>

In article <995tbo0oce at drn.newsguy.com>,
 Grant Griffin <not.this at seebelow.org> wrote:

> Now don't get me start on that: to me, it had looked more like _Perl's_ #!  
> (That has too many sharp edges!)

To me, it just looks like an octotherp.  ;-)  --John

-- 
John W. Baxter   Port Ludlow, WA USA  jwbnews at scandaroon.com


From sholden at holdenweb.com  Thu Mar 22 23:32:25 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 22 Mar 2001 23:32:25 -0500
Subject: PyUnit in Python 2.1 standard library
References: <3ABA8A23.85EA76CE@downright.com> <zZxu6.150312$__6.27726222@typhoon.southeast.rr.com>
Message-ID: <7pAu6.9141$G72.244697@e420r-atl1.usenetserver.com>

"Don Tuttle" <tuttledon at hotmail.com> wrote in message
news:zZxu6.150312$__6.27726222 at typhoon.southeast.rr.com...
> "Doug Fort"
> > Just recieved word that Steve Purcell has checked PyUnit  into the
> > standard library.  Aside from being a useful tool to me PyUnit typifies
> > the kind of development that ESR advocates in 'The Cathedral and the
> > Bazaar'.   A project started to 'scratch an itch', then shared with the
> > communityto evolve and grow.  None of the stuff I  work on seems to come
> > out so clean.  Congratulations Steve!  PyUnit is an example to us all.
>
> So what kind of itch does PyUnit scratch?
>
The one you get about an hour before you start receiving emails about a bug
in your software.

regards
 Steve





From cce at clarkevans.com  Sat Mar 24 22:22:20 2001
From: cce at clarkevans.com (Clark C. Evans)
Date: Sat, 24 Mar 2001 22:22:20 -0500 (EST)
Subject: Yet Another PEP:  Query Protocol Interface or __query__
In-Reply-To: <Pine.LNX.4.21.0103201837570.21343-100000@clarkevans.com>
References: <Pine.LNX.4.21.0103201837570.21343-100000@clarkevans.com>
Message-ID: <Pine.LNX.4.21.0103242220380.4332-100000@clarkevans.com>

Below is an updated version of the proposal Ive been 
a champion for.  It incorporates a great deal of feedback
that Ive received.  Changes include:

   * By default, adapt now throws an error again, however
     an alternate object can be provided upon failure much
     like getattr.

   * Now __adapt__ can either return None or throw a TypeError
     with the same effect.

   * This now includes reverse adaptation of the object
     by the protocol via __prodapt__ (better name?), which
     is very similar to Paul Prescods __check__

   * The text has been tightened (hopefully eliminating 
     much superfluous material)

   * The sample implementation code has been cleaned up
     and tested in more detail.

   * Ive received much comment to remove the type-checking
     stuff.  I have a question:  Is type-checking a component
     of adapting?  

I would very much appreciate any comments and feedback.

Especially if the intent/value of this proposal is not
clear as I seem to be having a difficult time with the
expression of this.

Further, I still do not have a PEP number.  Either I have 
not asked correctly, am being stupid, have pissed someone 
off, or am proposing something that has already been 
rejected.  In any case, what do I do?

Kind regards,

Clark

P.S.  Id like to thank Alex Martelli for his most helpful
      comments and hope that I have dutifully incorporated
      his suggestions appropriately. 


PEP: XXX
Title: Protocol Checking and Adaptation
Version: $Revision$
Author: Clark Evans
Python-Version: 2.2
Status: Draft
Type: Standards Track
Created: 21-Mar-2001
Updated: 23-Mar-2001

Abstract

    This proposal puts forth a light-weight explicit mechanism 
    for the adaptation (including verification) of an object to 
    a context where a specific type, class, interface, or other
    protocol is expected.  This proposal can leverage existing
    protocols such as the type system and class hierarchy and is
    orthogonal, if not complementary to the pending interface
    mechanism [1] and signature based type-checking system [2]

    This proposal allows a standard and extensible method to
    ask two types of questions:  Is a particular object
    compliant with a given protocol? (b) And if not, can the
    object be wrapped so that it is complaint?  This proposal
    does not limit what a protocol is, what compliance to the 
    protocol means, nor what a wrapper constitutes.

Motivation

    Currently there is no standardized mechanism in Python for
    asking if an object supports a particular protocol. Typically,
    existence of particular methods, particularly those that are
    built-in such as __getitem__, is used as an indicator of
    support for a particular protocol.  This technique works for
    protocols blessed by GvR, such as the new enumerator proposal
    identified by a new built-in __iter__.  However, this technique
    does not admit an infallible way to identify interfaces lacking
    a unique, built-in signature method.

    More so, there is no standardized way to obtain an adapter
    for an object.  Typically, with objects passed to a context
    expecting a particular protocol, either the object knows about
    the context and provides its own wrapper or the context knows
    about the object and wraps it appropriately.  The difficulty 
    with these approaches is that such adaptations are one-offs,
    are not centralized in a single place of the users code, and
    are not executed with a common technique, etc.  This lack of
    standardization increases code duplication with the same
    adapter occurring in more than one place or it encourages
    classes to be re-written instead of adapted.  In either case,
    maintainability suffers.

    It would be very nice to have a standard function that can
    be called upon to verify an objects compliance with a 
    particular protocol and provide for a wrapper if one is
    readily available -- all without having to hunt through 
    a librarys documentation for the appropriate incantation.

Requirements

    When considering an objects compliance with a protocol, 
    there are several cases to be examined:

     a) When the protocol is a type or class, and the object
        has exactly that type or is a member of the class.
        In this case compliance is automatic.

     b) When the object knows about the protocol and either
        considers itself compliant or knows how to wrap itself
        appropriately.

     c) When the protocol knows about the object and either
        it complies or can be wrapped accordingly.

     d) When the protocol is a class, and the object is a
        member of a subclass.  This is distinct from the first
        case (a) above, since inheritance does not necessarily
        imply substitutability and must be handled carefully.

     e) When the context knows about the object and the
        protocol and knows how to adapt the object so that
        the required protocol is satisfied.  This could use an
        adapter registry or similar method.

    For this proposal's requirements, the first case should be
    come for free and the next three cases should be relatively
    relatively easy to accomplish.  This proposal does not address 
    the last case, however it provides a base mechanism upon which 
    such an approach could be developed.  Further, with only minor
    implementation changes, this proposal should be able to 
    incorporate a new interface type or type checking system.

    The fourth case above is subtle. A lack of substitutability can
    occur when a method restricts an argument's domain or raises an 
    exception which a base class does not or extends the co-domain to
    include return values which the base class may never produce.  
    While compliance based on class inheritance should be automatic, 
    this proposal should allow an object to signal that it is not 
    compliant with a base class protocol.

Specification

    This proposal introduces a new built-in function, adapt, which
    is the basis for supporting these requirements. 

    The adapt function has four parameters: 1) the object to be
    adapted ("obj"), 2) the protocol requested of the object
    ("protocol"), 3) an optional object to return if the object 
    could not be adapted ("alternate"), and 4) an optional flag 
    which can be used to prevent a wrapper from being generated
    ("can_wrap") if provided.

    A successful result of the adapt function returns either
    the object passed ("obj") if the object is already compliant 
    with the protocol, or a secondary object ("wrapper"), which
    provides a view of the object compliant with the protocol.  
    The definition of wrapper is explicitly vague and a wrapper
    is allowed to be a full object with its own state if necessary.
    A failure to adapt the object to the protocol will raise a
    TypeError unless the alternate parameter is used, in this
    case the alternate argument is returned.

    To enable the first case listed in the requirements, 
    the adapt function first checks to see if the objects 
    type or the objects class are identical to the protocol.
    If so, then the adapt function returns the object directly
    without further ado.

    To enable the second case, when the object knows about the
    protocol, the object must have an __adapt__ method.  This 
    optional method takes three arguments, the object being adapted
    ("self"), the protocol requested ("protocol"), and a flag 
    ("can_wrap").  The object may return itself through this method 
    to indicate compliance.  Alternatively, if the can_wrap flag is
    true the object also has the option of returning a wrapper object
    compliant with the protocol.  Finally, if the object cannot 
    determine its compliance, it should either return None or 
    raise a TypeError to allow other mechanisms to be checked.

    To enable the third case, when the protocol knows about the object,
    the protocol must have a __prodapt__ method.  This optional method
    takes three arguments, the protocol requested ("self"), the
    object being adapted ("obj"), and a flag ("can_wrap").  If the 
    protocol finds the object to be compliant, it can return obj
    directly.  Alternatively, if the can_wrap flag is true, the 
    method may return a wrapper compliant with the protocol.  Finally,
    compliance cannot be determined, this method should either 
    return None or raise a TypeError so other mechanisms can be tried.

    The fourth case, when the objects class is a sub-class of the 
    protocol, is handled by the built-in adapt function.  Under normal
    circumstances, if isinstance(object,protcol) then adapt returns
    the object directly.  However, if the object is not substitutable,
    either the __adapt__ or __prodapt__ methods above may raise an
    adaptForceFailException to prevent this default behavior.

    This proposal could introduce one more built-in function, "isa",
    perhaps implemented as an operator.  If this is deemed useful, the
    function would take two arguments, the first is the object to be
    checked ("obj"), and the second is the protocol to check the
    object against ("protocol").  The return value of the operator
    will be either be the obj or None.  This function would be 
    implemented by calling the adapt function with the can_wrap being
    false and an alternate argument of None.

    Please note two important things.  First, this proposal does not
    preclude the addition of other protocols.  Second, this proposal
    does not preclude other possible cases where adapter pattern may
    hold, such as the context knowing the object and the protocol 
    (the last case in the requirements).  In fact, this proposal opens
    the gate for these other mechanisms to be added; while keeping the
    change in manageable chunks.

Reference Implementation and Example Usage

    -----------------------------------------------------------------
    adapter.py
    -----------------------------------------------------------------

    import types

    adaptRaiseTypeException = "(raise a type exception on failure)"
    adaptForceFailException = "(forced failure of adapt)"

    def adapt(obj, protocol, alternate = \
                   adaptRaiseTypeException, can_wrap = 1):

        # first check to see if object has the exact protocol
        if type(obj) is types.InstanceType and \
           obj.__class__ is protocol: return obj
        if type(obj) is protocol: return obj

        # procedure to execute on success
        def succeed(retval,obj,protocol,can_wrap):
            if can_wrap: return retval
            return obj

        # procedure to execute on failure
        def fail(obj,protocol,alternate):
            if alternate is adaptRaiseTypeException:
                raise TypeError("%s cannot be adapted to %s" \
                                 % (obj,protocol))
            return alternate

        # try to use the object's adapting mechanism
        adapt = getattr(obj, '__adapt__',None)
        if adapt:
            try:
                retval = adapt(protocol,can_wrap)
                if retval: return succeed(retval,obj,protocol,can_wrap)
            except adaptForceFailException:
                return fail(obj,protocol,alternate)
            except TypeError: pass

        # try to use the protocol's adapting mechanism
        adapt = getattr(protocol, '__prodapt__',None)
        if adapt:
            try:
                retval = adapt(obj,can_wrap)
                if retval: return succeed(retval,obj,protocol,can_wrap)
            except adaptForceFailException:
                return fail(obj,protocol,alternate)
            except TypeError: pass

        # check to see if the object is an instance
        try:
            if isinstance(obj,protocol):
                return obj
        except TypeError: pass

        # no-adaptation-possible case
        return fail(obj,protocol,alternate)

    # imagine binary operator syntax
    def isa(obj,protocol):
        return adapt(obj,protocol,None,0)

    -----------------------------------------------------------------
    test.py
    -----------------------------------------------------------------

    import adapter
    import types
    from adapter import adapt
    from adapter import isa
    from adapter import adaptForceFailException

    class KnightsWhoSayNi: pass

    class Eggs:  # an unrelated class/interface
        def eggs(self): print "eggs!"
        word = "Nee-womm"

    class Ham:  # used as an interface, no inhertance
        def ham(self): pass
        word = "Ping"

    class Spam: # a base class, inheritance used
        def spam(self): print "spam!"

    class EggsSpamAndHam (Spam,KnightsWhoSayNi):
        def ham(self): print "ham!"
        def __adapt__(self,protocol,can_wrap):
            if protocol is Ham:
                # implements Ham's ham, but does not have a word
                return self
            if protocol is KnightsWhoSayNi:
                # we are no longer the Knights who say Ni!
                raise adaptForceFailException
            if protocol is Eggs and can_wrap:
                # Knows how to create the eggs!
                return Eggs()
            raise TypeError("haha")

    class Bing (Ham):
        def __adapt__(self,protcol,can_wrap):
            raise adaptForceFailException

        # in this case, it is the protcol which determines
        # if the class is adapted
    class SacredWord:
        class HasSecredWord:
            def __call__(self, obj, can_wrap):
                if getattr(obj,'word',None): return obj
        __prodapt__= HasSecredWord()


    def test():
        x = EggsSpamAndHam()
        adapt(x,Spam).spam()
        adapt(x,Eggs).eggs()
        adapt(x,Ham).ham()
        adapt(x,EggsSpamAndHam).ham()
        print adapt(Eggs(),SacredWord).word
        print adapt(Ham(),SacredWord).word
        pass
        if adapt(x,KnightsWhoSayNi,None): raise "IckyIcky"
        if not isa(x,Spam): raise "Spam"
        if     isa(x,Eggs): raise "Eggs"
        if not isa(x,Ham): raise "Ham"
        if not isa(x,EggsSpamAndHam): raise "EggsAndSpam"
        if     isa(x,KnightsWhoSayNi): raise "NightsWhoSayNi"
        if     isa(x,SacredWord): raise "SacredWord"
        try:
            adapt(x,SacredWord)
        except TypeError: pass
        else: raise "SacredWord"
        try:
            adapt(x,KnightsWhoSayNi)
        except TypeError: print "Ekky-ekky-ekky-ekky-z'Bang, " \
                                + "zoom-Boing, z'nourrrwringmm"
        else: raise "NightsWhoSayNi"
        pass
        b = Bing()
        if not isa(b,Bing): raise "Not a Bing"
        if isa(b,Ham): raise "Not a Ham!"
        if isa(1,types.FloatType): raise "Not a float!"
        if isa(b,types.FloatType): raise "Not a float!"
        if isa(1,Ham):             raise "Not a Ham!"
        if not isa(1,types.IntType): raise "Is an Int!"

    -----------------------------------------------------------------
    Example Run
    -----------------------------------------------------------------
    >>> import test
    >>> test.test()
    spam!
    eggs!
    ham!
    ham!
    Nee-womm
    Ping
    Ekky-ekky-ekky-ekky-z'Bang, zoom-Boing, z'nourrrwringmm


Relationship To Paul Prescod and Tim Hochbergs Type Assertion method

    Paul and Tim had proposed an type checking mechanism, where the
    Interface is passed an object to verify.  The example syntax Paul
    put forth recently [2] was:

        interface Interface
            def __check__(self,obj)

    This could be re-written to avoid the "interface" keyword as:

        class Interface:
            class Checker:
                def __call__(self, obj): pass  #check the object
            __check__= Checker()

    As I understand it, __prodapt__ is essentially __check__, only 
    that the function is also given the ability to substitute a 
    wrapper for the object if can_wrap is set.  Further, __prodapt__
    returns an object instead of true/false, however, this seems
    a trivial difference.  Indeed, a bulk of the idea for this 
    proposal was built directly from Paul and Tims work.

    In short, the work put forth by Paul and company is great, and
    I dont see any problems why these two proposals couldnt work
    together in harmony, if not be completely complementary.

Relationship to Python Interfaces [1] by Michel Pelletier

    The relationship to this proposal to Michels proposal could
    also be complementary.  If approved, the first two paragraphs
    of the built-in adapt function could be changed as follows:

        # first check to see if object has the exact protocol
    >   if type(protocol) is types.InterfaceType and \
    >       instance(obj,protocol): return obj
        if type(obj) is types.InstanceType and \
           obj.__class__ is protocol: return obj
        if type(obj) is protocol: return obj

        # procedure to execute on success
        def succeed(retval,obj,protocol,can_wrap):
    >       if type(protocol) is types.InterfaceType:
    >           if not instance(retval,protocol):
    >               raise "Bad __adapt__ or __prodapt__!"
            if can_wrap: return retval
            return obj

Relationships To Microsofts Query Interface:

    Although this proposal may sounds similar to Microsofts
    QueryInterface, it differs by a number of aspects.  First,
    there is not a special "IUnknown" interface which can be used
    for object identity, although this could be proposed as one
    of those "special" blessed interface protocol identifiers.
    Second, with QueryInterface, once an object supports a particular
    interface it must always there after support this interface;
    this proposal makes no such guarantee, although this may be
    added at a later time. Third, implementations of Microsofts
    QueryInterface must support a kind of equivalence relation.
    By reflexive they mean the querying an interface for itself
    must always succeed.  By symmetrical they mean that if one
    can successfully query an interface IA for a second interface
    IB, then one must also be able to successfully query the
    interface IB for IA.  And finally, by transitive they mean if
    one can successfully query IA for IB and one can successfully
    query IB for IC, then one must be able to successfully query
    IA for IC.  Ability to support this type of equivalence relation
    should be encouraged, but may not be possible.  Further research
    on this topic (by someone familiar with Microsoft COM) would be
    helpful in further determining how compatible this proposal is.

Backwards Compatibility

    There should be no problem with backwards compatibility unless
    someone had used __adapt__ or __prodapt__, but this seems 
    unlikely.  Indeed this proposal, save an built-in adapt() 
    function, could be tested without changes to the interpreter.

Questions and Answers

    Q:  Why was the name changed from __query__ to __adapt__ ?

    A:  It was clear that significant QueryInterface assumptions were
        being laid upon the proposal, when the intent was more of an
        adapter.  Of course, if an object does not need to be adapted
        then it can be used directly and this is the basic premise.

    Q:  Why is the checking mixed with the adapter mechanism?

    A:  Good question.  They could be separated, however, there
        is significant overlap, if you consider the checking
        protocol as returning a compliant object (self) or
        not a compliant object (None).  In this way, adapting
        becomes a special case of checking, via the can_wrap.

        This could be separated out, but the two concepts are 
        very related so much duplicate work would be done, and 
        the overall mechanism would feel quite a bit less unified.

    Q:  I like/dont like the isa function/operator.

    A:  It is separable from this proposal, it just seems like
        a handy short-hand for checking rather than adapting.

    Q:  The name __prodapt__ sucks.

    A:  Suggest a better alternative; I suppose __check__ could
        be used for the protocol side, however I dont want to
        trample on Paul Prescods proposal.

Credits

    This proposal was created in large part by the feedback
    of the talented individuals on both the main mailing list
    and also the type signature list.  Specific contributors
    include (sorry if I missed someone).

    This proposal is based largely off the suggestions from
    Alex Martelli and Paul Prescod with significant feedback
    from Robin Thomas and borrowing ideas from Marcin 'Qrczak'
    Kowalczyk.  Other contributors (via comments) include:

        Michel Pelletier, Jeremy Hylton, Carlos Ribeiro,
        Aahz Maruch, Fredrik Lundh, Rainer Deyke,
        Timothy Delaney, and Huaiyu Zhu

Copyright

    This document has been placed in the public domain.


References and Footnotes

    [1] http://python.sourceforge.net/peps/pep-0245.html
    [2] http://mail.python.org/pipermail/types-sig/2001-March/001223.html
    [3] http://www.zope.org/Members/michel/types-sig/TreasureTrove
    [4] http://mail.python.org/pipermail/types-sig/2001-March/001105.html
    [5] http://mail.python.org/pipermail/types-sig/2001-March/001206.html
    [6] http://mail.python.org/pipermail/types-sig/2001-March/001223.html









From fredrik at pythonware.com  Fri Mar 16 12:40:23 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 16 Mar 2001 17:40:23 GMT
Subject: Python style questions
References: <98tb91$9t9$1@saltmine.radix.net> <mailman.984759577.27755.python-list@python.org>
Message-ID: <bess6.179$sk3.52516@newsb.telia.net>

"D-Man" <dsh8290 at rit.edu> wrote:
> Curiosly enough, print's format string is supposed to be identical to
> C's printf format string.  However the man page for printf says the %x
> prints 0x then the number using A-F for 10-15.

better check that page again; printf should add 0x only if you
use the # flag.

Cheers /F




From aleaxit at yahoo.com  Fri Mar 23 07:40:44 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 23 Mar 2001 13:40:44 +0100
Subject: PartialInterfaces ? Re: Checking the type
References: <mailman.985343914.18681.python-list@python.org>
Message-ID: <99fg8k01sr0@news2.newsguy.com>

"Clark C. Evans" <cce at clarkevans.com> wrote in message
news:mailman.985343914.18681.python-list at python.org...
> > # safer
> > def munge(alist):
> >     # first extract all bound-methods we'll need
> >     append = alist.append
> >     extend = alist.extend
    [snip]
> This is a very cool, "protocol", a partial interface.

Why, thanks.  Wish I could claim originality:-), but,
as I said, this (or a close analog thereof) is rather
standard fare in C++ templates.

> I was thinking about how this could be formalized as
> a "checker/adapter"... here is very imperfect code)
>
> class partial_list_checker:
>       def __init__(self, append=0,extend=0,index=0,...
>           self.check_append=append
>           self.check_extend=extend
>           self.check_index =index
>           ...
>
>       def check(self,alist)
>           try:
>               if self.check_append: append = alist.append
>               if self.check_extend: extend = alist.extend
>               if self.check_index:
>                  try: alist[0] = alist[0]
>                  except IndexError: pass
>               ...
>               return alist
>           except AttributeError
>              return None
>
> What do you think?  Kinda slow, hunh?

Rather.  Maybe slightly better:

class checker_of_methods:
    def __init__(self, *method_names_needed):
        self.method_names_needed = method_names_needed
    def check(self, anobj):
        return [ getattr(anobj, method)
            for method in self.method_names_needed ]

I'd let the AttributeError propagate up to the caller
of achecker.check if the object being checked does not
supply a needed method -- seems simples/smoothest.  And
I think returning the already-bound methods is also a
(very slight) optimization as well as a simplification.

Of course, this doesn't handle the specific check for
indexability (alist[0]=alist[0]), though I guess that
could be added as a special case (maybe in a derived
class of checker_of_methods specialized for sequence
checking).

Anyway, usage here might be, e.g.:
    append, extend = checker_of_methods(
        'append', 'extend').check(alist)


Alex





From news at dorb.com  Sat Mar  3 09:29:10 2001
From: news at dorb.com (Darrell)
Date: Sat, 03 Mar 2001 14:29:10 GMT
Subject: playing with IP's
References: <mailman.983524688.21332.python-list@python.org>
Message-ID: <Wc7o6.172046$v.15307723@typhoon.nyroc.rr.com>

"Langa Kentane" wrote:
> Greetings.
> Here is what I wanna do:
> Given a network number and mask ie:
> 192.168.1.0/32
> I want to create a list with all the valid hosts on that range.
> Problem is I don't know how to play with decimal.
>
> Where would I start.. any sample code?
> Thanks
>
I don't know a lot about this, but here's a starting point.
Be sure to post any improvements :)
It might be handy to print out the subnet boundary's.

--Darrell

#!/usr/bin/env python
import re

def shiftOctents(octents):
    """
    Convert dot separated IP format to a number
    This might exist in the socket module?
    """
    octents=octents.split(".")
    octents.reverse()
    v=0
    for x in range(len(octents)):
        v|= int(octents[x]) << (x*8)
    return v

    # Bytes used for local addresses in a class of IP
_netWorkClassDict={'a':3, 'b':2, 'c':1}

def getValues(input, netWorkClass='b'):
    """
    Return IP converted to a number
    subnet address size
    host address size
    """
    netWorkClass=netWorkClass.lower()

    ip, mask = input.split("/")
    mask = shiftOctents(mask)
    hosts = ~mask & 0xffffffffL -1

    val='0x'+'ff'*_netWorkClassDict[netWorkClass]
    netWorkClassMask = int(val, 16)
    subnet=max(0, (netWorkClassMask/(hosts+2)) -1)

    ip= shiftOctents(ip)
    return ip, subnet, hosts

def backToDotNotation(val):
    res=[]
    tempVal=val
    for x in range(4):
        res.append(tempVal & 0xff)
        tempVal = tempVal >> 8
    res.reverse()
    return "%d.%d.%d.%d"%tuple(res)

def test1():
    input=["192.168.1.0/255.255.0.0","192.168.1.0/255.255.240.0",
    "192.168.1.0/255.255.254.0","192.168.1.0/255.255.255.252"
    ]
    expected=[(0, 65534), (14, 4094),(126, 510), (16382, 2)]

    netWorkClass='b'
    for ix in range(len(input)):
        i=input[ix]
        print "Using:", i
        print "Class:", netWorkClass
        values=getValues(i, netWorkClass)
        assert(values[1:]==expected[ix])
        print 'Number of hosts for this subnet:', values[2]
        print 'Ending IP for this subnet:', backToDotNotation(values[0] |
values[2])
        print 'Number of subnets:', values[1]
        print

test1()





From egbert at bork.demon.nl  Thu Mar  8 13:40:29 2001
From: egbert at bork.demon.nl (Egbert Bouwman)
Date: Thu, 8 Mar 2001 19:40:29 +0100
Subject: Get terminal size in perl and python
Message-ID: <20010308194028.A1616@bork.demon.nl>

My perl correspondent uses:
    use Term::ReadKey;
    ($cols,$lines)=GetTerminalSize()
What is the python equivalent ?
If I don't come up with one, python looses ten points.
egbert
-- 
Egbert Bouwman - Keizersgracht 197 II - 1016 DS  Amsterdam - 020 6257991
========================================================================



From grante at visi.com  Thu Mar 22 16:10:01 2001
From: grante at visi.com (Grant Edwards)
Date: Thu, 22 Mar 2001 21:10:01 GMT
Subject: Good Editor?
References: <mailman.985291768.27561.python-list@python.org>
Message-ID: <JStu6.1735$SB2.192020@ruti.visi.com>

In article <mailman.985291768.27561.python-list at python.org>, Bruce Sass wrote:
>On Thu, 22 Mar 2001, fjs wrote:
>
>> Sorry folks,
>>
>> I use both windows and linux.  So, all the recomendations have been helpful.
>
>I'm wondering why JED wasn't mentioned... it does a good job with
>tweaking scripts, writing small programs, and the syntax
>highlighting doesn't get confused as easily as with emacs.

Jed is what I use under both Linux and MS-Windows.  Python mode
is pretty decent.

-- 
Grant Edwards                   grante             Yow!  I have a very good
                                  at               DENTAL PLAN. Thank you.
                               visi.com            


From kumo at bellsouth.net  Wed Mar 14 11:28:11 2001
From: kumo at bellsouth.net (David Rush)
Date: 14 Mar 2001 16:28:11 +0000
Subject: Einstein's Riddle
References: <oNPq6.396644$ge4.138479566@news2.rdc2.tx.home.com> <wi6elw1pg5y.fsf@merlin.irisa.fr> <3AAE9730.5A7A4249@brazee.net> <slrn9atjsc.5j.scarblac@pino.selwerd.nl> <3AAF8740.907533CC@brazee.net>
Message-ID: <okfae6oiack.fsf@bellsouth.net>

Howard Brazee <howard at brazee.net> writes:
> The original poster must have used a Spammer tool.  This thread is cross-posted
> elsewhere as well, and I have seen one other coded solution, in Forth.

I've seen far too many.

> It would be fun to see a bunch of language solutions, first to solve this problem,
> and then a more generic version to solve this class of problems.

Set up a website, then. Quit spamming the world.

david rush
-- 
Einstein said that genius abhors consensus because when consensus is
reached, thinking stops. Stop nodding your head.
	-- the Silicon Valley Tarot


From max at alcyone.com  Sun Mar 25 03:00:19 2001
From: max at alcyone.com (Erik Max Francis)
Date: Sun, 25 Mar 2001 00:00:19 -0800
Subject: Python Mega Widgets rpm spec file?
References: <vwgv6.1085$107.1233792@typhoon.ne.mediaone.net>
Message-ID: <3ABDA593.86CA0E02@alcyone.com>

rodrigc at mediaone.net wrote:

> Grayson's book refers to Python Mega Widgets (Pmw) quite
> a bit, but this does not come standard with Redhat distributions.
> Has anyone made an rpm spec file for Pmw, so that it
> can be installed as an rpm package?

It's available here.  There's nothing to compile, so it's really
trivial.  If you're so inclined, you could make your own rpm.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ Triumph cannot help being cruel.
\__/ Jose Ortega y Gasset
    Interstelen / http://www.interstelen.com/
 A multiplayer, strategic, turn-based Web game on an interstellar scale.


From claird at starbase.neosoft.com  Wed Mar 21 12:03:33 2001
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 21 Mar 2001 11:03:33 -0600
Subject: Accessing Oracle from Python remotely ?
References: <Xns906AB9C901807PaCmAnRDLM@161.48.76.20> <eE3u6.776$9%.203510@newsr1.maine.rr.com>
Message-ID: <F92141C233DA318B.A84C85D98997C694.A9951520D4A22E7C@lp.airnews.net>

In article <eE3u6.776$9%.203510 at newsr1.maine.rr.com>,
gkiffney <gkiffney at my-deja.com> wrote:
>There's http://www.zope.org/Products/DCOracle which should get you started.
>There was an article by Uche Ogbuji in Linuxworld that helped me get going
>with this, but it's apparently fallen off the Internet map into the bit
>bucket.  Using the dcoracle module (which is a bit of a pain to get going,
			.
			.
			.
Do you mean
<URL: http://www.linuxworld.com/lw-2000-04/lw-04-oracle8i.html >?
-- 

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


From kare at speech.kth.se  Thu Mar 15 11:43:15 2001
From: kare at speech.kth.se (Kare Sjolander)
Date: Thu, 15 Mar 2001 17:43:15 +0100
Subject: ANNOUNCE: The Snack Sound Toolkit v2.0.7
Message-ID: <3AB0F123.F1E2D432@speech.kth.se>

This release contains:

* Several enhancements

* Several important bug-fixes, e.g. related to
  - MP3 decoding
  - sound with > 2 channels

* Updated demos



Change log

2/6/01 (bug fix) Fixed bug in the Macintosh audio code. The sound stop
command now stops sound output without any delay.

2/25/01 (bug fix) Fixed bug in the Windows audio code which limited
output to two channels even though the hardware was capable of more.

3/1/01 (enhancement) The -fileformat option to the sound, read, data,
append, configure, and cget commands now accept an empty string as
argument indicating that no file format was specified.

3/1/01 (bug fix) A number of fixes which gives much more stable MP3
decoding. (Wilkason)

3/1/01 (enhancement) Enhanced file format detection to discriminate
between raw sound files and MP3 files.

3/1/01 (bug fix) Fixed endianess problem in tests/data.test

3/5/01 (configuration improvement) Fixed configure script for FreeBSD to
not use ossaudio library.

3/9/01 (bug fix) Fixed memory leak when playing -file sounds.

3/13/01 (bug fix) Fixed problem for sounds linked to disk files with
more than two channels.

3/13/01 (enhancement) Precomputed waveform shapes now handle
multi-channel sounds. The shape file is now stored in AIFF format with
Lin8 encoding.

3/13/01 (enhancement) The wrap.tcl demo is now more intelligent in
creating its temporary work directory on the Windows platform.

3/14/01 (bug fix) Bartlett analysis window corrected. (Rank)



The Snack Sound Toolkit is designed to be used with a scripting
language.
Currently it supports Tcl/Tk and Python.
Snack adds commands to play, record, and process sound and supports
in-memory sound objects, file based audio, and streaming audio.
It handles fileformats such as WAV, MP3, AU, AIFF, and NIST/Sphere.
Snack is extensible, new commands, filters, and sound file formats can
be added using the Snack C-library. An easy to build example extension
is contained in the source distribution.
Snack also does sound visualization, e.g. waveforms and spectrograms.
The visualization objects update in real-time and can output postscript.

Snack works with Tcl8.0 - Tcl8.4 and Python 1.5.2-2.0
Platforms: Linux, Solaris, HP-UX, IRIX, NetBSD, Macintosh, and
Windows 95/98/NT/2K.

Source and binaries can be downloaded from
http://www.speech.kth.se/snack/

Regards,
Kare Sjolander
kare at speech.kth.se




From mcalla at home.com  Thu Mar 15 16:56:30 2001
From: mcalla at home.com (Mike Callahan)
Date: Thu, 15 Mar 2001 21:56:30 GMT
Subject: Communicating with serial port
References: <ndNn6.5161$t21.152593@news3.oke.nextra.no>
Message-ID: <iUas6.7165$Q47.1945583@news1.rdc1.tn.home.com>

Peter,

I recently finished a program using Roger's latest version of Sio and it
worked well with Python 2.0.

Mike Callahan

"Peter Blom" <pet-blo at online.no> wrote in message
news:ndNn6.5161$t21.152593 at news3.oke.nextra.no...
> I shall write a python program for downloading data from a memory card in
a
> measuring instrument, via a serial port.
> For this purpose, I have downloaded SioModule.zip - a serial comm
extension
> for windows 95, upgraded to Python 1.5.1.
> This does not seem to work - when importing the module 'serial.py', my
> system hangs (Windows 98, Activestate Python 2.0).
> Could anybody give a hint?
>
> Peter (a newbie)
>
>




From jeremy at alum.mit.edu  Fri Mar 16 17:33:25 2001
From: jeremy at alum.mit.edu (Jeremy Hylton)
Date: Fri, 16 Mar 2001 17:33:25 -0500 (EST)
Subject: Type mismatchings and the rank newbie
In-Reply-To: <ylws6.2639$227.274111@newsread2.prod.itd.earthlink.net>
References: <ylws6.2639$227.274111@newsread2.prod.itd.earthlink.net>
Message-ID: <15026.38069.279933.21223@w221.z064000254.bwi-md.dsl.cnc.net>

>>>>> "JB" == John Brawley <jgbrawley at earthlink.net> writes:

  JB> Rank newbie here, trying to do something specific.  I need to
  JB> use a set of numbers (say, 1 to 25), and create 25 lists from
  JB> the numbers (1 thru 25), which lists have a character "P" and a
  JB> number (one of the 25).

  JB> The lists I want look like : 
      P1=[x, y, z] 
      P2=[x, y, z] 
      P3=[x, y, z] 
      (etcetera down to P25)

[...]

  JB> I need to write a loop that takes each number in sequence and
  JB> sticks it next to the ascii "P" so that when the loop is
  JB> finished, I end up with a set of lists each of whose names is
  JB> different by one number.

  JB> The loop is not a problem; I just can't figure how to stick a
  JB> different number form the user input total, into the second
  JB> place of the list name ( Pn=[x, y, z] where 'n' is the number
  JB> stuck in there by the loop) with every iteration of the loop.

  JB> The docs are confusing to me (I'm not a programmer and don't
  JB> want to be, but there is no other way for me to do what I need
  JB> to to see what this program will display for me).

  JB> Can anyone suggest a way out of this hole?

The point you're stuck at is: You have an integer, say, 12, and you
want to make an assignment to the name P12.  Is that right?

If so, then you need some a helper object that can direct you from the
number 12 to the variable P12.  There is no good way to invent the
variable name on the fly.  (It is technically possible using exec, but
this is fairly complicated and can be hard to understand.)

Is it absolutely necessary to use variables names P1, P2, etc?  It
would be simpler to use a dictionary.

    P = {}
    P[1] = [x, y, z]
    P[2] = [x, y, z]

If that's not possible, here is another, slightly more subtle
suggestion. 

    # These are place holders for the real values, which will be
    # created based on user input.
    P1 = []
    P2 = []
    ...
    P25 = []

    Plist = [P1, P2, P3, ..., P25]

    # get the user input here
    n = get_number()
    Plist[n].extend([x, y, z])
    # then Plist[n] => "Pn" => [x, y, z]

The above code works using references to the lists.  In Python, names
create references to objects.  Lists are mutable objects.  When you
call the extend method on a list, it modifies the list in place and
adds the elements of its argument.

>>> l = []
>>> l.extend([1])
>>> l
[1]
>>> l.extend([2, 3])
>>> l
[1, 2, 3]

Then you can bind the same object to a new name or put it in a list.

>>> l2 = l
>>> l2
[1, 2, 3]
>>> l.append(4)
>>> l
[1, 2, 3, 4]
>>> l2
[1, 2, 3, 4]
>>> list_of_lists = [l]
>>> list_of_lists
[[1, 2, 3, 4]]
>>> list_of_lists[0].append(5)
>>> list_of_lists
[[1, 2, 3, 4, 5]]
>>> l
[1, 2, 3, 4, 5]

Hope this helps.

Jeremy




From farmerk at pacbell.net  Tue Mar  6 04:29:39 2001
From: farmerk at pacbell.net (Keith Farmer)
Date: Tue, 6 Mar 2001 01:29:39 -0800
Subject: win32, GUI AppBar class
Message-ID: <a72p6.7151$qv3.3197162@nnrp5-w.sbc.net>

Has anybody implemented an AppBar class in Win32?  That is, a window that
docks to the edge of the screen (such as TaskBar, Launch Bar, ICQ, AIM,
Office Bar, etc)?

Looking through the PythonWin docs, no support seems inherent for this
(dockable control bars, but not full-fledged windows).  I was thinking the
only choice would be calldll, but being in Python 2.0, I discovered npstruct
(potentially necessary for this task) doesn't seem to be compiled for it
yet.

I'd be most interested in docking/auto-hide support from stock win32/python
libraries -- like Tkinter and PyWin -- though I'm thinking a mixin class
would be really nifty, and allow for the creation of Python-based wharfs,
etc.

Keith Farmer
SF, CA




From sheila at spamcop.net  Sun Mar 18 01:08:27 2001
From: sheila at spamcop.net (Sheila King)
Date: Sun, 18 Mar 2001 06:08:27 GMT
Subject: reading stdinput
References: <mailman.984885323.10683.python-list@python.org>
Message-ID: <a4k8bto792ealce3e382i9jtm8n2e65p4c@4ax.com>

On Sun, 18 Mar 2001 11:11:00 +0930 (CST), Michael Hall <admin at mulga.com.au>
wrote in comp.lang.python in article
<mailman.984885323.10683.python-list at python.org>:

:How do I read standard input into a Python script? I'm trying to do the
:same as 'read' in a shell script to create a variable. I've played about
:with various combinations of 'sys.stdin.read()' and such but no go.

Here's a really simple little script, that reads from stdin and writes the
result to stdout. I just re-tested it now, to make sure it works:

#!/usr/bin/env python

import sys
line = sys.stdin.readline()
print line


--
Sheila King
http://www.thinkspot.net/sheila/
http://www.k12groups.org/



From kens at sightreader.com  Fri Mar 16 12:18:25 2001
From: kens at sightreader.com (Ken Seehof)
Date: Fri, 16 Mar 2001 11:18:25 -0600
Subject: Cookies
References: <mailman.984707839.22839.python-list@python.org> <0Dhs6.576315$U46.17170213@news1.sttls1.wa.home.com>
Message-ID: <004d01c0ae3d$1cbfbde0$04090a0a@upcast.com>

Yeah, that's the closest I've come up with too, but the difficulty is making
the system reliably support all browsers.  Unfortunately from the point of
view of a standalone application, there is no standard for cookies.  Cookies
are standardized from the point of view of servers and javascript, but that
doesn't help me with a standalone app.

The best approach so far seems to be to have the server send back a page
with a 10 digit unique identifier in a large font.  The python app would
then ask the user to type in the number manually.  Seems kinda dumb that I
need the users fingers as part of my protocol. :-)

- Ken

----- Original Message -----
From: "Greg Jorgensen" <gregj at pobox.com>
Newsgroups: comp.lang.python
To: <python-list at python.org>
Sent: Thursday, March 15, 2001 11:35 PM
Subject: Re: Cookies


> A cookie is just a text file stored in a special place where the browser
can
> find it. Where exactly the cookies go depends on your OS and browser.
>
> Using a text file to store the key would seem the easiest solution, but
> client-side Javascript can't create or write a file.
>
> I think you should generate unique IDs on the server, not on the client.
> Here's one way to handle it:
>
> 1. Client browser sends form data to server.
> 2. Server processes form, generates a unique ID, stores the form data
keyed
> by the unique ID.
> 3. Server replies with a "thank you" or "database updated" page that also
> writes a cookie.
> 4. Python program starts up and looks for the cookie. Depending on your OS
> and browser the location and name of the cookie will vary, but with a
little
> experimenting you can easily find the most likely locations.
> 5. Python program reads the cookie. Open your cookie with a text editor to
> see what's in there--it's just name: value pairs.
> 6. Python program sends HTTP request to server with unique ID it got from
> the cookie.
>
> Obviously this won't work if the client has disabled cookies. With a
little
> more trouble you can detect this and perhaps put the unique ID on the
> response page and tell the user to save it or something. Or embed the
unique
> ID in the reply page and tell the user to save that page and give it a
> specific name, then have your Python program open that text file and find
> the unique ID in there.
>
> Depending on your requirements and security concerns, you can either have
> your database program generate the unique ID (my preference), use the
> Windows GUID functions (which are available to ASP on the server), or
> implement one of the GUID functions you can easily find on the web. Many
> programmers have solved this in one way or another; send email to me if
you
> want specific advice on generating unique IDs from your database engine.
>
> Good luck.
>
> <<<
> "Ken Seehof" <kens at sightreader.com> wrote in message
> news:mailman.984707839.22839.python-list at python.org...
> Short version:
>
> Is it possible for a stand-alone python application to read a cookie
without
> the help of a browser?"
>
> Long version:
>
> I have a web page W and a client side python script P.  I want W to
generate
> a GUID (perhaps using javascript) which is then sent to a server S, where
it
> is to be used as a key in a database.  Now I run P (on the client, and not
> in a browser) and I want P to somehow find the value of the GUID, so that
it
> can query the database on S.
>
> Alternate question:
>
> Is there a technique common to both javascript and python that will obtain
a
> unique identifier for a computer (e.g. IP address).  It must not be
browser
> specific.
>
> Motivation:
>
> I need to write a record of data from an html form, keyed by some kind of
> unique client id, to the server and then be able to query that record from
a
> stand alone python script running on the client.  The missing link is
> getting the html form and the python script to agree on a value for the
id;
> the rest is easy.
> >>>
>
> --
> Greg Jorgensen
> PDXperts
> Portland, Oregon, USA
> gregj at pobox.com
>
>
> --
> http://mail.python.org/mailman/listinfo/python-list
>




From bedge at troikanetworks.com  Thu Mar  1 19:28:23 2001
From: bedge at troikanetworks.com (Bruce Edge)
Date: Thu, 01 Mar 2001 16:28:23 -0800
Subject: define a new func on the fly?
Message-ID: <3A9EE927.8744154E@troikanetworks.com>

I'm trying to define a new function on the at runtime,

this fails:

>>> eval( "def global_func_name():\n\tpass" )
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
  File "<string>", line 1
    def global_func_name():
      ^
SyntaxError: invalid syntax

any idea how I can do this?



Thanks, Bruce.


From aleaxit at yahoo.com  Thu Mar 15 05:26:52 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 15 Mar 2001 11:26:52 +0100
Subject: Who's minister of propaganda this week?
References: <mailman.984595938.11127.python-list@python.org> <98otuq0nnf@news1.newsguy.com> <slrn9b0cc2.gsa.neelk@alum.mit.edu>
Message-ID: <98q6ro2uoa@news2.newsguy.com>

"Neelakantan Krishnaswami" <neelk at alum.mit.edu> wrote in message
news:slrn9b0cc2.gsa.neelk at alum.mit.edu...
> On Thu, 15 Mar 2001 00:09:30 +0100, Alex Martelli <aleaxit at yahoo.com>
wrote:
> >"Steven D. Majewski" <sdm7g at virginia.edu> wrote in message
> >news:mailman.984595938.11127.python-list at python.org...
> >    [snip]
> >> ( OK: not having multiple dispatch in Python means that some manual
> >> dispatching is sometimes required, but usually this is burried deep
> >> in some classes __coerce__ method. )
> >
> > "Usually" in arithmetic, perhaps.  For fancier stuff, lack of
> > multimethods sometimes cramps (not that it cramps any less in C++,
> > mind you), though some patterns (such as 'dynamic visitor') can help
> > a bit in certain (reasonably frequent) special cases.
>
> What's a dynamic visitor? Is it a dictionary keyed on class? That's
> what I usually do when emulating multimethods in Python.

Sorry, the pattern-name is _Acyclic_ Visitor -- it's fully
described by its author, Robert Martin, in the PDF at URL
http://www.objectmentor.com/publications/acv.pdf for example.

The 'dynamic' aspect is at its core (in the C++ version, it
uses dynamic_cast systematically, for example) but not in
the name (because the key _goal_ of it, vs regular Visitor,
is framed as removing a very troublesome cyclic dependency).

Keying a dict on class has the unfortunate side-effect of
not allowing polymorphism by inheritance -- objects of classes
_derived from_ the one used as the key will not match.  You
can preserve this basic idea by using as key, instead of
the class, some ad-hoc attribute which will be inherited (if
it IS acceptable to alter the class-objects of interest by
adding an attribute) except where specifically overridden,
an approach which can be used as a first-path (fast-path)
to be followed by an isinstance loop (slow but sure) if and
when the key attribute is not obtained (e.g. for types that
are not classes, or, classes that can't be altered).

I don't think the applicability cases of this approach to
hand-simulation of multiple dispatch will overlap much
with those of visitors (of any kind) or coercion approaches.


Alex





From phlip_cpp at my-deja.com  Mon Mar 12 15:15:31 2001
From: phlip_cpp at my-deja.com (Phlip)
Date: Mon, 12 Mar 2001 12:15:31 -0800
Subject: PyEval_CallObject and threads
References: <mailman.984426376.20428.python-list@python.org>
Message-ID: <aaar6.32$P26.59620@news.pacbell.net>

Tim Peters wrote:

> [Phlip]
>> My colleague came over and asked me to write this:
>>
>> "We got multiple C++ pthreads calling PyEval_CallObject. How threadsafe
>> is the Python API? A C++ module using Boost BPL creates multiple threads,
>> which then call back into Python. When two threads simultaneously call
>> PyEval_CallObject we go boom."
>>
>> I will now go examine his serialization primitives. But if anyone has any
>> better ideas...
> 
> With very few exceptions, all calls to Python C API functions must be made
> while holding the global interpreter lock.  The exceptions have mostly to
> do
> with a handful of functions related to initialization and shutdown.  See
> the section "Thread State and the Global Interpreter Lock" in the Python/C
> API Reference Manual for details.

Further inspection reveals our architect(s) think we need to call Python 
script asynchronously from multiple threads. By 'lock' we assume both you 
and the section you cite mean 'block'. Isn't this what Stackless Python is 
for?

-- 
  Phlip                          phlip_cpp at my-deja.com
============== http://phlip.webjump.com ==============
  --  Proud victim of the dreaded boomerang effect  --


From garabik-news at spam.melkor.dnp.fmph.uniba.sk  Tue Mar 13 13:00:42 2001
From: garabik-news at spam.melkor.dnp.fmph.uniba.sk (Radovan Garabik)
Date: 13 Mar 2001 18:00:42 GMT
Subject: Color curses and Python 2?
References: <slrn9amlnn.ktp.jimd@vega.starshine.org> <slrn9anibn.i3.amk@mira.erols.com> <slrn9apqvc.ktp.jimd@vega.starshine.org>
Message-ID: <984506329.356361@cdwork.cvt.stuba.sk>

Jim Dennis <jimd at vega.starshine.org> wrote:
 :  I can categorically guarantee that my terminal/console supports
 :  common ncurses color (since I use it with vim, slrn, w3m, lynx,
 :  xemacs, dialog, and the Perl Curses.pm module among others).

 :  Heck I can even use color in shell using tput!  (Please don't
 :  suggest that I use external calls to tput in my Python code;
 :  that's not the question here).

 :  As I said, I haven't seen a 1.6.x package for Debian.  I gather
 :  that the 1.5.x versions don't have ncurses/color support (I 
 :  compiled my own for that).  Part of the question at hand is whether
 :  the 2.x betas were supposed to have color support.  

well, I am using python2 as distributed with woody and color
definitely works....


-- 
 -----------------------------------------------------------
| Radovan Garabik http://melkor.dnp.fmph.uniba.sk/~garabik/ |
| __..--^^^--..__    garabik @ melkor.dnp.fmph.uniba.sk     |
 -----------------------------------------------------------
Antivirus alert: file .signature infected by signature virus.
Hi! I'm a signature virus! Copy me into your signature file to help me spread!


From claird at starbase.neosoft.com  Fri Mar 23 09:25:44 2001
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 23 Mar 2001 08:25:44 -0600
Subject: Tkinter - Widget Colours
References: <mailman.985345890.22443.python-list@python.org>
Message-ID: <3CD847F90A14AEA6.B799B72D55BF6880.28EDEA4DC41BBA47@lp.airnews.net>

In article <mailman.985345890.22443.python-list at python.org>,
O'Rourke Clodagh-corour01 <corour01 at motorola.com> wrote:
>Hi all,
>
>I was wondering does anyone know how the bag/fag colours can be changed after the widget is packed.  
			.
			.
			.
Or before, even:
  mybutton.configure(foreground = "red")
-- 

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


From jmarshal at mathworks.com  Tue Mar 20 14:33:27 2001
From: jmarshal at mathworks.com (Joshua Marshall)
Date: 20 Mar 2001 19:33:27 GMT
Subject: Sequence-spreading
References: <3AB7167F.2816AA21@student.liu.se> <99790u01c73@news2.newsguy.com> <99851p$ba6$2@neptun.beotel.net>
Message-ID: <998ba7$5j7$1@news.mathworks.com>

gzeljko <gzeljko at sezampro.yu> wrote:

> Alex Martelli <aleaxit at yahoo.com> wrote in message
>> offsets = [ [] for i in range(pm) ]
>> for i in range(len(arg)):
>>     offsets[i%pm].append(i)
>>
>>
>> Alex
>>

> Old syntax yet works:

> offsets = [[]]*pm

Careful here:

 >>> x = [[]]*2
 >>> y = [[] for i in range(2)]
 >>> x
 [[], []]
 >>> y
 [[], []]
 >>> x[0] is x[1]
 1
 >>> y[0] is y[1]
 0

The first way gives you the same value at each element ("[]" is
evaluated once), while the second way gives you distinct values ("[]"
is evaluated for each element).


From nspurrier  Fri Mar 16 22:52:52 2001
From: nspurrier (at)
Date: 17 Mar 2001 03:52:52 GMT
Subject: md5 passwords
References: <mailman.984779122.3483.python-list@python.org>
Message-ID: <98un2k62shv@news1.newsguy.com>

This is off the top of my head... buyer beware.

The passwords may have salt added to them. I know this is done
for crypt, but I'm not sure if this is done for all
other hash algorithms. With crypt the salt is the
first two characters (12 bits). 

On my OpenBSD system (which uses Blowfish) my passwords look like:
$2a$06$M43SrlFfnlLoFmK3ZAtnfquh3PsM/QZNdNNAwBEse1ReOrV4IjYeMs
The first 128 bits are defined as the salt -- which I think is
the first 4 characters if it's encoded in base 64.

If you want to send me some of your password file with
the plaintext passwords, then I will take a quick look
to see if I can figure out what they are doing.

Yours,
Noah Spurrier
P.S. Do you understand salt? You create a random salt 
before starting.  You add that to the plaintext password then run 
the hash. You store the salt and hash in the password file. 
To check a password you take the salt from the password file
and add it to the plaintext password you are testing; run
the hash algorithm and check against the encrypted password.
This just help to prevent people from taking a huge dictionary
and running the hash on each word then check to see if any match
the hashes in your password file.

> Can anyone explain to me how to use the md5 module to generate
> passwords for the shadow password file?
> 
> I would like to be able to be able to os.system('useradd -p
> md5/crypt username')
> 
> I can generate the required crypt hashes no problem, but can't
> seem to figure out how to generate and md5 hash that looks
> anything like what I see in /etc/shadow.
> 
> -- 
> Stand Fast,
>     tjg.
> 
> Timothy Grant                         tjg at exceptionalminds.com
> Red Hat Certified Engineer            www.exceptionalminds.com
> Avalon Technology Group, Inc.         <><       (503) 246-3630
> >>>>>>>>>>>>>Linux, because rebooting is *NOT* normal<<<<<<<<<
> >>>>This machine was last rebooted:  59 days  1:59 hours ago<<
> 


==================================
Posted via http://nodevice.com
Linux Programmer's Site


From emile at fenx.com  Tue Mar 27 08:56:26 2001
From: emile at fenx.com (Emile van Sebille)
Date: Tue, 27 Mar 2001 05:56:26 -0800
Subject: Changing every other instance of <B> in a file
References: <8e41ctkk4l38jmj34070j15iotgv4is2cg@4ax.com>
Message-ID: <99q6d7$2brvo$1@ID-11957.news.dfncis.de>

Here's one way (although I'm interested to see how re might do this):

txt = """Blablabla <B> talktalk <B> blabla blabla balbalblabla
Blablabla <B> talktalk <B> blabla blabla balbalblabla
Blablabla <B> talktalk <B> blabla blabla balbalblabla
Blablabla <B> talktalk <B> blabla blabla balbalblabla
Blablabla <B> talktalk <B> blabla blabla balbalblabla
Blablabla <B> talktalk <B> blabla blabla balbalblabla
Blablabla <B> talktalk <B> blabla blabla balbalblabla
Blablabla <B> talktalk <B> blabla blabla balbalblabla"""

parts = txt.split("<B>")

joiners = ("<B>","</B>")
toggle = 0

result = []

for part in parts:
 result.append(part)
 result.append(joiners[toggle])
 toggle = 1 - toggle

newText = "".join(result[:-1])

print newText

--

Emile van Sebille
emile at fenx.com

---------
"Lars Kl?boe" <larskl at klassekampen.no> wrote in message
news:8e41ctkk4l38jmj34070j15iotgv4is2cg at 4ax.com...
> I have quite a few (~70000) quark tagged files I'd like to HTMLize, and
> as my tool of choice has been python for every other script, I thought
> I'd stick to what I know. Or thought I knew.
>
> The original file.txt (quark tagged) :
>
> Blablabla <B> talktalk <B> blabla blabla balbalblabla
>
> The resulting file.html (html)
>
> Blablabla <B> talktalk </B> blabla blabla balbalblabla
>
> As you can tell, every other instance of <B> is to be changed into </B>.
>
> How on earth can I do this as easy as I did in sed ? I'm stuck
> mindwise, locked on a useless solution I'm too embarrased
> to show you ;)
>
> All help is appreciated !
>
> --
> Lars Kl?boe
>




From kens at sightreader.com  Mon Mar 26 12:04:19 2001
From: kens at sightreader.com (Ken Seehof)
Date: Mon, 26 Mar 2001 09:04:19 -0800
Subject: Good GUI for Python
References: <002a01c0b33f$9595ade0$340210ac@winson>
Message-ID: <005b01c0b616$cda12640$a325fea9@his>

wxPython is the best.

I use PythonWin as my IDE and wxPython as my GUI platform.  I find wxPython much easier than PythonWin mostly because MFC is hard to use (even though I've been using MFC in C++ and python (win32ui) for about 10 years).  Also, wxPython is portable, unlike win32ui.   I agree that win32ui is a masterpiece, but it is dependent on MFC, which is not a masterpiece :-(.

I am going to switch to Komodo next (which absorbs Marks expertise but without the MFC limitations).  Komodo looks really nice, but I haven't gotten it to work on my system yet (probably because I have both Python 1.5.2 and 2.0 installed).  I'm looking forward to the switch.  I will continue to use wxPython as my GUI platform.
  Hi, All
   
  I surveyed 3 GUI framework for Python including wxPython, PythonWin's included win32ui and Tkinter. My development environment is Mark Hammond's PythonWin downloaded from ActiveStates.
   
  These're my issues:
   
  First, ActiveStates' Python tool cannot support Tkinter and I confirmed with their stuff. So Tkinter maybe good, but not compatible with PythonWin.
   
  Second, Mark's masterpiece win32ui and total MFC solution is excellent, but I'd wanna bind on Windows platform.
   
  Third, wxPython sounds great but not so easy as win32ui.
   
  Is there any suggestions ? Change my develop environment or is there another good gui for me ??

  --
  Best Regards,
   
  Winson Chang
   
  3F, 2, 150 Lane, Sec. 5, Hsin I Rd., Taipei, Taiwan, R.O.C.
   
  Tel: +886-2-27225333 ext 376
  Fax: +886-2-27222330
   
  Email: winson at mdsserv.mds.com.tw
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20010326/0ec836c9/attachment.html>

From chrisw at nipltd.com  Wed Mar 28 18:14:29 2001
From: chrisw at nipltd.com (Chris Withers)
Date: Thu, 29 Mar 2001 00:14:29 +0100
Subject: Setting file attributes on file creation from ZipFiles (solved)
Message-ID: <3AC27055.B4AC2568@nipltd.com>

I posted this question a while ago, and now that I've found the answer I thought
I'd post it back ;-)

Anyway, the problem was setting the modified dates on files I was creating by
unzipping them from a ZipFile object.

You can do this using a function as follows:

from os import utime
from time import time,mktime

def unzip_with_date(zipefile,filename):
    f = open(filename,'wb')
    f.write(zipefile.read(filename))
    f.close()
    utime(filename,(time(),round(mktime(zipefile.getinfo(filename).date_time +
(0,0,-1)))))

Maybe that could be made into a method and included in ZipFile?

cheers,

Chris

PS: For the curious, I found out how to do this by looking at shutil.py; I love
a language where the source for the libraries is included :-)




From bryan at eevolved.com  Sat Mar  3 10:35:16 2001
From: bryan at eevolved.com (Bryan Mongeau)
Date: Sat, 03 Mar 2001 15:35:16 GMT
Subject: Adding a method to an instance: solution
References: <G2%n6.360714$JT5.11606234@news20.bellglobal.com>
Message-ID: <Ua8o6.302710$Pm2.4401349@news20.bellglobal.com>

In case anyone is wondering:

 import new

? def newMethod(self):
? ? ?# whatever

? f.newMethod = new.instancemethod(newMethod, f, f.__class__)


Thanks to Paul Foley on this one and thank you all for the advice.

Regards,
-- 
<=====================================>
Bryan Mongeau
Lead Developer, Director
eEvolved Real-Time Technologies Inc.
Website:    http://www.eevolved.com
Public key: http://eevolved.com/bcm.pk
<=====================================>

"To say that the mind is "just molecular machines" is like saying that the 
Mona Lisa is "just dabs of paint". Such statements confuse the parts with 
the whole, and confuse matter with the pattern it embodies. We are no less 
human for being made of molecules." -- K. Eric Drexler



From hamish_lawson at yahoo.co.uk  Tue Mar 27 11:49:04 2001
From: hamish_lawson at yahoo.co.uk (=?iso-8859-1?q?Hamish=20Lawson?=)
Date: Tue, 27 Mar 2001 17:49:04 +0100 (BST)
Subject: Why not Tcl/Tk?
Message-ID: <20010327164904.14603.qmail@web208.mail.yahoo.com>

After using Tcl on a project for a year, I found its biggest flaw to be
its relative lack of support for high-level data structures. While Tcl
does have an associative-array structure, it is hard work to make
nested structures using it.

In addition I missed decent OO support. There is an add-on OO package
for Tcl; but because it's not a standard part of Tcl, not many other
Tcl libraries make use of it. Compare this with Python, where f.read()
could be reading from any kind of object that supports the read
protocol - a disk file, a pipe to a process, an in-memory file, a
compressed file, a URL connection, or a user-defined object.

Hamish Lawson


____________________________________________________________
Do You Yahoo!?
Get your free @yahoo.co.uk address at http://mail.yahoo.co.uk
or your free @yahoo.ie address at http://mail.yahoo.ie



From fredrik at pythonware.com  Tue Mar 20 16:17:32 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Tue, 20 Mar 2001 21:17:32 GMT
Subject: File Descriptor Modes
References: <mailman.985118007.7016.python-list@python.org>
Message-ID: <MNPt6.2555$4N4.335487@newsc.telia.net>

O'Rourke Clodagh wrote:
> Does anyone where I would find a summary of the different
> modes it is possible to open a file descriptor in?

how about the library reference, under builtin functions?

http://www.python.org/doc/current/lib/built-in-funcs.html
=> open

summary: "r" opens for reading, "w" opens for writing
(zapping an existing file), and "a" opens for appending.
add a "+" to open for reading and writing (use "r+" to
update an existing file), and add a "b" to open in binary
mode.

Cheers /F




From emile at fenx.com  Fri Mar 16 23:26:26 2001
From: emile at fenx.com (Emile van Sebille)
Date: Fri, 16 Mar 2001 20:26:26 -0800
Subject: ZPT, a next-generation template technology
References: <98ufbu$kkh$1@uranium.btinternet.com> <slrn9b5j52.i70.mgjv@martien.heliotrope.home> <98ujtv$dcv$1@panix3.panix.com>
Message-ID: <98up78$3moic$1@ID-11957.news.dfncis.de>

"Aahz Maruch" <aahz at panix.com> wrote in message
news:98ujtv$dcv$1 at panix3.panix.com...
> In article <slrn9b5j52.i70.mgjv at martien.heliotrope.home>,
> Martien Verbruggen <mgjv at tradingpost.com.au> wrote:
> >On Sat, 17 Mar 2001 01:43:50 -0000,
> > Hamish Lawson <hamish_lawson at btinternet.com> wrote:
> >>
> >> * Zope is a leading open-source web application server and content-
> >> management framework, written in Python by Digital Creations.
> >
> >So why did you post it to cll.perl.misc and cl.java? At least alt.php is
> >slightly relevant to this, although you might piss them off as well. I
> >don't know what their policy on offtopic posts is.
> >
> >However, cl.perl.misc gets enough offtopic crap as it is. Don't post
> >future announcements here.
>
> Ah, the friendliness of Perl programmers that we've all come to know and
> love.
> --
>                       --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)
>
Especially now when Zope is supporting Perl...

Emile van Sebille
emile at fenx.com




From cce at clarkevans.com  Tue Mar 20 21:17:18 2001
From: cce at clarkevans.com (Clark C. Evans)
Date: Tue, 20 Mar 2001 21:17:18 -0500 (EST)
Subject: Yet Another PEP:  Query Protocol Interface or __query__
In-Reply-To: <99917n$2jf$1@panix6.panix.com>
References: <Pine.LNX.4.21.0103201837570.21343-100000@clarkevans.com> <99917n$2jf$1@panix6.panix.com>
Message-ID: <Pine.LNX.4.21.0103202107010.14634-100000@clarkevans.com>

On 20 Mar 2001, Aahz Maruch wrote:
> Clark C. Evans <cce at clarkevans.com> wrote:
> >
> >    A new built-in method, __query__ is proposed.  This method has
> >    a single argument, an interface protocol identifier, and either
> >    returns an object supporting the given protocol, or throws an
> >    unknown exception.
> >
> >    An interface protocol identifier is a lower-case string having
> >    a reverse DNS host name string:   TLD '.' DOMAIN ['.' NAME ]*
> >    Where TLD is a top level domain such as 'com' or 'uk.co'.  DOMAIN
> >    is a registered name in the given TLD.  And one or more optional
> >    NAME separated by a period.  NAME is a sequence of one or more
> >    alphabetic characters including the dash '-' character.  See the
> >    relevant ITEF specifications for specific details.
> 
> It's not at all clear to me why this domain thing is being suggested.

A unique identifier is needed.  One could use GUIDS, but I 
think that this would be less readable and maintainable.

> Is there supposed to be some kind of lookup mechanism 
> associated with it?

Yes.  I was planning that 

  query(obj,"org.python.example.eggs")

would behind the scenes call __query__ on the 
object obj.  This indirection allows for 
additional "hooks" to be added to the system 
via another PEP that could perform a wrapping
function.  See Alex's recent post.

> Most importantly, how does one find out what interfaces 
> an object *does* support?

I don't have this need -- so I didn't include it.  I suppose
we could ammend the proposal to have this functionality.
Perhaps obj.__query__() could return a sequence (or better 
yet a new iterator object).

What do you think?

> >Example Usage
> >
> >    >>> class EggsOnly:
> >            def eggs(self,str): print "eggs!" + str
> >            def __query__(self,protocol):
> >                if protocol == "org.python.example.eggs": return self
> >                raise "unknown: " + protocol
> >
> >    >>> class EggsAndHam:
> >            def ham(self,str): print "ham!"+str
> >            def __query__(self,protocol):
> >                if protocol == "org.python.example.ham": return self
> >                if protocol == "org.python.example.eggs":
> >                    return EggsOnly()
> >                raise "unknown" + protocol
> 
> This seems to be implying that it is appropriate to return a brand-new
> object with no association with the original object.  Is that correct?

Yes.  The __query__ mechansim is vague in this respect,
no use to have any artificial limitations.

> 
> Does it make sense to have something like:
> 
> class HamAndEggs:
>     def __query__(self, protocol):
>         if protocol.find("org.python.example.ham") == 0:
>             return self
> h = HamAndEggs()
> e = h.__query__("org.python.example.ham.eggs")


Let me see, a string goes in, and an object 
comes out. Sure.  Looks good to me!

Clark




From sean at digitome.com  Fri Mar  9 16:10:56 2001
From: sean at digitome.com (Sean Mc Grath)
Date: Fri, 09 Mar 2001 21:10:56 GMT
Subject: XML Module problems
References: <j4y9uhnb1m.fsf@informatik.hu-berlin.de> <memo.20010309203450.37329B@digitig.compulink.co.uk>
Message-ID: <3aa93f65.699574094@news.iol.ie>

(Tim Rowe)
>
>So it does! I only noticed that Pyxie (the DOS version on Sean McGrath's 
>book, not the rpm) broke because:
>
>a. It couldn't find pyexpat (though evidently xml/parsers/expat.py 
>manages) and
>
>b. When I changed pyxie to point to xml.parsers.expat, it found that the 
>attribute list was a dictionary, not a list.
>
>The version on the Pyxie site worked, though, and it's rather too 
>important for me at the moment for me to re-install the broken version to 
>investigate further!

Yes. The original pyexpat used a list for attributes on the start
element handler. A later version of pyexpat changed to a dictionary.
The code on http://www.pyxie.org was changed to worth with both
flavours of pyexpat.

Sean



From sdementen at hotmail.com  Fri Mar 30 04:12:14 2001
From: sdementen at hotmail.com (Sébastien de Menten)
Date: Fri, 30 Mar 2001 11:12:14 +0200
Subject: NumPy loop efficiency
Message-ID: <3ac44e27$0$16692$73bec57d@news.be.uu.net>

Hi,

I was wondering if writing the following statement with NumPy arrays z,a,b
and c:
z = a + b + c
translates into one loop like
for (i in indices)
    z[i] = a[i]+b[i]+c[i]

or two loops
for (i in indices)
    t[i]=a[i]+b[i]
    z[i]=t[i]+c[i]

And if
a[:]=cos(a)+sin(a)
creates temporary variables or assigns directly the result (like
a[i]=cos(a[i])+sin(a[i])

I guess I have too many optimistic expectations...

TX.
Seb




From tim.one at home.com  Sat Mar 24 17:46:55 2001
From: tim.one at home.com (Tim Peters)
Date: Sat, 24 Mar 2001 17:46:55 -0500
Subject: New PEP: The directive statement
In-Reply-To: <j43dc3o31m.fsf@informatik.hu-berlin.de>
Message-ID: <LNBBLJKPBEHFEDALKOLCEECBJIAA.tim.one@home.com>

[Martin von Loewis]
> ...
> My point is that even though you can nuke all obsolete future
> statements, it might not be desirable to do so - since the module then
> become silently incorrect in an older release, instead of outright
> failing.

Using the tool isn't required.  It's a tool I would write because it's a tool
*I* would use (I don't care about clinging to old releases in my own code,
and indeed am quite keen to get rid of then-useless cruft ASAP).  YMMV.

> ...
> I was asking you to predict the future: Do you think that people
> will prefer to write
>
>   if __future__.nested_scopes.getOptionalRelease() >= sys.version_info:
>
> instead of
>
>   if sys.version_info < (2,1):

As explained before, hardcoded version numbers can't be made to work *if*
you're interested in writing code that depends on a change's MandatoryRelease
(which is just a best guess under any release in which sys.version_info is
less than that).

> ...
> I have really no problem with the format of the _Feature members; I
> very much like using tuples for versioning, since it is quite elegant.
> My question is whether you think that people will use those computed
> version numbers instead of hard-coded ones, when comparing to
> sys.version_info.

I *expect* that-- just like now --they'll use hard-coded ones, until they
figure out that they can't get that to work as intended.  Hard-coded version
numbers are always a bad idea anyway -- even if "they worked", the *reader*
of

   if sys.version_info < (2,1):

hasn't a clue in hell what the true intent of that code is.  If the true
intent is to key off the presence or absence of nested_scopes semantics,
writing a test that *names* nested_scopes is clearly much better.

I've never used a hard-coded version number in my entire Python life.  Up
until now, I've been happy to do things like:

import tokenize
if hasattr(tokenize, "NL"):
    # all right!  I can make my parsing much faster by using NL.
    ...
else:
    # bummer!  It's a crufty old version of tokenize.py w/o NL.

But, in the presence of *incompatible* language changes, it's not at all
clear that a "just try it and see whether it works" runtime test is always
possible.  For example, if the incompatible change is the introduction of a
new keyword (like, say, "directive" <wink>), then code attempting to use the
new keyword will likely blow up with a SyntaxError under earlier releases,
and SyntaxError can't be caught at compile-time (at least not without a layer
of obfuscating indirection).  So being able to name the changed feature
explicitly wins big on that count too.




From loewis at informatik.hu-berlin.de  Fri Mar  2 09:00:57 2001
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 02 Mar 2001 15:00:57 +0100
Subject: A counter-proposal to __future__ in PEP 236
References: <j47l29lpo8.fsf@informatik.hu-berlin.de> <mrwn6.16173$Qb7.2637930@newsb.telia.net>
Message-ID: <j4y9uojmna.fsf@informatik.hu-berlin.de>

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

> > Even though this looks like an import statement, it is not: instead,
> > it changes the meaning of variable binding in the presence of nested
> > functions.
> 
> Really?  I'd say it imports an alternate compiler, and activates it
> for all modules that happens to contain a from __future__ state-
> ment.

Exactly. That makes it not to be an import statement; an import
statement essentially calles __import__, and you cannot arrange it to
import compilers.

> > directive nested_scopes
> 
> Note that we're not talking optional standard features here, we're
> talking untested experimental stuff, on a scale too large to handle
> in an ordinary prerelease cycle.

I understand that. Then let's call it 

experimental nested_scopes

or

directive experimental nested_scopes

> __future__ stuff may go away or change, if they turn out to be
> a bad idea.  If you're easily confused, you shouldn't use them at
> all.

The specific thing being "imported" (e.g. nested scopes) may go away
or change. I read PEP 236 to mean that __future__ is here to stay.

> But with your proposal, you make them look like optional features.

Is the the keyword that makes them look like that? My proposal is open
to picking a different keyword.

> That isn't just a bad idea, it makes Python look bad too.

Why is that?

> Smells like hypergeneralization to me (which is a worse sin than
> premature optimization).  Source code encodings and experimental
> compiler features are two entirely different things.

They are similar as source code encoding might also be an experimental
feature first. They are also similar 

> > In this implementation, directive is only considered as a keyword if
> > it appears at the beginning of the module
> 
> And you think from __future__ is a hack?  Gimme a break.

It's a transitional feature to introduce the keyword directive (or
whatever it is called). At some point, directive will be a proper
keyword, at which time it won't be a hack anymore. At that time,
future statements will still be spelled "from __future__ import
case_insensitive", and it will still be a hack.

There are two unrelated issues here: smooth introduction of nested
scopes, and smooth introduction of per-module semantical
changes. Apparently, there is no clear way of doing that, so it should
be done in a way that causes least confusion and least hackery.

Regards,
Martin


From whisper at oz.net  Thu Mar 29 15:33:51 2001
From: whisper at oz.net (Dave LeBlanc)
Date: 29 Mar 2001 20:33:51 GMT
Subject: CGI XML-RPC (xmlrpccgi.py) demo?
References: <mailman.985874918.20186.python-list@python.org>
Message-ID: <9a067f$png$1@216.39.170.247>

Isn't Bobo the older name for Zope? Perhaps something named bobo is in
an older zope distribution?

Regards,

Dave LeBlanc

On Thu, 29 Mar 2001 16:07:24 +0200 (MET DST), Eugene Leitl
<Eugene.Leitl at lrz.uni-muenchen.de> wrote:

>On Thu, 29 Mar 2001, Fredrik Lundh wrote:
>
>> sorry, it looks like I misread your original question -- I thought
>> you were looking for a way to pass arguments (i.e. do RPC calls)
>> to a server process via HTTP and CGI, not that you had to use
>> a certain syntax.
>
>Oh, I can use XML-RPC just fine, as my (apparently quite
>buzzword-compliant) company wants to give their products in future a SOAPy
>flavour (they kept mishearing me saying SOAP when I was saying Zope ;). I
>just thought that CGI XML-RPC would also handle parsing POST/GET
>parameters -- well, it would have been too easy then. Sigh.
>
>I guess I'll have to parse the parameters by hand, and using XML-RPC
>at a later stage of the processing pipeline. 
>
>> under XML-RPC, you send XML-RPC requests, not FORMs.  see
>> http://www.xmlrpc.com for more info.
>
>Thanks, have been there (though got no T-shirt yet).
>
>You as others been very helpful. Now if somebody could tell me where I can
>get Bobo for Medusa...
>
>



From greg at cosc.canterbury.ac.nz  Wed Mar  7 22:45:38 2001
From: greg at cosc.canterbury.ac.nz (Greg Ewing)
Date: Thu, 08 Mar 2001 16:45:38 +1300
Subject: I come to praise .join, not to bury it...
References: <mailman.983485923.1136.python-list@python.org> <97p7iq$kk2$1@nntp6.u.washington.edu> <97rka002k9@news1.newsguy.com> <3AA2F4B3.A7496E76@cosc.canterbury.ac.nz> <97vmoh061p@news2.newsguy.com> <3AA46134.D385208B@cosc.canterbury.ac.nz> <98292h0s0g@news1.newsguy.com> <3AA58BD0.3324C709@cosc.canterbury.ac.nz> <9850ge013ns@news1.newsguy.com>
Message-ID: <3AA70062.C3EC318C@cosc.canterbury.ac.nz>

Alex Martelli wrote:
> 
> The joiner-dispatch DESIGN *affords* efficiency gains;

I still don't see how. You have to dispatch on all the
arguments eventually somehow, whether by typeswitch or
something else.

> And *WHAT* 'standard interface' would apply on the joiner
> object, pray?

The same one that applies to all the strings being
joined, i.e. "make yourself into a string". Yes, yes,
I KNOW you want your joiner object to be able to be
not just a simple string but some esoteric entity
capable of doing anything it wants to any sequence
of anything at all. But I don't feel any great desire
for that whatsoever. I'm quite happy with the current
highly specialised special-case-the-snot-out-of-joining-
these-strings meaning of 'join'. And the new design is
no better at achieving *that*, as far as I can see.

It's just occurred to me that there may be some
natural-language ambiguity here in the use of the
term "joiner". To me, it's not "the joiner" in the
active sense of "the object which performs the 
computation of joining", but the passive sense of
"the object which ends up being in between the
other strings".

In other words, it's the difference between "pipe
joiner" being a person who carries out plumbing work,
and a short piece of pipe with threads in each end.

> 'Counting' technical advantages is as idiotic as 'counting'
> coins as a measure of wealth

Sure it is, but the point is that blindly ranking all 
technical considerations above all aesthetic ones is
just as silly. Otherwise we might as well all switch
to Perl, which I hear is faster at reading files. :-)

>     'comma-separate these values'

You do have a point there, sort of, but it still sounds
more terse and cryptic than "separate these values with
commas", and crypticness is generally not valued much
in the Python culture. Also, I don't normally think of
the '.' operator as corresponding to a hyphen used that
way in English. The Python translation of that to my
mind would be

   comma_separate(these_values)

and not

   comma.separate(these_values)

which to me means either "comma, separate these values" 
or "separate this comma with these values". The former
is slightly odd because I'm not used to giving commands
to punctuation symbols, and the latter is just plain...
what can I say... BACKWARDS! (Yes, there's that word
again.)

> (Please note that the CSV acronym IS an English one

Also interesting to note that it's *not* CJV. Even
in English, "comma-join these values" sounds strange,
to my ears, anyway. We regularly talk about commas as
being separators, but hardly ever as joiners. 

If I were putting "-" between things, I might 
say it was a joiner rather than a separator. But even
then I wouldn't say "hyphen-join these values", I'd
say "hyphenate these values".

> The string object and the string module are both in
> the Python core, so moving functionality between them
> can in no way be considered 'bloating'.

Not bloating the core, bloating the string object.

The string module may be statically linked in by
default, but it could be dynamically linked, or omitted
altogether, just by changing the config file. Chances
are I'll never want to do that, but it's nice to know
that I *could*. I couldn't cut out all the non-essential
parts of the string object so easily.

I like the idea of objects that concentrate on doing one
job and doing it well. In the case of strings, that
job is being a repository of characters, not being something
that can do everything you can think of that you might
possibly want to do to some characters.

Yes, that's an aesthetic consideration, but I don't
think you can separate technical and aesthetic considerations
completely. How much weight you give to a given technical
advantage (such as modularity of the core features) is
an aesthetic judgement.

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


From loewis at informatik.hu-berlin.de  Sat Mar 24 08:50:37 2001
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 24 Mar 2001 14:50:37 +0100
Subject: PEP 245: Python interfaces
References: <Pine.LNX.4.32.0103222253510.26302-100000@localhost.localdomain>
Message-ID: <j4hf0jz36a.fsf@informatik.hu-berlin.de>

Michel Pelletier <michel at digicool.com> writes:

> I'm the author of PEP 245, a framework and syntax for interface objects in
> Python.
> 
> http://python.sourceforge.net/peps/pep-0245.html
> 
> It has just recently been made a PEP, and I'd like to solicite some more
> input from the community (you).

Even though this is quite a long document, it is incomplete. This is
probably not surprising since it is the first draft, but to really
evaluate, I think a number of details should be added. In order of my
preference:

- proposal for a concrete syntax. The PEP should provide a detailed
  specification of the grammar changes, preferably in (A)EBNF or the
  Grammar/Grammar format. Currently, only a "Syntax Overview" is
  given; that doesn't answer all possible questions. In particular,
  the role of the "implements" reserved(?) word is never specified.

- detailed description of the contents of the module Interface.  For
  example, Interface.Exception.BrokenImplementation is used but not
  defined. For the moment, I assume that the proposal has the same
  definitions as the reference implementation.

- a reference implementation that works as a patch to Python. This is
  probably not needed at the moment, but certainly prerequisite for
  people to completely evaluate the PEP.

I like many aspects of the draft, but a few points should be
corrected, IMO:

- naming of the Interface package. Few standard modules are in
  uppercase, and I think this shouldn't be uppercase, either. Instead,
  I'd prefer a module "interfaces", in analogy to the types module.

- packaging. I think the hierarchy is too deep, everything should be
  contained in the interfaces module (there is not an awful lot of
  stuff in it, is there?).

- exceptions. I think the standard exception hierarchy should be used
  better, e.g. inheriting from NotImplemented (or directly using that
  exception) where appropriate. Also, BrokenImplementation seems a bad
  choice of wording; Smalltalk uses subclassResponsibility to indicate
  that an abstract method is not implemented.

- nesting of interface assertions. Why is it necessary to support

   implements FooInterface, (BarInterface, (BobInterface, MyClass.__implements__))

  Wouldn't it be sufficient (and more consistent with classes) to write

  implements (FooInterface, BarInterface, BobInterface)+MyClass.__implements__

- Access to fields. I think it is confusing that you won't write
  Iface.feature, e.g. in

  interface Iface:
    def feature(self):
      "does something"

  Instead of writing Iface.feature, it appears that you write
  Iface.getDescriptionFor("feature"). It would be better if attribute
  access could be used.

Regards,
Martin


From sholden at holdenweb.com  Fri Mar  9 09:36:18 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Fri, 9 Mar 2001 09:36:18 -0500
Subject: Can anyone recomend a good intoduction to C...
References: <mailman.983894288.19965.python-list@python.org> <3AA5EF59.507F1D91@olsen.ch> <983958878.976344@newsmaster-04.atnet.at> <989cvj$sba$1@panix3.panix.com> <984132875.222868@newsmaster-04.atnet.at>
Message-ID: <lX5q6.45269$1D5.2047774@e420r-atl1.usenetserver.com>

"Werner Schiendl" <ws-news at gmx.at> wrote in message
news:984132875.222868 at newsmaster-04.atnet.at...
> What's TMTOWTDI, please?
>
There's another programming language whose name begins with P, whose primary
philosophy is demonstrated in the well-known phrase"

    "There's More Than One Way to Do It"

This means that it's more difficult to establish recognisable programming
idioms, because for every task there are between three and seventeen ways to
accomplish it, and the choice is usually made on efficiency grounds rather
than those of program readability.

perl's-not-a-bad-language-if-i-don't-have-to-use-it-ly y'rs  - steve





From bedge at troikanetworks.com  Tue Mar  6 13:52:01 2001
From: bedge at troikanetworks.com (Bruce Edge)
Date: Tue, 06 Mar 2001 10:52:01 -0800
Subject: Clearing history from readline module
Message-ID: <3AA531D1.B9AC8CFE@troikanetworks.com>

I have two contexts from my interp, regular python CLI and my cmd based
interp.
I want to maintain separate command line histories for each. I can do a
readline.read_history_file, but that appends the current history. 
How can I reset the readline history data?
I tried reload(readline) but that has no effect.

TIA, Bruce.


From mwh21 at cam.ac.uk  Thu Mar 29 19:16:41 2001
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 30 Mar 2001 01:16:41 +0100
Subject: diff lists
References: <mailman.985908578.30079.python-list@python.org>
Message-ID: <m3wv98az6e.fsf@atrus.jesus.cam.ac.uk>

"Tim Peters" <tim.one at home.com> writes:

> [Michael Hudson]
> > Tangentially, does anyone know of any good algorithms for "edit
> > distance" between two sequences?  E.g. if I have
> >
> > "abcdef"
> >
> > and want to get to
> >
> > "abQUACKcde"
> >
> > I want to get the answer back "insert 'QUACK' at position 3 and delete
> > a character at position 11".
> 
> See std library module difflib.py (new in 2.1).  This packages the
> SequenceMatcher class from the heart of the Tools/scripts/ndiff.py tool for
> easy reuse.

Ah yes; I knew about that and thought it was probably too
heavy-weight, but admittedly didn't actually try it...

[snip]
> >>> for tag, i1, i2, j1, j2 in m.get_opcodes():

... or look close enough to find the get_opcodes methods.  I'll give
it a whirl.

Thanks!
M.

-- 
  ARTHUR:  But which is probably incapable of drinking the coffee.
                    -- The Hitch-Hikers Guide to the Galaxy, Episode 6


From wtanksle at dolphin.openprojects.net  Tue Mar 20 16:53:24 2001
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Tue, 20 Mar 2001 21:53:24 GMT
Subject: Wrong reasons to enhancing the language
References: <mailman.985055665.29647.python-list@python.org>
Message-ID: <slrn9bfmgp.9e.wtanksle@dolphin.openprojects.net>

On Mon, 19 Mar 2001 21:33:07 -0500, Tim Peters wrote:
>and-we-really-*need*-a-fifth-group-that-doesn't-agree-with-any-of-the-
>    others<wink>-ly y'rs  - tim

I'm offended by accurate, fast, or recognisable math.  Can we all switch
to INTERCAL and leave this Python stuff behind like a bad dream?

-- 
-William "Billy" Tanksley


From ws-news at gmx.at  Tue Mar 13 04:25:39 2001
From: ws-news at gmx.at (Werner Schiendl)
Date: Tue, 13 Mar 2001 10:25:39 +0100
Subject: This math scares me
References: <mailman.984458537.14769.python-list@python.org>
Message-ID: <984475826.262977@newsmaster-04.atnet.at>

I wonder why 17 digits are used to represent the numbers...
Think I can remember the precision of double was 15 to 16 decimal digits
(depends on the value)

Did I miss something?

regards
werner

Tim Peters <tim.one at home.com> wrote in message
news:mailman.984458537.14769.python-list at python.org...
> [Andrew Koenig]
> > I'm curious -- Does Python now follow either the IEEE or the more
> > restrictive Scheme rules for floating-point conversion?
> > ...
>
> All aspects of Python fp are inherited from whatever the platform C
compiler
> and libraries happen to do.  repr() conversion to string uses the platform
> sprintf's %.17g format; str() conversion %.12g; both are fiddled just a
> little to ensure that an exponent or a decimal point (or both) appears.
> string->float is a x-platform adventure, since some platforms can't read
back
> all the C doubles they print (thinking of infinities and NaNs here);
Python
> inherits that too, whenever it happens.
>
> it's-all-x3j11's-fault<wink>-ly y'rs  - tim
>
>




From cribeiro at mail.inet.com.br  Wed Mar 28 08:42:51 2001
From: cribeiro at mail.inet.com.br (Carlos Alberto Reis Ribeiro)
Date: Wed, 28 Mar 2001 10:42:51 -0300
Subject: types.AnyType - describing interfaces with sequences of
  types
In-Reply-To: <3ac1c5b3_1@goliath.newsfeeds.com>
References: <mailman.985756894.12348.python-list@python.org>
Message-ID: <5.0.2.1.0.20010328095224.021b4ec0@mail.inet.com.br>

At 02:56 28/03/01 -0800, you wrote:
>You get the idea. AnyType above will be eternally happy to be equivelent to
>anything you compare it to. Note that you can't do 'IntType is AnyType',
>that tells you actual object identity. Now, if you wanted it to only show up
>as 'equivelent' to TypeTypes, and not say, strings, (e.g. "string" ==
>AnyType) you could do a check to make sure what its being compared to is a
>TypeType.....

My first instinct it was to do something like this. However, it did not 
work, but because of a dumb mistake of mine :-) I tried again now, and this 
is the resulting code:

class CAnyType:
   def __cmp__(self, other):
         return (type(other) != TypeType)
   def __rcmp__(self, other):
         return (type(other) != TypeType)

AnyType = CAnyType()

That was my fault - I was trying to compare the class to IntType, but I 
need an instance for the code to work.

Anyway, I felt that should be a better way to make it, maybe be inheriting 
from some "UserType" object. However no such object exists. I don't know if 
"UserType" does not exist because it not make sense, or if it does not 
exist simply because no one else required it before. The way it is, it's 
inconsistent, because:

 >>> type(AnyType)        # should return "TypeType"
<type 'instance'>



Carlos Ribeiro





From agavrilov at home.com  Mon Mar 12 21:19:42 2001
From: agavrilov at home.com (Alexander Gavrilov)
Date: Tue, 13 Mar 2001 02:19:42 GMT
Subject: Extending Python: rewriting a single method in C
References: <g0wv9vxms0.fsf@scumbag.ecs.soton.ac.uk> <slrn9apn8f.382.jepler@potty.housenet>
Message-ID: <2tfr6.26915$zV3.1982567@news1.frmt1.sfba.home.com>

"Jeff Epler" <jepler at inetnebr.com> wrote in message
news:slrn9apn8f.382.jepler at potty.housenet...
> On 12 Mar 2001 11:13:51 +0000, Jacek Generowicz
>  <jmg at ecs.soton.ac.uk> wrote:
> >I have an object-oriented numerical code written in Python. It has
> >come to the stage where the speed of the code is seriously impeding
> >progress. Profiling reveals (surprise, surprise) that most of the time
> >is spent in only 2 methods.
> >
> >Is there any way of re-writing just these methods in C ?
>
> It's easy to write a function in C.  Other posters have mentioned
references
> to documentation about doing this.
>
> However, you cannot make a builtin function be a method of a
class/instance.
> Instead, you could write:
>
> from cmodule import fast_a, fast_b
>
> class fastKlass(Klass):
> def a(self, int1, int2):
> fast_a(self, int1, int2)
> def b(self, int1, int2):
> fast_b(self, int1, int2)
>

Actually, you can write the following:

class fastKlass(Klass):
    a = fast_a
    b = fast_b

Whenever the program calls 'a' or 'b' methods, 'fast_a' or 'fast_b' get
called instead. And you'll eliminate of 'double-calling' overhead.
I didn't check it with functions from C extension module, but it works with
plain Python functions. The following example works for me:

def func1(self):
    print self.a1

def func2(self,somearg):
    print somearg+self.a2

class C:
    def __init__(self):
        self.a1 = 'attr1'
        self.a2 = 'attr2'

    meth1 = func1
    meth2 = func2

c = C()
c.meth1()
c.meth2('Some string ')

Alexander




From edcjones at erols.com  Tue Mar 20 14:18:38 2001
From: edcjones at erols.com (Edward C. Jones)
Date: Tue, 20 Mar 2001 14:18:38 -0500
Subject: Small bug in PIL
Message-ID: <3AB7AD0E.566C29F7@erols.com>

In the PIL (Python Imaging Library) source code file "Image.py", line 937, in
function "merge", the error message is incorrect. Here is the code:

    for im in bands[1:]:
        if im.mode != getmodetype(mode) or im.size != bands[0].size:
            raise ValueError, "wrong number of bands"

Thanks,
Ed Jones




From fredrik at pythonware.com  Wed Mar 21 12:23:44 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 21 Mar 2001 17:23:44 GMT
Subject: Regular expressions and non-standard character set
References: <ka8y9u01htt.fsf@cheerful.cs.hut.fi> <0XMt6.2503$4N4.321668@newsc.telia.net>
Message-ID: <As5u6.909$sk3.295436@newsb.telia.net>

I wrote:
> are you sure?

turns out I wasn't sure what I were doing...

"(?L)\b" and "(?L)\B" doesn't work properly in 2.0 and 2.1b1, under
non-C locales.  \W and other locale-dependent operators work as
expected, though.

this will be fixed in 2.1b2 (out soon) and hopefully also in 2.0.1 (out
when it's done).

to work around this problem, use "pre" instead of "re".

Sorry /F




From shaleh at valinux.com  Mon Mar 12 15:31:08 2001
From: shaleh at valinux.com (Sean 'Shaleh' Perry)
Date: Mon, 12 Mar 2001 12:31:08 -0800 (PST)
Subject: continue/else functionality
In-Reply-To: <BD674CDC0A68C642B60038986FBB56DF0138B7@mailsrv.ASPATECH.COM.BR>
Message-ID: <XFMail.20010312123108.shaleh@valinux.com>

> 
> 
>       for line in f.readlines():
> 
>           datalist = string.split(line)
>           if datalist[0][0] == 1:
>              continue
>           elif datalist[0][0] == 0:
>              break
>           
>           ## Do a bunch of string manipulation here...
> 

why do read the first line, check it, if it works, go into the loop.

line = f.readline()
if line[0][0] == 1:
        for line in f.readlines():
                # do things




From josh at pobox.com  Sun Mar 11 07:02:49 2001
From: josh at pobox.com (Joshua Spoerri)
Date: Sun, 11 Mar 2001 07:02:49 -0500
Subject: supporting asyncronous messages
In-Reply-To: <j466hloqof.fsf@informatik.hu-berlin.de>; from loewis@informatik.hu-berlin.de on Wed, Mar 07, 2001 at 10:48:32AM +0100
References: <mailman.983793067.17928.python-list@python.org> <j466hloqof.fsf@informatik.hu-berlin.de>
Message-ID: <20010311070249.A69096@acis.mc.yu.edu>

On Wed, Mar 07, 2001 at 10:48:32AM +0100, Martin von Loewis wrote:
thank you for the pointer. i thought about it a little more, and for
this particular application, i'd like to use tristate logic to automatically
wait for only the relevant events. idea being that if an expression
evaluates to "unknown", it should block, but if it evaluates to true or
false it should continue. it means that you only need to worry about
a single expression to control flow and wait behaviour.

can python overload the logical operators, if not, are there any
plans to enable it to? do you see any gotchas in this approach?

thanks, josh




From emile at fenx.com  Sat Mar 10 08:15:11 2001
From: emile at fenx.com (Emile van Sebille)
Date: Sat, 10 Mar 2001 05:15:11 -0800
Subject: How to efficently build a nested dictionary
References: <3AAA2042.E69A53FC@gmx.de>
Message-ID: <98d9f1$1hopn$1@ID-11957.news.dfncis.de>

Any particular reason not to do:

ky = (dir, year, month, day)
p_ttl = traffic.get(key,0)
traffic[ky] = p_ttl + bytes

--

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


"Carsten Gaebler" <cgaebler at gmx.de> wrote in message
news:3AAA2042.E69A53FC at gmx.de...
> Hi there!
>
> I wrote a script for FTP traffic accounting, i.e. for each
"top level"
> FTP directory it counts the bytes transferred per day. The
traffic is
> stored in a dictionary that has the following structure:
>
> traffic = {dir: {year: {month: {day: 42}}}}
>
> for several dirs, years, months, days, of course.
>
> While parsing the FTP log files I am building up the
dictionary in this
> way:
>
> if not traffic.has_key(dir):
>     traffic[dir] = {year: {month: {day: bytes}}}
> elif not traffic[dir].has_key(year):
>     traffic[dir][year] = {month: {day: bytes}}
> elif not traffic[dir][year].has_key(month):
>     traffic[dir][year][month] = {day: bytes}
> elif not traffic[dir][year][month].has_key(day):
>     traffic[dir][year][month][day] = bytes
> else:
>     traffic[dir][year][month][day] += bytes
>
>
> Now, this looks a bit clumsy to me, especially if I
consider
> you-know-which-language-I-mean which would simply allow to
write
> $traffic{$year}{$month}{$day} += $bytes;
>
> Any ideas how to do this more elegantly?
>
> Regards
> Carsten.




From aahz at panix.com  Thu Mar 15 15:18:02 2001
From: aahz at panix.com (Aahz Maruch)
Date: 15 Mar 2001 12:18:02 -0800
Subject: Dynamic append in a dictionnary object ...
References: <3ab11de3$0$389$446d705c@news.skynet.be>
Message-ID: <98r81q$nts$1@panix3.panix.com>

In article <3ab11de3$0$389$446d705c at news.skynet.be>,
Furax <constant.dupuis at swing.be> wrote:
>
>I'm pretty new to Python.
>I cannot figure out, reading the documentation, how I can dynamicaly add a
>new pair in a dictionnary ?

>>> d = {}
>>> d[1] = 'foo'
>>> d
{1: 'foo'}
>>> d['bar'] = 20
>>> d
{1: 'foo', 'bar': 20}

I suggest you work through the tutorial at
http://www.python.org/doc/current/tut/tut.html
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

"The overexamined life sure is boring."  --Loyal Mini Onion


From jepler at inetnebr.com  Tue Mar 27 21:38:28 2001
From: jepler at inetnebr.com (Jeff Epler)
Date: Wed, 28 Mar 2001 02:38:28 GMT
Subject: close(), exceptions and problems
References: <slrn9bm5u5.klj.kuntsi@myntti.helsinki.fi> <99i3j7$lqr$1@news.cybercity.dk> <uzoe83s26.fsf@ctwd0143.fitlinxx.com>
Message-ID: <slrn9c2jl4.581.jepler@potty.housenet>

On 26 Mar 2001 20:43:13 -0500, David Bolen
 <db3l at fitlinxx.com> wrote:
>Note that close() is often is the point at which buffered data is
>flushed to a file, so there are a variety of ways in which that can
>fail (I/O error, out of disk space, etc...) which will cause the
>close() to fail.

GNU Libc's documentation is fairly explicit about this:
     The normal return value from `close' is 0; a value of -1 is
     returned in case of failure.  The following `errno' error
     conditions are defined for this function:

    `EBADF'
          The FILEDES argument is not a valid file descriptor.

    `EINTR'
          The `close' call was interrupted by a signal.  *Note
          Interrupted Primitives::.  Here is an example of how to
          handle `EINTR' properly:

               TEMP_FAILURE_RETRY (close (desc));

    `ENOSPC'
    `EIO'
    `EDQUOT'
          When the file is accessed by NFS, these errors from `write'
          can sometimes not be detected until `close'.  *Note I/O
          Primitives::, for details on their meaning.

Strangely enough, the section 2 manpage omits this detail and documents
the only errno value of close as EBADF.

Jeff


From pbrian at demon.net  Mon Mar 19 11:29:19 2001
From: pbrian at demon.net (Paul Brian)
Date: Mon, 19 Mar 2001 16:29:19 -0000
Subject: newbie - infinite loop
References: <985012513.18971.0.nnrp-12.c1c3e154@news.demon.co.uk> <slrn9bc8et.95q.scarblac@pino.selwerd.nl>
Message-ID: <985019137.15846.0.nnrp-09.c1c3e154@news.demon.co.uk>

All,

Thank you for such quick responses.

I shall take it as final that a for loop is expandable if the condition is
mutable.  I just thought I might have missed something else.

And thank you all for pointing out the deliberate :-) mistake over matching
and not matching

"Remco Gerlich" <scarblac at pino.selwerd.nl> wrote in message
news:slrn9bc8et.95q.scarblac at pino.selwerd.nl...
> Paul Brian <pbrian at demon.net> wrote in comp.lang.python:
> > I have solved the opriginal problem but am left with a serious question.
> >
> > Before I discovered the keywords in and not in, I tried to count unique
> > instances of names in a file.
> >
> Your code belows does the opposite: if it *is* already in the list, append
> it once more. And since you change the list you're currently looping over
> (which is a mortal sin, etc) you get an infinite loop.
> >
> It's a list. It's mutable. It's not a "for condition", for just tries to
get
> new elements from the list until that gives an error, then the loop is
> finished. If you append to the list during the loop, the new elements are
> also visited. If you add or delete things in the middle of the list,
> anything might happen.
>
> Do not taunt happy fun for loops. Do not change lists you are looping
over.





From quinn at riyal.ugcs.caltech.edu  Thu Mar 15 04:55:43 2001
From: quinn at riyal.ugcs.caltech.edu (Quinn Dunkan)
Date: 15 Mar 2001 09:55:43 GMT
Subject: Process details
References: <mailman.984439220.26863.python-list@python.org>
Message-ID: <slrn9b14cv.fae.quinn@riyal.ugcs.caltech.edu>

On Mon, 12 Mar 2001 15:22:16 -0800, Timothy Grant <tjg at exceptionalminds.com>
wrote:
>Hi all,
>
>I hope everyone is enjoying there first week back after the
>events of P9.
>
>I need to create a list of processes and process information. I
>have created a class that contains much of the information in
>/proc/'pid'/status.
>
>It works quite well, however, it takes about four seconds to
>populate a list of 90 procs. This seems a bit long. Is there a
>faster way to get ahold of the information than reading /proc?

You don't say what OS you're using, but from the sig I'm guessing Linux.  In
Linux, /proc is the official interface.  /bin/ps and friends use it.  Try
'cat /proc/[0-9]*/status >/dev/null'.  If that takes less than four seconds,
then the slowdown is likely in your code, not /proc.  On this linux machine,
it takes a few hundredths of a second for several hundred procs.


From moshez at zadka.site.co.il  Tue Mar 27 17:26:48 2001
From: moshez at zadka.site.co.il (Moshe Zadka)
Date: Wed, 28 Mar 2001 00:26:48 +0200
Subject: Threading legality/morality
In-Reply-To: <99qoje$o6q$1@troll.powertech.no>
References: <99qoje$o6q$1@troll.powertech.no>, <fsDv6.1247$NR.100201@news3.oke.nextra.no> <99nnbv$phs$1@panix6.panix.com> <h0%v6.1842$NR.144463@news3.oke.nextra.no> <99qa3v$6e0$1@panix6.panix.com>
Message-ID: <E14i1vM-0007TH-00@darjeeling>

[Aahz]
> There is a misfeature in urllib such that the standard FancyURLOpener
> class prompts for input if it gets an HTTP 401 (authorization needed).
> You probably pressed ENTER.

[Syver Enstad]
> Think so yes, how can it be fixed? I am pretty well versed in python but a
> pointer could be nice so I didn't use quite so much time on it.

1. No it can't be fixed. (Backwards compat.)
2. It doesn't need to be fixed -- urllib2 works much better

Python 2.1 has urllib2 with documentation. Find a howto in
http://www.lerner.co.il/~moshez/urllib2-howto.tex.
I'll try to find time to integrate it into the Python howto
project. 
-- 
"I'll be ex-DPL soon anyway so I'm        |LUKE: Is Perl better than Python?
looking for someplace else to grab power."|YODA: No...no... no. Quicker,
   -- Wichert Akkerman (on debian-private)|      easier, more seductive.
For public key, finger moshez at debian.org  |http://www.{python,debian,gnu}.org



From m.hadfield at niwa.cri.nz  Sun Mar  4 16:10:55 2001
From: m.hadfield at niwa.cri.nz (Mark Hadfield)
Date: Mon, 5 Mar 2001 10:10:55 +1300
Subject: Performance when working directory is slow network drive
References: <983411179.898268@clam-ext> <mailman.983425921.14373.python-list@python.org> <983480535.813990@clam-ext> <7uIn6.16286$Qb7.2659111@newsb.telia.net>
Message-ID: <983740313.603627@clam-ext>

"Fredrik Lundh" <fredrik at pythonware.com> wrote in message
news:7uIn6.16286$Qb7.2659111 at newsb.telia.net...
> Mark Hadfield wrote:
> > 1. Why having the current directory in the search path affects the
import
> > time for the os module much more than it does for other modules I tested
> > (like string). Is this because os is a package?
>
> try running the interpreter with -vv (very verbose) might
> give you additional clues:
>
>     python -vv -c "import os"
>
> (look for "trying" messages)

Yes. It's clear from the speed at which the output appears on the screen
that the delays occur during the following operations

# trying os.pyd
# trying os.dll
# trying os.py
# trying os.pyc

then later when it tries the same thing with ntpath, stat and UserDict.

Thank you.
---
Mark Hadfield
m.hadfield at niwa.cri.nz  http://katipo.niwa.cri.nz/~hadfield
National Institute for Water and Atmospheric Research




From mix77 at usa.net  Fri Mar  2 03:55:17 2001
From: mix77 at usa.net (Mix)
Date: Fri, 02 Mar 2001 10:55:17 +0200
Subject: CGI & Python (error)
Message-ID: <3A9F5FF5.34E01148@usa.net>

When accessing  the  script I get the error:

File "getdate2.py", line 12, in ?
    day=form["day"].value
  File "/usr/lib/python1.5/cgi.py", line 907, in __getitem__
    raise KeyError, key
KeyError: day

The script works from the command line.

My problems is:  the script I am trying to run is based on another one.
This
other script works. I can't  figure out the what is causing the problem
as
the 2 scripts are identical  in every aspect execpt for 1 variable. Both
scripts
are run in a page which has all the required parameters passed to it.
Help. Please...

I hope I'm understandable!





From see.signature at bottom.of.msg  Fri Mar  2 15:30:35 2001
From: see.signature at bottom.of.msg (Thomas Burns)
Date: 02 Mar 2001 20:30:35 GMT
Subject: Stream interfaces
References: <MPG.1505ab2efd31b1639896b4@news.earthlink.net> <memo.20010302083453.61911A@brangdon.madasafish.com>
Message-ID: <97ovtb$jjk@dispatch.concentric.net>

"Dave Harris" <brangdon at cix.co.uk> wrote in message
> [...]
> I expect we need more delegation. The main thing is to
> separate out the formatting front-end, which needs a rich and probably
> varied API, from the data-sink back-end which can use a much smaller
> and more uniform API. Then we can do buffering etc by means of
> pluggable adaptors, and construct pipelines like:
>     XmlStream -> CompressionAdaptor -> SocketStream

You can get one implementation of this from the VisualWorks Freebies site at
http://smalltalk.iwarp.com.  The free package "Streams & Bytes" contains a
generic StreamWrapper class.  I got sick of creating my own streams, then
realizing that there wasn't a good place to hang them.

The StreamWrapper has been used in some of the other free packages.  For
example, the Zlib (zip-style compression, for those that don't know) package
subclasses StreamWrapper to do compression and decompression on the fly via
streams.  There is also a ProgressReadStream that, when used in conjunction
with the WidgetExtensions, can open a progress bar showing progress reading
through fixed-length streams (such as file streams or internal streams).

> This is roughly the approach of the Java library. As far as I know it
> works reasonably well.

Don't get me started.  :-)

This is what Java provides through the availability of the BufferedWriter
and BufferedOutputStream.  But in Java these (string-writing vs.
byte-writing) are 2 separate hierarchies.  On top of that Writer and
OutputStream are abstract classes, not interfaces.  So you must root from
one of these hierarchies, no matter how badly they fit with your
implementation.  In all, I would say that providing the Buffered* classes is
a good thing, but it's no great shakes and there are other significant
problems magnified by typing issues.  Give me Smalltalk any day.

t
--
Thomas Burns
thomas.a.burns at usa.net
VisualWorks Freebies are available at http://smalltalk.iwarp.com




From lrl at ou.edu  Thu Mar 29 18:00:25 2001
From: lrl at ou.edu (Luke L)
Date: Thu, 29 Mar 2001 17:00:25 -0600
Subject: Python GUI with Swing
References: <001e01c0b812$43f00760$340210ac@winson> <mailman.985889320.22860.python-list@python.org>
Message-ID: <3AC3BE89.3F95C8DC@ou.edu>

> If speed is an issue, don't use Swing.  Swing is quite slow (from
> experience) as it is implemented entirely in Java.  If you only need
> to support a single platform you can use Jython+Swing+JNI+C using JNI
> to connect Java to the speed critical stuff written in C.  Otherwise I
> would recommend looking at other toolkits such as PyGTK and wxPython.

Yeah tell me about it...lightweight components are the wrong way to go....
Awt would be a step faster, but that's still using Java of course



From fredrik at pythonware.com  Thu Mar  1 13:40:36 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Thu, 01 Mar 2001 18:40:36 GMT
Subject: writing a C function to manipulate python lists
References: <jllmqpnclz.fsf@endor.bbb.caltech.edu>
Message-ID: <EIwn6.16175$Qb7.2638512@newsb.telia.net>

Michael Vanier wrote:
> I've been programming in python for many years but have never had to solve
> this particular problem.  I would like to write a short C function that will
> be part of a python-callable module.  The function will receive two python
> lists as arguments.  Currently I do this in python, but my code uses this
> function very heavily and it's the major speed bottleneck in the code.  The
> hope is that the C function will be much faster than iterating through the
> lists in python (which may not be true, but I'd like to see for myself).
>
> If anyone has some code lying around that does something like this, could
> they pass it on to me?

there's plenty of code in the Python source distribution...

searching for PyList and/or PySequence should help.  here's
one example, based on SRE's compile function:

static PyObject *
myfunc(PyObject* self_, PyObject* args)
{
    int i, n;

    /* accept a single list object */
    PyObject* code;
    if (!PyArg_ParseTuple(args, "O!", &PyList_Type, &code)
        return NULL;

    /* get the size */
    n = PyList_GET_SIZE(code);

    /* loop over the members */
    for (i = 0; i < n; i++) {
        int v;
        PyObject *o = PyList_GET_ITEM(code, i);
        v = PyInt_AsLong(o);
        if (v == -1 && PyErr_Occurred())
            goto oops; /* not a number */
        printf("%d\n", v);
    }

    ...

}

to handle any kind of sequence, use the PySequence abstract
API instead of the PyList API.

Cheers /F




From prairiesasquatch at crosswinds.net  Thu Mar 15 22:36:35 2001
From: prairiesasquatch at crosswinds.net (Kevin Russell)
Date: Thu, 15 Mar 2001 19:36:35 -0800
Subject: Spreadsheet-like array in Tkinter??
References: <3AB021B0.3F3DCA1F@att.net>
Message-ID: <3AB18A43.5BDC8B22@crosswinds.net>

An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20010315/a02392e6/attachment.html>

From ransen_spam_me_not at nemo.it  Sat Mar  3 12:09:52 2001
From: ransen_spam_me_not at nemo.it (Owen F. Ransen)
Date: Sat, 03 Mar 2001 17:09:52 GMT
Subject: Bug in Py_Finalize() ?
References: <mailman.983613305.18006.python-list@python.org>
Message-ID: <3aa10f5b.459867@news.newsguy.com>

On Sat, 3 Mar 2001 04:54:01 -0500, "Tim Peters" <tim.one at home.com>
wrote:

>You didn't give a complete program so it's impossible for me to say whether
>your problem is fixed.  I did put your loop in an extension module and
>boosted the loop limit to 1000.  No problem under Windows Python 2.1b1, under
>either release or debug builds.

That was the complete program, I had narrowed it down that
far to find out what was crashing.

I downloaded the Windows installer for 2.1b1 but there does not
appear to be any debug libraries...


--
Owen F. Ransen
http://www.ransen.com/
Home of Gliftic & Repligator Image Generators


From aleaxit at yahoo.com  Sat Mar 10 17:44:14 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sat, 10 Mar 2001 23:44:14 +0100
Subject: Can anyone recomend a good intoduction to C...
References: <mailman.983894288.19965.python-list@python.org> <ewwp6.1466$y6.241005@ruti.visi.com> <slrn9aeblb.cg1.scarblac@pino.selwerd.nl> <djagat4kk3cs9hg0c3k39cudk6goblhkg6@4ax.com> <989mbk$24nuv$1@fido.engr.sgi.com> <slrn9ah66m.k09.qrczak@qrnik.zagroda> <98auhn0270j@news2.newsguy.com> <slrn9aiaq7.rmu.qrczak@qrnik.zagroda> <98dpoe12ckn@news1.newsguy.com> <Pine.LNX.4.21.0103102137300.7964-100000@kev.ioc.ee>
Message-ID: <98eamb02ug2@news1.newsguy.com>

"Pearu Peterson" <pearu at cens.ioc.ee> wrote in message
news:Pine.LNX.4.21.0103102137300.7964-100000 at kev.ioc.ee...
>
> On Sat, 10 Mar 2001, Alex Martelli wrote:
>
> > "Expected" is correct; it's a social issue, NOT a technical one --
> > technically, C++ is a better choice for most Python extensions &c
    [snip]
> I must say that earlier I had also an opinion that no matter what I would
> not use C++ for extending Python --- the reasons of my opinion are
> now irrelevant after I learned about Boost. Though I had a gap about 2
> years of programming in C++, interfacing a C++ library to Python using
> Boost was so easy that I caught up with C++ and started to use it on

Glad to see somebody seconding my opinion -- particularly somebody
who was so adamantly set _against_ C++ before the Boost experience
(remember how astonished you were that I even _considered_ C++ for
gmpy...?  It would be done by now, and well on its way to a second
and richer release, if I had chosen C++ rather than C...:-).


> In conclusion, extending Python with C++ libraries in connection with
> Boost works very fine on Linux **if** you have plenty of memory available
> in your computer for compilation. At least, that's my experience.

Very good point.  Yes, depending on your compiler, building template-
rich C++ libraries can be a problem, if you're memory-scarce.  On my
old home Win98 machine, with VC++6, I get by with 64M without too
many tears -- but for a newer WinME box for my girlfriend I insisted
on 128M, and my own sooner-or-later Athlon/Linux box will have 256M
for sure (fortunately, that IS dirt-cheap today, _if_ one doesn't aim at
DDR or even Rambus...!-).


Alex





From shaleh at valinux.com  Wed Mar 28 11:16:10 2001
From: shaleh at valinux.com (Sean 'Shaleh' Perry)
Date: Wed, 28 Mar 2001 08:16:10 -0800 (PST)
Subject: list vs tuple
In-Reply-To: <hGlw6.782$p5.3210@news1.rivrw1.nsw.optushome.com.au>
Message-ID: <XFMail.20010328081610.shaleh@valinux.com>

On 28-Mar-2001 deadmeat wrote:
> Whats the practical difference between a list[] and a tuple() ?
> 
> A tuple seems to be a very basic list, so unless it's faster, what's the
> point of it?
> 

As others point out they are immutable.  This can lead to self documenting code.

dir_list = ('/path', '/foo/bar', '/baz/bat/files')

dir_list shouldn't change and this documents it.



From qrczak at knm.org.pl  Sat Mar 24 09:38:24 2001
From: qrczak at knm.org.pl (Marcin 'Qrczak' Kowalczyk)
Date: 24 Mar 2001 14:38:24 GMT
Subject: static methods
References: <Pine.LNX.4.21.0103231640070.5553-100000@clarkevans.com> <mailman.985386272.7736.python-list@python.org>
Message-ID: <slrn9bpcb0.bm7.qrczak@qrnik.zagroda>

Fri, 23 Mar 2001 14:27:42 -0800, Paul Prescod <paulp at ActiveState.com> pisze:

> Probably because they are simply not neccessary. 
> 
> def _static():
> 	XXX.foo=bar
> 	return XXX.foo
> 
> class Class:
> 	staticmethod = _static
> 	...

It doesn't work. Try to call it.

The fact that you are not the first person who thinks it works
suggests that the current Python's solution is not ideal :-)

-- 
 __("<  Marcin Kowalczyk * qrczak at knm.org.pl http://qrczak.ids.net.pl/
 \__/
  ^^                      SYGNATURA ZAST?PCZA
QRCZAK


From just at letterror.com  Tue Mar 13 10:34:14 2001
From: just at letterror.com (Just van Rossum)
Date: Tue, 13 Mar 2001 16:34:14 +0100
Subject: main in mac?
References: <20010313095339.19264.00000243@ng-dh1.aol.com> <3AAE3A1A.56124135@letterror.com>
Message-ID: <3AAE3DCA.8DBE482D@letterror.com>

I wrote:
> Hope this helped,

Thank you Netscape, for posting this three times. I keep forgetting that
an error doesn't neccesarily mean failure...

Sorry. I think I'll go back into lurk mode.

Just


From dyoo at hkn.eecs.berkeley.edu  Mon Mar 12 08:51:47 2001
From: dyoo at hkn.eecs.berkeley.edu (Daniel Yoo)
Date: Mon, 12 Mar 2001 05:51:47 -0800 (PST)
Subject: I have no idea
In-Reply-To: <E14cRZN-0005lZ-00@mail.python.org>
Message-ID: <Pine.LNX.4.21.0103120548030.16163-100000@hkn.eecs.berkeley.edu>

> In basic you use to be able to do this:
> 10 print "Hello World"
> 20 goto 10

In Python, you can write an infinite loop like this:

###
while 1:
    print 'Hello World'
###

Python follows the 'structured programming' philosophy and doesn't support
arbitrary goto's.  If you've programmed a lot in basic, this will probably
take some getting used to.




From nomad*** at ***freemail.absa.co.za  Wed Mar 14 09:06:54 2001
From: nomad*** at ***freemail.absa.co.za (nomad)
Date: Wed, 14 Mar 2001 14:06:54 GMT
Subject: [Q]: How can I get the logged in users username on Win32?
Message-ID: <3aaf7ab4.3152221@news.is.co.za>

Howzit,

After searching through the Python Docs, I found that on Un*x I could
use getpass.getuser(), but there is no equivalent for the Win32
platform.

In other words, how can I get the username in Windows?

As a side note, how can I enable Windows Scripting Host support for
python - there is mention of python in the original WSH docs - and if
I can use python scripts in WSH how would I call the built-in
Wscript.<methods>

TIA

nomad


From warren-postma at home.com  Fri Mar 23 14:38:13 2001
From: warren-postma at home.com (Warren Postma)
Date: Fri, 23 Mar 2001 19:38:13 GMT
Subject: Python Script to dial a VPN connection, automate VPN routing table updates.
Message-ID: <FCNu6.66711$A6.13912607@news4.rdc1.on.home.com>

Sorry for the nearly-off-topic nature of this post, but ....

I work from home now, and there's something in Windows 2000 that's
annoyingly manual:

I have a permanent internet connection via @home cable modem, and I "dial"
through that to create a VPN connection to the office.  The problems are
two:

(1) Windows won't accept or save my userid, password and the domain name, so
I have to enter it manually.

(2) After connecting, I have two default gateways in my routing tables
(ROUTE PRINT) and I need to update it so only the traffic that is destined
to the office goes through that interface.  The IP addresses I get given are
different each time, making a batch file insufficient, so voila, a perfect
use for python.

So after solving #2 above in Python, I wondered, does anyone know how, in
Python win32 extensions, one might dial an RAS connection, and if it's
possible to do so  through the win32 API somewhere or through COM? Then I
could solve #1 above as well.

Anyone else think this might make a nice little Python applet if I build in
some more VPN/RAS stuff and maybe even Gui-fy it a bit in wxPython or
tkInter.

Warren Postma





From lsommerer at ltec.net  Fri Mar  2 23:00:40 2001
From: lsommerer at ltec.net (L Sommerer)
Date: Fri, 02 Mar 2001 22:00:40 -0600
Subject: A counter-proposal to __future__ in PEP 236
References: <j47l29lpo8.fsf@informatik.hu-berlin.de> <97ll2u02qmp@news2.newsguy.com> <3A9FBE5C.95615241@sewardweb.com> <mailman.983578147.29040.python-list@python.org>
Message-ID: <3AA06C68.5D655E09@ltec.net>

> > I should also point out that the "from  __future__ import..." is also
> > confusing to a newbie, but others have said that already.
>
> And that's exactly as it should be: you shouldn't use the experimental
> feature(s) if you're a newbie :)

Yes, but we still look at your code, and we still try to make it fit into our model
(however limited it may be) of how python's syntax works.

Lloyd Sommerer



From claird at starbase.neosoft.com  Fri Mar 23 15:42:33 2001
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 23 Mar 2001 14:42:33 -0600
Subject: newbie question
References: <73125560B5E4EF16.CC3FC0DB3FD693E3.4AF0ACCBD2EFF29B@lp.airnews.net> <99fll8$996$1@rex.ip-plus.net> <mailman.985367791.3511.python-list@python.org> <mailman.985375195.17238.python-list@python.org>
Message-ID: <84EF7ECA6B21860E.2BACA551CEE41AA7.AE240BE7DBA69E93@lp.airnews.net>

In article <mailman.985375195.17238.python-list at python.org>,
Moshe Zadka  <moshez at zadka.site.co.il> wrote:
>On 23 Mar 2001, claird at starbase.neosoft.com (Cameron Laird) wrote:
>
><snipped a lot of wise things about CGI and Apache>
>> That's a lot more exciting
>> for them than hearing that, with a mere week or two of
>> study, they can install Zope on their own.
>
>That's pure FUD. Zope takes <depending on your internet connection> time
>to install, about 5 minutes to compile and 30 seconds to startup. And
>using it is not necessarily harder then learning CGI. I'm only mentioning
>it because "Zope is hard" seems to be a commonly uttered phrase, one that
>I believed until I actually had to get a Zope server up and running.
>
>Hooking up Zope to Apache (say, with reverse proxying) is very easy, though
>it does use some Apache configuration knowledge. This can all be learned
>in a day, and anyone seriously using Apache needs to know about this
>stuff.
			.
			.
			.
There are a lot of people who are frivolously using Apache.

That is, I run into a lot of people who aren't equipped to
install Apache on their own, let alone Zope.  They can *buy*
Apache, though (more precisely, its service equivalent), and
they want to know how far that'll take them, without any more
"programming" or system administration on their part.  What
use could someone uncomfortable with Zope installation pos-
sibly have with Python as a programming language?  Well, it's
a surprise to me, too, but often the answer is, "quite a lot."

Moshe, you're right in all you write here.  I've seen quite a
few folks, though, for whom it's all irrelevant.  They imagine
themselves to have better things to do with a day than learn
reverse proxying (for example).
-- 

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


From bos at hack.org  Sat Mar 31 07:23:04 2001
From: bos at hack.org (Rikard Bosnjakovic)
Date: Sat, 31 Mar 2001 14:23:04 +0200
Subject: Couple of simple questions
References: <3ac4854a.1170697@news.is.co.za>
Message-ID: <3AC5CC28.3796F8B2@hack.org>

nomad wrote:

> Does anybody know what size python variabel types are?
> Eg. how many bytes is and integer?  What about a long?  I have

Table 3.9 in "Python Essential reference" by Beazley:

Type (built-in)          Size
Integer                  12 bytes
Long Integer             12 bytes + (nbits/16 + 1)*2 bytes
Float                    16 bytes
Complex                  24 bytes
List                     16 bytes + 4 bytes for each item
Tuple                    16 bytes + 4 bytes for each item
String                   20 bytes + 1 byte per character
Dictionary               24 bytes + 12*2^n bytes, n = log2 (nitems)+1
Class Instance           16 bytes + a dictionary object
Xrange object            24 bytes



-- 
Rikard Bosnjakovic - http://a214.ryd.student.liu.se/cv/ - ICQ: 1158217

Anyone sending unwanted advertising e-mail to my address will be
charged $250 for network traffic and computing time. By extracting my
address from this message or its header, you agree to these terms.


From aahz at panix.com  Fri Mar  9 21:48:42 2001
From: aahz at panix.com (Aahz Maruch)
Date: 9 Mar 2001 18:48:42 -0800
Subject: how to make a function thread-safe ?
References: <3AA94A6A.50138001@noaa.gov> <98bl7n$cn6$1@panix2.panix.com>
Message-ID: <98c4ma$feo$1@panix2.panix.com>

>From robinjim at earthlink.net:
>
> So, I have probably been making things overly complicated for myself since I
> do not write functions that use global data.  I assume such a function is
> also thread-safe if it treats its parameter list as read-only (probably as
> any "good" function should).  Correct?

Exactly.

> If you have an opportunity to respond, please feel free to respond to the
> group rather than me personally -- whichever you prefer.

I generally prefer responses to public posts stay in public, so that
other people can learn.
-- 
                      --- Aahz (Copyright 2001 by aahz at pobox.com)

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

"..., and some of you may regard all women as evil traps that exist only
to tease, torture, and suck out your very soul."  --DrMax


From u263183593 at spawnkill.ip-mobilphone.net  Wed Mar  7 01:31:55 2001
From: u263183593 at spawnkill.ip-mobilphone.net (u263183593 at spawnkill.ip-mobilphone.net)
Date: Wed, 07 Mar 2001 06:31:55 GMT
Subject: Retrofitting 2.1b1 "No buffer space available" fix to 2.0
Message-ID: <l.983946715.1797119140@pool0776.cvx6-bradley.dialup.earthlink.net>

When using multiple threads, httplib, and timeoutsocket.py from the
Vaults of Parnassus under Win98SE, I get numerous "No buffer space
available" IOError exceptions.

This problem is fixed in the 2.1beta1 release. Is it possible to
retrofit 2.0 with the fix? I'd rather not wait for 2.1 to become
official.

I tried to rebuild 2.0 using the 1.30 revision of socketmodule.c from
sourceforge which fixes "closing sockets was not thread-safe", but I
still got the same IOErrors. Does anyone know which other changes
might have fixed the problem? socket.py didn't seem to have any
relevant changes.
 



-- 
Sent by  bortg from  hotmail piece  from  com
This is a spam protected message. Please answer with reference header.
Posted via http://www.usenet-replayer.com/cgi/content/new


From ngps at madcap.dyndns.org  Mon Mar 12 10:14:52 2001
From: ngps at madcap.dyndns.org (Ng Pheng Siong)
Date: 12 Mar 2001 15:14:52 GMT
Subject: Packages, modules and exceptions
Message-ID: <98ip5c$397$1@clematis.singnet.com.sg>

Hi,

In M2Crypto, I have the following (abridged) directory structure:

    M2Crypto/
            SSL/
                __init__.py
                Connection.py

Within M2Crypto/SSL/__init__.py, I have the following:

    class SSLError(Exception): pass


Now, various parts of M2Crypto/SSL/Connection.py wish to raise SSLError 
when it is appropriate to do so. 

For M2Crypto/SSL/Connection.py to get access to SSLError which is defined
in M2Crypto/SSL/__init__.py, it (Connection.py) contains the following:

    SSLError = 
     __import__('M2Crypto.SSL', globals(), locals(), 'SSLError').SSLError

This works, but looks kinda funny.

An equivalent(?) is:

    SSLError = 
     getattr('M2Crypto.SSL', globals(), locals(), 'SSLError'), 'SSLError')

Better, but still doesn't look very nice.

What's the best way to do this?

TIA for any input. Cheers.


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



From scarblac at pino.selwerd.nl  Tue Mar 27 04:27:00 2001
From: scarblac at pino.selwerd.nl (Remco Gerlich)
Date: 27 Mar 2001 09:27:00 GMT
Subject: math.choice attribute error
References: <Pine.LNX.4.21.0103270124290.3010-100000@moroni.pp.asu.edu>
Message-ID: <slrn9c0n1i.o22.scarblac@pino.selwerd.nl>

iddwb <iddwb at imap1.asu.edu> wrote in comp.lang.python:
> 
> from the ref man math.choice says it takes a sequence object and returns a
> random choice. perfect for my needs. the follow happens:
> 
> >>> import math
> >>> list = ['apple', 'orange', 'pear']
> >>> print math.choice(list)
> Traceback ....
> attributerror: choice
> 
> what have I missed?

You're confused, the function is in random, it's random.choice.

-- 
Remco Gerlich


From aahz at panix.com  Thu Mar 15 13:18:30 2001
From: aahz at panix.com (Aahz Maruch)
Date: 15 Mar 2001 10:18:30 -0800
Subject: This math scares me
References: <3aad4062.1290021@News.CIS.DFN.DE> <erRr6.659$lz1.20662@ruti.visi.com> <98qv55$f5d$1@news.udel.edu> <Iu7s6.897$lz1.80703@ruti.visi.com>
Message-ID: <98r11m$2pm$1@panix6.panix.com>

In article <Iu7s6.897$lz1.80703 at ruti.visi.com>,
Grant Edwards <grante at visi.com> wrote:
>
>If that's good enough for you, then go ahead and use FP.
>
>But, the people who actually write financial SW for a living seem to
>think it's not good enough.  You would have to talk to one of them to
>find out the exact situations where FP proved insufficient.  But, I
>don't think anybody would have gone to the trouble of implimenting BCD
>math just for the fun of it.  OTOH, I guess I know people who would do
>stuff like that just for fun.

Don't forget, most of these arguments took place pre-754.  It might be
interesting to revisit the whole discussion from scratch in light of the
advances in computing power etc.
-- 
                      --- Aahz  <*>  (Copyright 2001 by aahz at pobox.com)

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

"The overexamined life sure is boring."  --Loyal Mini Onion


From NoSpam at NoSpam.com  Sun Mar  4 12:33:13 2001
From: NoSpam at NoSpam.com (Tom)
Date: Sun, 04 Mar 2001 17:33:13 GMT
Subject: how to get the last interactive command enetered
References: <F2Tn6.15639$UZ4.4310299@news4.rdc1.on.home.com> <j4bsrjxjy3.fsf@informatik.hu-berlin.de> <I59o6.19418$UZ4.5669336@news4.rdc1.on.home.com> <j4zof17qp3.fsf@informatik.hu-berlin.de>
Message-ID: <t%uo6.25092$UZ4.7586037@news4.rdc1.on.home.com>

"Martin von Loewis" <loewis at informatik.hu-berlin.de> wrote in message
news:j4zof17qp3.fsf at informatik.hu-berlin.de...
> "Tom" <NoSpam at NoSpam.com> writes:
>
> > > No. Once the code is executed, the source code is dropped. All you can
> > > get at is the binary code.
> >
> > But when I catch a 'call' trace event the code has not been executed.  I
> > believe that it has been parsed and compiled but not executed (till I
get
> > the 'return' trace event).
>
> Yes, it has been parsed and compiled, the result of which was a code
> object (which in turn was put into a function object). After that, the
> source code, and the abstract syntax tree, is discarded (actually the
> source code is discarded in the process of parsing).
>
> The code object has a file name in it and line numbers, that's all
> that's left from the source.

I was afraid of that.

Thanks,

Tom.

> Regards,
> Martin




From tim.one at home.com  Sat Mar 24 02:35:50 2001
From: tim.one at home.com (Tim Peters)
Date: Sat, 24 Mar 2001 02:35:50 -0500
Subject: New PEP: The directive statement
In-Reply-To: <5.0.2.1.0.20010323183750.009f2510@mail.inet.com.br>
Message-ID: <LNBBLJKPBEHFEDALKOLCEEAGJIAA.tim.one@home.com>

[Carlos Ribeiro]
> ...
> c) Please keep the proposal *simple*.
>
>
> Man, this is turning into a nightmare...

That's why PEP 236 tries to address only one thing -- the immediate problem
at hand.

Beyond that, you're unlikely to end up with something simple; e.g., the C
compiler I use most often has 24 *kinds* of "pragma" statements, several of
which come in variations, even up to the point of having "push" and "pop"
flavors that maintain a bloody *stack* of "current pragmas in effect".  At
least "from __magicname__ import xyz" is ugly enough to discourage trying to
extend it <0.4 wink>.

like-the-skull-&-crossbones-on-a-jar-of-poison-ly y'rs  - tim




From gregj at pobox.com  Sat Mar 10 15:34:19 2001
From: gregj at pobox.com (Gregory Jorgensen)
Date: Sat, 10 Mar 2001 20:34:19 GMT
Subject: Problems writting a wrapper around linux' ssh command
References: <31rd89.lki.ln@127.0.0.1>
Message-ID: <fdwq6.1906$54.1822@www.newsranger.com>

ssh explicitly opens /dev/tty to read the password, so it's not looking on stdin
(see the readpass.c file in your ssh sources). In a normal terminal session
stdin is mapped to /dev/tty, but a child process started with popen2 will have
stdin mapped to the Popen3.tochild file descriptor. Can you use one of the other
secure authentication methods ssh supports?


In article <31rd89.lki.ln at 127.0.0.1>, gradha at iname.com says...
>
>Hi.
>
>I am trying to write a python wrapper around ssh. The last and only problem
>I have is password verification. I try to popen2.popen3 the ssh command,
>but no matter what I write to the input file, the ssh command ignores it
>and keeps requesting *keyboard* input.
>
>Is there any way to avoid this?


Greg Jorgensen
Deschooling Society
Portland, Oregon, USA
gregj at pobox.com


From alf at leo.logilab.fr  Thu Mar 15 05:01:08 2001
From: alf at leo.logilab.fr (Alexandre Fayolle)
Date: Thu, 15 Mar 2001 10:01:08 +0000 (UTC)
Subject: Regular Expressions and Python 1.5.2
References: <98q0k8$2b77$1@norfair.nerim.net>
Message-ID: <98q3t4$2d6o$1@norfair.nerim.net>

Alexandre Fayolle <alf at leo.logilab.fr> wrote:
> Is this a known bug in the module re of python 1.5.2, or am I missing 
> something obvious?

>>>> import re
>>>> t = "m&n"
>>>> re.sub('[^/_ -]','*',t)
> '***'
>>>> re.sub('[^_ -/]','*',t)
> '*&*'

[hits himself on the head with a dead fish]
OK, I was bit tired when I wrote the code. I had forgotten that '-' has 
a special meaning. Obviously, '&' lies somewhere between '_' and '/'.

Sorry for the noise.

Alexandre Fayolle
-- 
http://www.logilab.com 
Narval is the first software agent available as free software (GPL).
LOGILAB, Paris (France).


From wtanksle at dolphin.openprojects.net  Sat Mar 31 23:02:00 2001
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Sun, 01 Apr 2001 04:02:00 GMT
Subject: PythonLabs Team Moves to ActiveState
References: <mailman.986086501.5576.python-list@python.org>
Message-ID: <slrn9cda1n.l68.wtanksle@dolphin.openprojects.net>

On Sat, 31 Mar 2001 19:54:59 -0500, Guido van Rossum wrote:
>PythonLabs Team Moves to ActiveState

I'm glad we're finally getting some stability.

>   Barry has already approved PEP 357, Adding Block Delimiters to Python.

Okay, okay -- but these things are trivial.  If we really want to solve
these problems the right way, we have to behave completely like Pascal,
except with curly braces.

Perl has to as well.

The two languages should actually be exactly identical; the difference in
name can be explained as TMTOWTDI for one crowd, and The Best Way To Do it
for the other.

Oh, and get rid of those annoying "module" things.  They're so passe'.

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

-- 
-William "Billy" Tanksley


From phd at phd.pp.ru  Wed Mar  7 04:54:03 2001
From: phd at phd.pp.ru (Oleg Broytmann)
Date: Wed, 7 Mar 2001 12:54:03 +0300 (MSK)
Subject: GadFly and gfplus.py working with Python 2.0?
In-Reply-To: <Pine.GSO.4.03.10103071040190.13047-100000@sun1.lrz-muenchen.de>
Message-ID: <Pine.LNX.4.30.0103071253340.3221-100000@fep132.fep.ru>

On Wed, 7 Mar 2001, Eugene Leitl wrote:
> See subject.
> Aye, nay, maybe?

   Why not just try yourself? (Works for me, YMMV)

Oleg.
----
     Oleg Broytmann            http://phd.pp.ru/            phd at phd.pp.ru
           Programmers don't die, they just GOSUB without RETURN.




From crystalc at interlog.com  Fri Mar 16 23:34:14 2001
From: crystalc at interlog.com (Brian Alexander)
Date: Fri, 16 Mar 2001 23:34:14 -0500
Subject: ODBC for Python 2?
Message-ID: <0IBs6.29099$ZT5.789151@news.total.net>

Hello;

Is it possible to make use of Win32 ODBC drivers from Python 2? I would like
to run updates/synchronizations on a remote SQL database.

Brian.




From michel at digicool.com  Sun Mar 11 21:05:48 2001
From: michel at digicool.com (Michel Pelletier)
Date: Sun, 11 Mar 2001 18:05:48 -0800 (PST)
Subject: [Python-Dev] Revive the types sig?
In-Reply-To: <3AAC08DB.9D4E96B4@ActiveState.com>
Message-ID: <Pine.LNX.4.32.0103111745440.887-100000@localhost.localdomain>

On Sun, 11 Mar 2001, Paul Prescod wrote:

> Let's discuss my proposal in the types-sig. Executive summary:
>
>  * incremental development policy
>  * syntax for parameter type declarations
>  * syntax for return type declarations
>  * optional runtime type checking
>  * goals are better runtime error reporting and method documentation

I could be way over my head here, but I'll try to give you my ideas.

I've read the past proposals for type declarations and their
syntax, and I've also read a good bit of the types-sig archive.

I feel that there is not as much benefit to extending type declarations
into the language as their is to interfaces.  I feel this way because I'm
not sure what benefit this has over an object that describes the types you
are expecting and is associated with your object (like an interface).

The upshot of having an interface describe your expected parameter and
return types is that the type checking can be made as compile/run-time,
optional/madatory as you want without changing the language or your
implementation at all.  "Strong" checking could be done during testing,
and no checking at all during production, and any level in between.

A disadvantage of an interface is that it is a seperate, additional step
over just writing code (as are any type assertions in the language, but
those are "easier"  inline with the implementation).  But this
disadvantage is also an upshot when you immagine that the interface could
be developed later, and bolted onto the implementation later without
changing the implementation.

Also, type checking in general is good, but what about preconditions (this
parameter must be an int > 5 < 10) and postconditions and other conditions
one does now with assertions.  Would these be more language extensions in
your propsal?

As I see it, interfaces satify your first point, remove the need for your
second and third point, satify your fourth point, and meet the goals of
your fifth.

Nice to meet you at the conference,

-Michel





From thomas at xs4all.net  Fri Mar  9 16:24:46 2001
From: thomas at xs4all.net (Thomas Wouters)
Date: Fri, 9 Mar 2001 22:24:46 +0100
Subject: news from python9!!!
In-Reply-To: <uJRp6.46512$__6.10501931@typhoon.southeast.rr.com>; from tuttledon@hotmail.com on Thu, Mar 08, 2001 at 08:13:14PM +0000
References: <mailman.984074414.26983.python-list@python.org> <uJRp6.46512$__6.10501931@typhoon.southeast.rr.com>
Message-ID: <20010309222446.J404@xs4all.nl>

On Thu, Mar 08, 2001 at 08:13:14PM +0000, Don Tuttle wrote:
> "Fredrik Lundh"
> > (and the subject should probably read "news from active-
> > state's marketing department" ;-)

> Hey, you gotta run with what you can find, seeing how all you conventioneers
> where having way too much fun to drop into the newsgroup and give us an
> update! <wink>

> (Sorry about the award.  Maybe next year.)

Hey, Fredrik got a seat on the PSF Board of Directors, he should be damned
happy! And now he won't be eligible for any PSF-sponsored award next year,
of course <wink wink>.

-- 
Thomas Wouters <thomas at xs4all.net>

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



From shredwheat at mediaone.net  Mon Mar 19 10:17:51 2001
From: shredwheat at mediaone.net (Pete Shinners)
Date: Mon, 19 Mar 2001 15:17:51 GMT
Subject: newbie needs help installing NumPy, reference material.
References: <xeAq6.3544$R_6.353426@newsread2.prod.itd.earthlink.net> <Pine.HPX.4.33.0103182149420.10898-100000@sesame.ee.byu.edu>
Message-ID: <zqpt6.71925$Ch.9135330@typhoon.we.rr.com>

> > Just trying to get numeric python installed on a win2000 machine with python
> > 2.0, Idle, etc installed (but without a c compiler). I've downloaded
> > http://download.sourceforge.net/numpy/NumPy1711_20.zip, but am unclear about
> > whether it's the right version (there's an 18.0 version) and how to install
> > it.  The documentation at www.python.org was a little outdated.

hi, i also have a precompiled binary of Numeric 17.3 for win32 and python2.0
this isn't the latest version of Numeric, but afaik, the latest versions
only add features for python2.1 anyways, so it shouldn't make a difference
for python2.0

http://pygame.seul.org/ftp/contrib/Numeric-17.3-win32-2.0.zip

i've been doing all the precompiled win32 releases for the 16.x and
17.x. when 17.3 was released i also created this archive, but it was
never included into the sourceforge archive.

anyways, you can also use the link above for 17.1.1 fine, just letting
you know about this newer version i have. as with either version, just
unzip the archive into your C:\PYTHON20 directory (wherever it may be)





From cyberian_bear at hotmail.com  Sat Mar 17 20:08:27 2001
From: cyberian_bear at hotmail.com (cyberian bear)
Date: Sat, 17 Mar 2001 20:08:27 -0500
Subject: module to do euler path and topological sorting in python
Message-ID: <CUTs6.80$gB1.365@typhoon.nyu.edu>

can anyone suggest where i can find the modules which would do those two
things.
I have found a  website which had a link to topological sorting but it says
that access is forbidden. www.python.org doesn't have any matching reuslts.
thanx
cb




From scarblac at pino.selwerd.nl  Wed Mar 28 03:28:52 2001
From: scarblac at pino.selwerd.nl (Remco Gerlich)
Date: 28 Mar 2001 08:28:52 GMT
Subject: R: unique items in lists
References: <_nZv6.2878$yU4.468236@news.infostrada.it> <slrn9c0ph1.o5h.scarblac@pino.selwerd.nl> <t5_v6.3466$yU4.477532@news.infostrada.it> <slrn9c1424.o5h.scarblac@pino.selwerd.nl> <3AC0E010.3CDC94FC@mentor.com>
Message-ID: <slrn9c37vt.q06.scarblac@pino.selwerd.nl>

Bob Cannard <bob_cannard at mentor.com> wrote in comp.lang.python:
> > The dictionary method does one dictionary action for each word, and then it
> > does a dict.keys() which takes time proportional to each unique word.
> 
> I don't know how dictionaries are implemented, but hopefully they're
> a lot faster than that - one would expect a hash or tree implementation,
> giving a logarithmic run time for each lookup. A linear implementation
> would be an obvious cause of poor run time performance for Python
> generally.

Dictionaries are hashes, I'm assuming a dictionary setattr is about constant
time. The "dict.keys()" at the end must be linear, since its result is
linear...

-- 
Remco Gerlich


From syver at NOSPAMcyberwatcher.com  Wed Mar 14 10:01:55 2001
From: syver at NOSPAMcyberwatcher.com (Syver Enstad)
Date: Wed, 14 Mar 2001 16:01:55 +0100
Subject: Python interacting with Java Script webpage?
References: <KFur6.8651$Vg3.600799@bgtnsc06-news.ops.worldnet.att.net>
Message-ID: <98o19c$o9s$1@troll.powertech.no>

Use urllib or httplib, these are very simple to use and there are good
examples in the docuementation.

Find out what the names of the form fields are and use these in the
dictionary you use to send the form data

Both urllib and httplib returns a filelike object that is easy to stream out
to disk,

In addition urllib has got a method that writes the content to disk,
urllib.urlretrieve

Here is an example using urlopen:

The form fields used here would be syver, christine, marius.

file = urllib.urlopen('http://localhost/test/testform.asp',
data=urllib.urlencode({'syver': 30, 'christine': 33, 'marius': 30}))

The data parameter is where you pass in the form content. The easiest way is
as shown by using a dictionary and the urllib.urlencode function that
converts the dictionary into a format that you can use for query strings or
form posts.

an example using urlretrieve
ret = urllib.urlretrieve('http://localhost/test/testform.asp',
filename='form.txt', data=urllib.urlencode({'syver': 30, 'christine': 33,
'marius': 30}))

as you can see it's almost equal to urlopen, but you also specify a filename
that urlretrieve will write the content to.

Hope this helps you and doing what you want to do.


"William Dandreta" <wjdandreta at worldnet.att.net> wrote and wrote
> I would like to write a Python program that will be able to connect to a
> webpage that runs a Java Script, post data to the form displayed, and save
> the data retrieved to a file.





From tm at kernelconsult.com  Wed Mar 21 20:44:52 2001
From: tm at kernelconsult.com (TM)
Date: Thu, 22 Mar 2001 01:44:52 GMT
Subject: Please rate these Python books
Message-ID: <20010322.1445209@mis.configured.host>

Hi everybody,

I'm wondering if the last books on Python are good. Thanks for your 
points of view.
I'm especially thinking about :

1/Programming Python 2nd edition @ O'reilly
2/ Core Python Programming / Chun / Prentice Hall 
3/ Python Essential Reference / David Beazley / New Riders Publishing 

Thanks again




From nospam at mega-nerd.com  Thu Mar 15 15:43:18 2001
From: nospam at mega-nerd.com (Erik de Castro Lopo)
Date: Thu, 15 Mar 2001 20:43:18 GMT
Subject: decss
References: <Pine.LNX.4.21.0103151511300.7874-100000@tarpon.exis.net>
Message-ID: <3AB12769.6AF5ECDE@mega-nerd.com>

Jay Collins wrote:
> 
> Anyone every convert that perl script to python? 

Yep. I often get Perl code from the net. If I need to make anything
more than the most rudimentary changes to it I convert it to Python
first.

> I was just wondering how it would look

Python is FAR more readable.

> /perform.

The differences in speed between optimal Perl and optimal Python
are negligible.

Erik
-- 
-----------------------------------------------------------------
Erik de Castro Lopo     nospam at mega-nerd.com      (Yes its valid)
-----------------------------------------------------------------
"Two hands working can do more than a thousand clasped in prayer." 
-- anonymous


From gerhard.nospam at bigfoot.de  Thu Mar  1 01:29:06 2001
From: gerhard.nospam at bigfoot.de (Gerhard =?iso-8859-1?Q?H=E4ring?=)
Date: Thu, 01 Mar 2001 07:29:06 +0100
Subject: Python + MySQL + qt
References: <slrn99qh6f.7h0.jp@localhost.localdomain>
Message-ID: <3A9DEC32.941EFF71@bigfoot.de>

Jacek Pop?awski wrote:
> 
> I have a school project: create database and write application which will use
> it (I am not sure what exactly it will do, yet), it should have graphical menu
> and be user friendly. I can choose tool. I am thinking about MySQL or
> Postgresql (probably MySQL) and qt/gtk+ as toolkit (probably qt).

If the database is not very complex, it wouldn't matter which one you
choose. If the database server should run on Windows, too, I'd use MySQL
or Interbase, not Postgres.

> I am much more experienced in C than in Python, but I like this language and I
> think it will be next step to learn it more.
> And now questions:
> - is Python good choice here? what troubles will I have?

Python is sure an excellent choice here. Python is very good at
integrating things. Troubles? Probably lack of (good) documentation. For
Python and databases, you might want to check out the database topic
guide on the Python homepage.

> - what libraries will I need except pygtk or pyqt?

The appropriate database module for the database you choose. For MySQL,
there is MySQLdb (http://dustman.net/andy/python/MySQLdb/) For
PostgreSQL, there are several. I use PyGreSQL
(http://www.druid.net/pygresql/).

> - I am programming only in Linux, is it possible to run my application in
>   Windows, without any code fixes? are needed libraries available for Win?

The Python database modules MySQLdb and PyGreSQL (for PostgreSQL) are
available for Windows at my homepage (see below).
I have a bookmark for pyGTK on win32, for the case you want to go with
GTK: http://www.hans.breuer.org/ports/

I am quite a fan of wxPython. It's a very powerful library, and quite
easy to program. OK, it's not Visual Basic (yet), but who wants Visual
Basic, anyway ;-)

Gerhard

PS: Some of the modules you need might be already included in your Linux
distribution. I have SuSE, and it includes most of the above mentioned
Python modules.
-- 
Sorry for the fake email, please use the real one below to reply.
contact: g e r h a r d @ b i g f o o t . d e
web:     http://highqualdev.com


From insanc at cc.gatech.edu  Fri Mar 23 09:55:04 2001
From: insanc at cc.gatech.edu (Holland King)
Date: 23 Mar 2001 14:55:04 GMT
Subject: email address verification
References: <99d5qi$95l$1@solaria.cc.gatech.edu> <NFzu6.9043$G72.216162@e420r-atl1.usenetserver.com> <99fjv8$bhf$1@solaria.cc.gatech.edu> <E006A65FA23EFF2C.2D74AEB35552E0B6.065EC0B29EF187C6@lp.airnews.net>
Message-ID: <99fo48$d9h$1@solaria.cc.gatech.edu>

Cameron Laird <claird at starbase.neosoft.com> wrote:
: I don't understand your description.  In general,

ok, we run a server that holds a lot of accounts, once an account expires
then we run a script that parses the .forward file found in the home 
dir. the script then takes the addresses found and makes them into aliases
so that hopefully the user will not lose any mail. we need to make sure
that the aliases are real addresses so that at worst mail is not bounced 
and at best it doesn't go to a mailing list (i know i am being an optimist)

: comp.mail.misc might be a more productive place to
: ask these questions than c.l.p.

ok thank you i will try there to, the reason i asked here is that i tried
comp.linux.misc and they didn't help and i am writing it in python (i 
think my orignal post mention that i had found the smtp library and i 
was wondering if there was something similar that would work)

: Are you asking about verifying e-mail addresses *on
: hosts you control*?  That's entirely different!

no

-- 
Joseph Holland King  | "God whispers to us in our pleasures, speaks in our
                     |  conscience, but shouts in our pains: it is His
                     |  megaphone to rouse a deaf world." C. S. Lewis


From amk at mira.erols.com  Sun Mar 11 13:37:52 2001
From: amk at mira.erols.com (A.M. Kuchling)
Date: 11 Mar 2001 18:37:52 GMT
Subject: Color curses and Python 2?
References: <slrn9amlnn.ktp.jimd@vega.starshine.org>
Message-ID: <slrn9anibn.i3.amk@mira.erols.com>

On 11 Mar 2001 10:44:09 GMT, Jim Dennis <jimd at vega.starshine.org> wrote:
> I thought I'd heard that color ncurses support would be included
> in 1.6.  Was it?  I still haven't seen a 1.6.x package

Yes, it was.  Perhaps the Debian version of Python is miscompiled
(unlikely), or perhaps your terminal definition is set to something
that doesn't support colour.

--amk


From gzeljko at sezampro.yu  Tue Mar 20 12:44:29 2001
From: gzeljko at sezampro.yu (gzeljko)
Date: Tue, 20 Mar 2001 18:44:29 +0100
Subject: Sequence-spreading
References: <3AB7167F.2816AA21@student.liu.se> <99790u01c73@news2.newsguy.com>
Message-ID: <99851p$ba6$2@neptun.beotel.net>

Alex Martelli <aleaxit at yahoo.com> wrote in message
> offsets = [ [] for i in range(pm) ]
> for i in range(len(arg)):
>     offsets[i%pm].append(i)
>
>
> Alex
>

Old syntax yet works:

offsets = [[]]*pm

ly-y'rs-gzeljko




From aleaxit at yahoo.com  Sun Mar 11 03:04:57 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sun, 11 Mar 2001 09:04:57 +0100
Subject: Can anyone recomend a good intoduction to C...
References: <mailman.983894288.19965.python-list@python.org> <ewwp6.1466$y6.241005@ruti.visi.com> <slrn9aeblb.cg1.scarblac@pino.selwerd.nl> <djagat4kk3cs9hg0c3k39cudk6goblhkg6@4ax.com> <989mbk$24nuv$1@fido.engr.sgi.com> <slrn9ah66m.k09.qrczak@qrnik.zagroda> <98auhn0270j@news2.newsguy.com> <slrn9aiaq7.rmu.qrczak@qrnik.zagroda> <98dpoe12ckn@news1.newsguy.com> <3AAA85C1.81FB90AA@shore.net> <98eb1f02uqq@news1.newsguy.com> <3AAAE0BB.CBCC99EC@alcyone.com>
Message-ID: <98fbme0ta8@news1.newsguy.com>

"Erik Max Francis" <max at alcyone.com> wrote in message
news:3AAAE0BB.CBCC99EC at alcyone.com...
> Alex Martelli wrote:
>
> > It's no doubt possible that gcc has significantly worsened in the
> > last 5 or 6 years, but I'll confess that would really surprise me.
>
> No, it's actually gotten far better.  The pre-egcs versions of g++ had
> some really glaringly seriously problems with modern draft Standard (now
> actual Standard) compliance like exceptions, templates and the like.  If
> you weren't using these, though, g++ was just fine (which was probably
> the case for you).

Yes, this was before the Standard appeared, so we were not using
many of the Standard features -- templates being the only exception,
as it was obvious they bought us SO much it was worth taking a lot
of pain to use them.  (And it was indeed a lot:-).

> Note that modern version of egcs have all the
> Standard language _features_ all settled out, but there are a few
> Standard compliance problems in the library.  These are scheduled to be
> fixed in gcc 3.0.

One option we considered at one point was to purchase a library
somewhere (e.g., dinkumware) and use it instead of the one that
came with gcc.  Didn't pan out since we ended up on Windows
anyway, but early tests were promising.


Alex





From max at alcyone.com  Sun Mar 11 14:03:03 2001
From: max at alcyone.com (Erik Max Francis)
Date: Sun, 11 Mar 2001 11:03:03 -0800
Subject: Einstein's Riddle
References: <oNPq6.396644$ge4.138479566@news2.rdc2.tx.home.com>
Message-ID: <3AABCBE7.94C9BB61@alcyone.com>

Steve Graham wrote:

> Albert Einstein wrote this riddle this century [ed. 20th century].  He
> said
> 98% of the world could not solve it.

No, he didn't.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ Weakness of attitude becomes weakness of character.
\__/ Albert Einstein
    Interstelen / http://www.interstelen.com/
 A multiplayer, strategic, turn-based Web game on an interstellar scale.


From s713221 at student.gu.edu.au  Thu Mar 29 00:02:07 2001
From: s713221 at student.gu.edu.au (s713221 at student.gu.edu.au)
Date: Thu, 29 Mar 2001 15:02:07 +1000
Subject: warrior geeks
References: <3ABF72F1.98CB78F@sage.att.com> <mailman.985629094.30333.python-list@python.org>
Message-ID: <3AC2C1CF.4F70AF77@student.gu.edu.au>

Ken Seehof wrote:
> 
> Me too.  I'm an Aikidoist.
> 
 Me three, I'm an Aikidoka as well,

Joal Heagney/AncientHart


From NoSpam at NoSpam.com  Wed Mar 28 23:31:44 2001
From: NoSpam at NoSpam.com (Tom)
Date: Thu, 29 Mar 2001 04:31:44 GMT
Subject: Boost documentation.
References: <g0wv99x256.fsf@scumbag.ecs.soton.ac.uk>
Message-ID: <QUyw6.100011$A6.23429492@news4.rdc1.on.home.com>

The boost.org site has a link to the boost mailing list on Yahoo.  It has a
fair bit of boost.python traffic and would be a good place for your
question.

Tom.

"Jacek Generowicz" <jmg at ecs.soton.ac.uk> wrote in message
news:g0wv99x256.fsf at scumbag.ecs.soton.ac.uk...
> Are there any sources of documentation for the Boost Python Library
> besides the source examples, getting_started<1-5>.cpp ?
>
> How would I write a C++ function which receives a python list and
> converts it into a std::vector<int> for its own use ?
>
> Thanks,
>
> Jacek




From db3l at fitlinxx.com  Fri Mar  9 20:53:04 2001
From: db3l at fitlinxx.com (David Bolen)
Date: 09 Mar 2001 20:53:04 -0500
Subject: Remote script execution with Python ...
References: <7Ybp6.7729$r4.7336@news.indigo.ie> <ulmqil8qu.fsf@roundpoint.com> <Lpwp6.8059$r4.7712@news.indigo.ie> <98ank601u1b@news2.newsguy.com>
Message-ID: <u4rx28k5b.fsf@ctwd0143.fitlinxx.com>

"Alex Martelli" <aleaxit at yahoo.com> writes:

> "Robert Gahan" <gahanr at gofree.indigo.ie> wrote in message
> news:Lpwp6.8059$r4.7712 at news.indigo.ie...
> > Thanks Ben.
> >
> > os.popen2 is not available to windows programmers but
> 
> Works fine for me (NT4, Python 2.0 standard build).

Yes, the original post should be modified to say under Python 1.5.2,
where you do need to use the win32pipe module.

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


From jkraska1 at san.rr.com  Sat Mar 24 00:15:07 2001
From: jkraska1 at san.rr.com (Courageous)
Date: Sat, 24 Mar 2001 05:15:07 GMT
Subject: static methods
References: <mailman.985383828.2942.python-list@python.org> <99h8be$om7$1@panix3.panix.com>
Message-ID: <6ibobtoku1vb5m1p44k9fcd6bvdeipiprt@4ax.com>

>Without debating the rationale, the standard answer is to stick the
>class into a module with the same name and use module global functions
>to simulate static methods.

Yeah, it takes the uninitiated Python programmer a moment of pause
to realize that "global" variables and functions are okay in Python.
This is because every Python file is a "namespace", default. You
have automatic packaging, for free.

Python's packaging model succeeds at being elegant, easy to use,
syntactically lightweight, and conceptually the right thing.

C//



From chris.gonnerman at usa.net  Wed Mar 21 23:02:38 2001
From: chris.gonnerman at usa.net (Chris Gonnerman)
Date: Wed, 21 Mar 2001 22:02:38 -0600
Subject: Puzzling input problem
References: <3AB9449C.A1D14738@monitor.net>
Message-ID: <011801c0b284$f0abec00$a100000a@local>

----- Original Message -----
From: "Stephen R. Figgins" <fig at monitor.net>
Subject: Puzzling input problem


> I ran into a puzzling problem combining print with raw_input().
> Normally I would pass the prompt string to raw_input, but instead I
> tried something like this:
>
>   print "prompt string: ",
>   answer = raw_input()
>   print answer
>
> When it prints answer, it is indented one space.  Why is raw_input()
> putting a space in there?  Is it a bug, a feature, an unintended side
> effect?  input() does this too.   I am currently using Active Python 2.0
> build 202.

Actually, raw_input isn't the culprit, exactly.  The print statement's
format
handler doesn't know that raw_input() has modified the screen contents, so
it
still behaves like the cursor is just after the "prompt string: " on the
screen,
inserting a space as the ref guide says it will.

Do it this way instead:

    answer = raw_input("prompt string: ")
    print answer

and you should get what you want.






From urner at alumni.princeton.edu  Sat Mar  3 20:24:00 2001
From: urner at alumni.princeton.edu (Kirby Urner)
Date: Sat, 03 Mar 2001 17:24:00 -0800
Subject: Missing the functional mind set
References: <mlb2at8ptrr7sh5dgq8qlk2201kquh6tfl@4ax.com> <uDao6.335972$ge4.119457077@news2.rdc2.tx.home.com> <t6f2at870pk82q96bnd5a2od4n8ns5kf83@4ax.com> <ntg2atcemm3hp9rb26hn7c5uin4dfn8lc9@4ax.com> <97rqgb$lt2$1@tyfon.itea.ntnu.no>
Message-ID: <a663ats6ntl298rn1eusbas0vvq7rrrvg2@4ax.com>

"Magnus Lie Hetland" <mlh at idi.ntnu.no> wrote:
>
>This should be
>
>    >>> params = [None]*4

Yes, I seem to be constitutionally incapable of posting 
an error-free post.  Oh well...  As a certain USA 
ex-president might say 'I can feel my pain'.

Kirby



From paul at fxtech.com  Sun Mar 11 16:43:16 2001
From: paul at fxtech.com (Paul Miller)
Date: Sun, 11 Mar 2001 15:43:16 -0600
Subject: any interest in type-scoped static constants?
References: <3AAB9578.4C8754E4@fxtech.com> <mailman.984338774.14395.python-list@python.org>
Message-ID: <3AABF174.3C986EB4@fxtech.com>

> Why do you need a PEP when you can implement it for any extension you
> need. Just put a dictionary in the type and using type.getattr to look
> stuff up in your dictionary.

Correct me if I am wrong, but do you not need an *object* of the type
before you can call getattr on it? 

In other words, I would have to do this:

	constant_value = Point().CONSTANT_VALUE

Which would first CREATE an object of that type, then call its getattr()
method where I can look at the argument. I want to avoid creating
objects just to get a constant which is associated with the TYPE and not
the object.


From gmuller at worldonline.nl  Tue Mar  6 17:10:47 2001
From: gmuller at worldonline.nl (Gerrit Muller)
Date: Tue, 6 Mar 2001 23:10:47 +0100
Subject: Windows98+Idle0.6+ 32 bit color HW or SW problem?
Message-ID: <983n9c$mbn$1@nereid.worldonline.nl>

I recently acquired a new windows 98 computer and installed python 2.0 final
plus win32all-134. I now get horizontal disturbances in the idle and other
TKinter windows, when using 32 bit color mode and 1280*1024. When switching
between 16 bit and 32 bit color mode the Tkinter windows are apparantly not
updated, the contents of these windows becomes garbage.

Microsoft applications, such as IE and outlook express do not suffer from
these effects. Is there any known SW problem in Tkinter or related SW, which
may cause these problems. The alternative is suspect HW (AMD Thunderbird 900
MHz, ELSA Gladiac MX).

regards Gerrit




From stephen_purcell at yahoo.com  Thu Mar 15 05:31:34 2001
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Thu, 15 Mar 2001 11:31:34 +0100
Subject: reversing a dictionary (newbie)
In-Reply-To: <007201c0ad37$12571f80$6809a78f@dcs.shef.ac.uk>; from C.Brewster@dcs.shef.ac.uk on Thu, Mar 15, 2001 at 10:02:40AM -0000
References: <007201c0ad37$12571f80$6809a78f@dcs.shef.ac.uk>
Message-ID: <20010315113134.A5112@freedom.puma-ag.com>

Christopher Brewster wrote:
> I have written a program to extract pairs of words from a corpus (the BNC).
> It is my first Python program
> Once it has processed the files I "reverse" the dictionary so that the keys
> are frequencies and the values lists of word pairs of that frequency.

What if two words have the same frequency?  You probably want the 'reversing'
process to generate a list of (frequency, word) tuples.

But anyway... to reverse a dictionary, you can do:

>>> dict = {'a':1, 'b':2, 'c':3}
>>> reversed = {}
>>> for k, v in dict.items():
...   reversed[v] = k
>>> reversed
{3: 'c', 2: 'b', 1: 'a'}

Expect total memory usage to be triple that of 'dict' alone towards the end
of the loop, because of the 'items()' call.

-Steve

-- 
Steve Purcell, Pythangelist
Get testing at http://pyunit.sourceforge.net/
Any opinions expressed herein are my own and not necessarily those of Yahoo



From paul at fxtech.com  Wed Mar 14 18:58:11 2001
From: paul at fxtech.com (Paul T. Miller)
Date: Wed, 14 Mar 2001 17:58:11 -0600
Subject: Cameron Laird Quote on www.unixinsider.com -- Probably old news to 
 y'all, but new to me.
References: <bJOr6.9481$TW.38047@tor-nn1.netcom.ca>
Message-ID: <3AB00593.D1C4BBBF@fxtech.com>

> I apologize for trotting it out again, if I am repeating something earlier
> discussed, but I couldn't find any direct reference to it (my nntp server
> sucks, only goes back 7 days).
> 
> All I'll add is: Well said, Cameron!

I love Python. I especially love it as an embedded scripting language.
But it is not going to replace graphics applications where speed is
paramount. Many of these articles fail to take that into account. But
then again, many problem domains do not have the interaction constraints
that 3D or compositing applications have.


From tim.one at home.com  Sat Mar 24 00:38:50 2001
From: tim.one at home.com (Tim Peters)
Date: Sat, 24 Mar 2001 00:38:50 -0500
Subject: New PEP: The directive statement
In-Reply-To: <99dtt0$l6t$1@panix3.panix.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCIEAAJIAA.tim.one@home.com>

[James_Althoff]
> ok, how about:
>
> from  __readyOrNot__  import nested_scopes
>
> or perhaps,
>
> from  __likeItOrNot__  import nested_scopes

[Aahz Maruch]
> This is in some respects excessively flippant, but that very flippancy
> calls attention to it.  I'm not going to push this idea, but I do think
> it's worth considering.

Aahz, "__future__" was intended to be flippant enough to get peoples'
attention, and for some people that works (e.g., /F originally said "it makes
me smile", which was half the point).  But the people offended by an ounce of
irreverence aren't going to be happier with a full pound.

from__upyours__-ly y'rs  - tim




From greg at cosc.canterbury.ac.nz  Wed Mar  7 20:40:24 2001
From: greg at cosc.canterbury.ac.nz (Greg Ewing)
Date: Thu, 08 Mar 2001 14:40:24 +1300
Subject: sorry....never mind
References: <DZkp6.18811$pF4.3015652@news1.telusplanet.net> <fdlp6.16982$go.2157953@news0.telusplanet.net>
Message-ID: <3AA6E308.5B63ECC6@cosc.canterbury.ac.nz>

Victor Louie wrote:
> 
> <> is obsolescent.

Really? I hope it's not taken away, because that will
break ALL of my code! I hate != (looks too much like
C), so I always use <>. Nobody ever told me I was doing 
anything wrong!

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


From aleaxit at yahoo.com  Thu Mar 22 14:44:32 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 22 Mar 2001 20:44:32 +0100
Subject: python, perl, lisp, tcl / got a decision to make, can you give me a view point please ? and a joke for python users :)
References: <3aba3076.14395287@news.vif.com>
Message-ID: <99dkkd018e0@news1.newsguy.com>

"Yvon Boulianne" <mysticsguy at yahoo.ca> wrote in message
news:3aba3076.14395287 at news.vif.com...
> Hello,  i'm a bit confused with so many language ;-)
> theres the story, i like to learn a programming language for 2 reasons
> 1- i want to be able to make some script on my web site (i heard about
> zope, don't know what it have to do with python)

Zope is _written_ in Python; for the scripts, it lets you use python
itself, or, now, Perl as an alternative.  Anyway, Python is great
for that (and many other uses).


> 2- i want to learn to keep my mind trained, i'm a mystic and i want to
> make this mental grow in ability ;-)

Unfortunately, Python is far too good a language to serve this
purpose.  It's so simple you won't get much mental exercise
learning it, and so powerful it will make it too easy for you to
write all kinds of programs.  For your purpose [2], I suggest
Haskell (it starts out easy, but when you get to monads they
WILL give your mind a workout -- when, and if, you emerge
alive and still sane, you will have moved to a higher level of
consciousness) or javascript (chaotically and ramblingly hard
to learn, yet wimpy enough that using it to write significant
programs is going to be serious exercise each and every time).


Alex








From stevewilliams at wwc.com  Wed Mar  7 17:08:12 2001
From: stevewilliams at wwc.com (Steve Williams)
Date: Wed, 07 Mar 2001 22:08:12 GMT
Subject: I come to praise .join, not to bury it...
References: <kuofvfjb77.fsf@lasipalatsi.fi> <000001c0a64c$497df250$a9807018@cr706570a> <mailman.983997159.4842.python-list@python.org>
Message-ID: <3AA6B157.2B62BFA5@wwc.com>

D-Man wrote:

> On Tue, Mar 06, 2001 at 09:46:53AM -0500, Mike C. Fletcher wrote:
>

> [snip]

> Even with my newfound understanding, provided by the Martellibot, I
> agree with this.  Calling a method on an object is fine.  Calling it
> on a literal looks weird.  Take for example something like this
>
> three = 1.__add__( 2 )
>
> ;-)
>
> Continuing on into the darkness:

ten = +.join(1,2,3,4)

and

sixfactorial = *.join(range(1,7))

Since + and * are not quoted, it's clear that 'join' is an 'arithmetic
join'--polymorphism, doncha know.

And howsabout

   RatherLargeNumber ==  **.join(xrange(100))

The mind boggles.



From dinu at reportlab.com  Fri Mar 23 06:28:47 2001
From: dinu at reportlab.com (Dinu Gherman)
Date: Fri, 23 Mar 2001 11:28:47 GMT
Subject: Ruby will overtake Python within four years.
References: <mailman.985324111.22754.python-list@python.org>
Message-ID: <3abb31fb.184742956@news.vr.in-berlin.de>

On 23 Mar 2001 04:58:09 -0000, "prashant  patel"
<cognoscente at rediffmail.com> wrote:

>No I am not a ruby programmer not even using any scripting language other once used vbscript in ASP.

Then, how comes you predict "Ruby will overtake Python within 
four years." in your subject line?

>[...] But my inclination are towards python.

Likely the right conclusion... as in four years we'll be much closer
to the year 3000 than we're now, which probably means the type/
classes dichotomy will be gone, as well as Stacky Python. ;-)

Dinu



From not.this at seebelow.org  Thu Mar 22 13:44:11 2001
From: not.this at seebelow.org (Grant Griffin)
Date: 22 Mar 2001 10:44:11 -0800
Subject: New PEP: The directive statement
References: <mailman.985224089.1179.python-list@python.org>
Message-ID: <99dh5r013dg@drn.newsguy.com>

In article <mailman.985224089.1179.python-list at python.org>, "Tim says...
...
>I may like
>Martin's "transitional" better for this reason -- provided that it wakes
>people up enough to think about what they're reading.  Alas,
>
>    from __transitional__ import nested_scopes
>
>is not only ugly, it's not even funny, so would be worse than "from
>__future__ ...".  It works better with "directive" because it keeps it
>uniformly bland.

You know, if this "__future__" thing really goes through, I predict we'll
eventually see something like this:

   from __future__ import __transitional__

or maybe even

   import __future__ as __transitional__

(I'm not sure if the proposed syntax will support that, though. <wink>)

Now, I like a good joke as much as the next guy <wink, wink, say no more>, but
I'm not sure software is the place for jokes.

Problem is, computers seem to be inherently un-funny.  For example, imagine the
following from the old Star Trek series:

   Kirk: Computer?

   Computer: Working.

   Kirk: Why did the chicken cross the road?

   Computer: Specify chicken.

I once worked on a system which, whenver you made a typo in entering a command,
responded with, "Fat-finger typo error".  That was funny for the first day or
two, but it wore thin in about a week, and it became downright annoying after a
month.  Then, after close to a year, we finally got good at ignoring it.

Similarly, whenever I use "friend" in C++, I wince a little bit.  And not just
for the construct itself--for its name.  I don't think "friend" was ever meant
as a joke, but even so, it's just a _little_ too cute.

Oh, and don't get me started on Bill Gates' dancing paper clip...

>if-you-can't-laugh-at-your-tools-you're-taking-them-too-seriously-ly

Should we laugh at ourselves?  Definitely!  Should we laugh at each other? 
Hopefully!  But should we laugh at our tools?  Never!

(Sorry to be such a wet blanket. :-)

a-workman-is-known-by-his-tools-ly y'rs,

=g2

_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation            http://www.iowegian.com



From deblNonOspammy at theworld.com  Tue Mar 13 23:53:09 2001
From: deblNonOspammy at theworld.com (David Lees)
Date: Wed, 14 Mar 2001 04:53:09 GMT
Subject: Help doing simple threading
Message-ID: <3AAEF93E.F817D852@theworld.com>

I am trying to learn how to use threads.  I want multiple threads to be
able to communicate.  I thought that threads started by a common process
would share globals in that module.  I want thread to be able to store
results in a common array.  However the following returns all zeros, so
clearly I am not passing anything back to the top level list 'z'.  Can
someone tell me how I should be passing things back from threads?

Thanks in advance.

David Lees
---------------------------------------

import thread

#Create a top level list of length n
def create_len(x,n):
    for i in range(n):
        x.extend([0])
          

def save_stuff(k):
    z[k]=k*k

n=10
global z
z=[]
create_len(z,n)

for i in range(n):
    thread.start_new(save_stuff,(i,))

for i in range(n):
    print z[i]


From dougfort at downright.com  Mon Mar  5 06:15:03 2001
From: dougfort at downright.com (Doug Fort)
Date: Mon, 5 Mar 2001 06:15:03 -0500
Subject: [Advertisment] Re: Test Your Web Apps - article at LinuxQuality
References: <3AA34C6B.EC12B25B@goingware.com>
Message-ID: <XyKo6.20248$Sl.895586@iad-read.news.verio.net>

Hey! Don't forget about http://www.stressmy.com

"Michael D. Crawford" <crawford at goingware.com> wrote in message
news:3AA34C6B.EC12B25B at goingware.com...
> Tonight I posted:
>
> Use Validators and Load Generators to Test Your Web Applications
> http://linuxquality.sunsite.dk/articles/webapptesting/
>
> The article promotes the idea that one should check the HTML
> generated by a web application with a validator to ensure that
> it conforms to W3C standards, and that you should use a load
> generator to stress-test your server during development to
> detect performance problems early.
>
> Links to some load generators and validators are given, with
> tips on using them.
>
> I further suggest that if you generate XHTML documents rather
> than HTML, you can take advantage of the many available
> XML software packages to validate and process your documents.
>
> Comments, criticisms, corrections and suggested additions are
> appreciated.
>
> The Linux Quality Database at http://linuxquality.sunsite.dk/
> has the dual purpose of promoting higher quality among Free
> and Open Source software by publishing articles such as this,
> and also the development of an easy-to-use but powerful bug
> database to enable widespread public quality assurance of the
> linux kernel.
>
> You're welcome to contribute articles on the general topics
> of quality assurance and better programming practice.  I'm also
> looking for knowledgeable people who can help me design the schema
> for the bug database.
>
> --
> Michael D. Crawford
> GoingWare Inc. - Expert Software Development and Consulting
> http://www.goingware.com
> crawford at goingware.com
>
>   Tilting at Windmills for a Better Tomorrow.




From stephen_purcell at yahoo.com  Tue Mar 27 11:21:47 2001
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Tue, 27 Mar 2001 18:21:47 +0200
Subject: stripping cr/lf, lf, cr
In-Reply-To: <Lp2w6.705$p5.3213@news1.rivrw1.nsw.optushome.com.au>; from root@excalibur.iks-jena.de on Tue, Mar 27, 2001 at 03:34:03PM +0000
References: <Lp2w6.705$p5.3213@news1.rivrw1.nsw.optushome.com.au>
Message-ID: <20010327182147.B1260@freedom.puma-ag.com>

deadmeat wrote:
> Is there a platform independant function that will remove appropriate EOL
> characters from a string?  I don't want to use strip() since it will also
> remove other whitespace I want to keep.
> 

No, but it's easy to write one:

>>> import string, os
>>> def strip_eol(s):
...    pos = string.rfind(s, os.linesep)
...    if pos == -1: return s
...    if pos < len(s) - len(os.linesep):
...       return s
...    return s[:-len(os.linesep)]
... 
>>> os.linesep  
'\012'
>>> strip_eol('a line\n')
'a line'
>>> strip_eol('a line')
'a line'
>>> strip_eol('a \nline')
'a \012line'
>>> strip_eol('a \nline\n')
'a \012line'
>>> strip_eol('a line\r\n')
'a line\015'


-Steve

-- 
Steve Purcell, Pythangelist
Get testing at http://pyunit.sourceforge.net/
Any opinions expressed herein are my own and not necessarily those of Yahoo



From aleaxit at yahoo.com  Tue Mar  6 12:31:51 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Tue, 6 Mar 2001 18:31:51 +0100
Subject: Can Python replace Visual Basic? Should it?
References: <5DZo6.11126$hn5.1578410@news1.rdc1.mb.home.com> <98266k0p5h@news1.newsguy.com> <3q8p6.12338$hn5.1798569@news1.rdc1.mb.home.com>
Message-ID: <9836u70260g@news1.newsguy.com>

"Brad Bollenbach" <bbollenbach at homenospam.com> wrote in message
news:3q8p6.12338$hn5.1798569 at news1.rdc1.mb.home.com...
    [snip]
> > Why do you focus on 'used to use'?  One may perfectly well keep using
> > some VB (&c) for tasks it's particularly well-suited to, while using
> > Python wherever feasible (and, indeed, using both in the same project
>
> Certainly. And for all intents and purposes "used to use" isn't necessary,
> but the question happened to be directed (as its subject suggests) at
those
> who _have_ replaced VB with Python. :)

If only those who HAVE replaced answer, you'll get answers slanted
in favour of replacing -- which may be what you're after!-)


> But realistically, I just wanted to emphasize my interest in hearing
answers
> from those who have or currently are using VB or one of the other Win32
> visual development tools. Personally, I'm a former VB programmer now
working
> in Powerbuilder and am unhappy with both, again based on my dislike of
IDE's
> that obscure the programmer's view of what's going on.
>
> I could explain _why_ I don't like the 40,000 foot views of IDE's, but
> that's another thread.

You're definitely preaching to the choir, as far as I'm concerned,
on the 'IDEs which hide things' theme.


Alex





From jcollin at exis.net  Tue Mar 13 10:37:48 2001
From: jcollin at exis.net (Jay Collins)
Date: Tue, 13 Mar 2001 10:37:48 -0500
Subject: string.find() ?
In-Reply-To: <Pine.LNX.4.21.0103131027570.13200-100000@tarpon.exis.net>
References: <Pine.LNX.4.21.0103131027570.13200-100000@tarpon.exis.net>
Message-ID: <Pine.LNX.4.21.0103131037270.13200-100000@tarpon.exis.net>

Doh! I figured it out. had the arguments backward!




From qrczak at knm.org.pl  Sun Mar 18 09:13:47 2001
From: qrczak at knm.org.pl (Marcin 'Qrczak' Kowalczyk)
Date: 18 Mar 2001 14:13:47 GMT
Subject: Who's minister of propaganda this week?
References: <B98041329020E007.7DA84CAD9D0562AB.817BBC572F480869@lp.airnews.net> <98nu1m$12e@dispatch.concentric.net> <660ACCA1E4816F7E.C359CDF4168EF3FB.3B6B05538D619770@lp.airnews.net> <3AAFA13D.76E32C3F@divalsim.it> <98oa4o01i4v@news2.newsguy.com> <3AB0176D.8070903@yahoo.com> <mailman.984663621.25421.python-list@python.org>
Message-ID: <slrn9b9gkq.ahm.qrczak@qrnik.zagroda>

Thu, 15 Mar 2001 05:43:38 -0800, Paul Prescod <paulp at ActiveState.com> pisze:

> 3. Statically typed languages are absolutely not immune to type errors
> for a variety of reasons:
> 
>  * Null pointer exceptions are type errors. You've passed the null
> object when the code expected some other object. 

Not all languages have null pointers at all. (My favorites happen
not to have them.)

>  * Almost every statically typed language also has a set of features for
> working around the type system (dynamic casts). 

In many languages (Haskell, SML, OCaml, Clean, Mercury, Eiffel) they
are almost never used. Again, not all languages are like C++ and Java
where casting is essential.

>  * Even when the compiler thinks the "type" of your object is correct,
> there are a variety of type-related errors you can make such as passing
> a negative number where only positive ones are allowed,
[...]

Indeed. Nobody claims that static type system catch all type-related
errors. But they catch most of them, and it's better to catch some
than to catch none. The inability to catch all type-related errors
at compile time is hardly a disadvantage of static type systems wrt.
dynamic type systems, if dynamic type systems don't catch any of them
at compile time.

> 4. Yes, if you unit test every unit in your program, and do good
> coverage testing you WILL catch ever occurrence of some code calling
> foo. That's what good testing is about. And you need it, not to catch
> the basic type errors but to catch the subtle semantic errors. Catching
> the basic type errors comes for free.

This assumes that testing procedures are perfect and cover every
possible kind of data. Nobody's perfect. Tests still catch only
a proper subset of what static typing + tests catch together.

-- 
 __("<  Marcin Kowalczyk * qrczak at knm.org.pl http://qrczak.ids.net.pl/
 \__/
  ^^                      SYGNATURA ZAST?PCZA
QRCZAK


From aleaxit at yahoo.com  Mon Mar 12 04:01:17 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Mon, 12 Mar 2001 10:01:17 +0100
Subject: Static typing (was Re: Java guy interested in Python)
References: <mailman.984371806.20230.python-list@python.org>
Message-ID: <98i38q0nht@news2.newsguy.com>

"Tim Peters" <tim.one at home.com> wrote in message
news:mailman.984371806.20230.python-list at python.org...
    [snip]
> >     class int {
    [snip]
> > Isn't this perfectly legal C++, meaning something very different from
    [snip]
> As in the "Einstein's Riddle" thread, I was doing Rainer the favor of
> assuming he was trying to say something <ahem>.  If I don't extend you the
> same favor, then before I can answer your question I have to be assured of
a
> great many things, such as that e.g. this fragment isn't in the scope of
>
> #define char const

Right -- that was supposed to be the intended-to-be-humorous-but-telling
point (that one never can tell without a lot of context what 'only thing'
a piece of C++ can mean, due in part but not only to the preprocessor --
a snide-but-valid implied criticism of C++ [and C]).

But it does fall totally flat on re-reading.  My apologies.


Alex





From JollyRoger at hotmail.com  Sun Mar  4 21:42:08 2001
From: JollyRoger at hotmail.com (Bartok)
Date: Sun, 4 Mar 2001 19:42:08 -0700
Subject: Newbie question about nested lists and constructor functions...
Message-ID: <W1Do6.4683$ke1.121750@jekyl.ab.tac.net>

I need some help with nested lists when using class constructors to
initialize their contents.  I found the tutorial to be of not much help in
this area and have tried everything that I can think of.

I understand the process as I am an experienced C++ programmer but I seem to
be having some difficulty using the correct syntax.  Can anyone please help?

Thanks.

Norm





From thomas at xs4all.net  Fri Mar  9 22:51:54 2001
From: thomas at xs4all.net (Thomas Wouters)
Date: Sat, 10 Mar 2001 04:51:54 +0100
Subject: gah! I hate the new string syntax
In-Reply-To: <jwbnews-8E2915.18132009032001@corp.supernews.com>; from jwbnews@scandaroon.com on Fri, Mar 09, 2001 at 06:13:20PM -0800
References: <mailman.983615170.23313.python-list@python.org> <mailman.984014353.5058.python-list@python.org> <BcCp6.37817$1D5.1582482@e420r-atl1.usenetserver.com> <98amlj01sp6@news2.newsguy.com> <98asrq$645$1@panix6.panix.com> <G5eq6.47941$ML1.2142034@e420r-atl2.usenetserver.com> <jwbnews-8E2915.18132009032001@corp.supernews.com>
Message-ID: <20010310045154.K404@xs4all.nl>

On Fri, Mar 09, 2001 at 06:13:20PM -0800, John W. Baxter wrote:

> And at least one of the US Clipper ships finished a New York to San 
> Fransicso voyage under the command of the Captain's widow.  The captain 
> died en route.  The wife was aboard as navigator (being better at the 
> task than the mate), and the crew made it clear that things would go 
> much better with her in command than with the mate in command.  Sorry, I 
> don't have a reference handy (and the right Google search for this one 
> could be a bit tricky to create).

Not at all:

http://www.propellerclubhq.com/wpc/marys.htm

Recipe: type in "clipper captain wife navigator" and click on "I'm Feeling
Lucky".

At least, the story is so similar that I assume it's the same one :)

-- 
Thomas Wouters <thomas at xs4all.net>
The Python Software Foundation: http://www.python.org/psf.html
Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From erno-news at erno.iki.fi  Tue Mar  6 09:37:28 2001
From: erno-news at erno.iki.fi (Erno Kuusela)
Date: 06 Mar 2001 16:37:28 +0200
Subject: I come to praise .join, not to bury it...
References: <mailman.983485923.1136.python-list@python.org> <97p7iq$kk2$1@nntp6.u.washington.edu> <97rka002k9@news1.newsguy.com> <kuu257kb1n.fsf@lasipalatsi.fi> <982a4f0su3@news1.newsguy.com> <kuofvfjb77.fsf@lasipalatsi.fi> <1e6p6.27206$1D5.1119285@e420r-atl1.usenetserver.com>
Message-ID: <kuk863j74n.fsf@lasipalatsi.fi>

In article <1e6p6.27206$1D5.1119285 at e420r-atl1.usenetserver.com>,
"Steve Holden" <sholden at holdenweb.com> writes:

| "Erno Kuusela" <erno-news at erno.iki.fi> wrote in message
| news:kuofvfjb77.fsf at lasipalatsi.fi...

|| it could also be argued that python uses whitespace instead
|| of braces because of vague aesthetic shared by lots of people. :)

| Well it could be argued, but you could also look at the usability data which
| (I understand) was the basis of this choice when Python was originally
| designed.

aha! clearly, we need a usability study on string methods.

  -- erno



From ws-news at gmx.at  Tue Mar 13 17:14:04 2001
From: ws-news at gmx.at (Werner Schiendl)
Date: Tue, 13 Mar 2001 23:14:04 +0100
Subject: This math scares me
References: <3aad4062.1290021@News.CIS.DFN.DE> <20010312170356.B3513@better.net> <mailman.984480856.18097.python-list@python.org> <984486395.407379@newsmaster-04.atnet.at> <slrn9asf1d.ja3.qrczak@qrnik.zagroda> <984518199.127140@newsmaster-04.atnet.at> <jwbnews-09D75F.13313313032001@corp.supernews.com>
Message-ID: <984521930.27710@newsmaster-04.atnet.at>

> Bill Gates got this right in the early days, delivering two Basic
> interpreters:  binary floating point for speed, and decimal floating
> point for money and such.  (I never looked at how well his interpreters
> did...I just used them.)

Does this make a difference, as soon as arithmetics takes place?
I think not, it will not be _exact_ results in both ways.
Just storing and retrieving numbers is probably not a very common task
(why convert it for nothing? maybe for reduced size compared to a string...)

My work is not in financial software, but I would certainly not use floating
point for this kind of problem.
I think Windows supports for a 'Currency' data type with some APIs but never
used them...

regards
werner




From pfortin at pfortin.com  Wed Mar 14 09:50:42 2001
From: pfortin at pfortin.com (Pierre Fortin)
Date: Wed, 14 Mar 2001 09:50:42 -0500
Subject: PyQT-2.3 compile fails...
References: <3AAEBCCA.4E53671@pfortin.com> <98n3if$kol$1@news1.xs4all.nl>
Message-ID: <3AAF8542.B79A880@pfortin.com>

Boudewijn Rempt wrote:
> 
> Pierre Fortin <pfortin at pfortin.com> wrote:
> > Hi,
> 
> > Anyone have a clue as to why others can compile PyQT-2.3; but I get errors...?
> > I've also tried the CVS files...
> 
> > I'd hate to think this is a symlink related problem:
> > /usr/local/src --> /hd/b6/src
> 
> > Thanks,
> > Pierre
> 
> > # make
> > make  all-recursive
> > make[1]: Entering directory `/hd/b6/src/Python/PyQt-2.3'
> > Making all in qt
> > make[2]: Entering directory `/hd/b6/src/Python/PyQt-2.3/qt'
> > /bin/sh ../libtool --mode=compile c++ -DHAVE_CONFIG_H -I. -I. -I..
> > -I/usr/local/include/python2.0 -I/usr/local/include/sip -I/usr/local/qt/include
> > -I/usr/X11R6/include  -fno-exceptions -O2 -c sipqtQString.cpp
> > c++ -DHAVE_CONFIG_H -I. -I. -I.. -I/usr/local/include/python2.0
> > -I/usr/local/include/sip -I/usr/local/qt/include -I/usr/X11R6/include
> > -fno-exceptions -O2 -Wp,-MD,.deps/sipqtQString.pp -c  -fPIC -DPIC
> > sipqtQString.cpp -o sipqtQString.lo
> > sip/qstring.sip: In function `int sipConvertTo_QString(PyObject *, QString **,
> > int, int *)':
> > sip/qstring.sip:456: no matching function for call to `QString::setUnicodeCodes
> > (short unsigned int *&, int &)'
> > make[2]: *** [sipqtQString.lo] Error 1
> > make[2]: Leaving directory `/hd/b6/src/Python/PyQt-2.3/qt'
> > make[1]: *** [all-recursive] Error 1
> > make[1]: Leaving directory `/hd/b6/src/Python/PyQt-2.3'
> > make: *** [all-recursive-am] Error 2
> 
> Has sip been compiled with Qt support? Do you have a Qt > 2.x?

I would hope so...  ./configure --help gives:
  --enable-qt             enable Qt signal/slot support [default=yes]

I picked up all the *-2.3 stuff from TrollTech before starting any compiles...
sip & qt were made/installed without problem.  

Thanks,
Pierre          

> --
> 
> Boudewijn Rempt  | http://www.valdyas.org

-- 
Linux (Up 26 days) -- Reboots are for system upgrades... not Windows X^P
Last reboot reason:  02/14/01: testing startup changes for DSL


From vlouie at telusplanet.net  Sun Mar 25 16:46:00 2001
From: vlouie at telusplanet.net (Victor Louie)
Date: Sun, 25 Mar 2001 21:46:00 GMT
Subject: sorting dictionaries?
Message-ID: <sGtv6.59584$Lm2.6379598@news0.telusplanet.net>

Hi

I was just wondering if anyone knows how to sort a dictionary by
value...since the dictionary is <key: value> and where my value is a list.

Thanks in advance

Victor





From f8dy at yahoo.com  Fri Mar 30 14:40:27 2001
From: f8dy at yahoo.com (Mark Pilgrim)
Date: Fri, 30 Mar 2001 14:40:27 -0500
Subject: htmllib samples
References: <3AC2F48E.4EF77C71@aon.at>
Message-ID: <9a2nb4$3d8nd$1@ID-77331.news.dfncis.de>

"Oliver Vecernik" <vecernik at aon.at> wrote in message
news:3AC2F48E.4EF77C71 at aon.at...
> Hi,
>
> has anyone got some samples or point me to where to find them on how to
> implement this module?

http://diveintopython.org/dialect_divein.html

--
-M
You're smart; why haven't you learned Python yet?
http://diveintopython.org/




From duncan at rcp.co.uk  Thu Mar 15 05:09:05 2001
From: duncan at rcp.co.uk (Duncan Booth)
Date: Thu, 15 Mar 2001 10:09:05 +0000 (UTC)
Subject: [Q] Bug in Python?
References: <t6Zr6.92712$__6.18256422@typhoon.southeast.rr.com>
Message-ID: <Xns906566DC8AB40duncanrcpcouk@127.0.0.1>

"Daehyok Shin" <DSHIN at nc.rr.com> wrote in <t6Zr6.92712$__6.18256422
@typhoon.southeast.rr.com>:

>Recently, I met a strange behavior of Python 6.1.
>
Will someone please lock up that time machine!


From TCHUR at doh.health.nsw.gov.au  Sun Mar 11 19:23:24 2001
From: TCHUR at doh.health.nsw.gov.au (Tim CHURCHES)
Date: Mon, 12 Mar 2001 11:23:24 +1100
Subject: Subclass factories?
Message-ID: <saacb1ac.073@dohprimo.ccf.health.nsw.gov.au>

Is it Pythonic to use a factory function to create subclasses? The context is a system designed to be used by non-programmers. What I would like to do is to create a subclass factory function which accepts various parameters (sorry, arguments) from the end-user, does various sanity checks on these arguments, and then creates a new class which inherits from a predefined superclass and uses the arguments to overide various attributes and/or methods from the superclass, and finally to create an instance of the new class (and return that instance). In other words, allow end users to create subclasses without having to understand the class statement (remember, they are non-programmers). 

Or perhaps I should rephrase the question: "Is it possible to use a factory function to create instances of a class in which that instance's methods (as well as its attributes) are specialised? Such specialisation would normally be done by subclassing. Can that be wrapped in a function?"

If the subclass factory approach is reasonable, what is the best way to do it? So far I been assembling the necessary statements for the new class definition into a string then exec'ing that string, but that seems rather inelegant. I'm sure there must be a better way.

The overall design goal is to provide end-users with a set of functions which they can issue at the Python prompt (or in scripts) to get most things done (i.e. "most things" within the proscribed problem domain), without preventing them from using more general Python syntax (such as the class statement) if/when they progress to that.

BTW,  regarding the "Einstein's Riddle" thread (and not wanting in any way to promote off-topic traffic...), surely no-one owns the fish because Schrodinger's cat ate it?

Tim C





From phil at river-bank.demon.co.uk  Thu Mar  1 06:18:35 2001
From: phil at river-bank.demon.co.uk (Phil Thompson)
Date: Thu, 01 Mar 2001 11:18:35 +0000
Subject: COMMERCIAL: BlackAdder v1.0Beta2 Available
Message-ID: <3A9E300B.8EA0F0C2@river-bank.demon.co.uk>

The second Beta version of the BlackAdder Windows/Linux Python/Qt
development environment is now available from
http://www.thekompany.com/products/blackadder/. Revised screenshots are
at http://www.thekompany.com/products/blackadder/screenshots/.

The major changes since the previous Beta include:

- The user interface has been re-designed to make it much easier to
navigate and to make better use of the available screen space.

- The debugger now automatically displays local and global objects of
any selected frame. The objects displayed can be filtered according to
their type.

- Many features (particularly the Python editor) are now configurable
from the Preferences dialog. These include hard/soft tabs, tab width,
automatic indentation, fonts, colors, visible white space and
indentation guides.

- The inclusion of Qt v2.2.4.

Phil



From KatMouse at gmx.de  Mon Mar 19 16:05:31 2001
From: KatMouse at gmx.de (Eduard Hiti)
Date: Mon, 19 Mar 2001 22:05:31 +0100
Subject: How to increase PythonWin v2.0: COM browser limits?
References: <994psh$so5$1@news1.xs4all.nl>
Message-ID: <995sda$beu$06$1@news.t-online.com>

In my experience this is a bogus error message. It could be that the problem
lies elsewhere: maybe a registry corruption or a Pythonwin bug.

Eduard

"Twan van der Schoot" <twanvds at xs4all.nl> wrote in message
news:994psh$so5$1 at news1.xs4all.nl...
| I'm trying to use the COM Browser in Pythonwin version 2.0.  But I run
into
| limitations of some kind.  As a Python newby I dutifully scanned FAQs and
| newsgroups, but alas!
|
| To reproduce the problem: On a heavily used NT 4.0 system with a lot of
| registered COM stuff, start Pythonwin version 2.0 go to the menu entry:
COM
| Browser to start the COM browser and expand the "Registered Categories"
| branch in the tree view.
|
| Pythonwin shows in the interactive window a backtrace on an exception.
The
| culprit raising this exception was the response:
|
| File "D:\Python20\win32com\client\util.py", line 53, in __GetIndex
|     result = self._oleobj_.Next(1)
| pywintypes.com_error: (-2147024882, 'Not enough storage is available to
| complete this operation.', None, None)
| win32ui: Exception in OnNotify() handler
|
| Something suggests that the number of registered objects is too large for
| the COM browser.  As I've -of course- the sources available, I was wonder
if
| there is some parameter which I can increase to solve the problem.  Or am
I
| running into limitations of the treeview control of MS Windows?
|
| TIA
|
| Twan van der Schoot
| twanvds at NOSPAM.xs4all.nl
|
|




From mwh21 at cam.ac.uk  Fri Mar  9 07:47:01 2001
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 09 Mar 2001 12:47:01 +0000
Subject: Can anyone recomend a good intoduction to C...
References: <mailman.983894288.19965.python-list@python.org> <ewwp6.1466$y6.241005@ruti.visi.com> <slrn9aeblb.cg1.scarblac@pino.selwerd.nl> <djagat4kk3cs9hg0c3k39cudk6goblhkg6@4ax.com> <989mbk$24nuv$1@fido.engr.sgi.com> <slrn9ah66m.k09.qrczak@qrnik.zagroda>
Message-ID: <m3hf139kje.fsf@atrus.jesus.cam.ac.uk>

qrczak at knm.org.pl (Marcin 'Qrczak' Kowalczyk) writes:

> Language interoperability is fun.
> 
> working-on-Haskell<->Python-binding-ly y'rs

Really?  I've had thoughts about doing that, but drifted away from
haskell for a while.  Do you have anything to show for it yet?

Cheers,
M.

-- 
  ARTHUR:  Don't ask me how it works or I'll start to whimper.
                   -- The Hitch-Hikers Guide to the Galaxy, Episode 11


From jeremy at alum.mit.edu  Thu Mar  1 15:52:41 2001
From: jeremy at alum.mit.edu (Jeremy Hylton)
Date: Thu, 1 Mar 2001 15:52:41 -0500 (EST)
Subject: A counter-counter-proposal for PEP 236: #pragma( nested_scopes)
In-Reply-To: <epxn6.16181$Qb7.2640372@newsb.telia.net>
References: <j47l29lpo8.fsf@informatik.hu-berlin.de>
	<mrwn6.16173$Qb7.2637930@newsb.telia.net>
	<0Swn6.3324$TW.17061@tor-nn1.netcom.ca>
	<epxn6.16181$Qb7.2640372@newsb.telia.net>
Message-ID: <15006.46745.206128.996289@w221.z064000254.bwi-md.dsl.cnc.net>

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

  FL> But Python isn't supposed to grow lots of pragma-style settings.
  FL> There shouldn't be any "other things".  There Should Only Be One
  FL> Language.

  FL> We're talking about two kinds of "directives" here:

  [skipping #1]  

  FL> -- Experimental compiler features, which if successful will
  FL>    become standard in the next release.  Again, "directive"
  FL>    doesn't work, since it makes "experimental" stuff look
  FL>    "optional". 

A word of clarification: __future__ isn't about experimental
features.  Instead it is for bleeding edge users to use a new feature
in release X even though the feature won't be standard until release
Y, where X < Y.

If we want to do experimental stuff, we should probably do something
like:

from __possible_world__ import __future__
from __future__ import ...

Jeremy



From ws-news at gmx.at  Tue Mar  6 11:27:47 2001
From: ws-news at gmx.at (Werner Schiendl)
Date: Tue, 6 Mar 2001 17:27:47 +0100
Subject: Can anyone recomend a good intoduction to C...
References: <mailman.983894288.19965.python-list@python.org>
Message-ID: <983896344.701456@newsmaster-04.atnet.at>

Hi Laura,

The only reason I can imagine at this point to go with plain C is that there
are a lot more platforms that support C than do support C++. This is mainly
a topic in embedded development and realtime platforms.

Since you can do almost anything in C++ that can be done in C with even the
same syntax I think personally it is better to get the high level concepts
first and learn to use of-the-shelve components like the standard library.
The standard library of C++ is in my opinion much more intuitive and easy to
use for a novice programmer than all that traditional C stuff. So I think,
given a good guide like the one I recommended, one will be up and running
much faster. In addition, you will not get into a lot of common pitfalls
common in C programming.

I further think it is easier to move 'back' from C++ to C, when the need
arises, than vice versa. This is simply because you are forced to use a
different technique if the one you are used to does not work on that
platform.
Personally I started with procedural programming and moved to the object
oriented technique afterwards. This is why I think it is hard to get used to
objects when you are used to malloc, free, memcpy and friends.

Did I miss an important point?

regards
Werner

Laura Creighton <lac at cd.chalmers.se> wrote in message
news:mailman.983894288.19965.python-list at python.org...
> Werner Schiendl <ws-news at gmx.at> writes:
> >If you have a good reason to choose C over C++, take the K&R book the
others
> >recommend.
> >
> >Otherwise you should consider going straight to C++.
> >This will save you hours of headache with C memory stuff, string
handling,
> >and so on.
>
> This leaves the novice with the idea that C programmers, by and large
> all became happy C++ programmers and are pleased as punch with the
> new language.  The reason to learn C is to eventually learn C++.
> This is misleading.
>
> There are a very large number of people who think that C
> is a terrific language; extremely good at what it does, and well worth
> studying, in part to understand what it is that a language designer
> does when he *gets* *it* *right*, while also believing that studying C++
is
> of particular use when you want to see what happens when a language
> designer *gets* *it* *wrong*.  Or when forced to do so at gunpoint.
>
> Laura Creighton
>




From aleaxit at yahoo.com  Thu Mar 15 08:00:14 2001
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 15 Mar 2001 14:00:14 +0100
Subject: I come to praise .join, not to bury it...
References: <mailman.983485923.1136.python-list@python.org> <97p7iq$kk2$1@nntp6.u.washington.edu> <97rka002k9@news1.newsguy.com> <3AAF7E47.294617E1@sage.att.com>
Message-ID: <98qeco02hn7@news1.newsguy.com>

"Garry Hodgson" <garry at sage.att.com> wrote in message
news:3AAF7E47.294617E1 at sage.att.com...
> Alex Martelli wrote:
>
> > I have not heard ANY _technical_ arguments opposed to
> > this last time the discussion went around -- nothing but
> > vague aesthetic, "it should be the other way", "I find it
    [snip]
> don't be too quick to write off Aesthetics.  it's a significant
> part of what attracted me to python (i'd seen enough curly braces
> and semicolons to last me a lifetime).  avoiding scaring off newbies
> until they've had time to appreciate other virtues is not a Bad Thing.

And then again, the _appearance_ that too much emphasis
was being placed on syntax-sugary issues kept me *AWAY*
from Python for longer than it should have -- sure, sure,
I could save some punctuation, but, *WHAT DID I CARE*?!

Good thing (for me) that I eventually looked beyond the
pretty face, of course:-).  But, appearing to care about
looks more than about substance _is_ going to keep away
some people (just as it is going to attract others), so,
it's not a clear "win" either way, even if you just frame
it as a popularity-race.  Focus on *substance*, on the
other hand, is going to be a win for all who _are_ involved
with Python -- not just those who happen to be on the
"I like it" side vs the "I like it not" one (and I've
never seen any argument about _aesthetics_ [as opposed to
_usability_ -- human-factors studies DO have a substantial
side too:-)] that doesn't boil down to such trifles]).


Alex







From trentm at ActiveState.com  Fri Mar  9 18:34:50 2001
From: trentm at ActiveState.com (Trent Mick)
Date: Fri, 9 Mar 2001 15:34:50 -0800
Subject: IPC9 quotes wanted
In-Reply-To: <slrn9ai9ln.10q.amk@mira.erols.com>; from amk@mira.erols.com on Fri, Mar 09, 2001 at 06:38:59PM +0000
References: <slrn9ai9ln.10q.amk@mira.erols.com>
Message-ID: <20010309153450.D4002@ActiveState.com>

On Fri, Mar 09, 2001 at 06:38:59PM +0000, A.M. Kuchling wrote:
> I intended to put a note on the message board about this, but
> forgot...  If you were at IPC9 and heard someone say something
> quotable, please let me know who it was and what they said, so that it
> can be added to the quotation file.  Thanks!
> 
> (I reserve the right to not add quotations, so don't feel bad if your
> submitted quotation doesn't get in.)

Tim Peters:
    You can't stop people from complaining, but you *can* control what they
    complain about.  -- while moderating the "Overhaul Python's Numerical
    Model" seminar.

Bruce Eckel:
    Perl is like vise grips. You can do anything with it but it is the wrong
    tool for every job.


Trent


-- 
Trent Mick
TrentM at ActiveState.com



From grante at visi.com  Sat Mar 31 19:32:22 2001
From: grante at visi.com (Grant Edwards)
Date: Sun, 01 Apr 2001 00:32:22 GMT
Subject: Know of Substantial Apps Written in Python?
References: <mailman.986080180.25333.python-list@python.org>
Message-ID: <qGux6.9643$SB2.918159@ruti.visi.com>

In article <mailman.986080180.25333.python-list at python.org>, Tim Peters wrote:
>[Moshe Zadka]
>> ...
>> Python would never fit on a floppy.
>
>Sure it can!  Python + the entire base std library (top level
>of Lib/) still fits on a floppy with ease, if you use any
>common compression scheme.  Just for fun, I just built a zip
>file with that stuff for the Windows 2.1b2 distribution, and it
>came in at just over a tad above 1Mb (so about 70% of one
>floppy).

I assume that Moshe meant that it will never fit on a
single-floppy Linux system.  Python isn't much good with no OS
to run it.  Ash/bash is already there as a command line shell,
so there's no incremental cost in having startup scripts
written in shell.  A 1M incremental cost for writing them in
Python would be out of the question.

Other than increased maintenance costs, there's no reason you
can't stick with ash/bash/sh scripts for floppy systems and
Python scripts for full-up systems.

-- 
Grant Edwards                   grante             Yow!  With this weapon
                                  at               I can expose fictional
                               visi.com            characters and bring about
                                                   sweeping reforms!!


From andreas at andreas-jung.com  Wed Mar 21 17:13:59 2001
From: andreas at andreas-jung.com (Andreas Jung)
Date: Wed, 21 Mar 2001 17:13:59 -0500
Subject: os.system output
References: <B7C74E1A6FBAD411BFC60002B309DDF4E42F@mailbox.juilliard.edu> <20010321180207.A24260@father>
Message-ID: <000001c0b266$10274030$9865fea9@SUXLAP>

----- Original Message ----- 
From: "Kalle Svensson" <kalle at gnupung.net>

> 
> Check out os.popen().
> 
> >>> dirs = os.popen("ls").read()

or commands.getstatusoutput() which is actually a wrapper around os.popen()

Andreas 
Digital Creations




From s2mdalle at titan.vcu.edu  Thu Mar  1 18:53:33 2001
From: s2mdalle at titan.vcu.edu (David Allen)
Date: Thu, 01 Mar 2001 23:53:33 GMT
Subject: Tkinter:  image doesn't exist.
Message-ID: <1iBn6.1691$Ry6.649664@typhoon2.ba-dsg.net>

When I run this line of code:

self.symbol = self.widget.create_image(x, y, image=fooimage)

I get this error:

  File "./FooModule.py", line 87, in __init__
    self.symbol = self.widget.create_image(x, y, image=fooimg)
  File "/usr/lib/python1.5/lib-tk/Tkinter.py", line 1223, in create_image
    return self._create('image', args, kw)
  File "/usr/lib/python1.5/lib-tk/Tkinter.py", line 1216, in _create
    (self._w, 'create', itemType)
TclError: image "137289752" doesn't exist

I've talked to several people and they though this
might be because the image is getting garbage 
collected, but I've verified that GC is *not* the
issue. I replaced fooimg with an actual PhotoImage
constructor (fooimg is a PhotoImage) with valid
data, and it still does the exact same thing.

What kind of a crappy error message is "image doesn't
exist?"  How would I fix it?

Any help would be greatly appreciated.
--
David Allen
http://opop.nols.com/
----------------------------------------
Der Horizont vieler Menschen ist ein Kreis mit Radius Null -- und das 
nennen sie ihren Standpunkt.


From mkc at mathdogs.com  Mon Mar 26 19:42:21 2001
From: mkc at mathdogs.com (Mike Coleman)
Date: 26 Mar 2001 18:42:21 -0600
Subject: fixing map(None, [1,2,3]) ?
In-Reply-To: <XFMail.20010326151203.shaleh@valinux.com>
References: <XFMail.20010326151203.shaleh@valinux.com>
Message-ID: <87y9ts2gb6.fsf@mathdogs.com>

"Sean 'Shaleh' Perry" <shaleh at valinux.com> writes:
> On 26-Mar-2001 Mike Coleman wrote:
> > Shouldn't "map(None, [1,2,3])" return "[(1,), (2,), (3,)]"?
> > 
> > Is there any hope that this could be fixed?
> > 
> map runs the function on each element of the list it is passed and returns an
> element of a new list.  Why should it magically return a list of one-tuples?

The (documented) behavior of map when 'None' is given instead of a function is
special.  It would be very useful, too, except for the problem cited above:

   $ python
   Python 1.5.2 (#0, Apr  3 2000, 14:46:48)  [GCC 2.95.2 20000313 (Debian GNU/Linux)] on linux2
   Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
   >>> map(None, [1, 2, 3], [4, 5, 6])
   [(1, 4), (2, 5), (3, 6)]
   >>> map(None, [1, 2, 3], [4, 5, 6], [7, 8, 9])
   [(1, 4, 7), (2, 5, 8), (3, 6, 9)]
   >>> map(None, [1, 2, 3], [4, 5, 6], [7, 8, 9], [10, 11, 12])
   [(1, 4, 7, 10), (2, 5, 8, 11), (3, 6, 9, 12)]

   >>> map(None, [1, 2, 3])
   [1, 2, 3]
   # shouldn't that be [(1,), (2,), (3,)]?

   >>> map(None) 
   Traceback (innermost last):
     File "<stdin>", line 1, in ?
   TypeError: map() requires at least two args
   # returning [] in this case would be nice for completeness

(Certainly I could be missing something here.)

--Mike

-- 
Mike Coleman, mkc at mathdogs.com
  http://www.mathdogs.com -- problem solving, expert software development



From hamish_lawson at yahoo.co.uk  Tue Mar 27 05:38:55 2001
From: hamish_lawson at yahoo.co.uk (=?iso-8859-1?q?Hamish=20Lawson?=)
Date: Tue, 27 Mar 2001 11:38:55 +0100 (BST)
Subject: running code from "Scripting The Web with Python"
Message-ID: <20010327103855.23536.qmail@web206.mail.yahoo.com>

Eugene Leitl wrote:

> I've snipped the following code from Guido's
> http://www.w3j.com/6/s3.vanrossum.html presentation:
> ... I can't run it due to syntax errors


Wow, they really messed up the formatting on that! Below is the
properly formatted version.

Hamish Lawson

---

import StringIO
import SimpleHTTPServer

BaseClass = SimpleHTTPServer.SimpleHTTPRequestHandler

CounterTemplate = """ <H1>Server
Statistics</H1>

This <A HREF=/index.html>server</A> has been accessed
<b>%d</b> times.  """

count = 0

class MyRequestHandler(BaseClass):

    def do_GET(self):
        global count
        count = count + 1
        BaseClass.do_GET(self)

    def send_head(self):
        if self.path == "/counter.html":
            return self.send_counter()
        else:
            return BaseClass.send_head(self)

    def send_counter(self):
        self.send_response(200)
        self.send_header("Content-type", "text/html")
        self.end_headers()
        text = CounterTemplate % count
        return StringIO.StringIO(text)

def test():
    SimpleHTTPServer.test(MyRequestHandler)

test() 


____________________________________________________________
Do You Yahoo!?
Get your free @yahoo.co.uk address at http://mail.yahoo.co.uk
or your free @yahoo.ie address at http://mail.yahoo.ie



From fredrik at pythonware.com  Thu Mar 22 11:10:12 2001
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Thu, 22 Mar 2001 16:10:12 GMT
Subject: New PEP: The directive statement
References: <mailman.985085006.24236.python-list@python.org> <slrn9bi7q3.2sa.hzhu@rocket.knowledgetrack.com> <j4snk6yxiv.fsf@informatik.hu-berlin.de>
Message-ID: <Etpu6.3010$4N4.414632@newsc.telia.net>

Martin von Loewis wrote:
> Generalization is a dangerous thing: you can too easily cross the
> border to over-generalization.

also known as "hypergeneralization" in the Python universe:

http://www.google.com/search?q=hypergeneralization

Cheers /F




From cribeiro at mail.inet.com.br  Sat Mar 31 09:15:10 2001
From: cribeiro at mail.inet.com.br (Carlos Alberto Reis Ribeiro)
Date: Sat, 31 Mar 2001 11:15:10 -0300
Subject: Know of Substantial Apps Written in Python?
In-Reply-To: <3AC55CC8.44850ECF@seebelow.org>
References: <3ac4e4a8_1@news4.newsfeeds.com>
 <KE6x6.415$SR6.235983@typhoon1.ba-dsg.net>
 <3ac4ff12$1_2@news4.newsfeeds.com>
Message-ID: <5.0.2.1.0.20010331111123.021eb0d0@mail.inet.com.br>

At 22:27 30/03/01 -0600, Grant Griffin wrote:
>Rumor has it that Google is implemented in Python.  (OK, maybe not _all_
>of it. <wink>)

Not only Google, but substantial parts of Yahoo! site, including the entire 
yahoo mail interface (you can actually see th .py endings in many pages).

Also in Python:

- Infoseek/Ultraseek server (core library in C, bind to Python)
- The RedHat installer
- MojoNation

As for Sketch, the FAQ answer your question. The main problem is that the 
stable version of Tk was not ported to Windows, and there are some low 
level stuff that wasn't ported either.

Please also refer to the Software Carpentry contest.


Carlos Ribeiro





From moshez at zadka.site.co.il  Sat Mar 24 05:16:37 2001
From: moshez at zadka.site.co.il (Moshe Zadka)
Date: Sat, 24 Mar 2001 12:16:37 +0200
Subject: How 2 get rid of console
In-Reply-To: <tborq16qctlm42@corp.supernews.com>
References: <tborq16qctlm42@corp.supernews.com>
Message-ID: <E14gl65-0005zC-00@darjeeling>

On Sat, 24 Mar 2001 05:00:43 -0500, "Bill Walker" <bwalker at earth1.net> wrote:
> I'm using Tkinter to build a GUI application. How do I get rid of the
> distracting dos console under the Tk window?
 
Run it with "pythonw" and not with "python".
One way to do it is to rename the script "something.pyw".
Usually, it's better to say what platform you're working on instead of
letting people guess (Win32?)
-- 
"I'll be ex-DPL soon anyway so I'm        |LUKE: Is Perl better than Python?
looking for someplace else to grab power."|YODA: No...no... no. Quicker,
   -- Wichert Akkerman (on debian-private)|      easier, more seductive.
For public key, finger moshez at debian.org  |http://www.{python,debian,gnu}.org



From mlh at idi.ntnu.no  Fri Mar  2 16:31:33 2001
From: mlh at idi.ntnu.no (Magnus Lie Hetland)
Date: Fri, 2 Mar 2001 22:31:33 +0100
Subject: Thinking Pythonically (was Re: gah! I hate the new string syntax)
References: <mailman.983485923.1136.python-list@python.org> <B7In6.16281$Qb7.2658902@newsb.telia.net> <97ohh2$polt0$1@ID-77331.news.dfncis.de> <97opsb$nhu$1@nntp6.u.washington.edu> <97oupq$qijrh$1@ID-77331.news.dfncis.de>
Message-ID: <97p3mh$bkm$1@tyfon.itea.ntnu.no>

"Mark Pilgrim" <f8dy at yahoo.com> wrote in message
news:97oupq$qijrh$1 at ID-77331.news.dfncis.de...
[snip]
> >>> [line for line in output if line[:4]=='Tag:']
> ['Tag: /vobs/intranet\\012', 'Tag: /vobs/project\\012']
>
> Now I think: OK, I have a list of the lines I want, but I just want the
> name, not the "Tag:" prefix or all the whitespace.
>
> >>> vobs = [line.replace('Tag:', '').strip() for line in output if
> line[:4]=='Tag:']
> >>> vobs
> ['/vobs/intranet', '/vobs/project']

I can't resist fiddling with little snippets of code... Why don't you
use:

>>> vobs = [line[4:-1] for line in output if line[:4] == 'Tag:']

Oh, well...

--

  Magnus Lie Hetland          http://www.hetland.org

 "Reality is what refuses to disappear when you stop
  believing in it"                 -- Philip K. Dick





From thomas at xs4all.net  Fri Mar  9 16:20:34 2001
From: thomas at xs4all.net (Thomas Wouters)
Date: Fri, 9 Mar 2001 22:20:34 +0100
Subject: News from Python9!!!
In-Reply-To: <20010308161415.C30277@mediasupervision.de>; from gregor@mediasupervision.de on Thu, Mar 08, 2001 at 04:14:15PM +0100
References: <NUCp6.65206$Z8.11697516@typhoon.southeast.rr.com> <20010308161415.C30277@mediasupervision.de>
Message-ID: <20010309222034.I404@xs4all.nl>

On Thu, Mar 08, 2001 at 04:14:15PM +0100, Gregor Hoffleit wrote:
> On Thu, Mar 08, 2001 at 03:21:17AM +0000, Don Tuttle wrote:
> > 4)And a bugfix release of Python 2.0 has been released. It includes PPM
> > (Programmer's Package Manager ) to make finding and installing Python
> > modules simple.

> Shouldn't this read 'bugfix release of ActivePython 2.0' ?

It should. A bugfix release of 2.0 is forthcoming (for various reasons), and
Guido did more or less announce that intention at the conference, but it's
not released yet.

-- 
Thomas Wouters <thomas at xs4all.net>

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